Oene, 1(1976) 33--47 © Elsevier/North-Holland Biomedical Press, Amsterdam -- Printed in The Netherlands

33

SYNTHESIS OF HIGH MOLECULAR WEIGHT POLYPEPTIDES IN ESC,H£RICHIA COLI MINICELLS DIRECTED BY CLONED 8.4C,CHAROMYCE8 C£R£VISIAE 2.#m DNA (Cloned yeast 2-#m DNA; expression in E. coli minicells) CORNELIS P. HOLLENBERG, BIRGIT KUSTERMANN-KUHN and HANS-DIETER ROYER

Max-Planck-lnstitut f~r Biologie, Abteilung Beermann 8pemannstrasse 34, D-7400 TF~bingen (Federal Republic of Germany) (Received September 10th, 1976) (Accepted September 24th, 1976 )

SUMMARY

The minicell-producing Escherichia coil strain P 678-54 was transformed with a series of defined PTY chimeric plasmids consisting of yeast 2-#m DNA and K coil plasmid pCR1. In minicells the integrated 2-#m DNA from yeast directed specifically the synthesis of six polypeptides with apparent molecular weights of 15 000,17 500, 20 000, 22 000, 37 000, and 48 000. The specificity of five other polypeptides, which cover a molecular weight range of 19 000 to 28 000, has not yet been established with certainty. Neither the orientation of the integrated DNA, nor the inversion which distinguishes the two structural forms of 2-#m DNA affected the polypeptides synthesized. However, integration at a given £coRI site appeared to be correlated with the absence of one particular polypeptide band; this suggests that at least one of these sites is located in an expressed region of the DNA. L~TRODUCTION

Yeast 2-#m DNA is a circular extra-chromosomal element of which 50-100 copies ate normally found in several strains o f S ~ c ~ r o m y c e s cerevisiae (Hollenberg et al., 1970; Guerineau et al., 1971). Cloning of ~#m DNA in Eschevich~ coli plasmid pCR1 showed the e~i~tence of two types of 2-#m monomers which are differentiated by an inversion of about 40% of the molecule (Hollenberg et al., 1976). ~ inversion is surrounded on both sides by the sequences of a0;19 #mlong inverted duplication which are spaced 0.67 #m and 0 . ~ #mapart, Reciprocal recombination between the sequenc~ of the inverted duplication leads to inversion of the enclosed DNA segments and

34

can thus giv~ rise to the two types of monomers (Hollenberg et al., 1976; Guerineau, 1~76; Hollenberg and Royer, 1976). As yet no functions for yeast 2-pm DNA have been discovered, though some genetics] data which suggest the involvement o f ~ D N A ~ multiple drug resistance, have been presented (Guerineau et aL, 1974. Griffith e t ~ : i 1975), • Our present work describes the expression of cloned 2.pm DNA in K coli miuicells. Integrated in plasmid pCR1, the yeast DNA directed the synthesis of discrete polypeptides with m o l e ~ a r weights of up to 48 000. These results suggest that E. coli minicells may be able to express yeast 2-pro DNA faithfully and may provide a system for further study of its gene products. These suggestions are supported by the finding that a cloned yeast DNA segment could complement an E. coli histidine auxotroph (Struhl et al., 1976). Whereas the yeast DNA directs the synthesis of high molecular weight polypepfides, mostly small polypeptides were generated in minicelis by cloned mouse mito. chondrial DNA (Chang et al., 1975). These were much smaller than the polypeptides synthesized by the m!tochondria in vitro. Our observations suggest that certain types of foreign genes can be properly expressed in minicells, i.e. possibly those of yeast 2-pro DNA, whereas others cannot. MATERIALSAND METHODS

DNA preparations and construction of chimeric plasmids Closed circular DNA from Saccharomyces cerevisiae H1 was isolatecl from protoplasts. Protoplasts were prepared according to a modification of the procedure of Kov~ et al. (1968), based on the method of Duell et al. (1964). I g of cells (late logarithmic phase) was washed successively ~.:ith cold H=O and 0.5 M sodium thioglycollate--0.1 M Tris (titrated to pH 9.3 with NaOH). The cells were suspended in 2.5 ml of the thioglycoL!ate medium and incubated for 30 rain at 30°C. The cells were then washed!twice with H20 and once with medium A (1.5 M sorbitol, 1 mM EDTA, 10raM citric acid, 10 mM K2HPO4, titrated to pH 5.8 with NaOH). After suspension in 1.25 ml medium A and addition of 35 mg helicase (Industrie Biologique, Gennevilliers, France), the cells were incubated at 30°C until 90% ofthe cells had been converted to protoplasts (30--60 min). To determine the percentage of conversion, the number of cells in a 25-fold dilution in H~O was compared to the number present in a 25-fold dilution in medium A. The protoplasts were pelleted by 10 mLn centrifugation at 2500 g and the supemataut was decanted carefully. This supernatant can be frozen at --20°C and after addition of 50% fresh helicase be used again to make protoplasts. The protoplast pellet was suspended gently in 10 vol. of medium B (1.5 M sorbitol, 2 ~ EDTA, pH 7.6)and c e n ~ g e d for 1 0 ~ at 2500 g, This wMh procedure was repeated twice, followed by a centrifugation for10 n a t 4200g. 2-#m DNA was isolated from ~ e d pr0toplas~by one o f t h e following procedures: (1)frv~tionation o f a t o ~ protoplast ly~t~ in two succmive .... CsCl-ethidium bromide equilibrium 8radients as previoualy described (Hollen-

35

berg et al., I970); (2) modified Hirt procedure (Hirt, 1967; Smith et al., 1971). Washed pro~plast s were suspended in medium B at a concentration of I g protoplasts per ml. After addition of 5 ml 0.1 M EDTA, the suspension was briefly vortexed and lysed by the addition of 6 ml 2% sodium dodecyl sulfate. Thelysate Was kept for 15 rain ~ r o o m temperature after Which 2 m! of 7 M CsCI ~ added and mixed in gently. The lysate was kept for 1.5 h at 0°C and was then centrifuged in a Spinco 60 Ti rotor at 30 000 rpm for 30 rain at 4°C. 10.7 g CsCI was dissolved in the supernatant, which was then kept for I h at 0°C. The lysate was centrifuged for 15 rain at 10 000 g and the liquid phase was collected from under the protein top layer. Ethidium bromide (10 mg/ml) was added to a final concentration of 1.5 mg/ml. The solution, which by now should have a density of 1.55 g/ml, was centrifuged in a Spinco 60 Ti rotor at 44 000 rpm for 42 h at 20°C. The superc~iled DNA band, clearly visible at 366 nm light, was collected. Plasmid pCR1 DNA (Covey et al., 1976) was isolated as closed circular DNA (Katz et al., 1973). The construction of yeast 2-#m DNA/pCR1 DNA chimeric plasmids and tr~msformation of E. coli 490 (recA-, rk-, ink-) have been previously described (Hollenberg et al., 1976) and were performed in general according to the procedure published by Cohen et al. (1972). These procedures were also used for transformation of the minicell producing strain K coil P 67854, recA +, rk ~, mk÷ (Adler et al., 1967). Restriction enzyme reactions £eoRI and HindIII restriction endonuclease reactions were performed as previously described (Hollenberg et al., 1976). Analysis of proteins synthesized in K coil miniceUs Minicells from E. coli P 678-54 which had been transformed with plasmid pCR1 or a chimeric plasmid were isolated from stationaw phase cul~.~res by fractionation on two successive sucrose gradients, according to the procedures described by Roozen et al. (1971). The minicells were suspended at an absorbance at 620 nm of 0.2 or 1.0 and incubated in a mineral salt solution (Curtiss, 1965) containing 0.5% glucose, L-leucine (6.1 #g/ml), [3H]leucine (50 #Ci/ml; 59.5 Cilmmole) and a mixture of other amino acids as described by Roozen et al. (1971). After incubation at 37°C for 40 min, cells were collected and suspended in 0.01 M Tris--HCl (pH 8.0). Lysozyme was added to a final concentration of I mg/ml, EDTA to a final 0.05 M and the suspension was made 0.5 mM in phenylmethylsulfonyl fluoride by the addition of 0.01 vol. of a 0.05 M solution in isopropanol. The suspension was frozen and thawed rapidly three times, then was lysed by the addition of 20% sodium dodecyl sulfate to a final concentration of 2% as described by Chang et al. (1975). Electrophoresis was carried out in 15% polyacrylamide--0.2% sodium dodecyl sulfate slab gels with a 5% spacer gel (Laemmli, 1970), as described by Studier (1973). Lysate samples were mixed with I vol. of sample buffer con-

~6

raining 0.1 M Tris--HCl ~H8), 2% sodium dodecyl sulfate, 2% mer~ptoe~auol and 20% giycerol~ and kept in a boiling water bath f o r 2 min. 20-40,#1 o f sample were added per gel slot, Electrophoresk ~ performed at 55 V f o r 12 h at 18°C. Gels were tJtemed in Coomassie B Blue (2.5 gfl) in45% ethanol --10% acetic acid(v/v) for 30 min and.d~tain"ed in 5%ethanol-~..§% acetic acid (v/v) at 65°C, After .electrophoresis ...... (,and s ~ ) t h e ~ a b gel ~ p r e pared for fluorography as described by Bonnet and Laskey (1974) and exposed to flash-activated Kodak RP Royal X-Omat film (Laskeyand Milig, 1975) at-70°C. Analysis of RNA synthesized in minicells Isolated minicells were incubated with [3H]uridine (100 pCi/ml; 47 Ci/ mmole) and, after 30 miu incubation at 37°C, RNA was isolated as described by Roozen et al. (1971). Total yeast DNA was digested with £coRI and electrophoresed in horizonted 0.7% agarose gels (Shinnick et al., 1975). The DNA was transferred to nitrocellulose filters and hybridized to [aH]RNA asdescribed by Southern (1975). The filter was impregnated with PPO and exposed to flash.activated Kodak RP Royal X-Omat film (Southern, 1975) at-70°C. Saturation hybridization to DNA immobilized on membrane filterswas performed by the method of Gillespie and Spiegelman (1965) as described previously (Hollenberg, 1976). EcoRI fragments of yeast 2.#m DNA were isolated by gel electtophoresis in 0.7% agarose (Aay and Borst, 1972) of an £coRI digest of chimeric plasmid PTY21 as described previously (Hollenberg et al., 1976). DNA was extracted from the excised gel bands by the freeze4queeze method of Thuring et al. (1975). Electron microscopy Heteroduplexes were prepared and spread as described by Davis et al. (1971). , RESULTS

Recombinant plasmids The recombinant plasmids which we have used in this study all contained plasmid pCR1 as a cloning vehicle, pCR1 is a ColE1 derived plasmid which carries a kanamycin resistance marker and has only one EcoRI and one HindlII site (Covey et al., 1976). The struct-dre of the two types, 14 and 23, of the 2-#m monomer present in the yeast cell is schematically presented in Fig.1. After denaturation and reannealing the DNA occurs p r e d o ~ a n t l y in ~1dumb-bell structure with a small ~ d a lar,~e single-stranded loop, ~ d a ouble.stranded stem. ~ s.truc~e m formed b y ~ renatumtion of the sequenc.es ~of the inverted dupncation(id ~ e n ~ ) ; B o t h ~ p a oz monomers contain two EcORI sites, RIA in the ls~ge ioop (~loop) ancl i

87

Type23 ~

H

Type14

I

~ H 1 B

Fig.1. (A) Schematic outline of double stranded yeast monomer type 28 (consisting of £coRI fragments 2 and 8) and t y p e 14 (conskting of £coRI fragments I and 4). The inversion, which differentiates type 14 from type 28, is indicated by the half crescent. The id sequences are black, Numbers 1--4 refer to the £coRI fragments; RIA and RIB refer to E¢oRI sites; H1, H2 and H8 to HJndHI sites. (B) Schematic diagram_a of the reannealed dumb-bell structure of each type of monomer.

RIB in the small loop (S-loop) (Hollenberg et al., 1976). For the insertion in pCR1 we used a limited £¢oRI digest of 2-~m DNA. Each of the EcoRI sites, therefore, could have formed the link between the two DNA segments. The existence of two types of monomer each containing two possible joining sites and the two different orientations in which each yeast DNA segment can be inserted, give a total number of eight different chimeric plasmids. The analysis of each unknown cloned chimeric plasmid proceeded as follows. (1) The molecular weight of the plasmid was determined by electrophoresis of the su0ercofled DNA in 0.5% agarose plus 100 ~g ethidium bromide (Aay and Borst, 1972). The plasmids here analyzed all had a molecular weight of approx. 13 • 10 ~, as would be expected for the sum of plasmid pCRI and one molecule of 2 #m length. (2) Analysis of the EcoRI fragments indicated which type of monomer had been inserted (see Fig.2c and d for PTY7§ and 81 analysis). PTY21 and PTYSI contained fragments 2 and 3 (type 23) whereas PTY39 and "/5 contained fragments I and 4 (type 14). (3) Integration site: as we showed earlier (Hollenberg et al., 1976), heteroduplexes between a chimeric plasmid and pCR1 reveal immediately at which EcoRI site the yeast DNA has been inserted. The id sequences of the 2-~m DNA segment in the heteroduplex reassociate and give rise to the typical dumbben structure with the S- and IAoop. Fig.3 shows the heteroduplexes of some analyzed plssmids, (4) Orientation: Each yeast monomer is cleaved three

38

~ o

Mk

Fig.2. Agarosc gel analysis of restriction fragments of PTY75 and PTYS1 DNA before and after replication in minicell-producing E. co|i P678-§4. (a) £coRI digest of PTY75 isolated from mini(PTY75). (b) EcoRI digest of PTY81 from mini(PRY81). (e) £coRI digest of PTY75. (d) EcoRI digest of PTY81. (e) £coRI digest of phage ~, DNA. (f) HindIII digest of PTY75. (g) HindIII digest of PTY75 isolated from mini(PTY7§). (h) Same as in gel g plus markers from gel e. (i) HindIII digest of PTY81. (j) HindIII digest of PTY81 isolated from mini(PTY81). (k) Same as in gel j plus markers from gel e. The two minor bands migrating between pCR1 and fragment I in gels a--d, are due to overdigestion and originate from the pCR1 DNA. The molecular weights of the HindIH fragments are listed in Table I. Electrophoresis was carried out in 0.7% agarose containing 0.5 ~g cthidium bromide per ml (Aay and Borst, 1972) at 3 V/cm at 18°C, as previously described (Hollenberg et al., 1976). times by endonuclease HindIII. Both E c o R I sites are located in close proximity to a H i n d I I I site (Fig.l). pCR1 has only one H i n d I I I site. The molecular weight~ of the longest HindIII fragments of a r e c o m b i n a n t plasmid, therefore, indicate the orientation of the inserted yeast segment. The structure of PTY 21, 39, 75 and 81 is shown schematically in Fig.4 and the analytical data of these plasmids are summarized in Table I. Type 23 in PTY21 and t y p e 14 in PTY39 are b o t h integrated at the same £ c o R I site, b u t with opposite orientation. The same is true for PTY7§ and 81, which are both joined at site RIA b u t in which HI and H3 are respectively the H i n d I I I sites closest to the pCR1 HindIII site.

39

Fig.3. Heteroduplexes between pCR1 and PTY plasmids. The integrated yeast 2-~m DNA strand hangs out in the typical dumb-bell structure. A tracing of the dumb-bell and the joining site with pCR1 is shown in the inserts. (a) pCR1/PTY39. (b) pCR1/PTY21. (c) pCR1/PTY75. (d) pCR1/PTY81. Bar is 0.5 ~m. TABLE I ANALYTICAL DATA OF PTY PLASMIDS Orient- Molecular ationa weight (. 10 "6)

Recombinant Monomer plasmid type

Joining HindIII site fragments (molecular weight • 10 "~ )

PTY21

23

RIB

6.8

4.75 0.8

0.6

HI

13

PTY39

14

RIB

8.5

2.35 1.45 0.8

H3

13

PTY75

14

RIA

8.2

2.3

H1

13

PTY81

23

RIA

6.8

4.75 0.8

H3

13

1.70 0.8 0.6

• The orientation is indicated by the 2-~m DNA HindIII site closest to the pCR1 HindIII site.

,tO H2

2.S

HI

/q RIB u

pCRI

PTY21

IA5 225

PTY39

H

H

2.2S 2.5

H3

PTY 81

-IJ,S

PrY 75

Fig.4. Schematicdrawingsof chimeric pksmids PTY21, 39, 75 and 81. The Hindm fragment sizes are given in daltons • 10"~.

TranSformation o f a minieeU.producing K eoli strain K coli P678-54, reeA +, rk +, mk÷ (Adler et al., 1967) was transformed with isolated chimeric plasmids. The transformed strains will be indicated by mini followed in brackets by the plasmid present e.g. mini(pCR1) contains plasmid pCR1. The frequency of transformation was reduced by 200-fold compared with the transformation of K eoli 490 recA, rk", ink", probably due to restriction of the incomingunmodified DNA (Morrow et al., 1974). Kanamycin-resistant clones were isolated and the supercofled plasmid DNA analyzed. In all cases analyzed the £coRI and HindHI restriction fragments of chimeric plasmid DNA isolated from minicells had the same mobilities in agarose gels as the restriction fragments of native 2-pro DNA. There was no evidence that recombination had taken place d',tring the course of the experiment; this is in contrast to the extensive recombination among mtDNA-pSC 101 chimeric DNA observed by Chang et al. (1975) in several minicellproducing strains. As far as detectable with these restriction enzymes, the PTY chimeric plasmids were stable. Transcription o f 2-~m DNA in E. eoli minicells ~ Clones of K coli P67~;-54. containing e ~ e r i e p ~ m i d s were cultivated and minicelk were isolated ~ad incubated with [,H]uridine ~ c o r d i n g t o Roozen

41

Fig.§. Hybridization of [SH]RNA synthesized in minicells from mini(PTY21) with an £eoRl digest of total yeast DNA. (a) £eoRl digest of total yeast DNA fractionated in 0.7% agarose. (b) £eoRl digest of PTY21 DNA. (c) Fluorogram of [SH]RNA hybridized to a nitrocellulose strip that carries the transferred DNA from gel a. See text and METHODS. for further details.

et al. (1971). Purified RNA from these cells was hybridized with an EcoRI digest of total yeast DNA fractionated on an 0.7% agarose gel and immobilized on nitrocellulose membrane as described by Southern (1975). After fluorography, hybridization exclusively to the 2-/~m DNA EcoRI fragments could be observed (Fig,5) indicating that at least part of both EcoRI fragments had been tnmscribed in the minicelk. The absence of hybridization in other regions of the gel showed that this hybridization was specific. Of course, no conclusions could be drawn as to the fidelity of chain selection, of initiation or of termination.

Polypeptide synthesis in E. coil minicells directed by the presence of 2-1~m DNA Isolated minicells were incubated with [3H]leucine under the conditions described by Roozen et al. (1971). At a minicell density of 0.2 and 0.8 measured at A62onm, the rate of incorporation was almost linear for 30 min at 37°C. Roozen et al. (1971) observed the same kinetics for minicells containing other plasmids. In minicelk from E. co~i P 678-54 which had not been transformed witha plasmid DNA, no time-dependent incorporation of [3H]leucine could be detected. After 40 min of synthesis the miniceUs were lysed

42

Fig.6. FluoroiP~mzof [SH]leucine-labelledpoJypeptidu ~mthesized in ~ o m ~ e d minicells as ducribed under METHODS. (8) mini(PTYS1). (b) mini(PTY?5).(o)mini(PTY39). (d) mini(PTY21). (e) mini(l~RX). The polypeptid~ which were rynthuized exclusively in the pre~mce of inbreed 2-~m DNA are indicated, and their m o l ~ weifhte ( • 10"s) are given. The position of the markers b shown on the left side. BSA, bovine serum albumin, 67 000; OA, ovalbumin, 43 000; GPd, ~y~aldehyde phosplmte dehydrogenmm, 36 000; Chym, chymotrypsinogen, 25 000; Mb, myogiobin, 1"/200; Hb, hemoglobin, X5 500.

by sodium dodecyl sulfate treatment in the presence of phenylmethylsulfonyl fluoride and the proteins were fracfionated on 15% polyacrylamide slab gels with a 5% stacking gel as described in MATERIALSAND METHODS. A comparison of the stained gels of mini(pCltl) and mini(PTY21) proteins did not reveal any difference in the protein bands. As expected, the integrated yeast DNA did not give rise to a major proportion of the total proteins present in the minicells. If, however, the radioactive polypeptides synthesized in minicells with PTY chimeric plasmids were compared to those synth~ized in minicells with pCR1, a number of interesting differences could be ol~erved (Fig.6). At least six extra bands, covering a molecular weight r ~ g e of 15 000 to 48 000, are present in the gels of minicells carrying PTY pla~mids (Table ILl). In addition, five other bands may be PTY-specific. The nature of these bands could not yet be determined because either avery faint counterband was present in the mini(pCR1) gel, or the position of a possible counterbsnd was too close to a h e ~ y band for its absence to be clearly demonstrated, The molecular weights of these pepfides are listed in colunm 2 of Table H. The four different plasmids studied yielded only two different gel patterns.

43 TABLE 11 POLYPEPTIDE BANDS SYNTHESIZED UNDER DIRECTION OF INTEGRATED 2-/~m DNA Results of a eompsrison of the gel bands in Fig. 6. The molecular weights of bands which could not with certainty be identified as yeast DNA specific, have been listed in the second

column (su text). Symbols used: 0, absent; ---, undecided; ±, just detectable; +, band of lower intensity; ÷÷, +++ and ++++, indicate increasing intensities (compare Fig.6). Apparent molet~htr weilht (. 10"s)

Presence and intensity in PTY transformed minicells

Certain

PTY21

PTY39

PTY75

PTY81

pCR1

÷

÷

÷

÷

± B

Uncertain

4~

87

i

i

+÷+

++÷

28

÷÷

÷÷

+÷

+÷

24

m

~

÷

÷

±

++++

++++

0

0

0

÷

÷

±

±

D

÷

÷

±

±

u

+

+

+

+

0

÷÷

+÷

÷+

÷÷

--

+

+

+

+

0

÷+

÷+

++

++

0

22

21.5 21 20 19

17.5 15

The integration site determined the pattern type. Mini(PTY21)' and mini (PTY39) have a very heavy band of 22 000 molecular weight which is missing in the mini(PTY75) and mini(PTY81) protein gel. In contrast a 37 000 molecular weight band is present as a strong band in mini(PTY75) and mini(PTY 81), but its presence in mini(PTY21) and mini(PTY39) is hardly detectable. This band is listed as certain although the mini(pCR1) gel has ~ light band close to this position. The absence of the 22 000 polypeptide in mini(PTY75) and that of the 37 000 polypeptide in mini(PTY21) could mean that sites RIA and RIB are located in DNA regions coding for these proteins and proper translation is precluded by insertions at these sites. No differences between the polypeptide bands from m!ni(PTY21) and mini(PTY39) cells could be detected, which suggests that type 14 and type 23 DNA directed the synthesis of the same polypeptides. The ~ame conclusion can be drawn from the identicai gel patterns of mini(PTY75) and mini(PTY81). The orientation of 2-~m DNA in the chimeric plasm!ds also did not detectably affect the z~inicell proteins. PTY21 and PTY39 as well as PTY75 and PTY 81 carry the inserts with opposite orientation, but nevertheless gave rise to the same polypeptide pattern. The total molecular weight of the six polypeptides listed in Table II as

44

certain, is 159 -500. About 3.2 • 106 dMtons of DNA are required to code this ~mount of amino acids. Together with accessory sequences for regulation and replication, these proteins could more than fill up the coding capacity of a 2-~m DNA molecule. DISCUSSION

Chimeric plasmids PTY21, 39, 75 and 81, which harbor all differences in structural properties found in the eight possible configurations (see Table I), were chosen for expression in minicells. After it became clear that neither the type of 2-~m DNA integrated, nor the relative orientation of the insertion had any influence on which polypeptides were synthesized, further study could be confined to plasmidsPTY21 and PTY75. About twenty well detectable bands were resolved by electrophoresis of a mini(pCR1) lysate. The total molecular weight of the polypeptides is around 400 000; 8.106 daltons of DNA would be needed to code for them. Considering that a few stable mRNAs might be carried over from the mother cell and still be ~mslated in the minicells, this DNA value is i~. reasonable agreement with the 9 • 106 daitons of pCR1 DNA. Most 2-#m DNA specific bands contain the same or a lower amount of radioactivity than the majority of the pCR1 specific bands. Two exceptions, however, were observed. A band of 37 000 molecular weight is strongly represented in n~ni(PTY75) and uncertain in m ini(PTY39) gels. The absence of the polypeptide in mini(pCR1) could not definitely be established, as a hand wa~ close to this position, The difference between mini(PTY7§) and mini (PTY39) in relation to this 37 000 polypeptide, however, makes it a candidate for a product of the DNA region including site RIB. A band with a molecular weight of 22 000 is at least four times heavier than the other bands. The band is present in mini(PTY21) but absent in mini (PTY75) and mini(pCR1) gels. This suggests that the integration site of PTY75, site RIA, is located in the DNA region coding for this polypeptide. Since no strong band of lower molecular weight is present exclusively in mim(PTY75) gels it is possible that site RIA is located in the proximal half of the mRNA or that an intact RIA site is required for transcription. Proteins with a molecular weight of less than 10 000 would not be detected in our gel system. Alternatively, it is not excluded that the polypeptide of 22 000 molecular weight represents an interrupted product of the gene c ~ g for the 37 000 molecular weight peptide. Two other bands of relatively ~ h t intensity are present just below the 22 000 band. In mini(PTY75) gels two barely detec. table bands could be observed at about the same position. The two bands might be related to the 9.2 000 band, but identification h ~ to await further analysis. For lower molecular weight polypeptides the possibilit~ always exists that they are breakdown products of specific proteas~ acting ~onlonger polypep. tides in the minicells. We do not have evidence yet as to whether this process gave rise to some of the chimeri~ plasmid specific bands.

45

The orientation of the 2-/~m DNA insert in the chimeric plasmid did not influence the polypeptides synthesized. This observation suggests that none of the observed chimeric plasmid-specific polypeptides has been translated from the hybrid region of a read-through transcript running from one DNA segment into the other. In most cases read-through transcription would lead to a frame shift which would result in early termination of the polypeptide involved. Small early terminated polypeptides however, would not be detected in our gel system. Finally, we would like to mention a possible class of proteins that cannot yet be excluded. It is conceivable that the synthesis of pCR1 coded proteins could be induced by transcripts or translation products of the inserted 2-/~m DNA. These proteins would have been scored as 2-/~m DNA specific polypeptides. The function of yeast 2-pro DNA is unknown and the present experiments were designed to set up a system to obtain information on the polypeptides coded by dlis DNA. The data presented in the first stage of this ~study show that E. coli minicells synthesize a small number of high molecular weight polypeptides in specific response to the presence of integrated yeast 2-/zm DNA. Immunological snalysis of these proteins in comparison to proreins from the yeast cell will now be necessary to establish whether any of these polypeptides represents a faithful product of a 2-#m DNA gene. Two properties of the chimeric plasmid expression as analyzed here already argue in favor of a faithful expression. (1) Only a few discrete high molecular weight polypeptides which taken together are within the limits of the coding capacity were produced. (2) The inversion present in about half of the 2-#m DNA molecules in vivo did not influence the polypeptides synthesized. This implies that none of the DNA sequences coding for a polypeptide extends from one loop segment over the id sequences into the next loop segment; this is in agreement with an equal coding function for both types of monomers. £. coli minicells have been previously used to study expression of eukaryotic DNA integrated in chimeric plasmi'ds (Chang et al., 1975). The polypeptides synthesized in response to the presence of integrated mouse mitochondrial DNA in minicells were of low molecular weight, and thus different from those synthesized by these mitochondria in vivo. Our results suggest, however, that minicells may represent a suitable system in which to study the gene products of yeast 2-/1m DNA. ACKNOWLEDGEMENTS

The authors thank Mrs. Ljubica Sanader for excellent technical assistance and are indebted to Prof. Dr. W. Beermann for his continuous interest and support. We thank Dr. U. Schwarz for providing K coli P678-54.

46

nFA RSNCES :69

Cohen, S.N., Chang, A.C+Y. and HsU, L.,~Non ebromwomal antibiotic resist~,ee in bacteria: Genetic transformation +)fF_,seherichiaeoUby R-factorDNA, Proe. Natl. Aead,~Sei. USA, 69 (1972) o.110--2114. • r Covey, C., Richardson, D. and Carbon, J., A method for the deletion of restri~ion sites in bacterial plasmid deoxyri'bonudeie acid, )4ol. Gen. Oenet., 145 (1976) 15~r--158. Curtiss, R. HI, Chromosomal aberrations associated with mutations to bacteriophage resistance in £wheriehia col/, J. Bacteriol., 89 (1965) 28--40, Davis, R.W., Simon, M, and Davidson, N., Electron mieroacope heteroduplex methods for mapping regions of base sequence homology in nucleic acids, in Grossman, L. and Moldave, K. (Eds.), Methods in Enzymology, Vol. 21 D, Academic Press, New York, 1971, pp. 413--428. Duell, E.A., ][noue, S. and Utter, M.F., Isolation and properties of intact mitoehondria from spheroplasts of yeast, J. Bacteriol., 88 (I964) 1762-- 1773. Gfllespie, D. and 8piegelman, $., A quantitative essay for DNA-RNA hybrids with DNA immobilized on a membrane, J. Mol. Biol., 12 (1965) 629--842. Griffiths, D.E., Lancashire, W.E. and Zanders, E.D., Evidence for an extrachromosomal element involved in mitoehondr~al function: A mitoehonddal episome?FEl~S Letters, 53 (1975) 126--130. Guerineau, M., Yeast episome: Structure and function, in 81onimski, P. and B6eher, Th. (Eds.), Genetics and Biogenesis of Chloroplaste and Mitoehondria, Elsevier/NorthHolland, Amsterdam, 1976, in press. Guerineau, M., Grandehamp., C., Paoletti, J. and 81onimski, P., Characterization of a new class of circular DNA molecules in yeast, Bioehem. Biophys. Res. Commun., 42 (1971) 550--557. Guerineau, M., Slonimski, P.P. and Avner, P.R., Yeast episome: Oligomyein resistance associated with a small eovalently closed non-mitoehondrial circular DNA, Biochem. Biophys. Res. Commun., 61 (1974) 462--469. Hirt, B., Selective extraction of polyoma DNA from infected mouse cell cultures, J. Mol. Biol., 26 (1967) 365--369. Hollenberg, C.P., Proportionate representation of rDNA and Balbiani Ring DNA in polytene chromosomes of Chironomus tentam, Chromoeoma (Berl,), 57(1976) 185--197. Hol]enberg, C.P., Borst, 1). and van Bruggen, E.F.J., Mitoehondrial DNA, V. A 2b-~ dosed cireular duplex DNA molecule in wild-type yeast mitoehondria. Structure and genetic complexity, Bioehim. Biophys. Acta, 209 (1970) 1--15. Hollenberg, C.P., Degelmann, A., Kustermann-Kuhn, B. and Royer, H.D., Characterization of 2-~m DNA of 8oceharomyces cerevisioe by restriction fragment analysis and integration in an Escheriehio eoUplasnfid, Pr0e. Natl. Aead, Sei, USA,73 (1976) 2072--2076. Hollenberg, C,P.+and Royer, H,D., Electron pi~fla f nat!veSd elonedi'~2!~m DNA from 8aceharomyce8 cere~+'.ule~,.inSlon~,~P-~i~ a ~ue.~,r'.~z~n'( and Biogenesis of Chloroplasts and Mitochondria, EkevierlNorth-Hollana, ,,mwwmmu, 1976, in press.

47 Katz, L., Kingsbury, D.T. and He!inski, D.R., Stimulation by cyclic adenosine monophosphate of plasmid deoxyribonucleic acid replication and catabolite repression of the plasmid deoxyribonucleic acid-protein relaxation complex, J. Bacteriol., 114 (I 973) 577--591. Kov~, L,, Bedn~ov~[, H. and Greks~k, M., Oxidative phosphorylation in yeast, I. Isolation and properties of phosphorylating mitochondria from stationary phase ceils, Biochim. Biophys, Acta, 153 (1968) 32--42. Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage '1'4, Nature, 227 (1970) 680-685. Lmkey, R.A. and Mills, A.D., Quantitative f'dm detection of 3H and 14C in polyacrylamide gels by fluorography, Eur. J. Bioehem., 56 (1975) 335--341. Morrow, J.F., Cohen, S.N., Chang, A.C.Y., Boyer, H.W., Goodman, H.M. and Helling, R.B. Replication and transcription of eukaryotic DNA in Escherichm coli, Proc. Natl. Acad. k i . USA, 71 (1974) 1743--1747. Roozen, K.J., Fenwick, Jr. R.G., and Curtiss, R. III, Synthesis of ribonucleic acid and protein in plasmid contdinin~ minicells of Eseherichia ¢oli K-12. J. Bacteriol., 107 (1971) 21--33. Shinnick, T.M., Lund, E., Smithies, O. and Blattner, F.R., Hybridization of labelled RNA to DNA in agarose gels, Nucleic Acid Res., 2 (1975) 1911--1929. Smith, C.A., Jordan, J.M. and Vinograd, J., In vivo effects of intercalating drugs on the superhelix density of mitochondrial DNA isolated from human and mouse ceils in culture, J. Mol. Biol., 59 (1971) 255-272. Southern, E.M., Detection of specific sequences among DNA fragments separated by gel electrophoresis, J. Mol. Biol., 98 (1975) 503--517. Struhl, K., Cameron, J.R. and Davis, R.W., Functional genetic expression of eukaryotic DNA in £scherich~ coli, Proc. Natl. Acad. Sci. USA, 73 (1976) 1471--1475. Studier, F.W., Analysis of bacteriophage T7 early RNAs and proteins on slab gels, J.Mol. Biol., 79 (1973) 237--248. Thuring, R.W.J., Sanders, J.P.M. and Borst, P., A freeze-squeeze method for recovering long DNA from agarose gels, Anal. Biochem., 66 (1975) 213--220. Communicated by J. Carbon.