Eur. J. Biochem. 70, 523-529 (1976)

Replication of M-13 DNA in Plasmolysed Escherichia coli Cells Formation of Fully Synthetic Duplex DNA Barbara E. KESSLER-LIEBSCHER and Walter L. STAUDENBAUER Max-Planck-lnstitut fur Biochemie, Abteilung Hofschneider, Martinsried bei Munchen (Received April 30/September 2, 1976)

The replication of the double-stranded replicative-form DNA of bacteriophage M- 13 was studied in a cellular system in vitro prepared by plasmolysis of M-13-am.5-infected Escherichia coli cells. Newly synthesized DNA was density-labelled with bromodeoxyuridine triphosphate and analysed by equilibrium centrifugation in neutral CsCl. After a 60-min incubation at 30 "C 15 - 20 of the radioactive label incorporated from [32P]dGTPwas found in fully synthetic duplex DNA, corresponding to 7 - 9 replicative form molecules/cell. The plasmolysed cell system is therefore capable of re-initiating new rounds of replicative form replication in vitro. The kinetics of labelling indicate that molecules are selected for replication at random from an intracellular pool of approximately 150 replicative form molecules. Rifampicin and nalidixic acid, which interfere with the semiconservative replication of replicative form DNA, completely prevent the formation of fully synthetic duplex DNA.

The single-stranded viral DNA of phage M-13 is converted upon infection to a double-stranded replicative form, which then replicates semi-conservatively to form a pool of progeny replicative form molecules [1,2]. Later in the infection a switch from doublestrand replication to the synthesis of progeny single strands occurs. Single strand synthesis depends on the accumulation of the phage-coded gene V protein, since mutants deficient in gene V continue the replicative form replication for extended periods of time and do not enter into single strand synthesis. Although single strand synthesis does not involve all the same enzymes as replicative form replication [3,4], both types of DNA synthesis require the phage-specific gene I1 protein [5,6] and involve replicating intermediates containing elongated viral strands [7,8]. Gene I1 protein mediates the conversion of covalently closed replicative form I molecules to open circular replicative form I1 molecules having a viral-strandspecific discontinuity [5,9]. This nicking of the viral strand of the replicative form is a prerequisite for replicative form replication as well as for single strand Ahbreviurions. Replicative form, double-stranded circular replicative-form DNA; replicative form I, replicative form with both strands covalently closed; replicative form 11, replicative form with discontinuity(ies) in either strand; dBTP, 5'-brornodeoxyuridine sultriphosphate; Hepes, N-2-hydroxyethylpiperazine-N'-2-ethane phonate; EGTA, ethyleneglycol bis(2-aminoethyl)N,N'-tetraacetic acid.

synthesis and possibly provides a 3' primer terminus for viral strand synthesis. The further elucidation of the mechanism of phage DNA replication requires the development of appropriate in vitro systems. A start in this direction are studies employing nucleotide-permeable cells. Escherichia coli cells permeabilized by treatment with organic solvents or high concentrations of sucrose (plasmolysis) are no longer viable but have retained the capacity for macromolecule synthesis upon addition of ATP and low-molecular-weight precursors. Such cellular systems in vitro offer several experimental advantages for studies on DNA replication. Thus density-labelling experiments can readily be performed by substituting dBTP for dTTP. Furthermore, the effects of various agents can be investigated without complications due to limited permeability or indirect effects on overall metabolism. Hoffmann-Berling and co-workers have studied 4x174 replicative form replication in a cellular system in vitro prepared by ether treatment of phageinfected E. coli cells [lo]. It was shown by density labelling that the newly synthesized DNA was about equally distributed between two classes of hybrid replicative form molecules, in which either the viral or the complementary strand is labelled [ll].However, no label was detected in fully synthetic molecules, indicating that this system is defective in re-initiation.

524

A previous communication from this laboratory (121described a system in vitro prepared byplasmolysis of M-13-am5-infected E. coli cells, which can carry out M- 13 replicative form replication. Density labelling revealed besides molecules of hybrid density a small amount of heavier DNA. In this paper we present evidence that this material consists of fully synthetic replicative forms molecules. The effects of specific inhibitors on the formation of the various classes of density-labelled replicative form molecules are also reported. MATERIALS AND METHODS Bacteria and Bacteriophages E. coli 159 (F' thj'- hcr-) obtained from Dr D. Pratt was grown in M 9 medium [13] supplemented with 5 pg/ml thymidine at 37 'C. M-13 am5 was also kindly provided by Dr D. Pratt.

Preparation of Plasmolysed Cells

A 1-1 culture of E. coli 159 was grown to a density of 3 x 10' cells/ml and infected with M-13 am15 at a multiplicity of 20 phages/cell. Aeration was continued for 40 min at 37 "C. The culture was then poured on an equal volume of ice and harvested by low-speed centrifugation at 4 "C. The pellets were washed twice with cold 20 mM Hepes, pH 8.0, 10 mM MgCl2, 0.1 mM EDTA. Plasmolysis was carried out by resuspending the cells in 5 ml 20 mM Hepes, pH 8.0, 5 mM EGTA, 2 M sucrose. Aliquots of the plasmolysed cell suspension were quick-frozen in an acetone solid C02 bath and stored at - 20 "C. Stundard Incubation Mixture

Standard incubation mixtures (1 ml) contained 20 mM Hepes, pH 8.0, 100 mM KCl, 10 mM NH4C1, 1 mM dithiothreitol, 2 mM ATP, 20 mM phosphoenolpyruvate, 0.25 mM NAD, 0.5 mM each of CTP, GTP and UTP, 0.05 mM each of dATP, dCTP and dTTP, and 0.02 mM of [3zP]dGTP (specific activity 0.5 Ci/ mmol corresponding to about 1000 counts inin-' pmol-'). 100 p1 of the plasmolysed cell suspension (corresponding to 5 x lo9 cells) was added to the incubation mixture and the tubes incubated for 30 min at 30 'C. Incubation was stopped by adding 1 ml of cold 0.1 M EDTA and placing the tubes in an ice-bath. Density-labelling mixtures are standard incubation mixtures with dTTP replaced by 0.05 mM dBTP. Isolation of Phage-Specific D N A

Plasmolysed cells were washed twice with 50 mM Tris, pH 7.5, 5 mM EDTA, 0.1 M NaC1, resuspended

Replication of M-13 Duplex DNA

in 1 ml buffer and incubated with lysozyme (0.5 mg/ml) for 30 min at 0 "C. Cells were lysed by the addition of 0.1 m15 sarkosyl followed by a 15-min incubation at 37 "C. The highly viscous lysate was carefully layered on a 10- 30 % (w/v) sucrose gradient in 35 ml Tris/EDTA/NaCl buffer containing 1 M NaCI. Centrifugation was performed in a Spinco SW 27 rotor at 25000 rev./min for 17 h at 4 "C. 0.8-ml fractions were collected from the top of the gradient by pumping 50 sucrose into the bottom of the tube. Aliquots of each fraction were assayed for acid-insoluble radioactivity. Fractions containing phage-specific DNA (19- 30 S) were pooled and dialysed against Tris/EDTA/NaCl buffer. Equ ilihr ium Centr fuga t ion Centrifugations were performed in a Spinco Ti-50 rotor. Tubes containing 5.2 g CsCl and 4.0 g sample plus Tris/EDTA/NaCl buffer were filled with light mineral oil and centrifuged at 40000 rev./min for 40 h at 15 "C. 60 fractions were collected from the bottom of the tube. Aliquots of each fraction were spotted on filter-paper disks and assayed for radioactivity.

Agarose Gel Electrophoresis 1.5 '%, agarose gels were prepared by dissolving 1.5 g agarose (Seakem) in SO ml water in a boilingwater bath. After cooling the liquified agarose to 45 "C an equal volume of pre-warmed electrophoresis buffer (80 mM Tris, 40 mM sodium acetate, 4 mM EDTA, adjusted with acetic acid to pH 7.7) was added and the mixed solution was poured into 10-cm acrylic tubes (0.7 cm internal diameter). The gel tubes were cooled for 30 min at room temperature, and the top 1-cm portion of the gels was removed with a razor blade to ensure a flat gel surface. The tubes were then covered at the bottom with wet dialysis tubing and immediately mounted in the electrophoresis apparatus. Electrophoresis was carried out in electrophoresis buffer (diluted 1:1) containing 0.2 sodium dodecylsulfate. Gels were pre-run for 30 min at room temperature. DNA samples (100 pl) were mixed with 50 p1 diluted electrophoresis buffer containing 30 sucrose and 0.6 sodium dodecylsulfate and warmed at 37 "C for 5 min. Samples were then loaded into the top of the gel tubes and subjected to electrophoresis for 9 h at 5 mA/tube. After electrophoresis the gels were sliced into 1-mm disks, which were placed into scintillation vials containing 10 ml of a Triton-toluene mixture [2.5 1 of toluene, 1.25 1 of Triton X-100,0.95 g of dimethyl-1,4bis-2-(5-phenyloxazolyl)-benzene,and 31 g of 2,sdiphenyloxazole]. After keeping them at room temperature for 2 days the gel slices were counted in a liquid-scintillation counter.

'x

B. E. Kessler-Liebscher and W. L. Staudenbduer

525 1 .85

--

1.80

.!

1.75

,"

1

-

m

._ YI

.70 5 0

1.65

n -0

10

20 30 Fraction number

40

5c

Fig. 1. Pjcrzngruphii~unulj,sis of' hron?odeox~u,-idin~~-luh~llL~d wpli1.uf i v e fornz D N A . Plasinolysed cells were prepared as described in Methods except that 1 mCi [3H]thymidine (26 Ci/mmol) was added to a 1-1 culture prior to infection with M-13 um5. A I-mi densitylabelling mixture was incubated for 30 min at 30 "C. Incorporation was stopped by addition of 1 mlO.1 M EDTA. The cells were lysed, the phage D N A isolated by sucrose gradient centrifugation and analysed by buoyant density centrifugation in neutral CsCl as described in Methods. Densities (e)weredeterinined from refractometer readings using the data of Szybalski [26]. The values obtained were corrected by referring to the position of 3H-labelled replicative form D N A (density 1.701 g/ml). (-0) 32P-labelled newly synthesized D N A ; (0--0) 'H-prelabelled DNA

Materials [32P]dGTP (2.5 Ci/mmol) was from the Radiochemical Centre (Amersham). Rifamycin AF/ABDP and rifamycin AF/O13 were a gift from Dr S. Riva (Lepetit). The sources of all other reagents have been described previously [I 21.

RESULTS Identification of Fully Synthetic Replicative Form Molecules

A plasmolysed cell system was prepared from M-13-am5-infected E. coli cells as has been described previously [ 121. Progeny replicative form molecules were radioactively labelled by carrying out the infection in the presence of [3H]thymidine.In order to avoid possible deleterious effects [l], cells were not treated with mitomycin C prior to phage infection. For density labelling of the DNA synthesized in vitro, plasmolysed cells were incubated with a reaction mixture containing dBTP instead of dTTP employing [32P]dGTP for radioactive label. The phage-specific DNA was separated from chromosomal DNA by su-

Distance migrated (crn)

Fig. 2. Agarose gel el.crrr~/~horrsis.32P-labelled DNA banding at a density of 1.81 g/ml in the gradient shown in Fig. 1 was pooled and further analysed by electrophoresis in 1.5:; agarose gels for 9 h at 5 inAigel. Migration is from left ( - ) to right ( + ) . The gels were "P-labelled sliced and analysed as described in Methods. (0-0) D N A ; (0---O), 'H-labelled replicative form DNA used as reference. RF I, 11, replicative forms I and I 1

crose gradient centrifugation [8J and further analysed by equilibrium centrifugation in neutral CsCl (Fig. 1). While most of the 3H prelabel stayed in the position of light replicative form DNA (density 1.70 g/ml), two major peaks of 32P radioactivity are observed at the densities of 1.76 and 1.745 corresponding to heavylight replicative form (viral strand labelled with bromodeoxyuridine) and light-heavy replicative form (complementary strand labelled). These two classes of hybrid replicative form molecules can be separated by buoyant density centrifugation due to their different content of bromodeoxyuridine [14]. 172, of the 3Hlabelled DNA was co-banding with the hybrid replicative form molecules. This indicates that during a 30-min incubation at 30 "C one out of six intracellular replicative form molecules has undergone one round of replication. At a density of 1.81 g/ml a third peak of 32Pradioactivity is observed, which amounts to 14'%,of the total incorporated [32P]dGMPbut does not contain any 3H-labelled DNA. This material was pooled and further analysed by electrophoresis in 1.5 agarose gels [15,16]. As shown in Fig. 2, most of the 32Pradioactivity was found in the position of covalently closed replicative form I DNA, which migrates faster through the gel owing to its more compact configuration. A smaller portion of the label was co-migrating with open circular replicative form I1 DNA. This result was confirmed by velocity sedimentation of the 32Plabelled DNA in an alkaline CsCl gradient (Fig. 3).

526

Replication of M-13 Duplex DNA

1.80 I

10

45 s

Density (g/crn3) 1 .?5 1

1 .?C I

18s

L

"0

10

40

20 30 Fraction number

Fraction number

Fig. 3 . Alkuline CsCl velocity sedimentation. 0.1 ml of the densitylabelled DNA from fractions 8- 14 of the gradient shown in Fig. 1 was denatured in 0.2 M NaOH and layered on a 4.3-ml linear alkaline CsCl gradient (density 1.2- 1.4 g/ml) prepared as (described previously [27]. Centrifugation was performed in a Spinco SW 56 rotor at 45000 rev./min for 1 h at 8 "C. Fractions were collected on filter-paper disks and assayed for radioactivity. Sedimentation is from right to left.).--.( 3ZP-labelled DNA; (0- -0) 3Hlabelled replicative form DNA and single strands used as sedimentation markers

Table 1. Equilibrium centrifugation of' broPnodeoxyuridine-labelled replicutive form D N A Density-labelling mixtures (1 inl) were incubated at 30 "C for the times indicated. Replicative form DNA was isolated and analysed by equilibrium centrifugation in neutral CsCl as described in Methods Time

DNA -

~

heavyheavy min

Iightheavy

counts/min - ~-

15 30 45 60

heavylight

2100 5400 9700 15400

__ 37100 52800 69600 84700

~-

~~~

17100 24400 31500 36000

17900 23000 28400 33300

- -

total

-

heavyheavy

% 57 10 2 139 183

About two-thirds of the 32P-labelled DNA was sedimenting somewhat faster than denatured imarker replicative form I as expected for bromodeoxyuridinelabelled replicative form I DNA. The other third of the alkali-treated DNA was sedimenting slightly faster than M-13 single strands. From these data we infer that the heavy peak from the gradient shown in Fig.1 consists of fully synthetic replicative form molecules (heavy-heavy replicative forms).

Fig. 4. Transfer of lahe1,from hybrid tofully synthetic DNA. A density labelling mixture (2 ml) was incubated for 15 inin at 30 "C. A I-ml sample was removed and incorporation stopped by addition of 50 mM EDTA and cooling in an ice-bath. 0.05 ml 10 mM dGTP was added to the rest of the incorporation mixture and incubation continued for 45 min at 30 "C. The labelled phage DNA was isolated and subjected to buoyant density centrifugation as described in Methods. The radioactivity patterns of the pulse and chase are superimposed according to the density profile determined as described in Fig. 1. 92% of the pulse kdbel incorporated into phage DNA was recovered after the chase. (----O) DNA labelled during 15-min incubation; (0-0) labelled DNA after 45-min chase

Flow oJ'Labe1,frorn Hybrid to Fully Synthetic Replicative Form Next the distribution of radioactive label between the three classes of density-labelled replicative form molecules was determined after different times of incubation. Plasmolysed cells were incubated for 15- 60 min with density-labelling mixtures, the phagespecific DNA was isolated and further analysed by equilibrium centrifugation. The results are summarized in Table 3 . It can be seen that the relative amount of heavy-heavy replicative form increased markedly with the incubation time from 6 % at 15 min to 19% after 60 min. Since the incorporation leveled off after about 1 h, no further changes in the labelling pattern occurred afterwards. The observed distribution if radioactive label is consistent with a mode of replication involving the random selection of molecules for duplication and the return of the newly synthesized copies to the pool of replicative form molecules. This is also suggested by the experiment described in Fig. 4. [32P]dGMPincorporated during a 15-min incubation in the presence of dBTP was chased for 45 rnin by addition of an excess of unlabelled dGTP. Pycnographic analysis of the labelled replicative form DNA (Fig. 4) indicates that approximately 10 % of the incorporated radioactivity

B. E. Kessler-Liebscher and W. L. Staudenbauer

527 Density (g/crn3)

1.80

1.75

I

I

1

1 .70

1.80

I

1.75

I

%

1 .70

I

I

1.75

1 .70

Fraction number

Density (g /crn3)

1.75

1.80

10

20

1 .70

30

40

1

.8G

500 10 Fraction number

20

30

40

50

Fig. 5. Pycnugraphic unaly.sis ufphuge D N A synthesized in the presence of inhibitors. Density-labelling mixtures (1 ml) were incubated for 60 inin at 30 'C. Phage-specific DNA was isolated and analysed by buoyant density centrifugation in neutral CsCl as described in Methods. Densities were determined as described in Fig. 1. The following inhibitors were added: (A) 150 pg/ml chloramphenicol; ( B ) 5 pg/ml rifampicin (0-0) or rifamycin AFiABDP (0-0); (C) 20 pg/ml nalidixic acid (0-0) or 50 pg/ml nalidixic acid (O---O); (D) 0.25'5,, 2-phenylethanol

can be chased out of hybrid replicative form into fully synthetic replicative form molecules. In this and in several analogous experiments no preferential transfer of label from only one type of hybrid replicative form was observed.

Effects ojhhibitors To decide whether formation of fully synthetic replicative form molecules occurs by a semi-conservative replication process or by extensive DNA repair, its sensitivity towards specific inhibitors was investigated. It has been shown previously [I21 that replicative form replication in the plasmolysed cell system is inhibited by 70-75% in the presence of nalidixic acid or 2-phenylethanol. Furthermore, while chloramphenicol has no inhibitory effect, rifampicin reduces the incorporation of label into replicative

x.

form DNA by 40 Pycnographic analysis of density labelled phage DNA synthesized in the presence of these agents is shown in Fig. 5. It can be seen that the labelling pattern observed after incubation in chloramphenicol is practically indistinguishable from an untreated control (Fig. 5 A). However, upon addition of low concentrations ( 5 pg/ml) of rifampicin or rifamycin AF/ABDP no fully synthetic replicative form is formed and the relative amount of light-heavy replicative form is markedly reduced (Fig. 5 B). Synthesis of heavy-heavy replicative form is also completely blocked if the incubation is carried out in the presence of nalidixic acid (Fig.5C). It appears that low concentration of nalidixic acid (20 pg/ml) cause a preferential inhibition of light-heavy replicative form synthesis. The residual synthesis of heavy-light replicative form can be further reduced by increasing the concentration of nalidixic acid to 50 pg/ml. Contrary to

Replication of M-I 3 Duplex DNA

528

rifampicin and nalidixic acid, 2-phenylethanol(O.25 ”/, v/v) affects the synthesis of all types of replicative form molecules to a similar extent (Fig. 5D). These data indicate that fully synthetic replicative form is formed by a replicative process indistinguishable from semi-conservative replicative form replication. It should be noted that a considerable amount of label incorporated in the presence of rifamycin AF/ ABDP or nalidixic acid is found in DNA molecules banding in a density range between hybrid and light marker DNA. Such an intermediate density would be expected for partially replicated molecules in the process of replication. However, analysis by agarose gel electrophoresis and velocity sedimentation indicates that this material consists mostly of replicative form I molecules (data not shown). DlSCUSSION The data presented in this paper demonstrate that a cellular system in vitro prepared by plasniolysis of M-13-am.5-infected E. coli cells is capable of reinitiating new rounds of M-13 replicative form replication. After a 30-min incubation 10- 15 7; of the label incorporated into phage-specific DNA is found in fully synthetic replicative forms. This corresponds to 3 - 5 fully synthetic molecules/cell, since from the incorporation of labelled deoxynucleotides a synthesis rate of approximately 30 replicative form mslecules/ cell in 30 min can be calculated [12]. The majority of the label is therefore incorporated in about 50 hybrid molecules implying that 25 intracellular replicative form molecules have undergone one round of replication. Furthermore, since 17% of the prelabelled replicative form DNA was shifted to hybrid density during a 30-min incubation (see Fig.l), an intracellular pool of 150 replicative form niolecules/cell can be calculated. This is in good agreement with earlier estimates ranging from 130- 250 replicative form molecules/cell [17,18]. Formation of fully synthetic replicative form molecules is not dependent on protein synthesis but requires transcription as shown by its sensitivity to rifampicin. A direct role of RNA polymerase in replicative form replication has been indicated by data in vivo of Brutlag et al. [19]. Detailed studies by Fidanian and Ray [20] have revealed tha.t complementary strand synthesis is much more sensitive to rifampicin than viral strand synthesis. The preferential inhibition of the synthesis of replicative form molecules with density-labelled complementary strands observed upon addition of rifampicin to the plasmolysed cell system (Fig. 5B) is in full agreement with these results in v i w . It seems reasonable to assume that RNA polymerase is required for RNA priming of complementary strand synthesis. 2-Phenylethanol, which has been shown to inhibit the initiation

of bacterial DNA replication [21], interferes with replicative form replication by a mechanism different from that of rifampicin, since all types of densitylabelled replicative form molecules are equally affected (Fig. 5 D). Nalidixic acid, a specific inhibitor of the semiconservative replication of duplex DNA [9,22], completely prevents the synthesis of fully synthetic replicative form DNA. The formation of supercoiled replicative forms banding in an intermediate density range might indicate that molecules blocked in the process of replication are degraded and converted to replicative form DNA by a repair mechanism. Degradation of the parental viral strands has been observed previously during replicative form replication in vivo and involves the single-strand endonuclease activity of the recBC enzyme [23]. It may also account for the variable degree of asymmetry in the distribution of 32Plabel between the two classes of hybrid replicative form molecules leading to a slight excess of heavy-light replicative form (see Fig. 4). Although second rounds of replication are observed in the plasmolysed cell system, most replicative form molecules replicate only once and only a small fraction of the label can be chased from hybrid to fully synthetic molecules. After a round of replication the newly synthesized DNA is converted to covalently closed replicative form I molecules and there are indications that ring closure is required before a new round of replication can commence [14]. The labellin;; data shown in Table 1 and Fig.4 indicate that the newly synthesized replicative form molecules return to a pool of replicative form DNA and subsequent selection of a molecule for duplication occurs at random with respect to previous replication events. There is no evidence that M-13 replicative form replication is restricted to unique replicative form molecules attached to replication sites as has been suggested for 4 x 174 replicative form replication [24]. Our results are obviously inconsistent with any model assuming that the viral strand DNA is synthesized in a continuous and endless fashion on a circular complementary strand template [25]. However, the process which terminates viral strand synthesis after one round of replication is obscure [ 2 ] . Possibly nicking of the elongated viral strand by gene I1 protein at the origin/terminus might provide the necessary stop signal [8]. This work was supported by a grant from the Deutsche Forschungsgemeinschaft. We thank Miss E. Zehelein for technical assistance and Dr P. H. Hofschneider for stimulating discussions and helpful comments on the manuscript.

REFERENCES 1 . Marvin, D. A. & Hohn, B. (1969) Bacteriol. Rev. 33, 172-209. 2. Denhardt, D. T. (1975) CRC Crit. Rev. Microhiol. 4, 161-223.

529

B. E. Kessler-Liebscher and W. L. Staudenbauer 3. Staudenbauer, W. L., Olsen, W. L. & Hofschneider, P. H. (1973) Eur. J . Biochem. 32, 241-253. 4. Ray, D. S., Dueber, J. & Suggs, S. (1975) J . Virol. 16,348-355. 5. Lin, N. S.-L. & Pratt, D. (1972) J . Mol. Biol. 72, 37-49. 6. Tseng, B. Y. & Marvin, D. A. (1972) J . Virol. 10, 384-391. 7. Tseng, B. Y. & Marvin, D. A. (1972) J . Virol. 10,371-383. 8. Kessler-Liebscher, B. E., Staudenbauer, W. L. & Hofschneider, P. H. (1975) Nucleic Acids Res. 2, 131-141. 9. Fidanian, H. M. & Ray, D. S. (1972) J . Mol. Biol. 72, 51 -63. 10. Durwald, H . & Hoffmann-Berling, H. (1971) J . Mol. Biol. 58, 755-773. 11. Miiller-Wecker, H., Geider, K. & Hoffmann-Berling. H. (1972) J . Mol. B i d . 69, 319-331. 12. Staudenbauer, W. L. (1974) Eur. J. Biochem. 47, 353-363. 13. Miller, J. (1972) Ewperiments in Molecular Genetics, Cold Spring Harbor Laboratory. 14. Staudenbauer, W . L. & Hofschneider, P. H. (1973) Biochem. Biophys. Res. Commun. 54, 578- 584. 15. Aaij, C . & Borst, P . (1972) Biochim. Biophys. Acta, 26’9, 192200.

16. Tegtmeyer, P. & Macasaet, F. (1972) J . Virol. 10, 599-604. 17. Ray, D. S., Bscheider, H. P. & Hofschneider, P. H. (1966) J . Mol. Biol. 21, 473-483. 18. Hohn, B., Lechner, H. &Marvin, D. A. (1971) J . Mol. Biol. 56, 143- 154. 19. Brutlag, D., Schekman, R. W. & Kornberg, A. (1971) Proc. Nut2 Acad. Sci. U . S . A .68, 2826-2829. 20. Fidanian, H. M. & Ray, D. S. (1974) J . Mol. B i d . 83, 63-82. 21. Lark, K. G. (1969) Annu. Rev. Biochem. 38, 569-604. 22. Schneck, P. K., Staudenbauer, W. L. & Hofschneider, P. H. (1973) Eur. J . Biochem. 38, 130- 136. 23. Tseng, B. Y., Hohn, B. & Marvin, D. A. (1972) J . Virol. 10, 362 - 370. 24. Sinsheimer, R. L. (1968) Prog. Nuclric Acid Res. Mol. Biol. 8, 115- 169. 25. Gilbert, W. & Dressler, D. H. (1968) C,’oldSpring Harbor Symp. Quant. Biol. 33,473 - 484. 26. Szybalski, W. (1968) Methods Enzymol. 12, 346- 349. 27. Staudenbauer, W. L. & Hofschneider, P. H. (1971) Biochem. Biophys. Res. Cornmun. 42, 1035- 1041.

B. E. Kessler-Liebscher and W . L Staudenbauer, Max-Planck-Institut fur Biochemie. Am Klopferspitz, D-8033 Martinsried, Federal Republic of Germany