Gene, 64 (1988) 165-172

Elsevier

GEN 02329

165

Short Communications

Cloning and analysis of a yeast genomic DNA sequence capable of directing gene transcription in Escherichia cofi as well as in yeast

(Recombinant DNA; Succharomyces cerevisiue; promoters; plasmid vectors; CAT assay; primer extension;

Southern blot)

Ju Won Kwak”, Jiyoung Kim”, Ook Joon Yoo b and Moon Hi Hans

” Genetic Engineering Center, and b Department of Biological Science and Engineering, KAIST, Seoul (Korea) Tel. (02)967-8901

Received 8 June 1987

Revised 15 September 1987

Accepted 11 December 1987

Received by publisher 5 January 1988

SUMMARY

A DNA fragment was isolated from yeast genomic sequences by its ability to direct the transcription of

promoterless CmR (cat) gene in Escherichiu coli and in yeast. Nucleotide sequencing and primer extension

analysis showed that yeast DNA contains sets of consensus sequences pertaining to prokaryotic and yeast-type

promoter elements. It was designated as yeast- and E. coli-type promoter (YEPZ). Typical E. coli-type

promoter elements are found at appropriate positions: TATTTT from -12 to -7 and TTGTCC from -35 to

-30 with their spacing of 17 bp from the single mRNA start point determined by the primer extension. Analysis

of cat transcripts from yeast cells showed that the YEPI caused multiple transcription initiations at more than

20 different points that are spaced over a lOO-bp region. The DNA is composed of A + T-rich sequences and

putative TATA-like sequences are found at several places upstream from the transcription start points. Deletion

analysis showed that a 276-bp sequence between -872 and -596 from the initiating ATG codon was required

for the maximal promoter activity in yeast but not in E. coli.

Correspondence to: Dr. M.H. Han, Genetic Engineering Center,

the Korea Advanced Institute of Science and Technology,

Cheongryang P.O. Box 131, Seoul (Korea) Tel. (02)967-890 1,

ext. 3341.

ExoIII, exonuclease III; kb, kilobase or 1000 bp; nt, nucleo-

tide(s); PolIk, large (Klenow) fragment of E. coli DNA polymer-

ase I; poly(A)+ , polyadenylated; R, resistance; Trp + , trypto-

phan prototroph (phenotype); UAS, upstream activation site;

YEPI, yeast- and E. coli-type promoter; YEPGE medium, see

Abbreviations: Ap, ampicillin; bp, base pair(s); CAT, Cm acetyl

transferase; cat, gene coding for CAT; Cm, chloramphenicol;

EXPERIMENTAL AND DISCUSSION, section a.

0378-l 119/SS/$O3.50 0 1988 Elsevier Science Publishers B.V. (Biomedical Division)

INTRODUCTION EXPERIMENTAL AND DISCUSSION

Functional expression of some yeast genes in

E. coli has been reported (review, Hollenberg, 1980).

In order for the expression to occur, the tran-

scriptional machinery of the prokaryotic cell must

recognize and interact with some DNA elements

which are contained in the 5’-flanking sequences of

yeast genes or in the adjacent plasmid nucleotide

sequences. In the case of yeast HIS3 gene, a region

close to or within the yeast promoter was essential

for expression in E. coli (Struhl and Davis, 1980).

The 5’-flanking sequences of the yeast URA3 genes

were revealed to contain E. coli-type promoter se-

quences (Rose et al., 1984).

The random genomic DNA fragments of Saccha-

romyces cerevisiae were tested for their ability to

express the plasmid gene of tetracycline resistance in

E. coli (Neve et al., 1979). The frequency with which

promoter-active recombinants were found among

total recombinants was as high as 28 %. The yeast

DNA segments which are transcriptionally active in

E. coli as well as in yeast would be expected to exist

in yeast genome with significant frequencies. Except

a few cases of well-characterized yeast genes, iso-

lation and characterization of the yeast DNA

segments has been limited. As a means of searching

for such a DNA fragment in yeast genome, we have

developed promoter-probe, shuttle plasmid vectors

between S. cerevisiue and E. coli (pK15 and pKW4,

Kwak et al., 1987) utilizing the promoterless CmR

gene. The bacterial CmR (cat) gene has been reported

to be functionally expressed in yeast (Cohen et al.,

1980) and proved to be an efficient marker for both

S. cerevisiue and E. coli (Hadfield et al., 1986).

In this paper, we describe cloning and analysis of

a yeast genomic nucleotide sequence capable of

directing gene transcription in E. coli and in yeast.

Sequencing and transcriptional analysis of the yeast

genomic nucleotide sequence revealed novel features

of E. coli and yeast-type promoter regions which are

conserved in natural S. cerevisiae genome.

(a) Isolation of a yeast genomic DNA sequence

capable of directing gene transcription in Escheri-

chia coli and in yeast

A promoter-probe shuttle plasmid vector, pK15,

was derived from a yeast 2 pm-based shuttle plasmid

between yeast and E. coli, into which the bacterial

promoterless cat gene of pBR325 was inserted. In

pK15, a unique BamHI site was introduced just

upstream from the ribosome-binding, Shine-Dal-

garno sequences of the promoterless cat gene, and

utilized for cloning of random yeast genomic DNA

fragments to isolate the transcriptionally active

DNA fragments in both organisms.

Genomic DNA with high M, was prepared from

yeast cells as described (Rodriguez and Tait, 1983),

and partially digested with AluI + HueIII. The DNA

fragments in the range of 0.5-2.5 kb in size were

inserted into Bum HI-cleaved and end-filled pK 15,

and introduced into E. co/i LE392 by transfor-

mation. BamHI sites were generated at both ends of

insert DNA in recombinant plasmids. The E. coli

cells carrying promoter-inserted plasmids were se-

lected and amplified, directly, by growing the trans-

formed cells in a medium containing Ap and Cm (50

pg/ml each). Plasmid DNA was prepared from the

amplified culture and reintroduced into yeast

D13-1A (MATa, his3-532, trpl, ga12) by lithium

acetate transformation (Ito et al., 1983). Of 50 Trp +

transformants selected on SD(-trp) agar medium

(0.67% yeast nitrogen base w/o amino acids and 2%

dextrose supplemented with necessary minimal

ingredients except tryptophan), ten colonies showed

resistance to 1 mg Cm/ml in YEPGE medium (1%

yeast extract, 2% peptone, 3% glycerol and 2%

ethanol). After analysis for the inserts in the recom-

binant plasmids by BamHI digestion, it was found

that all the CmR clones turned out to contain the

identical insert of 2.2 kb in length. This is probably

due to selective amplification of a specific clone in

the initial amplification step in E. cofi and/or failure

of weaker promoter-active clones to grow in the

following selection step in yeast, under selective

pressure given by the high concentrations of Cm

used. The insert was designated as YEPI.

167

A B B

I 1

2p or i. lRPI

SE B CK II I

B

C

i i I I I 1

i t

* I

I I ‘L-------l I

AP

YEP 1 (560)

YEP 1 (840)

YEP 1 (2,200)

‘ Primer ’

‘Southern’

s C TE T A BR R RAfH H RRK B

I I I I I I

Fig. 1. Restriction map of YEpl and sequencing strategy. (A) The upper box represents the entire sequence of plasmid YEPI-pK15. The

restriction map of the 2200-bp insert was determined. AluI in parentheses was deduced from the sequence data of cur gene and used

to prepare the primer (see panel B). (B) DNA fragments used for primer extension (‘Primer’) and as probe for Southern-blot hybridization

(‘Southern’). (C) Restriction map of YEPZ (840) and sequencing strategy. The arrows indicate the direction of sequencing by the method

of Sanger et al. (1977). The abbreviations used for restriction sites are: A, Alul; B, BarnHI; C, &I; E, EcoRI; H, HueIII; K, KpnI;

P, PvuII; R, RsaI; S, SalI; T, TuqI.

16X

(b) Nucleotide sequence analysis of YEPI

A restriction map of YEPl in pK15 is shown in

Fig. 1A. To narrow down the DNA region required

for promoter activity, YEPl sequence was subcloned

using available restriction sites. Three different sizes

of DNA fragments (in bp), YEPl(560), YEPI (840)

and YEPI (2200) were tested for in vivo promoter

function by measuring CAT activities of crude

extracts prepared from cells containing each plas-

mid, where the numbers in parentheses refer to the

size of the insert starting from BumHI site upstream

from the Shine-Dalgarno sequences of cut gene. As

shown in Table I, YEPI (560) was able to promote

gene expression to the same levels as YEPl(840) and

YEPl(2200) in E. cob, while gene expression by

YEPl(560) was reduced by about five-fold in yeast

cells. The results indicate that the DNA fragment

YEPl(840) contains DNA elements that are required

for efficient initiation of transcription in E. coli and

yeast cells.

Fig. 1C shows the segments of YEPl(840) region

that were sequenced and the sequencing strategy.

The sequence determined is shown in Fig. 2. Com-

puter search for homology in the yeast gene bank

using BIONETTM system (Smith et al., 1986) identi-

TABLE 1

Expression analysis of the YEPl promoter in yeast and

Escherichia coli

Plasmids” Levels of CAT b

S. cerevisiae E. coli

tied the YEPl sequence as a new one. The tran-

scription start points in E. coli and in yeast have been

determined (see below) and are indicated as solid

circles and solid squares, respectively.

The sequence data show that TATTTT, a Pribnow

box, is contained in a position from -12 to -7, and

TTGTCC from -35 to -30 from the mRNA start

point determined in E. coli. The two hexanucleotide

elements are separated by 17 bp like typical E. coli

promoters (Hawley and McClure, 1983). Shine-

DaIgarno sequences upstream from the ATG codon

(underlined) were derived from the bacterial cat gene.

It is likely that the DNA region containing the

canonical sequences of E. coli promoters is responsi-

ble for promoting transcription initiation in E. coli.

In S. cerevisiae, transcription of YEPl-cat fusion

gene was initiated at multiple sites. TATA-like se-

quences are found at multiple positions upstream

from the transcription initiation sites. It is known

that yeast promoters often contain multiple TATA

elements within 120 nt upstream from the tran-

scription start point (Nagawa and Fink, 1985; Hahn

et al., 1985). In YEPl DNA, TATA-like sequences

are spread out over a region of up to 200 nt at far

upstream sequences in multiple copies for tran-

scription initiation in yeast. The region is composed

of extremely A + T-rich sequences and stretches of

T‘s are also found at several places. The region from

-498 to -362 is composed of up to 80% (A + T).

Nucleotide sequence analysis of YEPl reveals struc-

tural features of DNA sequence which are very

similar to those of well-characterized promoters of

E. coli genes as well as of yeast genes. Coexistence

of both types of promoter elements in YEPl sequence

makes it functional as a yeast-E. coli dual promoter.

YEP1(2200)-pK15 5.5 25

YEPI (840)-pK15 4.3 28

YEPI (560)-pK15 1.3 27

YEPl (O)-pK15 0 0

(c) Determination of transcription start points in

Escherichia cofi and in yeast

d The regions of YEPI denoted by YEPl(2200), YEPZ (840) and

YEPI(560) are represented in Fig. 1A. YEPI(pK15 is the

parental plasmid without any promoter insert.

’ For CAT assay, crude extracts were prepared from E. coli or

yeast cells using acid-washed glass beads and the CAT activities

were determined spectrophotometrically as described by

Rodriguez and Tait (1983). Total protein concentration was

determined by the method of Bradford (1976), using protein-

assay dye reagent concentrate (Bio-Rad). The specific activity of

CAT was expressed as nmol/min/mg.

Primer extension was carried out by hybridizing

RNA synthesized in vivo in E. coli harboring the

plasmid YEPl -pK15 to a labelled primer and subse-

quently treating it with reverse transcriptase. 119 bp

of AluI-PvuII fragment of CmR DNA (Fig. 1B) was

used as primer following 32P-labelling by ExoIII

digestion and PolIk filling-in as described previously

(Guo and Wu, 1983). The location of the tran-

scriptional start point was identified by analysis of

169

Sal1 -850EcoRI GT~GACACACTG~ACACCTTGAATTCTTCCAC~TATTTAAGC~TTGGCAATG~AAAGTGTGC~GGCACGG

-800 -750

GAtCGTTACTAAiCCAGTAGTTtCTGAATCCAiTCTTTCAAG~CAAACGGGA~ATGTGGGGA~TTCGACT

-700

TT~ACCAGTTCCiTTTTTTTcTiAACAGTGAAiGGATCAGTA~GCAGATAGG~GAAGTCCTC~TTAGCGG

-650 -600 B~~HI

CT~GTCTTTGGA~GAGCTTTTTiTGAAAAACTATTTCTTTTA~AGAGATCAC~TGGCATGTT~CGGGATC

-550 ------I

CC~ATATTCTATSTTGGGGGTA~GCTGTCTTAAGTTTGCCTG~AAGTATTATTGTTGTCTCA~GTGCAAT ______ -500

----T TT~CGTTTTACA~TGGGCTAAC~CAACTTAATACAACTTTTC~TTCCTTTGT~TAGAGTCTT~ATGAATG _____ -450 -400

AA~GTCTTATTC~TTTTTTTTT~TTTTTAGCAiAGTTT;A~~~TCCTTTTTT~TTCCAAGAC~~~~~~TC -- ___ _____ -350 _____

TC~TTATATTCT~TTTAGTTTC~CATGGAATT~CAAGTACAC~~~~~~CTAC~GCCTTACTT~AAAATTG _____ ______ -300 -250

CCiGTCTGGGTC~TATACTTTG~ACCCTTGTA~AGCTTTGTA~TTTTGCCAA~TGTTAATCC~GATACAA

ClaI -200

TAiCGATGGCTT~TACATAGAT~CCACTGGCC~ATTGTCCTA~GTAAGAATA~TCTAAGGAG~CTGTTTG

-150

TA~ATCAACTTTiTTTGTAAAAiTTGGCCCAGiATTTTCATA~TCAATACAG~CACGTAGTA~~GGTACC

T1 mm n = +1

ATCCGAGATTTTtAGGAGCTiiiiG:iGCTAAAiiTG

? Fig. 2. Nucleotide sequence of YEPl. The first nucleotide of the translational start codon, ATG, for the CmR (cat) gene was numbered

+ 1. The sequence between the BumHI site (-34) and the ATG (+ 1) is derived from the bacterial cal gene. Sequences bearing

resemblance to canonical promoter sequences of E. coii are boxed; they are found at expected locations in relation to the determined

transcription start point (solid circle over G). The multiple start points for mRNA transcription in yeast are denoted by three different

sizes of squares above each nucleotide, according to their transcriptional efficiencies, i.e., high, moderate, and low. Putative TATA-like

sequences for YEPI as a yeast promoter are marked with dashed lines. An upward arrow indicates the position of 3’-end of the primer

used for the primer extension experiment.

the extension product on a sequencing gel. As seen known DNAin parallel. All mRNA transcripts from

in Fig. 3A, a single, discrete, labelled DNA species YEPI in E. coli started at a G nucleotide position

149 nt in length was obtained. The start point could (-35) as denoted in Fig. 2. In E. coli, 93% of

be identified precisely by running a sequence ladder messages start with a purine nucleotide (Hawley and

obtained by enzymatic dideoxy sequencing of a McClure, 1983).

A E3

220

154

154

119

119

Fig. 3. Primer extension analysis of start points for WI transcripts from YEPI in E. co/i (A) and yeast (B). Total RNA (E. co/i) or

poly(A)’ RNA (yeast) extracted from clones harboring YEPI-introduced pK15 plasmid was used. The primer used is represented in

Fig. 1B and the 3’.end of the primer is indicated by an arrow in Fig. 2. Total RNA from E. coli cells was isolated by extraction with

hot phenol (Scherrer and Darnell, 1962). and treatment with RNasc-free DNaseI (100 units/mg) for 30 min at room temperaturepas

described by Gabain et al. (1983). RNA from yeast cells was prepared as described by Russell and Hall (1982) from cells grown in YEPGE

medium (see EXPERIMENTAL AND DISCUSSION, section a) supplemented with Cm (1 mg/ml) for the sample in lane 4.

Poly(A) + RNA was fractionated by an oligo(dT) cellulose column. The RNAs were combined with “P-labelled primer DNA and primer

extension was done as described by Henikoff et al. (1986). Lanes I and 3, primer extension done on RNA extracted from E. c,oli (lane

1) or yeast (lane 3) cells harboring the parental plasmid pKl5, as controls. Lanes 2 and 4, primer extension done on RNA extracted

from E. co/i (lane 2) or yeast (lane 4) harboring the plasmid, YEPI-pKI5. Lane M, pBR322 Hinfl fragment, “P-labelled by PolIk fill-in.

S, nucleotide sequence ladders obtained by enzymatic dideoxy sequencing of known DNAs. Numbers on either side denote the sizes

in nt.

The result of the primer extension experiment of promoter (Faye et al., 1981). The start points deter-

mRNA synthesized in vivo in yeast is shown in mined are denoted with different sizes of squares

Fig. 3B. In contrast to the mRNA initiation in E. coli, according to transcriptional efficiencies in Fig. 2.

YEPl caused multiple mRNA initiations at more Analysis of 18 yeast promoters previously done by

than 20 different points that are spaced over a lOO-bp Hahn et al. (1985) revealed two preferred sequences

region in S. cerevisiue; this appears to be quite similar for mRNA initiation in yeast, TC(G/A)A and

to the start of transcription from the yeast CYCI PuPuPyPuPu. Of the transcription start points of

171

YEPl, several correspond to the suggested se-

quences; AGTGG, T’CAA and AATE, where the

arrows indicate the position and direction of tran-

scription initiation with moderate efficiencies. But

the most frequent initiation occurred at a separate

sequence, 17 bp upstream from the ATG codon of

the CAT-coding sequence.

(d) Existence of the YEPl sequence in yeast genome

Existence of the YEPl sequence in yeast genome

is expected from the procedure used in its isolation.

It was directly demonstrated by Southern-blot

hybridization of yeast genomic DNA with the 560-bp

SclnzHI fragment of the YEPJ sequence (Fig. 1B).

The “P-labelled YEPJ DNA was hybridized to

chromosomal DNA of S. cerevisiae that had been

digested with either of the restriction endonucleases

EcoRI, C&I, or EcoRI + CfaI, whose cutting sites

are contained in YEPJ sequence. As shown in Fig. 4,

the YEPJ sequence hybridizes to at least three

different restriction fragments of yeast genomic

DNA. The double digestion of yeast genomic DNA

with EcoRI + CJaI produced the band of 612 bp

which had been expected from the sequence of YEPJ in Fig. 2. No positive bands were obtained for paren-

tal plasmid pK15 containing no insert, which was

used as a negative hybridization control. The results

suggest that YEPJ sequence was actually derived

from yeast genomic DNA and more than one copy

of YEPJ-related nucleotide sequence exist in the

S. cerevisiae genome. However, it is not known

whether the YEPJ sequence originated from a

regulatory region of a yeast gene or from another

sequence which happens to contain promoter ele-

ments such that it could be functional when it is

placed in the right position and orientation.

It was demonstrated that deletion of sequence

between nt -872 and -596 of YEPJ region caused

significant reduction of transcription initiation in

vivo as deduced from the measurement of CAT ac-

tivities (Table I). It seems likely that the upstream

sequence of YEPJ may act as an UAS, which is

found in 5’-flanking sequences of many yeast genes,

such as CYCJ, HIS4, GALl, GALJO and PHOS. A

TATA element and a UAS region are the two types

of regulatory sequences required for efficient tran-

scription of yeast genes (Guarente, 1984). The YEPJ sequence seems to contain a TATA element and a

kb 3.0

20 I.6

Fig. 4. Southern-blot analysis of yeast genomic DNA with the

YEPI sequence. Standard Southern-blot analysis was carried

out using yeast genomic DNA cleaved with the following restric-

tion endonucleases: lanes: 1, EcoRI; 2, EcoRI + Club; 3, CluI.

The DNA probe represented in Fig. 1B (B-B) was 32P-labelled

by nick-translation. Parental plasmid pK15 (with no insert) was

run (lane 4) as a negative hybridization control. Lane M is a I-kb

DNA ladder (BRL) 32P-labelled by PolIk.

UAS region which are characteristic of the regu-

latory region of yeast genes coding for mRNA. It is

unlikely that nucleotide sequences other than regula-

tory regions would contain both a TATA element

and an upstream sequence which increases the level

of transcription. Thus, on the basis of sequence

analysis and deletion mutant analysis, we suggest

that the YEPl sequence may be derived from a regu-

latory region of a yeast gene.

Detailed analysis of the YEPI sequence by a com-

bination of in vitro mutagenesis and in viva func-

tional test of mutants will be required to identify the

elements in YEPJ which are important for regulation

of transcription initiation of protein-coding genes in

yeast. It will also be interesting to know the origin

and function of the YEPJ sequence in the native state

172

in yeast genome, which remains to be further investi-

gated.

ACKNOWLEDGEMENTS

This work was supported by a grant from the

Ministry of Science and Technology of Korea. We

are grateful to Prof. Chi-Born Chae, Department of

Biochemistry, University of North Carolina, for

computer analysis of the sequence data and invalu-

able advices. Thanks are also given to Drs. Doe Sun

Kang, Dae Sil Lee, and colleagues at the Molecular

Biology Laboratory, Genetic Engineering Center,

KAIST, for helpful discussions and suggestions.

REFERENCES

Bradford, M.M.: A rapid and sensitive method for the quanti-

tation of microgram quantities of protein utilizing the prin-

ciple of protein dye binding. Anal. Biochem. 72 (1976)

248-254.

Cohen, J.D., Eccleshall, T.R., Needleman, R.B., Federoff, H.,

Buchferer, B.A. and Marmur, J.: Functional expression in

yeast of the Escherichiu coli plasmid gene coding for chlor-

amphenicol acetyhransferase. Proc. Natl. Acad. Sci. USA 77

(1980) 1078-1082.

Faye, G., Leung, D.N., Tatchell, K., Hall, B.D. and Smith, M.:

Deletion mapping of sequences essential for in vivo tran-

scription of the iso-1-cytochrome c gene. Proc. Natl. Acad.

Sci. USA 78 (1981) 2258-2262.

Gabain, A., Belasco, J.G., Schottel, J.L., Chang, A.C.Y. and

Cohen, S.N.: Decay of mRNA in Escherichia coli: investi-

gation of the fate of specific segments of transcripts. Proc.

Natl. Acad. Sci. USA 80 (1983) 653-657.

Guarente, L.: Yeast promoters: positive and negative elements.

Cell 36 (1984) 799-800.

Guo, L.H. and Wu, R.: Exonuclease III: use for DNA sequence

analysis and in specific deletions of nucleotides. Methods

Enzymol. 100 (1983) 60-96.

Hadheld, C., Cashmore, A.M. and Meacock, P.A.: An efficient

chloramphenicol-resistance marker for Saccharomyces cerevi-

siae and Escherichia coli. Gene 45 (1986) 149-158.

Hahn, S., Hoar, E.T. and Guarente, L.: Each of three ‘TATA

elements’ specifies a subset ofthe transcription initiation sites

at the CYC-I promoter of Saccharomyces cerevisiae. Proc.

Natl. Acad. Sci. USA 82 (1985) 8562-8566.

Hawley, D.K. and McClure, W.R.: Compilation and analysis of

Escherichia coli promoter DNA sequences. Nucl. Acids Res.

11 (1983) 2237-2255.

Henikoff, S., Keene, M.A., Fechtel, K. and Fristrom, J.W.: Gene

within a gene: nested Drosophila genes encode unrelated pro-

teins on opposite DNA strands. Cell 44 (1986) 33-42.

Hollenberg, C.P.: Recombinant DNA research. An update of

techniques and results. Prog. Bot. 42 (1980) 171-185.

Ito, H., Fukuda, Y., Murata, K. and Kimura, A.: Transformation

of intact yeast cells treated with alkali cations. J. Bacterial.

153 (1983) 163-168.

Kwak, J.W., Kim, J., Yoo. O.J. and Han, M.H.: A method to

isolate DNA sequences that are promoter-active in Escheri-

chia coli and in yeast. Biochem. Biophys. Res. Commun. 149

(1987) 846-851.

Nagawa, F. and Fink, G.R.: The relationship between the

‘TATA’ sequence and the transcription initiation sites at the

HIS4 gene of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci.

USA 82 (1985) 8557-8561.

Neve, R.L., West, R.W. and Rodriguez, R.L.: Eukaryotic DNA

fragments which act as promoters for a plasmid gene. Nature

277 (1979) 324-325.

Rodriguez, R.L. and Tait, R.C.: Recombinant DNA Techniques:

An Introduction. Addison-Wesley, Reading, MA, 1983.

Rose, M., Grisafi, P. and Botstein, D.: Structure and function of

the yeast U&43 gene: expression in Escherichia coli. Gene 29

(1984) 113-124.

Russell, P.R. and Hall, B.D.: Structure of the Schizosaccharo-

myces pombe cytochrome c gene. Mol. Cell. Biol. 2 (1982)

106-I 16.

Sanger, F., Nicklen, S. and Coulson, A.R.: DNA sequencing with

chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74

(1977) 5463-5467.

Scherrer, K. and Darnell, J.E.: Sedimentation characteristics of

rapidly labelled RNA from HeLa cells. Biochem. Biophys.

Res. Commun. 7 (1962) 486-490.

Smith, D.H., Brutlag, D., Friedland, P. and Kedes, L.H.:

BIONETTM: national computer resource for molecular

biology. Nucl. Acids Res. 14 (1986) 17-20.

Struhl, K. and Davis, R.D.: A physical, genetic and tran-

scriptional map of the cloned his3 gene region of Saccharo-

myces cerevisiae. J. Mol. Biol. 136 (1980) 309-332.

Communicated by G. Wilcox