Purification of the gene 43, 44, 45, and 62 proteins of the bacteriophage T4 DNA replication apparatus.

A procedure has been developed which allows the T4 bacteriophage proteins corresponding to the products of genes 43, 44, 45, and 62 to be purified to near homogeneity from a single T4-infected cell lysate (greater than 90% single species as judged by sodium dodecyl sulfate polyacrylamide elctrophoresis). In these preparations, the major problem of removing all contaminating nucleases has been overcome. Each of the above proteins is known from genetic analysis to be essential for phage DNA replication. The protein product of gene 43 is T4 DNA polymerase, and its recovery can be monitored using a standard DNA polymerase assay. The other three gene products have been designated as "polymerase accessory proteins," since they directly enhance polymerase function on both single- and double-stranded DNA templates. Their activities were monitored by an "in vitro complementation assay," which measures the stimulation of DNA synthesis observed in a concentrated lysate of T4 mutant-infected Escherichia coli cells when the missing T4 wild type protein is added. Starting from 300 g of infected cell paste, we obtained 9.3 mg of gene 43 protein, 21 mg of gene 45 protein, and 70 mg of a tight complex made up of 44 and 62 proteins; final yields were estimated at 30%, 14%, and 28%, respectively, of the initial activity present in the lysate. When the above purified proteins are incubated with preparations of two other T4 DNA replication proteins (gene 41 and gene 32 proteins) plus deoxyribonucleoside and ribonucleoside triphosphates, extensive DNA synthesis occurs on both single- and double-stranded DNA templates. As reported elsewhere, this synthesis mimicks that catalyzed by the T4 DNA replication apparatus in vivo.

A procedure has been developed which allows the T4 bacteriophage proteins corresponding to the products of genes 43,44,45, and 62 to be purified to near-homogeneity from a single T4-infected cell lysate (>90% single species as judged by sodium dodecyl sulfate polyacrylamide electrophoresis).
In these preparations, the major problem of removing all contaminating nucleases has been overcome.
Each of the above proteins is known from genetic analysis to be essential for phage DNA replication. There are at least six known T4 bacteriophage genes whose products are involved with DNA chain elongation and repli-43, 44,45, and 62. When purified preparations of the proteins specified by these six genes are incubated in the presence of deoxyribonucleoside and ribonucleoside triphosphates, extensive DNA synthesis is catalyzed on both single-and doublestranded DNA templates (17)(18)(19)(20).' Detailed characterization of the activities of these proteins and the in vitro DNA synthesis reactions they catalyze will be found elsewhere; (13)(14)(15)(16)(17)(18)(19)(20)(21)(22).' Here we report a purification procedure, developed and refined in this laboratory during the past several years, which allows reproducible preparation in high yield of the proteins made by T4 genes 43, 44, 45, and 62, starting from a single lysate. The procedures and results are outlined in the text, and are presented in detail in the miniprint supplement.'

MATERIALS AND METHODS
Bacteria, Bacteriophage, Enzymes, and Chemicals-The host strain used for all experiments WZIS Escherichia coli DllO (poZAl, endr.
thy-. su-1 obtained from Dr. C. Richardson (23). The T4 bacteriophage used were regA-amEl (45-j, regA-amN82 (44-l, and regA-amN55-amH39 (42-, 30-j, derived from strains constructed by J. Wiberg (24,25). The regA mutation (originally called SP62) results in at least lo-fold overproduction of the T4 DNA replication proteins corresponding to genes 44,62, and 45 (15,261. Our refined procedures for the purification of the T4 proteins corresponding to genes 32 and 41 will be found in adjoining papers" procedures were carried out at 4°C unless otherwise indicated. Note that the two polypeptides coded for by genes 62 and 44 are isolated together as a stable complex containing multiple copies of the two gene products (14). The first purification step is DEAE-cellulose chromatography of a cleared cell lysate, prepared as described under "Materials and Methods." All three protein species are separated by this step and are subsequently purified independently of each other.
DEAE-Cellulose Chromatography-A column of Whatman DE52 DEAE-cellulose (29 X 8.5 cm) with a packed volume of 1.7 liters was prepared as described by the manufacturer and equilibrated with A0 buffer (for buffer definitions, see "Materials and Methods"). All flow rates were 500 ml/h. Fraction I (575 ml prepared as described under "Materials and Methods") was loaded onto the column followed by washing with 2 liters of A0 buffer. A  from 0 to 0.2 M NaCl (A0 to A1 buffer) was run, and the protein elution profile shown in Fig. 1 was obtained. To find the 44/ 62 protein complex, a gene 44 complementation assay was performed on all column fractions (see "Materials and Methods"); as indicated, all of the activity was found in the fractions which did not bind to the column, slightly leading the major peak of protein in the breakthrough (Fig. 1). A DNA polymerase assay was performed to locate the gene 43 protein activity. This polymerase activity was found in fractions eluting during the early part of the gradient (60 to 90 mM NaCl), ahead of the major peak of salt-eluted protein (Fig. 1). To locate the 45 protein, a gene 45 complementation assay was performed; as indicated (Fig. l), this activity eluted late in the salt gradient along with the major protein peak (110 to 150 mM NaCl). In each case, peak fractions estimated to contain 80% of the total recovered activity were pooled, giving rise to the three different Fraction II preparations to be described (see Tables I, II, and III, below). Subsequent to this point, the three protein fractions of interest were purified separately. If necessary, unused fractions could be stored on ice while awaiting further purification, since 45 and 43 protein activities appear to be reasonably stable. However, with practice, all three preparations can be completed in less than a week starting from cell lysis, with two individuals sharing the work involved. If one person h,mdles the cell lysis and purification of 43 protein, while the other person purifies 44/62 and 45 proteins, a reasonable balance of time and effort is achieved and the proteins are recovered with activities comparable to those to be described. (A suitable time schedule is given in the miniprint supplement) .
Purification of 44/62 Protein-A summary of the 44/62 purification results is presented in Table I. A total of 70 mg of highly purified protein complex was obtained from the above DEAE-cellulose eluate (Fraction II) by a two-step procedure (hydroxylapatite chromatography followed by phosphocellulose chromatography).
Because of the overproduction of 44/ 62 protein induced by the regA mutation (15), the purification required from the cleared lysate was only 73-fold. The final protein obtained represented about a 30% yield of the starting 44/62 activity, as judged by the in vitro complementation assay (Table I). Fig. 2 presents typical data from which such quantitative estimates of activity are made; enzyme units are defined as described in the legend. It should be noted that, for reasons not understood, these complementation activity assays show a logarithmic, rather than a linear, response to added enzyme. (This is shown in Fig. 2 for 44/62 protein, and it is true for the other complementation activities as well (15)). Although we find that estimates made in successive assays are remarkably reproducible, this logarithmic response reduces the accuracy with which the recovery of complementation activity can be quantitated.
Note that it also artificially broadens the complementation activity peaks measured from column eluates (e.g. see Fig. 1 and the miniprint supplement), a    Table I.) unfrozen at -20°C in Buffer H. Its specific activity has remained essentially unchanged under these conditions for at least 6 months. Precise details of the purification procedures used for this and the other proteins are presented in the miniprint supplement to this paper. All of these procedures have proven to be highly reproducible, and have been in use for several years in three different laboratories.
Purification of 45 Protein-A summary of the 45 protein purification results is presented in Table II. A total of 21 mg of highly purified protein was obtained, starting from the DEAE-cellulose eluate previously described (Fig. 1). The additional purification steps used were ammonium sulfate precipitation, hydroxylapatite chromatography, hydrophobic chromatography on norleucine-Sepharose, and a second DEAE-cellulose fractionation.
The final preparation, representing 14% of the starting 45 protein complementation activity, was stored unfrozen at -20°C in Buffer 0. Its specific activity has not changed noticeably over the course of a year. Precise details of the purification procedures used are presented in the miniprint supplement to this paper.  Table III. A total of 9.3 mg of highly purified protein was obtained, starting from the DEAE-cellulose eluate in Fig. 1. The additional purification steps used were DNA-cellulose chromatography, hydroxylapatite chromatography, hydrophobic chromatography on norleucine-Sepharose, and a second DEAE-cellulose fractionation.
The final preparation is stored as a liquid in Buffer Do at -20°C at a concentration of 0.5 mg/ml or higher. Under these conditions, it appears to be quite stable.
For this particular preparation, on overall yield of about 30% of the original activity was obtained in the most purified fractions (Table III), as judged by the standard DNA polymerase assay described under "Materials and Methods." When this assay and the gene 43 complementation assay were compared side by side during a similar purification run, compa-   rable yields of polymerase activity were obtained at each step as judged by either assay. 5 We have therefore chosen to use the simpler direct assay for routine monitoring of DNA polymerase purifications, as in Table III. Physical Properties of the Purified Proteins- Table  IV summarizes some of the relevant properties of the 43,45, and 44/62 proteins purified here, as we presently understand them. As previously reported (14), the gene 44 and 62 proteins are isolated as a tight complex (180,000 daltons). Originally, from relative Coomassie blue staining intensities, this complex appeared to contain 4 molecules of 44 protein (34,000 daltons each) and 2 molecules of 62 protein (20,000 daltons each). Presently, with different purification and gel-staining procedures, a higher ratio of 44 protein to 62 protein is estimated; thus, the exact subunit composition of this complex needs to be reinvestigated with better methods. The two different polypeptide chains in the 44/62 complex have thus far been separable only by agents (such as SDS and guanidine) which  (Table III) This protein functions in such small amounts as not to be detectable as a band even on strongly overloaded gels. Nevertheless, its activity is required for the in vitro replication seen in Fig. 5 (Refs. 20, 37).", ' denature proteins. However, we now suspect that monitoring with the gene 62 complementation assay, as done previously (14), rather than with the gene 44 complementation assay used here, may better select against a partial dissociation of the 62 protein which occurs during the purification.
The other replication proteins contain only a single type of subunit, as indicated. However, the gene 45 protein appears to exist as a dimer. Likewise, while the T4 DNA polymerase sediments at low concentrations as a monomer (13), the sigmoidal response to added polymerase in some experiments raises the possibility that higher 43 protein oligomers are the active form in the complete replication system. It is interesting to note that the existence of these replication proteins as oligomers raises the possibility of multisite activity, or cooperativity, or both, which could be important in the replication process.
Catalytic Purification of T4 Gene 43,44,45, and 62 Proteins that all of them be present simultaneously, and there is physical evidence that they form a multienzyme complex (see Refs. 19 and 20). Since analysis of individual functions in such a multicomponent system is difficult, it is instructive to study in detail as many "partial reactions" as possible. Thus, Table  IV lists the well known DNA polymerase and proof-reading  exonuclease activities of 43 protein, both of which strongly  prefer single-stranded DNAs as substrates. Also, included in Table IV  DNA templates, in a reaction requiring ATP (16,21,22). In this situation, these proteins can be shown (16) to increase both the rate at which an individual polymerase molecule moves along the DNA template, and its "sticking distance" (or processiveness). Thus, the 44/62 and 45 proteins may be considered to be polymerase "accessory proteins," which increase the efficiency of the DNA polymerase reaction by an ATP-dependent mechanism whose details are not yet known. In addition to its crucial role as an accessory protein in T4 DNA replication, the gene 45 protein is essential for late T4 RNA transcription (33,34). For this second process, it is presumably relevant that the 45 protein has been observed to bind specifically to the T4-modified, host RNA polymerase (35). The 45 protein thus has two different essential roles during T4 infections, and in each role it interacts with a different polymerase.
Purity-Analysis of the purified proteins by polyacrylamide electrophoresis in the presence of SDS reveals only one very predominant protein band for each gene product, as illustrated in Fig. 3. Moreover, the mobility (and thus the molecular weight) of each of these bands corresponds to the value expected from previous studies, in which a single altered polypeptide was found after infection with T4 amber mutants in each respective gene (5,14,15,36). As determined by tracing the Coomassie blue-stained gel in Fig. 3 on a Joyce-Loebl densitometer, each protein preparation obtained by our methods is at least 90% homogeneous.
Absence of Nuclease-The original T4-infected cell lysate is very rich in nucleases. The purification steps listed in Tables  I, II, and III were designed to remove these activities, which otherwise severely distort the in vitro replication process. It is crucial that each batch of purified proteins be tested for endodeoxyribonuclease activity with both single-stranded and double-stranded DNA as substrates. This is most conveniently done with the electrophoresis assay shown in Fig. 4. With 20 min of incubation at 37°C under the salt conditions used for the in vitro DNA synthesis assay, none of the proteins caused detectable nuclease activity, whether present singly or in the complete replication protein mixture (Fig. 4). The lower limit of visual detection with this assay is estimated to be 10% of the double-stranded or single-stranded template being nicked; thus, with all proteins present, less than one nick should be introduced per 60,000 DNA nucleotides in a 20-min DNA synthesis reaction.
Contaminating exodeoxyribonuclease activities have been less of a problem in these preparations.
With the exception of the T4 DNA polymerase which has an inherent 3'-to 5'exonuclease activity (7-lo), the replication proteins prepared according to the above methods contain no detectable activity in standard exonuclease assays (less than 1% of the DNA rendered acid soluble in a 45-min incubation at 37°C with 20 pg/ml of each protein; for details, see "Materials and Methods" and Footnote 3). Stimulation of the in Vitro T4 DNA Replication Reaction-The protein products corresponding to genes 43,44,45, and 62 are each an essential component of the T4 DNA replication apparatus, as judged both by in viva (l-4) and in uitro studies (17)(18)(19)(20). In Fig. 5 (17,19,20)' have been published.
The reaction is a very efficient one, with many times as much DNA made as was originally added as template. Further detailed manuscripts, describing a variety of partial reactions, and the fidelity and the rate of the complete reaction, are in preparation and will be published elsewhere. In addition, many results comparable to ours have been obtained recently in the Nossal laboratory (37).

DISCUSSION
Using an in vitro complementation assay and a standard DNA polymerase assay, it has been possible to purify the T4 proteins corresponding to genes 43, 44, 45, and 62 to 90% or greater homogeneity, with no detectable exo-or endodeoxyribonuclease contamination. All the proteins are purified from a single T4-infected cell lysate in less than one weeks' time. In this particular preparation, 70 mg of 44/62 protein, 21 mg of 45 protein, and 9 mg of 43 protein were isolated from 300 g of T4-infected cell pellet. The 44/62 and 43 protein quantities obtained here are about usual, but the 45 protein was obtained in less than half of the expected amount. The 45 protein yield can be improved by increasing the time allowed for growth of the T4-infected bacterial culture used as the source of the purified proteins (to at least 40 min).
It is hoped that this paper and the adjoining paper in this journal (27) will provide a convenient purification scheme which will allow others to purify readily these T4 DNA replication proteins, and thus encourage a wide variety of studies on their physical and chemical properties in other laboratories.
Achnowledgments-During the past several years many persons in our laboratory (J. Barry