Bacteriophage P2 Tin Protein Inhibits Phage T4 DNA ... - ScienceDirect

11 downloads 0 Views 1MB Size Report
A Novel Mechanism of Virus–Virus Interactions: Bacteriophage P2 Tin ... Thereby Tin protein inhibits either assembly or function, or both, of the T4 replisome.
VIROLOGY

230, 72–81 (1997) VY978464

ARTICLE NO.

A Novel Mechanism of Virus–Virus Interactions: Bacteriophage P2 Tin Protein Inhibits Phage T4 DNA Synthesis by Poisoning the T4 Single-Stranded DNA Binding Protein, gp32 GISELA MOSIG,*,1 SIDNEY YU,† HEEJOON MYUNG,† ELISABETH HAGGA˚RD-LJUNGQUIST,‡ LAURA DAVENPORT,* KARIN CARLSON,§ and RICHARD CALENDAR† *Department of Molecular Biology, Vanderbilt University, Nashville, Tennessee; †Department of Molecular and Cell Biology, University of California, Berkeley, California; ‡Department of Genetics, University of Stockholm, Sweden; and §Department of Microbiology, University of Uppsala, Sweden Received November 9, 1996; returned to author for revision December 13, 1996; accepted January 24, 1997 P2 prophages have been known to inhibit DNA replication and growth of T-even phages. We show here that this inhibition is due to poisoning of the T-even single-stranded DNA binding protein gp32 by the product of the nonessential P2 tin gene. Synthesis of Tin protein from a gene cloned in a multicopy plasmid is necessary and sufficient to completely prevent de novo DNA replication and growth of wild-type T2 or T4 phage. We isolated more than 20 independent mutants that render T-even phages resistant to poisoning by the P2 Tin protein. In all of these mutants, which we call asp, Asp codon 163 of gene 32 is changed to a Gly or Asn codon. The mutant alleles are recessive; i.e., when wild-type and asp mutants coinfect the same host cells, most DNA replication is poisoned by P2 Tin protein. To explain our results, we propose that the P2 Tin protein interacts with T-even gp32 at position 163 and distorts the helical filament of gene 32 protein on single-stranded DNA. Thereby Tin protein inhibits either assembly or function, or both, of the T4 replisome. The inhibition of late gene expression by P2 Tin protein may be an indirect consequence of inhibition of DNA replication. q 1997 Academic Press

INTRODUCTION

al., 1979b; Tomizawa, 1967; Williams et al., 1994). Gp32 interacts both with DNA and with other proteins that drive these processes (Formosa and Alberts, 1984; Kreuzer and Morrical, 1994; Mosig, 1994a; Mosig and Breschkin, 1975; Mosig et al., 1979b; Mosig et al., 1984; Wheeler et al., 1996). It has been known that prophage P2, resident in Shigella dysenteriae, interferes with T-even phage development (Bertani, 1953; Lederberg, 1957) by preventing T2 DNA synthesis and limiting RNA and protein synthesis to a few minutes after infection (Smith et al., 1969). For unknown reasons this inhibition by P2 prophage is less severe in Escherichia coli B. The inhibiting P2 gene(s) remained unknown. Sequencing of the last unknown segment of P2 DNA (Calendar et al., 1997; Genbank Accession No. X99628) revealed an open reading frame, previously called orf94 (Linderoth et al., 1993). We found that this orf is responsible for the inhibition. Therefore we now call this P2 gene tin (for T-even inhibition). Here we describe the nature of the inhibition.

Many plasmids and prophages express genes whose products inhibit or restrict other replicons. Such inhibitory mechanisms confer upon the host bacterium protection against virulent viruses, either directly or by reducing the concentrations of such viruses in the environment (Molineux, 1991; Parma et al., 1992; Shub, 1994; Snyder and Kaufmann, 1994; Snyder, 1995). The best known inhibitory mechanisms are based on restriction-modification systems designed to degrade foreign DNA (Revel, 1983; Carlson et al., 1994; Raleigh et al., 1991). Other mechanisms inhibit translation (Yu and Snyder, 1994), transcription (Herman and Snustad, 1982; Kutter et al., 1984; Kutter et al., 1981), or lead to destruction of the host’s membrane potential (Parma et al., 1992; Snyder, 1995; Snyder and Kaufmann, 1994). Here we describe a novel mechanism whereby phage P2 lysogens inhibit replication of the virulent bacteriophage T4. Such lysogens produce a protein that targets the major singlestranded DNA binding protein of T4 (Alberts and Frey, 1970; Williams et al., 1994). This protein, the product of gene 32 (gp32), is essential for DNA replication, recombination, and repair (Alberts and Frey, 1970; Bernstein and Wallace, 1983; Epstein et al., 1964; Kozinski and Felgenhauer, 1967; Mosig, 1985; Mosig et al., 1979a; Mosig et

MATERIALS AND METHODS Phage and bacteria Wild-type T4D and the following T4 mutants: amN130 (gene 46, recombination nuclease); amN116 (gene 39, DNA topoisomerase); amH17 (gene 52, DNA topoisomerase); amC5 (gene 59, DNA-helicase-loading protein); amN134 (gene 33, RNA polymerase accessory protein);

1 To whom correspondence and reprint requests should be addressed. Fax: (615) 343-6707. E-mail: [email protected].

0042-6822/97 $25.00

72

Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

vira

AP: Virology

PHAGE P2 TIN POISONS PHAGE T4 gp32

amH39 (gene 30, DNA ligase); amE219 (gene 61, primase); amE727 and tsC9 (both gene 49, recombination endonuclease VII); amN81 (gene 41, DNA helicase); amE315 and amA453 (both gene 32, ssDNA binding protein); amE10 (gene 45, DNA sliding clamp); and amC64 (gene 55, sigma factor for late transcription), originally isolated and obtained from A. H. Doermann and R. S. Edgar, have been backcrossed to wild-type T4 and maintained in our labs. The gene 45 and gene 55 mutants had been backcrossed three or five times, respectively, by John Wiberg. The T4 gene 32 mutants asp1 (Asp163 Gly) and asp5 (Asp163Asn) and 25 other asp mutants of T4 and T2 phages with one or the other of the same base changes were isolated and characterized during the present studies based on their ability to grow in E. coli B strains containing the plasmids pYMD2 or pYMD3 (Fig. 1), on recombinant frequencies and on their DNA sequence. Proportions of recombinants were determined, and double mutants were isolated from the phage progenies of standard crosses (Mosig et al., 1977) between single mutants. E. coli B (supo) and UT481 (supD) (from C. Lark, University of Utah) were used to propagate T4 wild-type or am mutants, respectively. They have been maintained in our labs for many years. E. coli B was transformed with pYMD2, pYMD3, or pYMD4 and E. coli UT481was transformed with pYMD2 by standard methods (Sambrook et al., 1989). E. coli B or UT481 bearing pYMD2 or pYMD3 do not support growth of wild-type T-even phages, whereas those bearing pYMD4, in which translation initiation of tin is defective, permit T-even growth. Plasmids pYMD2 (Calendar et al., 1997) was generated by cloning a ScaI – Tth111 fragment of P2 DNA (positions 30,133 to 31,590; Fig. 1) into the polylinker of pUC19. This plasmid contains the P2 tin gene and an additional orf, orf91. The direction of transcription of orf91 is opposite to that of tin (Fig. 1). Orf 91 was removed in pYMD3 by deleting an internal StuI – EcoRV fragment from pYMD2 (Fig. 1). To generate pYMD4, in which tin is not translated, the tin-containing fragment of pYMD2 was transferred to pUC119, and single-stranded DNA was made by M13 infection. Mismatched oligonucleotide mutagenesis (Sambrook et al., 1989) was used to introduce a SmaI site at the translation initiation region of the tin gene. The oligonucleotide was 5*ATAAAAGGTGTTCCCGGGAATAACATGGAT3*. The SmaI site, 5*CCCGGG3*, replaces the nucleotides 5*GATATG3* and alters the initiation codon for tin.

73

quencing in a Perkin–Elmer thermocycler using reagents and protocols of the Promega fmol sequencing kit according to the vendor’s protocol, as described (Mosig and Colowick, 1995). Measuring T4 DNA synthesis Total de novo T4 DNA replication was monitored by incorporation of [3H]thymidine (ICN) into TCA precipitable material as described (Mosig and Colowick, 1995). Measuring T4 protein synthesis E. coli B carrying either pUC19 or pYMD2 were grown at 377 in M9 (Carlson and Miller, 1994) containing 0.5% glucose, 100 mg/ml casamino acids (CAA), 1 mg/ml thiamine–HCl, and 50 mg/ml carbenicillin to about 3 1 107 cells/ml (determined in a Petroff-Hauser cell under a microscope). The cells were then pelleted by centrifugation, resuspended in prewarmed fresh medium of the same composition, except that CAA were present at 20 mg/ml, and grown to a density of about 2 1 108 cells/ml. The cells were then infected with an m.o.i. of approximately six with T4 wild type, asp1, 32am315, or 45amE10– 55amC64 mutants. At the time of infection all cells carried the plasmid, as determined from their ability to form colonies in the presence or absence of ampicillin. As expected from the m.o.i., less than 1% of the bacteria were able to form colonies 4 min after infection. At different times after infection 2-ml portions of the infected bacteria were labeled for 4 min by adding 2 mCi [14C]-L-casamino acids (Dupont-New England Nuclear) together with 26 mg cold CAA per milliliter. Labeling was terminated by pipetting the cells into 2 ml ice-cold 0.6 M trichloroacetic acid. The samples were processed for electrophoresis as described (Cardillo et al., 1979), except that the reducing agent was not added until the samples were to be analyzed. Gels were 15% polyacrylamide–0.4% bisacrylamide gels in Tris–glycine (Sambrook et al., 1989). Acrylamide–bisacrylamide (30%) solution (Catalogue No. ELCR1DC01) was purchased from Millipore. Gels were dried and subjected to PhosphorImager analysis and autoradiography. Measuring the integrity of T4 DNA

DNA, obtained from T4 particles by boiling them in distilled water for 5 min and avoiding subsequent harsh pipetting, was used as template for repetitive primer se-

Samples of the same infected cells that were used for protein labeling were taken 30 min after infection. Intracellular DNA was isolated as described (Krabbe and Carlson, 1991) and electrophoresed in 1% agarose in 0.178 M Tris–borate, 0.178 M boric acid, 0.004 M Na2H2 EDTA, pH 8.0. Gels were blotted onto nylon membranes, which were hybridized with T4 and lambda DNA labeled by random-priming with [a-32P]dCTP (DuPont-New England Nuclear) and subjected to PhosphorImager analysis and autoradiography as described (Krabbe and Carlson, 1991).

AID

vira

DNA sequencing of T4 mutants

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology

74

MOSIG ET AL.

FIG. 1. Genetic map of P2 phage, updated from Lindqvist, Deho´, and Calendar (1993). Arrows represent the directions of transcription. The DNA content of tin-containing plasmids pYMD2 and pYMD3 is shown as striped bars.

RESULTS The phage P2 tin gene is necessary and sufficient to inhibit T-even phages

Rho-factor-dependent transcription termination, thus reducing tin transcription. The target of the P2 tin gene is the T-even gene 32 protein

The P2 tin gene is located between P2 genes A and old (Fig. 1; Calendar et al., 1997). A DNA segment encompassing tin was cloned into pUC19 to give plasmid pYMD2, which contains an additional open reading frame of unknown function, orf91, that can be transcribed from the opposite DNA strand. This additional DNA segment was deleted in plasmid pYMD3 (Fig. 1), which contains tin as the only P2 gene. Tin is predicted to encode a protein of 253 amino acid residues. In the prophage, tin is cotranscribed with the old gene, which is neither present in pYMD2 nor in pYMD3. We surmise that in these plasmids tin is transcribed from a plasmid promoter. The multicopy plasmids pYMD2 or pYMD3 completely prevent plaque formation of wild-type T6, T4, or T2, the so-called T-even phages, in E.coli B, indicating that expression of the tin gene is necessary and sufficient for inhibition of the T-even phages by P2. The essential role of tin in T4 inhibition was also confirmed by changing the initiation codon of tin, AUG, to GGG by oligonucleotidedirected mutagenesis as described under Materials and Methods to give plasmid pYMD4. As predicted, in pYMD4-bearing E. coli B, both wild-type T2 and wild-type T4 phage can grow. The different extent of inhibition exerted by P2 prophage as compared with multicopy tin clones suggests that inhibition depends on the level of tin expression in different situations. In the prophage state the copy number of the tin gene is less than in the plasmid-bearing cells. Moreover, tin has no promoter of its own, relying on the old promoter for expression. The 150 nucleotide spacer between old and tin could cause

T2 and T4 mutants that can grow in E. coli bearing pYMD2 were found with frequencies of less than 1007 in each of more than 20 independent wild-type T4 or T2 lysates which had been grown from single plaques. We call these T4 mutants that overcome the effects of the P2 tin gene, asp (aborts sensitivity to P2). Crosses of the first isolated asp1 mutant with am mutants in many T4 genes at different map positions revealed no linkage with asp, except for mutants in genes 38, 34, 33, and 30 (data not shown), suggesting that asp mutations are located in or near T4 gene 32. Subsequent crosses of several asp mutants with gene-32 am mutants A453 and E315 yielded less than 0.7% wild-type recombinants, i.e., am/ progeny that could not grow in pYMD2-bearing supo host bacteria (Table 1). Because the recombinant frequencies indicated that the asp mutations are located between the am mutations A453 at codon 116 and E315 at codon 206, or downstream of E315 (Table 1), we sequenced the entire gene 32 of asp1 and the DNA between A453 and the end of gene 32 in asp2 through asp5 and in A453 and E315. Remarkably, in each mutant the same Asp codon 163 of gene 32 is changed to an Asn or Gly codon. Subsequent sequencing between codons 135 and 254 of each one of the independent T2 or T4 mutants revealed that all asp mutants have one or the other of the same base changes at codon 163 (Table 2). We found only two additional differences between our sequences and the reported wild-type T4 gene 32 DNA sequence (Krisch and Allet, 1982): the reported Arg 138 codon CGC is CGT and

AID

vira

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology

PHAGE P2 TIN POISONS PHAGE T4 gp32

pUC19 or the tin plasmid pYMD2, contained similar proportions of high molecular weight T4 DNA (Fig. 3), indicating that phage T4 DNA was not fragmented as a result of tin expression. (Equal amounts of DNA were loaded in each lane of Fig. 3 to compensate for the lack of replication of wild-type T4 in tin-expressing host bacteria.) In contrast, and consistent with their ability to produce phage progeny, T4 asp mutants with either Gly or Asn substitutions at codon 163 of gene 32 have normal de novo DNA synthesis (Figs. 2A and 2B) and normal burst sizes (data not shown) in E. coli B carrying tin clones. In E. coli B without tin plasmids, T4 wild type or asp mutant DNA synthesis is indistinguishable (Fig. 3C), indicating that the asp mutations alone have no detrimental effects on interactions of gp32 with other T4 proteins required for DNA replication, recombination, or packaging (Mosig, 1994a; Mosig and Colowick, 1995). This conclusion is supported by the following results: we constructed double mutants containing asp5 in combination with second mutations in other DNA metabolism genes. In previous experiments (Mosig et al., 1979b), several of these second mutations had shown differential effects on DNA replication in combination with certain gene-32 ts mutations (at the permissive temperature for the single gene 32 mutants). We tested second mutations in gene 46 (proposed recombination nuclease), gene 49 (recombination endonuclease VII), gene 61 (primase), genes 52 and 39 (DNA topoisomerase), gene 41 (DNA helicase), and gene 59 (helicase-loader). The single mutants that we crossed with asp5 have apparent phenotypes called ‘‘DNA arrest, DA’’ (genes 46, 41, and 59); ‘‘DNA Delay, DD’’ (genes 52, 39, and 61) or ‘‘wild-type like’’ (gene 49) in E. coli B without tin plasmids (Mosig, 1994a; Mosig and Colowick, 1995). We found that these double mutants have the expected DNA arrest (DA) and wild-type like or DNA-delay (DD) phenotypes in E. coli B, both with or without tin plasmids (data not shown). As expected, none of the single mutants, except for the single asp mutant, showed de novo DNA synthesis in E. coli bearing tin plasmids. Together, these results suggested that the P2 Tin protein poisons T4 DNA replication by poisoning T4 gp32

TABLE 1 Recombinant Frequencies from Crosses between Gene 32 asp and am Mutants Progeny Parent 1

Parent 2

asp

am

wt

asp-am

asp1 asp5 asp7 asp8 asp10 asp11

amE315 amE315 amE315 amE315 amE315 amE315

248 205 111 98 223 216

151 149 134 150 174 183

1 1 2 1 1 1

0 1 2 1 2 0

1101

941

7

6 0.64%

asp1 asp5 asp7 asp8 asp10 asp11

amA453 amA453 amA453 amA453 amA453 amA453

416 153 257 226 195 247

278 259 151 174 217 161

2 1 0 0 0 2

1494

1240

5

75

0 0 0 0 0 2 2 0.26%

the Phe codon 139 TTT is TTC, both in our wild type and in the asp mutants. The T2 and T4 genes 32 and their products have nearly identical DNA and amino acid sequences (McPheeters et al., 1988). Since both mapping and DNA sequencing results agree in the positioning of the codon 163 asp mutations, we conclude that these mutations are responsible for the resistance to inhibition by P2 and that the corresponding wild-type T4 singlestranded DNA binding protein is the target of P2 Tin protein. The P2 Tin protein poisons T4 DNA replication of wild-type T4 but not of asp mutants There is little or no de novo wild-type T4 DNA synthesis in E. coli containing the tin plasmids pYMD2 (Fig. 2A) or pYMD3 (Fig. 2B). The lack of DNA synthesis is not due to fragmentation of the infecting T4 DNA. Phage T4 DNA, recovered from wild-type T4- infected cells carrying either

TABLE 2 Numbers of Different Independent Mutants with Changes at Nucleotide Positions 487 or 488, Altering Codon 163 from Asp to Asn or Gly, Respectively

Phage

No. of mutants

T4 T2 T4 T2

9 4 7 7

Position

Wild-type base pair

Mutant base pair

Wild-type codon

Mutant codon

487

GC

AT

GAT Asp

AAT Asn

488

AT

GC

GAT Asp

GGT Gly

Note. Numbering starts at the first base of the initiation codon. The sequences were obtained as described under Materials and Methods, using as primer 5*CCAAATCATCAGCCACTT3*.

AID

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

vira

AP: Virology

76

MOSIG ET AL.

FIG. 3. Integrity of phage T4 DNA in Tin-expressing cells. Samples of the same cultures analyzed in Fig. 4 were taken 30 min after phage infection, and total DNA was extracted and analyzed by agarose gel electrophoresis, blotting, and hybridization with a mixture of labeled T4 and lambda DNA. The lower content of phage T4 DNA in wild-typeinfected pYMD2-carrying cells (lane 3) was compensated for by loading more cell-equivalents onto the gel. The top band in the lambda BstEII digest (lane 5) corresponds to a 14.1-kb fragment. The bands seen in lanes 1–4 correspond to DNA ú20 kbp. The figure was prepared from PhosphorImager data through Adobe and Canvas software.

function in T4 DNA replication. Altering a critical aspartate residue at position 163 prevents the poisoning, without interfering with the interaction of gp32 with other proteins of T4 DNA replication or recombination. Because the known functions of gp32 depend on formation of multiprotein–nucleic acid helical filaments, we asked whether the asp mutants affected the sensitivity of wildtype gp32 to poisoning by P2 Tin protein. We coinfected tin-plasmid-bearing E. coli B hosts with both phages in varying ratios at the same total m.o.i.s. The data in Fig. 2B show clearly that in such bacteria, coinfected with wild-type and asp mutant T4, the P2 Tin protein poisons T4 DNA replication and, by implication, the gp32 helical filament, even when gp32 with a substitution at Asp 163 is present in the same cell; i.e., in genetic terms, the

FIG. 2. (A) Cumulative incorporation of [3H]thymidine into T4 DNA after infection of E. coli B, containing the multicopy tin plasmid pYMD2 (Fig. 1), with wild type (wt), asp1 or asp5 T4 phage at m.o.i.s of 5 at 377. (B) Cumulative incorporation of [3H]thymidine into T4 DNA after infection of E. coli B, containing the multicopy tin plasmid pYMD3 (Fig. 1), with wild-type T4 alone (6wt), asp5 alone (6 asp5) or mixtures of wt and asp5 T4 phage in different ratios as indicated, at total m.o.i.s of 6 at 377. (C) Cumulative incorporation of [3H]thymidine into T4 DNA after

infection of E. coli B, without any tin plasmid, with wild type (wt), asp1, or asp5 T4 phage at m.o.i.s of 5 at 377. Only approximately half as much radioactive thymidine was used in C as compared with A and B.

AID

vira

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology

PHAGE P2 TIN POISONS PHAGE T4 gp32

wild-type gp32 has a dominant negative effect on DNA replication in tin-containing E. coli. The residual DNA replication seen in the coinfection experiments at varying relative m.o.i.s (as compared with the lack of DNA replication of the same bacteria infected only with wild-type T4 at the same total m.o.i. of 6) is not much higher than the proportion of cells infected only with asp mutant phage (expected from the Poisson distribution). At most, a few cells, coinfected with a minority of wild-type and a majority of asp mutants, can synthesize some T4 DNA.

77

We have shown that the inhibition of growth of T-even phage in P2 lysogens is due to the poisoning of the Teven single-stranded DNA binding protein gp32 by the product of the P2 tin gene and, consequently, of T-even DNA replication and late gene expression. Mutations that alter the Asp codon 163 of gp32 render gp32 resistant to poisoning by P2 Tin protein. Our working model is that the gp32-DNA filament coordinates the interaction of other replication proteins (we envision this filament to function like a roller conveyor), that the P2 Tin protein disrupts this coordinating function, and that the asp mutants are insensitive to the disruption.

We suspect that P2 Tin protein would poison recombination-dependent DNA replication as well as origin-dependent replication, but because DNA replication is inhibited early, there is little recombination later on (Dannenberg and Mosig, 1981, 1983). Alternatively, one might consider that gp32 of the asp mutants would competitively inhibit wild-type gp32 from binding to DNA, thereby avoiding poisoning of the T4 protein–DNA filament by P2 Tin protein. We consider the latter alternative unlikely, because wild-type T4 is dominant or codominant over asp mutants in its susceptibility to P2 Tin. Similar dominant inhibitory effects of mutant derivatives of other proteins that form filaments on DNA (e.g., RecA protein) have been taken as evidence that the mutant proteins participate in the filaments and inactivate their function (Lauder and Kowalczykowski, 1993). The three simplest possible poisoning mechanisms of T4 gp32 by P2 Tin protein and the resistance of the T4 asp mutants include the following: (1) Asp163 of T4 gp32 is the direct target for interactions with P2 Tin, either because Tin is a DNA binding protein that can interact with gp32 or because Tin sits backpack on the filament. In either case, Tin would distort the gp32–DNA filament. If multiple conformations of T4 gp32 (as they are known to exist for E. coli Ssb or RecA proteins) are important for its functions, binding of Tin might inhibit transitions between different conformational states. (2) Asp163 can be phosphorylated by Tin and the phosphorylation affects interactions with other proteins or degradation. (3) Asp 163 is the target of a protease, which could be either Tin itself or be activated by Tin. These possibilities, which are not mutually exclusive, are under investigation. That Tin itself might be a singlestranded DNA binding protein is suggested by comparing the positions of critical basic and aromatic amino acids in a segment of the Tin protein and in several other single-stranded DNA binding proteins, aligned in Table 3 (modified from Wang and Hall, 1990). In one of these proteins the contacts with DNA are visible in the 3Dstructure. Possibility (3) is unlikely, because in pulse– chase experiments we found that at least as much gp32 is synthesized in E. coli B with or without tin-containing plasmids and that gp32 of wild type and asp mutants have equal stability (H. E. Olivey and G. Mosig, unpublished results). Our model is based on previous genetic and biochemical evidence that the T4 gp32, a zinc protein that binds cooperatively to single-stranded DNA (Chase and Williams, 1986; Gauss et al., 1987; Karpel, 1990; Pan et al., 1989a and b; Shamoo et al., 1995 and 1995a; Shamoo et al., 1994; Spicer et al., 1979; von Hippel et al., 1982; Williams et al., 1994), also interacts with numerous other proteins involved in DNA replication, recombination, and repair (Breschkin and Mosig, 1977a and b; Burke et al., 1980; Formosa and Alberts, 1984; Hosoda et al., 1980;

AID

vira

The P2 Tin protein inhibits late T4 gene expression Protein synthesis is much reduced at late times (Fig. 4A) after infection of tin-bearing E. coli B bacteria with wild-type T4 as compared with the same bacteria infected with T4 asp mutants. No late T4 proteins are made after infection of tin-bearing bacteria with wild-type T4 (Fig. 4B). This is evident by comparison with proteins synthesized in tin-less bacteria infected with T4 gene 45–55 mutants (lane1), which are defective in DNA synthesis and do not produce the T4 sigma factor required for late gene expression and therefore no late proteins (Williams et al., 1994). The gene 32 am mutant E315 (lane 11) does not synthesize full-length gene 32 protein and instead overproduces the corresponding am peptide (arrowhead). A weak band that appears at the same position as wild-type gp32 in this mutant is due to the rIIB, rnh, and gene 44 proteins, which have similar molecular weights and comigrate with gp32. Because of defective DNA synthesis, the gene 32 mutant E315 also synthesizes little or no late proteins. The effect of Tin on T4 late protein synthesis can be a direct consequence of Tin on T4 gene expression or an indirect effect of the poisoning of DNA synthesis, since it is known that T4 late gene transcription requires concomitant DNA replication (Williams et al., 1994). It is possible, that there is an additional direct effect of Tin on T4 transcription. After infection with T4 mutants which cannot replicate DNA, synthesis of early proteins is prolonged. In contrast, in tin-bearing cells, most early proteins are shut off at late times. DISCUSSION

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology

78

MOSIG ET AL.

FIG. 4. Protein synthesis in the presence of plasmid-encoded Tin. E. coli B carrying either pUC19 (without tin) or pYMD2 (Fig. 1) was infected with different T4 strains, as shown above B. At different times after infection newly synthesized proteins were pulse-labeled for 4 min. (A) Total acid-insoluble radioactivity (cpm per 0.2 ml infected culture) in pYMD2-carrying cells infected with T4 wild type (dotted columns) or T4 asp1 (grey columns). (B) SDS–PAGE analysis of the proteins. The arrows to the right point to the expected positions for gp23 and its processed derivative gp23*, the major late capsid proteins synthesized in T4-infected cells, and gp32; the am peptide of the gene-32 mutant E315 (arrowhead) is overproduced. Band positions for the nonradioactive size markers in lane 6 are shown by horizontal lines. From top to bottom these correspond to 97, 66, 45, 31, 22, and 15 kDa. The pattern seen in pYMD2-carrying cells infected with asp1 was the same as that seen upon infection of pUC19carrying cells or plasmid-free E. coli B with either T4 wild type or the asp1 mutant (data not shown). B was prepared from the PhosphorImager data through Adobe and Canvas software.

AID

VY 8464

/

6a2f$$8464

03-01-97 03:04:00

vira

AP: Virology

PHAGE P2 TIN POISONS PHAGE T4 gp32

79

TABLE 3 Common Patterns of Basic and Aromatic Amino Acids in Sequences of Single-Stranded DNA Binding Proteins from Different Organisms Protein (position)

Sequence

References

R-X3-K-X4 -Y-X15-F-X4-Y-0 -K-X4 -Y-X3- K K-X3-K-X12-Y-X14-Y-X6-Y-X3-K-X4 -Y-X7- K R-X4-R-X4 -Y-X7 -Y-X6-Y-X4-K-X26-F-X6- R R-X5-K-X4 -Y-X7 -Y-X6-Y-X19K-X12-F-X6- R K-X5-K-X4 -W-X5 -F-X9-Y-X2-K-X4 -W-X7- R R-X3-R-X9 -F-X4 -F-X4-F-X1-R-X12-W-X6- R K-X3-R-X9 -F-X4 -F-X4-F-X3-K-X10-W-X6- R K-X3-R-X12-F-X3 -Y-X4-F-X3-K-X6 -W-X6- R R-X3-K-X3 -F-X36-Y-X13F-O -K-X16-W-X11-R

P2 Tin (26–66) T4 gp32 (67–123) fd gp5 (16–80) P Ike (16–81) Eco Ssb (43–86) VZV Ssb (801–847) HSV1ICP8 (803–849) EBV Ssb (741–771) AD5 Ssb (410–499)

(Calendar et al., 1997) (Krisch and Allet, 1982) (Nakashima et al., 1974) (Peeters et al., 1983) (Sancar et al., 1981) (Davison and Scott, 1986) (Wang and Hall, 1990) (Baer et al., 1984) (Kruijer et al., 1981)

Note. Single-letter amino acid abbreviations are used. Xs followed by numbers indicate the number of unspecified amino acids between the aligned conserved basic or aromatic residues (bold).

Hosoda and Moise, 1978; Huberman et al., 1971; Kreuzer and Morrical, 1994; Mosig, 1985; Mosig, 1994b; Mosig et al., 1977; Mosig and Bock, 1976; Mosig and Breschkin, 1975; Mosig et al., 1979b; Mosig, Shaw, and Garcia, 1984). Cooperative binding to single-stranded DNA is a prerequisite for most, if not all, gp32 functions and translational autoregulation adjusts the concentration of gp32 to the abundance of single-stranded DNA segments during viral development (Chase and Williams, 1986; McPheeters et al., 1988; Karpel, 1990; Shamoo et al., 1995 and1995a; Shamoo et al., 1994; Spicer et al., 1979; von Hippel et al., 1982; von Hippel et al., 1983; Williams et al., 1994). The crystal structure of gp32 (Shamoo et al., 1995 and 1995a) has revealed a major DNAbinding cleft and the position of zinc. Neither the N-terminal domain, required for cooperativity, nor the flexible Cterminal domain (Burke et al., 1980; Hosoda et al., 1980; Hosoda and Moise, 1978; Spicer et al., 1979) are visible in the published 3D structure (Shamoo et al., 1995 and 1995a). Aa 163 is far away from any of these regions, making it plausible that it might be the direct target for interactions with Tin. Within the framework of our model, P2 lysogens in E. coli B inhibit T-even growth less severely than in E. coli with multicopy plasmids expressing P2 tin, because they produce less Tin protein, thus causing fewer distortions in the protein–DNA filaments. The reasons why P2 prophages are less detrimental for T4 growth in E. coli B than in Shigella are not known. One possible interpretation, consistent with our proposed model, is based on the presence of a defective P2-related prophage in E. coli B (Cohen, 1959). A defective Tin protein from this prophage might interact with T4 gp32 without poisoning it, thus preventing the detrimental effects of wild-type P2 Tin protein. Remarkably, in P2 lysogens tin is cotranscribed with the old gene, which interferes with growth of lambdoid phages albeit by a different mechanism. Both tin and old are dispensable for P2 growth. They are presumably of

AID

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

selective advantage by ‘‘altruistic death’’ (Parma et al., 1992; Shub, 1994; Snyder and Kaufmann, 1994). Although the infected bacteria are killed by T4 or lambda infection, they do not produce more phage to kill yet uninfected bacteria of a population. Moreover, old and tin have a much higher AT content than the essential P2 genes, and they are not present in all P2-related phages, suggesting that they have been acquired by P2 from other organisms via horizontal transfer (Calendar et al., 1997). The novel mechanism that we propose here for inhibition of the virulent T-even phages by prophage P2 expands the repertoire that viruses use in the evolutionary struggle for survival. ACKNOWLEDGMENTS This work was supported by NIH research grants from the National Institute of General Medicine GM13221 to G.M. and AI-08722 from the National Institute of Allergy and Infectious Diseases to R.C., and by grants 72 and B96-13X-11577-01A from the Swedish Medical Research Council to E.H.L. and K.C., respectively. We thank Caulley Fonvielle for help with the experiments shown in Fig. 2B and with the preparation of this manuscript and Helen Revel for constructive criticism.

REFERENCES Alberts, B. M., and Frey, L. (1970). Bacteriophage gene 32: A structural protein in the replication and recombination of DNA. Nature (London) 227, 1313–1318. Baer, R., Bankier, T., Biggin, M. D., Deininger, P. L., Farrell, P. J., Gibson, T. J., Hatfull, G., Hudson, G. S., Satchwell, S. C., Seguin, C., Tuffnell, P. S., and Barrell, B. G. (1984). DNA sequence and expression of the B95-8 Epstein–Barr virus genome. Nature (London) 310, 207–211. Bernstein, C., and Wallace, S. S. (1983). DNA repair. In ‘‘Bacteriophage T4’’ (C. K. Mathews, E. Kutter, G. Mosig, and P. B. Berget, Eds.), pp. 138–151. Am. Soc. Microbiol., Washington, DC. Bertani, G. (1953). Lysogenic versus lytic cycle of phage multiplication. Cold Spring Harbor Symp. Quant. Biol. 18, 65–70. Breschkin, A., and Mosig, G. (1977a). Multiple interactions of a DNAbinding protein in vivo. I. Gene 32 mutations of phage T4 inactivate different steps in DNA replication and recombination. J. Mol. Biol. 112, 279–294. Breschkin, A., and Mosig, G. (1977b). Multiple interactions of a DNA-

vira

AP: Virology

80

MOSIG ET AL.

binding protein in vivo. II. Effects of host mutations on DNA replication of phage T4 gene-32 mutants. J. Mol. Biol. 112, 295–308. Burke, R. L., Alberts, B. M., and Hosoda, J. (1980). Proteolytic removal of the COOH terminus of the T4 gene 32 helix-destabilizing protein alters the T4 in vitro replication process. J. Biol. Chem. 255, 11484– 11493. Calendar, R., Yu, S., Myung, H., Barreiro, V., Odegrip, R., Carlson, K., Davenport, L., Mosig, G., Christie, G. E., and Hagga˚rd-Ljundquist, E. (1997). The lysogenic conversion genes of coliphage P2 have unusually high AT content. In ‘‘Symposium on Horizontal Gene Transfer’’ (M. Syvaanen, Ed.), in press. Cardillo, T. S., Landry, E. F., and Wiberg, J. S. (1979). RegA protein of bacteriophage T4D: identification, schedule of synthesis, and autogenous regulation. J. Virol. 32, 905–916. Carlson, K., and Miller, E. S. (1994). Working with T4. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 421–426. Am. Soc. Microbiol., Washington, DC. Carlson, K., Raleigh, E. A., and Hattman, S. (1994). Restriction and modification. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 369– 381. Am. Soc. Microbiol., Wahington, DC. Chase, J. W., and Williams, K. R. (1986). Single-stranded DNA binding proteins required for DNA replication. Annu. Rev. Biochem. 55, 103– 136. Cohen, D. (1959). A variant of phage P2 originating in Escherichia coli, strain B. Virology 7, 112–126. Dannenberg, R., and Mosig, G. (1981). Semiconservative DNA replication is initiated at a single site in recombination-deficient gene 32 mutants of bacteriophage T4. J. Virol. 40, 890–900. Dannenberg, R., and Mosig, G. (1983). Early intermediates in bacteriophage T4 DNA replication and recombination. J. Virol. 45, 813–831. Davison, A. J., and Scott, J. E. (1986). The complete DNA sequence of varicella-zoster virus. J. Gen. Virol. 67, 1759–1816. Epstein, R. H., Bolle, A., Steinberg, C., Kellenberger, E., Boy de la Tour, E., Chevalley, R., Edgar, R., Susman, M., Denhardt, C., and Lielausis, I. (1964). Physiological studies of conditional lethal mutants of bacteriophage T4D. Cold Spring Harbor Symp. Quant. Biol. 28, 375–392. Formosa, T., and Alberts, B. M. (1984). The use of affinity chromatography to study proteins involved in bacteriophage T4 genetic recombination. Cold Spring Harbor Symp. Quant. Biol. 49, 363–370. Gauss, P., Krassa, K. S., McPheeters, M.A., Nelson, M.A., and Gold, L. (1987). Zinc(II) and the single-stranded DNA binding protein of bacteriophage T4. Proc. Natl. Acad. Sci. USA 84, 8515–8519. Herman, R. E., and Snustad, D. P. (1982). Plasmid pR386 renders Escherichia coli cells restrictive to the growth of bacteriophage T4 unf mutants. J. Virol. 41, 330–333. Hosoda, J., Burke, L., Moise, H., Kubota, I., and Tsugita, A. (1980). The control of T4 gene 32 helix-destabilizing protein activity in a DNA replication complex. ICN-UCLA Symp. Mol. Cell. Biol. 19, 505–514. Hosoda, J., and Moise, H. (1978). Purification and physicochemical properties of limited proteolysis products of T4 helix destabilizing protein (gene 32 protein). J. Biol. Chem. 253, 7547–7555. Huberman, J. A., Kornberg, A., and Alberts, B. M. (1971). Stimulation of T4 bacteriophage DNA polymerase by the protein product of T4 gene 32. J. Mol. Biol. 62, 39–52. Hurley, J. M., Chervitz, S. A., Jarvis, T. C., Singer, B. S., and Gold, L. (1993). Assembly of the bacteriophage T4 replication machine requires the acidic carboxy terminus of gene 32 protein. J. Mol. Biol. 229, 398–418. Karpel, R. L. (1990). T4 bacteriophage gene 32 protein. In ‘‘The Biology of Nonspecific DNA Protein Interactions’’ (A. Revzin, Ed.), pp. 103– 130. CRC Press, Boca Raton, FL. Kozinski, A. W., and Felgenhauer, Z. Z. (1967). Molecular recombination in T4 bacteriophage deoxyribonucleic acid. II. Single-strand breaks

and exposure of uncompleted areas as a prerequisite for recombination. J. Virol. 1, 1193–1202. Krabbe, M., and Carlson, K. (1991). In vivo restriction: Sequence and structure of endonuclease II-dependent cleavage sites in bacteriophage T4 DNA. J. Biol. Chem. 266, 23407–23415. Kreuzer, K. N., and Morrical, S. W. (1994). Initiation of DNA replication. ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.) Am. Soc. Microbiol., Washington, DC. Krisch, H. M., and Allet, B. (1982). Nucleotide sequences involved in bacteriophage T4 gene 32 translational self-regulation. Proc. Natl. Acad. Sci. USA 79, 4937–4941. Kruijer, W., van Schaid, F. M., and Sussenbach, J. S. (1981). Structure and organization of the gene coding for the DNA binding protein of adenovirus type 5. Nucleic Acids Res. 9, 4439–4457. Kutter, E., Drivdahl, R., and Rand, K. (1984). Identification and characterization of the alc gene product of bacteriophage T4. Genetics 108, 291–304. Kutter, E. M., Bradley, D., Schenck, R., Guttman, B. S., and Laiken, R. (1981). Bacteriophage T4 alc gene product: general inhibitor of transcription from cytosine-containing DNA. J. Virol. 40, 822–829. Lauder, S. D., and Kowalczykowski, S. C. (1993). Negative co-dominant inhibition of recA protein function. J. Mol. Biol. 234, 72–86. Lederberg, S. (1957). Suppression of multiplication of heterologous bacteriophages in lysogenic bacteria. Virology 3, 496–513. Linderoth, N. A., Christie, G. E., and Hagga˚rd-Ljundquist, E. (1993). Bacteriophage P2: Physical, genetic and restriction map. In ‘‘Genetic Maps-Locus Maps of Complex Genomes’’ (S. J. O’Brien, Ed.), 6th ed., pp. 1.62–1.69. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Lindqvist, B. H., Deho´, G., and Calendar, R. (1993). Mechanisms of genome propagation and helper exploitation by satellite phage P4. Microbiol. Rev. 57, 683–702. McPheeters, D. S., Stormo, G. D., and Gold, L. (1988). Autogenous regulatory site on the bacteriophage T4 messenger RNA. J. Mol. Biol. 201, 517–535. Molineux, I. J. (1991). Host-parasite interactions: recent developments in the genetics of abortive phage infections. New Biol. 3, 230–236. Mosig, G. (1985). Bacteriophage T4 gene 32 participates in excision repair as well as recombinational repair of UV damages. Genetics 110, 159–171. Mosig, G. (1994a). Homologous recombination. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 54–82. Am. Soc. Microbiol., Washington, DC. Mosig, G. (1994b). T4 bacteriophage and related bacteriophages. In ‘‘Encyclopedia of Virology’’ (R. G. Webster, and A. Granoff, Eds.), Vol. 3, pp. 1376–1383. Academic Press, San Diego. Mosig, G., Berquist, W., and Bock, S. (1977). Multiple interactions of a DNA-binding protein in vivo. III. Phage T4 Gene-32 mutations differentially affect insertion-type recombination and membrane properties. Genetics 86, 5–23. Mosig, G., and Bock, S. (1976). Gene 32 protein of bacteriophage T4 moderates the activities of the T4 gene 46/47-controlled nuclease and of the Escherichia coli recBC nuclease in vivo. J. Virol. 17, 756– 761. Mosig, G., and Breschkin, A. M. (1975). Genetic evidence for an additional function of phage T4 gene 32 protein: Interaction with ligase. Proc. Natl. Acad. Sci. USA 72, 1226–1230. Mosig, G., and Colowick, N. (1995). DNA replication of bacteriophage T4 in vivo. In ‘‘Methods in Enzymology’’ (J. Campbell, Ed.), Vol. 262, pp. 587–604. Academic Press, San Diego. Mosig, G., Dannnenberg, R., Ghosal, D., Luder, A., Benedict, S., and Bock, S. (1979a). General genetic recombination in bacteriophage T4. Stadler Genet. Symp. 11, 31–35.

AID

vira

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology

PHAGE P2 TIN POISONS PHAGE T4 gp32

81

Mosig, G., Luder, A., Garcia, G., Dannenberg, R., and Bock, S. (1979b). In vivo interactions of genes and proteins in DNA replication and recombination of phage T4. Cold Spring Harbor Symp. Quant. Biol. 43, 501–515. Mosig, G., Shaw, M., and Garcia, G. M. (1984). On the role of DNA replication, endonuclease VII, and rII proteins in processing of recombinational intermediates in phage T4. Cold Spring Harbor Symp. Quant. Biol 49, 371–382. Nakashima, Y., Dunker, A. K., Marvin, D. A., and Konigsberg, W. (1974). The amino acid sequence of a DNA-binding protein, the gene 5 product of fd filamentous bacteriophage. FEBS Lett. 43, 125. Pan, T., King, G. C., and Coleman, J. E. (1989a). 1H NMR studies of T4 gene 32 protein: effects of zinc removal and reconstitution. Biochemistry 28, 8828–8832. Pan, T., King, G. C., and Coleman, J. E. (1989b). Comparision of cooperative and isolated site binding of T4 gene 32 protein to ssDNA by 1H NMR. Biochemistry 28, 8833–8839. Parma, D. H., Snyder, M., Sobolevski, S., Nawroz, M., Brody, E., and Gold, L. (1992). The Rex system of bacteriophage lambda: tolerance and altruistic cell death. Genes Dev. 6, 497–510. Peeters, B. P. H., Nakashima, Y., and Konigsberg, W. H. (1983). Characterization of the DNA binding protein encoded by the N-specific filamentous Escherichia coli phage IKe. Binding properties of the protein and nucleotide sequence of the gene. J. Mol. Biol. 169, 197– 215. Raleigh, E. A., Benner, J., Bloom, F., Braymer, H. D., DeCruz, E., Dharmalingam, K., Heitman, J., Noyer-Weidner, M., Piekarowicz, A., Kretz, P. L., Short, J. M., and Woodcock, D. (1991). Nomenclature relating to restriction of modified DNA in Escherichia coli. J. Bacteriol. 173, 2707–2709. Revel, H. R. (1967). Restriction of non-glucosylated T-even bacteriophages: properties of permissive mutants of Escherichia coli B and K-12. Virology 31, 688–701. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). ‘‘Molecular Cloning: A Laboratory Manual,’’ 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sancar, A., Williams, K. R., Chase, J. W., and Rupp, W. D. (1981). Sequences of the ssb gene and protein. Proc. Natl. Acad. Sci. USA 78, 4274–7278. Shamoo, Y., Friedman, A. M., Parsons, M. R., Konigsberg, W. H., and Steitz, T. A. (1995). Crystal structure of a replication fork singlestranded DNA binding protein (T4 gp32) complexed to DNA. Nature 376, 362–366. Shamoo, Y., Friedman, A. M., Parsons, M. R., Konigsberg, W. H., and Steitz, T. A. (1995a). Crystal structure of a replication fork singlestranded DNA binding protein (T4 gp32) complexed to DNA (Erratum). Nature 376, 616. Shamoo, Y., Williams, K. R., and Konigsberg, W. H. (1994). The function of zinc(II) in gene 32 protein (gp32). In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall,

F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.). Am. Soc. Microbiol., Washington, DC. Shub, D. A. (1994). Bacterial altruism? Curr. Biol. 4, 555–556. Smith, H. S., Pizer, L. I., Pylkas, L., and Lederberg, S. (1969). Abortive infection of Shigella dysenteriae P2 by T2 bacteriophage. J. Virol. 4, 162–168. Snyder, L. (1995). Phage-exclusion enzymes: a bonanza of biochemical and cell biology reagents? Mol. Microbiol. 15, 415–420. Snyder, L., and Kaufmann, G. (1994). T4 phage exclusion mechanisms. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 391–396. Am. Soc. Microbiol., Washington, DC. Spicer, E. K., Williams, K. R., and Konigsberg, W. H. (1979). T4 gene 32 protein trypsin-generated fragments: Fluorescence measurement of DNA-binding parameters. J. Biol. Chem. 254, 6433–6436. Tomizawa, J. (1967). Molecular mechanisms of genetic recombination in bacteriophage: Joint molecules and their conversion to recombinant molecules. J. Cell Physiol. 70(Suppl.1), 201–214. von Hippel, P. H., Kowalczykowski, S. C., Lonberg, N., Newport, J. W., Paul, L. S., Stormo, G. D., and Gold, L. (1983). Autoregulation of expression of T4 gene 32: A quantitative analysis. In ‘‘Bacteriophage T4’’ (C. K. Mathews, E. M. Kutter, G. Mosig, and P. B. Berget, Eds.), pp. 202–207. Am. Soc. Virol., Washington, DC. von Hippel, P. H., Kowalczykowski, S. C., Lonberg, N., Newport, J. W., Paul, L. S., Stormo, G. D., and Gold, L. (1982). Autoregulation of gene expression. Quantitative evaluation of the expression and function of the bacteriophage T4 gene 32 (single-stranded DNA binding) protein system. J. Mol. Biol. 162, 795–818. Wang, Y., and Hall, J. D. (1990). Characterization of a major DNA-binding domain in the herpes simplex virus type 1 DNA-binding protein (ICP8). J. Virol. 64, 2082–2089. Wheeler, L. J., Ray, N. B., Ungermann, C., Hendricks, S. P., Bernard, M. A., Hanson, E. S., and Mathews, C. K. (1996). T4 phage gene 32 protein as a candidate organizing factor for the deoxyribonucleoside triphosphate synthetase complex. J. Biol. Chem. 271, 11156–11162. Williams, K. P., Kassavetis, G. A., Herendeen, D. R., and Geiduschek, E. P. (1994). Regulation of late-gene expression. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 161–175. Am. Soc. Microbiol., Washington, DC. Williams, K. R., Shamoo, Y., Spicer, E. K., Coleman, J. E., and Konigsberg, W. H. (1994). Correlating structure to function in proteins: gp32 as a prototype. In ‘‘Molecular Biology of Bacteriophage T4’’ (J. Karam, J. W. Drake, K. N. Kreuzer, G. Mosig, D. H. Hall, F. A. Eiserling, L. W. Black, E. K. Spicer, E. Kutter, K. Carlson, and E. S. Miller, Eds.), pp. 301– 304. Am. Soc. Microbiol., Washington, DC. Yu, Y.-T. N., and Snyder, L. (1994). Translational elongation factor Tu cleaved by a phage-exclusion system. Proc. Natl. Acad. Sci. USA 91, 802–806.

AID

vira

VY 8464

/

6a2f$$$101

03-01-97 03:04:00

AP: Virology