YEAST

VOL.

7: 347-356 (1 99 1)

DNA Insertions in the 'Silent' Regions of the 2 pm Plasmid of Saccharomyces cerevisiae Influence Plasmid Stability J. F. M. BIJVOET, A. L. VAN DER ZANDEN, N. GOOSEN, J. BROUWERt A N D P. VAN D E PUTTE Laboratory r?fMolecular Genetics, Department of Biochemistry, Leiden University, Gorlaeus Laboratories, Einsteinweg 5,2333 CC, Leiden, The Netherlands

Received 27 June 1990: revised 20 November 1990

The 2 pm plasmid of the yeast Saccharomyces cerevisiae is in principle a suitable vector for expression of foreign genes, due to its high copy number and extreme stability. However, the cloning of genes into 2 pm often results in a reduced copy number and/or reduced stability. One reason for this observed instability could be that the inserts in general were made in one of the several open reading frames (ORFs) of the plasmid. Therefore we studied the effect on stability of insertions in the silent regions of 2 pm without interrupting any known essential regions or ORFs. Using the SnaBI site, a yeast-integrating plasmid (Yips) was introduced into the region between the ARS and STB locus in two possible orientations. The resulting plasmids could be stably maintained in the cells without the need for complementation by the wild-type 2 pm plasmid. However, the stability of these plasmids in a cir" host was still one to two orders of magnitude lower (0.2% and 0.8% respectively) as reported for the wild-type 2 pm (0.01%). Removal of 2 kb of the bacterial sequences from Yip5 did not increase stability. The stability was dependent on the orientation of the insert. We found that in the less stable orientation, transcription originating from the insert was running into the STB region. DNA inserted in the XmaIII site located outside the ORFs in the REP2IFLP intergenic region influenced both stability and copy number of the plasmid. These effects are strongly dependent on the size of the insert. Insertion of a 2 kb DNA fragment increased the copy number, probably through an effect on FLP expression. KEY WORDS -2

pm-plasmid stability; copy number; regulation.

INTRODUCTION

(for a review see Futcher, 1988). These areas together occupy about 84% of the total 6318 bp The endogenous 2 pm plasmid of Saccharomyces plasmid. The cis-acting sequences ARS and STB are cerevisiae has been proposed as a suitable vector for essential for replication (Huberman et al., 1987) and the expression of heterologous genes in yeast proper partitioning during cell division respectively (Broach, 1983). The plasmid has a high copy (Jayaram et al., 1985; Murray and Cesarini, 1986). number and is very stable. For the production of FRT is the target sequence for the FLP recombinase interesting enzymes on a large scale these properties which catalyses recombination across the IRs are very attractive. resulting in inversion of one of the unique regions of The plasmid has two characteristic inverted the molecule. This recombination is believed to be repeats (IR), dividing the molecule into two involved in plasmid amplification by double rolling unique regions (Hartley and Donelson, 1980). circle replication (Futcher, 1986; Volkert and There are three cis-active regions, an autonomously Broach, 1986). The proteins REPl and REP2 have a replicating sequence (ARS) (Broach and Hicks, function in the partitioning process probably by I980), a partitioning locus (STB or REP3) (Murray binding to the STB sequence (Cashmore et al., 1986; and Cesarini, 1986) and an FLP recombination tar- Dobson et al., 1988). In addition REPl and REP2 get (FRT) (McLeod et al., 1986) which is located in play a role in the negative regulation of FLP and both IRs. The molecule contains four open reading REPl expression (Armstrong et al., 1988; Veit and frames (ORFs) (Hartley and Donelson, 1980) Fangman, 1988; Som et al., 1988). The RAFgene is encoding the FLP, REPl, REP2 and RAF proteins involved in the regulation of FL P expression by ?Author to whom correspondence should be sent antagonizing REPl/REP2 repression of the FLP 0749%503)3/91/040347-10 $05.00 0 1991 by John Wiley & Sons Ltd

348

J. F. M.BIJVOET ETAL.

Table 1 . Strains Strain

Genotype

Source

E. coli

GY4615 DH 1

endAI, gyrA9b, hsdRl7, recAl, supE44, thil cysB, mtl, pyrF, rpl, srL, thi

D. Hanahan, Harvard P. Morand, Gif-sur-Yvette

S. cerevisiae

MG171 MG171" MG228

MATa, his3-11,3-15,leu2-3,2-112, ura3-52,[cir+] Unilever Research Laboratories, Vlaardingen (cir"] derivative of MG171 cured by pJDB207

(a-DLI)

M A Ta, his3-11,3-15,leu2-3,2-112, ura3-52, cir" (cir+] derivative of a-DLI

MG228+

promoter (Murray et al., 1987). The regulation of the expression of the genes of the 2 pm molecule is of very complex nature and not elucidated in detail. Except for a relatively small reduction in growth rate of the host cells (Mead et al., 1986), there is no specific phenotype associated with the presence of the yeast plasmid. This brief summary of the knowledge about the 2pm plasmid shows that none of the abovementioned domains of the plasmid can be used for the insertion of DNA without affecting the gene regulation or partitioning system. This is illustrated by the fact that all 2pm derivatives studied until now, carrying additional DNA sequences in one of the mentioned regions, are affected in copy number maintenance or in the partitioning system. Hence these derivatives display a significantly lower stability than the wild-type plasmid (Futcher and Cox, 1983). Moreover, many of these plasmids can only be maintained using 2 pm as helper plasmid. To study the effect(s) of insertions into nonessential sequences ('silent regions') of the plasmid, we introduced either foreign DNA, containing bacterial and yeast sequences (Yips) or yeast DNA (URA3) in the SnaBl site, located at position 3609 (Hartley and Donelson, 1980)between the ARS and STB sequence and several other foreign DNA fragments into the Xmalll site located at position 5425, between the REP2 and FLP ORF. The characteristics of these 2 pm derivatives are presented in this paper. MATERIALS AND METHODS Strains and media The strains we used for our experiments are described in Table 1. S. cerevisiae was cultivated

Unilever Research Laboratories, Vlaardingen

non-selectively on YEPD containing 1% yeast extract (Difco), 2% bactopeptone (Difco) and 2% glucose. Selection for uracil proficiency was performed on YNB medium containing 0.67% yeast nitrogen base (Difco), 2% glucose and the required amino acids. Escherichia coli cells were grown in LB supplied with the required antibiotics. D N A techniques and plasmid constructions DNA manipulations were performed as described by Maniatis et al. (1989) unless stated otherwise. Yeast was transformed using the LiAc method (It0 et al., 1983) modified by Kuo and Campbell (1983). DNA was isolated from yeast by the method of Cryer et al. (1 975) and total RNA was isolated from yeast spheroplasts according to the protocol of Sripati (1978). For Northern analysis, RNA was transferred to Genescreen Plus nylon membranes (Dupont) and hybridizations were conducted according to the specifications of the manufacturer. Probes were labeled by random oligo primer extension (Feinberg and Vogelstein, 1983) using [a-32P]dCTP(Amersham). The 2 pm plasmid was isolated from MG171 and derivatives were constructed as shown in Figure 1. Plasmids pJB550 and pJB560 are derivatives of pJB501 and pJB502 respectively in which the 2 kb BgZI fragment is deleted. Hence both antibiotic markers are lost and therefore the plasmids were recovered in the E . coli pyrF strain (KA723) using uracil selection. For the construction of pJB5OO and pJB5OOR (Figure 2), we used double homologous in vivo recombination, based on the method of Ma et al. (1987). The linear EcoRI-SspI fragment of pJB162 was transformed to the S. cerevisiae cir' strain MG228 selecting for URA'. The recombinant plasmid was shuttled back to E. coli (DHI)

349

DNA INSERTIONS OF SACCHAROMYCES CEREVISIAE

(PISn)

Figure 1. Schematic representation of the various vectors. All vectors contain the total 2 pm sequence. In the 2 pm parts of the molecules the STB sequence and the origin of replication are indicated and the dashed boxes represent the inverted repeats. The yeast LIRA3 gene is the chromosomal 1170 bp fragment. Plasmids pJB501 and pJB502 were constructed by ligation of PvuII-digested Yip5 and SnaBI-digested 2 p n DNA. Deletion of the 2 kb Bgn restriction fragment from pJB501 and pJB502 resulted in pJB550 and pJB560 respectively. The Bgn sites were made blunt by T4 polymerase. Plasmid pJB506 is a derivative of pJB502 containing a 2 kb NotI fragment including a (Tn9) chloramphenicol-resistance cassette bordered by a polylinker sequence in the XmaIII site. The plasmid pJB5062 is a derivative of pJB506 in which the chloramphenicol-resistancemarker was removed using the two bordering Sac1 sites, resulting in a 110 bp insertion in the XmaIII site. Bacterial markers: Ap, ampicillin, Tc, tetracycline; Cam, chloramphenicol resistance. The indicated symbols for restriction enzymes: B, Bgn; N, NorI; P, PvuII; Sa, Sad; Sn, SnaBI; X, XmaIII. The plasmids are not drawn to scale.

selecting for kanamycin resistance. The bacterial sequence of pJB563 was deleted by removing the NotI fragment, after which the plasmid was reintroduced into the yeast cir" recipient selecting for uracil proficiency. Stability measurement

For stability experiments colonies were resuspended in water and a minimum of 100 single cells were put onto a selective plate using micromanipulation. The remaining cell suspension was used to inoculate a non-selective 10 ml YEPD culture and

incubated overnight at 30°C (approx. ten generations). A sample of the overnight culture was taken and put on a selective plate and another sample was diluted 1000-fold into fresh YEPD (10 ml) and incubated for ten generations. This procedure was repeated until the required number of generations was obtained. Colonies were counted after 3 days of incubation at 30°C. By examining the selective agar plates under the microscope, cells without plasmid (colonies varying from 10 to 1500 cells) could be distinguished from inviable cells (single cells). The plasmid maintenance was calculated as the number of cells containing plasmid divided by the

J. F. M. BIJVOETEr AL.

350

E

IE/Snl

..3%

4

\

i I

!

B

=s

pJ8500lRI 76 OR1

\

OR1

I

I€/%)

Figure 2. Homologous in vivo recombination was used for the construction of pJB5OO and pJB5OOR. pUC sequences are indicated as double-filled lines, pHSS6 sequences as double open lines and 2 pm sequences as single lines. The (IRA3 sequence is the I170 bp Hind111 chromosomal fragment. In the 2 pm sequence the STB region and the origin of replication (ORI)as well as the inverted repeats (boxes) are indicated. The 2240 bp EcoRI 2 pm fragment of pCV20 was subcloned into pUC7. Plasmid pJBl61, a pHSS6 derivative containing the I170 bp Hind111 yeast (IRA3 gene, was linearized at the EcoRl site, made blunt with Klenow and introduced into the unique 2 pm SnuBI site resulting in pJB162. This plasmid was linearized with Sspl and EcoRI and introduced into a S. cercvisioecir ' strain. Double homologous recombination with the endogenous 2 pm plasmid resulted in pJBS63, a shuttle vector containing the total 2 pm sequence. Deletion of the bacterial sequence by restriction with Nor1 resulted in a 2 pm-based yeast vector containing the yeast (IRA3 gene bordered by a small I 10 bp polylinker sequence. The bacterial resistance genes are symbolized as follows: Ap, ampicillin; Kn, kanamycin. Restriction sites are indicated as: E, EcoRI; H, HindlII; N, Norl, Sn, SnuBI; Ss, SspI. Plasmid molecules are not drawn to scale.

total number of viable cells and the number of generations of growth in the absence of selection pressure.

RESULTS AND DISCUSSION We first assessed the influence of insertion of Yip5 into the SnaBI site of 2 pm, located at position 3609 between the ARS and STB sequences, on plasmid stability. This site was selected using the following criteria: (1) the site had to be unique and (2) the site had to be located in a non-essential region of the molecule. To select for the presence of this 2 p m derivative in yeast, we used the URA3 gene which is in the Yip5 sequence. Several methods can be used to determine plasmid loss from cultures growing

under non-selective conditions (Futcher and Cox, 1983; Murray and Szostack, 1983). We used micromanipulation as described in Materials and Methods to detect the loss of the URA3 marker during cultivation on non-selective medium. This enabled us to follow the growth of singlecells, taken from non-selective grown cultures, which were directly placed onto selective plates. Introduction of Yip5 in the unique SnaBI of 2 pm resulted in plasmids pJB501 and pJB502 (Figure 1) which only differ in the orientation of the Yip5 insert. Figure 3 shows the results of a typical stability experiment carried out with these two plasmids. In a colony pregrown on a selective plate (generations=O) a fraction of uracil-deficient cells is already present. Hence the initial fraction of uracil-

35 1

DNA INSERTIONS OF SACCHAROMYCES CEREVISIAE

B % URA+

% URA+

100

0

5

10

15

20

25

30

35

40

generations

0

5

10

15

20

25

30

35

40

generations

Figure 3. Stability experiment for the pJB501 and pJB502 plasmids in MG171 cir" (loss frequency 0.6% and 0.2%) and MG17 I cir' (loss frequency 0.8% and 0.4%). Panel A shows the values obtained in the cir" strain; ( x ) pJBSO1, ( 0 )pJB502. Panel B shows the results in the cir' strain, symbols are the same as in panel A. Error bars indicate the variation of the average value for at least two independent experiments.

proficient cells is never 100%. Apparently during colony formation even under selection for URA+ some of the cells have lost the plasmid. The two plasmids can be stably maintained without a 2pm helper plasmid. However the plasmid loss as determined for the pJB501 and pJB502 plasmids in a cir" host, 0.6% and 0.2% per cell per generation respectively, is significantly higher than for the wild-type 2 pm molecule: 0.01 YOper cell per generation (Futcher, 1986; Mead et al., 1986). We also see an effect of the cir genotype of the host on plasmid stability. Probably due to plasmid incompatibility, the frequency of loss is slightly higher in a cir' host and is 0.8% and 0.4% per cell per generation for pJB501 and pJB502 respectively. The rate of plasmid loss under non-selective conditions, as measured for both recombinant plasmids, even in a cir" host will generate 2&60% of plasmid-free cells respectively upon prolonged growth for approximately 100 generations. From the difference in the frequency of plasmid loss between pBJ501 and pJB502 it follows that the plasmid stability is influenced by the orientation of the

inserted DNA. Possibly the instability is caused by transcription originating from the bacterial sequence and proceeding into the 2 pm part of the molecules. Both Murray and Cesarini (1986) and Kikuchi (1983) showed that the STB region has to be free of transcription to ensure proper partitioning. In yeast, transcripts originating from pBR322 sequences have been observed by several investigators (Marcynski and Jaehning, 1985; Ehrhart and Hollenberg, 1981). Figure 4 shows a Northern analysis of total RNA isolated from pJB501 and pJB502 containing yeast cells using the 302 bp Hpd-AvaT fragment of 2 pm which covers the largest part of the 62 bp repeat region of STB as a probe. The pJB5Ol clone shows a strong signal, whereas for the pJB502 clone no transcript is observed. For the wild-type 2 pm plasmid (lane 2) no signal can be seen, even after longer exposure (panel B). The difference in transcription for the plasmids pJB501 and pJB502 was also observed when the cir+ derivative strain was used as host (results not shown). Obviously the wild-type 2 pm plasmid has no influence on transcription of the

352

J . F. M. BIJVOET E T A L .

still has a lower stability than the wild-type plasmid. A possible explanation for this phenomenon might be that the promoter activity originating from the fragment in this case directs transcription towards the 2 pm origin of replication and thus might affect plasmid stability. In an attempt to locate the origin of this transcription we deleted a 2 kb Bgn fragment of the Yip5 sequence covering the bacterial ampicillin and tetracycline promoters (Figure I). This had neither an effect on the transcription over the STB sequence (Figure 4, lanes 5, 6) nor on the stability of the resulting plasmids (pJB550, pJB560). Apparently the promoter causing the transcription over the STB sequence is not located on the 2 kb fragment that was deleted. Based on the fact that there is no significant difference in length of the transcript for pJB5OI and pJB550, we assume that the promoter activity is originating from a site located between the SnaBI/ PvuII junction and the proximal Bgfl site. However, we cannot exclude the possibility that the promoter was created by the fusion of yeast and bacterial sequences in the pJB501 plasmid. Clearly the transcription we detect differs from the transcriptional activity coming from pBR322 sequences described earlier (Marcynski and Jaehning, 1985; Ehrhart and Hollenberg, 1981). Because the nature of the inserted DNA could affect plasmid maintenance, we also determined the stability of 2 pm derivatives which contained the 1170 bp URA3 gene of S. cerevisiae in the SnaBI site. Plasmid pJB5OO constructed by double homologous recombination in vivo as described in Materials and Methods, contains the yeast URA3 gene bordered by a synthetic polylinker sequence of 1 10 bp. This plasmid displays a frequency of loss of approximately 1.5% per cell per generation in a cir" host which is significantly higher than for a Yips/ Figure 4. Northern analysis of total RNA, using the 0.3 kb SnaBI derivative. To examine whether the orientation of the URA3 gene has any influence on 2 pm HpaI/Aval fragment as probe to detect transcription over the STB region. Panels A and B show normal and prolonged plasmid stability, we reversed the gene but not exposure of PsrI/Hpal-hybridized filter, panel C shows a r R N A the polylinker sequence. The resulting plasmid control. Lane 1,MG171 cir0;lane2,MG171cir+;lane3,MG171 pJB5OOR has a frequency of plasmid loss of cir", pJB501; lane 4, MG171 cir", pJB502; lane 5, MG171 cir", approximately 1.0% per cell per generation. In pJB550; lane 6 , MG171 cir", pJB560. Each lane was loaded with 20 pg of total RNA. accordance with the results of Murray and Cesarini (1986) we find an effect of the orientation of the recombinant plasmids in this region. These results URA3 gene on 2 pm stability which again could be clearly indicate that the transcription originates the consequence of the direction of transcription. from the bacterial insert and is also restricted to one Recently Chinery and Hinchliffe (1989) found indeorientation. Most likely this transcription is the pendently that the SnaBI site of 2 pm can be used for cause of the difference in stability for the two recom- introduction of the yeast 1 170 bp URA3 DNA fragbinant plasmids. However the plasmid pJB502, ment. According to their results this insertion did which carries the insert in the opposite orientation, not affect the stable maintenance of the plasmid. In

DNA INSERTIONS OF SACCHAROMYCES CEREVISIAE

their view the SnaBI site could therefore be very promising for heterologous gene expression in yeast. However, we found that the plasmid stability is lower after introduction of the URA3 gene bordered by a synthetic polylinker sequence. The plasmids pJB5OO and pJB5OOR are at least two orders of magnitude less stable that the wild-type 2 pm. Transcription arising from the polylinker sequences bordering the URA3 gene could be the cause of the observed instability. However, we could not detect any transcription over the STB region. Furthermore pJB502 does not contain this polylinker sequence and this plasmid is significantly less stable than the ones described by Chinery and Hinchliffe (1989). The discrepancy with the results of Chinery might be due to differences in the genetic background of the strains used. In the case of the SnaBI site transcription originating from the insert has an influence on plasmid stability. Therefore any foreign DNA inserted into the SnaBI site should contain active yeast transcription termination signals. Since plasmids containing either a complete, a partially deleted (minus 2 kb) or an enlarged (plus 1.7 kb, results not shown) Yip5 insertion show no significant difference in stability, we have to conclude that disturbance of 2 pm plasmid maintenance does depend on the sequence or structure rather than on the size of the inserted DNA. We also investigated the effect on plasmid stability of introduction of DNA sequences into the XmaIII site located between the REP2 and FLP ORFs of plasmid pJB502. The effect on plasmid stability was monitored for two different insertions. One contained a DNA fragment of approximately 2 kb harbouring a chloramphenicol resistance marker (pJB506), the other a polylinker sequence of only 110 bp (pJB5062). The plasmid with the 2 kb insert in both possible orientations causes heterogenous colony morphology in a cir' host. Besides colonies with a normal smooth morphology, sectored or nibbled colonies are seen at a high frequency on YEPD plates. Analysis of these colonies shows that cells from smooth colonies have lost the plasmid, whereas the nibbled colonies contain cells with plasmid. Nibble phenotype has been described by Holm (1 982) as a consequence of a chromosomal mutation in the nib allelle and is only seen in the presence of 2 pm. Nibbled colonies would arise from a subfraction of atypical cells (Holm, 1982; Sweeney and Zakian, 1989) containing an unusually high amount of plasmid which causes a reduced division potential and can lead to cell death. Apparently the

353 nib mutation disturbs the copy number of control of the yeast plasmid. Sectored colonies were also described as a consequence of FLP overproduction (Reynolds et al., 1987) which results in an overamplification of 2 pm molecules. In our experiments with the plasmid containing a 2 kb insert in the XmaIII site, we found that the average plasmid contents of cells from sectored colonies is extremely high. This is illustrated by the fact that plasmid bands can easily be detected on ethidium bromidestained agarose gels of total DNA (results not shown). Total RNA blots probed with FLP indicate a strong increase in FLP messenger (Figure 5). Other plasmid-specific mRNA levels (REPZ, URA3 and pBR322), however, have increased as well. Apparently the insertion strongly disturbs the plasmid copy number control irrespective of the orientation. Furthermore we observed slow but gradually increasing division rates for these clones during stability experiments. After three subsequent 1000-fold dilutions and overnight growth the cultures grow at normal rates but from Figure 6 it can be seen that by that time the culture consists of only plasmid-free cells. Because in this case plasmid loss leads to a higher division rate during non-selective cultivation, the plasmid loss per cell per generation cannot be determined accurately but is in any case very high in comparison with pJB502. Since plasmids with extremely high copy numbers result in a reduced division potential of the host cells, nonselectivegrowth unintentionally selects for plasmidfree cells. Therefore the actual stability for pJB506 in a cir" host will be higher than presented in Figure 6. The destabilization was also independent of the orientation of the 2 kb insert (results not shown). A slow division rate, accompanied by heterologous colony morphology and high instability, was also observed when the a-galactosidase gene (2 kb) of Cyamopsis tetragonoloba (Overbeeke et al., 1989) was inserted. This suggests that the aberrent colony morphology is not due to a particular sequence. An insertion of a sequence of llObp in the XmaIII site (pJB5062) does not result in nibbled or sectored colonies and the associated high copy numbers. Northern analysis revealed levels of transcription of plasmid-encoded genes which are comparable to the wild-type, indicating a normal copy number for this plasmid. Apparently the introduction of 110 bp in the XmaIII site does not influence copy number control to a large extent although there is still a negative effect on plasmid stability (Figure 6). In comparison with the 2 kb insert, this plasmid is much more stable, although significantly

354

J. F. M. BIJVOET ETAL.

1 2 3 4 5 6 7 8

% URA+

A

B

0

C

D Figure 5 . Northern analysis of total RNA using different probes. Panel A, FLP; panel B, REP2; panel C, URA3; panel D, pBR322. Lanes 1,3,5,7, MG 228 cir" strain containing: no plasmid, pJB501, pJB502 and pJB506 respectively. Lanes 2,4, 6, 8, MG 228 cir+ strain containing: (wild-type). 2 pm, pJB501, pJB502 and pJB506 respectively. Each lane was loaded witn 20pgofRNA.

less so that the pJB502 plasmid. For our results we have to conclude that small insertions result in plasmid destabilization whereas large inserts influence copy number as well. The discrepancy between the small (1 10 bp) and the big (2 kb) insert in this site with respect to interference with copy number control is very intriguing and deserves further exploration. Clearly the XmaIII site cannot be used for

5

10

15

20

25

30

35

40

days (generations x 0.1) Figure 6. Stability experiment for XmaIII derivatives of pJB502, pJB506 (2 kb) and pJB5062 (1 10 bp) in MG228 cir". The stability of pJB502 (dashed line) is given as reference. Because of the change in growth rate during the experiments for the pJB506 plasmid, the time is indicated in days instead of generations. Symbols: ( 0 )pJB506, ( x ) pJB5062.

the introduction of additional DNA sequences without severe consequences for plasmid stability. As for the reason of this observed instability, analysis of the sequences of 2 p m which might interact with the REPl/REP2 complex revealed a consensus sequence element consisting of the nanonucleotide 5'-TGCATTTTT-3' and its complement 5'-AAAAATGCA-3' separated by 6 or 7 bp (Murray et al., 1987; Armstrong et al., 1988). This sequence element is found twice in the FLPIREP2 intergenic region. The XmaIII site is located in the 7 b p spacing region separating the double nonamer proximal to the FLP gene. Armstrong et al. (1988) suggested that the double nonamer sequence in the FLPIREPZ intergenic region is involved in the negative regulation of FLP expression by REPl/ REP2 repression. According to this model, FLP expression is determined by an active negative regulation target sequence and by the REP1 and REP2 protein levels in the cell. Our results show that insertions of 2 kb in the XmaIII site setting the double nonamer sequences very far apart have a

DNA

ixsmm)Ns OF SACCHAROMYCES

CEREVISIAE

strong influence on the copy number, while insertion o f 1 10 bp does not. Apparently the double nonamer can be separated for an additional 110 bp and still function in the negative regulation of the FLPgene. When. however. these sequences are separated b y 2 k b insertions. the negative control is completely impaired, leading t o increased copy numbers. Summarizing, it can be concluded that insertions made in the SnuBI site o f 2 pm situated in the silent region between the STB region a n d the O R I r e s u l t in plasmids which are rather stable a n d can be maintained without a 2 p m helper plasmid. T h e stability of these plasmids is still somewhat lower in comparison with the wild-type 2 p m , also when an almost complete yeast D N A insert was used. T h e reason is not evident but the instability might be due t o a very low a m o u n t of transcription over STB. In a n y case we showed that a detectable transcription over STB leads t o a higher instability. W h e n the STB region is protected on both sides by strong transcription termination signals, the SnuBI site is probably very suitable for cloning foreign D N A fragments of various size a n d source. T h e region surrounding the Xmulll site is very important f o r t h e plasmid copy n u m b e r regulation a n d stability and therefore not suitable for insertion o f additional D N A is a high plasmid stability is required. ACKNOWLEDGEMENTS This research was supported by the D u t c h Ministry o f Economic Affairs a n d the Unilever Research Laboratories in Vlaardingen (grant n u m b e r

ITP 680400). REFERENCES Armstrong. K . A.. Som, T., Volkert, F. C. and Broach, J . R. ( 1988). Regulation of yeast plasmid amplification. Cuncer Cells. Eukur,rotic D N A Replicution, vol. 6. Cold Spring Harbor Laboratories, New York, pp. 213 -223. Broach. J . R . and Hicks. J . B. (1980). Replication and recombination functions associated with the yeast plasmid 2 pm circle. CelI21,501-508. Broach. J . R. (1983). Construction of high copy yeast vcctors using 2-pm circlc sequences. Methods of En-,~molog~~ 101, 307-325. Cashmorc. A. M.. Albury. M. S.. Hadfield, C. and Mcacock. P. A. ( 1986). Genetic analysis of partitioning functions encoded by the 2 pm circle of Saccharomyes c~c~rc~i.sitre. !Mol. Gen. Genet. 203, I54 162. Chincry. S. A. and Hinchliffe. E. (1989). A novel class of vcctor for ycast transformation. Curr. Genetics 16,21 -25. Crycr. D. R.. Eccleshall, R. and Marmur, J. (1975). Isolation of Yeast DNA. ed. Prescott D. M. Acadcmic Prcss: New York. Methods Cell Biol. 12.3945.

355 Dobson, M. J., Yull. F. E., Molina, M., Kingsman, S. M. and Kingsman, A. J . (1988). Reconstruction of the yeast 2 pm plasmid partitioning mechanism. Nucleic A c i h Re.?. 16,7103-71 17. Ehrhart, E. and Hollenbcrg, C. P. (1981). Curing of Succharomyes cerevisiae 2-pm DNA by transformation. Curr. Genetics 3,83-89. Feinberg, A. P. and Vogelstein, B. (1983). A technique for radiolabcling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, &l3. Futcher, A. B. and Cox, B. S. (1983). Maintenance of the 2 pm circle plasmid in populations of Sacchn-om~ces cerevi.?iae.J . Bucteriol. 154.61 2-622. Futcher, A. B. (1986). Copy number amplification of the 2 pm circle plasmid of Suecharomyes cerevisiue. J . Theoretical Biol. 119, 197- 204. Futcher, A. B. (1988). The 2 p m circle plasmid of Saccharomyces cerevisiae. Yeast 4 2 7 - 40. Hartley, J. L. and Donelson, J . E. (1980). Nucleotide sequence of the yeast plasmid. Nature 286,860-865. Huberman, J . A., Spolita, L. D.. Nawotka. K . A,. ElAssouli, S. M. and Davis, L. R . (1987). The in VIW replication origin of the yeast plasmid. C ’ d l 51, 4 7 3 4 8 1. Holm, C. (1982). Clonal lethality caused by the yeast plasmid 2 pm DNA. Cell 29,585 -594. 110, H., Fukuda, Y.. Murata. K. and Kimurd. A. (1983). Transformation of intact ycast cells treated with alkali cations. J . Bucteriol. 153, 163- 168. Jayaram, M.. Sutton, A. and Broach, J . R. (1985). Properties of REP3: a cis acting locus required for stable propagation of the Saccharomyces cerevislue plasmid 2 pm circle. Mol. Cell. Biol. 5, 24662475. Kikuchi, Y. (1983). Ycast plasmid requires a cis acting locus and two plasmid proteins for its stable maintenance. CeII35.487493. Kuo, C. L. and Campbell, J. L. (1983). Cloning of Sacchuromjws cerevisiue DNA replication genes: Isolation of the C D C 8 gene and two gencs that compensate for the cdc8-I mutation. Mol. Cell. Biol. 3, 1730- 1737. Ma, H., Kunes, S.. Schatz, P. J . and Botstein, D. (1987). Plasmid construction by homologous recombination in yeast. Gene 58,2 10-2 16. Maniatis,T., Fritsch. E. F. andSambrook. J. (1989). M o l ecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Prcss, New York. Marcynski, G . T. and Jaehning. J . A. (1985). A transcription map of a yeast centromere plasmid: unexpected transcripts and altered gene expression. Nucleic Acids Re.?. 13,8487- 8506. McLeod, M.. Craft. S. and Broach. J. R. (1986). Identification of the crossovcr site during FLP-mediated recombination in the Saccharomyes cerevisiae plasmid 2 pm circle. Mol. Cell. Biol. 6, 3357 -3367. Mead. D. J., Gardner. D. C. J . and Oliver, S. G. ( I 986). The yeast 2 plasmid: strategies for the survival of a selfish DNA: Mol. Gen. Gene& 205,41 7 4 2 I .

356 Murray, J. A. H. and Cesareni, G. (1986). Functional analysis of the yeast plasmid partition locus STB. EMBO J. 5,3391-3399. Murray, J. A. H., Scarpa, M., Rossi, N. and Cesareni, G. (1987). Antagonistic controls regulate copy number of the yeast 2 p plasmid. EMBO J. 13,42054212. Murray, A. W. and Szostack, J. W. (1983). Pedigree analysis of plasmid segregation in yeast. Cell 34, 961-970. Overbeeke, N., Fellinger, A. J., Toonen, M . Y., van Wassenaar, D. and Verrips, T. (1989). Cloning and nucleotide sequence of the a-galactosidase cDNA from Cyamopsis tetragonoloba (guar). Plant Molecular Biology 13,541-550. Reynolds, A. E., Murray, A. W. and Szostack, J. W. (1987). Roles of the 2 Fm gene products in stable maintenance of the 2pm plasmid of Saccharomyces cerevisiae. Mol. Cell. Biol. 7,3566-3573.

J. F. M. BIJVOET ETAL.

Som, T., Armstrong, K. A,, Volkert, F. C. and Broach, J. R. (1988).Autoregulationof2 pmcirclegeneexpression provides a model for maintenance of stable plasmid copy levels. CeN52,27-37. Sripati, C. E. (1978).Isolation, characterization and translation of mRNA from yeast. Methods in Cell Biology, vol. 20. Academic Press, New York, pp. 61-81. Sweeney, R. and Zakian, V. (1989). Extrachromosomal elements cause a reduced division potential in nib1 strains of Saccharomyces cerevisiae. Genetics 122, 749-757. Veit, B. E. and Fangman, W. L. (1988).Copy number and partition of the Saccharomyces cerevisiue 2 pm plasmid controlled by transcription regulators. Mol. Cell. Biol. 8,49494957. Volkert, F. C . and Broach, J. R. (1986). Site specific recombination promotes plasmid amplification in yeast. Cel146,541-550.

DNA insertions in the 'silent' regions of the 2 microns plasmid of Saccharomyces cerevisiae influence plasmid stability.

The 2 microns plasmid of the yeast Saccharomyces cerevisiae is in principle a suitable vector for expression of foreign genes, due to its high copy nu...
764KB Sizes 0 Downloads 0 Views