Anatomy of the stimulative sequences flanking the ARS consensus sequence of chromosome VI in Saccharomyces cerevisiae

Gene, 150(1994)213-220 0 1994 Elsevier Science B.V. All rights reserved. 0378-l 119/94/$07.00

213

GENE 08284

Anatomy of the stimulative sequences flanking the ARS consensus sequence of chromosome VI in Saccharomyces cerevisiae (Plasmid replication efficiency; linker mutation; DUE; single-strand-specific nuclease hypersensitivity)

Mohammad B. Rashid”, Katsuhiko Shirahigeb, Naotake Ogasawarab and Hiroshi Yoshikawab ‘Department of Genetics, Osaka University Medical School, 2-2 Yamadaoka. Suita 565. Japan. Tel. (81-6)-879-3311;

and bGraduate School of Biological

Sciences, Nara Institute of Science and Technology, Ikoma 630-01, Japan

Received by A. Nakazawa: 25 May 1994; Revised/Accepted: 4 July/S July 1994; Received at publishers: 21 July 1994

SUMMARY

We have analyzed the relationship between autonomously replicating sequence (MS) structure and function for three ARS (ARS605, ARS607 and ARS609) from chromosome VI of Saccharomyces cerevisiae by systematic XhoI-linker mutation in the ARS consensus sequence (ACS) and flanking sequences. All mutations that encroached upon the ACS destroyed ARS activity. DNA sequences stimulative for ARS function were identified on either side of the ACS of ARS605 and only on the 3’-side of the ACS of ARS607. In ARS609, however, no such stimulative sequences were observed. Base substitutions complementary to the wild-type sequence of those stimulative regions, in ARS60.5 and ARS607, that did not change the AC of unwinding nor affected ARS activity suggests that these regions have, at least, a function as DNA-unwinding elements (DUE). ARS605, ARS607 and ARS609 DNA are of low AG value and showed hypersensitivity to single-strand-specific nuclease when inserted in negatively supercoiled plasmid. Linker mutations inhibitory for ARS activity (5Lll and 7L14) also caused significant changes in local nucleotide (nt) sensitivity within the ACS and its adjoining regions. Complementary base substitutions, however, did not affect these changes in local nt sensitivity. These results imply that the stimulative regions flanking the ACS are necessary to produce an optimum conformation around the ACS which may be important for full ARS activity.

INTRODUCTION

Hsiao and Carbon, 1979). By sequencing and analyzing elements from the yeast genome, a ll-bp AT-rich consensus sequence (ACS), 5’-WTTTAYRTTTW-3’, was found in all ARS (Van Houten and Newlon, 1990; Shirahige et al., 1993). We have reported the identification and characterization of nine ARS on chromosome VI of S. cerevisiae (Shirahige et al., 1993) exhibiting the general features of an ARS, i.e., an ll-bp ACS flanked by stimulative sequences. This is in accord with the previously characterized ARS (H4ARS, ARS307, 2 PARS, rDNA-ARS, ARSl, and ARSl21). The ACS were either a 11-of-11, lo-of-11 or 9-of-11 match to the consensus (Newlon and Theis, 1993). The ACS, by analogy to other conserved sequences essential for prokaryotic and viral ori (Kornberg and Baker, 1992), is thought to serve as the recognition site ARS

Autonomously replicating sequences (ARS) isolated from Saccharomyces cerevisiae confer to plasmids the ability to replicate autonomously (Stinchomb et al., 1979; Correspondence to: Dr. H. Yoshikawa, Graduate School of Biological Sciences, Nara Institute of Science and Technology, 8916-5 Takayamacho, Ikoma 630-01, Japan. Tel. (81-7437)2-5400; Fax (81-7437)2-5409; e-mail: hyoshika@bs. aist-nara. ac. jp.

Abbreviations: AC& ARS consensus sequence(s); ARS, autonomously replicating sequence(s); bp, base pair(s); CBS, complementary base substitution(s); 2-D, two-dimensionak DUE, DNA-unwinding element(s); kb, kilobase or 1000 bp; nt, nucleotide(s); oligo, oligodeoxyribonucleotide; ori, origin of DNA replication; ORC, ori recognition complex; PCR, polymerase chain reaction; R, A or G; ss, single strand(ed); wt, wild type; W, A or T; Y, C or T. SSDI 0378-1119(94)00495-l

214 for initiator protein(s). Bell and Stillman (1992) have identified a six subunit protein complex called the ori recognition complex (ORC), a strong candidate for the initiator protein(s) in yeast. For full ARS activity the ACS alone is not sufficient, flanking sequences are required (Kearsey, 1984; Srienc et al., 1985; Stritch et al., 1986). But the exact role of these stimulative sequences has yet to be determined. Some evidence suggesting binding sequences for transcription factors (Diffley and Stillman, 1988; Marahrens and Stillman, 1992; Sweder et al., 1988; Walker et al., 1989, 1990 and 1991) and A +T-rich DNA unwinding elements (DUE) within the stimulative sequences (Umek and Kowalski, 1988; Natale et al., 1992) have been reported. The process of replication requires unwinding of a DNA molecule as a prerequisite. The presence of an A +T-rich sequence within the ARS elements is consistent with the necessary melting of the double helix at the replication origin to facilitate initiation by creating an entry site for helicase and other replication proteins. This notion is supported by experiments which show that DUE is essential for the function of H4ARS (Umek and Kowalski, 1988) and ARS307 (Natale et al., 1992). The DUES are easily unwound sequences and are operationally defined as ss-specific nuclease hypersensitive regions (Umek et al., 1989). It has recently been shown by 2-D gel electrophoresis that the DUE, required for ori activity in the chromosome, is the same as that required for the function of the ARS in a plasmid (Huang and Kowalski, 1993). Analyses of &functional elements of nine ARS in chromosome VI of S. cerevisiae by deletion and mutation revealed that stimulative regions are variable in size and location relative to ACS. Among them three ARS, ARS605, ARS607 and ARS609 require only 101 bp, 111 bp and 155 bp, respectively for full ARS activity (Shirahige et al., 1993). It is expected that all information required for full ARS activity are supposed to be contained within these short DNA fragments. The aim of this study was to investigate at the nt level the relationship between structure and function in these regions

Fig. 1. Positions of XhoI-linker ARS607, and 457-bp ScaI-XbaI

flanking the ACS with emphasis on the role of the DUE in ARS activity.

RESULTS

AND DISCUSSION

(a) Effect of linker mutations on ARS activity of ARS605,

ARS607 and ARS609 By deletion analysis a minimum length of ARS of full activity was identified (Shirahige et al., 1993). Further analysis by this technique is difficult because vector sequences contribute to the environment of the ARS and therefore its activity, particularly when deletions very closely approach the ACS. Therefore, we undertook XhoI (CTCGAG) linker mutation analysis using larger ARS fragments which maintain the same environment for 260-457 bp around ACS. ARS605, ARS607 and ARS609 wt and mutant plasmids were subcloned into pKS2 vector containing CEN4, which confers to plasmids the ability to segregate correctly. Therefore, mitotic stability of the plasmid should directly reflect the efficiency of initiation of plasmid replication. ARS activity and mitotic stability of these mutant plasmids as well as wt were measured by following previous protocol (Shirahige et al., 1993; Murray and Szostak, 1983) and then expressed as plasmid loss rate (Fig. 1) and illustrated graphically in Fig. 2. In ARS605, we constructed seventeen XhoI linker mutations along the ARS (Fig. 1A). Three linker mutants (5L6, 5L7 and 5L8) which have mutations in the ACS, failed to transform yeast at high frequency. All other mutant plasmids produced transformants at high frequency but some of them produced transformants that grew more slowly than the wt transformants on selective media. These plasmids (5L5, 5L9, 5Lll and 5L12) from slow growing transformants had higher plasmid loss rates (Fig. 1A and Fig. 2A). Therefore, four sequences were identified by these assays: an essential ACS region and three sequences stimulative to ARS activity. In ARS607, we made twenty one linker mutations along the ARS (Fig. 1B). Three linker mutants (7L8, 7L9

mutations and their ARS activity. The 440-bp Pstl-XbaI fragment of ARS605, 260-bp BamHI-SspI fragment of fragment of ARS609 were subcloned in the SmnI sites of the shuttle vector pKS2 (Shirahige et al., 1993). Linker

mutations were constructed by PCR using mutagenic oligo primers containing XhoI sites. Plasmids were transformed into S. cereuisiae strain DKDSDH (MATa, Ieu2, his3, trpl) by Liacetate method (Ito et al., 1983) and selected on selective (without leucine) plates. ARS activity and mitotic stability assay were performed on at least two or three independent transformants for each plasmid construct following the previous protocol (Shirahige et al., 1993). The wt ARS605, ARS607 and ARS609 sequence are shown (A, B and C, respectively). Each dotted line below the wt sequences represents an XhoI linker mutant. Linker mutants are named as 5Ll,SL2,5L3 and so on. SCBS and 7CBS are for complementary base substitutions at positions shown by bold underlines. For each mutant, mutated nt are shown by upper case. ARS activity (+ indicates ability to transform yeast at high frequency, - indicates failure to transform yeast at high frequency) and plasmid loss rate (in X) which is 100 minus mitotic stability are shown on the right margin. The range of variations of plasmid loss rate is shown in parenthesis (k). The ACS sequences are boxed. The nt numbers as shown above 605, 607 and 609 ARS sequences are according to registered sequences in GenBank/EMBL with accession Nos. D13942, D13940 and D13938 respectively.

215

ARS

+

Plasmid Loss Rate (%) 20 wfi?) 18 21 21 18 49

(il.) lkl) wf;?) W) (f5)

46 (25) 19 (*II) 39 W) _____-_-_----_--------_----_-_-___-__-------------_____---_-~---_--___~cGaG____-___----..__-_._________________ + 31 (i5) 20 (ia?) + 22 iltli + 19 ($2) + 17 (iI) + 20 W)

+

5L12

+ +

6

24 ii2) 36 kk2) 25 kM.1

ARS Activity

Plasmid Loss Rate

+

12 (k22)

+ + + + + + +

14 12 13 11 9 13 16

if11 tt3) WI 1121 e2f (fl) (io)

+ + + + + + * + + + +

11 9 14 23 17 11 23 13 24 17 15

mu) ml) ctl) t-u) e33) (kli (+il) (fl) (k22) WI (i2)

+ + +

15 w2) 11 W) 16 Wf

C ARS Activity

Plasmid Loss Rate fclr)

+ 9L2

9L3 YL4 YL5 9L6 9L7 9LB 9L9 9LlO 9Lll YL12 9L13 YL14 YL15 YL16 YL17 9L18 YLl.9

+ + + + _

+ + +

+ + +

17 13 13 11 18 _

ctl) (fl) (fl) (f4) (f2)

12 13 11 16 15 19

(?lI (fl) (til iM) ifl) Lfl)

216 and 7LlO), which have mutations in the ACS, failed to produce any transformants. All other mutant plasmids produced transfo~ants but some of them (7L14, 7L17 and 7L19) produced transformants that grew slower than the wt plasmid on selective media. The plasmid loss rate (in %) was determined to be markedly increased in contrast to that of the wt plasmid (Fig. 2B). Therefore in ARS607, four sequences were identified: the ACS region essential for ARS activity, and three sequences in the 3’-flanking side stimulative to ARS activity. We have also examined ARM09 sequence by linker mutations (Fig. lC, 2C). Nineteen XhoI linker mutations were performed. Three linker mutant plasmids (9L6,9L7 and 9L8) failed to transform yeast at high frequency as was the case in AR%05 and ARS607. Four linker mutants (9L1, 9L5, 9L12 and 9L19) showed small but significant increase in plasmid loss rates (less than two fold of the wt loss rates). Graphic representation of the plasmid loss rates showed a wavy pattern of effect on ARS activity suggesting the presence of weak stimulative sequences at those regions. (b) Analysis of the stimulative region by complementary base su~titutions We then examined how the stimulative regions influence efficient ARS activity. Since the region was substituted with a G + C-rich XhoI linker, the A+ T-content of the region was slightly changed. An A + T-rich region has been considered necessary to melt the double helix. To determine whether the inhibitory effect by linker mutations was due either to the change in sequence specificity or to the change in DUE we made base substitutions of the stimulative regions complementary to the wt sequence (Fig. IA SCBS-1 to -3 and 7CBS-1 to -3). Interestingly, in SCBS-1, 5CBS-3 and “ICBS-1, 7CBS-2 and 7CBS-3, we observed that these mutant plasmids did not differ from the wt in ARS activity. In 5CBS2, however, there was still detrimental effect on plasmid stability. We conclude that the regions whose functions were affected by linker mutations but not by CBS mutations may contribute to DUE activity. One region in ARS605 (5CBS-2) may function as sequence specific signal for protein factors stimulative to ARS activity. (c) Evidence for a DUE in the negatively supercoiied ARStM5, ARS607 and AR5509 plasmids To examine the presence of a DUE element within the ARS605, ARS607 and ARS609, we performed ss-specific

nuclease (Pl nuclease) hypersensitivity assay using a YIp5 plasmid in which all nuclease hypersensitive sites have been mapped. Pl nuclease supposed to generate only one nick per molecule of negatively supercoiled plasmid as nicking the plasmid relaxes the DNA negative supercoiling therefore making it resistant to Pl nuclease. We mapped nicks made by the Pl nuclease in ARS605 and ARS607 relative to unique restriction sites (NcoI and PuuII) in YIp5 by agarose gel electrophoresis after glyoxal denaturation of the DNA. For each nuclease hypersensitive site, two bands are expected to be generated. In both NcoI and PuuII digested samples, two hypersensitive regions were observed in the plasmid containing AR%07 (Fig. 3A). One pair corresponds to a site in the vector, the 3’-terminal region of the ampicillin-resistance gene (Kowalski et al., 1988) and the other pair corresponds to a region within the ARS607 insert. The latter is more sensitive to nuclease than the former. Similar two pairs of bands were also detected in the plasmid containing ARS605, although sensitivity in ARS60.5 is slightly weaker than that in the vector. The third pair corresponding to the region near 3’-end of URA-3 gene in the vector was seen only in the plasmid containing AR%05 (Fig. 3A). ARS609 was also found to be hypersensitive to Pl nuclease (data not shown). We also performed Pl nuclease assays with mutant plasmids of ARS605 and ARS607 in comparison with the wt plasmids with the expectation of finding variable intensities of bands from mutant plasmids in parallel to their variations in ARS activity (Fig. 3B). No such variation was detected.

(d) Nucleotide level mapping of Pl cleavage sites To determine whether these mutations cause local change of nuclease hypersensitivity, we performed mapping of the nick sites at the nt level. In both ARS605 and ARS607, a broad hypersensitive region was detected. In ARS605 (Fig. 4A), wt or mutants, the broad hypersensitive region includes major sensitive sites (nt 300 to 350) located in the 3’-side of the ACS and other sensitive sites distributed within 155 bp (nt 146 to 300) around the ACS. These two regions together have an A ST content of 70%. The cleavage pattern of a mutant derivative (5Lll) compared with the wt plasmid show decrease in nuclease hypersensitivity in and around the ACS. mutations in 5CBS-3, however, did not affect nuclease hypersensitivity at those positions. In ARS607 (Fig. 4B,C), the broad

Fig. 2. Graphic representation of the plasmid loss rates. Data in Fig. 1 was used to construct the figure. The horizontal axis shows the relative positions (not according to scale) of linkers (Fig. 1) in ARS605 (panel A) ARS607 (panel B) and ARS609 (panel C) together with wt plasmids at the far right side of each graph. Plasmid loss rates of CBS plasmids (SCBS l-3 and 7CBS l--3} are shown by hatched bars. Above each graph, the ACS (boxed) and the orientations (S- 3’) of the ARS are shown.

217

A

ARS605 5’

B

ACS

ARS607 ACS

S’

C

1’

1’

ARS609 5’

soI

ACS

3’

218

A r

B

Ncol

Ncol

Ncol

I

kb

- 2.3 - 2.0

Q,

- 0.6

Fig. 3. Nuclease hypersensitivity assay of wt and mutant derivatives. Panel A is for wt plasmids and panel B is for mutant plasmids as compared with wt. Names of mutants are as in Fig. 1. Panel B was used for quantitative analysis using a Bioimage analyzer (Fuji). Methods: EcoRI-Hind111 fragments from pKS2 containing ARS fragments were inserted in the EcoRI-Hind111 sites of YIpI. Isolation of negatively supercoiled plasmids was essentially according to the protocol followed by Sheflin and Kowalski (1985). Pl nuclease reactions were executed as previously described (Natale et al., 1993) except that about ten-times more enzyme was used to make single nick in most of the plasmids. Plasmids after Pl nuclease reactions, linearized at a restriction endonuclease site (NcoI or PuuII), 32P-end-labeled and denatured before 1% agarose gel electrophoresis. Denatured DNA fragments of known size served as markers (lane M). The bands that are generated from ARS fragments are shown by arrows. The bands shown by asterisks correspond to nuclease sensitive sites in the YIpI. ARS605 or ARS607 plasmid treated in the same way without addition of Pl nuclease is shown in -Pl lanes

hypersensitive region spanning 161 bp (nt 245 to 400) is located around the ACS and in the 5’-flanking side. The region has 72% A+T content. The cleavage pattern of a mutant plasmid (7L14) as compared with the wt plasmid show a significant decrease in sensitivity in the proximity of the ACS, i.e., bands at nt 366 to 377 disappeared in lane 7L14. Furthermore, additional sensitive sites were created (at nt 345 and 348) in the region immediately upstream from ACS. Base substitutions complementary to the wt sequence in the corresponding sites, 5CBS-3 to 5Lll and 7CBS-1 to 7L14, did not affect local sensitivity. In contrast to these mutations, other mutations (7L17 and 7L19) which affected ARS activity did not show significant changes in nuclease sensitivity except for the sites of mutations themselves (Fig. 4C). Mutation within the ACS (7L8) also showed a wt cleavage pattern. These changes in hypersensitivity observed in 5Lll and 7L14 may be correlated with the inhibitory effect of these linker mutations on ARS activity. From these results we conclude that these stimulative regions are necessary to produce an optimum unwound conformation around the ACS for efficient ARS activity of ARS605 and ARS607. The observation that some mutants (7L17 and 7L19) exhibit a wt pattern of nuclease sensitivity whilst having a reduced ARS activity suggest that the nuclease hypersensitivity assay does not always fully reflect the DUE

activity. The wt cleavage pattern exhibited by the ACS mutant (7L8) may indicate that the ACS, which is a putative initiator protein binding site, does not contribute to DUE activity. Although we have observed a broad region of nuclease hypersensitivity, a large portion is not required for full ARS activity as deduced by deletional analysis (Shirahige et al., 1993); deletions upto 52 bp in the 3’-side of the ACS of ARS605 and upto 21 bp in the 5’-side of the ACS of ARS607 have no effect on the ARS activity. However, the cleavage patterns of the two mutants (5Lll and 7L14) show clearly that a small hypersensitive region around the ACS seems to be important for full ARS activity. We name these small regions as micro DUE which may modulate the structure of the ACS through DUE activity as compared with conventional DUE of considerable size. It will be interesting to know how these micro DUE in and around the ACS actually take part in the initiation of DNA replication in vivo. (e) Conclusions (I) By linker mutation analysis, the essential nature of ACS was further confirmed for all ARS, and stimulative sequences were identified in both ARS605 and ARS607. (2) Three stimulative regions were identified by XhoIlinker mutation and mapped in ARS605 as well as in

219

-320

-280

'220

-310

-290 150

Fig. 4. Mapping 4.5 pg of plasmid

of Pl nuclease

hypersensitive

DNA in a volume

sites at the nt level. Pl nuclease

of 60 11. Pl nicked

products of Maxam-Gilbert sequencing reactions wt and mutant plasmids. Panel B shows patterns without

PI nuclease

ACS (arrow)

treatment

are shown

as -Pl.

plasmids

were digested

reactions

were as in Fig. 3 in a larger scale. The reaction contained with 32P and analyzed along side the

with EcoRI and end-labeled

(A+ G, C+T). Panel A is for ARS605 wt and mutant plasmids. Panels B and C are for ARS607 of electrophoresis of the same samples with two different lengths of running time. The wt plasmids The nt numbering

beside the AS-G

lanes is shown

as in Fig. 1. Positions

and sequence

of the

and the linkers (bar) are shown.

All stimulative regions except one were unaffected by base substitutions complementary to the wt. These results suggest strongly that these regions function as DUE. (3) Regions hypersensitive to Pl nuclease were found in all three ARS. These regions are of about 150 bp (ARS607) to 200 bp (ARS605) in and around the ACS. Two linker mutations that affected ARS activity also caused change in local nuclease sensitivity in or around ACS. CBS mutations at the corresponding sites in these two plasmids do not affect ARS activity as well as local nuclease sensitivity. This may indicate that those regions function as micro DUE and are necessary for an optimum conformation important for efficient ARS activity. ARS607.

ported by grants-in-aid for Cooperative Research and for Special Project Research from the Ministry of Education, Science and Culture, Japan.

REFERENCES Bell, S.P. and Stillman, B.: ATP-dependent recognition of eukaryotic origins of DNA replication by a multiprotein complex. Nature 357 (1992) 1288134. Diffley, J.F.X.

and

Stillman,

B.: Purification

of a yeast

protein

binds to origins of DNA replication and a transcriptional Proc. Natl. Acad. Sci. USA 85 (1988) 2120-2124. Hsiao, CL. and Carbon,

J.: High frequency

transformation

that

silencer. of yeast by

plasmids containing the cloned yeast ARC4 gene. Proc. Nat]. Acad. Sci. USA 76 (1979) 382993833.

ACKNOWLEDGMENTS

Huang, R.Y. and Kowalski, D.: A DNA unwinding element and an ARS consensus comprise a replication origin within a yeast chromosome. EMBO J. 12 (1993) 4521-4531.

We are grateful to Dr. S. Harashima for YIp5 plasmid and DKDSDH strain of S. cereuisiae and to P.R. Smith for editorial reading of this paper. This work was sup-

Ito, H., Fukuda, Y., Murata, K. and Kimura, A.: Transformation of intact yeast cells treated with alkali cations. J. Bacterial. 153 (1983) 163-168. Kearsey, S.: Structural requirements for the function of a yeast chromosomal replicator. Cell 37 (1984) 2999307.

220 Kornberg,

A. and Baker,

T.A.: DNA

Replication,

2nd edition.

Freeman and Company, New York, 1992. Kowalski, D., Natale, D.A. and Eddy, M.J.: Stable

DNA

not “breathing,” accounts for single-strand-specific sensitivity of specific A+T-rich sequences. Proc.

W.H.

unwinding,

nuclease hyperNatl. Acad. Sci.

USA 85 (1988) 9464-9468. Marahrens,

Y. and Stillman,

replication

defined

(1992) 817-823. Murray, A.W. and Szostak, gation Natale,

B.: A yeast chromosomal

by multiple

functional

origin of DNA

elements.

Science

255

J.W.: Pedigree

analysis

of plasmid

A.E. and Kowalski,

segre-

D.: DNA helical stability

accounts for mutational defects in a yeast replication Natl. Acad. Sci. USA 89 (1992) 2654-2658.

origin.

Proc.

Natale, D.A., Umek, R.M. and Kowalski, D.: Ease of DNA unwinding is a conserved property of yeast replication origins. Nucleic Acids Res. 21 (1993) 5555560. Newlon,

elements. by mung

Curr. Opin. Genet. bean nuclease

of supercoiled

pBR322

and function

of yeast ARS

Dev. 3 (1993) 752-758.

D.: Altered

H.:

occur in promoter DNA.

Nucleic

detected

and terminator Acids

and

M. and Scott, J.F.: Mutations

terization

in ARSlincrease

J.L.: Purification

and charac-

that bind to yeast ARSs. J. Biol. Chem.

of proteins

on

in Saccharomyces cereuisiae. Yeast

Sweder, KS., Rhode, P.R. and Campbell,

Umek, R.M. and Kowalski, determinant of initiation

263

Res.

regions 13 (1985)

characterization

N. and Yoshikawa,

of autonomously

sequences from chromosome VI of Sacchnromyces Cell. Biol. 13 (1993) 5043-5056.

replicating

cereuisiae. Mol.

Stinchcomb, D.T., Struhl, K. and Davis, R.W.: Isolation and characterization of a yeast chromosomal replicator. Nature 282 (1979) 39-43.

D.: The ease of DNA at replication origins.

unwinding as a Cell 52 (1988)

559-567. Umek, R.M., Linskens,

M.H., Kowaiski,

D. and Huberman,

beginnings in studies of DNA replication Acta 1007 (1989) 1-14. Van Houten,

J.V. and Newlon,

from

yeast

C.S.: Mutational

chromosome

III.

J.A.: New

origins. Biochem. Biophys.

Mol.

analysis

of a replication

Cell. Biol.

10 (1990)

3917-3925. protein

S.C., Tye, B.-K. and Eisenberg,

and its DNA-binding

an autonomously

replicating

site are important sequence

S.: The OBFl

for the function

of

in Saccharomyces cereuisiae.

Mol. Cell. Biol. 9 (1989) 2914-2921. Walker,

K., Iwasaki, T., Rashid, M.B., Ogasawara, Location

R., Woontner,

J.L.: Effect of ARSI mutations

in Saccharomyces cereuisiae. Mol. Cell. Biol.

the rate of simple loss of plasmids 2 (1986) 169-178.

Walker, S.S., Francesconi,

DNA conformations

613776154. Shirahige,

Strich,

origin

C.S. and Theis, J.F.: The structure

Sheflin, L.G and Kowalski,

chromosome stability 5 (1985) 167661684.

(1988) 17270-17277.

in yeast. Cell 34 (1983) 961-970.

D.A., Schubert,

Srienc, F. Bailey, J.E. and Campbell,

S.S., Francesconi,

S.C. and Eisenberg,

S.: A DNA replication

enhancer in Saccharomyces cereuisiue. Proc. Natl. Acad. Sci. USA 87 (1990) 4665-4669. Walker, S.S., Malik, A.K. and Eisenberg,

S.: Analysis

of the interactions

of functional domains of a nuclear origin of replication from Saccharomyces cereuisiae. Nucleic Acids Res. 19 (1991) 6255-6262.