Development of an in Vitro Bacteriophage N4 DNA Replication System*

0 downloads 0 Views 3MB Size Report
Aug 15, 2018 - products (dnp, dbp, and exo) required in vivo for N4 ... the wild type and CR63 (Ff lac' supD) the suppressor strain used for growth of phages (5) ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 261, No. 23,Issue of August 15,pp. 10506-10510,1986 Printed in U.S.A.

Development of an in Vitro Bacteriophage N4 DNA Replication System* (Received for publication, October 28, 1985)

J. Kevin RistSB, Margaret PearleSB,Akio SuginoV, and Lucia B. Rothman-DenesSII** From the 11Department of Molecular Geneticsand Cellular Biology and of $Biochemistryand Molecular Biology, The University of Chicago, Chicago, Illinois 60637and llhboratory of Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, NorthCarolina 27709

An in vitro DNA replication system from bacteriophage N4-infected Escherichia coli has been developed. It requires MgCl,, all four deoxyribonucleoside triphosphates, and exogenously added N4 phage DNA; other DNAs are used inefficiently or not at all. Ribonucleoside triphosphates are not required, although they stimulate DNA synthesis. Invitro replication starts at the ends of the N4 genome and moves progressively inward. Initiation occurs through hairpin priming at the 3’ ends of the genome, but shows a strong preference for the right end. Three N4 gene products (dnp, dbp, and exo) required in vivo for N4 DNA synthesis are absolutely required in the in vitro system. These findings are discussed with respect to the mode of N4 DNA replication.

RNA polymerase is responsible for early transcription (6), it is required for all events during the N4 growth cycle. With the use of temperature-sensitive mutants and temperature shift-up experiments, we have shown that its activity is also directly required for DNA replication (4). The unique structure of the genome and the involvement of the virion RNA polymerase in in vivo DNA synthesis prompted us to investigate the mode and mechanism of DNA replication. In this paper, we present the development and characterization of an in vitro replication system from N4infected E. coli. This system specifically uses N4 DNA, initiates at the ends of the genome by “hairpin priming,” and requires at least three phage-coded functions necessary for in vivo N4 DNA replication. MATERIALS ANDMETHODS

Coliphage N4 differs significantly from other lytic coliphages in its genomic structure and developmental requirements (1). The N4 genome consists of a linear doublestranded DNA, 72 kilobase pairs in length, flanked by 400450-base pair direct repeats(2). The left (or early) endof the genome contains a seven-base 3‘ overhang (3).’ The right end is heterogeneous; at least five ends, each differing from each other by approximately 10 bases in length, can be detected. Short, two- or three-base 3‘ single-stranded ends, which are not complementary to the single-stranded left terminus, are present at these ends. We have previously shown that N4DNA replication is independent of the Escherichia coli gene products involved in initiation and elongation of the host genome (4). The N4 genome is indeed large enough to code for its own replication functions. N4 DNA replication, however, depends upon the host machinery responsible for the processing of Okazaki fragments: DNA ligase and thepolymerase I 5’-3’-exonuclease (4). To date, we have identified five N4 geneproducts that are required for in vivo replication: the virion-associated DNA-dependent RNA polymerase (5), a 5‘-3’-exonuclease,* and three others of unknown function. Because the virion

* This work was supported in part by Grants CA 19265 and GM 35170 from the National Institutes of Health (to L. B. R.-D.). The costs of publication of this article were defrayed in part by the payment of page charges. This articlemust therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. §Supported by United States Public Health Service Training Grant GM 07281. ** To whom correspondence should be addressed. H. Ohmori, L. Haynes, and L. B. Rothman-Denes, unpublished data. Guinta, D., Lindberg, G., and Rothman-Denes, L. B. (1986) J. Biol. Chem., in press.



Bacteria and Phages-E. coli W3350 (F- cou’ r+m+gal- lac-)was ac’supD) the suppressor strain used for the wild type and CR63 (Ff l growth of phages (5). E. coli DllO (poD11end1 thyA) (7) was used for extract preparation. N4 wild type and amber mutants (8) were grown as previously described. To isolate [3H]N4 DNA, phage was grown in E. coli DllO in the presence of [3H]thymidine.E. coli DllO was grown with vigorous aeration at 37 “C to an ODszoof 0.5 in M9 media supplemented with 100 pg/ml thymine. Cells were pelleted, resuspended in fresh media, and infected a t a multiplicity of infection of 10. Uridine (150 pg/ml) and [3H]thymidine (4 pCi/ml) were added at 10 min after infection. After 3 h, phage was harvested as described previously (5). Media-LB media (9) supplemented with 10 mg/liter thymine was used to grow E. coli Dl10 for extract preparation. ChemicaZ~-[~H]dTTP (65 Ci/mmol) and [3H]thymidine (72 Ci/ mmol) were purchased from ICN Pharmaceuticals. [cY-~’P]~ATP (410 Ci/mmol) and [w3’P]dTTP (3000 Ci/mmol) were purchased from Amersham Corp. Deoxynucleoside triphosphates and ribonucleoside triphosphates were purchased from P-L Biochemicals. Enzymes-Restriction endonucleases HpaI and AvaII were purchased from Bethesda Research Laboratories and XbaI from IBI. Enzymes were used according to manufacturers’ specifications. Preparation of DNAs”N4 DNAwas isolated from virions by phenol extraction (10). pBR322 DNA and related plasmids were isolated by CsC1-ethidium bromide centrifugation (9). pBR322 plasmids containing N4 DNA sequences were constructed in this laboratory by Cherie Malone. The HpaIfragment C was inserted into the BamHI site after BamHI linker ligation yielding pBR(C). The terminal fragments HpaI, K, and G were CG-tailed and inserted into the PstIsite, yielding plasmids pBR(K) and pBR(G),respectively. Preparation of the Replication Extract from N4-infected Cells-E. coli DllO was grown to an ODszoof 0.5 with vigorous aeration at 37 “C.Cells were infected with phage at a multiplicity of infection of 10 and incubation was continued for 20 min. The cells were collected by centrifugation and resuspended in %w the original volume of 50 mM Tris-HC1, pH 8, 10% w/v sucrose. The cells were frozen in a dry ice/ethanol bath, thawed a t room temperature, and lysed with 500 pg/ml lysozyme, 10 mM EDTA. The cells were incubated on ice for 15 min, and Brij 58 and NaCl were added to concentrations of 0.5% and 1 M, respectively. The lysates were gently mixed, incubated on ice for 20 min, and then centrifuged at 100,000 x g for 45 min. An

10506

Development of an in Vitro Bacteriophage N4 DNA Replication System equal volume of saturated ammonium sulfate in buffer A (50 mM Tris-HC1, pH 8, 10% v/v glycerol, 10 mM 2-mercaptoethanol, 1 mM EDTA) was added dropwise with stirring a t 4 'C and stirred for another 20 min. The precipitate was collected by centrifugation a t 10,000 rpm for 30 min in a Sorvall SS-34 rotor. The precipitate was resuspended in '/so0volume (of the original culture) of buffer A and dialysed 2 h against buffer A. The extract was divided into small aliquots and stored a t -70 "C or used immediately. ReplicationAssay-The reaction mixture(100 pl) contained 66 mM Tris-HC1, pH 7.8, 10 mM MgClZ,2 mM dithiothreitol, 1 mM spermidine, 5 mM ATP, 0.1 mM each CTP, UTP, and GTP, 100 pg/ml bovine serum albumin, 10 p~ each dATP, dCTP, dGTP, and [3H] dTTP (2000 cpm/pmol), extract (1p1 unless otherwise stated), and DNA template (10 pg/ml unless stated otherwise). The reactions were incubated for 20 min a t 30 "C and stopped with 1 ml of 5% trichloroacetic acid. The acid-insoluble material was collected on GF/A filters (Whatman) andretained radioactivity was measured. Nitrocellulose Filter Hybridization-DNAs were restricted, electrophoresed, and transferred to nitrocellulose paper. DNA-DNA hybridization was carried out asdescribed previously (2). Two-dimensional Gel Electrophoresis-One per cent agarose gels were run atneutral pH in the first dimension. Lanes of interest were cut out and applied to a 1%alkaline agarose gel prepared and run in the second dimension according to Maniatis et d.(9). Isolation of DNA from the Replication Reaction-The reaction mixture was phenol-extracted and theaqueous phase passed through a 1-ml Sephadex G-50 column. The excluded volume was ethanolprecipitated and resuspended in the appropriate restriction enzyme buffer.

10507

dition ofKC1 above 25 mM inhibits the reaction, with 50% inhibition at 55 mM. Addition of glycerol does not affect DNA synthesis; however, the presence of polyethyleneglycol 8000 greatly stimulates incorporation. At concentrations greater than 15%, addition of polyethyleneglycol 8000 is inhibitory. The optimal pH for the reaction is between 7.7 and 8.2 (data not shown); pH 7.8 is routinely used. The rate of the in vitro replication reaction is linear for approximately 30 min at 30 "C after which time incorporation ceases (Fig. 1).Finally, the reaction is totally dependenton the addition of exogenous N4 DNA (Table I). Template Specificity-The template specificity was studied to determine the relationship between the in vitro reaction and thein vivo process of replication. Fig. 2 shows the capacity of various DNAs to serve as templates in the in vitro reaction. N4 DNA is the preferred template. Maximal incorporation is achieved at 10 pg/ml DNA when 150 pgof total protein is used. Under these conditions, after 30 min incubation, up to 10% of the template was replicated. X, 4x174 RF I, pBR322, PstI-digested pBR322 DNAs did not serve as templatesin the replication reaction (Fig. 2). Single-stranded DNAs such as viral 4x174and heat-denatured N4 or X DNAs did not support DNA synthesis (data not shown). Protein Requirements-Mutants in five N4 genes (vrp, exo, dbp, dnp, and dm) have been shown to affect in vivo N4 DNA replication (4). Three of these mutants (dbp, dnp, and dm)

RESULTS

An in Vitro System Capable of N4 DNA Synthesis Details for the preparation of the replication extract are given under "Materials and Methods." E. coli D110, a strain that lacks DNA polymerase I-polymerizing activity, is the host for wild type N4 infection. The cells are harvested 2030 min after infection, at which time the rate of N4 DNA synthesis has reached maximal levels (4). The presence of high salt in the extraction buffer is critical for attaining maximal activity in the in vitro replication system. The bacterial cell wall and the E. coli chromosome, which is not 20 40 60 80 degraded after N4 infection ( l l ) , are removed by centrifugation at 100,000 x g (12). The soluble proteins are fractionated TIME.min and concentrated by the addition of 50% ammonium sulfate. FIG. 1. Time course of in vitro N4 DNA synthesis. DNA The precipitate iscollected by centrifugation andresuspended replication was measured as described under "Materials and Methods.'' Aliquots were removed a t the indicated times. to a protein concentration of 15 mg/ml. Characterization of the System Reaction Requirements-As shown in Table I, DNA synthesis in vitro is totally dependent upon MgClz and deoxyribonucleoside triphosphates. Omission of ATP alone or UTP, CTP, or GTP results in lower replication activity. However, ribonucleoside triphosphates are not absolutely required. AdTABLE I Reauirements of N4 DNA revlieation in vitro Omission and additions %

Complete 100" -N4 DNA 3 0.6 -M&L -ATP 53 -dATP, dCTP, and dGTP 3.4 -CTP, UTP, and GTP 79 +KC1,50 mM 55 +KCl, 200 mM 5 +Glycerol, 20% 115 +Polvethvlene elvcol. 15% 200 100% corresponds to 80 pmol of dTMP incorporated under standard conditions.

I 10

20

30

DNA. pg/mi

FIG. 2. Template specificity of N4 DNA replication in vitro. Replication activity was measured under standard reaction conditions. Templates used were: 0 - 4 , N4 DNA A-A, X DNA; E-U, PstI-treated pBR322 DNA; U, 6x174 RFI DNA.

Development of an in Vitro Bacteriophage N4 DNA Replication System

10508

TABLE I1 Requirement of N4 gene products for in vitro DNA replication Infecting phage

am33A7

Activity

Exueriment 1

Exueriment 2

pmol dTMPincorporated 0.8

None N4+ dnp am25 dnp am32A257 ex0 amDll dbp d m am12 31.0

23.0 0

0.3 1.5 1.3 24.0

E

A,CD

19.0

FG

IJK

H

n

Hpa I

a

L MNO P ’



-10

15

5 1

)in

0.9 Ava II 1

I

IO

25-

5

FIG.4. Localization of the origin and direction of in vitro N4 DNA replication. Top, AuaII and HpaI restriction endonuclease maps of N4 DNA (2). Bottom, hybridization of DNA synthesized in vitro to Southern blots containing AuaII- or HpaI-restricted DNA. Conditions as described under “Materialsand Methods.” I

0.3

0.6 Extract

pl 33am7

FIG.3. Complementation of the in vitro replication defect of am25-infected cell extract witham33A7-infectedcell extract. Replication was measured in the standard activity reaction mixture as described under“Materials and Methods.” Two pl of am25-infected cell extract and the indicated amount of am33A7 extract were used in a 0.1-ml reaction.

E a

I /

! 16C /

5

show a DNA negative (DO) phenotype (4). Mutants in the other two cistrons required for DNA replication exhibit a complex phenotype. One of them, a suppressor-sensitive mutant, shows a DNA arrest (DA) phenotype at 37 “C but a DO phenotype a t 42 0C.3 This mutant (exo) lacks an N4-coded 5’-3’-exonuclease.* The fifth gene product involved in DNA replication is the virion-associated, DNA-dependent RNA polymerase (vrp) (5). To testwhether the in vitro replication system requires these factors, crude extracts were prepared from mutant-infected cells. Table I1 shows that extracts prepared from am33A7 (dbp), am25 (dnp), and exoDll (exo) mutant-infected cells are incapable of supporting N4 DNA replication. Incontrast,extracts from am12 (dm)mutant infection show considerable replication activity. The lack of activity of am33A7- and am25-infected cell extracts is not due to the presence of an inhibitor of replication since addition of mutant-infected extracts to wild type extracts does not affect the latter’s ability to support replication (data not shown). However, addition of am33A7-infected extract to am25-infected extract restores in vitro DNA synthesis activity (Fig. 3), suggesting that mutant extracts can be complemented efficiently for the missing function. Although we have been unable to show dependence on the dm gene product, the in vitro system requires at least three of the gene products, indicating that the system reflects at least some but perhaps not all in vivo replication reactions. Characterization of the in Vitro Products The strict template specificity of the in vitro system suggested to us that replication might be initiating at specific D. Guinta, S. Spellman, and L. B. Rothman-Denes, unpublished observations.

10

15

20

p Q / m l , DNA

FIG.5. Template activity of cloned N4 DNA fragments in the in vitrosystem. Replication activity was measured under standN4DNA; ard reaction conditions. Templates usedwere: O - 0 , A-A, pBR322 DNA; U, pBR(G); C . , pBR(K); X-X, pBR(C). sites on the N4 DNA molecule. To localize the site(s) of in vitro initiation, the labeled replication products were isolated after different lengths of incubation and digested with HpaI or AvaII restriction endonucleases (2). The digests were hybridized to Southern blots of N4 DNA treated with the respective restriction nucleases (Fig. 4). After 1 min of incubation, newly synthesized DNA hybridizes only to HpaI K and G fragments, with stronger hybridization to HpaI G than K. At longer times of incubation, the labeled DNA hybridizes to restriction fragments progressively more internal within the genome (see upper portion of the Fig. 4). Relatively more radioactivity hybridizes to right-end than the left-end proximal fragments. The fact that hybridization occurs to HpaI fragments M, N, etc. demonstrates that hybridization to HpaI K is not due to the presence of the 450 base pairs terminal repeats. Analysis of the AuaII-restricted products confirms the above results. Moreover it shows that after 35 min of incubation DNA synthesis has proceeded through all regions of the genome. This analysis indicates that the in vitro replication system is not only specific for N4 DNA, but that in vitro replication of this DNA initiates a t or near the ends of the genome. The availability of N4 HpaI fragments cloned in pBR322 allowed us to testthem as templatesfor N4 DNAreplication. Specifically, three clones containing terminal HpaIfragments G and K and internal HpaIfragment C were tested (see Fig. 4, top). The level of DNA synthesis obtained using plasmids

Development of an in Vitro Bacteriophage N 4 DNA Replication System

10509

neutral

FIG. 6. Two-dimensional gel electrophoresis of the product of the in vitro replication reaction.The standard replication reactioncontained 10 pg/ ml N4 DNA, [cY-~’P]~ATP (250 pCi/ml), and 1 pl of wild type N4-infected crude extract. The DNA was analyzed as described under “Materialsand Methods” before ( B ) andafter (C) S1 treatment. A, uniformly labeled t3H]N4 DNA was cleavedwithrestrictionendonuclease XbaI and analyzed by two-dimensional gel electrophoresis.Bottom, XbaI restriction endonuclease map ofN4 DNA.

i

DE’

F-

containing HpuIG or K fragments is significantly higher than that achieved with the vector alone or theHpuI C recombinant (Fig. 5). HpaI G plasmid appears to be more efficient than HpuI K, reminiscent of the higher activity of that end of genomic DNA (Fig. 4). However, HpuI K and G containing plasmids show drastically reduced activity when compared to genomic N4 DNA. This might be explained by a different topo!ogyof the DNA or by the lack of the structures (3’ overhangs) naturally present at theends of the N4 genome.’ Linearization of the plasmids resultsin theloss of differential template activity. In order to study the involvement of the ends of the N4 genome in initiation of in vitro replication, the products of the replication reaction were extracted and restricted with XbuI endonuclease. Two-dimensional gel electrophoresis, neutral in the first dimension and alkaline in thesecond, was used for analysis of the products. The results of such an experiment are shown in Fig. 6. The products of restriction endonuclease cleavage of mature N4 DNA migrate on a diagonal (Fig. 6A). In contrast to mature DNA, the two terminal fragments from the in vitro replication reaction migrate off the diagonal, specifically as bands of twice the expected size, under denaturing conditions (second dimension) (Fig. 6B). These results suggest that in vitro replication initiates at the ends of the genome through hairpin priming. This was confirmed by treatment of the replicated DNA with S1 nuclease which regenerates end fragments of mature size (Fig. 6C). DISCUSSION

We are attempting toinvestigate the mechanism of initiation of N4 DNA replication and the protein factors required in the process. To that end we have developed the in vitro systempresented in this paper. Several lines of evidence suggest that it reflects some, butnot all, of the in vivo processes and requirements of N4 replication. First, the in vitro system is exquisitely specific for N4 DNA, other DNAs are used poorly, if at all. Initiation on N4 DNA is restricted to theends of the genome (see Fig. 4). Indeed, it must occur at the ends since the endfragments from replicating DNA migrate anomalously inalkaline gels(Fig. 6). A normal migration pattern isrestored when the endfragments are treatedwith S1 nuclease prior to electrophoresis, suggesting the presence of hairpins at the ends of the in vitroreplicated DNA. Electron microscopic analysis of in vivo-replicating N4 DNA has failed to reveal eye structures (replication bubbles): K. Rist and M. Pearle, unpublished data.

Y-shaped molecules and molecules with single-stranded tails of different lengths are commonly seen, reminiscent of the structures observed during Bacillus subtilis $29 (13, 14) or adenovirus (15,16) replication. These resultssuggest that the in vivo origin of N4 DNA replication lies close to or at the ends of the genome. Replication of $29 and adenovirus DNAs is primed by “terminal” proteins, which remain covalently linked to the 5’ ends of the mature DNA molecules (17-19). We have been unable to detect such structures in the N4 genome? The structure of the in vitro product, if it is a reflection of the invivo process, would suggest that initiation of N4 DNA replication occurs instead through DNA priming at the3‘ single-stranded ends of the N4 genome. The results presented in Table I1 demonstrate that the activities of at least three gene products required for N4 in vivo DNA replication, dbp, dnp, and exo, are required in vitro. Furthermore, the replication activity of DNA synthesis-deficient crude extracts can be restored by mixing two replicationdeficient crude extracts (Fig. 3) or by the addition of fractionated N4 wild type-infected cell extracts (i.e. in vitro complementation). These resultshave enabled us topurify the products of the dbp, dnp, and ex0 genes. The dbp gene codes for a single-stranded DNA-binding protein while dnp codes for a DNA polymerase.6 These two proteins provide part of the elongation machinery for DNA replication. The independence of the in vitro reaction of the activity of the dns gene product, an M, 78,000 protein (4), may indicate that the invitro system does not mimic all aspects of the in vivo process of replication, i.e. maximal rate, processivity of the replication process, lagging strand synthesis, and unwinding of the template strands. Alternatively, the dm gene product may be required for a reaction which is not being studied in oursystem, such as formation or resolution of concatemers. We have been unable to study the role of the virion RNA polymerase in the in vitro replication reaction as only temperature-sensitive mutants of this gene product are available and the wild type N4-infected replication system is thermolabile. This enzyme, which has been purified to homogeneity from N4 virions (5), transcribessingle-stranded N4 DNA with in vivo specificity (20). The N4 genome contains four N4 virion RNA polymerase promoters. P1, P2, and P3 are located in the early (left 10%) area of the genome, with P1 located in the terminalredundancy. P4 is a copy of P1 and is present in the right terminal redundancy (20). All N4 virion RNA polymerase promoters are located on one strand of the

‘K. Rist, unpublished data.

K. Rist, A. Sugino, and L. B. Rothman-Denes,unpublished data.

10510

Development of an in Vitro Bacter.iophageN4 DNA Replication System

N4 DNA(3). Both in vivoand in vitro transcription of doublestranded DNA require supercoiled, promoter-containing template and E. coli single-stranded DNA-binding protein7 (21). The N4 virion RNA polymerase promoters PI, P2, and P3 direct the synthesis of early N4 RNAs (3, 20). On singlestranded templates, the virion RNA polymerase also synthesizes small (less than 20 bases) heterogeneous RNAs.’ Based on these properties of the N4 virion RNA polymerase, two possible mechanisms can be suggested to explain its involvement in DNA replication. First, the N4 virion RNA polymerase could synthesize the primer for initiation of replication at the origin, perhaps at the terminal redundancy. Such a mechanism has been demonstrated to occur in T7 DNA replication (22, 23). If that is indeed the case, the promoter sequences would be required for replication. It is not known, at present, whether these sequences are involved in N4 DNA replication. A second possibility relies on the ability of the virion RNA polymerase to synthesize small RNAs which could serve as primersfor discontinuous replication. Further experiments are required to answer these questions. The requirement for the activity of the N4 exonuclease in the in vitro system is puzzling. Only one suppressor-sensitive mutant in this gene is available. It shows a DNA arrest phenotype under nonsuppressing conditions but a DNA negative phenotype at high temperatures3 Unlike the X and T7 exonucleases, which also degrade a 5’-3’ direction, the N4 exonuclease is not required for recombination (24, 25).3 Perhaps the exonuclease is required in the in uitro reaction to provide a single-stranded template for elongation. Two lines of evidence suggest that is not the case. N4-denatured DNA is not acompetent template. Moreover, X and T7exonucleases cannot effectively substitute for the N4 enzyme in the in vitro reaction.*,’ Reconstitution of the N4 replication system from purified components, analysis of partial reactions, and elucidation of the in vivo mechanism of replication are required for a precise description of the events contributingto thereplication of the N4 genome. P. Markiewicz, C. Malone, J. Chase, and L. B. Rothman-Denes, unpublished data. P. Markiewicz, unpublished data. M. Pearle, unpublished data.

REFERENCES 1. Rothman-Denes, L.B. (1984) in Microbiology-1984 (Leive, L., and Schlessinger, D., eds) pp. 116-119, American Society for Microbiology, Washington, D. C. 2. Zivin, R., Malone, C., and Rothman-Denes, L. B. (1980) Virology 104,205-218 3. Zivin, R., Zehring, W., and Rothman-Denes, L. B. (1981) J. Mol. BWl. 152,335-356 4. Guinta, D., Stambouly, J., Falco, S. C., Rist, J. K., and RothmanDenes, L. B. (1986) Virology 1 5 0 , 3 3 4 4 5. Falco, S. C., Zehring, W. A., and Rothman-Denes, L. B. (1980) J. BWl. C h m . 255,4339-4347 6. Falco, S. C., Vander Laan, K., and Rothman-Denes, L. B. (1977) Proc. Natl. Acad. Sci. U. S. A. 74,520-523 7. Moses, R. E., and Richardson, C. C. (1970) Proc. Natl. Acad. Sci. U. S. A. 6 7 , 674-681 8. Schito, G. C. (1973) Virology 55,254-265 9. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 10. Vander Laan, K., Falco, S. C., and Rothman-Denes, L. B. (1977) Virology 7 6 , 596-601 11. Schito, G. C., Pesce, A., and Satta, G. (1969) G. Microbiol. 1 7 , 141-150 12. Falco, S. C., and Rothman-Denes, L.B. (1979) Virology 96,466475 13. Harding, N. E., and Ito, J. (1980) Virology 1 0 4 , 323-338 14. Inciarte, M. R., Salas, M., and Sogo, J. M. (1980) J. Virol. 3 4 , 187-199 15. Lechner, R. L., and Kelly, T.J., Jr. (1977) Cell 12, 1007-1020 16. Winnacker, E. L. (1978) Cell 1 4 , 761-773 17. Salas, M., Mellado, R. F., Vinuela, E., and Sogo, J. M. (1978) J. Mol. Biol. 119, 269-291 18. Ito, J. (1978) J. Virol. 28, 895-904 19. Rekosh, D. M., Russell, W. C., Bellet, A. J. D., and Robinson, A. (1977) Cell 11,283-295 20. Haynes, L. L., and Rothman-Denes, L. B. (1985) Cell 4 1 , 597605 21. Rothman-Denes, L. B., Haynes, L., Markiewicz, P., Glucksmann, A., Malone, C., and Chase, J. (1985) UCLA Symp. Mol. Cell. Biol. New Ser. 30, 41-53 22. Hinkle, D.C. (1980) J. Virol. 3 4 , 136-141 23. Romano, L. J., Tamanoi, F., and Richardson, C. C. (1981) Proc. Natl. Acad. Sci. U. S. A. 7 8 , 4107-4111 24. Schulman, M. J., Hallick, L. M., Echols, H., and Signer, E. R. (1970) . . J. Mol. Biol. 52. 501-520 25. Powlings, A., and Knippers, R. (1974) Mol. Gen. Genet. 134, 173-180