Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Loss of Heterozygosity Drives Adaptation in Hybrid Yeast
Caiti S Smukowski Heil1 Christopher G DeSevoDagger2 Dave A Paisect2 Cheryl M Tuckerpara2 Margaret LHoangk34 and Maitreya J Dunham1
1Department of Genome Sciences University of Washington Seattle WA2Lewis-Sigler Institute for Integrative Genomics Princeton University Princeton NJ3Department of Embryology Howard Hughes Medical Institute Carnegie Institution Baltimore MD4Department of Biology Johns Hopkins University Baltimore MDDaggerPresent address Exelixis Dallas TXsectPresent address BioNano Genomics San Diego CAparaPresent address Wall High School Wall Township NJkPresent address Nanostring Seattle WA
Corresponding author E-mail maitreyauwedu
Associate editor Jianzhi Zhang
Abstract
Hybridization is often considered maladaptive but sometimes hybrids can invade new ecological niches and adapt tonovel or stressful environments better than their parents The genomic changes that occur following hybridization thatfacilitate genome resolution andor adaptation are not well understood Here we examine hybrid genome evolutionusing experimental evolution of de novo interspecific hybrid yeast Saccharomyces cerevisiae Saccharomyces uvarumand their parentals We evolved these strains in nutrient-limited conditions for hundreds of generations and sequencedthe resulting cultures identifying numerous point mutations copy number changes and loss of heterozygosity (LOH)events including species-biased amplification of nutrient transporters We focused on a particularly interesting examplein which we saw repeated LOH at the high-affinity phosphate transporter gene PHO84 in both intra- and interspecifichybrids Using allele replacement methods we tested the fitness of different alleles in hybrid and S cerevisiae strainbackgrounds and found that the LOH is indeed the result of selection on one allele over the other in both S cerevisiae andthe hybrids This is an example where hybrid genome resolution is driven by positive selection on existing heterozygosityand demonstrates that even infrequent outcrossing may have lasting impacts on adaptation
Key words hybrid adaptation loss of heterozygosity experimental evolution Saccharomyces uvarum Saccharomycescerevisiae
Introduction
Hybridization is now recognized as a common phenomenonacross the tree of life Historically however the detection ofhybrids has been difficult and its incidence may be under-reported for both plants and animals and almost certainly forcertain eukaryotes like insects and fungi (Bullini 1994 Albertinand Marullo 2012) Its importance as an evolutionary forcehas thus been maligned as hybrids appeared both rare andtypically at a reduced fitness In addition to potential postre-productive barriers the hybrid is theorized to be ill-adaptedto its environment and will also suffer minority cytotype dis-advantage because other hybrids are uncommon and back-crosses to parental species may be unfit (Mallet 2007)However hybrids can have a variety of advantages over theirparents including heterozygote advantage extreme pheno-typic traits and reproductive isolation (usually resulting frompolyploidy) and can thus facilitate adaptation to novel orstressful conditions invade unoccupied ecological nichesand even increase biodiversity
Some hybridization events lead to new hybrid species(Rieseberg 1997 Nolte et al 2005 Mavarez et al 2006
Meyer et al 2006 Soltis and Soltis 2009 Schumer et al2014) whereas most result in introgression from hybrid back-crosses to the more abundant parental species (Dowling et al1989 Taylor and Hebert 1993 Wayne 1993 Grant et al 2005Dasmahapatra et al 2012) Hybridization introduces geneticvariation into a population at orders of magnitude greaterthan what mutation alone can achieve in a sense operating asa ldquomulti-locus macro-mutationrdquo (Grant and Grant 1994Barton 2001 Mallet 2007 Abbott et al 2013) Therefore hy-bridization via introgression polyploidy or homoploid hybridspeciation may offer a rapid strategy for adaptation to chang-ing environmental conditions For example in Darwinrsquosfinches adaptive introgression supplied the morphologicalvariation that allowed the species to survive following an ElNi~no event (Grant and Grant 2010 2002) and in ancienthumans introgression allowed adaptation to high altitudes(Huerta-Sanchez and Casey 2015) among other traits(Racimo et al 2015) The most iconic example comes fromthe hybrid sunflower species Helianthus anomalus Helianthusdeserticola and Helianthus paradoxus from the parentsHelianthus annuus and Helianthus petiolaris These three hy-brid species are locally adapted to extreme desert salt marsh
Article
The Author 2017 Published by Oxford University Press on behalf of the Society for Molecular Biology and EvolutionThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby40) which permits unrestricted reuse distribution and reproduction in any medium provided the original work isproperly cited Open AccessMol Biol Evol 0(0)1ndash17 doi101093molbevmsx098 Advance Access publication March 29 2017 1
and dune habitats respectively and show traits such as in-creased drought or salt tolerance relative to their parents(Heiser 1954 Rieseberg 1991 Schwarzbach et al 2001Rosenthal et al 2002)
Agriculture and industry use both intra- and interspecifichybrids as a tool to increase yield or robustness introduceresistance to pests and create novel phenotype or flavorprofiles For example plant breeders have crossed domesti-cated species to wild species to introduce resistance to avariety of pathogens in wheat potato and canola (Masonand Batley 2015) and almost all maize grown in the UnitedStates is grown from intraspecific hybrid seeds which hasincreased yield and provided improved resistance to bioticand abiotic factors (Crow 1998) Vintners and brewers havecreated interspecific hybrids to select for traits such as loweracetic acid concentration (Bellon et al 2015) and many inci-dental fungal hybrids have been discovered in brewing andindustry including Pichia sorbitophila (Louis et al 2012) andvarious hybrids across the Saccharomyces clade (Gonzalezet al 2006 2008 Muller and McCusker 2009 Hittinger2013 Bellon et al 2015) most notably the lager-brewing yeastSaccharomyces pastorianus (Tamai et al 1998 Dunn andSherlock 2008 Walther et al 2014 Baker et al 2015 Gibsonand Liti 2015 Peris et al 2016) It is presumed that the severeselection pressures exerted during industrial processes haveselected for interspecific hybrid genomes that may be moreable to cope with the extreme environments
At the genomic level hybridization induces chromosomelossaneuploidy chromosomal rearrangements gene losschanges in gene expression changes in epigenetic modificationstransposable element mobilization and large-scale loss of het-erozygosity (LOH) in which the allele of one species is lost andthe allele of the other species is retained and may even beduplicated via gene conversion or break-induced replication(Masly et al 2006 Landry et al 2007 Doyle et al 2008Michalak 2009 Ainouche and Jenczewski 2010 Albertin andMarullo 2012 Abbott et al 2013 Soltis 2013 Borneman et al2014 Soltis et al 2014) These extensive changes can result in achimeric stabilized hybrid although the period of time for ge-nome stabilization to occur can range dramatically (Soltis et al2014) It is unknown whether there are structural and functionalbiases in the ways in which genesalleles are lost or modifiedBoth drift and selection influence the resolution of the hybridgenome but their contributions are difficult to untangle
Researchers have long been exploring the genetics of hy-brid traits in the lab particularly in agricultural crops al-though this is often slowed by infertility and reducedviability in many interspecific hybrids (Perez-Prat and vanLookeren Campagne 2002 Hajjar and Hodgkin 2007Ouyang et al 2010) The Saccharomyces genus which in-cludes the budding yeast Saccharomyces cerevisiae lends itselfparticularly well to experimental study Many hybrids of thisgenus have been discovered in brewing industrial and natu-ral environments indeed the genus itself is speculated tohave been founded by the product of an ancient hybridiza-tion event (Hittinger 2013 Marcet-Houben and Gabaldon2015 Barbosa et al 2016 Leducq et al 2016) Viable interspe-cific hybrids can be created de novo in the lab (Marinoni et al
1999 Greig et al 2002) and their ability to grow mitoticallymeans that the catastrophic postzygotic barriers to speciationthat generally doom other obligate sexually reproducing hy-brids can be avoided This experimental system allows us toobserve evolution in real time in the laboratory environmentand the genetic and genomic tools available in this modelgenus facilitate characterization of the connection betweengenotype and phenotype including fitness
Previous work in our lab group has utilized experimentalevolution to investigate adaptive events in haploid andhomozygous diploid S cerevisiae (Gresham et al 2008Payen et al 2014 Sunshine et al 2015) To investigate genomeevolution post hybridization we utilize an interspecific hybridS cerevisiae Saccharomyces uvarum and its parentals ahomozygous diploid S uvarum and an intraspecific hybridS cerevisiae GRF167 S cerevisiae S288C This allows us tounderstand the impact of varying levels of heterozygosity onadaptation and genome evolution ranging from none(S uvarum and previous S cerevisiae experiments) to intra-specific heterozygosity (S cerevisiae GRF167 S cerevisiaeS288C) to the most extreme case of interspecific hybridsSaccharomyces uvarum is one of the most distantly relatedspecies of S cerevisiae in the Saccharomyces clade separatedby 20 My and 20 sequence divergence at coding sites (Kelliset al 2003 Cliften et al 2006) Despite this extensive diver-gence S cerevisiae and S uvarum are largely syntenic andcreate hybrids though less than 1 of spores are viable(Greig 2009) The two species differ in their stress tolerancesfor example S cerevisiae being more thermotolerant S uva-rum being cryotolerant (Almeida et al 2014) Previous evolu-tion experiments using lab-derived hybrids have revealednovel andor transgressive phenotypes for ammonium limi-tation ethanol tolerance and growth on xylose (Belloch et al2008 Wenger et al 2010 Piotrowski et al 2012 Dunn et al2013) Notably Dunn et al (2013) revealed several LOHevents and a repeatable nonreciprocal translocation that pro-duces a gene fusion at the high-affinity ammonium permeaseMEP2 after selection in ammonium limitation offering insightinto potential mutational events in the adaptation andorstabilization of S cerevisiae S uvarum hybrids
Here we evolved these hybrids and diploids in replicate inthree nutrient-limited conditions for hundreds of genera-tions Using whole genome sequencing we found wholechromosome aneuploidy genome rearrangements copynumber variants de novo point mutations and LOH Wesought to determine how initial heterozygosity affects adap-tation to novel conditions and explore whether neutral orselective forces are influencing the resolution of the hybridgenome over time In particular we investigated a reoccur-ring LOH event observed in both intra- and interspecifichybrids and found support for the hypothesis that LOH atthis locus is due to selection
Results
Experimental Evolution of Hybrid and Parental SpeciesAn interspecific hybrid was created by crossing S cerevisiaeand S uvarum (strains in supplementary table S1
Smukowski Heil et al doi101093molbevmsx098 MBE
2
Supplementary Material online) and evolved in continu-ous culture in the chemostat (Monod 1949 Novick andSzilard 1950a 1950b) In parallel homozygous diploidS uvarum and heterozygous diploid S cerevisiae(GRF167 S288C) were also evolved Each strain wasgrown in two or more replicate independent culturesunder three different nutrient limitationsmdashglucose phos-phate and sulfatemdashfor 85ndash557 generations (median 158)at 30 C except for S uvarum which was unable toachieve steady state in all conditions at 30 C and sowas evolved at 25 C The population sizes were largelysimilar across strains species and conditions
Evolved clones were isolated from each population andsubsequently competed individually against the appropriategreen fluorescent protein (GFP)-tagged ancestor to gaugerelative fitness As expected evolved hybrid and parentalclones generally exhibit higher fitness than their unevolvedancestor with typical relative fitness gains between 20 and30 (tables 1 and 2) To explore whether these fitness gainsare general or condition specific we additionally competedeach hybrid clone in the two nutrient-limited conditions inwhich the clone was not evolved Results are variable withsome clones having negative or neutral fitness in the alternateconditions suggesting condition-specific adaptation andsome clones experiencing fitness gains in multiple conditionssuggesting more general growth benefits (table 1) Only oneclone exhibited fitness gains in all three nutrient environ-ments and no clones have a greater fitness gain in an alter-nate condition than the condition it was evolved in signifyingthat clones are largely specifically adapted to the particularcondition in which they were evolved
Mutations Are Recovered in Both Novel andPreviously Observed Gene Targets in InterspecificHybridsTo identify mutations in the evolved hybrids we generatedwhole genome sequencing data for 16 clones from the endpoints of the evolution experiments (table 1) We thus cap-tured data from a range of nutrient limitations (phosphate 6glucose 3 sulfate 7) and generations (100ndash285 median 154)Each clone had an average of 24 point mutations a numberof which have been previously identified in prior S cerevisiaeevolution experiments For example a nonsynonymous mu-tation in the S cerevisiae allele of the glucose-sensing geneSNF3 has been identified in glucose-limited experiments inS cerevisiae (Kvitek and Sherlock 2013 Selmecki et al 2015)To our knowledge 2027 coding point mutations are uniqueto these experiments (Payen et al 2016)
In evolved parentals we again sequenced one clone fromthe end point of each population In total we sequenced 16clones 6 from each of the 3 nutrients (2 S uvarum diploidsand 4 S cerevisiae diploids) except in glucose limitation inwhich only 2 S cerevisiae populations were sampled Thegenerations ranged from 234 to 557 (median 477) in S uva-rum with an average of 283 mutations per clone and from127 to 190 (median 1665) in S cerevisiae with an average of09 point mutations per clone (table 2) This discrepancy in
point mutations between S cerevisiae and S uvarum may beexplained by differences in generation number
With the limited number of samples we have from hybridand parental clones it is difficult to draw any conclusionsregarding unique point mutations in hybrids Furthermorewe have not tested the fitness of these mutations to provethey are adaptive However one class of mutations that maybe of particular interest in hybrids are genomic mutationsthat may interact with the mitochondria as previous workhas shown that nuclearndashmitochondria interactions can un-derlie hybrid incompatibility (Lee et al 2008 Chou and Leu2010 Meiklejohn et al 2013) Other studies have found thatonly the S cerevisiae mitochondria are retained in S cerevisiae S uvarum hybrids (Antunovics et al 2005) and we reca-pitulate these findings potentially setting the stage for con-flicting interactions between the S uvarum nuclear genomeand the foreign mitochondria We observe several mitochon-dria-related mutations in hybrids in both S cerevisiae andS uvarum alleles For example one point mutation a nonsyn-onymous mutation in the S cerevisiae allele of the mitochon-drial ribosomal protein gene MHR1 was seen in two separateclones independently evolved in phosphate limitation Thisgene may be of particular interest as it was discovered in aprevious screen as being haploproficient (increased fitness of19) in hybrids in which the S cerevisiae allele is missing andthe S uvarum allele is retained (Lancaster S Dunham MJunpublished data) suggesting that this mutation may alteror disable the S cerevisiae protein in some way Another ex-ample involves the gene IRC3 a helicase responsible for themaintenance of the mitochondrial genome which has a non-synonymous mutation in the S uvarum allele in clone Gh3and is deleted in clone Gh2 potentially suggesting that theS uvarum allele is deleterious in the hybrid backgroundAlthough our sample size is small 427 point mutations inhybrids are related to mitochondria function compared with026 in parentals and may represent interesting targets forfurther exploration
Copy Number Variants Frequently Involve theAmplification of Nutrient TransportersYeasts in both natural and artificial environments are knownto frequently experience changes in copy number rangingfrom single genes to whole chromosomes (Dunham et al2002 Gresham et al 2008 Dunn et al 2012 Kvitek andSherlock 2013 Payen et al 2014 Selmecki et al 2015Sunshine et al 2015 Zhu et al 2016) This holds true in ourevolution experiments We observe copy number changesacross all genetic backgrounds (fig 1 supplementary figsS1ndashS3 Supplementary Material online) Clones were com-pared with array comparative genomic hybridization of pop-ulations to confirm that clones are representative ofpopulations (see Materials and Methods) The evolved hybridclones displayed an average of 15 copy number variants(CNVs) per clone (fig 1 table 1 Supplementary fig S3Supplementary Material online) as defined by the numberof segmental or whole chromosome amplificationsdeletions(though it is likely that some of these CNVs were created inthe same mutational event) The evolved S cerevisiae clones
Adaptation in Hybrids doi101093molbevmsx098 MBE
3
Tab
le1
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edH
ybri
dC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Gh
1ch
rXII
I85
2028
Inte
rgen
icce
r12
526
80
60
98(G
)0
356
160
(S)
118
(P)
chrI
I91
1866
917
272
HX
T6
7C
NV
(am
plifi
cati
on
)u
vaG
h2
chrI
V1
1191
9SN
F3N
on
syn
on
ymo
us
D11
4Yce
r10
028
17
62
18(G
)10
48
60
78(S
)11
23
(P)
chrI
II5
1593
GLK
1Sy
no
nym
ou
sT
252T
cer
chrI
V8
8480
19
1211
913
gen
esin
clu
din
gIR
C3
LOH
CN
Vu
valo
stch
rII
9121
439
1747
0H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
chrI
V83
6ge
nes
CN
V(a
mp
lifica
tio
n)
cer
Gh
3ch
rII
8894
21IR
C3
no
nsy
no
nym
ou
sM
333I
uva
124
186
56
047
(G)
176
86
367
(S)
104
6(P
)ch
rII
9124
169
1777
8H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
Ph
1ch
rV2
6939
2In
terg
enic
cer
103
291
86
137
(P)
168
60
78(G
)0
086
043
(S)
chrX
IV7
4668
8In
terg
enic
cer
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
rch
rIX
241
gen
esLO
HC
NV
uva
lost
cer
amp
Ph
2ch
rV4
3277
8G
LC7
Intr
on
cer
124
253
46
024
(P)
151
26
466
(G)
245
61
16(S
)ch
rVII
952
4P
DR
11N
on
syn
on
ymo
us
L383
u
vach
rXV
I23
2879
MR
PL4
0N
on
syn
on
ymo
us
V14
9Eu
vach
rXII
I19
4496
YM
L037
CN
on
syn
on
ymo
us
P30
6Su
vach
rIV
244
399
YD
L114
WN
on
syn
on
ymo
us
G11
9Cu
vach
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h3
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
r16
730
03
64
31(P
)21
39
66
25(G
)N
A(S
)ch
rIX
308
303
3084
YIL
166C
CN
V(a
mp
lifica
tio
n)
cer
chrX
III
02
4562
10ge
nes
incl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h4
chrV
II5
5588
5R
PL2
6BIn
tro
nce
r13
127
02
63
62(P
)0
686
410
(G)
204
66
860
(S)
chrX
246
208
PH
S1N
on
syn
on
ymo
us
K20
6Nce
rch
rXII
I32
4121
EIS1
No
nsy
no
nym
ou
sE3
49
uva
chrI
II0
826
8749
gen
esLO
HC
NV
cer
lost
chrX
III0
221
753
112
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pP
h5
chrX
III
2317
31P
PZ
1N
on
syn
on
ymo
us
A63
Su
va12
230
24
68
32(P
)
820
60
34(G
)18
20
62
91(S
)ch
rXII
I0
234
112
120
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIX
370
117
439
888
45ge
nes
LOH
CN
Vce
rlo
stP
h6
chrV
II9
7281
3P
FK1
No
nsy
no
nym
ou
sG
308S
cer
111
255
26
332
(P)
522
62
81(G
)N
A(S
)ch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rSh
1ch
rII5
1136
26
4497
469
6397
81
3184
74ge
nes
63
gen
esin
clu
din
gSU
L1LO
HC
NV
cer
lost
cer
amp
126
338
66
460
(S)
381
61
17(G
)
154
86
855
(P)
chIV
680
386
866
667
8666
67
9837
7410
4ge
nes
63
gen
esLO
HC
NV
uva
amp
uva
lost
chrX
VI
8470
00
9480
6649
gen
esLO
HC
NV
cer
lost
Sh2
chrV
II9
3638
4M
RP
L9N
on
syn
on
ymo
us
D16
7Gce
r26
819
64
64
30(S
)
619
61
21(G
)
143
66
37(P
)ch
rXV
I57
2308
ICL2
No
nsy
no
nym
ou
sM
247I
uva
chrV
III
1166
61ER
G11
No
nsy
no
nym
ou
sS2
86C
uva
chrI
I787
389
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh3
chrV
I16
2998
GC
N20
No
nsy
no
nym
ou
sD
171Y
cer
132
218
46
153
(S)
607
61
11(G
)5
556
481
(P)
chrX
IV4
9589
0FK
H2
Syn
on
ymo
us
S418
Su
vach
rII7
8658
48
131
8411
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
4
Tab
le1
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Sh4
chrX
IV6
6667
5A
RE2
No
nsy
no
nym
ou
sI4
46T
cer
285
271
96
433
(S)
617
60
51(G
)
205
56
330
(P)
chrX
V8
0083
2A
PC
55rsquo
-up
stre
amce
rch
rIV
259
17T
RM
3Sy
no
nym
ou
sS2
01S
cer
chrV
342
563
Inte
rgen
icu
vach
rX7
6976
8SP
O77
No
nsy
no
nym
ou
sD
418G
uva
chrX
990
873
LEU
35rsquo
-up
stre
amu
vach
rXII
192
491
Inte
rgen
icu
vach
rXIV
251
38EG
T2
Syn
on
ymo
us
T16
8Tu
vach
rII
7703
118
1318
422
gen
esi
ncl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
chrV
III
321
gen
esC
NV
(am
plifi
cati
on
)u
vaSh
5ch
rIV
310
881
RX
T3
No
nsy
no
nym
ou
sP
87T
uva
263
465
26
494
(S)
679
61
35(G
)3
726
723
(P)
chrV
III
1691
1In
terg
enic
uva
chrI
I78
6040
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh6
chrV
269
392
Inte
rgen
icce
r27
347
52
63
69(S
)2
606
125
(G)
447
66
04(P
)ch
rXIV
746
688
Inte
rgen
icce
rch
rIV
413
046
Inte
rgen
icu
vach
rII7
7894
28
131
8414
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
rSh
7ch
rII
2388
75In
terg
enic
cer
129
314
46
049
(S)
187
61
99(G
)8
746
922
(P)
chrX
VI
4906
31SV
L3N
on
syn
on
ymo
us
A24
5Vce
rch
rXV
I86
106
YP
L245
WN
on
syn
on
ymo
us
A17
4Dce
rch
rII
2732
96In
terg
enic
uva
chrI
I737
875
813
184
42ge
nes
in
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
Vs)
an
dlo
sso
fhet
ero
zygo
sity
even
ts(L
OH
)ar
ere
cord
edfo
rea
chev
olv
edh
ybri
dcl
on
eC
lon
esar
eid
enti
fied
by
nu
trie
nt
(Gg
luco
selim
itat
ion
Pp
ho
sph
ate
limit
atio
na
nd
Ssu
lfate
limit
atio
n)
anldquoh
rdquod
eno
tes
hyb
rid
an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
Gen
esin
un
der
line
hav
eb
een
fou
nd
toh
ave
po
int
mu
tati
on
sin
pri
or
exp
erim
ents
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fseq
uen
cin
gre
ads
and
are
thu
sap
pro
xim
ate
Rel
ativ
efi
tnes
sis
rep
ort
edw
ith
stan
dar
der
ror(
SE)a
nd
the
con
dit
ion
the
clo
ne
was
evo
lved
inlis
ted
firs
tfo
llow
edb
yth
etw
oal
tern
ativ
eco
nd
itio
ns
seve
ralc
lon
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
5
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
and dune habitats respectively and show traits such as in-creased drought or salt tolerance relative to their parents(Heiser 1954 Rieseberg 1991 Schwarzbach et al 2001Rosenthal et al 2002)
Agriculture and industry use both intra- and interspecifichybrids as a tool to increase yield or robustness introduceresistance to pests and create novel phenotype or flavorprofiles For example plant breeders have crossed domesti-cated species to wild species to introduce resistance to avariety of pathogens in wheat potato and canola (Masonand Batley 2015) and almost all maize grown in the UnitedStates is grown from intraspecific hybrid seeds which hasincreased yield and provided improved resistance to bioticand abiotic factors (Crow 1998) Vintners and brewers havecreated interspecific hybrids to select for traits such as loweracetic acid concentration (Bellon et al 2015) and many inci-dental fungal hybrids have been discovered in brewing andindustry including Pichia sorbitophila (Louis et al 2012) andvarious hybrids across the Saccharomyces clade (Gonzalezet al 2006 2008 Muller and McCusker 2009 Hittinger2013 Bellon et al 2015) most notably the lager-brewing yeastSaccharomyces pastorianus (Tamai et al 1998 Dunn andSherlock 2008 Walther et al 2014 Baker et al 2015 Gibsonand Liti 2015 Peris et al 2016) It is presumed that the severeselection pressures exerted during industrial processes haveselected for interspecific hybrid genomes that may be moreable to cope with the extreme environments
At the genomic level hybridization induces chromosomelossaneuploidy chromosomal rearrangements gene losschanges in gene expression changes in epigenetic modificationstransposable element mobilization and large-scale loss of het-erozygosity (LOH) in which the allele of one species is lost andthe allele of the other species is retained and may even beduplicated via gene conversion or break-induced replication(Masly et al 2006 Landry et al 2007 Doyle et al 2008Michalak 2009 Ainouche and Jenczewski 2010 Albertin andMarullo 2012 Abbott et al 2013 Soltis 2013 Borneman et al2014 Soltis et al 2014) These extensive changes can result in achimeric stabilized hybrid although the period of time for ge-nome stabilization to occur can range dramatically (Soltis et al2014) It is unknown whether there are structural and functionalbiases in the ways in which genesalleles are lost or modifiedBoth drift and selection influence the resolution of the hybridgenome but their contributions are difficult to untangle
Researchers have long been exploring the genetics of hy-brid traits in the lab particularly in agricultural crops al-though this is often slowed by infertility and reducedviability in many interspecific hybrids (Perez-Prat and vanLookeren Campagne 2002 Hajjar and Hodgkin 2007Ouyang et al 2010) The Saccharomyces genus which in-cludes the budding yeast Saccharomyces cerevisiae lends itselfparticularly well to experimental study Many hybrids of thisgenus have been discovered in brewing industrial and natu-ral environments indeed the genus itself is speculated tohave been founded by the product of an ancient hybridiza-tion event (Hittinger 2013 Marcet-Houben and Gabaldon2015 Barbosa et al 2016 Leducq et al 2016) Viable interspe-cific hybrids can be created de novo in the lab (Marinoni et al
1999 Greig et al 2002) and their ability to grow mitoticallymeans that the catastrophic postzygotic barriers to speciationthat generally doom other obligate sexually reproducing hy-brids can be avoided This experimental system allows us toobserve evolution in real time in the laboratory environmentand the genetic and genomic tools available in this modelgenus facilitate characterization of the connection betweengenotype and phenotype including fitness
Previous work in our lab group has utilized experimentalevolution to investigate adaptive events in haploid andhomozygous diploid S cerevisiae (Gresham et al 2008Payen et al 2014 Sunshine et al 2015) To investigate genomeevolution post hybridization we utilize an interspecific hybridS cerevisiae Saccharomyces uvarum and its parentals ahomozygous diploid S uvarum and an intraspecific hybridS cerevisiae GRF167 S cerevisiae S288C This allows us tounderstand the impact of varying levels of heterozygosity onadaptation and genome evolution ranging from none(S uvarum and previous S cerevisiae experiments) to intra-specific heterozygosity (S cerevisiae GRF167 S cerevisiaeS288C) to the most extreme case of interspecific hybridsSaccharomyces uvarum is one of the most distantly relatedspecies of S cerevisiae in the Saccharomyces clade separatedby 20 My and 20 sequence divergence at coding sites (Kelliset al 2003 Cliften et al 2006) Despite this extensive diver-gence S cerevisiae and S uvarum are largely syntenic andcreate hybrids though less than 1 of spores are viable(Greig 2009) The two species differ in their stress tolerancesfor example S cerevisiae being more thermotolerant S uva-rum being cryotolerant (Almeida et al 2014) Previous evolu-tion experiments using lab-derived hybrids have revealednovel andor transgressive phenotypes for ammonium limi-tation ethanol tolerance and growth on xylose (Belloch et al2008 Wenger et al 2010 Piotrowski et al 2012 Dunn et al2013) Notably Dunn et al (2013) revealed several LOHevents and a repeatable nonreciprocal translocation that pro-duces a gene fusion at the high-affinity ammonium permeaseMEP2 after selection in ammonium limitation offering insightinto potential mutational events in the adaptation andorstabilization of S cerevisiae S uvarum hybrids
Here we evolved these hybrids and diploids in replicate inthree nutrient-limited conditions for hundreds of genera-tions Using whole genome sequencing we found wholechromosome aneuploidy genome rearrangements copynumber variants de novo point mutations and LOH Wesought to determine how initial heterozygosity affects adap-tation to novel conditions and explore whether neutral orselective forces are influencing the resolution of the hybridgenome over time In particular we investigated a reoccur-ring LOH event observed in both intra- and interspecifichybrids and found support for the hypothesis that LOH atthis locus is due to selection
Results
Experimental Evolution of Hybrid and Parental SpeciesAn interspecific hybrid was created by crossing S cerevisiaeand S uvarum (strains in supplementary table S1
Smukowski Heil et al doi101093molbevmsx098 MBE
2
Supplementary Material online) and evolved in continu-ous culture in the chemostat (Monod 1949 Novick andSzilard 1950a 1950b) In parallel homozygous diploidS uvarum and heterozygous diploid S cerevisiae(GRF167 S288C) were also evolved Each strain wasgrown in two or more replicate independent culturesunder three different nutrient limitationsmdashglucose phos-phate and sulfatemdashfor 85ndash557 generations (median 158)at 30 C except for S uvarum which was unable toachieve steady state in all conditions at 30 C and sowas evolved at 25 C The population sizes were largelysimilar across strains species and conditions
Evolved clones were isolated from each population andsubsequently competed individually against the appropriategreen fluorescent protein (GFP)-tagged ancestor to gaugerelative fitness As expected evolved hybrid and parentalclones generally exhibit higher fitness than their unevolvedancestor with typical relative fitness gains between 20 and30 (tables 1 and 2) To explore whether these fitness gainsare general or condition specific we additionally competedeach hybrid clone in the two nutrient-limited conditions inwhich the clone was not evolved Results are variable withsome clones having negative or neutral fitness in the alternateconditions suggesting condition-specific adaptation andsome clones experiencing fitness gains in multiple conditionssuggesting more general growth benefits (table 1) Only oneclone exhibited fitness gains in all three nutrient environ-ments and no clones have a greater fitness gain in an alter-nate condition than the condition it was evolved in signifyingthat clones are largely specifically adapted to the particularcondition in which they were evolved
Mutations Are Recovered in Both Novel andPreviously Observed Gene Targets in InterspecificHybridsTo identify mutations in the evolved hybrids we generatedwhole genome sequencing data for 16 clones from the endpoints of the evolution experiments (table 1) We thus cap-tured data from a range of nutrient limitations (phosphate 6glucose 3 sulfate 7) and generations (100ndash285 median 154)Each clone had an average of 24 point mutations a numberof which have been previously identified in prior S cerevisiaeevolution experiments For example a nonsynonymous mu-tation in the S cerevisiae allele of the glucose-sensing geneSNF3 has been identified in glucose-limited experiments inS cerevisiae (Kvitek and Sherlock 2013 Selmecki et al 2015)To our knowledge 2027 coding point mutations are uniqueto these experiments (Payen et al 2016)
In evolved parentals we again sequenced one clone fromthe end point of each population In total we sequenced 16clones 6 from each of the 3 nutrients (2 S uvarum diploidsand 4 S cerevisiae diploids) except in glucose limitation inwhich only 2 S cerevisiae populations were sampled Thegenerations ranged from 234 to 557 (median 477) in S uva-rum with an average of 283 mutations per clone and from127 to 190 (median 1665) in S cerevisiae with an average of09 point mutations per clone (table 2) This discrepancy in
point mutations between S cerevisiae and S uvarum may beexplained by differences in generation number
With the limited number of samples we have from hybridand parental clones it is difficult to draw any conclusionsregarding unique point mutations in hybrids Furthermorewe have not tested the fitness of these mutations to provethey are adaptive However one class of mutations that maybe of particular interest in hybrids are genomic mutationsthat may interact with the mitochondria as previous workhas shown that nuclearndashmitochondria interactions can un-derlie hybrid incompatibility (Lee et al 2008 Chou and Leu2010 Meiklejohn et al 2013) Other studies have found thatonly the S cerevisiae mitochondria are retained in S cerevisiae S uvarum hybrids (Antunovics et al 2005) and we reca-pitulate these findings potentially setting the stage for con-flicting interactions between the S uvarum nuclear genomeand the foreign mitochondria We observe several mitochon-dria-related mutations in hybrids in both S cerevisiae andS uvarum alleles For example one point mutation a nonsyn-onymous mutation in the S cerevisiae allele of the mitochon-drial ribosomal protein gene MHR1 was seen in two separateclones independently evolved in phosphate limitation Thisgene may be of particular interest as it was discovered in aprevious screen as being haploproficient (increased fitness of19) in hybrids in which the S cerevisiae allele is missing andthe S uvarum allele is retained (Lancaster S Dunham MJunpublished data) suggesting that this mutation may alteror disable the S cerevisiae protein in some way Another ex-ample involves the gene IRC3 a helicase responsible for themaintenance of the mitochondrial genome which has a non-synonymous mutation in the S uvarum allele in clone Gh3and is deleted in clone Gh2 potentially suggesting that theS uvarum allele is deleterious in the hybrid backgroundAlthough our sample size is small 427 point mutations inhybrids are related to mitochondria function compared with026 in parentals and may represent interesting targets forfurther exploration
Copy Number Variants Frequently Involve theAmplification of Nutrient TransportersYeasts in both natural and artificial environments are knownto frequently experience changes in copy number rangingfrom single genes to whole chromosomes (Dunham et al2002 Gresham et al 2008 Dunn et al 2012 Kvitek andSherlock 2013 Payen et al 2014 Selmecki et al 2015Sunshine et al 2015 Zhu et al 2016) This holds true in ourevolution experiments We observe copy number changesacross all genetic backgrounds (fig 1 supplementary figsS1ndashS3 Supplementary Material online) Clones were com-pared with array comparative genomic hybridization of pop-ulations to confirm that clones are representative ofpopulations (see Materials and Methods) The evolved hybridclones displayed an average of 15 copy number variants(CNVs) per clone (fig 1 table 1 Supplementary fig S3Supplementary Material online) as defined by the numberof segmental or whole chromosome amplificationsdeletions(though it is likely that some of these CNVs were created inthe same mutational event) The evolved S cerevisiae clones
Adaptation in Hybrids doi101093molbevmsx098 MBE
3
Tab
le1
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edH
ybri
dC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Gh
1ch
rXII
I85
2028
Inte
rgen
icce
r12
526
80
60
98(G
)0
356
160
(S)
118
(P)
chrI
I91
1866
917
272
HX
T6
7C
NV
(am
plifi
cati
on
)u
vaG
h2
chrI
V1
1191
9SN
F3N
on
syn
on
ymo
us
D11
4Yce
r10
028
17
62
18(G
)10
48
60
78(S
)11
23
(P)
chrI
II5
1593
GLK
1Sy
no
nym
ou
sT
252T
cer
chrI
V8
8480
19
1211
913
gen
esin
clu
din
gIR
C3
LOH
CN
Vu
valo
stch
rII
9121
439
1747
0H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
chrI
V83
6ge
nes
CN
V(a
mp
lifica
tio
n)
cer
Gh
3ch
rII
8894
21IR
C3
no
nsy
no
nym
ou
sM
333I
uva
124
186
56
047
(G)
176
86
367
(S)
104
6(P
)ch
rII
9124
169
1777
8H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
Ph
1ch
rV2
6939
2In
terg
enic
cer
103
291
86
137
(P)
168
60
78(G
)0
086
043
(S)
chrX
IV7
4668
8In
terg
enic
cer
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
rch
rIX
241
gen
esLO
HC
NV
uva
lost
cer
amp
Ph
2ch
rV4
3277
8G
LC7
Intr
on
cer
124
253
46
024
(P)
151
26
466
(G)
245
61
16(S
)ch
rVII
952
4P
DR
11N
on
syn
on
ymo
us
L383
u
vach
rXV
I23
2879
MR
PL4
0N
on
syn
on
ymo
us
V14
9Eu
vach
rXII
I19
4496
YM
L037
CN
on
syn
on
ymo
us
P30
6Su
vach
rIV
244
399
YD
L114
WN
on
syn
on
ymo
us
G11
9Cu
vach
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h3
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
r16
730
03
64
31(P
)21
39
66
25(G
)N
A(S
)ch
rIX
308
303
3084
YIL
166C
CN
V(a
mp
lifica
tio
n)
cer
chrX
III
02
4562
10ge
nes
incl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h4
chrV
II5
5588
5R
PL2
6BIn
tro
nce
r13
127
02
63
62(P
)0
686
410
(G)
204
66
860
(S)
chrX
246
208
PH
S1N
on
syn
on
ymo
us
K20
6Nce
rch
rXII
I32
4121
EIS1
No
nsy
no
nym
ou
sE3
49
uva
chrI
II0
826
8749
gen
esLO
HC
NV
cer
lost
chrX
III0
221
753
112
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pP
h5
chrX
III
2317
31P
PZ
1N
on
syn
on
ymo
us
A63
Su
va12
230
24
68
32(P
)
820
60
34(G
)18
20
62
91(S
)ch
rXII
I0
234
112
120
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIX
370
117
439
888
45ge
nes
LOH
CN
Vce
rlo
stP
h6
chrV
II9
7281
3P
FK1
No
nsy
no
nym
ou
sG
308S
cer
111
255
26
332
(P)
522
62
81(G
)N
A(S
)ch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rSh
1ch
rII5
1136
26
4497
469
6397
81
3184
74ge
nes
63
gen
esin
clu
din
gSU
L1LO
HC
NV
cer
lost
cer
amp
126
338
66
460
(S)
381
61
17(G
)
154
86
855
(P)
chIV
680
386
866
667
8666
67
9837
7410
4ge
nes
63
gen
esLO
HC
NV
uva
amp
uva
lost
chrX
VI
8470
00
9480
6649
gen
esLO
HC
NV
cer
lost
Sh2
chrV
II9
3638
4M
RP
L9N
on
syn
on
ymo
us
D16
7Gce
r26
819
64
64
30(S
)
619
61
21(G
)
143
66
37(P
)ch
rXV
I57
2308
ICL2
No
nsy
no
nym
ou
sM
247I
uva
chrV
III
1166
61ER
G11
No
nsy
no
nym
ou
sS2
86C
uva
chrI
I787
389
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh3
chrV
I16
2998
GC
N20
No
nsy
no
nym
ou
sD
171Y
cer
132
218
46
153
(S)
607
61
11(G
)5
556
481
(P)
chrX
IV4
9589
0FK
H2
Syn
on
ymo
us
S418
Su
vach
rII7
8658
48
131
8411
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
4
Tab
le1
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Sh4
chrX
IV6
6667
5A
RE2
No
nsy
no
nym
ou
sI4
46T
cer
285
271
96
433
(S)
617
60
51(G
)
205
56
330
(P)
chrX
V8
0083
2A
PC
55rsquo
-up
stre
amce
rch
rIV
259
17T
RM
3Sy
no
nym
ou
sS2
01S
cer
chrV
342
563
Inte
rgen
icu
vach
rX7
6976
8SP
O77
No
nsy
no
nym
ou
sD
418G
uva
chrX
990
873
LEU
35rsquo
-up
stre
amu
vach
rXII
192
491
Inte
rgen
icu
vach
rXIV
251
38EG
T2
Syn
on
ymo
us
T16
8Tu
vach
rII
7703
118
1318
422
gen
esi
ncl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
chrV
III
321
gen
esC
NV
(am
plifi
cati
on
)u
vaSh
5ch
rIV
310
881
RX
T3
No
nsy
no
nym
ou
sP
87T
uva
263
465
26
494
(S)
679
61
35(G
)3
726
723
(P)
chrV
III
1691
1In
terg
enic
uva
chrI
I78
6040
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh6
chrV
269
392
Inte
rgen
icce
r27
347
52
63
69(S
)2
606
125
(G)
447
66
04(P
)ch
rXIV
746
688
Inte
rgen
icce
rch
rIV
413
046
Inte
rgen
icu
vach
rII7
7894
28
131
8414
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
rSh
7ch
rII
2388
75In
terg
enic
cer
129
314
46
049
(S)
187
61
99(G
)8
746
922
(P)
chrX
VI
4906
31SV
L3N
on
syn
on
ymo
us
A24
5Vce
rch
rXV
I86
106
YP
L245
WN
on
syn
on
ymo
us
A17
4Dce
rch
rII
2732
96In
terg
enic
uva
chrI
I737
875
813
184
42ge
nes
in
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
Vs)
an
dlo
sso
fhet
ero
zygo
sity
even
ts(L
OH
)ar
ere
cord
edfo
rea
chev
olv
edh
ybri
dcl
on
eC
lon
esar
eid
enti
fied
by
nu
trie
nt
(Gg
luco
selim
itat
ion
Pp
ho
sph
ate
limit
atio
na
nd
Ssu
lfate
limit
atio
n)
anldquoh
rdquod
eno
tes
hyb
rid
an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
Gen
esin
un
der
line
hav
eb
een
fou
nd
toh
ave
po
int
mu
tati
on
sin
pri
or
exp
erim
ents
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fseq
uen
cin
gre
ads
and
are
thu
sap
pro
xim
ate
Rel
ativ
efi
tnes
sis
rep
ort
edw
ith
stan
dar
der
ror(
SE)a
nd
the
con
dit
ion
the
clo
ne
was
evo
lved
inlis
ted
firs
tfo
llow
edb
yth
etw
oal
tern
ativ
eco
nd
itio
ns
seve
ralc
lon
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
5
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Supplementary Material online) and evolved in continu-ous culture in the chemostat (Monod 1949 Novick andSzilard 1950a 1950b) In parallel homozygous diploidS uvarum and heterozygous diploid S cerevisiae(GRF167 S288C) were also evolved Each strain wasgrown in two or more replicate independent culturesunder three different nutrient limitationsmdashglucose phos-phate and sulfatemdashfor 85ndash557 generations (median 158)at 30 C except for S uvarum which was unable toachieve steady state in all conditions at 30 C and sowas evolved at 25 C The population sizes were largelysimilar across strains species and conditions
Evolved clones were isolated from each population andsubsequently competed individually against the appropriategreen fluorescent protein (GFP)-tagged ancestor to gaugerelative fitness As expected evolved hybrid and parentalclones generally exhibit higher fitness than their unevolvedancestor with typical relative fitness gains between 20 and30 (tables 1 and 2) To explore whether these fitness gainsare general or condition specific we additionally competedeach hybrid clone in the two nutrient-limited conditions inwhich the clone was not evolved Results are variable withsome clones having negative or neutral fitness in the alternateconditions suggesting condition-specific adaptation andsome clones experiencing fitness gains in multiple conditionssuggesting more general growth benefits (table 1) Only oneclone exhibited fitness gains in all three nutrient environ-ments and no clones have a greater fitness gain in an alter-nate condition than the condition it was evolved in signifyingthat clones are largely specifically adapted to the particularcondition in which they were evolved
Mutations Are Recovered in Both Novel andPreviously Observed Gene Targets in InterspecificHybridsTo identify mutations in the evolved hybrids we generatedwhole genome sequencing data for 16 clones from the endpoints of the evolution experiments (table 1) We thus cap-tured data from a range of nutrient limitations (phosphate 6glucose 3 sulfate 7) and generations (100ndash285 median 154)Each clone had an average of 24 point mutations a numberof which have been previously identified in prior S cerevisiaeevolution experiments For example a nonsynonymous mu-tation in the S cerevisiae allele of the glucose-sensing geneSNF3 has been identified in glucose-limited experiments inS cerevisiae (Kvitek and Sherlock 2013 Selmecki et al 2015)To our knowledge 2027 coding point mutations are uniqueto these experiments (Payen et al 2016)
In evolved parentals we again sequenced one clone fromthe end point of each population In total we sequenced 16clones 6 from each of the 3 nutrients (2 S uvarum diploidsand 4 S cerevisiae diploids) except in glucose limitation inwhich only 2 S cerevisiae populations were sampled Thegenerations ranged from 234 to 557 (median 477) in S uva-rum with an average of 283 mutations per clone and from127 to 190 (median 1665) in S cerevisiae with an average of09 point mutations per clone (table 2) This discrepancy in
point mutations between S cerevisiae and S uvarum may beexplained by differences in generation number
With the limited number of samples we have from hybridand parental clones it is difficult to draw any conclusionsregarding unique point mutations in hybrids Furthermorewe have not tested the fitness of these mutations to provethey are adaptive However one class of mutations that maybe of particular interest in hybrids are genomic mutationsthat may interact with the mitochondria as previous workhas shown that nuclearndashmitochondria interactions can un-derlie hybrid incompatibility (Lee et al 2008 Chou and Leu2010 Meiklejohn et al 2013) Other studies have found thatonly the S cerevisiae mitochondria are retained in S cerevisiae S uvarum hybrids (Antunovics et al 2005) and we reca-pitulate these findings potentially setting the stage for con-flicting interactions between the S uvarum nuclear genomeand the foreign mitochondria We observe several mitochon-dria-related mutations in hybrids in both S cerevisiae andS uvarum alleles For example one point mutation a nonsyn-onymous mutation in the S cerevisiae allele of the mitochon-drial ribosomal protein gene MHR1 was seen in two separateclones independently evolved in phosphate limitation Thisgene may be of particular interest as it was discovered in aprevious screen as being haploproficient (increased fitness of19) in hybrids in which the S cerevisiae allele is missing andthe S uvarum allele is retained (Lancaster S Dunham MJunpublished data) suggesting that this mutation may alteror disable the S cerevisiae protein in some way Another ex-ample involves the gene IRC3 a helicase responsible for themaintenance of the mitochondrial genome which has a non-synonymous mutation in the S uvarum allele in clone Gh3and is deleted in clone Gh2 potentially suggesting that theS uvarum allele is deleterious in the hybrid backgroundAlthough our sample size is small 427 point mutations inhybrids are related to mitochondria function compared with026 in parentals and may represent interesting targets forfurther exploration
Copy Number Variants Frequently Involve theAmplification of Nutrient TransportersYeasts in both natural and artificial environments are knownto frequently experience changes in copy number rangingfrom single genes to whole chromosomes (Dunham et al2002 Gresham et al 2008 Dunn et al 2012 Kvitek andSherlock 2013 Payen et al 2014 Selmecki et al 2015Sunshine et al 2015 Zhu et al 2016) This holds true in ourevolution experiments We observe copy number changesacross all genetic backgrounds (fig 1 supplementary figsS1ndashS3 Supplementary Material online) Clones were com-pared with array comparative genomic hybridization of pop-ulations to confirm that clones are representative ofpopulations (see Materials and Methods) The evolved hybridclones displayed an average of 15 copy number variants(CNVs) per clone (fig 1 table 1 Supplementary fig S3Supplementary Material online) as defined by the numberof segmental or whole chromosome amplificationsdeletions(though it is likely that some of these CNVs were created inthe same mutational event) The evolved S cerevisiae clones
Adaptation in Hybrids doi101093molbevmsx098 MBE
3
Tab
le1
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edH
ybri
dC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Gh
1ch
rXII
I85
2028
Inte
rgen
icce
r12
526
80
60
98(G
)0
356
160
(S)
118
(P)
chrI
I91
1866
917
272
HX
T6
7C
NV
(am
plifi
cati
on
)u
vaG
h2
chrI
V1
1191
9SN
F3N
on
syn
on
ymo
us
D11
4Yce
r10
028
17
62
18(G
)10
48
60
78(S
)11
23
(P)
chrI
II5
1593
GLK
1Sy
no
nym
ou
sT
252T
cer
chrI
V8
8480
19
1211
913
gen
esin
clu
din
gIR
C3
LOH
CN
Vu
valo
stch
rII
9121
439
1747
0H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
chrI
V83
6ge
nes
CN
V(a
mp
lifica
tio
n)
cer
Gh
3ch
rII
8894
21IR
C3
no
nsy
no
nym
ou
sM
333I
uva
124
186
56
047
(G)
176
86
367
(S)
104
6(P
)ch
rII
9124
169
1777
8H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
Ph
1ch
rV2
6939
2In
terg
enic
cer
103
291
86
137
(P)
168
60
78(G
)0
086
043
(S)
chrX
IV7
4668
8In
terg
enic
cer
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
rch
rIX
241
gen
esLO
HC
NV
uva
lost
cer
amp
Ph
2ch
rV4
3277
8G
LC7
Intr
on
cer
124
253
46
024
(P)
151
26
466
(G)
245
61
16(S
)ch
rVII
952
4P
DR
11N
on
syn
on
ymo
us
L383
u
vach
rXV
I23
2879
MR
PL4
0N
on
syn
on
ymo
us
V14
9Eu
vach
rXII
I19
4496
YM
L037
CN
on
syn
on
ymo
us
P30
6Su
vach
rIV
244
399
YD
L114
WN
on
syn
on
ymo
us
G11
9Cu
vach
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h3
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
r16
730
03
64
31(P
)21
39
66
25(G
)N
A(S
)ch
rIX
308
303
3084
YIL
166C
CN
V(a
mp
lifica
tio
n)
cer
chrX
III
02
4562
10ge
nes
incl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h4
chrV
II5
5588
5R
PL2
6BIn
tro
nce
r13
127
02
63
62(P
)0
686
410
(G)
204
66
860
(S)
chrX
246
208
PH
S1N
on
syn
on
ymo
us
K20
6Nce
rch
rXII
I32
4121
EIS1
No
nsy
no
nym
ou
sE3
49
uva
chrI
II0
826
8749
gen
esLO
HC
NV
cer
lost
chrX
III0
221
753
112
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pP
h5
chrX
III
2317
31P
PZ
1N
on
syn
on
ymo
us
A63
Su
va12
230
24
68
32(P
)
820
60
34(G
)18
20
62
91(S
)ch
rXII
I0
234
112
120
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIX
370
117
439
888
45ge
nes
LOH
CN
Vce
rlo
stP
h6
chrV
II9
7281
3P
FK1
No
nsy
no
nym
ou
sG
308S
cer
111
255
26
332
(P)
522
62
81(G
)N
A(S
)ch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rSh
1ch
rII5
1136
26
4497
469
6397
81
3184
74ge
nes
63
gen
esin
clu
din
gSU
L1LO
HC
NV
cer
lost
cer
amp
126
338
66
460
(S)
381
61
17(G
)
154
86
855
(P)
chIV
680
386
866
667
8666
67
9837
7410
4ge
nes
63
gen
esLO
HC
NV
uva
amp
uva
lost
chrX
VI
8470
00
9480
6649
gen
esLO
HC
NV
cer
lost
Sh2
chrV
II9
3638
4M
RP
L9N
on
syn
on
ymo
us
D16
7Gce
r26
819
64
64
30(S
)
619
61
21(G
)
143
66
37(P
)ch
rXV
I57
2308
ICL2
No
nsy
no
nym
ou
sM
247I
uva
chrV
III
1166
61ER
G11
No
nsy
no
nym
ou
sS2
86C
uva
chrI
I787
389
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh3
chrV
I16
2998
GC
N20
No
nsy
no
nym
ou
sD
171Y
cer
132
218
46
153
(S)
607
61
11(G
)5
556
481
(P)
chrX
IV4
9589
0FK
H2
Syn
on
ymo
us
S418
Su
vach
rII7
8658
48
131
8411
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
4
Tab
le1
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Sh4
chrX
IV6
6667
5A
RE2
No
nsy
no
nym
ou
sI4
46T
cer
285
271
96
433
(S)
617
60
51(G
)
205
56
330
(P)
chrX
V8
0083
2A
PC
55rsquo
-up
stre
amce
rch
rIV
259
17T
RM
3Sy
no
nym
ou
sS2
01S
cer
chrV
342
563
Inte
rgen
icu
vach
rX7
6976
8SP
O77
No
nsy
no
nym
ou
sD
418G
uva
chrX
990
873
LEU
35rsquo
-up
stre
amu
vach
rXII
192
491
Inte
rgen
icu
vach
rXIV
251
38EG
T2
Syn
on
ymo
us
T16
8Tu
vach
rII
7703
118
1318
422
gen
esi
ncl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
chrV
III
321
gen
esC
NV
(am
plifi
cati
on
)u
vaSh
5ch
rIV
310
881
RX
T3
No
nsy
no
nym
ou
sP
87T
uva
263
465
26
494
(S)
679
61
35(G
)3
726
723
(P)
chrV
III
1691
1In
terg
enic
uva
chrI
I78
6040
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh6
chrV
269
392
Inte
rgen
icce
r27
347
52
63
69(S
)2
606
125
(G)
447
66
04(P
)ch
rXIV
746
688
Inte
rgen
icce
rch
rIV
413
046
Inte
rgen
icu
vach
rII7
7894
28
131
8414
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
rSh
7ch
rII
2388
75In
terg
enic
cer
129
314
46
049
(S)
187
61
99(G
)8
746
922
(P)
chrX
VI
4906
31SV
L3N
on
syn
on
ymo
us
A24
5Vce
rch
rXV
I86
106
YP
L245
WN
on
syn
on
ymo
us
A17
4Dce
rch
rII
2732
96In
terg
enic
uva
chrI
I737
875
813
184
42ge
nes
in
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
Vs)
an
dlo
sso
fhet
ero
zygo
sity
even
ts(L
OH
)ar
ere
cord
edfo
rea
chev
olv
edh
ybri
dcl
on
eC
lon
esar
eid
enti
fied
by
nu
trie
nt
(Gg
luco
selim
itat
ion
Pp
ho
sph
ate
limit
atio
na
nd
Ssu
lfate
limit
atio
n)
anldquoh
rdquod
eno
tes
hyb
rid
an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
Gen
esin
un
der
line
hav
eb
een
fou
nd
toh
ave
po
int
mu
tati
on
sin
pri
or
exp
erim
ents
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fseq
uen
cin
gre
ads
and
are
thu
sap
pro
xim
ate
Rel
ativ
efi
tnes
sis
rep
ort
edw
ith
stan
dar
der
ror(
SE)a
nd
the
con
dit
ion
the
clo
ne
was
evo
lved
inlis
ted
firs
tfo
llow
edb
yth
etw
oal
tern
ativ
eco
nd
itio
ns
seve
ralc
lon
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
5
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Tab
le1
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edH
ybri
dC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Gh
1ch
rXII
I85
2028
Inte
rgen
icce
r12
526
80
60
98(G
)0
356
160
(S)
118
(P)
chrI
I91
1866
917
272
HX
T6
7C
NV
(am
plifi
cati
on
)u
vaG
h2
chrI
V1
1191
9SN
F3N
on
syn
on
ymo
us
D11
4Yce
r10
028
17
62
18(G
)10
48
60
78(S
)11
23
(P)
chrI
II5
1593
GLK
1Sy
no
nym
ou
sT
252T
cer
chrI
V8
8480
19
1211
913
gen
esin
clu
din
gIR
C3
LOH
CN
Vu
valo
stch
rII
9121
439
1747
0H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
chrI
V83
6ge
nes
CN
V(a
mp
lifica
tio
n)
cer
Gh
3ch
rII
8894
21IR
C3
no
nsy
no
nym
ou
sM
333I
uva
124
186
56
047
(G)
176
86
367
(S)
104
6(P
)ch
rII
9124
169
1777
8H
XT
67
CN
V(a
mp
lifica
tio
n)
uva
Ph
1ch
rV2
6939
2In
terg
enic
cer
103
291
86
137
(P)
168
60
78(G
)0
086
043
(S)
chrX
IV7
4668
8In
terg
enic
cer
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
rch
rIX
241
gen
esLO
HC
NV
uva
lost
cer
amp
Ph
2ch
rV4
3277
8G
LC7
Intr
on
cer
124
253
46
024
(P)
151
26
466
(G)
245
61
16(S
)ch
rVII
952
4P
DR
11N
on
syn
on
ymo
us
L383
u
vach
rXV
I23
2879
MR
PL4
0N
on
syn
on
ymo
us
V14
9Eu
vach
rXII
I19
4496
YM
L037
CN
on
syn
on
ymo
us
P30
6Su
vach
rIV
244
399
YD
L114
WN
on
syn
on
ymo
us
G11
9Cu
vach
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h3
chrI
V1
0558
64M
HR
1N
on
syn
on
ymo
us
T21
8Rce
r16
730
03
64
31(P
)21
39
66
25(G
)N
A(S
)ch
rIX
308
303
3084
YIL
166C
CN
V(a
mp
lifica
tio
n)
cer
chrX
III
02
4562
10ge
nes
incl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rP
h4
chrV
II5
5588
5R
PL2
6BIn
tro
nce
r13
127
02
63
62(P
)0
686
410
(G)
204
66
860
(S)
chrX
246
208
PH
S1N
on
syn
on
ymo
us
K20
6Nce
rch
rXII
I32
4121
EIS1
No
nsy
no
nym
ou
sE3
49
uva
chrI
II0
826
8749
gen
esLO
HC
NV
cer
lost
chrX
III0
221
753
112
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pP
h5
chrX
III
2317
31P
PZ
1N
on
syn
on
ymo
us
A63
Su
va12
230
24
68
32(P
)
820
60
34(G
)18
20
62
91(S
)ch
rXII
I0
234
112
120
gen
esi
ncl
ud
ing
PH
O84
LOH
CN
Vu
valo
stc
eram
pch
rIX
370
117
439
888
45ge
nes
LOH
CN
Vce
rlo
stP
h6
chrV
II9
7281
3P
FK1
No
nsy
no
nym
ou
sG
308S
cer
111
255
26
332
(P)
522
62
81(G
)N
A(S
)ch
rIV
836
gen
esC
NV
(am
plifi
cati
on
)ce
rSh
1ch
rII5
1136
26
4497
469
6397
81
3184
74ge
nes
63
gen
esin
clu
din
gSU
L1LO
HC
NV
cer
lost
cer
amp
126
338
66
460
(S)
381
61
17(G
)
154
86
855
(P)
chIV
680
386
866
667
8666
67
9837
7410
4ge
nes
63
gen
esLO
HC
NV
uva
amp
uva
lost
chrX
VI
8470
00
9480
6649
gen
esLO
HC
NV
cer
lost
Sh2
chrV
II9
3638
4M
RP
L9N
on
syn
on
ymo
us
D16
7Gce
r26
819
64
64
30(S
)
619
61
21(G
)
143
66
37(P
)ch
rXV
I57
2308
ICL2
No
nsy
no
nym
ou
sM
247I
uva
chrV
III
1166
61ER
G11
No
nsy
no
nym
ou
sS2
86C
uva
chrI
I787
389
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh3
chrV
I16
2998
GC
N20
No
nsy
no
nym
ou
sD
171Y
cer
132
218
46
153
(S)
607
61
11(G
)5
556
481
(P)
chrX
IV4
9589
0FK
H2
Syn
on
ymo
us
S418
Su
vach
rII7
8658
48
131
8411
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
4
Tab
le1
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Sh4
chrX
IV6
6667
5A
RE2
No
nsy
no
nym
ou
sI4
46T
cer
285
271
96
433
(S)
617
60
51(G
)
205
56
330
(P)
chrX
V8
0083
2A
PC
55rsquo
-up
stre
amce
rch
rIV
259
17T
RM
3Sy
no
nym
ou
sS2
01S
cer
chrV
342
563
Inte
rgen
icu
vach
rX7
6976
8SP
O77
No
nsy
no
nym
ou
sD
418G
uva
chrX
990
873
LEU
35rsquo
-up
stre
amu
vach
rXII
192
491
Inte
rgen
icu
vach
rXIV
251
38EG
T2
Syn
on
ymo
us
T16
8Tu
vach
rII
7703
118
1318
422
gen
esi
ncl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
chrV
III
321
gen
esC
NV
(am
plifi
cati
on
)u
vaSh
5ch
rIV
310
881
RX
T3
No
nsy
no
nym
ou
sP
87T
uva
263
465
26
494
(S)
679
61
35(G
)3
726
723
(P)
chrV
III
1691
1In
terg
enic
uva
chrI
I78
6040
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh6
chrV
269
392
Inte
rgen
icce
r27
347
52
63
69(S
)2
606
125
(G)
447
66
04(P
)ch
rXIV
746
688
Inte
rgen
icce
rch
rIV
413
046
Inte
rgen
icu
vach
rII7
7894
28
131
8414
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
rSh
7ch
rII
2388
75In
terg
enic
cer
129
314
46
049
(S)
187
61
99(G
)8
746
922
(P)
chrX
VI
4906
31SV
L3N
on
syn
on
ymo
us
A24
5Vce
rch
rXV
I86
106
YP
L245
WN
on
syn
on
ymo
us
A17
4Dce
rch
rII
2732
96In
terg
enic
uva
chrI
I737
875
813
184
42ge
nes
in
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
Vs)
an
dlo
sso
fhet
ero
zygo
sity
even
ts(L
OH
)ar
ere
cord
edfo
rea
chev
olv
edh
ybri
dcl
on
eC
lon
esar
eid
enti
fied
by
nu
trie
nt
(Gg
luco
selim
itat
ion
Pp
ho
sph
ate
limit
atio
na
nd
Ssu
lfate
limit
atio
n)
anldquoh
rdquod
eno
tes
hyb
rid
an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
Gen
esin
un
der
line
hav
eb
een
fou
nd
toh
ave
po
int
mu
tati
on
sin
pri
or
exp
erim
ents
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fseq
uen
cin
gre
ads
and
are
thu
sap
pro
xim
ate
Rel
ativ
efi
tnes
sis
rep
ort
edw
ith
stan
dar
der
ror(
SE)a
nd
the
con
dit
ion
the
clo
ne
was
evo
lved
inlis
ted
firs
tfo
llow
edb
yth
etw
oal
tern
ativ
eco
nd
itio
ns
seve
ralc
lon
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
5
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Tab
le1
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE(c
on
dit
ion
)
Sh4
chrX
IV6
6667
5A
RE2
No
nsy
no
nym
ou
sI4
46T
cer
285
271
96
433
(S)
617
60
51(G
)
205
56
330
(P)
chrX
V8
0083
2A
PC
55rsquo
-up
stre
amce
rch
rIV
259
17T
RM
3Sy
no
nym
ou
sS2
01S
cer
chrV
342
563
Inte
rgen
icu
vach
rX7
6976
8SP
O77
No
nsy
no
nym
ou
sD
418G
uva
chrX
990
873
LEU
35rsquo
-up
stre
amu
vach
rXII
192
491
Inte
rgen
icu
vach
rXIV
251
38EG
T2
Syn
on
ymo
us
T16
8Tu
vach
rII
7703
118
1318
422
gen
esi
ncl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
chrV
III
321
gen
esC
NV
(am
plifi
cati
on
)u
vaSh
5ch
rIV
310
881
RX
T3
No
nsy
no
nym
ou
sP
87T
uva
263
465
26
494
(S)
679
61
35(G
)3
726
723
(P)
chrV
III
1691
1In
terg
enic
uva
chrI
I78
6040
813
184
11ge
nes
incl
ud
ing
SUL1
CN
V(a
mp
lifica
tio
n)
cer
Sh6
chrV
269
392
Inte
rgen
icce
r27
347
52
63
69(S
)2
606
125
(G)
447
66
04(P
)ch
rXIV
746
688
Inte
rgen
icce
rch
rIV
413
046
Inte
rgen
icu
vach
rII7
7894
28
131
8414
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
rSh
7ch
rII
2388
75In
terg
enic
cer
129
314
46
049
(S)
187
61
99(G
)8
746
922
(P)
chrX
VI
4906
31SV
L3N
on
syn
on
ymo
us
A24
5Vce
rch
rXV
I86
106
YP
L245
WN
on
syn
on
ymo
us
A17
4Dce
rch
rII
2732
96In
terg
enic
uva
chrI
I737
875
813
184
42ge
nes
in
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)ce
r
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
Vs)
an
dlo
sso
fhet
ero
zygo
sity
even
ts(L
OH
)ar
ere
cord
edfo
rea
chev
olv
edh
ybri
dcl
on
eC
lon
esar
eid
enti
fied
by
nu
trie
nt
(Gg
luco
selim
itat
ion
Pp
ho
sph
ate
limit
atio
na
nd
Ssu
lfate
limit
atio
n)
anldquoh
rdquod
eno
tes
hyb
rid
an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
Gen
esin
un
der
line
hav
eb
een
fou
nd
toh
ave
po
int
mu
tati
on
sin
pri
or
exp
erim
ents
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fseq
uen
cin
gre
ads
and
are
thu
sap
pro
xim
ate
Rel
ativ
efi
tnes
sis
rep
ort
edw
ith
stan
dar
der
ror(
SE)a
nd
the
con
dit
ion
the
clo
ne
was
evo
lved
inlis
ted
firs
tfo
llow
edb
yth
etw
oal
tern
ativ
eco
nd
itio
ns
seve
ralc
lon
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
5
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Tab
le2
Mu
tati
on
san
dFi
tnes
so
fEv
olv
edPa
ren
talC
lon
es
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Gc1
chrX
IV0
561
000
6322
507
8433
329
8ge
nes
79
gen
esC
NV
(am
plifi
cati
on
of
chr
14L
favo
rin
gG
RF1
67
del
etio
no
fch
r14R
)ce
r16
316
42
63
42
chrV
160
000
576
874
220
gen
esLO
H(f
avo
rsG
RF1
67)
Gc2
chrV
431
750
576
874
71ge
nes
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
167
103
66
058
chrX
V7
1000
01
0912
9119
6ge
nes
LOH
CN
V(m
on
oso
my
favo
rin
gS2
88C
)G
u1
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
468
180
36
212
chrI
I911
925
917
281
HX
T6
7C
NV
(am
plifi
cati
on
)ch
rXV
385
930
NEL
1N
on
syn
on
ymo
us
N12
9Ich
rII9
1190
9In
terg
enic
par
to
fth
eH
XT
67
amp
lifica
tio
nG
u2
chrX
V59
7ge
nes
CN
V(w
ho
lech
rom
oso
me
amp
lifica
tio
n)
uva
486
131
2ch
rII9
1192
59
1728
1H
XT
67
CN
V(a
mp
lifica
tio
n)
chrI
V1
0029
3R
GT
2N
on
syn
on
ymo
us
G10
7Vch
rV4
2093
FRD
1N
on
syn
on
ymo
us
G12
8Ach
rII9
1719
1H
XT
7Sy
no
nym
ou
sH
53H
chrX
I155
787
Inte
rgen
icP
c1ch
rXII
I03
9000
(LO
H)
01
9662
8(C
NV
3co
pie
s)
1966
283
7300
0(C
NV
2co
pie
s)LO
H1
5ge
nes
incl
ud
ing
PH
O84
C
NV
201
gen
esLO
HC
NV
(am
plifi
cati
on
fav
ori
ng
GR
F167
)ce
r15
221
22
60
81
Pc2
chrX
III0
411
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
16
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
149
181
36
103
chrV
III5
2034
9In
terg
enic
Pc3
chrX
III0
390
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
15
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
127
194
9
Pc4
chrX
III0
855
00(L
OH
)0
196
628
(CN
V3
cop
ies)
19
6628
373
000
(CN
V2
cop
ies)
LOH
40
gen
esin
clu
din
gP
HO
84
CN
V2
01ge
nes
LOH
CN
V(a
mp
lifica
tio
nf
avo
rin
gG
RF1
67)
cer
132
209
66
141
chrX
II2
6400
01
0781
7743
7ge
nes
LOH
(fav
ori
ng
S288
C)
chrX
V1
0231
97P
IP2
No
nsy
no
nym
ou
sE6
QP
u1
uva
240
1
686
110
Pu
2ch
rIX
144
80Y
PS6
5rsquo-u
pst
ream
uva
234
213
06
073
chrI
X2
2531
4SE
C6
No
nsy
no
nym
ou
sI1
84L
chrX
III
1295
67T
CB
3N
on
syn
on
ymo
us
E625
GSc
1ch
rXIV
01
0200
0(C
NV
3co
pie
s)6
3200
07
8433
3(C
NV
1co
py)
LO
H1
0000
07
8433
348
gen
es7
9ge
nes
367
gen
esLO
HC
NV
(am
plifi
cati
on
of
chr
14L
del
etio
no
fch
r14R
LO
Hfa
vori
ng
S288
C)
cer
182
380
66
175
chrV
III2
0796
7SM
F2N
on
syn
on
ymo
us
W10
5Sch
rXII
I190
000
196
500
RR
N11
CA
T2
VP
S71
LOH
CN
V(d
elet
ion
fav
ori
ng
GR
F167
)ch
rII7
8718
07
9735
0V
BA
1SU
L1P
CA
1C
NV
(am
plifi
cati
on
)Sc
2ch
rXII
578
gen
esC
NV
(wh
ole
chro
mo
som
eam
plifi
cati
on
fa
vori
ng
GR
F167
)ce
r17
640
21
61
33
chrX
II6
9200
01
0781
7719
3ge
nes
LOH
(fav
ori
ng
GR
F167
)ch
rII7
7322
08
1318
418
gen
esin
clu
din
gSU
L1C
NV
(am
plifi
cati
on
)Sc
3ch
rVI9
4104
FRS2
No
nsy
no
nym
ou
sV
303I
cer
201
413
46
677
chrV
III3
0890
3T
RA
1N
on
syn
on
ymo
us
V20
48A
chrX
IV2
3226
6P
OP
1N
on
syn
on
ymo
us
S477
ch
rXV
291
219
TLG
2N
on
syn
on
ymo
us
D28
6Ych
rXV
309
86H
PF1
Syn
on
ymo
us
T20
7Tch
rII7
8180
07
9223
05
gen
esin
clu
din
gB
SD2
and
SUL1
CN
V(a
mp
lifica
tio
n)
(co
nti
nu
ed)
Smukowski Heil et al doi101093molbevmsx098 MBE
6
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Tab
le2
Co
nti
nu
ed
Clo
ne
Loca
tio
nG
ene(
s)M
uta
tio
nSp
ecie
sG
ener
atio
ns
Rel
ativ
eFi
tnes
s6
SE
Sc4
chrI
I275
000
813
184
289
gen
esLO
H(f
avo
rin
gG
RF1
67)
cer
190
312
56
613
chrI
I788
608
795
833
SUL1
PC
A1
CN
V(a
mp
lifica
tio
n)
chrX
I517
650
666
816
68ge
nes
CN
V(a
mp
lifica
tio
n)
chrX
III1
9000
01
9650
0R
RN
11C
AT
2V
PS7
1LO
HC
NV
(del
etio
nf
avo
rin
gG
RF1
67)
chrX
IV6
3200
07
8433
379
gen
esLO
HC
NV
(del
etio
n)
chrX
V3
3670
03
4200
034
2000
109
1291
2ge
nes
384
gen
esLO
H(f
avo
rin
gG
RF1
67f
avo
rin
gS2
88C
)ch
rIX
233
67C
SS1
No
nsy
no
nym
ou
sD
914N
Su1
chrX
177
350
345
680
96ge
nes
incl
ud
ing
SUL2
CN
V(a
mp
lifica
tio
n)
uva
557
218
62
37(S
anch
eze
tal
201
7)ch
rXV
I466
649
DIG
1N
on
syn
on
ymo
us
E49Q
chrV
188
548
Inte
rgen
icSu
2ch
rX1
7735
03
4568
096
gen
esin
clu
din
gSU
L2C
NV
(am
plifi
cati
on
)ch
rIV
803
704
KT
R3
5rsquo-u
pst
ream
chrI
I121
779
PIN
4N
on
syn
on
ymo
us
N26
3Sch
rVII
165
902
MP
T5
No
nsy
no
nym
ou
sQ
618K
chrI
I836
169
RSC
3Sy
no
nym
ou
sR
4Rch
rIV
107
948
UFD
2Sy
no
nym
ou
sG
691G
chrI
II2
8761
8In
terg
enic
Poin
tm
uta
tio
ns
cop
yn
um
ber
vari
ants
(CN
V)
and
loss
ofh
eter
ozy
gosi
tyev
ents
(LO
H)
are
reco
rded
for
each
evo
lved
par
enta
lclo
ne
Clo
nes
are
iden
tifi
edb
yn
utr
ien
t(G
glu
cose
limit
atio
nP
ph
osp
hat
elim
itat
ion
an
dS
sulfa
telim
itat
ion
)b
ysp
ecie
s(ldquo
crdquod
eno
tes
Sce
revi
siae
ldquou
rdquod
eno
tes
Suv
arum
)an
dth
en
um
ber
ind
icat
esit
sd
eriv
atio
nfr
om
ind
epen
den
tp
op
ula
tio
ns
No
teth
atm
uta
tio
ns
inth
eS
uvar
umge
no
me
use
Suv
arum
chro
mo
som
esan
dco
ord
inat
esA
llb
reak
po
ints
wer
eca
lled
by
visu
alin
spec
tio
no
fse
qu
enci
ng
read
san
dar
eth
us
app
roxi
mat
eR
elat
ive
fitn
ess
isre
po
rted
wit
hst
and
ard
erro
r(S
E)s
ever
alcl
on
esar
ere
po
rted
wit
ho
ut
SEd
ue
tote
chn
ical
diffi
cult
ies
wit
hre
plic
ates
Adaptation in Hybrids doi101093molbevmsx098 MBE
7
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
had an average of 15 CNV per clone and the evolvedS uvarum clones had an average of 1 CNV per clone (table 2supplementary figs S1 and S2 Supplementary Material online)The most common event across nutrient limitations in theinterspecific hybrids was an amplification of the S cerevisiaecopy of chromosome IV which occurred in four
independent hybrid clones (three in phosphate limitationone in glucose limitation supplementary fig S3Supplementary Material online) Several other characteris-tic rearrangements occurred in the evolved S cerevisiaeclones including the amplification of the left arm of chro-mosome 14 accompanied by segmental monosomy of the
FIG 1 Evolved hybrids exhibit changes in copy number and loss of heterozygosity Copy number variants are displayed for selected evolved hybridclones from three nutrient-limited conditions Gh2 glucose Ph4 phosphate and Sh4 sulfate See additional figures in supplementary fig S3Supplementary Material online Hybrid copy number determined by normalized sequencing read depth per open reading frame (ORF) is plottedacross the genome according to S cerevisiae ORF coordinates to account for three reciprocal translocations between S cerevisiae and S uvarumChromosomes are plotted in alternating light and dark purple red indicates a S cerevisiae copy number variant and blue indicates a S uvarumcopy number variant Gh2 has a whole chromosome amplification of S cerevisiae chrIV a small segmental deletion of S uvarum chrIV (non-copyneutral loss of heterozygosity) and an amplification of S uvarum HXT67 Ph4 has a small segmental deletion of S cerevisiae chrIII (non-copyneutral loss of heterozygosity) and an amplification of S cerevisiae chrXIII with corresponding deletion of S uvarum chrXIII (copy neutral loss ofheterozygosity) Sh4 has an amplification of S cerevisiae SUL1 and a whole chromosome amplification of S uvarum chrVIII (note there is areciprocal translocation between chrVIII and chrXV in S uvarum relative to S cerevisiae) Note that Sh4 is plotted on a different scale For specificcoordinates of copy number variants see table 1
Smukowski Heil et al doi101093molbevmsx098 MBE
8
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
right arm of chromosome 14 an event seen previously inother evolved populations (Dunham et al 2002 Greshamet al 2008 Sunshine et al 2015) All copy number events in
S cerevisiae had break points at repetitive elements knownas Ty elements except those located on chrII which maybe mediated by another mechanism (Brewer et al 2015) In
FIG 2 Repeated loss of heterozygosity at the PHO84 locus in intra- and interspecific hybrids (A) The 25-kb region extending from the left telomereof chromosome XIII to the high-affinity phosphate transporter gene PHO84 (B) Copy number is plotted across part of chromosome XIII in thehybrid ancestor and three evolved hybrid clones in phosphate limitation (clone indicated in upper right corner) Red shows the S cerevisiae alleleblue shows the S uvarum allele and purple shows where both species exhibit the same copy number Note 8 kb of telomere sequence is removeddue to repetitive sequence (C) Alternate allele frequency is plotted for a portion of chromosome XIII in the ancestor and four evolved S cerevisiaeclones in phosphate limitation (clone indicated in upper right corner) All evolved S cerevisiae clones exhibit a loss of heterozygosity at thetelomeric portion of chromosome XIII (loss of S288C amplification of GRF167) as illustrated by an allele frequency of zero compared with theancestor Regions of heterozygosity are interspersed with regions of homozygosity as one of the parents of the diploid was itself the product of across between strains FL100 and S288C and the other parent was S288C Regions of heterozygosity are due to FL100 haplotypes S cerevisiae copynumber for the four evolved clones is shown below the ancestor is diploid across the chromosome (also see table 2 supplementary fig S1Supplementary Material online)
Adaptation in Hybrids doi101093molbevmsx098 MBE
9
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
contrast copy number variants in the hybrid were rarely facil-itated by repetitive elements perhaps in part because S uva-rum has no full length Ty elements however why S cerevisiaeTy elements and long terminal repeat (LTR) sequences fromeither genome were not utilized remains unknown
Frequently in nutrient-limited evolution experimentscopy number variants involve amplification of the nutrient-specific transporter and indeed we also observed amplifica-tion of these transporters in many of the clones In sulfatelimitation the S cerevisiae allele of the high-affinity sulfatetransporter gene SUL1 is amplified in 77 hybrid clones and 44 S cerevisiae clones (fig 1 tables 1 and 2 supplementary figsS1 and S3 Supplementary Material online) InterestinglySUL2 is the preferred sulfate transporter in S uvarum(Sanchez et al 2017) and was not observed to be amplifiedin the evolved hybrids (supplementary fig S2 and table S2Supplementary Material online) In glucose limitation previ-ous S cerevisiae evolution experiments found frequent am-plification of the high-affinity glucose transporter genesHXT67 (Brown et al 1998 Dunham et al 2002 Greshamet al 2008 Kao and Sherlock 2008 Kvitek and Sherlock2011) In our experiments the S uvarum alleles of theHXT67 transporters are amplified in 33 hybrid clones andboth S uvarum clones but are not amplified in evolved Scerevisiae clones This is suggestive that the S uvarum HXT67alleles confer a greater fitness advantage compared with Scerevisiae alternatively the genomic context could be morepermissive to amplification in S uvarum (fig 1 tables 1 and 2supplementary figs S1ndashS3 Supplementary Material online)Finally in phosphate limitation the S cerevisiae copy of thehigh-affinity phosphate transporter gene PHO84 is amplifiedand the S uvarum allele is lost in 36 hybrid clones in an eventknown as LOH (figs 1 and 2 table 1 supplementary fig 3Supplementary Material online) Intriguingly the evolved Scerevisiae clones also display LOH and accompanied amplifi-cation favoring the allele derived from strain GRF167 over theS288C allele in 44 clones (fig 2 table 2) All hybrid clonescarry the ldquopreferredrdquo GRF167 S cerevisiae allele as this was theallele used to create the de novo hybrid
Loss of Heterozygosity Is a Common Event inHeterozygous Evolving PopulationsSelection on heterozygosity as a LOH event could representis an underappreciated source of adaptation in microbial ex-perimental evolution as typical experiments evolve a haploidor homozygous diploid strain asexually and as a result havelittle variation to select upon LOH is observed in natural andindustrial hybrids (Albertin and Marullo 2012 Louis et al2012 Pryszcz et al 2014 Wolfe 2015 Schroder et al 2016)but here we document its occurrence in both intra- andinterspecific hybrids in the laboratory as a result of short-termevolution (also see (Dunn et al 2013 Burke et al 2014)) LOHis observed across all nutrient conditions with 13 independentLOH events detected in S cerevisiae and 9 independent eventsdocumented in the hybrids (figs 1 and 2 tables 1 and 2 sup-plementary figs S1 and S3 Supplementary Material online) Itthus appears that this type of mutational event is both com-mon and can occur over short evolutionary timescales
The LOH event can result in copy-neutral (where one alleleis lost and the other allele is amplified) or non-neutral chro-mosomal segments (where one allele is lost rendering thestrain hemizygous at that locus) and can favor the retentionof either allele In S cerevisiae there is a bias in resolutionwhere LOH events favor retaining the GRF167 allele over theS288C allele (1013 events Pfrac14 00169 table 2 supplementaryfig S4 Supplementary Material online) One unique case inclone Sc4 has a small approximately 5 kb LOH event onchrXV favoring GRF167 which switches to favoring S288Cfor the rest of the chromosome The retention of S cerevisiaeis slightly more common in the hybrids (59 events Pfrac14 10table 1 supplementary fig S3 Supplementary Material on-line) though not as drastic as the observed genome resolu-tion in the hybrid S pastorianus where LOH favorsS cerevisiae over S eubayanus (Nakao et al 2009) The sizeof the event ranges from approximately 25 kb to the wholechromosome level in the hybrids and from 5 kb to 540 kbin S cerevisiae Where LOH is accompanied by an amplifi-cation event the LOH event always occurs first and doesnot share break points with the amplification event Unlikemany CNV events most LOH events do not appear to bemediated by existing repetitive sequence such as a trans-posable element in the hybrid or S cerevisiae and are mostlikely a product of break-induced replication or mitoticgene conversion with break-induced replication as the fa-vored method as all events extend to the telomere (Hoanget al 2010) The exceptions are in hybrid clones Ph4 Ph5 andSh1 where there is a non-copy neutral loss of S cerevisiaemediated by a Ty element or LTR and S cerevisiae clonesSc1 and Sc4 where there is a 65-kb deletion of the S288Callele flanked by two Ty elements
LOH events in hybrids could signify several ongoing pro-cesses in hybrid genome evolution LOH regions may repre-sent 1) loci with incompatibilities 2) selection on existingvariation or 3) genetic drift eroding genomic segments It isimpossible to definitively rule out any of these hypotheseswithout further experimentation however there are severalarguments disfavoring the incompatibility hypothesis Firstalthough our sample size is modest failing to see repeatedLOH events across nutrient conditions may indicate a lack ofgeneral incompatibility between species (although we cannotpreclude condition-specific incompatibility Hou et al 2015Piatkowska et al 2013) This is consistent with previous stud-ies in yeast which suggest that classic DobzhanksyndashMullerincompatibilities are rare (Liti et al 2006 Greig 2009 Houet al 2014) Furthermore LOH events observed inevolved S cerevisiae suggest that this mutation type isnot unique to interspecific hybrids Instead repeatedevents within a particular condition such as the repeatedLOH at PHO84 in phosphate limitation or the 65 kbsegment on chrXIII in sulfate limitation suggest thatthese events are beneficial and are indeed selection onone allele over the other
LOH Is Driven by Selection on One AlleleTo test the hypothesis that LOH events provide a selectiveadvantage we used allele replacement in which the allele of
Smukowski Heil et al doi101093molbevmsx098 MBE
10
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
one speciesstrain is replaced with the allele of the otherspeciesstrain in an otherwise isogenic background We testedthis hypothesis using the most commonly seen LOH eventLOH at PHO84 Although the region extends from 25 kb to234 kb in length in the hybrids and from 40 kb to 85 kb in Scerevisiae and thus includes many genes PHO84 was a primecandidate driving this event PHO84 is one of only ten genesencompassed in the region extending from the telomere tothe break point of the shortest LOH event and is included inevery other LOH event on chromosome XIII (fig 2) It is ahigh-affinity phosphate transporter responsible for inorganicphosphate uptake in high and low phosphate conditions(Wykoff and OrsquoShea 2001) and previous work identified apoint mutation in PHO84 (an alanine to valine substitution atthe 50 end of the gene) which increased fitness by 23 inphosphate-limited conditions (Sunshine et al 2015) Finallyprior work with other nutrient transporters has shown am-plification of nutrient transporters to be a key event in adapt-ing to nutrient-limited conditions
We thus selected a region of approximately 25 kb encom-passing the PHO84 ORF its promoter and 30-UTR(Nagalakshmi et al 2008 Yassour et al 2009 Cherry et al2012) We created allele replacement strains using the twoalleles of S cerevisiae in a S cerevisiae diploid background thetwo alleles are 991 identical in this region (supplementaryfig S5 Supplementary Material online) and each strain isidentical to the ancestral strain used in our evolution exper-iments except at the PHO84 locus The S cerevisiae ancestorcarries one copy of GRF167 (preferred) and one copy ofS288C (ldquoun-preferredrdquo) so named due to which allele wasretained and amplified in the evolved clones To measureany resultant changes in fitness we competed each strainindividually against a fluorescent ancestral strain and mea-sured which strain overtook the culture Relative fitness isthus defined as the growth advantage per generation Twocopies of the un-preferred allele decreased relative fitness by531 (6186) whereas two copies of the preferred alleleincreased relative fitness by 993 (6027) This displays anoverall difference in relative fitness of 1524 between the un-preferred and preferred alleles By comparing the fitness ofthese allele replacement strains with the evolved clones (table 2)the allele replacement does not fully recapitulate the fit-ness gain observed in the evolved clone One explanation isthat the additional mutations present in the evolvedstrains also contribute to their total fitness Another expla-nation could be the increased copy number of the PHO84region that we see in these evolved clones To further ex-plore this fitness difference we cloned the GRF167 alleleonto a low copy number plasmid and transformed theallele replacement strain carrying two preferred S cerevi-siae alleles to simulate increased copy number of PHO84and saw only a minimal fitness increase of 176 (6106note all fitness competitions involving plasmids were donein conjunction with empty plasmids to take into accountany fitness effects from the plasmid itself) This supportsthe conclusion that relative fitness gains in the evolvedclone are largely due to the loss of the S288C allele andselection and amplification of the GRF167 allele with little
additional benefit from further amplification It could also bethe case that co-amplification of other genes in the segmentis required to attain the full benefit as previously observedby the contribution of BSD2 to the SUL1 amplicon in sulfatelimitation (Sunshine et al 2015 Payen et al 2016)
To understand the fitness effects of LOH and amplificationin the hybrids we generated hybrid strains with varying num-bers of S cerevisiae GRF167 PHO84 alleles Unfortunately wewere unable to obtain a successful strain carrying the pre-ferred S cerevisiae allele in a S uvarum background howeverwe were able to generate a S uvarum PHO84 knockout strainthus creating a hybrid with one copy of S cerevisiae PHO84When combined with a low copy plasmid carrying the sameallele this strain effectively has two or more copies ofS cerevisiae PHO84 in a hybrid background This hybrid straincan serve as a proxy for the LOH event observed in the evo-lution experiments and has a relative fitness gain of 2557(6288) The ancestral hybrid with the same plasmid (effec-tively 1 S uvarum allele and 2 or more S cerevisiae alleles) hasa relative fitness gain of 1253 (6131) The difference be-tween these two hybrids suggests that while amplification isbeneficial the highest fitness is achieved with the loss of theS uvarum allele (Pfrac14 00061)
Together these results support the conclusion that theS cerevisiae GRF167 allele is preferred over the S288C alleleand that S cerevisiae alleles are preferred over the S uvarumallele in the hybrid and hence that the LOH events seen inboth intra- and interspecific hybrids are the product ofselection
DiscussionIn summary we sought to understand forces underlying ge-nome stabilization and evolution in interspecific and intra-specific hybrids as they adapt to novel environments Weevolved and sequenced clones from 16 hybrid populationsand 16 parental populations to reveal a variety of mutationalevents conferring adaptation to 3 nutrient-limited conditionsOf particular note we find LOH in both evolved intraspecificand interspecific hybrid clones in all nutrient environmentspotentially signifying areas where selection has acted on pre-existing variation present in the ancestral clone We used anallele replacement strategy to test this hypothesis for a com-monly repeated LOH event and show that selection is indeeddriving the homogenization of the genome at this locusAlthough other studies in natural industrial and lab-evolved isolates have observed LOH we present the first em-pirical test of the causal evolutionary forces influencing theseevents This work can begin to help us understand past hy-bridization events and subsequent genome resolution in hy-brids in natural and artificial systems
Similarities and Differences between Intra- andInterspecific HybridsAlthough our sample size is modest we observe several in-teresting trends when comparing S uvarum clones (homo-zygous) S cerevisiae clones (intraspecific hybrid andpreviously published homozygous) and interspecific hybrid
Adaptation in Hybrids doi101093molbevmsx098 MBE
11
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
clones First theory and previous research predict the inter-specific hybrid may experience more genome instability in theform of chromosomal rearrangements and CNV events(Xiong et al 2011 Lloyd et al 2014 Chester et al 2015Mason and Batley 2015) Instead in our work the interspecifichybrid and S cerevisiae clones experience the same numberof CNV events (both 15 CNVsclone) However the mech-anism of CNV formation seems to differ between the hybridand S cerevisiae clones Whereas S cerevisiae CNV breakpoints typically occur at transposable elements in our studyand previous studies (Dunham et al 2002 Gresham et al2008 Fedoroff 2012) the interspecific hybrid CNVs do notutilize transposable elements although they obviously sharethe same sequence background as the S cerevisiae clones(albeit in one copy) Whether this difference is due to theabsence of full-length transposable elements in S uvarum isunknown but this could potentially explain the lower num-ber of CNV events in S uvarum clones (1 CNVclone) andpresents an intriguing direction for future study
The Predictability of EvolutionWe now have many examples of predictable evolution innatural systems (Losos et al 1998 Rundle et al 2000 Elmerand Meyer 2011 Conte et al 2012 Jones et al 2012 Martinand Orgogozo 2013 Wessinger and Rausher 2014) and inlaboratory experimental evolution in which there often ap-pears to be a limited number of high fitness pathways thatstrains follow when adapting to a particular condition (Fereaet al 1999 Woods et al 2006 Gresham et al 2008 Burke et al2010 Salverda et al 2011 Kawecki et al 2012 Kvitek andSherlock 2013 Lang and Desai 2014) For example it is wellestablished that amplifications of nutrient transporters aredrivers of adaptation in evolution in nutrient-limited condi-tions Previous work in our group has particularly focused onthe amplification of the high-affinity sulfate transporter geneSUL1 in sulfate-limited conditions which occurs in almostevery sulfate-limited evolution experiment and confers a fit-ness advantage of as much as 40 compared with the un-evolved ancestor strain The amplification of phosphatetransporters has been markedly less common and thus driv-ers of adaptation in this condition have been less clearGresham et al (2008) identified a whole chromosome ampli-fication of chrXIII in one population In a follow-up studySunshine et al (2015) found whole or partial amplification ofchrXIII in 38 populations A genome-wide screen for segmen-tal amplifications found a slight increase in fitness for a smalltelomeric segment of chromosome XIII and a A49V pointmutation in PHO84 was observed to increase fitness by 23However screens by Payen et al (2016) showed that althoughPHO84 is recurrently mutated in various experiments itshowed no benefit when amplified or deleted in phos-phate-limited conditions Finally additional evolution exper-iments recapitulated the point mutation seen in Sunshineet al (2015) in 2432 populations and saw amplification ofPHO84 in 832 populations (Miller A Dunham MJ unpub-lished data) It is important to note that all these experimentsused a strain background derived from S288C or CENPKboth of which carry the same (un-preferred) PHO84 allele
In our work we observed amplification of the S cerevisiaeGRF167 allele of PHO84 in 44 S cerevisiae clones from 4populations and 36 hybrid clones from 6 populations Thisamplification was always preceded by the loss of the S288Callele in S cerevisiae clones and the LOH break points arenever shared with the amplification break points There is a15 fitness difference between carrying two copies of theS288C allele of PHO84 compared with carrying two copiesof the GRF167 allele of PHO84 and additional copies of theGRF167 allele do not provide substantial further fitness gainsThe two alleles differ by several noncoding changes and threenonsynonymous substitutions a mutation from glutamicacid to aspartic acid (E229D) leucine to proline (L259P)and leucine to glutamine (L556Q) the latter two of whichare considered ldquononconservativerdquo protein mutations due tochanges in hydrophobicity and structure (supplementary figS5 Supplementary Material online) Intriguingly the L259Pmutation has actually been previously identified as being re-sponsible for resistance to tetrachloroisophthalonitrile andpartial resistance to pentachlorophenol in a QTL study ofsmall-molecule drugs (Perlstein et al 2007) IndeedPerlstein et al found proline at residue 259 to be conservedacross fungal species and even in orthologous human xeno-biotic transporters likely because proline-induced kinks intransmembrane spans have been shown to be essential toprotein function (Cordes et al 2002 Perlstein et al 2007) Itthus appears that amplification of PHO84 has been less pre-dictable as the S288C allele does not confer a fitness advan-tage unless mutated per Sunshine et al (2015) Togetherthese results imply that strain background can constrainadaptive pathways
In hybrids the amplification of the S cerevisiae segmentoccurred in conjunction with the loss of the S uvarum alleleHybrid strains with the LOH had a 2557 relative fitness gainwhereas hybrid strains with amplification of the S cerevisiaePHO84 allele without the LOH had a 1253 fitness gain ThusLOH confers an additional 1304 fitness gain showing thatselection for LOH has a larger impact on fitness than ampli-fication alone Note that S uvarum does have proline at res-idue 259 like the preferred GRF167 allele and differs fromboth S cerevisiae alleles at the other two coding substitutions(229N and 556K) but the amino acid and noncoding diver-gence is too high to speculate what substitutions are respon-sible for the selection of the GRF167 allele (supplementary figS6 Supplementary Material online) Why the loss of one alleleis more beneficial remains unclear as PHO84 is thought tofunction as a monomer (Bun-Ya et al 1991) but it may bedue to competition for cell wall space or negative interactionswith other genes in the PHO pathway (Mouillon and Persson2006)
The infusion of variation created by hybridization providesnew templates for selection to act upon which can be moreimportant than either point mutations or copy number var-iants alone Our work shows that outcrossing need not becommon to have long-lasting effects on adaptation This im-plication is particularly relevant in yeast where outcrossingmay occur quite rarely followed by thousands of asexual gen-erations (Ruderfer et al 2006 Greig and Leu 2009 Liti 2015)
Smukowski Heil et al doi101093molbevmsx098 MBE
12
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Applications to Other Hybrids and CancerThe observation that LOH occurs in hybrid genomes is in-creasingly documented (Louis et al 2012 Borneman et al2014 Soltis et al 2014 Marcet-Houben and Gabaldon 2015Pryszcz et al 2014 Schroder et al 2016) although the rea-son(s) for this type of mutation has been unresolved As mostexamples stem from allopolyploid events that occurred mil-lions of years ago understanding why LOH is important inhybrid genome evolution is difficult Cancer cells are alsoknown to experience LOH sometimes involved in the inac-tivation of tumor suppressor genes leaving only one copy ofthe gene that may be mutated or silenced (Thiagalingam et al2001 Tuna et al 2009 Lapunzina and Monk 2011) Datasupport the conclusion that LOH events are selected for dur-ing tumor development as many LOH events involve specificchromosomal segments (Thiagalingam et al 2001) althoughthe underlying molecular and genetic reasons for selection isan open debate (Ryland et al 2015)
Here we experimentally demonstrate that LOH can occurin homoploid interspecific hybrids as well as in intraspecifichybrids These events occur within a few hundred generationsand are common mutations more common on average inthe intraspecific hybrid (13 eventsclone) than the interspe-cific hybrid (056 eventsclone) Interestingly the LOH eventsdo not share break points with the CNV events in S cerevisiaeinstead they appear to occur independently and to precedeany subsequent amplification (amplification occurs following913 LOH events) Competition assays with the PHO84 locusprovide support that LOH itself may be more beneficial thanamplification or at least increase the selective benefit of am-plification events The observation that LOH in intraspecifichybrids occurs independently from copy number change pro-vides different opportunities for adaptation to novelconditions
Other cases of LOH like the copy number neutral eventsobserved in Ph4 Ph5 and Sh4 all of which favor the retentionof the S uvarum allele (supplementary fig S3 SupplementaryMaterial online) may be due to hybrid incompatibility withina particular protein complex other epistatic interactions(Piatkowska et al 2013) or neutral processes We furthermorediscover examples where one species allele appears to bepreferred over the other without LOH such as the repeatedamplification of the S uvarum high-affinity glucose transpor-ters HXT67 When one species allele is amplified and theother is not amplified one explanation is that the local se-quence context can permit or deny amplification In the caseof HXT67 previous experiments in S cerevisiae have shownthat amplification of HXT67 is quite common (Brown et al1998 Dunham et al 2002 Gresham et al 2008 Kao andSherlock 2008 Kvitek and Sherlock 2011) thus suggestingthat when given a choice between this locus in S cerevisiaeor S uvarum in the hybrid the preferred allele is indeed Suvarum though more subtle differences in rate cannot yet beruled out A similar scenario is observed with S cerevisiaeSUL1 (Sanchez et al 2017) Together our results show thatthe heterozygosity supplied by hybridization is an importantcontributor to adaptive routes explored by populations asthey adapt to novel conditions
Although we cannot generalize our results from thePHO84 locus across the many other LOH events discoveredin our hybrids and S cerevisiae in the future we can usesimilar methodology to explore whether positive selectionalways drives LOH or whether other explanations such asincompatibility resolution contribute as well Future experi-ments might also utilize a high throughput method to ex-plore segmental LOH in hybrids at a genome-wide scalesimilar to ongoing experiments at the gene level (LancasterS Dunham MJ unpublished data) Although our sample sizeis moderate this is a novel and necessary step in understand-ing forces underlying hybrid genome stabilization and high-lighting an underappreciated mechanism of hybridadaptation
ConclusionsThe mutation events we observe in our experimentallyevolved hybrids are in many ways quite representative ofmutations observed in ancient hybrid genomes suggestingthat hybrid genome stabilization and adaptation can occurquite rapidly (within several hundred generations)Furthermore our results illustrate that the infusion of varia-tion introduced by hybridization at both the intra- and inter-species level can increase fitness by providing choices of allelesfor selection to act upon even when sexual reproduction israre This may be particularly important for leveraging existingvariation for agricultural and industrial processes and as cli-mate change potentially increases natural hybridization (Kellyet al 2010 Hoffmann and Sgro 2011 Muhlfeld et al 2014)
Materials and Methods
StrainsA list of strains used in this study is included in supplementarytable S1 Supplementary Material online All interspecific hy-brids were created by crossing a ura3 LYS2 haploid parent to aURA3 lys2 haploid parent of the other mating type plating onmedia lacking both uracil and lysine and selecting for proto-trophs The S cerevisiae strain background known asldquoGRF167rdquo is itself a cross between FL100 and the genomictype strain S288C (data not shown) GRF167 was chosen as astrain background for simultaneous work investigating trans-posable elements during experimental evolution which willbe addressed in a future study
Evolution ExperimentsContinuous cultures were established using media and con-ditions previously described (Gresham et al 2008 Sanchezet al 2017) Detailed protocols and media recipes are availableat httpdunhamgswashingtoneduprotocolsshtml (lastaccessed March 6 2017) Samples were taken daily and mea-sured for optical density at 600 nm and cell count micros-copy was performed to check for contamination andarchival glycerol stocks were made daily An experimentwas terminated when contamination growth in tubing orclumping appeared (number of generations at the endpoint for each population are presented in tables 1 and
Adaptation in Hybrids doi101093molbevmsx098 MBE
13
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
2) Samples from each end point population were colonypurified to yield two clones for further study
Array Comparative Genomic HybridizationPopulations from the end point of each evolution were ana-lyzed for copy number changes using array comparative ge-nomic hybridization following the protocol used in Sanchezet al (2017)
SequencingDNA was extracted from overnight cultures using theHoffmanndashWinston protocol (Hoffman and Winston 1987)and cleaned using the Clean amp Concentrator kit (ZymoResearch) Nextera libraries were prepared following theNextera library kit protocol and sequenced using paired end150 bp reads on the Illumina NextSeq 500 machine (sequenc-ing coverage in supplementary table S2 SupplementaryMaterial online) The reference genomes used were S cerevi-siae v3 (Engel et al 2014) S uvarum (Scannell et al 2011) anda hybrid reference genome created by concatenating the twogenomes Sequence was aligned to the appropriate referencegenome using bwa v062 (Li and Durbin 2009) and mutationswere called using GATK (McKenna et al 2010) and samtools0119 (Li et al 2009) Mutations in evolved clones were filteredin comparison with the ancestor to obtain de novo mutationsAll mutations were first visually inspected using IntegrativeGenomics Viewer (Robinson et al 2011) Subsequently pointmutations in the hybrids were confirmed with Sanger se-quencing (supplementary table S3 Supplementary Materialonline) Copy number variants were visualized usingDNAcopy for S cerevisiae and S uvarum (Seshan andOlshen 2016) LOH events were called based on sequencingcoverage in the hybrids and by identifying homozygous variantcalls in S cerevisiae All break points were called by visual in-spection of sequencing reads and are thus approximate
Fitness AssaysThe pairwise competition experiments were performed in 20ml chemostats (Miller and Dunham 2013) Each competitorstrain was cultured individually until steady state was reachedand then was mixed 5050 with a GFP-tagged ancestor Eachcompetition was conducted in at least two biological repli-cates for approximately 15 generations after mixing Sampleswere collected and analyzed twice daily The proportion ofGFPthorn cells in the population was detected using a BD AccuriC6 flow cytometer (BD Biosciences) The data were plottedwith ln [dark cellsGFPthorn cells] versus generations The rela-tive fitness coefficient was determined from the slope of thelinear region
Strain ConstructionAllele replacements for the PHO84 locus were done followingthe protocol of the Caudy lab with further modifications de-scribed here The native locus was replaced withKluyveromyces lactis URA3 The pho84DURA3 strain wasgrown overnight in 5 ml of C-URA media then inoculatedin a flask of 100 ml yeast extract peptone dextrose (YPD) andgrown to an optical density of 06ndash08 Cells were washed then
aliquoted 275 ml of transformation mix (35 ml 1 M lithiumacetate 240 ml of 50 3500 polyethylene glycol) 10 ml ofsalmon sperm and approximately 3 mg of polymerase chainreaction product were added to the cell pellet It was incu-bated at 37 C (S uvarum) or 42 C (S cerevisiae) for 45 minthen plated to YPD It was replica plated to 5-fluorooroticacid the following day and colonies were tested for the gain ofthe appropriate species allele The GRF167 allele was clonedinto the pIL37 plasmid using Gibson assembly (Gibson et al2009) Correct assembly was verified by Sanger sequencing Allprimers used can be found in supplementary table S3Supplementary Material online
Supplementary MaterialSupplementary data are available at Molecular Biology andEvolution online
AcknowledgmentsWe thank Noah Hanson and Erica Alcantara for technicalassistance We thank Monica Sanchez and three anonymousreviewers for helpful comments on this manuscript JosephSchacherer kindly confirmed the contribution of FL100 to theGRF167 strain background Thanks to Yixian Zheng and DougKoshland for contributing to the initial experimental designcreating yeast strains and purchasing the oligonucleotidesused for the microarrays This work was supported by theNational Science Foundation (grant number 1516330) andthe National Institutes of Health (grant number R01GM094306) MD is a Senior Fellow in the GeneticNetworks program at the Canadian Institute for AdvancedResearch and a Rita Allen Foundation Scholar CSH wassupported in part by National Institutes of Health (grantnumber T32 HG00035) This work was also supported bythe National Institutes of Health (grant number P50GM071508) to the Lewis-Sigler Institute and from theHoward Hughes Medical Institute to Doug Koshland andYixian Zheng Sequencing read data have been deposited atthe NCBI under BioProject accession PRJNA374049Microarray data are deposited in the Gene ExpressionOmnibus (GEO) repository under accession numbersGSE95086 and GSE87401 and in the Princeton MicroarrayDatabase
ReferencesAbbott R Albach D Ansell S Arntzen JW Baird SJ Bierne N Boughman J
Brelsford A Buerkle CA Buggs R et al 2013 Hybridization andspeciation J Evol Biol 26229ndash246
Ainouche ML Jenczewski E 2010 Focus on polyploidy New Phytol1861ndash4
Albertin W Marullo P 2012 Polyploidy in fungi evolution after whole-genome duplication Proc Biol Sci 2792497ndash2509
Almeida P Goncalves C Teixeira S Libkind D Bontrager M Masneuf-Pomarede I Albertin W Durrens P Sherman DJ Marullo P et al2014 A Gondwanan imprint on global diversity and domesticationof wine and cider yeast Saccharomyces uvarum Nat Commun54044
Antunovics Z Nguyen HV Gaillardin C Sipiczki M 2005 Gradual ge-nome stabilisation by progressive reduction of the Saccharomyces
Smukowski Heil et al doi101093molbevmsx098 MBE
14
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
uvarum genome in an interspecific hybrid with Saccharomyces cer-evisiae FEMS Yeast Res 51141ndash1150
Baker E Wang B Bellora N Peris D Hulfachor AB Koshalek JA AdamsM Libkind D Hittinger CT 2015 The Genome sequence ofSaccharomyces eubayanus and the domestication of lager-brewingyeasts Mol Biol Evol 322818ndash2831
Barbosa R Almeida P Safar SVB Santos RO Morais PB Nielly-Thibault LLeducq JB Landry CR Goncalves P Rosa CA Sampaio JP 2016Evidence of natural hybridization in Brazilian wild lineages ofSaccharomyces cerevisiae Genome Biol Evol 8317ndash329
Barton NH 2001 The role of hybridization in evolution Mol Ecol10551ndash568
Belloch C Orlic S Barrio E Querol A 2008 Fermentative stress adapta-tion of hybrids within the Saccharomyces sensu stricto complex Int JFood Microbiol 122188ndash195
Bellon JR Yang F Day MP Inglis DL Chambers PJ 2015 Designingand creating Saccharomyces interspecific hybrids for improvedindustry relevant phenotypes Appl Microbiol Biotechnol998597ndash8609
Borneman AR Zeppel R Chambers PJ Curtin CD 2014 Insights into theDekkera bruxellensis genomic landscape comparative genomics re-veals variations in ploidy and nutrient utilisation potential amongstwine isolates PLoS Genet 10e1004161
Brewer BJ Payen C Di Rienzi SC Higgins MM Ong G Dunham MJRaghuraman MK 2015 Origin-dependent inverted-repeat amplifi-cation tests of a model for inverted DNA amplification PLoS Genet11e1005699
Brown CJ Todd KM Rosenzweig RF 1998 Multiple duplications of yeasthexose transport genes in response to selection in a glucose-limitedenvironment Mol Biol Evol 15931ndash942
Bullini L 1994 Origin and evolution of animal hybrid species Trends EcolEvol 9422ndash426
Bun-Ya M Nishimura M Harashima S Oshima Y 1991 The PHO84 geneof Saccharomyces cerevisiae encodes an inorganic phosphate trans-porter Mol Cell Biol 113229ndash3238
Burke MK Dunham JP Shahrestani P Thornton KR Rose MR Long AD2010 Genome-wide analysis of a long-term evolution experimentwith Drosophila Nature 467587ndash590
Burke MK Liti G Long AD 2014 Standing genetic variation drives re-peatable experimental evolution in outcrossing populations ofSaccharomyces cerevisiae Mol Biol Evol 313228ndash3239
Cherry JM Hong EL Amundsen C Balakrishnan R Binkley G Chan ETChristie KR Costanzo MC Dwight SS Engel SR et al 2012Saccharomyces genome database the genomics resource of buddingyeast Nucleic Acids Res 40D700ndashD705
Chester M Riley RK Soltis PS Soltis DE 2015 Patterns of chromosomalvariation in natural populations of the neoallotetraploid Tragopogonmirus (Asteraceae) Heredity (Edinb) 114309ndash317
Chou JY Leu JY 2010 Speciation through cytonuclear incompatibilityinsights from yeast and implications for higher eukaryotes Bioessays32401ndash411
Cliften PF Fulton RS Wilson RK Johnston M 2006 After the duplicationgene loss and adaptation in Saccharomyces genomes Genetics172863ndash872
Conte GL Arnegard ME Peichel CL Schluter D 2012 The probability ofgenetic parallelism and convergence in natural populations Proc BiolSci 2795039ndash5047
Cordes FS Bright JN Sansom MS 2002 Proline-induced distortions oftransmembrane helices J Mol Biol 323951ndash960
Crow JF 1998 90 years ago the beginning of hybrid maize Genetics148923ndash928
Dasmahapatra KK Walters JR Briscoe AD Davey JW Whibley A NadeauNJ Zimin AV Hughes DST Ferguson LC Martin SH et al 2012Butterfly genome reveals promiscuous exchange of mimicry adap-tations among species Nature 48794ndash98
Dowling TE Smith GR Brown WM 1989 Reproductive isolation andintrogression between Notropis-Cornutus and Notropis-Chrysocephalus (Family Cyprinidae)mdashcomparison of morphologyallozymes and mitochondrial-DNA Evolution 43620ndash634
Doyle JJ Flagel LE Paterson AH Rapp RA Soltis DE Soltis PS Wendel JF2008 Evolutionary genetics of genome merger and doubling inplants Annu Rev Genet 42443ndash461
Dunham MJ Badrane H Ferea T Adams J Brown PO Rosenzweig FBotstein D 2002 Characteristic genome rearrangements in experi-mental evolution of Saccharomyces cerevisiae Proc Natl Acad Sci U SA 9916144ndash16149
Dunn B Paulish T Stanbery A Piotrowski J Koniges G Kroll E LouisEJ Liti G Sherlock G Rosenzweig F 2013 Recurrent rearrange-ment during adaptive evolution in an interspecific yeast hybridsuggests a model for rapid introgression PLoS Genet9e1003366
Dunn B Richter C Kvitek DJ Pugh T Sherlock G 2012 Analysis of theSaccharomyces cerevisiae pan-genome reveals a pool of copy num-ber variants distributed in diverse yeast strains from differing indus-trial environments Genome Res 22908ndash924
Dunn B Sherlock G 2008 Reconstruction of the genome origins andevolution of the hybrid lager yeast Saccharomyces pastorianusGenome Res 181610ndash1623
Elmer KR Meyer A 2011 Adaptation in the age of ecological genomicsinsights from parallelism and convergence Trends Ecol Evol26298ndash306
Engel SR Dietrich FS Fisk DG Binkley G Balakrishnan R Costanzo MCDwight SS Hitz BC Karra K Nash RS et al 2014 The referencegenome sequence of Saccharomyces cerevisiae then and now G3(Bethesda) 4389ndash398
Fedoroff NV 2012 Presidential address Transposable elements epige-netics and genome evolution Science 338758ndash767
Ferea TL Botstein D Brown PO Rosenzweig RF 1999 Systematicchanges in gene expression patterns following adaptive evolutionin yeast Proc Natl Acad Sci U S A 969721ndash9726
Gibson B Liti G 2015 Saccharomyces pastorianus genomic insightsinspiring innovation for industry Yeast 3217ndash27
Gibson DG Young L Chuang RY Venter JC Hutchison CA Smith HO2009 Enzymatic assembly of DNA molecules up to several hundredkilobases Nat Methods 6343ndash5
Gonzalez SS Barrio E Gafner J Querol A 2006 Natural hybrids fromSaccharomyces cerevisiae Saccharomyces bayanus andSaccharomyces kudriavzevii in wine fermentations FEMS Yeast Res61221ndash1234
Gonzalez SS Barrio E Querol A 2008 Molecular characterization of newnatural hybrids of Saccharomyces cerevisiae and S kudriavzevii inbrewing Appl Environ Microbiol 742314ndash2320
Grant PR Grant BR 1994 Phenotypic and genetic consequences ofhybridization in Darwinrsquos finches Evolution 48297ndash316
Grant PR Grant BR 2002 Unpredictable evolution in a 30-year study ofDarwinrsquos finches Science 296707ndash711
Grant PR Grant BR 2010 Natural selection speciation and Darwinrsquosfinches Proc Calif Acad Sci 61245ndash260
Grant PR Grant BR Petren K 2005 Hybridization in the recent past AmNat 16656ndash67
Greig D 2009 Reproductive isolation in Saccharomyces Heredity (Edinb)10239ndash44
Greig D Leu JY 2009 Natural history of budding yeast Current Biol19R886ndashR890
Greig D Louis EJ Borts RH Travisano M 2002 Hybrid speciation inexperimental populations of yeast Science 2981773ndash1775
Gresham D Desai MM Tucker CM Jenq HT Pai DA Ward A DeSevoCG Botstein D Dunham MJ 2008 The repertoire and dynamics ofevolutionary adaptations to controlled nutrient-limited environ-ments in yeast PLoS Genet 4e1000303
Hajjar R Hodgkin T 2007 The use of wild relatives in crop improvementa survey of developments over the last 20 years Euphytica 1561ndash13
Heiser CB 1954 Variation and subspeciation in the common sunflowerHelianthus annuus Am Midl Nat 51287ndash305
Hittinger CT 2013 Saccharomyces diversity and evolution a buddingmodel genus Trends Genet 29309ndash317
Hoang ML Tan FJ Lai DC Celniker SE Hoskins RA Dunham MJ ZhengY Koshland D 2010 Competitive repair by naturally dispersed
Adaptation in Hybrids doi101093molbevmsx098 MBE
15
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
repetitive DNA during non-allelic homologous recombination PLoSGenet 6e1001228
Hoffmann AA Sgro CM 2011 Climate change and evolutionary adap-tation Nature 470479ndash485
Hoffman CS Winston F 1987 A 10-minute DNA preparation from yeastefficiently releases autonomous plasmids for transformation ofEscherichia coli Gene 57267ndash272
Hou J Friedrich A de Montigny J Schacherer J 2014 Chromosomalrearrangements as a major mechanism in the onset of reproductiveisolation in Saccharomyces cerevisiae Curr Biol 241153ndash1159
Hou J Friedrich A Gounot JS Schacherer J 2015 Comprehensive surveyof condition-specific reproductive isolation reveals genetic incom-patibility in yeast Nat Commun 67214
Huerta-Sanchez E Casey FP 2015 Archaic inheritance supporting high-altitude life in Tibet J Appl Physiol (1985) 1191129ndash1134
Jones FC Grabherr MG Chan YF Russell P Mauceli E Johnson JSwofford R Pirun M Zody MC White S et al 2012 The genomicbasis of adaptive evolution in threespine sticklebacks Nature48455ndash61
Kao KC Sherlock G 2008 Molecular characterization of clonal interfer-ence during adaptive evolution in asexual populations ofSaccharomyces cerevisiae Nat Genet 401499ndash1504
Kawecki TJ Lenski RE Ebert D Hollis B Olivieri I Whitlock MC 2012Experimental evolution Trends Ecol Evol 27547ndash560
Kellis M Patterson N Endrizzi M Birren B Lander ES 2003 Sequencingand comparison of yeast species to identify genes and regulatoryelements Nature 423241ndash254
Kelly B Whiteley A Tallmon D 2010 The Arctic melting pot Nature468891ndash891
Kvitek DJ Sherlock G 2011 Reciprocal sign epistasis between frequentlyexperimentally evolved adaptive mutations causes a rugged fitnesslandscape PLoS Genet 7e1002056
Kvitek DJ Sherlock G 2013 Whole genome whole population sequenc-ing reveals that loss of signaling networks is the major adaptivestrategy in a constant environment PLoS Genet 9e1003972
Landry CR Hartl DL Ranz JM 2007 Genome clashes in hybrids insightsfrom gene expression Heredity (Edinb) 99483ndash493
Lang GI Desai MM 2014 The spectrum of adaptive mutations in ex-perimental evolution Genomics 104412ndash416
Lapunzina P Monk D 2011 The consequences of uniparental disomyand copy number neutral loss-of-heterozygosity during human de-velopment and cancer Biol Cell 103303ndash317
Leducq JB Nielly-Thibault L Charron G Eberlein C Verta JP Samani PSylvester K Hittinger CT Bell G Landry CR 2016 Speciation drivenby hybridization and chromosomal plasticity in a wild yeast NatMicrobiol 115003
Lee HY Chou JY Cheong L Chang NH Yang SY Leu JY 2008Incompatibility of nuclear and mitochondrial genomes causes hy-brid sterility between two yeast species Cell 1351065ndash1073
Li H Durbin R 2009 Fast and accurate short read alignment withBurrows-Wheeler transform Bioinformatics 251754ndash1760
Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R 2009 The Sequence AlignmentMap formatand SAMtools Bioinformatics 252078ndash2079
Liti G 2015 The fascinating and secret wild life of the budding yeast Scerevisiae Elife 4e05835
Liti G Barton DB Louis EJ 2006 Sequence diversity reproductiveisolation and species concepts in Saccharomyces Genetics174839ndash850
Lloyd AH Ranoux M Vautrin S Glover N Fourment J Charif D ChouletF Lassalle G Marande W Tran J et al 2014 Meiotic gene evolutioncan you teach a new dog new tricks Mol Biol Evol 311724ndash1727
Losos JB Jackman TR Larson A de Queiroz K Rodriguez-Schettino L1998 Contingency and determinism in replicated adaptive radia-tions of island lizards Science 2792115ndash2118
Louis VL Despons L Friedrich A Martin T Durrens P Casaregola SNeuveglise C Fairhead C Marck C Cruz JA et al 2012 Pichia sorbi-tophila an interspecies yeast hybrid reveals early steps of genomeresolution after polyploidization G3 (Bethesda) 2299ndash311
Mallet J 2007 Hybrid speciation Nature 446279ndash283Marcet-Houben M Gabaldon T 2015 Beyond the whole-genome du-
plication phylogenetic evidence for an ancient interspecies hybrid-ization in the bakerrsquos yeast lineage PLoS Biol 13e1002220
Marinoni G Manuel M Petersen RF Hvidtfeldt J Sulo P Piskur J 1999Horizontal transfer of genetic material among Saccharomyces yeastsJ Bacteriol 1816488ndash6496
Martin A Orgogozo V 2013 The loci of repeated evolution a catalog ofgenetic hotspots of phenotypic variation Evolution 671235ndash1250
Masly JP Jones CD Noor MAF Locke J Orr HA 2006 Gene transpositionas a cause of hybrid sterility in Drosophila Science 3131448ndash1450
Mason AS Batley J 2015 Creating new interspecific hybrid and polyploidcrops Trends Biotechnol 33436ndash441
Mavarez J Salazar CA Bermingham E Salcedo C Jiggins CD Linares M2006 Speciation by hybridization in Heliconius butterflies Nature441868ndash871
McKenna A Hanna M Banks E Sivachenko A Cibulskis K Kernytsky AGarimella K Altshuler D Gabriel S Daly M et al 2010 The genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNA sequencing data Genome Res 201297ndash1303
Meiklejohn CD Holmbeck MA Siddiq MA Abt DN Rand DMMontooth KL 2013 An Incompatibility between a mitochondrialtRNA and its nuclear-encoded tRNA synthetase compromises de-velopment and fitness in Drosophila PLoS Genet 9e1003238
Meyer A Salzburger W Schartl M 2006 Hybrid origin of a swordtailspecies (Teleostei Xiphophorus clemenciae) driven by sexual selec-tion Mol Ecol 15721ndash730
Michalak P 2009 Epigenetic transposon and small RNA determinants ofhybrid dysfunctions Heredity (Edinb) 10245ndash50
Miller AW Dunham MJ 2013 Design and use of multiplexed chemostatarrays J Vis Exp 72e50262
Monod J 1949 The growth of bacterial cultures Annu Rev Microbiol3371ndash394
Mouillon JM Persson BL 2006 New aspects on phosphate sensing andsignalling in Saccharomyces cerevisiae FEMS Yeast Res 6171ndash176
Muhlfeld CC Kovach RP Jones LA Al-Chokhachy R Boyer MC Leary RFLowe WH Luikart G Allendorf FW 2014 Invasive hybridization in athreatened species is accelerated by climate change Nat ClimChange 4620ndash624
Muller LAH McCusker JH 2009 A multispecies-based taxonomic mi-croarray reveals interspecies hybridization and introgression inSaccharomyces cerevisiae FEMS Yeast Res 9143ndash152
Nagalakshmi U Wang Z Waern K Shou C Raha D Gerstein M SnyderM 2008 The transcriptional landscape of the yeast genome definedby RNA sequencing Science 3201344ndash1349
Nakao Y Kanamori T Itoh T Kodama Y Rainieri S Nakamura NShimonaga T Hattori M Ashikari T 2009 Genome sequence ofthe lager brewing yeast an interspecies hybrid DNA Res 16115ndash129
Nolte AW Freyhof J Stemshorn KC Tautz D 2005 An invasive lineage ofsculpins Cottus sp (Pisces Teleostei) in the Rhine with new habitatadaptations has originated from hybridization between old phylo-geographic groups Proc Biol Sci 2722379ndash2387
Novick A Szilard L 1950a Description of the chemostat Science112715ndash716
Novick A Szilard L 1950b Experiments with the Chemostat on spon-taneous mutations of bacteria Proc Natl Acad Sci U S A 36708ndash719
Ouyang YD Liu YG Zhang QF 2010 Hybrid sterility in plant storiesfrom rice Curr Opin Plant Biol 13186ndash192
Payen C Di Rienzi SC Ong GT Pogachar JL Sanchez JC SunshineAB Raghuraman MK Brewer BJ Dunham MJ 2014 The dynam-ics of diverse segmental amplifications in populations ofSaccharomyces cerevisiae adapting to strong selection G3(Bethesda) 4399ndash409
Payen C Sunshine AB Ong GT Pogachar JL Zhao W Dunham MJ 2016High-throughput identification of adaptive mutations in experimen-tally evolved yeast populations PLoS Genet 12e1006339
Perez-Prat E van Lookeren Campagne MM 2002 Hybrid seed produc-tion and the challenge of propagating male-sterile plants TrendsPlant Sci 7199ndash203
Smukowski Heil et al doi101093molbevmsx098 MBE
16
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17
Peris D Langdon QK Moriarty RV Sylvester K Bontrager M Charron GLeducq JB Landry CR Libkind D Hittinger CT 2016 Complex ances-tries of lager-brewing hybrids were shaped by standing variation inthe wild yeast Saccharomyces eubayanus PLoS Genet 12e1006155
Perlstein EO Ruderfer DM Roberts DC Schreiber SL Kruglyak L 2007Genetic basis of individual differences in the response to small-molecule drugs in yeast Nat Genet 39496ndash502
Piatkowska EM Naseeb S Knight D Delneri D 2013 Chimeric proteincomplexes in hybrid species generate novel phenotypes PLoS Genet9e1003836
Piotrowski JS Nagarajan S Kroll E Stanbery A Chiotti KE Kruckeberg ALDunn B Sherlock G Rosenzweig F 2012 Different selective pressureslead to different genomic outcomes as newly-formed hybrid yeastsevolve BMC Evol Biol 1246
Pryszcz LP Nemeth T Gacser A Gabaldon T 2014 Genome comparisonof Candida orthopsilosis clinical strains reveals the existence of hy-brids between two distinct subspecies Genome Biol Evol61069ndash1078
Racimo F Sankararaman S Nielsen R Huerta-Sanchez E 2015 Evidencefor archaic adaptive introgression in humans Nat Rev Genet16359ndash371
Rieseberg LH 1991 Homoploid reticulate evolution in Helianthus(Asteraceae)mdashevidence from ribosomal genes Am J Bot781218ndash1237
Rieseberg LH 1997 Hybrid origins of plant species Annu Rev EcolSystemat 28359ndash389
Robinson JT Thorvaldsdottir H Winckler W Guttman M Lander ESGetz G Mesirov JP 2011 Integrative genomics viewer NatBiotechnol 2924ndash26
Rosenthal DM Schwarzbach AE Donovan LA Raymond O RiesebergLH 2002 Phenotypic differentiation between three ancient hybridtaxa and their parental species Int J Plant Sci 163387ndash398
Ruderfer DM Pratt SC Seidel HS Kruglyak L 2006 Population genomicanalysis of outcrossing and recombination in yeast Nat Genet381077ndash1081
Rundle HD Nagel L Boughman JW Schluter D 2000 Natural selectionand parallel speciation in sympatric sticklebacks Science287306ndash308
Ryland GL Doyle MA Goode D Boyle SE Choong DY Rowley SM Li JBowtell DD Tothill RW Campbell IG et al 2015 Australian OvarianCancer Study Group Loss of heterozygosity what is it good for BMCMed Genomics 845
Salverda MLM Dellus E Gorter FA Debets AJM van der Oost J HoekstraRF Tawfik DS de Visser JAGM 2011 Initial mutations direct alter-native pathways of protein evolution PLoS Genet 7e1001321
Sanchez MR Miller AW Liachko I Sunshine AB Lynch B Huang MAlcantara E DeSevo CG Pai DA Tucker CM et al 2017 Differentialparalog divergence modulates genome evolution across yeast spe-cies PLoS Genet 13(2)e1006585
Scannell DR Zill OA Rokas A Payen C Dunham MJ Eisen MB Rine JJohnston M Hittinger CT 2011 The awesome power of yeast evo-lutionary genetics new genome sequences and strain resources forthe Saccharomyces sensu stricto genus G3 (Bethesda) 111ndash25
Schroder MS Martinez de San Vicente K Prandini TH Hammel SHiggins DG Bagagli E Wolfe KH Butler G 2016 Multiple originsof the pathogenic yeast Candida orthopsilosis by separate hybridiza-tions between two parental species PLoS Genet 12e1006404
Schumer M Rosenthal GG Andolfatto P 2014 How common is homo-ploid hybrid speciation Evolution 681553ndash1560
Schwarzbach AE Donovan LA Rieseberg LH 2001 Transgressivecharacter expression in a hybrid sunflower species Am J Bot88270ndash277
Selmecki AM Maruvka YE Richmond PA Guillet M Shoresh NSorenson AL De S Kishony R Michor F Dowell R et al 2015Polyploidy can drive rapid adaptation in yeast Nature 519349ndash52
Seshan VE Olshen A 2016 DNAcopy DNA copy number data analysisVersion R package version 1460
Soltis DE Visger CJ Soltis PS 2014 The polyploidy revolution then andnow Stebbins revisited Am J Bot 1011057ndash1078
Soltis PS 2013 Hybridization speciation and novelty J Evol Biol26291ndash293
Soltis PS Soltis DE 2009 The role of hybridization in plant speciationAnnu Rev Plant Biol 60561ndash588
Sunshine AB Payen C Ong GT Liachko I Tan KM Dunham MJ 2015The fitness consequences of aneuploidy are driven by condition-dependent gene effects PLoS Biol 13e1002155
Tamai Y Momma T Yoshimoto H Kaneko Y 1998 Co-existence of twotypes of chromosome in the bottom fermenting yeastSaccharomyces pastorianus Yeast 14923ndash933
Taylor DJ Hebert PDN 1993 Habitat-dependent hybrid parentage anddifferential introgression between neighboringly sympatric Daphniaspecies Proc Natl Acad Sci U S A 907079ndash7083
Thiagalingam S Laken S Willson JK Markowitz SD Kinzler KWVogelstein B Lengauer C 2001 Mechanisms underlying losses ofheterozygosity in human colorectal cancers Proc Natl Acad Sci US A 982698ndash2702
Tuna M Knuutila S Mills GB 2009 Uniparental disomy in cancer TrendsMol Med 15120ndash128
Walther A Hesselbart A Wendland J 2014 Genome sequence ofSaccharomyces carlsbergensis the worldrsquos first pure culture lageryeast G3 (Bethesda) 4783ndash793
Wayne RK 1993 Molecular evolution of the dog family Trends Genet9218ndash224
Wenger JW Schwartz K Sherlock G 2010 Bulk segregant analysis byhigh-throughput sequencing reveals a novel xylose utilization genefrom Saccharomyces cerevisiae PLoS Genet 6e1000942
Wessinger CA Rausher MD 2014 Predictability and irreversibility ofgenetic changes associated with flower color evolution inPenstemon barbatus Evolution 681058ndash1070
Wolfe KH 2015 Origin of the yeast whole-genome duplication PLoSBiol 13e1002221
Woods R Schneider D Winkworth CL Riley MA Lenski RE 2006 Testsof parallel molecular evolution in a long-term experiment withEscherichia coli Proc Natl Acad Sci U S A 1039107ndash9112
Wykoff DD OrsquoShea EK 2001 Phosphate transport and sensing inSaccharomyces cerevisiae Genetics 1591491ndash1499
Xiong Z Gaeta RT Pires JC 2011 Homoeologous shuffling andchromosome compensation maintain genome balance inresynthesized allopolyploid Brassica napus Proc Natl Acad SciU S A 1087908ndash7913
Yassour M Kapian T Fraser HB Levin JZ Pfiffner J Adiconis X Schroth GLuo SJ Khrebtukova I Gnirke A et al 2009 Ab initio construction ofa eukaryotic transcriptome by massively parallel mRNA sequencingProc Natl Acad Sci U S A 1063264ndash3269
Zhu YO Sherlock G Petrov DA 2016 Whole genome analysis of 132clinical Saccharomyces cerevisiae strains reveals extensive ploidy var-iation G3 (Bethesda) doi101534g3116029397
Adaptation in Hybrids doi101093molbevmsx098 MBE
17