The Number, Speed, and Impact of Plastid Endosymbioses in ...PP64CH24-Keeling ARI 25 March 2013 11:0 The Number, Speed, and Impact of Plastid Endosymbioses in Eukaryotic Evolution

PP64CH24-Keeling ARI 25 March 2013 11:0

The Number, Speed,and Impact of PlastidEndosymbioses in EukaryoticEvolutionPatrick J. KeelingCanadian Institute for Advanced Research and Department of Botany, University of BritishColumbia, Vancouver, Canada V6T 1Z4; email: [email protected]

Annu. Rev. Plant Biol. 2013. 64:583–607

First published online as a Review in Advance onFebruary 28, 2013

The Annual Review of Plant Biology is online atplant.annualreviews.org

This article’s doi:10.1146/annurev-arplant-050312-120144

Copyright c© 2013 by Annual Reviews.All rights reserved

Keywords

photosynthesis, chloroplast, algae, organelle, phylogeny

Abstract

Plastids (chloroplasts) have long been recognized to have originated byendosymbiosis of a cyanobacterium, but their subsequent evolutionaryhistory has proved complex because they have also moved between eu-karyotes during additional rounds of secondary and tertiary endosym-bioses. Much of this history has been revealed by genomic analyses,but some debates remain unresolved, in particular those relating tosecondary red plastids of the chromalveolates, especially cryptomon-ads. Here, I examine several fundamental questions and assumptionsabout endosymbiosis and plastid evolution, including the number ofendosymbiotic events needed to explain plastid diversity, whether thegenetic contribution of the endosymbionts to the host genome goes farbeyond plastid-targeted genes, and whether organelle origins are bestviewed as a singular transition involving one symbiont or as a gradualtransition involving a long line of transient food/symbionts. I also dis-cuss a possible link between transporters and the evolution of proteintargeting in organelle integration.

583

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Plastid: generic termfor organelles derivedfrom endosymbiosis ofa cyanobacterium,including chloroplasts,leukoplasts,rhodoplasts,apicoplasts, etc.

Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . 584THE ENDOSYMBIOTIC

BUILDING BLOCKS OFPLASTID EVOLUTION . . . . . . . . . 584

MULTIPLE SECONDARYENDOSYMBIOSES—BUTHOW MANY? . . . . . . . . . . . . . . . . . . . . 587

THE HACROBIAN HYPOTHESIS . . 588THE CHROMALVEOLATE

HYPOTHESIS. . . . . . . . . . . . . . . . . . . . 588The Plastid Lineage and the

Chromalveolate Hypothesis . . . . . 590The Host Lineage and the

Chromalveolate Hypothesis . . . . . 591WHEN IS AN ENDOSYMBIOSIS

HYPOTHESIS PROVEN? THECOMMON ANCESTRY OF THEALVEOLATE-STRAMENOPILEPLASTIDS. . . . . . . . . . . . . . . . . . . . . . . . 591

TERTIARY AND SERIALSECONDARYENDOSYMBIOSIS . . . . . . . . . . . . . . . 593

THE PROCESS OFENDOSYMBIOSIS ANDPLASTID ORIGINS . . . . . . . . . . . . . . 594How Much Does an Endosymbiont

Contribute to Its Host? . . . . . . . . . 594Which Comes First, Gene Transfer

or Cellular Fixation? . . . . . . . . . . . . 596CONCLUDING REMARKS: WHAT

ARE THE NEXT OBVIOUSQUESTIONS? . . . . . . . . . . . . . . . . . . . . 600

INTRODUCTION

It is a common misperception that evolution isgoing somewhere in particular, when insteadmuch about the history of life has been com-plicated and unpredictable. The evolutionaryhistory of plastids—particularly what we havelearned in the past 10 years through advancesas diverse as improved high-throughputsequencing and a growing reappreciation ofspecies discovery and natural history—is a good

example: Major transitions in evolution doinclude adaptations of remarkable importance,but they also include reversals, redundancyand waste, improbable complexities, and oftenmore chance than necessity.

Many of the complexities of plastid evolu-tion stem from the fact that it is also a story ofsymbiosis. That plastids arose through the en-dosymbiotic uptake of a cyanobacterium is nowpast any serious debate. This initial endosym-biosis was one of the more important eventsin the evolution of life; by itself, however, it ex-plains only a fraction of plastid diversity becauseplastids then spread endosymbiotically to othereukaryotes. Here, I briefly review the main en-dosymbiotic events that led to the current di-versity of plastids, including some of the morecontentious outstanding questions, but also fo-cus on the actual process of endosymbiosis andthe implications of different models of plastidorigin in plant and algal genomes.

THE ENDOSYMBIOTICBUILDING BLOCKS OFPLASTID EVOLUTION

The single most confounding factor in plastidevolution is the simple fact that, although plas-tids may all be closely related, the algae thatcontain them are not (Figure 1). This is be-cause plastid evolution has included multiplelayers of endosymbiotic events.

Primary endosymbiosis, the ultimate sourceof photosynthesis in eukaryotes, refers to theuptake and retention of a cyanobacterium by aeukaryote. The plastids of land plants and greenalgae, red algae, and glaucocytophyte algae allderive from primary endosymbiosis (Figure 2).Much debate has led to a consensus that pri-mary plastids originated from a common an-cestor. This issue is not covered here because ithas been discussed in detail elsewhere (62, 84)(see also sidebar, Paulinella).

Primary endosymbiosis led to a plastidbounded by two membranes. It is difficult totrace the history of membranes through thisprocess, but based on the presence of certainlipids and membrane proteins (in particular

584 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Heterolobosea

Fungi

Land plants

Acantharia

Polycystines

Foraminifera

Ascetosporids

PhytomyxidsChlorarachniophytes

Euglyphids

Protaspids

Malawimonads

Jakobids

Diplomonads

Diphylleids

Retortamonads

Parabasalids

Oxymonads

Euglenids

Diplonemids

Kinetoplastids

Chlorophytes

Charophytes

Thaumatomonads

Phaeodarea

Microsporidia

Alv

eo

late

s

Animals

Choanoflagellates

Nucleariids

Ichthyosporea

Mycetozoa

Archamoebae

Ce

rco

zoa

Mesostigma

Glaucophytes

Cyanidiophytes

Floridiophytes

Bangiophytes

Apicomplexans

Bicosoecids

Haptophytes

Cryptomonads

Diatoms

Syndinians

Colponemids

Ciliates

Perkinsids

Oxyrrhis

Dinoflagellates

Actinophryids

Bolidophytes

Opalinids

Labyrinthulids

Thraustochytrids

Oomycetes

Raphidophytes

Chrysophytes

Phaeophytes

Capsaspora

Tubulinids

AcanthopodinidsVannellinids

Eustigmatophytes

Am

oe

bo

zoa

Op

istho

ko

nts

Stra

me

no

pile

s

Apusomonads

Breviates

Ancyromonads

Corallochytrids

Biliphytes

Katablepharids

Telonemids

Centrohelids

CercomonadsVitrella

ChromeraColpodellids

Prokaryotes(root)

Arc

ha

ep

last

ids

Rh

iza

ria

Un

iko

ntsD

isco

ba

Gre

en

alg

ae

Ex

cav

ate

s

Re

tari

a

“Ha

cro

bia

ns”

Re

da

lga

e

SAR

Figure 1Tree of eukaryotes with plastid lineages. This hypothesis for relationships between major eukaryotic lineagesis a synthesis of many studies based on large phylogenomic analyses or rare genomic characteristics. Dottedlines denote particularly contentious relationships, and in some cases two conflicting possibilities are shown.Colored boxes denote supergroups, including the more disputed hacrobians, which are shown because thereis not yet a viable alternative. Chromalveolates would be equivalent to the Hacrobia and SAR(stramenopile-alveolate-rhizarian) groups combined. The membrane structure and origin of plastids (whereknown) are shown at the periphery. The primary plastids with two membranes are found in thearchaeplastids (green, red, and cyan), and a second independent primary plastid is found in the rhizarianeuglyphid Paulinella (see sidebar, Paulinella). Secondary plastids are found in SAR organisms, cryptomonadsand haptophytes (both hacrobian), and euglenid excavates. The secondary plastids show the membranenumber and whether nucleomorphs are present, and the color indicates whether they are derived from greenor red algae. Plastids in apicomplexans and perkinsids are derived from a red alga, but are shown in white toindicate they are now all nonphotosynthetic.

www.annualreviews.org • Endosymbiosis and Plastid Evolution 585

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


PAULINELLA

The long debate over whether plastids arose from a single pri-mary endosymbiosis has focused almost exclusively on the plastidsof glaucophytes, red algae, and green algae, but one strange andfascinating amoeba has much to say on this issue. Paulinella chro-matophora was characterized more than a century ago (85, 100).It is a euglyphid testate amoeba (a shelled rhizarian amoeba) butis different from its kin in possessing two bean-shaped photosyn-thetic compartments called chromatophores. Chromatophoresare vertically inherited and tied to host division (75, 76). Thesewere recognized as functionally like plastids but more closelyresemble the cyanobacterium Synechococcus (75, 76), which wasborne out by molecular analyses (90). The key question ofwhether the chromatophore was genetically integrated with itshost (e.g., are there nucleus-encoded chromatophore-targetedproteins?) was suggested by the absence of certain genes fromthe chromatophore genome (113), and recent studies suggest thatprotein targeting may well occur (106, 112, 114). If this is con-clusively demonstrated, then there is no argument to not refer tothis as a second independent origin of plastids.

Primaryendosymbiosis:plastid origin in whicha eukaryote takes up aphotosyntheticcyanobacterium,resulting in a primaryplastid

Transit peptide:N-terminal proteinextension that isrecognized by theTIC/TOC complexesfor posttranslationalprotein import intoplastids

Secondaryendosymbiosis:plastid origin in whicha eukaryote takes upanother eukaryotealready containing aprimary plastid,resulting in asecondary plastid

beta barrel proteins, such as Toc75) (69, 134),it is most likely that the two membranesof primary plastids correspond to the twomembranes of the gram-negative cyanobac-terial cell (21). Assuming the endosymbiontwas taken up by phagocytosis, this in turnsuggests that the phagosomal membrane thatwould have originally surrounded the newendosymbiont was lost, which was likely anextremely important event in the transitionfrom “food” to “organelle.” This transitionis often erroneously represented in textbookdiagrams of endosymbiosis, which show thecyanobacterium with a single membrane andthe outer membrane of the plastid derivingfrom the phagosome (e.g., 39).

The dominant mode of evolution in thesymbiont in the early stages of the integra-tion was reduction, and many genes that wereunnecessary in the new intracellular environ-ment were simply lost. However, many othergenes were transferred from the symbiont tothe host nucleus (25). Once expressed, the prod-

ucts of some of these genes were targetedback to the integrating organelle in a processnearly always mediated by an N-terminal ex-tension called a transit peptide, which is rec-ognized by protein complexes in the outer andinner membranes of the plastid called TIC andTOC (for translocon inner-membrane chloro-plast and translocon outer-membrane chloro-plast, respectively) (136). Based on the observedpresence of Chlamydia-like genes in plant andalgal lineages, it has also been suggested thatthe ancestor of the host harbored another bac-terial symbiont related to Chlamydiales (63,105, 142), but there are many interpretationsof these data.

Up to this point, the origin of plastidsproceeded much as we imagine the origin ofmitochondria did, with one important excep-tion: Whereas mitochondria originated veryearly in eukaryotic evolution and the resultingorganelle is still found in all known eukaryoticlineages (132, 148), plastids originated some-what later. This has one important implication,namely that some eukaryotes had plastids andothers did not, which is significant becauseit creates conditions favoring the spread ofplastids between lineages, a process called sec-ondary endosymbiosis. In this case, a primaryalga is itself swallowed by another eukaryoteand then reduced and integrated within its newhost in ways superficially similar to primaryendosymbiosis (74, 97). There is, however, onedifference that proves critical for how the inte-gration proceeds, which is that the phagosomalmembrane surrounding the secondary plastidis not lost, leading to plastids with four mem-branes: the two original plastid membranes,the plasma membrane of the endosymbioticalga, and the phagosomal membrane of thesecondary host (96). In two lineages, the di-noflagellates and the euglenids, one membranehas been lost (most likely the third from the in-side, corresponding to the plasma membrane ofthe algal endosymbiont, because targeting mostobviously requires the other three membranes),resulting in three membranes (23, 107, 137).

As with the primary endosymbiont,many genes were lost from the secondary

586 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Bipartite targetingpeptide: leader onproteins targeted tosecondary plastids,consisting of a signalpeptide followed by atransit peptide

Signal peptide:N-terminal proteinextension that isrecognized by thesignal recognitionparticle andcotranslationallyinserted into theendomembrane system

endosymbiont and many others were trans-ferred to the host nucleus. Because most genesfor plastid-related proteins had already beenrelocated to the primary algal nucleus, in sec-ondary algae genes for plastid-targeted proteinshave generally moved and their products havebeen retargeted twice (3, 4, 27, 50). However,the retention of the phagosomal membranearound secondary plastids puts secondary plas-tids on a completely different footing with theirhost: Primary plastids are in the cytoplasm,but secondary plastids are topologically withinthe lumen of the endomembrane system.Accordingly, nucleus-encoded plastid-targetedproteins in secondary algae have a bipartitetargeting system. Plastid-targeting leaders arecomposed of a signal peptide (which targetsthe proteins to the endomembrane system) fol-lowed by a transit peptide–like sequence (whichtargets them to the plastid) (2, 82, 137, 146).

MULTIPLE SECONDARYENDOSYMBIOSES—BUTHOW MANY?

Secondary endosymbiosis is known to havetaken place more than once, but the exactnumber of secondary endosymbiotic eventsremains contentious. We know that it oc-curred multiple times because both green andred algae have been taken up in secondaryendosymbiotic events (Figure 2) (74, 81, 97).From an evolutionary perspective, this is fortu-itous because it allows us to compare the effectsof multiple independent secondary endosym-biotic events to differentiate general principlesfrom isolated events. In the case of green algalsymbionts, there is now broad consensus fromall available evidence that the two lineageswith green secondary plastids—the euglenidsand the chlorarachniophytes—acquired theseplastids independently. Indeed, phylogenetictrees based on whole plastid genomes also showthat they are not specifically related within thegreen algae (126), and the hosts are also phy-logenetically distant: Euglenids are related totrypanosomes in Excavata, whereas chlorarach-niophytes are cercozoan amoebae in Rhizaria

(Figure 1) (9, 73, 80, 87). The membranestructures of these green plastids are also differ-ent: Euglenid plastids have three membranes,whereas chlorarachniophyte plastids have fourand also retain a relict green algal nucleuscalled a nucleomorph (43), which is absent ineuglenids.

The evolution of red algal secondary plastidsis more contentious, mostly because the hostphylogeny is still unsettled but also becausethere are more lineages and likely more ancientevents to consider. Red secondary plastids arefound in cryptomonads (which also contain ared nucleomorph), haptophytes, stramenopiles,and dinoflagellates, all of which are also charac-terized by the unique presence of chlorophyllc (22, 71). Cryptomonad, haptophyte, andstramenopile plastids also share a commonstructure with four membranes, the outermostof which is connected to the endomembraneand the nuclear envelope (23, 24), but dinoflag-ellate plastids are bounded by three membranes(107). The last group with a red algal plastidis the apicomplexans, a lineage of obligateintracellular parasites including importantpathogens (e.g., Plasmodium and Toxoplasma).The discovery of the four-membrane apicom-plexan plastid nearly 20 years ago (83, 98, 150,151) and the ensuing debate over whether itderived from a green alga or a red alga (150)catalyzed advances in the field by acceleratingcomparative plastid genomics. This debate wasultimately resolved by the discovery of newlineages of still-photosynthetic apicomplexanrelatives (68, 103) (see below), underscoring theimportance of continuing field-based speciesdiscovery.

The debate over even this most basic featureof apicomplexan plastid evolution is a relativelysimple problem compared with the larger issueof how many endosymbiotic events are neededto explain the whole diversity of red secondaryplastids. In its current form, this debate centerson two independent but potentially compatiblehypotheses: the hacrobian hypothesis and thechromalveolate hypothesis. Although both areextremely contentious, they serve a useful pur-pose as null hypotheses.


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


THE HACROBIAN HYPOTHESIS

The core point of the hacrobian hypothesis(116) is that two major algal lineages, the cryp-tomonads and the haptophytes, share a com-mon ancestor, as do their plastids. This ideastemmed from the shared unique presence of ahorizontally transferred rpl36 in the plastids ofcryptomonads and haptophytes (125) togetherwith phylogenomic analysis that united theselineages in both nuclear and plastid trees (49,68, 119). These data appeared to offer a well-supported and consistent view, but this concepthas since become more complex.

First, a number of heterotrophic lineageswere added to this hypothetical grouping basedon nuclear phylogenomic analyses (16, 18): thekatablepharids, telonemids, centrohelids, andbiliphytes [the biliphytes, or picobiliphytes,were originally described as a photosynthetictaxon (111), but single-cell genomic data nowsuggest that they lack plastids (155)]. Thestrength of these associations is variable, withonly the katablepharids being unquestionablyrelated to cryptomonads (18, 77, 116). Theabsence of plastids in at least some of theselineages (18, 155) complicates the simple viewthat the hacrobian plastids are monophyletic.Second, recent phylogenomic analyses ofnuclear genes have called the monophyly of thehost lineages into question. Specifically, and incontrast to initial analyses that supported thehacrobians as a group (49, 119), recent analysessplit them into two subgroups: the cryptomon-ads (together with the katablepharids and

perhaps the biliphytes) and the haptophytes(perhaps together with telonemids and cen-trohelids), with the cryptomonads branchingwith archaeplastids without significant statis-tical support (18). This conflicts with plastidphylogenies, which put cryptomonads andhaptophytes together with strong support (68).

Altogether, the hacrobian hypothesisremains unresolved, and—getting back tothe core issue of the common ancestry ofcryptomonads and haptophytes and theirplastids—the implications of potentially differ-ent signals from plastid and nuclear sequencesrequire further investigation. All of this focusesattention on the position of the cryptomonadsas one of the greater mysteries in the tree ofeukaryotes (18).

THE CHROMALVEOLATEHYPOTHESIS

The chromalveolate hypothesis refers to theidea that all red secondary plastids can betraced back to a single endosymbiotic event(23). In practice, this has been tested mostlyby examining whether all lineages with redalgal secondary plastids (and all lineagesdemonstrated to be closely related to any ofthem) are closely related in both plastid andnuclear gene trees. Although this is a good firstapproach, a better test is to examine whethertheir plastid-targeting systems are homolo-gous, as was done to test the monophyly ofmitochondria and hydrogenosomes (122).

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 2Schematic depicting major endosymbiotic events in plastid evolution. At the top, primary plastids originated through theendosymbiotic integration of a cyanobacterium, resulting in two-membrane plastids in glaucophytes (which retain the peptidoglycanwall between the two plastid membranes), red algae, green algae, and their close relatives, the land plants. Green algae were integratedin two independent secondary endosymbioses, giving rise to plastid-bearing euglenids and chlorarachniophytes (middle right). Theorigin of red algal secondary plastids (middle left) is more contentious, in particular regarding whether the cryptomonad and haptophyteplastids (dashed red lines) originated from the same endosymbiosis as those of stramenopiles and alveolates (the latter consisting ofciliates, apicomplexans, and dinoflagellates). An additional round of endosymbiosis occurred in dinoflagellates (bottom). Serial secondaryendosymbiosis involved the integration of another primary plastid (a green alga), resulting in Lepidodinium. Tertiary endosymbiosisinvolved the integration of another secondary alga. The greatest degree of integration involves a haptophyte, but other putative tertiaryplastids were acquired from cryptomonads and diatoms. Independent of this entire history, another primary endosymbiotic origin ofplastids in the euglyphid testate amoeba Paulinella appears to be under way involving a cyanobacterium related to Synechococcus (bottomright).

588 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


mary endosymbiosisendoPrimary endosymoPrimar endos mbiosis

mary endosymbiosisendosPrimary endosymosPrimar endos mbiosis

ary endosymbiosisy endSecondary endosymbiosis

ary endosymbiosisendoSecondarycondary endosySSecondary endosymbiosisary endosymbiosisy endSecondary endosymbiosis

Serial secondary al seriaSerial seciiendosymbiosisendosyend s mbiosos bi ii i

en alga)reeg ga

ary endosymbiosisy enaryTertiaaryertiary endosymbiosistomd ti

StramenopilesStramenopenoSt il Ciliates

Dinophysis Lepidodinium

Euglenids Chlorarachniophytes

Paulinella

DinoDinoflag ttDiApicomplexans

Green algaeen alGreen algae

DurinskiaKarloddKarlodinium

d al aeRedd

Glaucophytes

ary endosymbiosisy enaryTertiaaryertiary endosymbiosis)yptocryc tcry omonadt

ndosymbiosisy enTertiiaryertiary endosymbiosisii i iitophyte)hap( tophytehhhh h t

hytespptHaptophytopH t h t

pptomCryptomonadtomtC t d

?

Land plants

Primary endosymbiosis

Primary endosymbiosis

Secondary endosymbiosis

Secondary endosymbiosisSecondary endosymbiosis

Serial secondary endosymbiosis

(green alga)

Tertiary endosymbiosis(diatom)

Stramenopiles Ciliates

Dinophysis Lepidodinium

Euglenids Chlorarachniophytes

Paulinella

DinoflagellatesApicomplexans

Green algae

DurinskiaKarlodinium

Red algae

Glaucophytes

Tertiary endosymbiosis(cryptomonad)

Tertiary endosymbiosis(haptophyte)

Haptophytes

Cryptomonads

?

Land plants


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


The chromalveolate hypothesis has provedto be extraordinarily provocative, but in polar-izing the field it has also provoked a good dealof constructive work on the evolution of algaldiversity. Many in the field have never acceptedthe monophyly of these plastids; however,it is useful to discuss both from a historicalperspective and because it has emerged as thenull hypothesis (like it or not, this group hasappeared in most “tree of eukaryotes” diagramsin the past decade, with Figure 1 beingan exception). Moreover, even with muchskepticism, there is no single well-supportedalternative, and most criticism stems from datafailing to support the hypothesis rather thanactively supporting an alternative.

The Plastid Lineage and theChromalveolate Hypothesis

The chromalveolate group was initially pro-posed to reduce the number of secondaryendosymbiotic events as much as possible (23).The plastids also uniquely share the presence ofchlorophyll c, and cryptomonad, haptophyte,and stramenopile plastids share a strikinglysimilar membrane topology in relation to theendomembrane and nuclear envelope (22, 24).From molecular data, trees based on wholeplastid genomes have generally united cryp-tomonads, haptophytes, and stramenopiles,especially when analyzing genes related to pho-tosynthesis, which tend to be more conserved(51, 68, 126). But it has been nearly impossibleto evaluate how the alveolates (dinoflagellatesand apicomplexans) fit into this tree becauseof the strange nature of their plastid genomes.Apicomplexans, as intracellular parasites,unsurprisingly have highly reduced plastidgenomes that lack photosynthesis-relatedgenes (78, 151). Dinoflagellates have evenstranger genomes, having moved most oftheir genes to the nucleus and convertedthe rest of their genome to small minicircles(47, 156). The upshot is that apicomplexansand dinoflagellates have almost no genes incommon, and neither group’s plastids arereadily comparable to those of other groups.

This dead end was initially sidestepped byanalyzing nucleus-encoded genes for plastid-targeted proteins, which served as stand-ins forthe plastids themselves. Two enzymes providedthe first molecular support for the inclusion ofalveolate plastids: glyceraldehyde-3-phosphatedehydrogenase (GAPDH) and fructose-bisphosphate aldolase (FBA) (35, 54, 120). Thephylogenetic null hypothesis for chromalve-olate plastid-targeted proteins would be thatthey are related to red algal plastid-targeted ho-mologs that themselves derived from cyanobac-teria. However, chromalveolate GAPDH andFBA show different phylogenetic patterns.Plastid-targeted GAPDH derived from a geneduplication of an ancestral cytosolic copy(35, 54), and plastid FBA in chromalveolatesderived from a class of FBA nonhomologousto that found in plastids of green and redalgae (120). The history of GAPDH and FBAwithin the chromalveolate groups is morecomplex than originally thought (1, 139),however, so how these characteristics relate tothe evolution of the organisms is no longer soclear.

A somewhat more satisfying approach be-came possible in recent years owing to the dis-covery of new algal lineages that have providedcritical new plastid data. A survey of dinoflagel-late symbionts of coral isolated two new photo-synthetic species, and when their phylogeneticposition was determined, they were found tobe deep-branching relatives of apicomplexans(103, 115). The plastid genomes of these organ-isms, Chromera and Vitrella, were completelysequenced and found to represent a much lessreduced and derived state than those of ei-ther apicomplexans or dinoflagellates (68). In-deed, these plastids bring together features ofboth groups (including some nucleus-encodedcharacteristics, like the proteobacteria-derivedRuBisCO formerly restricted to dinoflagel-lates) and have fewer oddities than any otherknown alveolate plastid (68). Accordingly, thesegenomes allowed the first direct test of the com-mon ancestry of alveolate plastids and those ofother chromalveolates, and analyses supportedthe union of alveolate and stramenopile plastids

590 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


but only weakly supported the chromalveolatesas a whole (68).

It is also noteworthy that the analysis ofChromera and Vitrella effectively ended oneof the most long-standing and confoundingdebates surrounding the chromalveolates: thered versus green origin of the apicomplexanplastid. Unsurprisingly, the discovery thatparasitic apicomplexans have a plastid initiateda debate on what it was doing there. What wassurprising was how easy it was to explain itsfunctions (38, 123) but how difficult it was toexplain its origins. Although it was generallyaccepted that the apicoplast originated bysecondary endosymbiosis (99, 150), there wereconflicting hypotheses as to whether this was ared alga, and possibly related to dinoflagellateplastids (42, 149), or a green alga, arising froman independent endosymbiosis (41, 78). Asnoted above, apicomplexan and dinoflagellateplastid genomes are effectively impossible todirectly compare, so this deadlock was onlyreally resolved by the discovery of Chromeraand Vitrella. Both phylogenetic and characteranalyses of these genomes together with thoseof either apicomplexans or dinoflagellates indi-vidually have shown beyond any serious doubtthat the apicoplast is indeed derived from thesame red alga as the dinoflagellate plastid (68).

The Host Lineage and theChromalveolate Hypothesis

Testing the chromalveolate hypothesis usingdata from nuclear genomes has proved evenmore difficult. Based on early single-gene trees(144), concatenations of small numbers ofgenes (55), and the distribution of the Rab1aGTPase (31), the only consistent feature wasthe monophyly of the alveolates and stra-menopiles. With the advent of large phyloge-nomic analyses, hundreds of nuclear genes werebrought to bear on the question, and one ma-jor advance quickly emerged: the inclusion ofRhizaria. Rhizaria is a large group of primar-ily amoebae and amoeboflagellates of stunningdiversity—so diverse, in fact, that the super-group was established solely on the grounds

of molecular phylogenetic data (19, 108) be-cause no unique and common morphologi-cal characteristic has been described. The firstphylogenomic analyses including rhizarian rep-resentatives showed a strong relationship tothe alveolates and stramenopiles (19, 20, 49),and this relationship has remained consistentin subsequent studies that include many moredata from many additional rhizarian taxa (14,16–18). Some variations in exactly how thesethree groups are related have been observed,but the most strongly supported analyses indi-cate that alveolates and stramenopiles are sis-ters, with rhizarians branching at their base.This group has not been formally named butis widely known as the SAR (stramenopile-alveolate-rhizarian) clade.

The chromalveolate hypothesis is not sub-stantially altered by the inclusion of rhizarians,because it simply requires one more loss of anancestral red algal plastid (see below); however,phylogenomic analyses of nuclear data fromhacrobian taxa have challenged the hypothesis.As noted above, the first phylogenomic analy-ses of nuclear data initially strongly supported aunion of cryptomonads and haptophytes but didnot resolve their relationship with other chro-malveolates. Since then, the monophyly of thehacrobians has also been challenged by nucleargene data (see above), but, more critically, so hastheir relationship to chromalveolates: In someanalyses, haptophytes go with SAR but cryp-tomonads do not (18), whereas in other analysesneither goes with SAR (7). This debate remainsunresolved, because alternative positions havealso failed to gain any support, but it can besaid that large-scale phylogenomic data fromthe nuclear lineage fail to support the chromal-veolate hypothesis.

WHEN IS AN ENDOSYMBIOSISHYPOTHESIS PROVEN? THECOMMON ANCESTRY OF THEALVEOLATE-STRAMENOPILEPLASTIDS

The alveolate-stramenopile group that is sup-ported by all analyses of plastid and cytosolic


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


data (see above) captures most of the diversityand controversy of the chromalveolate hy-pothesis. For example, if one accepts that thedemonstrably related secondary red plastids ofthese demonstrably related host lineages sharea common endosymbiotic ancestry, then theancestors of major nonphotosynthetic lineageslike ciliates (alveolates) and oomycetes (stra-menopiles) must have had a plastid. Indeed,many of the surprising implications of thechromalveolate hypothesis are already manifestin this reduced, “chromalveolate-light” versionof the hypothesis, but this relationship andits implications are not widely discussed andthe implications are not entirely accepted.This leads to the question, what would it taketo prove a common ancestral endosymbioticorigin of such plastids?

Given the consistency of plastid and hostphylogenies, there are few specific reasons toquestion the common ancestry of alveolate-stramenopile plastids. But one issue that doesrequire discussion is the abundance of plastid-lacking lineages within alveolates and stra-menopiles and the fact that they appear to beclustered basal to plastid-containing lineages(e.g., ciliates at the “base” of alveolates andoomycetes at the “base” of stramenopiles). Thiscan be interpreted as evidence for two indepen-dent plastid origins late in the evolution of eachlineage (10, 34). However, any interpretationhinges on a complete understanding of two is-sues: the distribution itself and the likelihood ofevents that lead to the distribution.

At first glance, the distribution seemsobvious—organisms either have plastids or donot. But this is not so simple, because crypticplastids have been found in a variety of lin-eages [e.g., perkinsids and apicomplexans (37,98, 140, 151)]. The best-studied aplastidal alve-olates and stramenopiles [e.g., the best-studiedciliates and oomycetes (30, 142)] really do lackplastids, but most others have not been investi-gated in enough detail to be sure. At the sametime, when interpreting the distribution it isimportant to think about the number of events,and not the number of organisms with or with-out plastids. Many ciliates may lack plastids, but

this could represent a single event of plastid loss,and as such is no more unlikely than a plastidloss from a single species.

Examining the distribution of plastid lossevents (or putative plastid loss, given that wemostly cannot distinguish loss from severe re-duction) on the alveolate-stramenopile tree alsoreveals that the notion that they are basal isquestionable. In the stramenopiles, plastid lossevents within plastid-containing lineages areknown (as is the photosynthesis loss event), andthe number of these events is perhaps equal tothe number needed to explain the basal lineages(8, 23, 45, 101, 131). In alveolates the situa-tion is even more unbalanced: Ciliates are notreally basal to apicomplexans and dinoflagel-lates so much as sister to them (Figure 1), andplastid loss has happened several times in thelineages leading to apicomplexans (e.g., Cryp-tosporidium and Colpodella) and many more timesin the dinoflagellates (128). With alveolates andstramenopiles related in both plastid and hosttrees, why the one additional loss event to ac-count for ciliates should be given any specialweight is unclear.

Interpreting the distribution of events alsorequires weighting the likelihood of thoseevents, but we suffer from a near-complete lackof understanding about the process of plastidloss. Although there has been a great deal ofspeculation about how plastids originate, littleconcrete has been said about why or how theyare lost or what traces might be left behind.Moreover, there are few cases where we can becertain of the two prerequisites for concludingthat plastid loss occurred: that an organism’sancestor had a plastid and that that organismnow does not—one or both of which are oftenmurky questions.

Lastly, if one accepts the phylogenetic rela-tionship between both host and plastid com-ponents of the alveolates and stramenopiles,then what is the alternative to concluding thattheir common ancestor had the same plastid?The only alternative model that is consistentwith the data at face value is that the plastidoriginated in one lineage and was subsequentlytransferred to the other by a process called

592 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


tertiary endosymbiosis, which is discussedbelow.

TERTIARY AND SERIALSECONDARY ENDOSYMBIOSIS

Tertiary and serial secondary endosymbiosesadd yet more layers of associations after sec-ondary endosymbiosis. Both are known only indinoflagellates, which ancestrally had a red sec-ondary plastid, but a few lineages seemingly lostor reduced this plastid and either took up an-other secondary alga to form a tertiary plastidor took up another primary alga to form a serialsecondary plastid.

Serial secondary endosymbiosis is gen-erally thought to have occurred only once,represented by the dinoflagellate genus Lepido-dinium, where the ancestral red algal secondaryplastid has been replaced by a green algalone (147). The Lepidodinium plastid genomeconfirms its ancestry from a chlorophyte, andanalysis of its nucleus-encoded plastid-targetedgenes shows that it acquired genes from avariety of sources (94, 102).

Truly unambiguous tertiary endosymbiosisis only really known in one lineage of dinoflag-ellates, represented by the genera Karenia andKarlodinium, which have lost their ancestralplastid and acquired a new one from a hap-tophyte (141). The haptophyte is completelyreduced so that all that remains is the plastid,but unfortunately the structure of the plastid ispoorly known, so the number and nature of themembranes are uncertain. Neither is the natureof protein targeting known, but, intriguingly,the targeting peptides include a signal peptide(indicating the plastid is in the endomembranesystem) followed by a sequence with little ornothing in common with any known transitpeptide (121). It appears that this system uses atargeting mechanism with at least some uniqueelements, and it would be a worthwhile systemto use in studying possible alternative solutionsto the challenges of protein targeting.

Two other dinoflagellate lineages have un-dergone similar processes, but neither is quiteas straightforward. Dinophysis has what may bea tertiary plastid derived from a cryptomonad

(130), but there remains some debate as towhether it is totally integrated or a transientresidue of feeding called a kleptoplast (48,138). A different situation is again found inanother lineage of dinoflagellates, representedby Kryptoperidinium and Durinskia, which haveacquired a plastid from diatoms (26, 29). Thereis no question that these are permanent andcompletely integrated into the dinoflagellatehost cell cycle, but in this case the diatomretained its own nucleus and even intactmitochondria (29, 65). It is unclear whetherthere is any integration at the level of genetransfer and protein targeting; if so, they defyhow organelles are most commonly defined.

Although tertiary endosymbiosis may beevolutionarily rare and restricted to a handfulof dinoflagellate lineages, it is nonethelessa potentially important process for our un-derstanding of organelle integration. This isthe case because tertiary endosymbionts offertwo practical advantages as model systems.First, because the events took place relativelyrecently, they might offer intermediate stagesin the integration process, and fewer clueswill have been erased by time, as is inevitablewith more ancient endosymbioses. Second,the phylogenetic identities of both hostsand endosymbionts are well known (26, 66,67, 141). This is important in allowing theunambiguous identification of endosymbiont-derived genes, and in testing hypotheses aboutthe endosymbiont’s genetic contribution andthe order of events that take place during theintegration process (see below).

A better understanding of known cases oftertiary endosymbiosis is also potentially im-portant to explain incongruences between plas-tid and nuclear gene phylogenies. We currentlyhave an unsatisfying situation where one getsthe sense that many people working in the fieldbelieve that certain secondary plastids mightreally be cryptic tertiary plastids, but there arestrikingly few cases where this has been specif-ically spelled out in the literature, and eventhose attempt to explain different plastids fordifferent reasons (5, 12, 154). These hypothesesgenerally stem not from actively contradictory


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


phylogenies but rather from phylogenies thatfail to resolve the position of the host lineage.Tertiary endosymbiosis has also never beenobserved outside of dinoflagellates, and eventhere it is rare, so extending it to other lineagesshould be based on observed similarities.However, what we do know about the natureof tertiary plastids is inconsistent with what weknow about secondary plastids: For example,tertiary plastids do not appear to possess plastidmembrane structures or plastid-targeting sys-tems like those of secondary plastids (29, 121).

So although cryptic tertiary endosymbiosisremains a formal possibility to be explored withactual data, it is not a quick fix. Known tertiaryplastids need to be studied in much greaterdetail, and direct evidence for contradictionsbetween plastid and host phylogenies that pointto a specific tertiary event needs to be tested.Most important, the evolutionary origins ofplastid-targeting systems should be explored inmuch more detail. Neither phylogenies of hostgenes nor those of plastid genes can formallydisprove the monophyly or polyphyly of chro-malveolates: Phylogenies that favor cryptictertiary endosymbioses can infinitely accountfor monophyletic plastids, whereas those thatfavor plastid loss can infinitely account forpolyphyletic (or paraphyletic) hosts. Targetingsystems might offer the clearest data on theorigins of diverse plastids.

THE PROCESS OFENDOSYMBIOSIS ANDPLASTID ORIGINS

It is easy to say plastids originated by “en-dosymbiosis,” but we seldom ask what thatreally means. There are many possible ways tointegrate two cells, and as is often the case withevolution, the order of events in a transitionis critical to the details of its outcome. Below,I question several common assumptions andtheir implications.

How Much Does an EndosymbiontContribute to Its Host?

It has long been clear that the genetic contribu-tion of an endosymbiont to its host goes beyond

just genes for proteins that are targeted back tothe plastid (e.g., 93), but exactly how significantthis additional contribution may be is a majorquestion. This issue becomes more complexin secondary (and tertiary) endosymbioses,because nonplastid genes ultimately derivedfrom the cyanobacterial symbiont may havebeen acquired in addition to genes from the en-dosymbiont (green or red algal) nuclear lineage.Understanding this process is key to both howwe understand the process of endosymbiosisand how we reconstruct its evolutionary his-tory. This is because the existence of significantnumbers of such endosymbiotic genes in thehost nuclear genomes have been not only in-ferred but also used as evidence for a number ofevolutionary scenarios that fall broadly into twocategories: that known endosymbiotic eventsmay be older than they seem, and that othercryptic endosymbiotic events also took place.These are reviewed in the next two sections.

How old are known secondary symbioses?It has been argued that a number of nonpho-tosynthetic lineages were once photosynthetic,based on the inferred presence of nucleus-encoded genes believed to be derived from anow-lost or cryptic symbiont (37, 95, 124, 133,142). Typically, the nonphotosynthetic lineageis closely related to an algal group, so the argu-ment is not for a previously unrecognized en-dosymbiosis but rather for a known endosym-biosis having taken place earlier than previouslythought. It is also important to note that thesegenes are typically not argued to function in anexisting plastid, or even to have once functionedin a plastid; instead, they are thought to besimple genes with some phylogenetic affinity tocyanobacteria, plastids, or the nuclear lineageof a secondary symbiont (i.e., “red” or “green”).

One of the first such suggestions wasthat the largely parasitic kinetoplastids (e.g.,Trypanosoma) had once contained the greenalgal–derived plastid still found in their sis-ter group, the euglenids (53, 92). This wasbased on the conclusion that genes in try-panosomes showed some kind of relationship to“green” genes in phylogenetic reconstructions.

594 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


However, closer inspection of most of thesegenes showed that they did not support thisconclusion (i.e., the trypanosome gene was notactually closely related to a green homolog)or were misleading owing to poor sampling(32, 127). In addition, this ancient origin ofthe green plastid was not supported by thebiology and distribution of plastids within theeuglenid lineage, which strongly suggested thatthe plastid was acquired relatively recently,probably owing to the evolution of eukary-otrophy occurring within the diversification ofeuglenids rather than before (86).

The discredited idea that trypanosomesonce contained a plastid should be kept in mindwhen interpreting such phylogenetic data, be-cause a number of similar cases have been madefor various chromalveolate subgroups. As notedabove, the chromalveolates include a numberof nonphotosynthetic lineages. Within theframework of the chromalveolate hypothesis(or the “chromalveolate-light” hypothesis),the ancestor of these lineages once containeda red algal plastid. In what appeared to be adramatic consequence of this symbiosis, thefirst genome of an oomycete was reported tocontain more than 700 symbiont-derived genes(142), a substantial proportion of the genome.The first analysis of a ciliate genome concludedthat no significant number of plastid geneswere to be found in Tetrahymena (30); how-ever, a reanalysis concluded that Tetrahymenaand Paramecium do in fact contain symbiont-derived genes, although only 16 were identified(124). Similarly, genomes of the apicomplexanCryptosporidium and the nonphotosyntheticdinoflagellate Crypthecodinium were found tocontain genes phylogenetically inferred to bederived from the red lineage (64, 129).

Two other cases—the deep-branching andnonphotosynthetic dinoflagellate relativesPerkinsus and Oxyrrhis—stand out because inboth genera, putatively plastid-derived geneswere found but were also suggested to be tar-geted to cryptic plastids. In the case of Oxyrrhis,this was based on the observation that somegenes that were phylogenetically related toplastid-targeted genes in other dinoflagellates

still retained the distinctive N-terminal plastid-targeting leaders (133). In Perkinsus, genes en-coding N-terminal leaders were also identified(37, 46, 95), but protein products of some geneshave also been localized to discrete membrane-bounded structures—altogether constitutingstrong evidence for the maintenance of arelict plastid, which in this case appears to beinvolved in isoprenoid biosynthesis (37, 46, 95).

The strength of evidence for these conclu-sions varies dramatically, and unfortunately, themost interesting are supported by the weak-est evidence. In lineages where their phyloge-netic position already makes a strong case foran ancestral plastid, the evidence for relict plas-tid genes is relatively strong. For example, inPerkinsus, where there is now even evidence forthe organelle, and in Oxyrrhis and Cryptheco-dinium, where the genes in question are ac-tually related to plastid-targeted homologs, aplastid-containing ancestor can justifiably beconcluded. In the major nonphotosyntheticoomycete and ciliate lineages, however, thegenes in question are generally not obviouslyrelated to plastid function. If these genes are tobe interpreted in the context of the chromalve-olate symbiosis, then there is a critical assump-tion with implications that are seldom recog-nized: specifically, that the red algal ancestordonated nucleus-encoded cytosolic proteins toits new host. The implication of this is that suchtransfers must have happened in the early stagesof the symbiosis, and the genes would have ac-quired a function that led to their retentionin the ancestor of all chromalveolates. Accord-ingly, genes that are truly red algal relicts of thechromalveolate symbiosis should be widespreadwithin extant chromalveolates, including thephotosynthetic ones. In other words, if the redgenes in these nonphotosynthetic lineages arecompared, many should be the same genes,they should be related to one another, and theyshould have homologs in other chromalveolategroups. Alternatively, a lack of significant over-lap in the red genes identified in different lin-eages would suggest that the current pattern isunrelated to the possible presence of an ances-tral endosymbiont. Remarkably, however, these


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


results have not been examined as a whole, sothe degree of overlap is unknown.

All this leads to the question, how manygenes does it take to distinguish between hori-zontal gene transfer and endosymbiosis? Mostof the phylogenies that have been suggestedto support an influx of genes from secondaryendosymbiont nuclei to the host nuclei (apartfrom those that encode proteins actually tar-geted to the plastid) appear to be based moreon phylogenetic noise than on signal, and wheredetailed analyses have been done, relatively fewputatively endosymbiont-derived genes remain(15, 28). Without more information on theoverall proportion of genes from these genomesthat appear to be phylogenetically incongruentwith the known position of the host, it is im-possible to say whether these genes represent alegitimate spike in phylogenetic signal indica-tive of some relict of the endosymbiotic event,or background noise from unresolved phyloge-nies and low-level horizontal gene transfer.

Were there other cryptic secondaryendosymbioses? There is a second issuethat has emerged from similar data, whichis the question of whether there were oldercryptic endosymbiotic events that took placeprior to the establishment of extant plastids inphotosynthetic lineages. Specifically, a numberof analyses have suggested that the red plastidin chromalveolates was preceded by a greenone, which has since been lost except for somerelict nuclear genes. This idea goes back toearly genomic surveys of plastid-containingchromalveolates where genes for a numberof plastid-targeted proteins were found to bephylogenetically green rather than red (40, 50,109, 145). At the same time, phylogenomicanalyses began to reveal the now-acceptedrelationship between rhizarians, alveolates, andstramenopiles (see above). This is significantbecause chlorarachniophytes are rhizarians,and they have a secondary plastid of greenalgal origin that contains many genes forplastid-targeted proteins derived from the redlineage (3). Ultimately, whole-genome analysesof diatoms concluded that a large contingent

of green genes were encoded in the diatomgenome (104), and similar results were foundfor a photosynthetic relative of apicomplexans,Chromera, where the phylogenetic signals forgreen and red genes were concluded to beapproximately equal, at 250 genes each (152).

Once again, however, if these green genesin chromalveolates were derived from anancient cryptic green plastid, we would expectthe various chromalveolate lineages to retainmany of the same green genes today, but thishas not been demonstrated. Most important,however, a careful reanalysis of the green geneshas also suggested that the contribution maybe dramatically overestimated by automaticphylogenetic analyses. In the case of thediatoms, reanalysis of the 1,700 reported greengenes showed that only a handful could beconfidently attributed to the green lineage(28). The remainder were found to be poorlyresolved, missing key taxa (such as red algae),or more probably a result of horizontal genetransfer. Similarly, in the case of Chromera, theoriginal study pointed out that the green signalis best interpreted as the manifestation of arti-facts due to many more data being available forgreen lineages than red (152); a reanalysis usinga slightly different automated search strategyand manual inspection of the trees went evenfurther and concluded that only 23 genes wereclearly red and 9 clearly green (15). Overall,the data supporting a cryptic green plastid inthe evolution of chromalveolates are rapidlydiminishing. Indeed, at this point, the dataare indistinguishable from a small number ofhorizontal gene transfer events, leading to theprediction that similar levels of foreign genesthat are phylogenetically affiliated with otherlineages will also be found in these genomes.

Which Comes First, Gene Transferor Cellular Fixation?

Although we all might agree that endosymbio-sis requires that a symbiont be ingested, be per-manently retained by the host, transfer genes tothe host, and evolve a protein-targeting system,digging into the specifics shows that the order

596 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


in which these events take place is not so obvi-ous. Though generally unstated, the textbookview of endosymbiosis also implies that plas-tids originate rather suddenly—a heterotrophingests an algal cell but fails to digest it. Suchautotrophy-by-indigestion might be shockingfor both parties, so it is reasonable to comparethe implications of the order of events to seewhat else might result in a fair description ofendosymbiosis. Below are two of several possi-bilities, in an attempt to show how changing thehypothesized order of events changes both thenature of the transformation and our expecta-tions about how the cells are affected.

Endosymbiosis as a rapid transition. In thismodel, a heterotrophic predator engulfs a pho-tosynthetic prey cell, but rather than being di-gested, the autotroph is retained and surviveswithin the heterotroph. This singular event setsoff a long-term transformation of both cells, butprimarily of the autotroph. One could imagineit adapting to its intracellular environment bydividing with its host cell, losing functions thatare now redundant or unnecessary (like motil-ity), and evolving transporters to exchange nu-trients and energy with the host. Over time,genes flowing from the endosymbiont to thehost (perhaps by lysis, if there are multiple sym-bionts within a host) find their way into the nu-cleus and are incorporated into nuclear chro-mosomes and expressed on cytosolic ribosomes.

The most complex step in such a schemewould be the establishment of a protein-targeting system. In primary endosymbionts,this corresponds to transit peptides and theTIC/TOC system; the origin of this system isobscure, but it appears that some parts comefrom the host and others from the endosym-biont (6, 70, 88). In secondary endosymbionts,the second half of the protein-targeting system(based on transit peptides and the TIC/TOCsystem) is already in place in the endosymbiont.The first step requires sorting to the endomem-brane mediated by a signal peptide, a systemalready in place in the host. The major hur-dle in secondary endosymbiosis would there-fore be to link these two targeting systems to-

gether, which requires sorting proteins that arealready in the endomembrane system specifi-cally to the plastid. It is now clear that in allsecondary plastids, this step is dependent on in-formation encoded in the transit peptide (2, 44,57, 59, 60, 79, 82) and some protein complexthat mediates passage across the third mem-brane. In chromalveolates, the protein complexis the symbiont-specific ERAD-like machin-ery (SELMA) system, which is derived fromthe endoplasmic reticulum–associated degra-dation (ERAD) complex (13, 36, 56, 135). Atleast in chlorarachniophytes, some other so-lution seems to have been found, because noplastid-specific SELMA-like system is encodedin the genome (58). It is also unclear how theloss of the third membrane in euglenids and di-noflagellates would have affected this step, andtheir transit peptides are notably different.

The existence of a system to specifically im-port proteins from the cytosol to the endosym-biont sets up an evolutionary ratchet becauseit allows an increasing number of proteins tobe targeted relatively easily. Even if the rateof transfer from symbiont to host remains un-changed, the rate at which host-encoded genesmake endosymbiont-encoded homologs func-tionally redundant would increase dramaticallysimply because there is a mechanism to targetthe products of transferred genes back to theorganelle. At the genomic level, the expectedoutcome of this order of events is that the hostnucleus contains a large number of symbiont-derived genes, all of which should share a com-mon phylogenetic signal tracing back to a singleendosymbiont.

Endosymbiosis as a slower transitionand the importance of transporters: thetargeting-ratchet model. Although there isnothing impossible about the above order ofevents, subtle changes that lead to differentexpectations are just as plausible, or arguablymore so. Consider, for example, the implica-tions of establishing protein targeting before theendosymbiont is permanently fixed in the host.In this model (Figure 3), a grazing heterotrophbegins to transiently retain photosynthetic


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Eating

Transient association

Transporters targeted

Gene transfer andmore targeting

Result: many genes derived frommany transient symbionts

Dig

estio

n

Dig

estio

n

DigestionFigure 3A targeting-ratchet model for the endosymbiotic origin of plastids as a long-term cyclic association withphotosynthetic food. This illustration shows a secondary endosymbiosis, but a similar model could bedeveloped for primary or tertiary endosymbiosis. In the cyclic portion (top), a heterotrophic predator (purple)eats and digests various algal prey (bottom left). Over time, the host predator begins to transiently retain preycells before digesting them (top left). For longer associations to benefit the host, it would need to acquireenergy from photosynthesis without digesting the prey. This could be achieved by specifically targetingtransporter proteins to membranes of the transient symbionts, allowing nonlethal energy and nutrientuptake (top right). The existence of this targeting creates a ratchet, allowing an increasing number of proteinsto be targeted to transient symbionts, which in turn allows proteins encoded by genes acquired fromsymbionts to be targeted to subsequent symbionts (bottom right), driving the integration process. Ultimately,the retention period lengthens to the point that one particular food alga (or many within a population) isnever digested and goes on to become the plastid organelle. However, the history of gene transfer andtargeting from transient symbionts means that “algal” genes in the host are not necessarily all derived fromthe same symbiont as the plastid (indicated by the many colored genes on the host chromosome), even ifthey encode plastid-targeted proteins.

598 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


prey before digesting them [as is known tooccur in nature (33, 52, 72, 89)]. Over time,the retention period could increase as the hostdevelops control over signaling degradation.But if the predator has no means of extractingthe energy and nutrients of its meal withoutdigesting it, then there is no advantage toincreasing the retention time. To reap the fullbenefits of prolonging such an association,it would need transporters that could extractthe energy and nutrients from the transientsymbiont without digesting it.

The importance of transporters in the ori-gin of plastids has been recognized (143) butis typically overshadowed by the importance ofprotein targeting (11, 23, 91). Both are impor-tant, but their evolutionary origins might alsobe tied to one another: If the predator can tar-get transporters to its prey, it would not onlyhave greater, nonlethal access to the energyproduced by photosynthesis, but would also setup a powerful evolutionary ratchet where theacquisition of additional genes from transientsymbionts could provide positive feedback if thetargeting of their products prolongs the usefulretention of and/or control over symbionts. Inother words, targeting would ratchet the sys-tem toward fixation, but fixation does not nec-essarily ratchet toward establishing targeting.This cycle could continue with transient sym-bionts remaining in the cytoplasm for increas-ing periods of time before being digested andwith the collection of host-encoded genes forsymbiont-targeted proteins growing in the nu-cleus, until the process is ultimately fixed whensome prey cell is never digested and becomesthe organelle.

This targeting-ratchet model shares severalsteps in common with Larkum’s “shoppingbag” model for plastid evolution (62, 84) andshares Tyra’s emphasis on the early importanceof transporters in plastid origins (143). But theevolutionary link between targeting and trans-porters (which can be applied to primary or sec-ondary plastids, or to mitochondria, for thatmatter) and the ratchet it creates early in theprocess provide a plausible first push toward ge-netic integration.

The order of events also has other implica-tions at the cellular and genomic levels. At thecellular level, it suggests that the plastids withinan algal lineage might derive from differentsymbionts. In practice, this would be difficultto recognize unless very different kinds of algaewere fixed in a related lineage of hosts, but thereis one intriguing case. The diatom-derived ter-tiary plastids are constrained to a small groupof related dinoflagellates, but molecular phy-logenies have shown that the plastids are de-rived from both pennate and centric diatoms.They might have been serial replacements (61)but might also have been fixed independentlyin different but closely related diatom-eatingdinoflagellate hosts.

At the genomic level, it is possible that thegenes for plastid-targeted proteins within a sin-gle cell need not all be derived from the samelineages as the organelle to which they are tar-geted [as pointed out in the shopping bag model(62, 84)]. If the host has a narrow prey prefer-ence, then the distinction might be impossibleto discern, but if the host is grazing on a widevariety of prey, then genes for plastid-targetedproteins would be expected to be derived froman equally wide variety of sources. Interest-ingly, most secondary algal lineages have nowbeen observed to contain a substantial numberof genes for plastid-targeted proteins that areclearly not derived from the same lineage as theplastid itself (3, 40, 50, 109, 110). Generally thisis attributed to a single cryptic endosymbiosis,or more realistically to horizontal gene transfer.Indeed, it is probably impossible to entirely dis-tinguish the process outlined above from hori-zontal gene transfer, but it is worth noting thatrandom horizontal gene transfer should con-tinue throughout the evolution of a given lin-eage, so we should see such genes shared by allmembers of a lineage in some cases (an ancientorigin) but also narrowly distributed in othercases (a more recent origin). Either way, com-ponents of key plastid pathways come from dif-ferent lineages, but whether these were over-laid on ancestrally homogeneous pathways byhorizontal gene transfer or represent a gradualbuilding up of the plastid proteome from a slow


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


and cyclic integration of different prey cells isan interesting question to ponder. At the sametime, it is worth remembering that each pri-mary, secondary, and tertiary plastid origin isa unique event, so different models might bestexplain different plastid endosymbioses.

CONCLUDING REMARKS: WHATARE THE NEXT OBVIOUSQUESTIONS?

Investigation of the complex evolutionaryhistory of plastids has benefited hugely fromsuccessive technical revolutions that have madesequence analysis of diverse lineages at the ge-nomic level possible. Indeed, we may even beapproaching the point where we have substan-tially characterized the data available to us from

representatives of all major algal lineages. How-ever, if recent years have revealed any blindspots in this field, it is our ignorance of thenatural diversity of these lineages, and partic-ularly the oddballs that branch close to them.The discovery of photosynthetic relatives ofapicomplexans (103), for example, has allowedformerly impossible leaps forward in our un-derstanding of this chapter in plastid evolu-tion. Equally illuminating new nonphotosyn-thetic relatives of algal lineages have also beenidentified (111, 116–118, 153, 155). Althoughless appreciated, each of these discoveries hastremendous potential to upset our establishedviews of plastid evolution. At this point, thegreatest potential for transformative change isfrom a renaissance of natural history and speciesdiscovery.

SUMMARY POINTS

1. Plastids, or chloroplasts, originated by the endosymbiotic uptake of a cyanobacteriumleading to primary plastids (like those in plants and green and red algae) but then spreadto other eukaryotic lineages through progressive rounds of secondary and even tertiaryendosymbiosis.

2. The evolutionary history of these subsequent endosymbiotic events is becoming clearer,but there is still controversy over the number of times secondary red algal plastidsoriginated.

3. During endosymbiosis, many genes are transferred from the symbiont to its new host,but the extent of this contribution (beyond genes for plastid proteins) is unclear. Recentclaims for substantial numbers of “algal” genes in host nuclei now appear to be the resultof sampling biases and difficulties with automated phylogenetic analyses.

4. In one model of plastid origins, protein targeting evolves early in the process to targettransporters to transient photosynthetic symbionts; once established, it creates a genetransfer ratchet that leads to progressive integration and finally fixation of the organellein the cell.

DISCLOSURE STATEMENT

The author is not aware of any affiliations, memberships, funding, or financial holdings that mightbe perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

I wish to thank Jan Janouskovec and Fabien Burki in particular for useful discussions and commentson the manuscript; David Bass and Alastair Simpson for comments on Figure 1; and the TulaFoundation, the Natural Sciences and Engineering Research Council, and the Canadian Institutefor Advanced Research for long-term support.

600 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


LITERATURE CITED

1. Allen AE, Moustafa A, Montsant A, Eckert A, Kroth PG, Bowler C. 2012. Evolution and functionaldiversification of fructose bisphosphate aldolase genes in photosynthetic marine diatoms. Mol. Biol. Evol.29:367–79

2. Apt KE, Zaslavkaia L, Lippmeier JC, Lang M, Kilian O, et al. 2002. In vivo characterization of diatommultipartite plastid targeting signals. J. Cell Sci. 115:4061–69

3. Archibald JM, Rogers MB, Toop M, Ishida K, Keeling PJ. 2003. Lateral gene transfer and the evolutionof plastid-targeted proteins in the secondary plastid-containing alga Bigelowiella natans. Proc. Natl. Acad.Sci. USA 100:7678–83

4. Bachvaroff TR, Concepcion GT, Rogers CR, Herman EM, Delwiche CF. 2004. Dinoflagellate expressedsequence tags data indicate massive transfer of chloroplast genes to the nuclear genome. Protist 155:65–78

5. Bachvaroff TR, Sanchez Puerta MV, Delwiche CF. 2005. Chlorophyll c–containing plastid relationshipsbased on analyses of a multigene data set with all four chromalveolate lineages. Mol. Biol. Evol. 22:1772–82

6. Balsera M, Stengel A, Soll J, Bolter B. 2007. Tic62: a protein family from metabolism to protein translo-cation. BMC Evol. Biol. 7:43

7. Baurain D, Brinkmann H, Petersen J, Rodriguez-Ezpeleta N, Stechmann A, et al. 2010. Phylogenomicevidence for separate acquisition of plastids in cryptophytes, haptophytes, and stramenopiles. Mol. Biol.Evol. 27:1698–709

8. Belcher JH, Swale EMF. 1976. Spumella elongata nov. comb., a colourless flagellate from soil. Arch.Protistenkd. 118:215–20

9. Bhattacharya D, Helmchen T, Melkonian M. 1995. Molecular evolutionary analyses of nuclear-encodedsmall subunit ribosomal RNA identify an independent rhizopod lineage containing the Euglyphidae andthe Chlorarachniophyta. J. Eukaryot. Microbiol. 42:64–68

10. Bodyl A. 2005. Do plastid-related characters support the chromalveolate hypothesis? J. Phycol. 41:712–1911. Bodyl A, Mackiewicz P, Stiller JW. 2009. Early steps in plastid evolution: current ideas and controversies.

Bioessays 31:1219–3212. Bodyl A, Stiller JW, Mackiewicz P. 2009. Chromalveolate plastids: direct descent or multiple endosym-

bioses? Trends Ecol. Evol. 24:119–2113. Bolte K, Gruenheit N, Felsner G, Sommer MS, Maier UG, Hempel F. 2011. Making new out of

old: recycling and modification of an ancient protein translocation system during eukaryotic evolution.Mechanistic comparison and phylogenetic analysis of ERAD, SELMA and the peroxisomal importomer.Bioessays 33:368–76

14. Brown MW, Kolisko M, Silberman JD, Roger AJ. 2012. Aggregative multicellularity evolved indepen-dently in the eukaryotic supergroup Rhizaria. Curr. Biol. 22:1123–27

15. Burki F, Flegontov P, Obornik M, Cihlar J, Pain A, et al. 2012. Re-evaluating the green versus red signalin eukaryotes with secondary plastid of red algal origin. Genome Biol. Evol. 4:623–35

16. Burki F, Inagaki Y, Brate J, Archibald JM, Keeling PJ, et al. 2009. Large-scale phylogenomic analysesreveal that two enigmatic protist lineages, Telonemia and Centroheliozoa, are related to photosyntheticchromalveolates. Genome Biol. Evol. 1:231–38

17. Burki F, Kudryavtsev A, Matz MV, Aglyamova GV, Bulman S, et al. 2010. Evolution of Rhizaria: newinsights from phylogenomic analysis of uncultivated protists. BMC Evol. Biol. 10:377

18. Burki F, Okamoto N, Pombert JF, Keeling PJ. 2012. The evolutionary history of haptophytes andcryptophytes: phylogenomic evidence for separate origins. Proc. R. Soc. B 279:2246–54

19. Burki F, Shalchian-Tabrizi K, Minge M, Skjaeveland A, Nikolaev SI, et al. 2007. Phylogenomics reshuf-fles the eukaryotic supergroups. PLoS ONE 2:e790

20. Burki F, Shalchian-Tabrizi K, Pawlowski J. 2008. Phylogenomics reveals a new “megagroup” includingmost photosynthetic eukaryotes. Biol. Lett. 4:366–69

21. Cavalier-Smith T. 1982. The origins of plastids. Biol. J. Linn. Soc. 17:289–30622. Cavalier-Smith T. 1986. The kingdom Chromista: origin and systematics. Prog. Phycol. Res. 4:309–4723. Cavalier-Smith T. 1999. Principles of protein and lipid targeting in secondary symbiogenesis: euglenoid,

dinoflagellate, and sporozoan plastid origins and the eukaryote family tree. J. Eukaryot. Microbiol. 46:347–66


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


24. Cavalier-Smith T. 2004. Chromalveolate diversity and cell megaevolution: interplay of membranes,genomes and cytoskeleton. In Organelles, Genomes and Eukaryotic Evolution, ed. RP Hirt, D Horner,pp. 71–103. London: Taylor and Francis

25. Cavalier-Smith T, Lee JJ. 1985. Protozoa as hosts for endosymbioses and the conversion of symbiontsinto organelles. J. Protozool. 32:376–79

26. Chesnick JM, Hooistra WH, Wellbrock U, Medlin LK. 1997. Ribosomal RNA analysis indicates abenthic pennate diatom ancestry for the endosymbionts of the dinoflagellates Peridinium foliaceum andPeridinium balticum (Pyrrhophyta). J. Eukaryot. Microbiol. 44:314–20

27. Deane JA, Fraunholz M, Su V, Maier UG, Martin W, et al. 2000. Evidence for nucleomorph to hostnucleus gene transfer: light-harvesting complex proteins from cryptomonads and chlorarachniophytes.Protist 151:239–52

28. Deschamps P, Moreira D. 2012. Reevaluating the green contribution to diatom genomes. Genome Biol.Evol. 4:683–88

29. Dodge JD. 1969. Observations on the fine structure of the eyespot and associated organelles in thedinoflagellate Glenodinium foliaceum. J. Cell Sci. 5:479–93

30. Eisen JA, Coyne RS, Wu M, Wu D, Thiagarajan M, et al. 2006. Macronuclear genome sequence of theciliate Tetrahymena thermophila, a model eukaryote. PLoS Biol. 4:e286

31. Elias M, Patron NJ, Keeling PJ. 2009. The RAB family GTPase Rab1A from Plasmodium falciparumdefines a unique paralog shared by chromalveolates and rhizaria. J. Eukaryot. Microbiol. 56:348–56

32. El-Sayed NM, Myler PJ, Blandin G, Berriman M, Crabtree J, et al. 2005. Comparative genomics oftrypanosomatid parasitic protozoa. Science 309:404–9

33. Esteban GF, Fenchel T, Finlay BJ. 2010. Mixotrophy in ciliates. Protist 161:621–4134. Falkowski PG, Katz ME, Knoll AH, Quigg A, Raven JA, et al. 2004. The evolution of modern eukaryotic

phytoplankton. Science 305:354–6035. Fast NM, Kissinger JC, Roos DS, Keeling PJ. 2001. Nuclear-encoded, plastid-targeted genes suggest a

single common origin for apicomplexan and dinoflagellate plastids. Mol. Biol. Evol. 18:418–2636. Felsner G, Sommer MS, Gruenheit N, Hempel F, Moog D, et al. 2011. ERAD components in organ-

isms with complex red plastids suggest recruitment of a preexisting protein transport pathway for theperiplastid membrane. Genome Biol. Evol. 3:140–50

37. Fernandez Robledo JA, Caler E, Matsuzaki M, Keeling PJ, Shanmugam D, et al. 2011. The search forthe missing link: a relic plastid in Perkinsus? Int. J. Parasitol. 41:1217–29

38. Foth BJ, McFadden GI. 2003. The apicoplast: a plastid in Plasmodium falciparum and other apicomplexanparasites. Int. Rev. Cytol. 224:57–110

39. Freeman S, Herron JC. 2007. Evolutionary Analysis. San Francisco: Pearson40. Frommolt R, Werner S, Paulsen H, Goss R, Wilhelm C, et al. 2008. Ancient recruitment by chromists

of green algal genes encoding enzymes for carotenoid biosynthesis. Mol. Biol. Evol. 25:2653–6741. Funes S, Davidson E, Reyes-Prieto A, Magallon S, Herion P, et al. 2002. A green algal apicoplast ancestor.

Science 298:215542. Gardner MJ, Goldman N, Barnett P, Moore PW, Rangachari K, et al. 1994. Phylogenetic analysis of

the rpoB gene from the plastid-like DNA of Plasmodium falciparum. Mol. Biochem. Parasitol. 66:221–3143. Gilson PR, Su V, Slamovits CH, Reith ME, Keeling PJ, McFadden GI. 2006. Complete nucleotide

sequence of the chlorarachniophyte nucleomorph: nature’s smallest nucleus. Proc. Natl. Acad. Sci. USA103:9566–71

44. Gould SB, Sommer MS, Hadfi K, Zauner S, Kroth PG, Maier UG. 2006. Protein targeting into thecomplex plastid of cryptophytes. J. Mol. Evol. 62:674–81

45. Grant J, Tekle YI, Anderson OR, Patterson DJ, Katz LA. 2009. Multigene evidence for the placement of aheterotrophic amoeboid lineage Leukarachnion sp. among photosynthetic stramenopiles. Protist 160:376–85

46. Grauvogel C, Reece KS, Brinkmann H, Petersen J. 2007. Plastid isoprenoid metabolism in the oysterparasite Perkinsus marinus connects dinoflagellates and malaria pathogens—new impetus for studyingalveolates. J. Mol. Evol. 65:725–29

47. Green BR. 2004. The chloroplast genome of dinoflagellates—a reduced instruction set? Protist 155:23–31

602 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


48. Hackett JD, Maranda L, Yoon HS, Bhattacharya D. 2003. Phylogenetic evidence for the cryptophyteorigin of the plastid of Dinophysis (Dinophysiales, Dinophyceae). J. Phycol. 39:440–48

49. Hackett JD, Yoon HS, Li S, Reyes-Prieto A, Rummele SE, Bhattacharya D. 2007. Phylogenomic analysissupports the monophyly of cryptophytes and haptophytes and the association of Rhizaria with chroma-lveolates. Mol. Biol. Evol. 24:1702–13

50. Hackett JD, Yoon HS, Soares MB, Bonaldo MF, Casavant TL, et al. 2004. Migration of the plastidgenome to the nucleus in a peridinin dinoflagellate. Curr. Biol. 14:213–18

51. Hagopian JC, Reis M, Kitajima JP, Bhattacharya D, de Oliveira MC. 2004. Comparative analysis of thecomplete plastid genome sequence of the red alga Gracilaria tenuistipitata var. liui provides insights intothe evolution of rhodoplasts and their relationship to other plastids. J. Mol. Evol. 59:464–77

52. Handeler K, Wagele H, Wahrmund U, Rudinger M, Knoop V. 2010. Slugs’ last meals: molecularidentification of sequestered chloroplasts from different algal origins in Sacoglossa (Opisthobranchia,Gastropoda). Mol. Ecol. Resour. 10:968–78

53. Hannaert V, Saavedra E, Duffieux F, Szikora J-P, Rigden DJ, et al. 2003. Plant-like traits associated withmetabolism of Trypanosoma parasites. Proc. Natl. Acad. Sci. USA 100:1067–71

54. Harper JT, Keeling PJ. 2003. Nucleus-encoded, plastid-targeted glyceraldehyde-3-phosphate dehydro-genase (GAPDH) indicates a single origin for chromalveolate plastids. Mol. Biol. Evol. 20:1730–35

55. Harper JT, Waanders E, Keeling PJ. 2005. On the monophyly of the chromalveolates using a six-proteinphylogeny of eukaryotes. Int. J. Syst. Evol. Microbiol. 55:487–96

56. Hempel F, Bullmann L, Lau J, Zauner S, Maier UG. 2009. ERAD-derived preprotein transport acrossthe second outermost plastid membrane of diatoms. Mol. Biol. Evol. 26:1781–90

57. Hirakawa Y, Burki F, Keeling PJ. 2011. Nucleus- and nucleomorph-targeted histone proteins in achlorarachniophyte alga. Mol. Microbiol. 80:1439–49

58. Hirakawa Y, Burki F, Keeling PJ. 2012. Genome-based reconstruction of the protein import machineryin the secondary plastid of a chlorarachniophyte alga. Eukaryot. Cell 11:324–33

59. Hirakawa Y, Gile GH, Ota S, Keeling PJ, Ishida K. 2010. Characterization of periplastidal compartment-targeting signals in chlorarachniophytes. Mol. Biol. Evol. 27:1538–45

60. Hirakawa Y, Nagamune K, Ishida K. 2009. Protein targeting into secondary plastids of chlorarachnio-phytes. Proc. Natl. Acad. Sci. USA 106:12820–25

61. Horiguchi T, Takano Y. 2006. Serial replacement of a diatom endosymbiont in the marine dinoflagellatePeridinium quinquecorne (Peridinales, Dinophyceae). Phycol. Res. 54:193–200

62. Howe CJ, Barbrook AC, Nisbet RE, Lockhart PJ, Larkum AW. 2008. The origin of plastids. Philos.Trans. R. Soc. Lond. B 363:2675–85

63. Huang J, Gogarten JP. 2007. Did an ancient chlamydial endosymbiosis facilitate the establishment ofprimary plastids? Genome Biol. 8:R99

64. Huang J, Mullapudi N, Lancto CA, Scott M, Abrahamsen MS, Kissinger JC. 2004. Phylogenomic evi-dence supports past endosymbiosis, intracellular and horizontal gene transfer in Cryptosporidium parvum.Genome Biol. 5:R88

65. Imanian B, Pombert JF, Dorrell RG, Burki F, Keeling PJ. 2012. Tertiary endosymbiosis in two dino-toms has generated little change in the mitochondrial genomes of their dinoflagellate hosts and diatomendosymbionts. PLoS ONE 7:e43763

66. Inagaki Y, Dacks JB, Doolittle WF, Watanabe KI, Ohama T. 2000. Evolutionary relationship betweendinoflagellates bearing obligate diatom endosymbionts: insight into tertiary endosymbiosis. Int. J. Syst.Evol. Microbiol. 50:2075–81

67. Ishida K, Green BR. 2002. Second- and third-hand chloroplasts in dinoflagellates: phylogeny of oxygen-evolving enhancer 1 (PsbO) protein reveals replacement of a nuclear-encoded plastid gene by that of ahaptophyte tertiary endosymbiont. Proc. Natl. Acad. Sci. USA 99:9294–99

68. Janouskovec J, Horak A, Obornik M, Lukes J, Keeling PJ. 2010. A common red algal origin of theapicomplexan, dinoflagellate, and heterokont plastids. Proc. Natl. Acad. Sci. USA 107:10949–54

69. Jarvis P, Dormann P, Peto CA, Lutes J, Benning C, Chory J. 2000. Galactolipid deficiency and abnormalchloroplast development in the Arabidopsis MGD synthase 1 mutant. Proc. Natl. Acad. Sci. USA 97:8175–79

70. Jarvis P, Soll J. 2001. Toc, Tic, and chloroplast protein import. Biochim. Biophys. Acta 1541:64–79


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


71. Jeffrey SW. 1976. Occurrence of chlorophyll c1 and c2 in algae. J. Phycol. 12:349–5472. Johnson MD. 2011. Acquired phototrophy in ciliates: a review of cellular interactions and structural

adaptations. J. Eukaryot. Microbiol. 58:185–9573. Keeling PJ. 2001. Foraminifera and Cercozoa are related in actin phylogeny: two orphans find a home?

Mol. Biol. Evol. 18:1551–5774. Keeling PJ. 2010. The endosymbiotic origin, diversification and fate of plastids. Philos. Trans. R. Soc.

Lond. B 365:729–4875. Kies L. 1974. Elektronenmikroskopische Untersuchungen an Paulinella chromatophora Lauterborn, einer

Thekamobe mit blau-grunen Endosymbionten (Cyanellen). Protoplasma 80:69–8976. Kies L, Kremer BP. 1979. Function of cyanelles in the thecamoeba Paulinella chromatophora. Naturewis-

senschaften 66:578–7977. Kim E, Graham LE. 2008. EEF2 analysis challenges the monophyly of Archaeplastida and Chromalve-

olata. PLoS ONE 3:e262178. Kohler S, Delwiche CF, Denny PW, Tilney LG, Webster P, et al. 1997. A plastid of probable green

algal origin in apicomplexan parasites. Science 275:1485–8979. Kroth PG. 2002. Protein transport into secondary plastids and the evolution of primary and secondary

plastids. Int. Rev. Cytol. 221:191–25580. Kumazaki T, Hori H, Osawa S. 1983. Phylogeny of protozoa deduced from 5S rRNA sequences. J. Mol.

Evol. 19:411–1981. Lane CE, Archibald JM. 2008. The eukaryotic tree of life: endosymbiosis takes its TOL. Trends Ecol.

Evol. 23:268–7582. Lang M, Apt KE, Kroth PG. 1998. Protein transport into “complex” diatom plastids utilizes two different

targeting signals. J. Biol. Chem. 273:30973–7883. Lang-Unnasch N, Reith ME, Munholland J, Barta JR. 1998. Plastids are widespread and ancient in

parasites of the phylum Apicomplexa. Int. J. Parasitol. 28:1743–5484. Larkum AW, Lockhart PJ, Howe CJ. 2007. Shopping for plastids. Trends Plant Sci. 12:189–9585. Lauterborn R. 1895. Protozoenstudien II. Paulinella chromatophora nov. gen., nov. spec., ein beschalter

Rhizopode des Sußwassers mit blaugrunen chromatophorenartigen Einschlussen. Z. Wiss. Zool. 59:537–44

86. Leander BS, Triemer RE, Farmer MA. 2001. Character evolution in heterotrophic euglenids. Eur. J.Protistol. 37:337–56

87. Leedale GF, Vickerman K. 2000. Phylum Euglenozoa. In An Illustrated Guide to the Protozoa, ed. JJ Lee,GF Leedale, P Bradbury, pp. 1135–85. Lawrence, KS: Soc. Protozool. 2nd ed.

88. Lv HX, Guo GQ, Yang ZN. 2009. Translocons on the inner and outer envelopes of chloroplasts sharesimilar evolutionary origin in Arabidopsis thaliana. J. Evol. Biol. 22:1418–28

89. Maeda T, Hirose E, Chikaraishi Y, Kawato M, Takishita K, et al. 2012. Algivore or phototroph?Plakobranchus ocellatus (Gastropoda) continuously acquires kleptoplasts and nutrition from multiple algalspecies in nature. PLoS ONE 7:e42024

90. Marin B, Nowack EC, Melkonian M. 2005. A plastid in the making: evidence for a second primaryendosymbiosis. Protist 156:425–32

91. Martin W. 2010. Evolutionary origins of metabolic compartmentalization in eukaryotes. Philos. Trans.R. Soc. Lond. B 365:847–55

92. Martin W, Borst P. 2003. Secondary loss of chloroplasts in trypanosomes. Proc. Natl. Acad. Sci. USA100:765–67

93. Martin W, Rujan T, Richly E, Hansen A, Cornelsen S, et al. 2002. Evolutionary analysis of Arabidopsis,cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genesin the nucleus. Proc. Natl. Acad. Sci. USA 99:12246–51

94. Matsumoto T, Shinozaki F, Chikuni T, Yabuki A, Takishita K, et al. 2011. Green-colored plastids inthe dinoflagellate genus Lepidodinium are of core chlorophyte origin. Protist 162:268–76

95. Matsuzaki M, Kuroiwa H, Kuroiwa T, Kita K, Nozaki H. 2008. A cryptic algal group unveiled: a plastidbiosynthesis pathway in the oyster parasite Perkinsus marinus. Mol. Biol. Evol. 25:1167–79

96. McFadden GI. 1999. Plastids and protein targeting. J. Eukaryot. Microbiol. 46:339–46

604 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


97. McFadden GI. 2001. Primary and secondary endosymbiosis and the origin of plastids. J. Phycol. 37:951–59

98. McFadden GI, Reith M, Munholland J, Lang-Unnasch N. 1996. Plastid in human parasites. Nature381:482

99. McFadden GI, Waller RF. 1997. Plastids in parasites of humans. Bioessays 19:1033–40100. Melkonian M, Mollenhauer D. 2005. Robert Lauterborn (1869–1952) and his Paulinella chromatophora.

Protist 156:253–62101. Mignot JP. 1977. Etude ultrastructurale d’un Flagelle du genre Spumella Cienk. ( = Heterochromonas

Pascher, = Monas O. F. Muller), Chrysomonadine leucoplastidiee. Protistologica 13:219–31102. Minge MA, Shalchian-Tabrizi K, Torresen OK, Takishita K, Probert I, et al. 2010. A phylogenetic mosaic

plastid proteome and unusual plastid-targeting signals in the green-colored dinoflagellate Lepidodiniumchlorophorum. BMC Evol. Biol. 10:191

103. Moore RB, Obornik M, Janouskovec J, Chrudimsky T, Vancova M, et al. 2008. A photosynthetic alveolateclosely related to apicomplexan parasites. Nature 451:959–63

104. Moustafa A, Beszteri B, Maier UG, Bowler C, Valentin K, Bhattacharya D. 2009. Genomic footprintsof a cryptic plastid endosymbiosis in diatoms. Science 324:1724–26

105. Moustafa A, Reyes-Prieto A, Bhattacharya D. 2008. Chlamydiae has contributed at least 55 genes toPlantae with predominantly plastid functions. PLoS ONE 3:e2205

106. Nakayama T, Ishida K. 2009. Another acquisition of a primary photosynthetic organelle is underway inPaulinella chromatophora. Curr. Biol. 19:R284–85

107. Nassoury N, Cappadocia M, Morse D. 2003. Plastid ultrastructure defines the protein import pathwayin dinoflagellates. J. Cell Sci. 116:2867–74

108. Nikolaev SI, Berney C, Fahrni JF, Bolivar I, Polet S, et al. 2004. The twilight of Heliozoa and rise ofRhizaria, an emerging supergroup of amoeboid eukaryotes. Proc. Natl. Acad. Sci. USA 101:8066–71

109. Nosenko T, Bhattacharya D. 2007. Horizontal gene transfer in chromalveolates. BMC Evol. Biol. 7:173110. Nosenko T, Lidie KL, Van Dolah FM, Lindquist E, Cheng JF, Bhattacharya D. 2006. Chimeric plastid

proteome in the Florida “red tide” dinoflagellate Karenia brevis. Mol. Biol. Evol. 23:2026–38111. Not F, Valentin K, Romari K, Lovejoy C, Massana R, et al. 2007. Picobiliphytes: a marine picoplanktonic

algal group with unknown affinities to other eukaryotes. Science 315:253–55112. Nowack EC, Grossman AR. 2012. Trafficking of protein into the recently established photosynthetic

organelles of Paulinella chromatophora. Proc. Natl. Acad. Sci. USA 109:5340–45113. Nowack EC, Melkonian M, Glockner G. 2008. Chromatophore genome sequence of Paulinella sheds

light on acquisition of photosynthesis by eukaryotes. Curr. Biol. 18:410–18114. Nowack EC, Vogel H, Groth M, Grossman AR, Melkonian M, Glockner G. 2011. Endosymbiotic gene

transfer and transcriptional regulation of transferred genes in Paulinella chromatophora. Mol. Biol. Evol.28:407–22

115. Obornik M, Modry D, Lukes M, Cernotikova-Stribrna E, Cihlar J, et al. 2012. Morphology, ultrastruc-ture and life cycle of Vitrella brassicaformis n. sp., n. gen., a novel chromerid from the Great Barrier Reef.Protist 163:306–23

116. Okamoto N, Chantangsi C, Horak A, Leander BS, Keeling PJ. 2009. Molecular phylogeny and descrip-tion of the novel katablepharid Roombia truncata gen. et sp. nov., and establishment of the Hacrobia taxonnov. PLoS ONE 4:e7080

117. Okamoto N, Horak A, Keeling PJ. 2012. Description of two species of early branching dinoflagellates,Psammosa pacifica n. g., n. sp. and P. atlantica n. sp. PLoS ONE 7:e34900

118. Okamoto N, Inouye I. 2006. Hatena arenicola gen. et sp. nov., a katablepharid undergoing probableplastid acquisition. Protist 157:401–19

119. Patron NJ, Inagaki Y, Keeling PJ. 2007. Multiple gene phylogenies support the monophyly of cryp-tomonad and haptophyte host lineages. Curr. Biol. 17:887–91

120. Patron NJ, Rogers MB, Keeling PJ. 2004. Gene replacement of fructose-1,6-bisphosphate aldolase (FBA)supports a single photosynthetic ancestor of chromalveolates. Eukaryot. Cell 3:1169–75

121. Patron NJ, Waller RF, Keeling PJ. 2006. A tertiary plastid uses genes from two endosymbionts. J. Mol.Biol. 357:1373–82


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


122. Rada P, Dolezal P, Jedelsky PL, Bursac D, Perry AJ, et al. 2011. The core components of organellebiogenesis and membrane transport in the hydrogenosomes of Trichomonas vaginalis. PLoS ONE 6:e24428

123. Ralph SA, Van Dooren GG, Waller RF, Crawford MJ, Fraunholz MJ, et al. 2004. Tropical infectiousdiseases: metabolic maps and functions of the Plasmodium falciparum apicoplast. Nat. Rev. Microbiol.2:203–16

124. Reyes-Prieto A, Moustafa A, Bhattacharya D. 2008. Multiple genes of apparent algal origin suggestciliates may once have been photosynthetic. Curr. Biol. 18:956–62

125. Rice DW, Palmer JD. 2006. An exceptional horizontal gene transfer in plastids: gene replacement bya distant bacterial paralog and evidence that haptophyte and cryptophyte plastids are sisters. BMC Biol.4:31

126. Rogers MB, Gilson PR, Su V, McFadden GI, Keeling PJ. 2007. The complete chloroplast genome ofthe chlorarachniophyte Bigelowiella natans: evidence for independent origins of chlorarachniophyte andeuglenid secondary endosymbionts. Mol. Biol. Evol. 24:54–62

127. Rogers MB, Keeling PJ. 2003. Lateral gene transfer and re-compartmentalisation of Calvin cycle enzymesin plants and algae. J. Mol. Evol. 58:367–75

128. Saldarriaga JF, Taylor FJR, Keeling PJ, Cavalier-Smith T. 2001. Dinoflagellate nuclear SSU rRNAphylogeny suggests multiple plastid losses and replacements. J. Mol. Evol. 53:204–13

129. Sanchez-Puerta MV, Lippmeier JC, Apt KE, Delwiche CF. 2007. Plastid genes in a non-photosyntheticdinoflagellate. Protist 158:105–17

130. Schnepf E, Elbrachter M. 1988. Cryptophycean-like double membrane-bound chloroplast in the di-noflagellate, Dinophysis Ehrenb.: evolutionary, phylogenetic and toxicological implications. Bot. Acta101:196–203

131. Sekiguchi H, Moriya M, Nakayama T, Inouye I. 2002. Vestigial chloroplasts in heterotrophic stra-menopiles Pteridomonas danica and Ciliophrys infusionum (Dictyochophyceae). Protist 153:157–67

132. Simpson AG, Roger AJ. 2002. Eukaryotic evolution: getting to the root of the problem. Curr. Biol.12:R691–93

133. Slamovits CH, Keeling PJ. 2008. Plastid-derived genes in the nonphotosynthetic alveolate Oxyrrhismarina. Mol. Biol. Evol. 25:1297–306

134. Sommer MS, Daum B, Gross LE, Weis BL, Mirus O, et al. 2011. Chloroplast Omp85 proteins changeorientation during evolution. Proc. Natl. Acad. Sci. USA 108:13841–46

135. Sommer MS, Gould SB, Lehmann P, Gruber A, Przyborski JM, Maier UG. 2007. Der1-mediatedpreprotein import into the periplastid compartment of chromalveolates? Mol. Biol. Evol. 24:918–28

136. Strittmatter P, Soll J, Bolter B. 2010. The chloroplast protein import machinery: a review. Methods Mol.Biol. 619:307–21

137. Sulli C, Fang Z, Muchhal U, Schwartzbach SD. 1999. Topology of Euglena chloroplast protein precursorswithin endoplasmic reticulum to Golgi to chloroplast transport vesicles. J. Biol. Chem. 274:457–63

138. Takishita K, Koike K, Maruyama T, Ogata T. 2002. Molecular evidence for plastid robbery (Klepto-plastidy) in Dinophysis, a dinoflagellate causing diarrhetic shellfish poisoning. Protist 153:293–302

139. Takishita K, Yamaguchi H, Maruyama T, Inagaki Y. 2009. A hypothesis for the evolution of nuclear-encoded, plastid-targeted glyceraldehyde-3-phosphate dehydrogenase genes in “chromalveolate” mem-bers. PLoS ONE 4:e4737

140. Teles-Grilo ML, Tato-Costa J, Duarte SM, Maia A, Casal G, Azevedo C. 2007. Is there a plastid inPerkinsus atlanticus (phylum Perkinsozoa)? Eur. J. Protistol. 43:163–67

141. Tengs T, Dahlberg OJ, Shalchian-Tabrizi K, Klaveness D, Rudi K, et al. 2000. Phylogenetic analysesindicate that the 19′hexanoyloxy-fucoxanthin-containing dinoflagellates have tertiary plastids of hapto-phyte origin. Mol. Biol. Evol. 17:718–29

142. Tyler BM, Tripathy S, Zhang X, Dehal P, Jiang RH, et al. 2006. Phytophthora genome sequences uncoverevolutionary origins and mechanisms of pathogenesis. Science 313:1261–66

143. Tyra HM, Linka M, Weber AP, Bhattacharya D. 2007. Host origin of plastid solute transporters in thefirst photosynthetic eukaryotes. Genome Biol. 8:R212

144. Van de Peer Y, Van der Auwera G, De Wachter R. 1996. The evolution of stramenopiles and alveolatesas derived by “substitution rate calibration” of small ribosomal subunit RNA. J. Mol. Evol. 42:201–10

606 Keeling

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


145. Waller RF, Patron NJ, Keeling PJ. 2006. Phylogenetic history of plastid-targeted proteins in theperidinin-containing dinoflagellate Heterocapsa triquetra. Int. J. Syst. Evol. Microbiol. 56:1439–347

146. Waller RF, Reed MB, Cowman AF, McFadden GI. 2000. Protein trafficking to the plastid of Plasmodiumfalciparum is via the secretory pathway. EMBO J. 19:1794–802

147. Watanabe MM, Suda S, Inouye I, Sawaguchi I, Chihara M. 1990. Lepidodinium viride gen. et sp. nov.(Gymnodiniales, Dinophyta), a green dinoflagellate with a chlorophyll a- and b-containing endosym-biont. J. Phycol. 26:741–51

148. Williams BAP, Keeling PJ. 2003. Cryptic organelles in parasitic protists and fungi. Adv. Parasitol. 54:9–67149. Williamson DH, Gardner MJ, Preiser P, Moore DJ, Rangachari K, Wilson RJ. 1994. The evolutionary

origin of the 35 kb circular DNA of Plasmodium falciparum: new evidence supports a possible rhodophyteancestry. Mol. Gen. Genet. 243:249–52

150. Wilson RJ. 2002. Progress with parasite plastids. J. Mol. Biol. 319:257–74151. Wilson RJ, Denny PW, Preiser DJ, Rangachari K, Roberts K, et al. 1996. Complete gene map of the

plastid-like DNA of the malaria parasite Plasmodium falciparum. J. Mol. Biol. 261:155–72152. Woehle C, Dagan T, Martin WF, Gould SB. 2011. Red and problematic green phylogenetic signals

among thousands of nuclear genes from the photosynthetic and Apicomplexa-related Chromera velia.Genome Biol. Evol. 3:1220–30

153. Yabuki A, Inagaki Y, Ishida K. 2010. Palpitomonas bilix gen. et sp. nov.: a novel deep-branching het-erotroph possibly related to Archaeplastida or Hacrobia. Protist 161:523–38

154. Yoon HS, Hackett JD, Bhattacharya D. 2002. A single origin of the peridinin- and fucoxanthin-containingplastids in dinoflagellates through tertiary endosymbiosis. Proc. Natl. Acad. Sci. USA 99:11724–29

155. Yoon HS, Price DC, Stepanauskas R, Rajah VD, Sieracki ME, et al. 2011. Single-cell genomics revealsorganismal interactions in uncultivated marine protists. Science 332:714–17

156. Zhang Z, Green BR, Cavalier-Smith T. 1999. Single gene circles in dinoflagellate chloroplast genomes.Nature 400:155–59


Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.

PP64-frontmatter ARI 25 March 2013 10:21

Annual Review ofPlant Biology

Volume 64, 2013Contents

Benefits of an Inclusive US Education SystemElisabeth Gantt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Plants, Diet, and HealthCathie Martin, Yang Zhang, Chiara Tonelli, and Katia Petroni � � � � � � � � � � � � � � � � � � � � � � � � �19

A Bountiful Harvest: Genomic Insights into Crop DomesticationPhenotypesKenneth M. Olsen and Jonathan F. Wendel � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �47

Progress Toward Understanding Heterosis in Crop PlantsPatrick S. Schnable and Nathan M. Springer � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �71

Tapping the Promise of Genomics in Species with Complex,Nonmodel GenomesCandice N. Hirsch and C. Robin Buell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �89

Understanding Reproductive Isolation Based on the Rice ModelYidan Ouyang and Qifa Zhang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 111

Classification and Comparison of Small RNAs from PlantsMichael J. Axtell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 137

Plant Protein InteractomesPascal Braun, Sebastien Aubourg, Jelle Van Leene, Geert De Jaeger,

and Claire Lurin � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 161

Seed-Development Programs: A Systems Biology–Based ComparisonBetween Dicots and MonocotsNese Sreenivasulu and Ulrich Wobus � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 189

Fruit Development and RipeningGraham B. Seymour, Lars Østergaard, Natalie H. Chapman, Sandra Knapp,

and Cathie Martin � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 219

Growth Mechanisms in Tip-Growing Plant CellsCaleb M. Rounds and Magdalena Bezanilla � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 243

Future Scenarios for Plant PhenotypingFabio Fiorani and Ulrich Schurr � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 267

v

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Microgenomics: Genome-Scale, Cell-Specific Monitoring of MultipleGene Regulation TiersJ. Bailey-Serres � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 293

Plant Genome Engineering with Sequence-Specific NucleasesDaniel F. Voytas � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 327

Smaller, Faster, Brighter: Advances in Optical Imagingof Living Plant CellsSidney L. Shaw and David W. Ehrhardt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 351

Phytochrome Cytoplasmic SignalingJon Hughes � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 377

Photoreceptor Signaling Networks in Plant Responses to ShadeJorge J. Casal � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 403

ROS-Mediated Lipid Peroxidation and RES-Activated SignalingEdward E. Farmer and Martin J. Mueller � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 429

Potassium Transport and Signaling in Higher PlantsYi Wang and Wei-Hua Wu � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 451

Endoplasmic Reticulum Stress Responses in PlantsStephen H. Howell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 477

Membrane Microdomains, Rafts, and Detergent-Resistant Membranesin Plants and FungiJan Malinsky, Miroslava Opekarova, Guido Grossmann, and Widmar Tanner � � � � � � � 501

The EndodermisNiko Geldner � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 531

Intracellular Signaling from Plastid to NucleusWei Chi, Xuwu Sun, and Lixin Zhang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 559

The Number, Speed, and Impact of Plastid Endosymbioses inEukaryotic EvolutionPatrick J. Keeling � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 583

Photosystem II Assembly: From Cyanobacteria to PlantsJorg Nickelsen and Birgit Rengstl � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 609

Unraveling the Heater: New Insights into the Structure of theAlternative OxidaseAnthony L. Moore, Tomoo Shiba, Luke Young, Shigeharu Harada, Kiyoshi Kita,

and Kikukatsu Ito � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 637

Network Analysis of the MVA and MEP Pathways for IsoprenoidSynthesisEva Vranova, Diana Coman, and Wilhelm Gruissem � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 665

vi Contents

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.


Toward Cool C4 CropsStephen P. Long and Ashley K. Spence � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 701

The Spatial Organization of Metabolism Within the Plant CellLee J. Sweetlove and Alisdair R. Fernie � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 723

Evolving Views of Pectin BiosynthesisMelani A. Atmodjo, Zhangying Hao, and Debra Mohnen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 747

Transport and Metabolism in Legume-Rhizobia SymbiosesMichael Udvardi and Philip S. Poole � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 781

Structure and Functions of the Bacterial Microbiota of PlantsDavide Bulgarelli, Klaus Schlaeppi, Stijn Spaepen, Emiel Ver Loren van Themaat,

and Paul Schulze-Lefert � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 807

Systemic Acquired Resistance: Turning Local Infectioninto Global DefenseZheng Qing Fu and Xinnian Dong � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 839

Indexes

Cumulative Index of Contributing Authors, Volumes 55–64 � � � � � � � � � � � � � � � � � � � � � � � � � � � 865

Cumulative Index of Article Titles, Volumes 55–64 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 871

Errata

An online log of corrections to Annual Review of Plant Biology articles may be found athttp://www.annualreviews.org/errata/arplant

Contents vii

Ann

u. R

ev. P

lant

Bio

l. 20

13.6

4:58

3-60

7. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by U

nive

rsity

of

Bri

tish

Col

umbi

a on

08/

21/1

3. F

or p

erso

nal u

se o

nly.

Documents

The Number, Speed, and Impact of Plastid Endosymbioses in ...PP64CH24-Keeling ARI 25 March 2013 11:0 The Number, Speed, and Impact of Plastid Endosymbioses in Eukaryotic Evolution