Current Genetics
Current Genetics 2, 193-200 (1980)
© bySpringer-Verlag 1980
Control of Recombination within and between DNA Plasmids of Saccharomyces cerevisiae Melanie J. Dobson, A. Bruce Futcher and Brian S. Cox Botany School, South Parks Road, Oxford, OX1 3RA, England
Summary. [2/am + ] and [2/am° ] yeast were transformed to stable leucine prototrophy with the hybrid yeast - E. coli plasmid, pJDB219. This plasmid contains the entire sequence of the endogenous 2/am yeast DNA plasmid in addition to the yeast nuclear L E U 2 + gene and the ColE1 derivative, pMB9. In the [2/am + ] transformants, a new wholly yeast L E U 2 + plasmid, pYX, was generated, probably by a recombination event between pJDB219 and 2 /am DNA. The plamid, pYX, in the absence of 2/am DNA, was found to exist in equimolar amounts of two forms, A and B, which probably arise by intramolecular recombination across the inverted repeat sequences of the 2 /am DNA portion of the plasmid, pJDB219 was found to require the presence of 2/am DNA to undergo this intramolecular recombination. The results suggest that 2 /am DNA and pYX code for a gene product required in this recombination event which pJDB219 cannot produce. Key words: Recombination - Plasmids - Transformation - Yeast.
Introduction Most strains of S a c c h a r o m y c e s cerevisiae contain 5 0 100 copies of a circular 1 /am double-stranded DNA plasrnid (Sinclair et al. 1967, Clark-Walker and MiNos 1974). This plasmid contains two non-tandem inverted repeated sequences (I.R.S.) each of about 0.64 kilobases (Livingston and Klein 1977). Intramolecular recombination across the I.R.S. has been proposed as the mechanism by which the two forms of the plasmid seen in vivo, A and B, are maintained in equimolar amounts (Guerineau et al. 1976). The A and B forms differ in the orientation of the non-repeated segments with respect to one another (Fig. 1).
2~mDNA
3.7
A
B
Fig. 1. Structure of the two forms of 2 t~m DNA. (v) EcoRI sites, (~7) PstI sites, (tvTr~) inverted repeat sequences. EcoRI fragment sizes are given in kilobases. The EcoRI site closest to the PstI site is referred to as the first of the two EcoRI sites
Several hybrid yeast - E. eoli plasmids containing the entire 2/am DNA molecule have been used to transform yeast (Beggs 1978, Struhl et al. 1979). In these cases, the recipients contained endogenous 2 /am DNA and the hybrid plasmids were reported to undergo intramolecular recombination across the 2/am DNA I.R.S. to give equimolar amounts of r h e A and B forms. Most workers, including Livingston and Klein (1977), have assumed that the general recombinational system of the yeast was responsible for this interconversion. However, Blanc et al. (1979) transformed a [2/am ° ] yeast recipient, DR19T7, and found that their hybrid plasmid showed very infrequent intramolecular recombination. In this study, the hybrid plasmid pJDB219 of Beggs (1978) was used to transform [2/am + ] and [2 #m ° ] yeast recipients. 2/am DNA was also introduced into [2/am ° ] transformants in an attempt to elucidate the role of the 2 #m DNA in this plasmid intramolecular recombination process and to examine more closely interactions between the hybrid and endogenous plasmids. 0172-8083/80/0002/0193/$01.60
194
M. Dobson et al.: Control of Recombination in Yeast Plasmids Transformation. Bacterial transformation was carried out by the procedure of Lederberg et al. (1974). Yeast transformation was carried out by the procedure of Hinnen et al. (1978).
oJDB219 -B
Restriction Endonueleases. EcoRI and BamHI were purchased from Boehringer Mannheim and digestion of DNA was carried out as recommended by the supplier.
•~
2-4
5.3 Fig. 2. Structure of the B form of the hybrid plasmid pJDB219. (v~ EcoRI sites, (o) BamHI site, (V--q) 2 #m DNA, (~27271)inverted repeat sequences of 2 #m DNA, ( ~ ) yeast chromosomal DNA carrying the LEU2 + gene, ( ~ ) pMB9 DNA. EcoRI fragment sizes are given in kilobases
Materials and Methods Yeast Strains. Two strains were used as recipients for transformation with the hybrid plasmid pJDB219, S. eerevisiae MC16 (a ade2-1 leu2 his4-712 SUF2 [2 #m+] from G. Fink) and S. earlsbergensis CBll/63 (a leu2 adel [2 #rn°]) derived from CBll (a adel [2 ttrn° ]) in this study by standard ethyl methanesulphonate mutagenesis. Other S. cerevisiae strains used were 701/2c (a ade2-1 leu2 his4-712 lys2 tad1-1 [2 tzrn+]), 701/2b (a leu2 his4712 tad1-1 karl-1 [2 tzm+]) 700/5e (a leu2 his4-712 karl-1 [2/~m+]) and 699/8e (a leu2 ade2-1 [2 #m+]). Bacterial Strain. E. coli JA221 (fecAl leuB6 trp~E5 h s d R hsdM+ lacY from J. Carbon) was used as a recipient for transformation by the hybrid plasmid pJDB219. Manipulations. Standard microbiological techniques were used in plating and culturing strains. Cultures containing recombinant plasmids were handled as outlined for containment category I under the UK Genetic Manipulation guidelinea. Chimaeric Plasrnid. The yeast-E, coli ptasmid pJDB219 (Beggs 1978) was used. pJDB219 consists of the entire 2 #m DNA plasmid with the ColE1 derivative pMB9 inserted at the second of the two EcoRI cut sites and a 1.1 kilobase yeast nuclear DNA fragment coding for the LEU2 gene inserted at the unique PstI restriction site (Fig. 2). In this study the B form was found to give EcoRI fragments of the following sizes, given in kilobases: 5.3, 3.7, 2.4, 0.7 as compared to the 5.3, 3.8, 2.5, 0.8 kilobase fragments reported by Beggs. Media. For yeast the general growth medium was YEPD (1% yeast extract, 1% peptone (Difco), 2% glucose) and the defined selective medium was YNB plus 1% glucose containing 0.67% yeast nitrogen base without amino adds (Difco). Appropriate amino acids were added to 20-30 /~g/ml. The sporulation medium was that of Hurst and Fogel (1964). The E. eoli general growth medium was L-broth (Lennox 1975) and the defined medium was M9 plus 0.2% glucose (Anderson 1946).
DNA Analysis. Yeast DNA was extracted by the method of Cryer et al. (1975) and E. coli DNA was extracted by the 'cleared lysate' procedure of Clewell and Helinski (1969). DNA samples were electrophoresed through 0.7 or 1% (w/v) agarose horizontal slab gels (11 x 17 x 0.6 cm) in 15 mM tris, 18 mM sodium dihydrogen orthophosphate, 0.5 mM EDTA, pH 7.8 at 190 V per 35 mA for 5 h or at 60 V per 15 mA for 16 h. Gels were stained by soaking in a solution of ethidium bromide (1 #g/ml) and photographed on a long wavelength ultraviolet transiUuminator. Molecular weights of unknown fragments of double-stranded DNA were calculated on the basis of a plot of molecular weight versus the reciprocal of electrophoretic mobility (distance the DNA species travelled from the origin in millimeters) for the EcoRI fragments of lambda and calf thymus DNA of known molecular weight (Thomas and Davis 1975, Philippsen et al. 1975) run in the same gel. Hybridization Conditions. DNA was transferred from agarose gels to nitrocellulose filtres (Millipore HAWP) by the procedure described by Cameron et al. (1979) except that the gels-were exposed to about 2,400 J- m- 2 of short wavelength ultraviolet light before the denaturation step. Transferred filters were presoaked as described by Denhardt (1966) before hybridization to radioactive probe DNA, carried out in 5 x SSC, 50% formamide at 42 °C for 20 h. 32p-labelled plasmid DNA for probes was prepared by the nick-translation method with E. coli DNA polymerase I as outlined by Schachat et al. (1974). After hybridization, the transfers were given two 3 h washes in 2 x SSC at 65 °C before being set up for autoradiography.
Results Transformation o f [2 # m +] Yeast pJDB219 in the B form was used to transform MC16 to stable leucine prototrophy (Futcher et al., in preparation). The hybrid plasmid was found to be present in the yeast transformants in an autonomously replicating form only, on the basis of four criteria: 1. When MC16 L E U + transformants were crossed to l e u 2 - strains of yeast, leucine prototrophy generally segregated in a 4 : 0 (LEU + : l e u - ) p a t t e r n in tetrads (Table 1). 2. Unrestricted DNA of the yeast transformants run on agarose gels showed a band which co-migrated with purified supercoiled pJDB219 DNA. 3. Unrestricted DNA from the yeast transformants could be used to transform leuB6 E. coli to leucine prototrophy with the DNA of the E. coli transformants showing a band on agarose gels which co-electrophoresed with supercoiled pJDB219 DNA. 4. Hybridization of 32p-labeUed pJDB219 DNA to nitrocellulose transfers of gels of BamHI restricted and
M. Dobson et al.: Control of Recombination in Yeast Plasmids Table 1. Segregation of leucine prototrophy when LEU + MC16 transformants are crossed to leu2- yeast strains. In each cross, two other nuclear markers were found to segregate 2: 2 in each tetrad. Germination was 40%-95% Transformant
MC16 L+2 MC15 L+3 MC16 L+8
Number of tetrads segregating LEU+:leu4:0
3:1
2:2
1:3
0:4
41 5 12
1 0 1
0 0 0
0 0 0
9 1 0
Table 2. The final plasmid content of several clones of pJDB219transformed MC16 after many generations of growth on selective medium Transformed clone
MC16 L+I MC16 L+2 MC16 L+2/5a MC16 L+3 MC16 L+8
Plasmidspresent 2 #m
pYX
pJDB219(A)pJDB219(B)
0 0 0 0 +
+ + 0 0 0
0 + + + +
0 + 0 + +
195 unrestricted DNA of the transformants did not pick up any bands homologous to pJDB2i9 other than those expected for the autonomously replicating hybrid plasmid and the leu2 chromosomal gene (Dobson et al., 1980). Five transformed clones of MC16 were grown for many generations on selective (leucine-minus) medium and at various intervals, the DNA of each of these clones was analysed on agarose gels. Several observations were made and are summarised below: 1. A new plasmid, pYX, intermediate in the size between 2/~m DNA and pJDB219 had arisen in two of the clones. 2. Of the clones examined, all but one had lost the endogenous 2/lm DNA in favour of pJDB219 or pYX (Dobson et al., 1980). 3. In one clone, MC16 L+ 1, pYX had displaced both pJDB219 and 2 ~tm DNA to become the only plasmid present in the cells and this clone still exhibits nonMendelian leucine prototrophy. Seven out of seven tetrads in a cross of MC16 L+I to a leu2- strain of yeast showed a 4 : 0 (LEU +:leu-) pattern of inheritance for leucine prototrophy. 4. Intramolecular recombination of the hybrid plasmid had occurred since A forms of pJDB219 could be recovered from the yeast transformants. In one clone, MC16 L+2/5a, not only 2/am DNA but the B form of pJDB219 had been lost leaving only the A form of pJDB219. Table 2 shows the final plasmid content of the five transformed clones of MC16 under study.
Restriction A nalysis o f p Y X
Fig. 3. Electrophoretic separation on a I% agarose gel of (1-3) EcoRI digested DNA and (4-5) uarestrieted DNA. 1, MC16 L+2/5a [pJDB219-A+];' 2 and 4, MC16 L+I [pYX+]; 3 and 5, pJDB219-B DNA. m = multimerie forms of the plasmids, rJ and sJ = relaxed circular and supercoiled forms of pJDB219. IX and sX = linear and supercoiled forms of pYX. The 0.7 kb EcoRI fragment of pJDB219 and pYX has run off the gel
The LEU + transformant MC16 L+I had lost both pJDB219 and 2:pm DNA in favour of pYX. EcoRI restricted total DNA of MC16 L+I was electrophoresed through an agarose gel to analyze this new plasmid. Unrestricted and EcoRI restricted pJDB219 DNA in the B form and EcoRI restricted MC16 L+2/5a DNA were run in parallel (Fig. 3). MC16 L+2/5a DNA with pJDB219 in the A form only gives four EcoRI fragments homologous to pJDB219 of 5.3, 4.0, 2.1 and 0.7 kilobase pairs. EcoRI digested pYX in MC16 L+ 1 gives bands homologous to pJDB2]9 which comigrate with all the EcoRI fragments of both the A and B forms of pJDB219 except for the 5.3 kb fragment which represents the pMB9 portion of the hybrid plasmid. The sum of the lengths of the five EcoRI fragments of pYX, 12.9 kb, is almost twice as large as the length of the linear unrestricted pYX, 6.8 kb, so it must be assumed that pYX, like the 2/lm DNA and pJDB219 can exist in two forms, A and B, which probably arise from intramolecular recombination across the I.R.S. of the 2/~m DNA portion of the plasmid. If one assumes that the 0.7 kb fragment is pres-
196
M. Dobson et al.: Control of Recombination in Yeast Plasmids pYX 0-7
0-7
E. coli to leucine prototrophy, pYX has only arisen when there was still 2/am DNA present in the MC16 transformed clone. Clones such as MC16 L+2/5a and MC 16 L+3 which lost 2/.tm DNA without the appearance of pYX, never acquired pYX. A possible origin of pYX is as a result of a recombination event between the hybrid plasmid pJDB219 and 2 #m DNA.
4.11
Transformation o f a [2/am °] Yeast A
B
Fig. 4. Proposed structure of the two forms of pYX. (v) EcoRI sites, (V----q) 2 #m DNA, (cr/777~) inverted repeat sequences of 2/~m DNA, (V----q)yeast chromosomal DNA carrying the LEU2+ gene. EcoRI fragment sizes are given in kilobases
pJDB219 in the B form was used to transform Saccharomyces carlsbergensis CB11/63 to stable leucine prototroplay (Futcher et al., in preparation). As was found by Blanc et al. (1979) for the transformation of a [2/am °] strain of yeast, the hybrid plasmid did not undergo intramolecular recombination. Only the B form of pJDB219 could be recovered from the transformants.
Attempts to Induce Intramolecular Recombination o f pJDB219 in a [2 gm ° ] [p YX°] LEU + Transformant
Fig. 5. Autoradiogram of EcoRI restricted DNA electrophoresed
through a 0.7% agarose gel where 32p-pJDB219 DNA has been hybridized to the nitrocellulose transfer of the gel. 1 and 2, DNA from two diploids of the cross CBll/63 L+I [pJDB219-B+] x MC16 [2 tzm+]; 3, DNA of a diploid fromthe cross CBll/63 L+I [pJDB219-B+] x MC16 L+1-39-8 [2 #rn° ] [pYX°]; 4, DNA of a diploid from the cross MC16 L+2/5a [pJDB219-A+] x CBll [2 ~m° ]. The 0.7 kb EcoRI fragment has been electrophoresed off the gel ent in a 2 : 1 ratio to each of the other four EcoRI fragments, Fig. 4 gives the proposed structures for the A and B forms of pYX. pYX is a completely yeast plasmid. It shows no hybridization to probe pMB9 DNA and unrestricted DNA from MC16 L+I will not transform leuB6
Two pJDB219-transformed clones of yeast, MC 16 L+2/5a and CB11/63 L + 1, were used for further analysis of the plasmid intramolecular recombination system. In both cases, pJDB219 was present in only one of its two possible forms, A in the former and B in the latter, and 2/~m DNA is absent. Several crosses were performed to test whether the presence of 2/~m DNA was indeed necessary for the conversion of pJDB219 from one form to another. These crosses are summarised in Table 3. Total DNA was extracted from the diploids of the first five crosses, EcoRI restricted and electrophoresed through an agarose gel. Autoradiograms of the gel tr/tcks for the first three diploids; MC16 L+2/5a [pJDB219-A +] x CBll [2/~m°], CB11/63 L + 1 [pJDB219-B + ] x MC16 L + 1-39-8 [2/~m ° ] [pYX °] and CBll/63 L+I [pJDB219-B +] x MC16 [2 gm + ], are shown in Fig. 5 where 32p-labelled pJDB219 DNA was hybridized to the nitrocellulose transfer of the gel. It can be seen that, when the LEU + transformant is crossed to a [2 #m °] strain, the pJDB219 does not undergo intramolecular recombination. In contrast, in the three crosses where CBll/63 L+I and MC16 L+2/5a were mated to [2/lm + ] strains of yeast, the diploids had four EcoRI fragments corresponding to the two forms of the 2/~m DNA introduced from the [2 #m + ] parent as well as the 4.0 and 2.4 kb EcoRI fragments which are characteristic of the A and B forms of pJDB219 respectively. The two forms of pJDB219 were not found to be present in equimolar amounts. The form of pJDB219 present originally was always present in greater quantities in the diploid.
M. Dobson et al.: Control of Recombination in Yeast Plasmids
197
Table 3. Plasmid content of parental yeast strains, diploid products, tetrads and haploid kar products for various crosses constructed in an attempt to analyze the plasmid intramolecular recombination event Cross
Plasrnid content Parental strain
Diploid
2 um
2 um
pJDB219 A
B
0
+
0
CBll
0
0
0
CBll/63 L+I x MC16 L+1-39-8a
0
0
+
0
0
0
CBll/63 L+I
0
0
-1-
MC16
+
0
0
CBll/63 L+I x 701/2c
0
0
+
+
0
0
MC16 L+2/5a x 699/8c
0
+
0
+
0
0
CBll/63 L+I
0
0
+
701/2b (karl-l)
+
0
0
MC16 L+2/5a x 700/5c (karl-l)
0
+
0
+
0
0
MC16 L+2/5a ×
×
pJDB219
Haploid/car product of nuclear genotype of transformed parent
Diploid after meiosis
2 ~m
2/~m
pJDB219
A
B
0
+
0
-
-
0
0
+
-
-
-I-
+
q-
--
+
+
-t-
--
+
+
+
--
X
A
B
0
+
+
0
+
0
pJDB219 A
B
0
+
0
0
0
+
a MC16 L+1-39-8 is a spontaneous Ieu2- clone of MC16L+1 [2 #m° ] [pYX+] which has lost the pYX plasmid (Dobson et al,, 1980)
To see whether the higher rate of recombination during meiosis could cause the intramolecular recombination of the hybrid plasmid in the absence o f 2/am DNA, the MC16 L+2/5a [pJDB219-A +] x C B l l [ 2 # m °] and CB11/63 L + 1 [pJDB219-B + ] x MC16 L + 1-39-8 [2/am °] [pYX °] mating mixes were streaked on sporulation medium and when more than 90% of the cells had sporulated, the ascogenous mixtures were grown in liquid YEPD. Total DNA was extracted, EcoRI restricted and electrophoresed through an agarose gel (Fig. 6). The ascogenous mixtures show p JDB219 only in the form present in the transformed parent so the higher rate of recombination during meiosis does not result in conversion of pJDB219 from one form to the other. The final two crosses in Tab. 3; CB 11/63 L+I [pJDB219B+] x 701/2b [2 #m+], MCl6 L+2/5a [pJDB219-A +] x 700/5c [2/am+], were attempts to introduce 2/am DNA
into the transformants by means o f a k a r - cross (Conde and Fink 1976). Yeast carrying the k a r - m u t a t i o n will mate and undergo cellular fusion with a haploid yeast of the opposite mating type but they are blocked in nuclear fusion. The resulting heterokaryons tend to produce haploid products with the nuclear genotype of one of the parents and cytoplasmic contributions from both parents. Haploid products with the nuclear genotype of the transformed parent were isolated. Total DNA was extracted from one clone of each cross, EcoRI restricted and electrophoresed through an agarose gel (Fig. 7). Neither haploid k a r product appeared to have acquired 2/am DNA from the k a r - parent and this is not unexpected since the 2/am DNA is transmitted to the haploid product of the non-nuclear genotype in only 50% o f the crosses (Livingston 1977). The two crosses gave different results with the A form of pJDB219 having been gener-
198
M. Dobson et al.: Control of Recombination in Yeast Plasmids
Fig. 6. Electrophoretic separation on a 1% agarose gel of EcoRI restricted DNA. 1, [2/~m+] yeast; 2, MC16 L+2/5a [pJDB219-A+]; 3, MCI6 L+3 [pJDB219-A+B]; 4, ascogenousmixture of the cross MC16 L+2/5a [pJDB219-A+] x CBll [2/~m°]; 5, pJDB219-B DNA. Fragment sizes of bands homologous to pJDB219 are indicated. Other bands in the yeast DNA digests represent ribosomal DNA
Fig. 7. Electophoretic separation of EcoRI restricted DNA on a 1% agarose gel (2, 3) and matching autoradiogram tracks (1, 4) where 32p-pJDB219 DNA has been hybridized to a nitrocellulose transfer of the gel. Haploid Par products with the nuclear genotype of the tranformed parent; 1 and 2, MC16 L+2/Sa [pJDB219-A+] x 700/5c [2gm+]; 3 and 4, CBll/63 L+I [pJDB219-B+] x 701/2b [2 gm+]
ated from the B form originally present in the CB11/63 L + 1 such that the two forms were now present in roughly equimolar amounts whereas in the MC16 L+2/5a haploid kar product, pJDB219 was still present only in the A form as in the parent transformant.
Discussion
Ratio o f A to B forms for Different Plasmids
2/am DNA appears always to exist in equimolar amounts of its A and B forms in vivo (Guerineau et al. 1976, Livingston and Klein 1977). Similarly in this study, pYX was always found in equimolar amounts of its A and B forms in the absence of 2/am DNA. In contrast, only one of the yeast LEU + transformants studied, MC16 L+3, was found to have both the A and B forms of pJDB219 in the absence of 2/am DNA and the two forms were not present in equimolar amounts (see Figure 6).
[2/am + ] and [2 gm ° ] yeast recipients were transformed with the hybrid yeast-E, coli plasmid pJDB219. In two of the MC 16 LEO+ transformants a new plasmid, pYX, containing no E. coli DNA appeared, pYX probably did not arise by a deletion ofpJDB219 since it was never generated in [2 #m ° ] yeast transformants. It was probably generated by recombination event between the 2/am DNA and pJDB219. The most likely mechanism by which this could have occurred was two sequential recombination events. The first recombination would produce a 2/.tm DNApJDB219 concatenate and the second would resolve the concatenate into pYX and a 2/am DNA moleucle carrying the pMB9 DNA insert (Fig. 8). Such a mechanism has been proposed by Guerineau et al. (1976) to account for the formation of 2/am DNA dimers and Storms et al. (1979) have reported concatenate formation between endogenous and hybrid plasmids where the latter contain 2 gm
M. Dobson et al.: Control of Recombination in Yeast Plasmids
d
QG 4t
e
Fig. 8. Proposed mechanism for production of pYXby sequential recombination events between pJDB219 and 2 ~m DNA. (v) EcoRI sites, (V---I) 2/~mDNA, ( ~ ) pMB9 DNA, (I----q) yeast chromosomal DNA carrying the LEU2 + gene. d, e = large and small EeoRI fragments of the 2/~m DNA. The initial recombination event between homologous regions of 2 ~m DNA and pJDB219 would produce a concatenate molecule. Intramolecular recombination between the direct repeats marked "e' would resolved the molecule into pYX and 2 t~m DNA with a pMB9 insert
DNA sequences. The 2/am DNA-pMB9 plasmid has not been observed in this study but it would be a cryptic plasmid in yeast and may have been quickly displaced by pYX of pJDB219 since the MC16 transformants were grown on selective medium favouring the maintenance of the L E U 2 + plasmids. Asymmetrical gene conversion of the type that operates at the mitochondrial loci co (Perlman and Birky 1974, Faye et al. 1979) and v a r l (Strausberg et al. 1978) could also account for the production of pYX. It has been shown in this study that 2/am DNA is necessary for the generation of pYX and for the intramolecular recombination of the pJDB219 hybrid plasmid. In contrast, pYX is able to undergo intramolecular recombination in the absence of 2/am DNA since it is found in equimolar amounts of A and B forms in [2/am °] yeast transformants. It is possible and consistent with the data that the insertion of the pMB9 DNA into the second of the EcoRI restriction sites of the 2/am DNA sequences in pJDB219 has interrupted the sequence of a gene which codes for a product needed for the plasmid intramolecular recombination which 2 #m DNA and pYX can still produce. Further support for this hypothesis comes from the evidence that hybrid y e a s t - E . c o l i plasmids similar to pJDB219 but with the bacterial DNA introduced at the first rather than the second EcoRI re-
199 striction site, are able to interconvert between the A and B forms in the absence of 2/am DNA (R. Storms, personal communication). The finding that one of the MC16 L E U + transformants, MC16 L+3, still maintains both the A and B forms of pJDB219 in the absence of 2/am DNA or pYX would seem to rule out the above hypothesis. However, in this transformant, the two forms are not present in equimolar amounts. This could be similar to the case reported by Livingston (1977), where strains of yeast maintained, in the same clone, two types of 2/am DNA with slightly different restriction patterns for at least 100 generations of growth although the two types were not maintained in equimolar ratios. Equimolar amounts indicate that interconversion between the two forms is frequent enough to maintain the equilibrium. The maintenance of different amounts of two plasmids in the yeast reflects the segregation pattern for the plasmid. For the 2/am DNA plasmid which exists in high copy number, it is clear that two types of the plasmid can be maintained in a single clone for many generations before segregation produces pure clones with only one type of plasmid. Another difficulty with the hypothesis that 2/am DNA codes for a gene product needed in intramolecular recombination of the plasmid, is that when CBll/63 L+I [pJDB219-B +] or MC16 L+2/5a [pJDB219-A +] were crossed to a [2/am +] strain of yeast, one would have expected equimolar amounts of the two forms of pJDB219 in the diploid once 2/am DNA was present. This did not seem to be the case. The input form of pJDB219, was always present in greater amounts than the other form. This may reflect heterogeneity of plasmid content among the diploid cells. 2/am DNA may have been rapidly displaces from the diploid such that only a small proportion of the cells contain 2/am DNA with pJDB219 in both its A and B forms while many of the cells contain only the form of pJDB219 originally present in the transformed parent. The k a r - cross, where an attempt was made to cytoduce 2/am DNA into CBll/63 L+I [pJDB219-B+], generated the A form of pJDB219 although no 2/am DNA was present in the haploid product. This result does not rule out the possibility that the 2/am DNA might be needed to interact physically in some way with the pJDB219 to produce the conversion from one form to another since the 2/am DNA may have been present in the haploid product but might have been displaced by the pJDB219 before the DNA extraction was made. However, it suggests that the 2/am DNA could code for a gene product needed for the intramolecular recombination which the pJDB219 cannot produce. The other k a r cross between MC 16 L+2/5a [pJDB219-A + ] and 700/5c also produced a haploid product with the nuclear genotype of the transformed parent which had not acquired
200 2 p m DNA but, in this case, the pJDB219 remained in the form o f the parental strain. More haploid products will have to be analyzed to see whether this is a consistent phenomenon. Gene products specific for a particular recombinational event are known in other genetic systems. Bacteriophage lambda requires the protein product o f the int gene for the site-specific recA-independent recombination which leads to integration (Gottesman and Weisberg 1971) and the transposons T n l and Tn3 code for a 19,000 dalton protein involved in the site-specific recAindependent recombination which leads to transposition (Arthur and Sherratt 1979). It would be o f great interest if the yeast 2/am DNA plasmid codes for a similar type of protein product. This raises the question o f the functional significance o f the two forms o f the 2/am DNA in vivo. Broach et al. (1979) suggested that the intramolecular recombination might act as a genetic switch allowing the transcription o f two different sets o f genes but to date, nothing is known about the functional role o f the 2 lain DNA in yeast. The finding that either the A or the B form o f pJDB219 can be maintained independently in the yeast transformant does suggest that replication and maintenance of the 2/am DNA plasmid are not dependent on such a genetic switch.
Acknowledgements. We wish to acknowledge the technical advice of Mr. Stephen Kearsey, gifts of radioactive nucleotides from Drs. I. W. Craig and A. J. Kingsman, and of the plasmid pJDB219, together with advice and encouragement, from Dr. Jean Beggs.
References Anderson EH (1946) Proc Natl Acad Sci USA 32:120-122 Arthur A, Sherratt D (1979) Mol Gen Genet 175:267-274 Beggs JD (1978) Nature 275:104-109 Blanc H, Gerbaud C, Slonimski P, Guerineau M (1979) Mol Gen Genet 176:335-342
M. Dobson et al.: Control of Recombination in Yeast Plasmids Broach J, Atkins J, MeGill C, Chow L (1979) Cell 16:827-839 Cameron J, Loh E, Davis R (1979) Cell 16:739-751 Clark-Walker G, MiNos G (1974) Eur J Biochem 41:359-365 Clewell DB, Helinski DR (1969) Proc Natl Acad Sei USA 62: 1159-1166 Conde J, Fink GR (1976) Proc Natl Acad Sci USA 73:36513655 Cryer DR, Eccleshall R, Marmur J (1975) Meth Cell Biol 12:3944 Denhardt DT (1966) Biochem Biophys Res Commun 23:641646 Dobson MJ, Futcher AB, Cox BS (1980) Curt Gen 2:201-205 Faye G, Dennebouy N, Kujawa C, Jacq C (1979) Mol Gen Genet 168:101-109 Gottesman M, Weisberg R (1971) In: Hershey AD (ed) The bacteriophage lambda. Cold Spring Harbor Laboratory, New York, p 113 Guerineau M, Grandehamp C, Slonimski P (1976) Proe Natl Acad Sci USA 73:3030-3034 Hinnen A, Hicks J, Fink G (1978) Proc Natl Acad Sci USA 75: 1929-1933 Hurst DD, Fogel S (1964) Genetics 50:335-358 Lederberg EM, Cohen SN (1974) J Bacteriol 119:1072-1074 Lennox ES (1975) Virology 1:190-206 Livingston D (1977) Genetics 86:73-84 Livingston D, Klein H (1977) J Bacteriol 129:472-481 Perlman PS, Birky CW (1974) Pro c Natl Acad Sci USA 71:4612 4616 Philippsen P, Streeck R, Zachau H, Muller W (1975) Eur J Biochem 57:55-68 Schachat F, Hogness D (1974) Cold Spring Harbor Symp Quant Biol 38:371-381 Sinclair J, Stevens B, Sanghavi P, Rabinowitz M (1967) Science 156:1234-1237 Storms R, McNeil J, Khandekar P, Parker J, An G, Friesen J (1979) J Bacteriol 140:73-82 Strausberg RL, Vincent RD, Perlman PS, Butow RA (1978) Nature 276:577-583 Struhl K, Stinchcomb D, Scherer S, Davis R (1979) Proc Natl Acad Sci USA 76:1035-1039 Thomas M, Davis R (1975) J Mol Biol 91:315-328 Communicated b y R. J. Schweyen Received June 19, 1980
Current Genetics 2, 193-200 (1980)
© bySpringer-Verlag 1980
Control of Recombination within and between DNA Plasmids of Saccharomyces cerevisiae Melanie J. Dobson, A. Bruce Futcher and Brian S. Cox Botany School, South Parks Road, Oxford, OX1 3RA, England
Summary. [2/am + ] and [2/am° ] yeast were transformed to stable leucine prototrophy with the hybrid yeast - E. coli plasmid, pJDB219. This plasmid contains the entire sequence of the endogenous 2/am yeast DNA plasmid in addition to the yeast nuclear L E U 2 + gene and the ColE1 derivative, pMB9. In the [2/am + ] transformants, a new wholly yeast L E U 2 + plasmid, pYX, was generated, probably by a recombination event between pJDB219 and 2 /am DNA. The plamid, pYX, in the absence of 2/am DNA, was found to exist in equimolar amounts of two forms, A and B, which probably arise by intramolecular recombination across the inverted repeat sequences of the 2 /am DNA portion of the plasmid, pJDB219 was found to require the presence of 2/am DNA to undergo this intramolecular recombination. The results suggest that 2 /am DNA and pYX code for a gene product required in this recombination event which pJDB219 cannot produce. Key words: Recombination - Plasmids - Transformation - Yeast.
Introduction Most strains of S a c c h a r o m y c e s cerevisiae contain 5 0 100 copies of a circular 1 /am double-stranded DNA plasrnid (Sinclair et al. 1967, Clark-Walker and MiNos 1974). This plasmid contains two non-tandem inverted repeated sequences (I.R.S.) each of about 0.64 kilobases (Livingston and Klein 1977). Intramolecular recombination across the I.R.S. has been proposed as the mechanism by which the two forms of the plasmid seen in vivo, A and B, are maintained in equimolar amounts (Guerineau et al. 1976). The A and B forms differ in the orientation of the non-repeated segments with respect to one another (Fig. 1).
2~mDNA
3.7
A
B
Fig. 1. Structure of the two forms of 2 t~m DNA. (v) EcoRI sites, (~7) PstI sites, (tvTr~) inverted repeat sequences. EcoRI fragment sizes are given in kilobases. The EcoRI site closest to the PstI site is referred to as the first of the two EcoRI sites
Several hybrid yeast - E. eoli plasmids containing the entire 2/am DNA molecule have been used to transform yeast (Beggs 1978, Struhl et al. 1979). In these cases, the recipients contained endogenous 2 /am DNA and the hybrid plasmids were reported to undergo intramolecular recombination across the 2/am DNA I.R.S. to give equimolar amounts of r h e A and B forms. Most workers, including Livingston and Klein (1977), have assumed that the general recombinational system of the yeast was responsible for this interconversion. However, Blanc et al. (1979) transformed a [2/am ° ] yeast recipient, DR19T7, and found that their hybrid plasmid showed very infrequent intramolecular recombination. In this study, the hybrid plasmid pJDB219 of Beggs (1978) was used to transform [2/am + ] and [2 #m ° ] yeast recipients. 2/am DNA was also introduced into [2/am ° ] transformants in an attempt to elucidate the role of the 2 #m DNA in this plasmid intramolecular recombination process and to examine more closely interactions between the hybrid and endogenous plasmids. 0172-8083/80/0002/0193/$01.60
194
M. Dobson et al.: Control of Recombination in Yeast Plasmids Transformation. Bacterial transformation was carried out by the procedure of Lederberg et al. (1974). Yeast transformation was carried out by the procedure of Hinnen et al. (1978).
oJDB219 -B
Restriction Endonueleases. EcoRI and BamHI were purchased from Boehringer Mannheim and digestion of DNA was carried out as recommended by the supplier.
•~
2-4
5.3 Fig. 2. Structure of the B form of the hybrid plasmid pJDB219. (v~ EcoRI sites, (o) BamHI site, (V--q) 2 #m DNA, (~27271)inverted repeat sequences of 2 #m DNA, ( ~ ) yeast chromosomal DNA carrying the LEU2 + gene, ( ~ ) pMB9 DNA. EcoRI fragment sizes are given in kilobases
Materials and Methods Yeast Strains. Two strains were used as recipients for transformation with the hybrid plasmid pJDB219, S. eerevisiae MC16 (a ade2-1 leu2 his4-712 SUF2 [2 #m+] from G. Fink) and S. earlsbergensis CBll/63 (a leu2 adel [2 #rn°]) derived from CBll (a adel [2 ttrn° ]) in this study by standard ethyl methanesulphonate mutagenesis. Other S. cerevisiae strains used were 701/2c (a ade2-1 leu2 his4-712 lys2 tad1-1 [2 tzrn+]), 701/2b (a leu2 his4712 tad1-1 karl-1 [2 tzm+]) 700/5e (a leu2 his4-712 karl-1 [2/~m+]) and 699/8e (a leu2 ade2-1 [2 #m+]). Bacterial Strain. E. coli JA221 (fecAl leuB6 trp~E5 h s d R hsdM+ lacY from J. Carbon) was used as a recipient for transformation by the hybrid plasmid pJDB219. Manipulations. Standard microbiological techniques were used in plating and culturing strains. Cultures containing recombinant plasmids were handled as outlined for containment category I under the UK Genetic Manipulation guidelinea. Chimaeric Plasrnid. The yeast-E, coli ptasmid pJDB219 (Beggs 1978) was used. pJDB219 consists of the entire 2 #m DNA plasmid with the ColE1 derivative pMB9 inserted at the second of the two EcoRI cut sites and a 1.1 kilobase yeast nuclear DNA fragment coding for the LEU2 gene inserted at the unique PstI restriction site (Fig. 2). In this study the B form was found to give EcoRI fragments of the following sizes, given in kilobases: 5.3, 3.7, 2.4, 0.7 as compared to the 5.3, 3.8, 2.5, 0.8 kilobase fragments reported by Beggs. Media. For yeast the general growth medium was YEPD (1% yeast extract, 1% peptone (Difco), 2% glucose) and the defined selective medium was YNB plus 1% glucose containing 0.67% yeast nitrogen base without amino adds (Difco). Appropriate amino acids were added to 20-30 /~g/ml. The sporulation medium was that of Hurst and Fogel (1964). The E. eoli general growth medium was L-broth (Lennox 1975) and the defined medium was M9 plus 0.2% glucose (Anderson 1946).
DNA Analysis. Yeast DNA was extracted by the method of Cryer et al. (1975) and E. coli DNA was extracted by the 'cleared lysate' procedure of Clewell and Helinski (1969). DNA samples were electrophoresed through 0.7 or 1% (w/v) agarose horizontal slab gels (11 x 17 x 0.6 cm) in 15 mM tris, 18 mM sodium dihydrogen orthophosphate, 0.5 mM EDTA, pH 7.8 at 190 V per 35 mA for 5 h or at 60 V per 15 mA for 16 h. Gels were stained by soaking in a solution of ethidium bromide (1 #g/ml) and photographed on a long wavelength ultraviolet transiUuminator. Molecular weights of unknown fragments of double-stranded DNA were calculated on the basis of a plot of molecular weight versus the reciprocal of electrophoretic mobility (distance the DNA species travelled from the origin in millimeters) for the EcoRI fragments of lambda and calf thymus DNA of known molecular weight (Thomas and Davis 1975, Philippsen et al. 1975) run in the same gel. Hybridization Conditions. DNA was transferred from agarose gels to nitrocellulose filtres (Millipore HAWP) by the procedure described by Cameron et al. (1979) except that the gels-were exposed to about 2,400 J- m- 2 of short wavelength ultraviolet light before the denaturation step. Transferred filters were presoaked as described by Denhardt (1966) before hybridization to radioactive probe DNA, carried out in 5 x SSC, 50% formamide at 42 °C for 20 h. 32p-labelled plasmid DNA for probes was prepared by the nick-translation method with E. coli DNA polymerase I as outlined by Schachat et al. (1974). After hybridization, the transfers were given two 3 h washes in 2 x SSC at 65 °C before being set up for autoradiography.
Results Transformation o f [2 # m +] Yeast pJDB219 in the B form was used to transform MC16 to stable leucine prototrophy (Futcher et al., in preparation). The hybrid plasmid was found to be present in the yeast transformants in an autonomously replicating form only, on the basis of four criteria: 1. When MC16 L E U + transformants were crossed to l e u 2 - strains of yeast, leucine prototrophy generally segregated in a 4 : 0 (LEU + : l e u - ) p a t t e r n in tetrads (Table 1). 2. Unrestricted DNA of the yeast transformants run on agarose gels showed a band which co-migrated with purified supercoiled pJDB219 DNA. 3. Unrestricted DNA from the yeast transformants could be used to transform leuB6 E. coli to leucine prototrophy with the DNA of the E. coli transformants showing a band on agarose gels which co-electrophoresed with supercoiled pJDB219 DNA. 4. Hybridization of 32p-labeUed pJDB219 DNA to nitrocellulose transfers of gels of BamHI restricted and
M. Dobson et al.: Control of Recombination in Yeast Plasmids Table 1. Segregation of leucine prototrophy when LEU + MC16 transformants are crossed to leu2- yeast strains. In each cross, two other nuclear markers were found to segregate 2: 2 in each tetrad. Germination was 40%-95% Transformant
MC16 L+2 MC15 L+3 MC16 L+8
Number of tetrads segregating LEU+:leu4:0
3:1
2:2
1:3
0:4
41 5 12
1 0 1
0 0 0
0 0 0
9 1 0
Table 2. The final plasmid content of several clones of pJDB219transformed MC16 after many generations of growth on selective medium Transformed clone
MC16 L+I MC16 L+2 MC16 L+2/5a MC16 L+3 MC16 L+8
Plasmidspresent 2 #m
pYX
pJDB219(A)pJDB219(B)
0 0 0 0 +
+ + 0 0 0
0 + + + +
0 + 0 + +
195 unrestricted DNA of the transformants did not pick up any bands homologous to pJDB2i9 other than those expected for the autonomously replicating hybrid plasmid and the leu2 chromosomal gene (Dobson et al., 1980). Five transformed clones of MC16 were grown for many generations on selective (leucine-minus) medium and at various intervals, the DNA of each of these clones was analysed on agarose gels. Several observations were made and are summarised below: 1. A new plasmid, pYX, intermediate in the size between 2/~m DNA and pJDB219 had arisen in two of the clones. 2. Of the clones examined, all but one had lost the endogenous 2/lm DNA in favour of pJDB219 or pYX (Dobson et al., 1980). 3. In one clone, MC16 L+ 1, pYX had displaced both pJDB219 and 2 ~tm DNA to become the only plasmid present in the cells and this clone still exhibits nonMendelian leucine prototrophy. Seven out of seven tetrads in a cross of MC16 L+I to a leu2- strain of yeast showed a 4 : 0 (LEU +:leu-) pattern of inheritance for leucine prototrophy. 4. Intramolecular recombination of the hybrid plasmid had occurred since A forms of pJDB219 could be recovered from the yeast transformants. In one clone, MC16 L+2/5a, not only 2/am DNA but the B form of pJDB219 had been lost leaving only the A form of pJDB219. Table 2 shows the final plasmid content of the five transformed clones of MC16 under study.
Restriction A nalysis o f p Y X
Fig. 3. Electrophoretic separation on a I% agarose gel of (1-3) EcoRI digested DNA and (4-5) uarestrieted DNA. 1, MC16 L+2/5a [pJDB219-A+];' 2 and 4, MC16 L+I [pYX+]; 3 and 5, pJDB219-B DNA. m = multimerie forms of the plasmids, rJ and sJ = relaxed circular and supercoiled forms of pJDB219. IX and sX = linear and supercoiled forms of pYX. The 0.7 kb EcoRI fragment of pJDB219 and pYX has run off the gel
The LEU + transformant MC16 L+I had lost both pJDB219 and 2:pm DNA in favour of pYX. EcoRI restricted total DNA of MC16 L+I was electrophoresed through an agarose gel to analyze this new plasmid. Unrestricted and EcoRI restricted pJDB219 DNA in the B form and EcoRI restricted MC16 L+2/5a DNA were run in parallel (Fig. 3). MC16 L+2/5a DNA with pJDB219 in the A form only gives four EcoRI fragments homologous to pJDB219 of 5.3, 4.0, 2.1 and 0.7 kilobase pairs. EcoRI digested pYX in MC16 L+ 1 gives bands homologous to pJDB2]9 which comigrate with all the EcoRI fragments of both the A and B forms of pJDB219 except for the 5.3 kb fragment which represents the pMB9 portion of the hybrid plasmid. The sum of the lengths of the five EcoRI fragments of pYX, 12.9 kb, is almost twice as large as the length of the linear unrestricted pYX, 6.8 kb, so it must be assumed that pYX, like the 2/lm DNA and pJDB219 can exist in two forms, A and B, which probably arise from intramolecular recombination across the I.R.S. of the 2/~m DNA portion of the plasmid. If one assumes that the 0.7 kb fragment is pres-
196
M. Dobson et al.: Control of Recombination in Yeast Plasmids pYX 0-7
0-7
E. coli to leucine prototrophy, pYX has only arisen when there was still 2/am DNA present in the MC16 transformed clone. Clones such as MC16 L+2/5a and MC 16 L+3 which lost 2/.tm DNA without the appearance of pYX, never acquired pYX. A possible origin of pYX is as a result of a recombination event between the hybrid plasmid pJDB219 and 2 #m DNA.
4.11
Transformation o f a [2/am °] Yeast A
B
Fig. 4. Proposed structure of the two forms of pYX. (v) EcoRI sites, (V----q) 2 #m DNA, (cr/777~) inverted repeat sequences of 2/~m DNA, (V----q)yeast chromosomal DNA carrying the LEU2+ gene. EcoRI fragment sizes are given in kilobases
pJDB219 in the B form was used to transform Saccharomyces carlsbergensis CB11/63 to stable leucine prototroplay (Futcher et al., in preparation). As was found by Blanc et al. (1979) for the transformation of a [2/am °] strain of yeast, the hybrid plasmid did not undergo intramolecular recombination. Only the B form of pJDB219 could be recovered from the transformants.
Attempts to Induce Intramolecular Recombination o f pJDB219 in a [2 gm ° ] [p YX°] LEU + Transformant
Fig. 5. Autoradiogram of EcoRI restricted DNA electrophoresed
through a 0.7% agarose gel where 32p-pJDB219 DNA has been hybridized to the nitrocellulose transfer of the gel. 1 and 2, DNA from two diploids of the cross CBll/63 L+I [pJDB219-B+] x MC16 [2 tzm+]; 3, DNA of a diploid fromthe cross CBll/63 L+I [pJDB219-B+] x MC16 L+1-39-8 [2 #rn° ] [pYX°]; 4, DNA of a diploid from the cross MC16 L+2/5a [pJDB219-A+] x CBll [2 ~m° ]. The 0.7 kb EcoRI fragment has been electrophoresed off the gel ent in a 2 : 1 ratio to each of the other four EcoRI fragments, Fig. 4 gives the proposed structures for the A and B forms of pYX. pYX is a completely yeast plasmid. It shows no hybridization to probe pMB9 DNA and unrestricted DNA from MC16 L+I will not transform leuB6
Two pJDB219-transformed clones of yeast, MC 16 L+2/5a and CB11/63 L + 1, were used for further analysis of the plasmid intramolecular recombination system. In both cases, pJDB219 was present in only one of its two possible forms, A in the former and B in the latter, and 2/~m DNA is absent. Several crosses were performed to test whether the presence of 2/~m DNA was indeed necessary for the conversion of pJDB219 from one form to another. These crosses are summarised in Table 3. Total DNA was extracted from the diploids of the first five crosses, EcoRI restricted and electrophoresed through an agarose gel. Autoradiograms of the gel tr/tcks for the first three diploids; MC16 L+2/5a [pJDB219-A +] x CBll [2/~m°], CB11/63 L + 1 [pJDB219-B + ] x MC16 L + 1-39-8 [2/~m ° ] [pYX °] and CBll/63 L+I [pJDB219-B +] x MC16 [2 gm + ], are shown in Fig. 5 where 32p-labelled pJDB219 DNA was hybridized to the nitrocellulose transfer of the gel. It can be seen that, when the LEU + transformant is crossed to a [2 #m °] strain, the pJDB219 does not undergo intramolecular recombination. In contrast, in the three crosses where CBll/63 L+I and MC16 L+2/5a were mated to [2/lm + ] strains of yeast, the diploids had four EcoRI fragments corresponding to the two forms of the 2/~m DNA introduced from the [2 #m + ] parent as well as the 4.0 and 2.4 kb EcoRI fragments which are characteristic of the A and B forms of pJDB219 respectively. The two forms of pJDB219 were not found to be present in equimolar amounts. The form of pJDB219 present originally was always present in greater quantities in the diploid.
M. Dobson et al.: Control of Recombination in Yeast Plasmids
197
Table 3. Plasmid content of parental yeast strains, diploid products, tetrads and haploid kar products for various crosses constructed in an attempt to analyze the plasmid intramolecular recombination event Cross
Plasrnid content Parental strain
Diploid
2 um
2 um
pJDB219 A
B
0
+
0
CBll
0
0
0
CBll/63 L+I x MC16 L+1-39-8a
0
0
+
0
0
0
CBll/63 L+I
0
0
-1-
MC16
+
0
0
CBll/63 L+I x 701/2c
0
0
+
+
0
0
MC16 L+2/5a x 699/8c
0
+
0
+
0
0
CBll/63 L+I
0
0
+
701/2b (karl-l)
+
0
0
MC16 L+2/5a x 700/5c (karl-l)
0
+
0
+
0
0
MC16 L+2/5a ×
×
pJDB219
Haploid/car product of nuclear genotype of transformed parent
Diploid after meiosis
2 ~m
2/~m
pJDB219
A
B
0
+
0
-
-
0
0
+
-
-
-I-
+
q-
--
+
+
-t-
--
+
+
+
--
X
A
B
0
+
+
0
+
0
pJDB219 A
B
0
+
0
0
0
+
a MC16 L+1-39-8 is a spontaneous Ieu2- clone of MC16L+1 [2 #m° ] [pYX+] which has lost the pYX plasmid (Dobson et al,, 1980)
To see whether the higher rate of recombination during meiosis could cause the intramolecular recombination of the hybrid plasmid in the absence o f 2/am DNA, the MC16 L+2/5a [pJDB219-A +] x C B l l [ 2 # m °] and CB11/63 L + 1 [pJDB219-B + ] x MC16 L + 1-39-8 [2/am °] [pYX °] mating mixes were streaked on sporulation medium and when more than 90% of the cells had sporulated, the ascogenous mixtures were grown in liquid YEPD. Total DNA was extracted, EcoRI restricted and electrophoresed through an agarose gel (Fig. 6). The ascogenous mixtures show p JDB219 only in the form present in the transformed parent so the higher rate of recombination during meiosis does not result in conversion of pJDB219 from one form to the other. The final two crosses in Tab. 3; CB 11/63 L+I [pJDB219B+] x 701/2b [2 #m+], MCl6 L+2/5a [pJDB219-A +] x 700/5c [2/am+], were attempts to introduce 2/am DNA
into the transformants by means o f a k a r - cross (Conde and Fink 1976). Yeast carrying the k a r - m u t a t i o n will mate and undergo cellular fusion with a haploid yeast of the opposite mating type but they are blocked in nuclear fusion. The resulting heterokaryons tend to produce haploid products with the nuclear genotype of one of the parents and cytoplasmic contributions from both parents. Haploid products with the nuclear genotype of the transformed parent were isolated. Total DNA was extracted from one clone of each cross, EcoRI restricted and electrophoresed through an agarose gel (Fig. 7). Neither haploid k a r product appeared to have acquired 2/am DNA from the k a r - parent and this is not unexpected since the 2/am DNA is transmitted to the haploid product of the non-nuclear genotype in only 50% o f the crosses (Livingston 1977). The two crosses gave different results with the A form of pJDB219 having been gener-
198
M. Dobson et al.: Control of Recombination in Yeast Plasmids
Fig. 6. Electrophoretic separation on a 1% agarose gel of EcoRI restricted DNA. 1, [2/~m+] yeast; 2, MC16 L+2/5a [pJDB219-A+]; 3, MCI6 L+3 [pJDB219-A+B]; 4, ascogenousmixture of the cross MC16 L+2/5a [pJDB219-A+] x CBll [2/~m°]; 5, pJDB219-B DNA. Fragment sizes of bands homologous to pJDB219 are indicated. Other bands in the yeast DNA digests represent ribosomal DNA
Fig. 7. Electophoretic separation of EcoRI restricted DNA on a 1% agarose gel (2, 3) and matching autoradiogram tracks (1, 4) where 32p-pJDB219 DNA has been hybridized to a nitrocellulose transfer of the gel. Haploid Par products with the nuclear genotype of the tranformed parent; 1 and 2, MC16 L+2/Sa [pJDB219-A+] x 700/5c [2gm+]; 3 and 4, CBll/63 L+I [pJDB219-B+] x 701/2b [2 gm+]
ated from the B form originally present in the CB11/63 L + 1 such that the two forms were now present in roughly equimolar amounts whereas in the MC16 L+2/5a haploid kar product, pJDB219 was still present only in the A form as in the parent transformant.
Discussion
Ratio o f A to B forms for Different Plasmids
2/am DNA appears always to exist in equimolar amounts of its A and B forms in vivo (Guerineau et al. 1976, Livingston and Klein 1977). Similarly in this study, pYX was always found in equimolar amounts of its A and B forms in the absence of 2/am DNA. In contrast, only one of the yeast LEU + transformants studied, MC16 L+3, was found to have both the A and B forms of pJDB219 in the absence of 2/am DNA and the two forms were not present in equimolar amounts (see Figure 6).
[2/am + ] and [2 gm ° ] yeast recipients were transformed with the hybrid yeast-E, coli plasmid pJDB219. In two of the MC 16 LEO+ transformants a new plasmid, pYX, containing no E. coli DNA appeared, pYX probably did not arise by a deletion ofpJDB219 since it was never generated in [2 #m ° ] yeast transformants. It was probably generated by recombination event between the 2/am DNA and pJDB219. The most likely mechanism by which this could have occurred was two sequential recombination events. The first recombination would produce a 2/.tm DNApJDB219 concatenate and the second would resolve the concatenate into pYX and a 2/am DNA moleucle carrying the pMB9 DNA insert (Fig. 8). Such a mechanism has been proposed by Guerineau et al. (1976) to account for the formation of 2/am DNA dimers and Storms et al. (1979) have reported concatenate formation between endogenous and hybrid plasmids where the latter contain 2 gm
M. Dobson et al.: Control of Recombination in Yeast Plasmids
d
QG 4t
e
Fig. 8. Proposed mechanism for production of pYXby sequential recombination events between pJDB219 and 2 ~m DNA. (v) EcoRI sites, (V---I) 2/~mDNA, ( ~ ) pMB9 DNA, (I----q) yeast chromosomal DNA carrying the LEU2 + gene. d, e = large and small EeoRI fragments of the 2/~m DNA. The initial recombination event between homologous regions of 2 ~m DNA and pJDB219 would produce a concatenate molecule. Intramolecular recombination between the direct repeats marked "e' would resolved the molecule into pYX and 2 t~m DNA with a pMB9 insert
DNA sequences. The 2/am DNA-pMB9 plasmid has not been observed in this study but it would be a cryptic plasmid in yeast and may have been quickly displaced by pYX of pJDB219 since the MC16 transformants were grown on selective medium favouring the maintenance of the L E U 2 + plasmids. Asymmetrical gene conversion of the type that operates at the mitochondrial loci co (Perlman and Birky 1974, Faye et al. 1979) and v a r l (Strausberg et al. 1978) could also account for the production of pYX. It has been shown in this study that 2/am DNA is necessary for the generation of pYX and for the intramolecular recombination of the pJDB219 hybrid plasmid. In contrast, pYX is able to undergo intramolecular recombination in the absence of 2/am DNA since it is found in equimolar amounts of A and B forms in [2/am °] yeast transformants. It is possible and consistent with the data that the insertion of the pMB9 DNA into the second of the EcoRI restriction sites of the 2/am DNA sequences in pJDB219 has interrupted the sequence of a gene which codes for a product needed for the plasmid intramolecular recombination which 2 #m DNA and pYX can still produce. Further support for this hypothesis comes from the evidence that hybrid y e a s t - E . c o l i plasmids similar to pJDB219 but with the bacterial DNA introduced at the first rather than the second EcoRI re-
199 striction site, are able to interconvert between the A and B forms in the absence of 2/am DNA (R. Storms, personal communication). The finding that one of the MC16 L E U + transformants, MC16 L+3, still maintains both the A and B forms of pJDB219 in the absence of 2/am DNA or pYX would seem to rule out the above hypothesis. However, in this transformant, the two forms are not present in equimolar amounts. This could be similar to the case reported by Livingston (1977), where strains of yeast maintained, in the same clone, two types of 2/am DNA with slightly different restriction patterns for at least 100 generations of growth although the two types were not maintained in equimolar ratios. Equimolar amounts indicate that interconversion between the two forms is frequent enough to maintain the equilibrium. The maintenance of different amounts of two plasmids in the yeast reflects the segregation pattern for the plasmid. For the 2/am DNA plasmid which exists in high copy number, it is clear that two types of the plasmid can be maintained in a single clone for many generations before segregation produces pure clones with only one type of plasmid. Another difficulty with the hypothesis that 2/am DNA codes for a gene product needed in intramolecular recombination of the plasmid, is that when CBll/63 L+I [pJDB219-B +] or MC16 L+2/5a [pJDB219-A +] were crossed to a [2/am +] strain of yeast, one would have expected equimolar amounts of the two forms of pJDB219 in the diploid once 2/am DNA was present. This did not seem to be the case. The input form of pJDB219, was always present in greater amounts than the other form. This may reflect heterogeneity of plasmid content among the diploid cells. 2/am DNA may have been rapidly displaces from the diploid such that only a small proportion of the cells contain 2/am DNA with pJDB219 in both its A and B forms while many of the cells contain only the form of pJDB219 originally present in the transformed parent. The k a r - cross, where an attempt was made to cytoduce 2/am DNA into CBll/63 L+I [pJDB219-B+], generated the A form of pJDB219 although no 2/am DNA was present in the haploid product. This result does not rule out the possibility that the 2/am DNA might be needed to interact physically in some way with the pJDB219 to produce the conversion from one form to another since the 2/am DNA may have been present in the haploid product but might have been displaced by the pJDB219 before the DNA extraction was made. However, it suggests that the 2/am DNA could code for a gene product needed for the intramolecular recombination which the pJDB219 cannot produce. The other k a r cross between MC 16 L+2/5a [pJDB219-A + ] and 700/5c also produced a haploid product with the nuclear genotype of the transformed parent which had not acquired
200 2 p m DNA but, in this case, the pJDB219 remained in the form o f the parental strain. More haploid products will have to be analyzed to see whether this is a consistent phenomenon. Gene products specific for a particular recombinational event are known in other genetic systems. Bacteriophage lambda requires the protein product o f the int gene for the site-specific recA-independent recombination which leads to integration (Gottesman and Weisberg 1971) and the transposons T n l and Tn3 code for a 19,000 dalton protein involved in the site-specific recAindependent recombination which leads to transposition (Arthur and Sherratt 1979). It would be o f great interest if the yeast 2/am DNA plasmid codes for a similar type of protein product. This raises the question o f the functional significance o f the two forms o f the 2/am DNA in vivo. Broach et al. (1979) suggested that the intramolecular recombination might act as a genetic switch allowing the transcription o f two different sets o f genes but to date, nothing is known about the functional role o f the 2 lain DNA in yeast. The finding that either the A or the B form o f pJDB219 can be maintained independently in the yeast transformant does suggest that replication and maintenance of the 2/am DNA plasmid are not dependent on such a genetic switch.
Acknowledgements. We wish to acknowledge the technical advice of Mr. Stephen Kearsey, gifts of radioactive nucleotides from Drs. I. W. Craig and A. J. Kingsman, and of the plasmid pJDB219, together with advice and encouragement, from Dr. Jean Beggs.
References Anderson EH (1946) Proc Natl Acad Sci USA 32:120-122 Arthur A, Sherratt D (1979) Mol Gen Genet 175:267-274 Beggs JD (1978) Nature 275:104-109 Blanc H, Gerbaud C, Slonimski P, Guerineau M (1979) Mol Gen Genet 176:335-342
M. Dobson et al.: Control of Recombination in Yeast Plasmids Broach J, Atkins J, MeGill C, Chow L (1979) Cell 16:827-839 Cameron J, Loh E, Davis R (1979) Cell 16:739-751 Clark-Walker G, MiNos G (1974) Eur J Biochem 41:359-365 Clewell DB, Helinski DR (1969) Proc Natl Acad Sei USA 62: 1159-1166 Conde J, Fink GR (1976) Proc Natl Acad Sci USA 73:36513655 Cryer DR, Eccleshall R, Marmur J (1975) Meth Cell Biol 12:3944 Denhardt DT (1966) Biochem Biophys Res Commun 23:641646 Dobson MJ, Futcher AB, Cox BS (1980) Curt Gen 2:201-205 Faye G, Dennebouy N, Kujawa C, Jacq C (1979) Mol Gen Genet 168:101-109 Gottesman M, Weisberg R (1971) In: Hershey AD (ed) The bacteriophage lambda. Cold Spring Harbor Laboratory, New York, p 113 Guerineau M, Grandehamp C, Slonimski P (1976) Proe Natl Acad Sci USA 73:3030-3034 Hinnen A, Hicks J, Fink G (1978) Proc Natl Acad Sci USA 75: 1929-1933 Hurst DD, Fogel S (1964) Genetics 50:335-358 Lederberg EM, Cohen SN (1974) J Bacteriol 119:1072-1074 Lennox ES (1975) Virology 1:190-206 Livingston D (1977) Genetics 86:73-84 Livingston D, Klein H (1977) J Bacteriol 129:472-481 Perlman PS, Birky CW (1974) Pro c Natl Acad Sci USA 71:4612 4616 Philippsen P, Streeck R, Zachau H, Muller W (1975) Eur J Biochem 57:55-68 Schachat F, Hogness D (1974) Cold Spring Harbor Symp Quant Biol 38:371-381 Sinclair J, Stevens B, Sanghavi P, Rabinowitz M (1967) Science 156:1234-1237 Storms R, McNeil J, Khandekar P, Parker J, An G, Friesen J (1979) J Bacteriol 140:73-82 Strausberg RL, Vincent RD, Perlman PS, Butow RA (1978) Nature 276:577-583 Struhl K, Stinchcomb D, Scherer S, Davis R (1979) Proc Natl Acad Sci USA 76:1035-1039 Thomas M, Davis R (1975) J Mol Biol 91:315-328 Communicated b y R. J. Schweyen Received June 19, 1980
