Isolation and Characterization of a Putative Bacteriophage T5 Transcription Replication Enzyme Complex from Infected Escherichia coli*

A well defined enzyme complex of approximately 5 X 10‘ daltons that contains phage and host cell com- ponents known to be required for the processes of phage transcription and DNA replication has been isolated from bacteriophage T5-infected Escherichia coli cells. In addition to the RNA polymerase of the host cell, the complex contains the phage-encoded: gpC2 which has been implicated genetically as a controlling element of late transcription; gpD9, the DNA polymer- ase required for T5 DNA replication; the proteins gpD5 (DNA-binding protein), and gpD15 (nuclease) which are both known to be essential for T5 DNA replication and for the initiation of late transcription. The viral gpD5 derived from the purified complex is a phosphoprotein. The enzyme complex also contains, protected from the action of nuclease, double-stranded DNA with an approximate molecular weight of 1 to 2 X 10’ (2 to 3% of the size of the T5 genome) which is derived preferen-tially from the center of the T5 DNA molecule. The composition of the enzyme complex suggests that the processes of transcription and replication are inte- grated in T5-infected cells. All phage T5 appears to achieved through the host Escherichia coli RNA polymerase (EC 2.7.7.6) (1); however, several T5 gene products, gpDI5 nuclease (2), gpD5 DNA-binding protein (3, and gpC2 gene products (5), the switch from early to late viral transcription. control exerted components

large quantities of the phage-encoded DNA polymerase and some T5 DNA. These observations lead us to suggest that all these gene products interact directly, or through templatemediated processes, or both, to result in an overall integration of the viral replicative and transcriptional programs.

The Purification of the Multienzyme Complex
Step I : Preparation of the Cell Extract-All operations were performed at 4°C. A frozen 4-g cell paste derived from T5-infected cells labeled with both sodium ["%]sulfate and sodium [3ZP]phosphate was used in the experiments to be depicted here unless otherwise noted. Since T5 infection leads to the rapid inhibition of host macromolecular synthesis, only viral macromolecules were labeled under these conditions. The cell paste was thawed and resuspended in 8 ml of buffer containing 0.01 M Tris-HCl, pH 7.9, 25% (w/v) sucrose (ribonuclease free) and 0.1 M NaC1, homogenized in a Dounce homogenizer and incubated for 15 min. Lysozyme, 2 ml of a 4 mg/ml solution in 0.3 M Tris. HC1, pH 7.9, and 0.1 M EDTA, was then added to the cell homogenate, followed 4 min later by the addition of 0.8 ml of PMSF2 (10 mg/ml) to retard proteolysis. The cells were then lysed by the addition of 10 ml of a solution containing 1 M NaCl, 0.02 M EDTA, and 0.08% (w/v) sodium deoxycholate with incubation for 10 min at 10°C. The mixture was then frozen using a dry ice-acetone bath, thawed, and diluted with 188 ml of TEGD buffer that contained 10 mM MgC12 and 5 pg/ml of pancreatic DNase. After a 30-min incubation period, ribosomes and other cellular debris were removed by centrifugation for 2 h at 160,000 X g in a Beckman 50 Ti rotor.
Step 2: DEAE-cellulose Chromatography-The high speed supernatant obtained in Step I above was diluted 2-fold with TEGD buffer that contained 50 mM NaCl and 1 mM PMSF and applied to a DEAE-cellulose column (1.5 X 30 cm) a t a flow rate of 80 ml/h that had been pre-equilibrated in TEGD buffer containing 50 m~ NaCl. The column was eluted with an 800-ml linear 50 mM to 0.5 M NaCl gradient in the same buffer. Column fractions of 12 ml were collected and 0.02-ml samples were assayed for both E . coli RNA polymerase and T 5 DNA polymerase activities. Assays of the T5 DNA polymerase consistently revealed two peaks of activity eluting from the DEAE-cellulose column (Fig. 1). Two peaks of T 5 DNA pol-ymerase which resolve by DEAE-cellulose chromatography have been noted before (26). Further purification of the second peak of DNA polymerase (Fractions 38 to 43) yields a single T5 phage-encoded polypeptide of 96,OOO daltons, monomeric in nature which possesses activities typical of other DNA polymerases (27). The first, or leading peak elutes at a position nearly coincident with a peak of E . coli RNA polymerase activity. Normally, the host RNA polymerase from uninfected cells elutes in Fractions 20 to 30 from DEAEcellulose. The fractions containing both activities (Fractions 25 to 37) were pooled and concentrated to a volume of 3 ml by ultrafiltration with either an Amicon PMlO or PM75 membrane filter. The bimodal elution pattern of the T5 DNA polymerase activity plus the alteration in the elution position of the majoritv of the E . coli RNA polymerase after infection to a position nearly coincident with that of a portion of T 5 DNA polymerase suggested a possible interaction between these two enzymes.
Step 3: Gel Filtration Chromatography-The pooled and concentrated DEAE-cellulose fractions (25 to 37) were clarified by centrifugation a t 10, OOO X g for 20 min and applied to a Bio-Gel A-1.5m column (1.5 X 30 cm) pre-equiJibrated with TEGD buffer containing 0.2 M NaCl. The column was developed with the same solution at a flow rate of 5 ml/h and 1-ml fractions were collected. Samples of the proteins that elute in each fraction were separated by electrophoresis on sodium dodecyl sulfate-polyacrylamide slab gels (Fig. 2). Although the pattern is complex, a discrete class of proteins elutes in the void volume of the column (Fractions 18 to 22), ahead of the bulk protein that begins to emerge in Fraction 29. Additional samples were either precipitated with trichloroacetic acid to monitor for the presence of labeled phage-specific protein and nucleic acid (Fig. 3 A ) or assayed for enzymatic activities (Fig. 3B). The data show that phage-specific proteins are present in both the void volume and in the fractions of the column containing the bulk of the total protein. The material in the void volume also contains some '"P-labeled nucleic acid even though the cell extract had been treated extensively with DNase prior to the beginning of the purification procedure. The enzymatic assay of column fractions for host RNA pol-ymerase and T 5 DNA polymerase reveals that the majority of both activities are found in the void volume. (Fig. 3R). The protein material in the void volume behaves 72-

-
12 -1 0 -4 with an aggregate molecular weight of at least 1.5 X 10'' (the exclusion limit of the column), even though none of the individual polypeptide components approach that molecular weight (Fig. 2). Therefore, this material would appear to represent a discrete complex of interacting protein and nucleic acid components.
Step 4: Hydroxyapatite Chromatography-Material in the excluded volume of the Bio-Gel A-1.5m column (Fractions 18 to 22) was pooled and applied to an hydroxyapatite column (0.9 X 15 cm) that had been pre-equilibrated with 3 volumes of Buffer P (0.07 M potassium phosphate, pH 7.5, 0.1 mM dithiothreitol, and 10% (v/v) glycerol). Following sample application, the column was washed with 2 volumes of Buffer P and eluted with a 200-ml linear gradient of 0.07 M to 0.7 M potassium phosphate, pH 7.5. Almost immediately upon beginning the linear gradient ( Fig. 4.4, Fraction 30). a single peak of material containing both phage-specific "%-labeled protein and '"P-labeled nucleic acid emerges. Assays in individual column fractions for both E. coli RNA polymerase and T5-specific DNA polymerase reveal a single peak containing both enzymes ( Fig. 4 B ) , exactly coincident with the elution profile of the phage-specific proteins and nucleic acid. We have also shown that this peak of protein is free of any polynucleotide phosphorylase activity (data not shown), a contaminating enzyme effectively removed in this step of the purification.
We then determined the protein composition of the material eluting from the hydroxyapatite column by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of individual column fractions (Fig. 5). The data of Figs. 4 and 5 show that both the E. coli RNA polymerase and the T5 DNA polymerase are part of a relatively homogeneous nucleoprotein complex that contains approximately 11 proteins (some of which are ""S-labeled and, hence, phage), and, in addition, some '"P-labeled phage nucleic acid. Based on either the enzymatic assays of the T5 DNA polymerase and the RNA polymerase, or in the case of the latter, also by quantitating the stain intensity of the p,p' subunits, roughly 25% of the cellular total of each of these enzymes is found in the peak of the material which elutes from hydroxypatite.
The column fractions containing the nucleoprotein complex were pooled and dialyzed against two l-liter changes of storage buffer (50 mM Tris.HC1, pH 7.9, 0.1 mM EDTA, 0.1 M dithiothreitol, 0.2 M NaCl, and 50% (v/v) glycerol). The complex was then stored in 0.15-ml aliquots at -70°C and was stable for at least 6 months.
In order to demonstrate that the material eluting from hydroxyapatite still represents a true "complex" rather than the fortuitous co-purification of components, we subjected a portion of the pooled material contained in the hydroxyapatite peak fractions (32 to 34) to chromatography on Bio-Gel A-5m (Fig. 6). Again both RNA polymerase and T 5 DNA polymerase enz-me activities, as well as the viral-specific "'S-labeled protein and :"P-labeled nucleic acid all co-elute. The penetration of the material into the gel matrix of the column also permits a rough evaluation of the overall molecular weight which we estimate to be 4 to 5 X 10'.

The Host Cell and Phage Polypeptide Constituents of the Enzyme Complex
The addition of [:''S]sulfate from 6 to 18 min after infection will label only phage-encoded proteins. A preparation of the multienzyme complex was "'S-labeled, purified as described above, and the protein components separated by polyacrylamide gel electrophoresis. A simple comparison of the radiolabeled protein pattern ( Fig. 7B) to the total proteins visualized by Coomassie blue staining ( Fig. 7A) allows one to determine which proteins are of phage or host origin. By this criterion, the complex contains six T5-specific pol-ypeptides ranging in molecular weight from 10,000 to 96,000. We have indicated in Fig. 7 the proposed identities of some of the gene products. The evidence leading to these designations is more definitive for some of these assignments than for others and will be summarized in the following section. The proposed T5 gene products are: gpD9, DNA polymerase (28); gpC2, an early gene product required for late transcription (5); gpD15, a 5' 3' exo-endonuclease required for both late transcription and normal replication (2, 29); and gpD5, a double-stranded DNA-binding protein required for both DNA replication and late transcription (3, 4). Two major viral proteins (72,000 and 10,000 daltons) and one host protein (12,000 daltons) are from unidentified genes of unknown function. The rerhaining polypeptides which are contributed by the host are the p$',(~, and n subunits of RNA polymerase.
We have quantitated the relative amount of each protein present by a densitometric evaluation of the Coomassie blue stained gels containing different concentrations of proteins.
These results are presented in Table I. Two interesting features of the complex are the submolar amount of the host n factor and the high molar ratio of the T5-encoded gpD5. On a molar basis, the gpD5 DNA-binding protein is the major protein constituent of the complex and is present in amounts which appear related to the level of T5 nucleic acid that is part of the complex. We have also found that when infected cells were exposed to sodium ["P]phosphate, during infection, that a subsequent gel analysis of the proteins of the complex showed that gpD5 (DNA-binding protein) is phosphorylated (Fig. 7 0 . The source of the phosphate and the nature of the chemical linkage to the protein are under investigation.

The Identity of the Individual Host and Viral Components Contained within the Complex
The Host RNA Polymerase-The presence within the complex of host-derived polypeptides of sizes expected for the subunits of the E. coli polymerase together with co-purification of a rifampicin-sensitive RNA polymerase enzymatic activity make the polypeptides assigned as B,p',o, and a in Fig.  7 highly probable. These proteins of the complex also comigrate together with subunits of purified RNA polymerase during electrophoresis.'' The Polypeptide Assignment for gpD9 (DNA Polymerase) of T5-Following T 5 infection a viral DNA polymerase activity is induced, which after DEAE-cellulose chromatography, resolves into two distinct peaks of activity well separated from the DNA polymerase activity of the host cell (Fig. 1 ) . These activities are absent when extracts from uninfected cells were chromatographed under identical conditions.:' The first of these two viral-induced peaks of DNA polymerase co-purifies with the complex throughout the remaining steps of the purification. The second of the two peaks of viral-induced '' T. A. Ficht and R. W. Moyer. unpublished results DNA polymerase is likely to correspond to the nonassociated enzyme which has been purified by Fujimura and his collaborators (26). Furthermore, the DNA polymerase of the purified complex is active in 0.2 M NaCl and also shows other known requirements expected for T5 DNA polymerase (14, 26) as opposed to those of the host DNA polymerase (15).
When cell extracts prepared from both wild type T5 and a T5amD9 mutant are chromatographed on DEAE-cellulose and compared, no DNA polymerase activity can be detected from the T5amD9 extract in either the region of free T5 DNA polymerase or of the multienzyme complex (Fig. 8). However, some RNA polymerase activity from the T5amD9 extract is found in the region expected for the complex, but subsequent chromatography of this material on Bio-Gel A-1.5m yields no material that elutes in the void volume, suggesting that the T5 DNA polymerase, gpD9, is needed for the overall integrity or stability of the multienzyme complex. Finally, we make the polypeptide assignment of 96,000 to gpD9 as depicted in Fig.  7 because we have shown that all the viral proteins of the complex are "early" proteins of T5,3 of which there are only two with molecular weights greater than 90,000 gpD9 and gpC2 (12, 26).
Since the gpC2 molecular weight has been determined to be slightly less than that of the host a subunit (12) and the known molecular weight of gpD9, 96,000 agrees with that of the largest viral polypeptide in the complex we have assigned gpD9 to the viral polypeptide which is larger than the host (I factor. The Polypeptide Assignment for ViralgpD5 (DNA-binding Protein)-The purified T5-encoded gpD5 DNA-binding protein (28,500 daltons) exhibits preferential and cooperative binding to duplex DNA and has been characterized extensively in our laboratory (4). The polypeptide in the complex, which we have identified as gpD5 (Fig. 7), co-migrates on sodium dodecyl sulfate-polyacrylamide gels with samples of the purified gpD5 protein (Fig. 9). Further evidence consistent with the presence of the gpD5 as an integral part of the complex comes from immunoprecipitation experiments with antisera directed against E. coli RNA polymerase. We have found that the multienzyme complex can be immunoprecipitated by this antisera directly from crude cell extracts of wild type T5-infected cells if conditions of high salt are avoided. Analysis of the immunoprecipitated labeled proteins by electrophoresis on sodium dodecyl sulfate-polyacrylamide gels (Fig. 1OA) yields a pattern very similar to that of purified complex (Fig. 7B). When immunoprecipitation is attempted using extracts of T5amD5 infected cells, the amount of viral proteins precipitated per equivalent number of cells is reduced by 90%. When the residual material which does precipitate from T5amD54nfected cells is analyzed, a pattern of precipitated viral proteins similar to that from wild type T5-infected cells is observed, except that the polypeptide-labeled gpD5 is missing (Fig. 10B). Mutants of T5 in gene D5 overproduce the remaining "early" proteins which are normally found as constituents of the complex, so that these proteins are not limiting. Therefore, the overall reduced level of complex immunoprecipitated from T5amD5-infected cells presumably reflects the inhibition of T5 DNA synthesis that is characteristic of these mutants. Based on this evidence, we propose the assignment of gpD5 to the major protein in the complex. The identification of this protein is of particular interest because this is also the only polypeptide of the complex which is phosphorylated (Fig. 7C).
The Polypeptide Assignment for the Viral gpDl5 (5' -+ 3' Exo-Endonuclease)-The gpD15 has been purified and extensively studied in our laboratory (17). It has a molecular weight of approximately 35,000, exhibits maximal enzyme activity at pH 9.3 and is physiologically required for both normal DNA replication and late transcription (2, 29). The complex contains a viral polypeptide of this molecular weight and also exhibits a potent exonuclease maximally active at pH 9.3 which is not inhibited by deoxyribonucleotide triphosp h a t e~.~ These are known properties of purified gpD15 and are distinct from those of the exonuclease present as an intrinsic part of the viral gpD9 (DNA polymerase). The latter enzyme is maximally active at pH 8.6, and is markedly inhibited by the presence of deoxyribonucleotide triphosphates (27).
A role for gpD15 in the formation of the complex is strongly suggested by our inability to isolate the complex from T5amD15-infected cells. Cells infected with T5amD15 mutants initiate T5 DNA replication and for 10 min show a normal pattern of DNA synthesis (29). They also show an extended synthesis of T5 early proteins (2) and, therefore, T5 DNA and all host and viral protein constituents of the complex except gpD15 should be present. Yet complex cannot be isolated from T5amD15-infected cells (Fig. 11). Cells were infected with either T5 or T5amD15 in the presence of sodium ["'Plphosphate in order to label the viral nucleic acid. The extracts were prepared and chromatographed on DEAE-cellulose. The fractions from DEAE-cellulose that normally contain complex (see Fig. 1) were then chromatographed on Bio-Gel A-1.5m (Step 2 of our purification scheme), where the complex should elute in the void volume. Assays for the presence of viral nucleic acid, T5 DNA polymerase or E. coli RNA polymerase activities (Fig. 11) from the Bio-Gel column show that no complex is present in T5amD15-infected cells.
These results suggest that a functional gpD15 product is required for either the formation and structural integrity of the complex or to create a modified DNA template which serves as a site of nucleation for the other components. We have, therefore, assigned the viral protein in the complex of 36,000 daltons to gpD15. The Polypeptide Assignment for the Viral gpC2 (A Transcriptional Factor for Late Genes)-There is no direct assay for C2 polypeptide and it has never been independently purified, so the identification of this protein in the complex is tentative. It has been previously shown by Chinnadurai and McCorquodale (5) that there are only two "early" viral polypeptides with molecular weights of approximately 9 0 , O O O and that one of these polypeptides is gpD9 and the other gpC2. Since we know that the viral proteins contained within the complex are all early proteins and that one of the polypeptides of this approximate molecular weight is likely to be gpD9 (DNA polymerase), we have, assigned the other to gpC2. In addition, cells infected with T5amC2 mutants yield no complex by direct purification.3 When extracts of T5amC2-infected cells are immunoprecipitated with anti-RNA polymerase antisera, only RNA polymerase and a viral polypeptide of 10,000 daltons precipitate (12), suggesting that a functional gpC2 is required for complex formation and integrity. The Viral Nucleic Acid Contained within the Complex Consists of Nonrandom DNA Sequences-When the enzyme complex was purified from infected cells exposed to sodium [32P]phosphate, a portion of the added label was found in nucleic acid even after extensive nuclease treatment. After extraction of the purified complex with phenol to remove the protein, we examined the physical properties of the "P-labeled nucleic acid. The material behaves as double-stranded DNA since it is resistant ( S O % ) to the action of the single-stranded nuclease S1. It sediments in a fairly homogenous fashion from 10 to 12 S in neutrzl sucrose gradients which corresponds to a molecular weight of 1 to 2 X lo6 (2 to 3% of the intact T5 genome). Treatment with the S1 nuclease (30) rapidly solubilizes about 10% of this material without any appreciable Relative enrichment calculated as described in the text.
change in the sedimentation profile, which suggests that there are single-stranded regions located near or at the termini of the DNA. If the DNA of the complex is heat-denatured, purified on sucrose gradients and then treated with pancreatic ribonuclease or alkali about 2% of the material is solubilized, while the remaining material behaves as single-stranded (S1 nuclease susceptible) chains of 10 to 11 S, a sedimentation value similar to that observed for the untreated heat-denatured material alone. These data suggest that perhaps some RNA is present in the terminal regions of the DNA derived from the complex. In summary, the nucleic acid core of the complex consists of double-stranded DNA of approximate molecular weight 1 to 2 X 10" with single-stranded regions consisting, in part, of RNA a t or near the termini of the molecules.
We next determined from where within the T5 genome the DNA of the complex is derived. The purified '"P-labeled DNA of the complex was isolated, denatured, and then hybridized to T5 DNA restriction fragments which were generated by either Sma I or Sal I enzyme digestion, separated by agarose gel electrophoresis and then immobilized on nitrocellulose strips by the technique of Southern (18). The per cent of the total counts hybridized per immobilized restriction fragment from the sample of complex DNA is then divided by the per cent of the counts hybridized to the same fragment for the "control" DNA, which is the total "P-labeled T5 DNA isolated from the same batch of infected cells. If the complex DNA were "random" in sequence, then the quotient of these two numbers should yield the same constant for each fragment; however, the higher the quotient derived from a n individual fragment compared to the remaining fragments, the more the complex DNA is enriched for these sequences. By this criterion, the regions of T5 DNA contained in Fragments C and D of the Sma I digest and the sequences contained within the Sal I A and B fragments are "overrepresented" in the DNA isolated from the purified complex (Table 11). If one examines this data and the Sma I and Sal I T 5 DNA restriction maps (Fig. 12), these results are best explained if the DNA derived from the complex contains a specific, nonrandom enrichment of T5 DNA sequences from near the center of the genome. This region of the T5 genome contains the "strongest" promoters for E. coli RNA polymerase (31, 32) and the preferred origin of T5 DNA replication (33).4

DISCUSSION
We report here the purification and characterization of a well defined 4 to 5 x 10" dalton nucleoprotein complex from phage T5-infected cells which consists of a viral DNA core and a limited number of host and viral proteins. Each polypeptide of the complex that we have tentatively identified has a known role in either phage transcription, replication, or, in some cases, both processes.
We believe that the nucleic acid is an essential part of the ' N. Hamlett and M. Rhoades, personal communication.
complex and that the interactions between the DNA and proteins are specific rather than the result of a random association. The DNA of the complex is normally resistant to the action of DNase. However, exposure of the complex to high salt, a condition which disrupts protein-DNA interactions, leads to an irreversible dissociation of the complex into its individual components and renders all the DNA susceptible to DNase treatment. Hybridization of the DNA of the complex to restriction fragments of T5 DNA further reveals a considerable enrichment for sequences from the center of the T5 genome. An enrichment of certain T5 DNA sequences within the complex not only argues for the specific association of components but also for essential protein-DNA interactions in order to maintain the overall structural integrity of the complex. If some form of T5 DNA is necessary as a structural requirement for the complex, then conditions which limit T 5 DNA synthesis would be expected to eliminate the formation of the complex. We have shown in this paper that this is the case since infections with amber mutants in both gpD9 (DNA polymerase) and gpD5 (DNA-binding protein) both of which are DO mutants and are required to initiate T5 DNA synthesis yield little or no complex either by immunoprecipitation (<5%, Fig. 10) in the case of T5amD5 phage or by direct isolation (Fig. 8) for T5amD9 phage.
Although protein-DNA interactions are important, the mere presence of T5 DNA is not sufficient to allow the stable assembly of a complete or partial complex. T5amD15 mutants which are defective in the 5' + 3' exo-endonuclease, synthesize considerable amounts of DNA yet no stable complex complete except for the gpD15 can be isolated (Fig. 11). This observation suggests two possible roles for gpD15; one role as an essential protein in order to maintain the structural integrity of the complex or a role as an enzyme to modify the T5 DNA to a form which could serve as a nucleation center for complex assembly.
The center region of the T5 genome found within the complex contains many features of interest (Fig. 12). The origin of T5 DNA replication has been shown, both by electron microscopy (33) and by biochemical means: to lie 50 to 60% from the left-hand end of the molecule. This region of phage T5 DNA has also been shown to possess promotor regions that exhibit an unusually high but unequal affinity for RNA polymerase (31,32). It is the region that is closest to the origin of DNA replication that contains the promotors with the highest affinity. We believe that the unusually strong interaction between RNA polymerase and these promotors may be one of the major stabilizing factors which convey sufficient structural stability upon the overall T5 complex to permit its purification. In this regard, we have attempted to isolate a similar complex from T4 infected cells by the procedure developed here and were unsuccessful. 5 A biological role(s) for the complex is suggested by the various protein constituents. Two of the T5 proteins believed to be components of the complex, gpD15 and gpD5, have been shown to be required for both replication and the initiation of late transcription (2)(3)(4)29). Another, gpC2, is thought to function to initiate late T5 transcription (5), whereas gpD9 (DNA polymerase) is thought to function mainly in replication. The presence of all of these components, as well as RNA polymerase, within a single complex argues persuasively for an integrated program of T5 DNA and RNA synthesis. The main host protein contribution to the complex is RNA polymerase, although this enzyme is deficient in the u subunit (Table  I). This enzyme is known to be responsible for the active transcription of all three classes of T5 RNA yet no modified T. A. Ficht, G. Mosig, and R. W. Moyer, unpublished results.
RNA polymerase that contains only phage proteins known to control transcription has yet been isolated despite considerable effort (6). Perhaps, then, late T5 transcription is regulated in a more intricate fashion by the components of the T5 complex such as we have isolated. In this regard, the subunits gpD15 (endo-exonuclease) and gpD5 (DNA-binding protein) are both known to interact with DNA. These polypeptides could act as a substitute for the ( I factor and allow for late transcription by either providing a nicked and gapped DNA template directly as it is synthesized by D9 as part of one single transcription. replication complex. Alternatively, the proteins gpC2 and gpD9 could bind to RNA polymerase and act as a rather complex viral "(I factor" in lieu of the displaced host u factor. Either mechanism, or both, could confer an enhanced recognition of the relatively weak late promotors. Similarly DNA synthesis could be regulated by the transcriptional activity near the center of the T5 genome derived from the high strength promotors near or at the origin of replication. Transcription could serve to activate this region to serve as the origin of replication, similar to the mechanism proposed for phage h (37). RNA polymerase could recognize these promotors and then initiate replication by providing the essential primers. This proposal is consistent with our observation that the host dnaG gene product is not required for T5 growth.s The gpD5, a double-stranded DNA-binding protein, is the most abundant protein of the complex. One of our most potentially intriguing findings is that only the portion of the gpD5 found within the complex is phosphorylated (Fig. 7 0 . The gpD5 protein is made in excess of 500,000 copies/cell (4) and when purified separately only a small percentage (5%), roughly equivalent to that associated with the complex, is found to be phosphorylated.'j Genetic evidence suggests that the gpD5 protein is multifunctional (38) and appears in many ways to be similar to the gene 45 product of T4 (39). Phosphorylation of a portion of the total protein could be one means of providing functional compartmentalization for the protein in either transcription or replication. A stable phosphorylated gpD5 may act as the "nucleus" about which the protein and DNA of the complex assemble.
Detailed enzymatic studies of the RNA and DNA polymerase activities of the T5 complex will be presented elsewhere. The complex possesses both endogenous RNA and DNA polymerase activities presumably mediated through the DNA of the complex and both activities are markedly stimulated by the addition of exogenous DNA templates. The sequences copied by both enzymes of the complex of either endogenous or exogenously added templates will be of interest and these studies are in progress. One reaction catalyzed by the complex is the efficient use of exogenously added fd DNA as a template for DNA synthesis in response to deoxyribonucleotide triphosphates which have been pretreated to eliminate any contaminating ribonucleotide triphosphates. No degradation of added fd DNA has been observed in this rifampicin-resistant reaction. The synthesis of fd DNA in vitro by the DNA polymerase of the T5 complex raises an interesting paradox, since T5 DNA polymerase, like all other DNA polymerases, requires a primed template for DNA synthesis. An independent priminp event must, then, take place prior to overall chain extension. Priming might be achieved through utilization of either endogenous ribonucleotides or RNA present within the complex. However, if the host RNA polymerase of the complex primes the fd-stimulated reaction in uitro, then the enzyme must do so through a rifampicin-resistant mechanism and be able to utilize deoxyribonucleotide triphosphates di- rectly. Experiments are now in progress to determine the priming mechanism used by the complex in uitro and the nature of the products generated.
In summary, we believe that the purified T5 enzyme complex presents a unique experimental system in which to study the assembly and organization of multienzyme systems as well as some of the partial and coupled reactions associated with DNA transcription and replication and how these processes are regulated and integrated.  infected E. c o l l W3llOth exposed to sodium l32Pl-phasphate a s procedure were performed at 4%. The c e l l pellet was resuspended in 3.6 ml Of lysing SolUtlon containinq 0.10 X Trls.XC1, pH 7.3, 10 M NaCN and 10 mM EDTA and incubated far 15 mlns. An addrtlanal 0 . 4 ml of l y~m g s o l u t m n containing lysozyme I10 mg/ml) was then added, followed 4 mlns later by 220 u l of 20% lv/vl sarkosyl Sol"tlan. Predigested pronase (10 mg/mll ~n pronase buffer 150 mM Tris.HC1, pH 8.0. 50 mM NaCl and 25 mM EDTA and heated for 10 mlns at 8OOc1 was added to a flnal concentration Of 2 mg/rnl. and the mrxture was incubated for 4 hours at 37OC. The rnlxtnre was adlusted to 0.3 N XOH and incubated for an addxtlonal 16 hours at 37oC. After nevtrallratlon with B 501nt10n Of 1 N XC1 prepared l n 50 mN T I~s . H C~, p H 7.9, the m1Xture was extracted tWlCe wlth phenol-chloroform I 1 : l ) .