Proc. Nat. Acad. Sci. USA Vol. 71, No. 5, pp. 1758-1762, May 1974

Synthesis of Bacteriophage Lambda DNA In Vitro: Requirement for 0 and P Gene Products (Escherichia coli/DNA initiation/DNA polymerase) HIROAKI SHIZUYA* AND CHARLES C. RICHARDSON Department of Biological Chemistry, Harvard Medical School, Boston, Massachusetts 02115

Communicated by Eugene P. Kennedy, January 31, 1974 We have developed a cell-free system to ABSTRACT study bacteriophage X DNA replication. Maximal DNA synthesis in vitro requires the four deoxynucleoside triphosphates, ATP, and exogenous X DNA. DNA synthesis requires the products of the phage 0 and P genes but is not inhibited by X repressor. The kinetics of synthesis is linear for 10-15 min; however, the product of synthesis amounts to only 0.5-1% of the added template DNA. As judged by isopycnic analysis, extensive regions of the template are copied. Sedimentation analysis indicates that all of the product consists of short (11 S) DNA chains. Fractions partially purified from XO+P+-infected cell extracts will complement extracts prepared from XC- or XP--infected cells.

Recently several systems have been developed which are capable of replicating DNA in vitro (1-9). These in vitro systems have proved to be particularly useful as in vitro complementation assays for host and phage proteins required for DNA replication but of unknown enzymatic activity. However, only the systems using extracts of phage T7-infected Escherichia coli are capable of replicating exogenous duplex DNA (8, 9). In this communication we describe an in vitro system to study the replication of another duplex DNA, bacteriophage X DNA. In vivo X DNA replication is initiated at a specific site, ori, located about 18% from the right end of a mature X DNA (10, 11). Lambda DNA synthesis proceeds bidirectionally during the first round of replication (10) and is discontinuous. Short lengths of DNA are synthesized and then joined to form DNA of high molecular weight to extend the growing DNA chain (12-14). Mutants of E. coli have been isolated which are defective in the elongation of the growing chain of bacterial DNA at high temperature (dnaB and dnaG) (15). Lambda DNA replication is also blocked at the nonpermissive temperature in these elongation mutants. However, X DNA is replicated in mutants of W1. coli (dnaA and dnaC) defective in initiation (15), suggesting that X codes for its own initiation function, but requires host functions for the elongation of DNA chains. Using density labeling experiments Ogawa and Tomizawa (16) have implicated the products of the 0 and P genes of X as essential for initiation. Furthermore, transcription of the ori region is also required for initiation, suggesting a role for RNA in initiation of DNA synthesis (17). Recently Klein and Powling (18) have shown that in vitro X DNA replication depends on the presence of the 0 and P gene products and a functional RNA polymerase.

MATERIALS AND METHODS

Bacterial Strains and Baderiophages. The strains used are derivatives of E. coli K-12. D110 is an endonuclease I deficient derivative of P3478 polAl. C600 and Ymel carrying supE and supF, respectively, are used as indicators for lambda amber mutants. Lambda CI857029 and XcId7P3 were obtained from A. Campbell, and Xc187S7 was obtained from M. Chamberlin. Strain 866 carrying plasmid XdVgal,2o was provided by R. Davis. Stocks of XcIg57sus phages were prepared by heat induction of the lysogens. Strains were routinely grown in T-broth supplemented with thymine (10 jMg/ml) and 0.2% maltose.

Chemicals. Unlabeled nucleotides were obtained from Schwarz BioResearch. [8H]dTTP and [a-32P]dATP were purchased from New England Nuclear Corp. BrdUTP was a gift from A. Kornberg. Lysozyme, rifampicin, and actinomycin D were purchased from Worthington Biochemical Co.

Growth of Cells and Preparation of Cell Extract. E. coli D10 was grown at 370 in T-broth containing 0.2% maltose and 10 Ag/ml of thymine. At OD59o = 0.5 (2 X 108 cells per ml) cells were infected with X phage at a multiplicity of 7. Thirtyfive minutes after infection, unless otherwise noted, the infected cells were collected by centrifugation and resuspended in 0.002 volume of 50 mM Tris * HCl buffer (pH 7.6) containing 10% sucrose. The suspension was quickly frozen in a dry ice-isopropanol bath and then stored at -30° for up to 1 month. Frozen cells were thawed and lysed according to the method described by Wickner et al. (7). The cell extract (Fraction I) was kept at 00 and was used within 3 hr. The cell extract has an A2w0 of 160-180 and an A280 of 80-90. Assay of DNA Synthesis In Vitro. Unless otherwise indicated, each assay (100 ul) contains 10 mM Tris- HCl buffer (pH 7.6), 2.5 mM MgCl2, 75,MM each of dATP, dCTP, dGTP, and [8H]dTTP (100 cpm/pmole), 700 MM ATP, 10 nmoles (nucleotide phosphorus) of linear duplex X DNA and 10-30 ,Ml of Fraction I. The reaction mixtures were incubated at 30°. The reaction was stopped by adding 2.5 ml of cold iN HCl-0.1 M sodium pyrophosphate, and the acid-insoluble radioactivity was determined as previously described (3).

Preparation of DNA. Lambda DNA and T7 DNA were isolated from phage particles as previously described (19). Twisted X DNA and XdVgal DNA molecules were prepared according to the modified procedure described by P. Sharp et al. (20). DNA concentration is expressed as nucleotide equivalents.

Abbreviations: BrdUTP, 5-bromodeoxyuridine 5'-triphosphate; EDTA, ethylenediaminetetraacetate. * Present address: Department of Biochemistry, Baylor College of Medicine, Houston, Texas 77025.

1758

Proc. Nat. Acad. Sci. USA 71

Lambda DNA Synthesis In Vitro

(1974)

1759

30

KN.

2O0

lb

1o

I0 10 15 zu AtWA /REACT/OVN M/KflW (nmoks)

U

5

FIG. 1. Effect of DNA concentration on the rate of DNA synthesis. Reactions were as described in Materials and Methods with the indicated amounts of X DNA.

The requirements for lambda DNA replication in vitro are shown in Table 1. The complete system contains 4 dNTPs, ATP, Mg++, X DNA, and cell extract. If one or three deoxynucleoside triphosphates are omitted from the system, DNA synthesis is reduced to 50% of the control. This small reduction is probably due to the presence of deoxynucleoside triphosphates in the highly concentrated extract. The addition of ATP, however, stimulates the rate of incorporation. At the optimum concentration of ATP (0.7 mM), the rate of DNA synthesis is 10 times greater than it is in the absence of ATP. Higher concentrations of ATP result in inhibition; 1.7 mM ATP reduces the optimal rate of incorporation 8-fold. The addition of the other three rNTPs had no effect on the rate or extent of synthesis, and they could only partially replace ATP. Rifampicin, a specific inhibitor of the initiation step in the reaction catalyzed by RNA polymerase (21), does not inhibit synthesis of X DNA in vitro (Table 1). Actinomycin D completely inhibits the reaction in vitro. At a concentration of 10 repressor molecules per operator, X repressor has no effect on DNA synthesis. TABLE 1. Requirements for X DNA replication in vitro

Complete - X DNA - Mg++ - dATP - dATP, dCTP,

dGTP

-ATP + Rifampicin (2-40 /g/ml) + Actinomycin D (5;,g/ml) + X Repressor (10 molecules/operator)

3

7

12

18

25

MINUTES

FIG. 2. Kinetics of DNA synthesis in vitro. Extracts were prepared from cells infected with XcI857S7 or XcI857029 or XcI857P3 as described in Materials and Methods.

RESULTS Requirements for X DNA replication in vitro

Additions

0

Activity (% complete) 100 3.8 2.7 63 44 8.8 90 3.5 80

Activity was measured as described in Materials and Methods. In the complete system 48 pmoles of dTMP was incorporated in 15 min at 300 in 20 41 of extract obtained from Xcl&57Srinfected cells. XS- was used to prevent cell lysis during preparation of the extract. Lambda repressor was kindly provided by Dr. T. Maniatis.

The effect of DNA concentration on the rate of DNA synthesis is shown in Fig. 1. The incorporation of dTMP increases linearly with increasing amount of DNA added up to 12 nmoles/100 Al of reaction mixture. The kinetics of DNA synthesis is shown in Fig. 2. The incorporation of dTMP proceeds for at least 15 min. In the initial stage the apparent rate of incorporation of dTMP is 8.1 pmoles/min. The actual rate is probably higher, however, due to the endogenous nucleoside triphosphates. Within 25 min the total amount of dTMP incorporated in a reaction mixture is 65 pmoles, equivalent to 1.4 X 109 X DNA molecules. Since each reaction mixture contains 6.7 X 1010 molecules of X DNA template, 2% of the DNA has been replicated. Closed circular X and XdVgal DNA (22) could not replace linear X DNA (Table 2). T7 DNA, on the other hand, was 60% as effective as X DNA. Requirement for products of 0 and P genes Two lambda genes, 0 and P, are involved in X DNA replication in vivo, and may be necessary for initiation of DNA synthesis (13). When extracts were prepared from cells infected with 029 or P3, mutations in genes 0 and P, respectively, little if any DNA synthesis was observed in the standard assay (Fig. 2). Therefore, the products of the 0 and

TAnl.E 2. Template specificity of DNA synthesis Activity

Template DNA X linear X twisted X dVgal twisted T7

(%) 100 3 3 63

DNA synthesis was measured as described in Materials and Methods except that the above DNAs were used. Linear X DNA was replaced with 10 nmoles of other template DNA indicated. The incorporation of dTMP with linear X DNA was 48 pmoles in 15 min at 300 in 20 Al of extract prepared from cells infected with XS7.

1760

Proc. Nat. Acad. Sci. USA 71 (1970/

Biochemistry: Shizuya and Richardson

18mg

3mIn.

g/ml 1.80

6 1.1.0 4-

_1.75

0.i.5

2

_1.70

10

0

1.4

20

30

40

50

60

W)

7mIn.

25 mg

I.4

g/ml

I N.4

F2

3

- 1.80

.0

2

_

11_

F

.75

.5

.14

1.70

0

0.5

29 mg

15min.

g

0 20 40 60 80 100 0 20 40 60 80 100

2

.0

NORMALIZED FRACTION NUMBER

FIG. 3. Sucrose gradient analysis of X DNA synthesized in vitro. DNA synthesis was carried out in 700 ,l of reaction mixture containing 2.5 mM MgCl2, 10 mM Tris * HCl buffer (pH 7.6), 700 MuM ATP, 75 MuM each of dCTP, dGTP, and dTTP, 90,MM each of [a-32P]dATP (63 cpm/pmole), 105 moles of X [3H]DNA (3.8 cpm/pmole), and 140 Ml of the extract prepared from cells infected with XcI857S7. After 3-, 7-, and 15-min incubation at 300, 200 Ml of the samples was taken, and 100 Ml of cold 0.1 M ethylenediaminetetraacetate (EDTA) (pH 8.0) was added. One hundred microliters of each sample was then applied on 5 ml of a neutral and an alkaline sucrose density gradient. Centrifugation was carried out at 50,000 rpm in a Spinco type SW50.1 rotor for 150 min at 40 in the Spinco L3-50. Twenty-five to 30 fractions were collected from the bottom of the tubes. After adding 50,Ml of 1 mg/ml of salmon sperm DNA to each fraction, acid-insoluble radioactivity was measured as described in Materials and Methods. Five to 20% sucrose gradients contain 1.0 M NaCl, 10 mM Tris-HCl buffer (pH 8.0), 1 mM EDTA and 0.015% Sarkosyl NL97 for neutral, 0.7 M NaCl, 0.3 M NaOH, 1 mM EDTA, and 0.015% Sarkosyl NL97 for alkaline. A, B, and C are the neutral sedimentation profiles of samples taken at 3, 7, and 15 min, respectively; D, E, and F are the alkaline sedimentation profiles of samples taken at 3, 7, and 15 min, respectively. o--o (32P); A--A (3H). To obtain experimental values, numbers on the ordinate should be multiplied by the given factor.

P

genes

of X

are

required for DNA synthesis in vitro

/mlI 1.80

_

as

well

as in vivo.

Characterization of X DNA synthesized in vitro

A. Sedimentation Analysis. DNA synthesized in vitro using linear X [3H]DNA as template and [a-32P]dATP were analyzed in both neutral and alkaline sucrose gradients (Fig. 3). Samples taken after 3, 7, and 15 min of incubation at 300 had incorporated 83, 160, and 243 pmoles of dAMP, respectively. In neutral sucrose, after 3-min incubation a small but detectable amount of parental [3H ]DNA was found in a slowly sedimenting component which increased and reached about 40% of the total parental [3H ]DNA after 15 min, indicating that significant degradation of the parental DNA occurred during the incubation. All of the radioactive label (32P) in-

_ 1.75

.5 _1.70 0 10 0

20

30

40

50

~~~~~~~~0

FRACTTION NUMBER FIG. 4. Pycnographic analysis of X DNA synthesized in vitro. DNA synthesis was carried out in 700 ,l of reaction mixture containing 2.5 mM MgCl2, 10 mM Tris * HCl buffer (pH 7.6), 700 MM ATP, 75 MAM dCTP and dGTP, 90 MM BrdUTP, 90 tiM [a- 32P] dATP (67 cpm/pmole), 105 nmoles X [3H] DNA (3.8 cpm/pmole), and 140 Ml of the extract prepared from cells infected with XcI857S7. After 3-, 7-, and 15-min incubation at 300, 200-,ul samples were taken and 100 ,ul of cold 0.1 M EDTA (pH 8.0) was added. To 200 Mul of sample, 80 Ml of 1.5% Sarkosyl NL97, 1.72 ml of standard saline-citrate (0.15 M NaCl, 0.015 M sodium citrate, pH 7) and 6.0 ml of 65% (w/w) CsCl were added 1.713 g/ml). Centrifugation was in a nitrocellulose tube (p performed at 37,000 rpm for 48 hr in Spinco type 40 rotor at 50 in a Spinco L3-50. The incorporation of dAMP at 3, 7, and 15 min was 84, 155, and 189 pmoles in 100 M1u of the reaction mixture. The recovery of the labeled DNA was more than 75% in all cases. Density increases from right to left. o- o (32P); A--A (3H). =

corporated into DNA after 3, 7, and 15 min of incubation sedimented considerably more slowly than did the bulk of the parental DNA. The average sedimentation rate of the newly synthesized DNA was 16 S, equivalent to duplex polynucleotides of 3.4 X 106 daltons. Analysis of the DNA by alkaline sedimentation revealed that extensive hydrolysis of the template DNA occurred during the incubation. After 3 min of incubation most of the template DNA is still intact, but after 15 min the majority of the molecules have been extensively cleaved. At the same time the radioactivity incorporated into newly synthesized DNA accumulates as short DNA chains with an average sedimentation rate of 11 S. At 3 min the 1lS product is first detected, and by 15 min virtually all the radioactivity is found in this material.

Proc. Nat. Acad. Sci. USA 71

(1974)

Lambda DNA Synthesis In Vitro

1761

TABLE 3. Complementation in vitro Recipient (Fraction I)

Donor (Fraction IIB)

O-P+

O+P_

Q,

K~

2.3

O-P+orO+Po +PO-P+ O+P+

dTMP incorporated, pmoles

Uninfected Uninfected 0+P-

O+P+ O+P

omPe

1.1 0. 7 2.2 47.2 1.2 8.3 16.8

The incorporation of dTMP was measured as described in Materials and Methods. 0-P+ 0 +P-, or 0 +P+ indicate extracts prepared from D110 infected withX 029, P., or S7, respectively. Each fraction contains 20 ml of the recipient or donor extract, or both. Reaction mixtures were incubated for 10 min at 300.

B. Isopycnic Analysis. In order to analyze the DNA product in isopycnic CsCl gradients, newly synthesized DNA was labeled with BrdUTP and [a-32P]dATP using X [3H]DNA as a template. The replacement of dTTP with BrdUTP had no influence on the rate of DNA synthesis. The results of equilibrium centrifugation in neutral CsCl gradients are shown in Fig. 4. In comparison to the parental 3H peak the material synthesized over a 3-min period using BrdUTP as a substrate shifted its density by 18 mg/ml, while the shifts after 5- and 15-min incubation were 25 and 29 mg/ml, respectively. When the material at 15 min was extensively broken by sonic irradiation, the distribution of the radioactivity in the gradient was broad, but the density shift at the peak was not more than 30 mg, indicating that the heavier newly snythesized DNA consisted of one heavy and one light strand. The rather small density shift at full hybrid position is due to endogenous dNTP in the extract, resulting in the decrease of the bromouracil/thymine ratio in the reaction mixture. Therefore, in the heavy strand only 50% of the total thymine was replaced by bromouracil (23). From these results we conclude that extensive regions of the X DNA template are copied. Partial purification of the 0 and P gene products It is known from in vivo studies that 0- and P- phage complement when bacteria are infected simultaneously, and both phages can multiply (24). Extracts prepared from cells infected with either 0 or P- were mixed in equal amounts in order to test complementation in vitro. However, the activity found in the mixed extract was not more than additive, suggesting that the gene 0 and P products cannot complement effectively in this in vitro system. Tomizawa (25) has shown that the two gene products cooperate in vivo to initiate replication of X DNA. One type of cooperation would be the formation of a complex of the 0 and P gene products. Therefore, if one can supply the 0 and P gene products as a complex to the cell extract lacking normal 0 and P protein, DNA synthesis may occur. In order to test this possibility the extract obtained from S- infected cells was fractionated by ammonium sulfate to separate the 0 and P gene products from the synthesizing activity. To 3 ml of Fraction I, 0.63 g of ammonium sulfate was added. After 20 min at 00 the suspension was centrifuged for 20 min at 27,000 X g, and the pellet was dissolved in 3 ml of 25 m M potassium phosphate buffer (pH 7.4)-0.1 mM\ EDTA-0.1 mAM dithiothreitol-

0

10

20

30

DONOR EXTRACTru/) FIG. 5. Complementation in extracts from X-infected cells by donor extract. The recipient extract was prepared from cells infected with Xch857P3 as described except that cells were resuspended in 0.001 volume of 10% sucrose in 50 mM Tris-HCl buffer (pH 7.6). The donor extracts were prepared as described in Materials and Methods except that cells were harvested at 60 min after infection with XcI857S7 (e--e) or XC1957029 (0E), and 3.0 ml of the donor extract was treated with ammonium sulfate to yield Fraction JIB as described in the text. Each fraction contained 5 Il of the recipient extract and the indicated amount of Fraction JIB. Reaction mixtures were incubated for 15 min at

30o.

10% glycerol (Fraction IJA). Ammonium sulfate (0.49 g) was added to the supernatant fluid. After centrifugation the pellet was suspended in 1.5 ml of the same buffer. The sample was then dialyzed for 3-4 hr at 0° against two changes of 2 liters of the same buffer (Fraction IIB). The synthesizing activity was found exclusively in Fraction IIA. Fraction JIB has no synthesizing activity by itself, but when this fraction was added to the extract from 0- and P- infected cells, stimulation of DNA synthesis was observed (Table 3 and Fig. 5). When Fraction IIB is added to the mutant extracts, DNA synthesis can be stimulated as much as 10-fold (Fig. 5). The comparable fractions prepared from 0- and P- infected or uninfected cell extracts were found to have no stimulatory activity (Table 3). Therefore, the stimulation factor found in S- infected cell extracts is dependent on the 0 and P gene products. DISCUSSION

Several lines of evidence indicate that the DNA synthesis observed in this in vitro system using X DNA as a template closely mimics DNA replication: extensive regions of the X template are copied; 11S fragments of DNA are produced; ATP is required as in other in vitro systems (1-7); and X gene products, as well as E. coli components, are required. The rate of incorporation of nucleotides in vitro is 320 nucleotides per sec per cell equivalent, an estimate which has been corrected for the pool of endogenous deoxynucleoside triphosphates. Synthesis is dependent on the addition of linear X DNA, and isopycnic analysis confirms that the X DNA is used as a template in the reaction. Although T7 DNA can partially replace X DNA, it has not been shown that the synthesis observed is dependent on X-specific proteins. For example, in the case of an in vitro replication system using phage-T7-infected cells, other DNAs such as X DNA are also effective in promoting synthesis, although T7-specific proteins do not appear to be required (8). However, in the present study phage-size X DNA molecules are not synthesized, and the synthesis observed is not affected by X repressor.

1762

Biochemistry: Shizuya and Richardson

Perhaps the strongest evidence that X-specific DNA replication occurs in vitro is the requirement for the products of the 0 and P genes of X. Both of these gene products are known to be required for X DNA replication in vivo. In vitro, 0- and Pextracts complement O+P+ extracts, but do not complement O+P- or O-P+ extracts, respectively. Klein and Powling (18), however, have reported partial complementation in their in vitro system. The explanation for this observation may be that only nascent 0 and P gene products can form a functional complex. In support of this interpretation Tomizawa (25) has presented evidence which suggests that the two gene products cooperate in vivo to initiate replication of X DNA. The complementation assay, however, does allow a partial purification of the 0 and P gene products from the extract of S- infected cells. Chromatography on DEAEcellulose revealed that the stimulation separates from DNA polymerase III and ATP-dependent exonuclease V. Precisely which E. coli components are necessary for X DNA replication in vitro is not known. Although we have not examined dnaB and dnaG mutants, mutants which do not support X DNA synthesis in vivo, we have found that DNA synthesis in extracts prepared from poiC mutants infected with X is temperature-sensitive, suggesting that DNA polymerase III is required. Recent evidence (7, 26) indicates that RNA synthesis is important in the initiation of DNA replication and in the initiation of "Okazaki fragments." While we have not been able to obtain any direct evidence for a role of RNA synthesis in vitro, the presence of rNTP in the concentrated extract could explain the apparent lack of requirement for these substrates. We have found that X DNA synthesis in vitro results in the accumulation of 11S fragments. It is unlikely that these fragments are a result of nuclease action, since most of the X DNA template is of considerably higher molecular weight. If these 11S fragments correspond to Okazaki fragments observed in vivo, our results suggest that the joining of these fragments is inefficient in the in vitro reaction. In another in vitro replication system using T7-infected cells, 11S fragments also accumulate (8). The absence of DNA polymerase I in vivo results in a more than 10-fold decrease in the rate of joining of the newly replicated DNA (27, 28) and suggests that this enzyme has a role in the completion of Okazaki fragments. Since we have used polAl-infected cells in these studies, the accumulation of 11S fragments in vitro may reflect the absence of DNA polymerase I. The presence of small 16S DNA in neutral sucrose gradients might be explained by a sensitivity of the DNA containing gaps to an endonuclease or to a displacement of duplex fragments as a result of re-initiations of synthesis.

Proc. Nat. Acad. Sci. USA 71

(1974)

This investigation was supported by United States Public Health Service Grant AI-06045, Grant NP-1C from the American Cancer Society, Incorporated, and N.I.H. Research Career Development Award GM13, 534 to C.C.R. 1. Smith, D. W., Schaller, H. E. & Bonhoeffer, F. J. (1970) Nature 226, 711-713. 2. Knippers, R. & Stratling, W. (1970) Nature 226, 713-717. 3. Moses, R. E. & Richardson, C. C. (1970) Proc. Nat. Acad. Sci. USA 67, 674-681. 4. Vosberg, H. P. & Hoffmann-Berling, H. (1971) J. Moi. Biol. 58, 739-753. 5. Ganesan, A. T. (1971) Proc. Nat. Acad. Sci. USA 68, 12961300 6. Schaller, H., Otto, B., Nusslein, V., Huf, J., Herrmann, R. & Bonhoeffer, F. (1972) J. Mol. Biol. 63, 183-200. 7. Wickner, W., Brutlag, D., Schekman, R. & Kornberg, A. (1972) Proc. Nat. Acad. Sci. USA 69, 965-969. 8. Hinkle, D. C. & Richardson, C. C. (1973) J. Biol. Chem., in press. 9. StrAtling, W., Ferdinand, F. J., Krause, E. & Knippers, R. (1973) Eur. J. Biochem. 38, 160-169. 10. Schnos, M. & Inman, R. B. (1970) J. Mol. Biol. 51, 61-73. 11. Stevens, W. F., Adhya, S. & Szybalski, W. (1971) in The Bacteriophage Lambda, ed. Hershey, A. D. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 515533. 12. Okazaki, R., Okazaki, T., Sakabe, K., Sugimoto, K., Kainuma, R., Sugino, A. & Iwatsuki, N. (1968) Cold Spring Harbor Symp. Quant. Biol. 33, 129-143. 13. Tomizawa, J. & Ogawa, T. (1968) Cold Spring Harbor Symp. Quant. Biol. 33, 533-551. 14. Ginsburg, B. & Hurwitz, J. (1970) J. Mol. Biol. 52, 265280. 15. Gross, J. (1972) Curr. Top. Microbiol. Immunol. 57, 39-74. 16. Ogawa, T. & Tomizawa, T. (1968) J. Mol. Biol. 38, 217225. 17. Dove, W. F., Hargrove, E., Ohashi, M., Haugli, F. & Guha, A. (1969) Jap. J. Genet. 44, suppl. 1, 11-22. 18. Klein, A. & Powling, A. (1972) Nature New Biol. 239, 71-73. 19. Weiss, B., Live, T. R. & Richardson, C. C. (1968) J. Biol. Chem. 243, 4530-4542. 20. Sharp, P. H., Hsu, M-T, Ohtsubo, E. & Davidson, N. (1972) J. Mol. Biol. 71, 471-497. 21. Sippel, A. & Hartmann, G. (1968) Biochim. Biophys. Acta 157, 218-219. 22. Matsubara, K. & Kaiser, A. D. (1968) Cold Spring Harbor Symp. Quant. Biol. 33, 769-775. 23. Pettijohn, D. E. & Hanawalt, P. C. (1964) J. Mol. Biol. 8, 170-174. 24. Campbell, A. (1961) Virology 14, 22-32. 25. Tomizawa, J. (1971) in The Bacteriophage Lambda, ed. Hershey, A. D. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 549-552. 26. Sugino, A., Hirose, S. & Okazaki, R. (1972) Proc. Nat. Acad. Sci. USA 69, 1863-1867. 27. Kuempel, P. L. & Veomett, G. E. (1970) Biochem. Biophys. Res. Commun. 41, 973-980. 28. Okazaki, R., Arisawa, M. & Sugino, A. (1971) Proc. Nat. Acad. Sci. USA 68, 2954-2957.