Identification, cloning, and characterization of the bacteriophage N4 gene encoding the single-stranded DNA-binding protein. A protein required for phage replication, recombination, and late transcription.

The coliphage N4-coded single-stranded DNA-binding protein (N4SSB) is essential for phage replication and for expression of the phage late genes, which are transcribed by the Escherichia coli sigma 70 RNA polymerase. As a first step in investigating the role of N4SSB in replication and transcriptional activation, we have identified and sequenced the N4SSB gene. The gene encodes a 265-amino acid protein with no apparent sequence homology to other single-stranded DNA-binding proteins. We present data indicating that N4SSB is also essential for phage recombination. Mutational analysis of the carboxyl terminus of the protein indicates that this region is required for protein-protein interactions with the N4 replication, N4 recombination, and E. coli transcriptional machineries, while the rest of the protein contains the determinants for single-stranded DNA binding.

The coliphage N4-coded single-stranded DNA-binding protein (N4SSB) is essential for phage replication and for expression of the phage late genes, which are transcribed by the Escherichia coli 70 RNA polymerase. As a first step in investigating the role of N4SSB in replication and transcriptional activation, we have identified and sequenced the N4SSB gene. The gene encodes a 265-amino acid protein with no apparent sequence homology to other single-stranded DNA-binding proteins. We present data indicating that N4SSB is also essential for phage recombination. Mutational analysis of the carboxyl terminus of the protein indicates that this region is required for protein-protein interactions with the N4 replication, N4 recombination, and E. coli transcriptional machineries, while the rest of the protein contains the determinants for single-stranded DNA binding.
Proteins that bind nonspecifically to single-stranded DNA with high affinity have been purified and characterized from several sources (1,2). Single-stranded DNA-binding proteins are present in high concentration in vivo and are essential components in a variety of DNA metabolism processes. They are required for DNA replication and are also involved in repair and recombination (1). They bind to single-stranded DNA stoichiometrically and, in most cases, with positive cooperativity (3).
The N4-coded single-stranded DNA-binding protein (N4SSB) 1 was originally detected as an activity capable of complementing the defect of mutant phage N4am7 in replication (4). The purified protein has a monomer M r of 32,000 and binds single-stranded DNA more tightly than RNA, with a binding site size of 11 nucleotides, an intrinsic binding constant of 3.8 ϫ 10 4 M Ϫ1 , and a cooperativity of 300 () in 0.22 M NaCl at 37°C (5). Although N4SSB is able to lower the melting transition of poly(dA)⅐poly(dT) by at least 60°C, it cannot lower the melting transition or assist in the renaturation of natural DNAs. In vitro, N4SSB specifically stimulates the N4 DNA polymerase by increasing its processivity 300-fold and by melting out hairpin structures that block polymerization (5). Surprisingly, N4SSB is also required for the synthesis of N4 late RNAs, which is catalyzed by the Escherichia coli 70 RNA polymerase (6). In vitro, N4SSB allows E. coli 70 RNA polymerase to utilize efficiently N4 late promoters (6). Therefore, in addition to playing an essential role in N4 DNA replication, N4SSB is a transcriptional activator, suggesting that it is a multifunctional protein. To dissect the domains of N4SSB responsible for activation of N4 DNA polymerase on a primed template and of E. coli 70 RNA polymerase at N4 late promoters, we cloned and sequenced the gene encoding N4SSB and constructed an N4SSB expression vector. Expression of the cloned protein complemented N4am7 phage, carrying a mutation in the N4SSB gene, for N4 DNA replication and late transcription. Additionally, we have found that N4SSB is essential for N4 DNA recombination. Analysis of the different phenotypes of carboxyl-terminal deletion mutants indicates that determinants of protein-protein interactions reside in a short, basic, carboxyl-terminal domain, while the rest of the protein contains the determinants of single-stranded DNA binding. Moreover, the phenotype of certain mutations indicates that the different functions of N4SSB are carried out by separate determinants.

MATERIALS AND METHODS
Bacterial Strains and Phages-E. coli strains W3350 (F Ϫ , thi, Cou R , r ϩ , gal, lac) and W3350supF were used for N4 and N4am7 infection, respectively. Wild-type and amber N4 mutants were grown as described previously (7). E. coli W3350pcnB (8), provided by Dr. D. Kiino, was used as the host for the construction of the N4SSB expression plasmid. E. coli W3350pcnB(DE3), constructed by transduction of E. coli W3350(DE3) (John Dunn, Brookhaven National Laboratory (9)) to tetracycline resistance with a P1 lysate grown on a strain carrying pcnB linked to Tn10, was used for expression and in vivo testing of wild-type and mutant cloned N4SSBs. E. coli B strain BL21(DE3) (F Ϫ , ompT, r Ϫ B , m Ϫ B ) was used for purification of N4SSB. Preparation and Manipulations of DNAs-N4 DNA was prepared according to Vander Laan et al. (10). For large-scale preparation of double-stranded circular plasmid and M13 replicative form DNAs, cells were lysed with alkali and purified by equilibrium centrifugation in CsCl-ethidium bromide gradients or by precipitation with polyethylene glycol according to Sambrook et al. (11). Small-scale, quick isolation of plasmid DNA was carried out using the alkaline lysis method (11). Oligonucleotides used for DNA sequencing, primer extension, oligonucleotide site-directed mutagenesis, and PCR were synthesized at the Oligonucleotide Facility at The University of Chicago or purchased from Operon Technologies, Inc. Preparation of N4 DNA blots and their hybridization to labeled DNA were as described previously (7).
PCR amplification was performed on template DNAs (20 ng of plasmid DNA and 200 ng of wild-type N4 and am7 genomic DNAs) suspended in a mixture containing 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl 2 , 0.01% gelatin, 0.2 mM dNTPs, 1 mM primers, and 5 units of Taq polymerase (Perkin-Elmer) in a total volume of 100 l, in an amplifying thermocycler (Perkin-Elmer). The DNA sequence of all PCR products was confirmed by the dideoxy chain termination method of Sanger et al. (12) after cloning of the fragment into M13.
Cloning of N4SSB and Construction of Expression Plasmids-N4 DNA HpaI restriction fragments were separated by electrophoresis on a 1% vertical agarose gel (7), and the N4 HpaID fragment was isolated by electroelution and inserted into the SmaI site of pKK232-8 (provided by Jü rgen Brosius, Mount Sinai School of Medicine) (13). The ability of cloned DNA fragments to rescue phage carrying the N4SSBam7 mutation was determined as described previously (14).
The DNA fragment containing the N4SSBam7 open reading frame (ORF) and the distal 100 noncoding bases was amplified by PCR using N4am7 DNA as a template and the following oligonucleotides as primers.
NcoI N4SSB-N: 5Ј-GGAATTCCATGGTGAGCAATTTATTCGGAAATCTG-3Ј EcoRI valine N4SSB-C: 5Ј-CAAGCTTCAGGTCTAATCCAGAGAGTATTCCG-3Ј HindIII N4SSB-N generates an additional valine residue as the second amino acid (underlined). After confirmation of the DNA sequence, the desired DNA fragment was inserted into pT5 (provided by Steve Eisenberg, Synergen) to yield pMC3, which carries the N4SSBam7 ORF under T7 RNA polymerase promoter control. Clones expressing the wild-type ORF were not recovered by this procedure (see "Results").
To clone the wild-type N4SSB ORF, a DNA fragment carrying the Shine-Dalgarno (SD) sequence present in pT5 and the amino-terminal region of the N4SSB gene (120 amino acids) was generated by PCR, using N4 DNA and the following oligonucleotides as primers.
The PCR fragment was restricted with XbaI and HindIII and inserted into the same sites of mp19 to verify the DNA sequence. The XbaI/BstBI fragment (N4SSB XB ) was used to replace the corresponding fragment (carrying the am7 mutation) in pMC3, yielding pMC5. Finally, pMC5 was treated with BglII and XbaI to release the fragment containing the T7 RNA polymerase promoter. This fragment was replaced with a 66-base-long DNA fragment containing the T7 minimal promoter and lac operator, isolated from pET-11a (Novagen) by restriction with the same enzymes, to yield pMC6. N4SSB was purified as described (5) with some modifications. 2 Deletion mutants were generated by oligonucleotide-directed, site-specific mutagenesis of pMC6. Isolation and Analysis of Intracellular DNAs, Measurement of N4 DNA, and Late RNA Synthesis-E. coli W3350pcnB(DE3)/pLysE bearing the N4SSB expression vector pMC6 was grown at 37°C to A 620 ϭ 0.4 in minimal salt medium (15) supplemented with 0.2% casamino acids (Difco) and 0.01 mg/ml thiamine. Cells were collected by centrifugation and resuspended in fresh medium. After a 15-min preincubation and a 30-min induction with 0.4 mM IPTG, cells were infected with N4am7 phage at a multiplicity of infection of 10. At different times after infection, 50 ml of cells were removed and processed as described (5).
To measure N4 DNA synthesis, at the indicated times after infection, 100-l samples of cells were removed and incubated for 2 min with 10 Ci of [ 3 H]thymidine (Amersham Corp.). Samples were processed as described previously (16). Late RNA synthesis was measured by primer extension as described (6).
Labeling of Proteins after Induction-Cells were grown as described above. After induction with 2 mM IPTG for 30 min at 37°C, cells (1 ml) were incubated for 90 min in the presence of 200 g/ml rifampicin at 37°C, and proteins were labeled with 10 Ci of [ 35 S]methionine (Amersham Corp.) or Tran 35 S-label (ICN) for 5 min at 37°C. Cells were collected by centrifugation and processed as described previously (16).
N4SSB Activation of N4 DNA Recombination-E. coli W3350pcnB(DE3)/pLysE bearing wild-type or mutant N4SSB expres-sion plasmids were grown at 37°C to A 620 ϭ 0.4 in LB medium and plated with the indicated amount of IPTG and appropriate antibiotics. N4am7 phage stocks were diluted to 10 2 -10 8 plaque-forming units/ml, and 10 l were spotted on lawns of wild-type and mutant N4SSBexpressing cells and cells carrying the vector plasmid. Individual plaques were resuspended and titered on W3350 and W3350supF. In all cases, the increase in the number of plaques upon IPTG induction was due to the generation of wild-type N4 recombinants.
N4SSB Stimulation of N4 DNA Polymerase Activity-In vitro stimulation of N4 DNA polymerase activity by wild-type or mutant N4SSB was determined as described previously (17).

RESULTS
Cloning and Sequencing of the N4 DNA Fragment Containing the Gene Encoding N4SSB-Previous experiments suggested that the gene encoding N4SSB mapped to the HpaID (14). Since the 6.3-kb HpaID fragment is toxic to E. coli, it was inserted into the SmaI site of the multiple cloning site of pKK232-8, a pBR322 derivative containing strong E. coli transcription terminator sites (rrnB T 1 and T 2 ) flanking a multiple cloning site (13). Clones were isolated that hybridized to the HpaID fragment in N4 HpaI genomic blots and to the HpaID-a fragment of HaeIII-restricted HpaID blots (data not shown) (14). However, the inserts were all smaller than the original 6.3-kb HpaID fragment.
Positive clones were tested for rescue of N4am7, which carries a mutation in N4SSB, by recombination between the N4am7 genomic DNA and the cloned fragment (14,16). One clone, pMC1, containing a 2.5-kb fragment was able to rescue N4am7. The frequency of wild-type plaque-forming units/total plaque-forming units of a lysate from pMC1-containing W3350 cells was 1000-fold higher than the frequency in hosts carrying pKK232-8, indicating that at least part of the N4SSB gene was present in pMC1 (data not shown). Although pMC1 contains only 2.5 kb of the N4 HpaID fragment, the cloned fragment is not the result of a rearrangement since the same PCR products were generated using pMC1, wild-type N4, or N4am7 genomic DNA as templates for amplification with two oligonucleotides that hybridize to each end of the 2.5-kb fragment (data not shown). The presence of the N4SSB gene in pMC1 was confirmed using an oligonucleotide probe (oligonucleotide a in Fig.  1) derived from the sequence of the amino-terminal 35 amino acids of the N4SSB protein purified from N4-infected cells. This probe hybridized to both pMC1 and the N4 HpaID-a fragment on genomic blots (data not shown).
The same oligonucleotide probe was used as a primer for double-stranded DNA sequencing of pMC1, wild-type N4, and N4am7 genomic DNAs. The sequence revealed a 265-amino acid ORF (Fig. 1). The N4am7 mutation mapped at amino acid 54, a CAG (glutamine) to TAG (amber) transition, verifying that this ORF codes for N4SSB. The DNA sequence is in agreement with the determined amino acid sequence of the aminoterminal 34 amino acids of the purified N4SSB protein, underlined in Fig. 1. 3 The predicted sequence agrees well with the determined amino acid composition of the purified protein (5). Analysis of the derived protein sequence does not reveal sequence homologies to T4 gp32 or any other single-stranded DNA-binding proteins.
Construction of the N4SSB Expression Vector-Attempts to overproduce N4SSB using the T7 expression system (9)  sulted in polypeptides smaller than N4SSB due to premature termination of translation caused by mutations within the coding sequence for N4SSB. The largest cloned fragment coded for a 28-kDa polypeptide (pMC6 28 ), missing the carboxyl-terminal 35 amino acids due to the deletion of two adenines in a run of adenines at positions 688 -691 (Fig. 1). These mutations occurred even in the absence of active T7 RNA polymerase. In this case, expression of N4SSB originated from cryptic E. coli RNA polymerase promoters present between the T7 promoter and Shine-Dalgarno sequence of N4SSB (data not shown).
These results indicate that expression of N4SSB is lethal to E. coli. The N4am7 ORF was cloned (pMC3) to overcome this problem (see "Materials and Methods"). The successful isolation of pMC3 indicates that low levels of N4SSB protein are lethal to the cell. Cloning of the wild-type N4SSB ORF (pMC6) was successfully achieved using a tightly regulated system in which the N4SSB gene is under the control of a T7 RNA polymerase promoter and the lac operator. In addition, the recombinant plasmid was introduced into a host strain carrying the pcnB mutation, which reduces the copy number of pBR322 derivatives (8), and pLysE, which synthesizes the T7 lysozyme and is an inhibitor of T7 RNA polymerase. Fig. 2 shows the expression of N4SSB in E. coli W3350pcnB(DE3)/pLysE, which was used as a host for N4 phage infection to test the effect of cloned N4SSB on N4 DNA recombination, N4 DNA replication, and N4 late transcription in vivo, and in BL21(DE3)/pLysE, which was used for the overexpression and purification of cloned N4SSB. The size of the N4SSB protein produced by the T7 RNA polymerase-directed, expressing clone was the same as that from N4-infected cells. The amount of N4SSB is higher in E. coli BL21(DE3)/pLysE than in E. coli W3350pcnB(DE3)/pLysE due to a higher copy number of pMC6 in the former strain.
The ability of the cloned and expressed N4SSB protein to complement N4am7 for N4 DNA replication was examined. Fig. 3 (left panel) shows the rate of [ 3 H]thymidine incorporation into DNA after N4am7 infection of the following cells: suppressor (W3350supF), non-suppressor (W3350), W3350pcnB(DE3)/ pLysE carrying pMC6, or pMC6 28 without or with preincubation with 0.4 mM IPTG for 30 min. Even though the rate of [ 3 H]thymidine incorporation in cells expressing cloned N4SSB was lower than in suppressor-containing cells, it was ϳ4-fold higher than in noninduced or N4SSB 28 -expressing cells. The increased rate of thymidine incorporation occurred in N4SSBexpressing cells when cells had been preincubated with 0.2-0.5 mM IPTG for 20 -45 min. The amount of N4SSB, synthesized from the T7-directed clone induced with 0.1-1 mM IPTG for 15-90 min, was found to be comparable to that produced in N4am7-infected suppressor-containing cells (data not shown).
Southern blot analysis was performed to test whether the increased rate of thymidine incorporation in N4SSB-expressing cells was due to active N4 DNA replication, recombination, or repair. Total intracellular DNAs (host chromosomal, N4 genomic, and plasmid DNAs) were prepared from wild-type N4SSB-expressing cells that had been preincubated in the absence or presence of 0.4 mM IPTG for 30 min at three different times (8,18, and 35 min) after N4am7 infection. DNAs were restricted with XbaI and SalI, and fragments were separated on an agarose gel, blotted, and hybridized to an excess amount of 32 P-labeled N4 genomic DNA (see "Materials and Methods"). Fig. 3 (right panel) shows that the amount of N4 DNA increased (as indicated by the increased amount of N4 XbaI fragments and the joint fragment, a product of N4 DNA replication (18)) only when IPTG was present, i.e. when N4SSB was induced. The decrease or absence of N4 DNA XbaI fragments in noninduced cultures at times late after infection is indicative of DNA degradation. These results demonstrate that cloned and expressed wild-type N4SSB protein can complement N4am7 for N4 DNA replication. In contrast, N4SSB 28 is inactive.
N4SSB enhances the rate of DNA synthesis catalyzed by N4 DNA polymerase by increasing the processivity of N4 DNA polymerase and by melting out hairpin structures that block polymerization (5). The ability of N4SSB purified from overproducing cells to activate N4 DNA polymerase on a primed template was examined. Fig. 4 shows that N4SSB purified from overproducing cells activates N4 DNA polymerase on a primed template as efficiently as the protein purified from N4 phageinfected cells, while N4SSB 28 is inactive.
N4SSB Is Required for N4 DNA Recombination-In our attempts to map the N4SSB gene, we observed that plasmids carrying the wild-type N4SSB gene rescued N4am7, while plasmids carrying the 28-kDa N4SSB protein, lacking the carboxylterminal 35 amino acids, did not. These results suggest that active N4SSB is required for N4 recombination. To test this hypothesis, we measured the ability of different concentrations of cloned and expressed N4SSB to rescue N4am7 phage. Wildtype N4 phage were generated from recombination between N4am7 genomic DNA and the N4SSB ORF on the plasmid only when the resident plasmid expressed the wild-type N4SSB gene product (Table I, third line). No significant increase in wild-type phage was observed in the absence of induction or when N4SSBam7 or N4SSB 28 was expressed (third (no IPTG), fifth, and seventh lines). Expressed N4SSB increased N4 DNA recombination 10 4 -10 5 -fold. This increase is not due to increased replication since N4SSB mutants (⌬264 -265,K259R and ⌬245-259) are able to support recombination, but are deficient in DNA replication (Table II). These results indicate that N4SSB activity is required for N4 DNA recombination.
The Carboxyl-terminal Region of N4SSB Is Required for N4 DNA Replication and Activation of Recombination and Late Transcription-N4SSB 28 , which is missing the carboxyl-terminal 35 amino acids, is unable to complement N4am7 phage for N4 DNA replication and activation of N4 late transcription and N4 DNA recombination. However, N4SSB 28 binds to singlestranded DNA, although with reduced affinity. These results suggest that the carboxyl-terminal region of N4SSB might be required for interactions with the replication, recombination, and transcription machineries.
To define the role of the carboxyl terminus, we generated two types of deletions by site-specific mutagenesis: carboxyl-terminal truncations and internal, in-frame deletions. The sequence of the mutant N4SSB ORFs in each plasmid was confirmed by double-stranded DNA sequencing. Mutant proteins were cloned for overexpression in W3350pcnB(DE3)/pLysE. The size and amount of expressed mutant protein were determined, following IPTG induction and [ 35 S]methionine labeling, by SDS-polyacrylamide gel electrophoresis and autoradiography. All mutants were of the expected size, were expressed to the same degree, and were as stable as the wild-type protein (Fig. 5).
The ability of N4SSB mutants to complement N4am7 for N4 late transcription, to support N4 DNA recombination, and to activate N4am7 DNA replication was measured. The results of these experiments are presented in Table II.
Deletion of the three carboxyl-terminal amino acids generated a protein (N4SSB ⌬263-265 ) active in supporting replication, but inactive in recombination or late transcription. Deletion of an additional residue (N4SSB ⌬262-265 ) abolished all three activities. These results suggest that the carboxyl-terminal region of N4SSB is required for interactions with the N4 DNA replication, recombination, and transcriptional machineries. Furthermore, the properties of N4SSB ⌬263-265 indicated that it is possible to isolate N4SSB mutants in which the determinants for catalyzing all three activities can be separated. This hypothesis was confirmed when we characterized two internal, in-frame deletions: N4SSB ⌬245-254 and N4SSB ⌬245-259 . N4SSB ⌬245-254 shows reduced ability to support late transcription, whereas N4SSB ⌬245-259 is fully proficient in supporting recombination, while it cannot complement N4am7 in replication and late transcriptional activation to wild-type levels.
The defect in all mutants discussed above can be explained by impaired single-stranded DNA binding activity and/or a differential involvement of the single-stranded DNA binding activity of the protein in recombination, replication, and activation of late transcription. To rule out this possibility, the N4SSB ⌬262-265 protein, which is deficient in supporting replication, recombination, and late transcription, was purified, and its ability to bind to single-stranded DNA was estimated from gel shift experiments. N4SSB ⌬262-265 eluted at the same salt concentration (1.5 M NaCl) from single-stranded DNA-agarose as the wild-type protein. Since the binding site size of N4SSB is 11 Ϯ 2 nucleotides, a single-stranded DNA oligomer containing one binding site size (12-mer) was used as a template in gel shift experiments. To stabilize the interaction of the proteins and 12-mer DNA, complexes were covalently cross-linked by irradiation at 300 ergs/mm 2 . Single-stranded DNA-bound N4SSB complexes were analyzed on a native polyacrylamide gel (Fig. 6). N4SSB binding to a 12-mer produces a major retarded species and a minor species due to protein-protein interactions. N4SSB ⌬262-265 binds more efficiently than the wild type to the single-stranded 12-mer, but fails to form the second complex. These and other results (data not shown) indicate that N4SSB ⌬262-265 is deficient in N4SSB-N4SSB interactions. DISCUSSION We have succeeded in sequencing and cloning the gene for N4SSB, a protein required for viral DNA replication, activation of late transcription, and, as we demonstrate in this paper, phage recombination. N4SSB is 265 amino acids in length, and no sequence similarity to other single-stranded DNA-binding proteins is evident. Specifically, the acidic carboxyl-terminal region present in several single-stranded DNA-binding proteins is absent in the predicted N4SSB sequence (19). However, both N4SSB and T4 gp32 contain a series of similarly spaced aromatic and charged residues in the first 130 amino acids (20).
Our inability to clone the N4SSB gene suggests that it is highly toxic to E. coli. While it is not yet clear what the determinants of lethality are, two SSB functions might be involved: its single-stranded DNA binding activity and its ability to activate RNA polymerase at the N4 late promoters. Successful cloning of N4SSB required a tightly regulated expression system encompassing (a) deletion of weak E 70 promoters present upstream of the N4SSB translational start site, (b) use of the chromosomal pcnB mutation to reduce the plasmid copy number, (c) introduction of the T7 lysozyme carried on plasmid pLysE to inhibit T7 RNA polymerase, and (d) introduction of the lac operator sequence immediately downstream of the T7 promoter to prevent T7 RNA polymerase from binding to its promoter until IPTG is added.
The cloned and expressed protein was able to complement N4am7 in vivo for N4 DNA replication and N4 late transcription. In addition, expression of wild-type N4SSB from T7-directed expressing clones increased N4 DNA recombination 10 4 -10 5 -fold, indicating that N4SSB is required for N4 DNA recombination.
Even though cloned N4SSB was able to complement N4am7 for N4 DNA replication and N4 late transcription, the level of activation did not reach those observed after N4am7 infection of suppressor-carrying cells. We have considered several alternative explanations that can account for the lower level (20%) of replication and late transcription activation in cloned N4SSB-expressing cells. N4am7 is not a dominant negative mutation. The N4am7 mutation could exert polarity, affecting the expression of downstream gene products that might be required for N4 DNA replication and late transcription. The downstream region (2 kb) of the N4SSB-coding region was sequenced. Two ORFs (185 and 147 amino acids in length) are present. An expression plasmid (pMC8) carrying the wild-type N4SSB or am7 allele and the two downstream ORFs was constructed. The N4SSB protein and the expected products from the two downstream ORFs were expressed in E. coli W3350pcnB(DE3)/pLysE. These proteins were also expressed from pMC8am7 in E. coli W3350pcnB(DE3)/pLysE. 4 These results suggest that the am7 mutation does not affect the expression of the two downstream genes. The ability of cloned N4SSB and the two downstream gene products to complement N4am7 for N4 DNA replication and N4 late transcription was examined (data not shown). The levels of both activities were similar to those observed when only N4SSB was expressed. We suspect that the inability of cloned N4SSB to fully activate DNA replication and late transcription is due to the timing of N4SSB expression during N4 development under these conditions. The lower level of late transcription might also be the result of lower levels of template.
The N4SSB expression vector pMC6 was able to rescue N4am7, while pMC6 28 was not. Increased expression of wildtype N4SSB from expressing clones increased N4 DNA recombination 10 4 -10 5 -fold. This observation indicates that N4SSB plays an essential role in N4 DNA recombination. Two other lines of evidence indicate that N4SSB is essential for N4 recombination independently of its requirement for N4 DNA replication. First, N4SSB ⌬263-265 supports DNA replication, while it is defective in activating N4 recombination (Table II). Second, two additional N4SSB mutants (N4SSB ⌬264 -265,K259R and N4SSB ⌬245-259 ), although deficient in supporting replication, can activate recombination to wild-type or nearly wildtype levels. N4 DNA recombination is independent of host recombination genes, suggesting that N4 encodes its own recombination functions. 5 T4 gp32 is also required for T4 DNA recombination. 6 In contrast, E. coli SSB activates host DNA recombination; its absence in ssb mutant strains reduces recombination only 7-fold (21).
N4SSB and E. coli SSB differ from other well characterized single-stranded DNA-binding proteins in that, in addition to    Reaction mixtures containing an excess of labeled 12-mer oligonucleotide and increasing concentrations of wild-type (wt) N4SSB or N4SSB ⌬262-265 were treated and applied to a 8% native polyacrylamide gel as described under "Materials and Methods." Labeled free probe is not shown. their involvement in replication and recombination, they are transcriptional activators (6,22). N4SSB and E. coli SSB are unique among transcriptional activators in that they do not bind to a specific double-stranded DNA site for activation to occur. How do they accomplish their transcriptional activation tasks in the absence of a cognate DNA-binding site? We have shown that E. coli SSB is an activator of the N4 virion-encapsulated DNA-dependent RNA polymerase by providing the correct DNA structure, a DNA hairpin on the template strand, for N4 virion RNA polymerase recognition (22,23). 7 DNA-binding sites for transcriptional activators serve at least two functions: 1) to increase the local concentration of the activator at its target and/or 2) to position the activator so as to make proper contacts with the transcriptional machinery. A DNA-binding site for N4SSB as a transcriptional activator might not be required since it is expressed at high levels during infection: ϳ11,000 molecules of N4SSB/cell (9 -18 M) (5). Indeed, we have recently proposed that an N4SSB-E. coli RNA polymerase complex forms, which then binds to N4 late promoters (24). Preliminary results indicate that the single-stranded DNA binding activity of N4SSB is not required for transcriptional activation. 8 The isolation of N4SSB mutants differentially affected in DNA replication, recombination, and activation of late transcription (Table II) suggests that different determinants of N4SSB are important for these different activities. The isolation and characterization of additional N4SSB mutants specifically affected in DNA replication, recombination, activation of late transcription, single-stranded DNA binding, and cooperativity are required to understand the role of N4SSB in these processes.