Identification of Bacteriophage T4 Gene 60 Product and a Role for This Protein in DNA Topoisomerase*

Bacteriophage T4 DNA topoisomerase has been iso- lated and shown to contain the proteins coded by the DNA-delay genes 39 and 52 (Liu, L. F., Liu, C.-C., and Alberts, B. M. (1979) Nature (Lond) 281, 456-461 and Stetler, G. L., King, G. J., and Huang, W. M. (1979) Proc. Natl. Acad. Sei. U. S. A. 76, 3737-3741). From comple- mentation measurements in vitro and from earlier ge-netic evidence, these workers suggested that the prod- uct of gene 60 (p60) was also a component of the DNA topoisomerase complex. This paper now establishes the identity of p60 and unequivocally shows that this protein is a component of the enzyme complex. T4 DNA topoisomerase was purified by a simplified two-column procedure and found to be a stable complex of p39, p52, and a protein with a relative molecular weight of 18,000. The 18,000-dalton chain has been unambiguously shown to be the product of gene 60 through the use of an amber mutant of gene 60 with Sup+ and Sup- hosts and analyses by two-dimensional gel electrophoresis. While p39 and p52 were tightly associated in the wild type enzyme complex, they were readily separated on a hydroxylapatite column from extracts of cells infected by an amber mutant of gene 60. These findings suggest that p60 plays a structural/ functional role in the enzyme complex by holding the larger p39 and p52 in juxtaposition. Liu et and Stetler et

shown to be part of the purified enzyme complex (2, 3), the gene 60 product was not clearly identified nor was it unequivocally shown to be part of the complex.
Concurrently with these reports, this laboratory had found that mutants in the T4 DNA-delay genes cause a decrease in t h e rate of synthesis of deoxyribonucleotides (5). As part of an investigation of this phenomenon and to examine the interaction of DN14 topoisomerase with the membrane (6, 7), we embarked on a study of this enzyme.
In this paper, the protein product of gene 60 has been identified by two-dimensional electrophoresis using an amber mutant of the gene and examining cell extracts after infection of Sup+ and Sup-hosts. The phage T4 DNA topoisomerase complex has been isolated by a simplified procedure and shown to contain 1360 as an integral part of t h e enzyme with p39 and p52. In the absence of p60, the other two component proteins do not remain tightly associated.

EXPERIMENTAL PROCEDURES
Biologicals-The Escherichia coli B strain was described earlier (8). E. coli B40, a B strain carrying supD suppressor, was obtained from Larry Snyder, Michigan State University, East Lansing, MI.
The bacteriophage 1'4 mutants used in this study and their sources are listed in Table I. The amHA9 amN82 double mutant was constructed by a standard phage cross in E. coli B40 (9).
Materials-Hydro'xylapatite, Bio-Gel HTP, was obtained from Bio-Rad; DEAE-cellulose was DE-52 from Whatman; agarose was SeaKem from Marine Colloids Div., FMC Corp., Rockland, ME; ethidium bromide was from Aldrich; bovine pancreas trypsin, soybean trypsin inhibitor, ,&lactoglobulin E, egg-white lysozyme, Brij 58, and PMSF were from Sigma; sperm whale myoglobin was from Schwarz/ Mann; and CHAPS was from Pierce Chemical Co. The other molecular weight marker proteins and all the materials for electrophoresis and autoradiography were obtained as previously described (10).

Bacteriophage T4.lnfection and Radioactive Labeling of T4 Early
Proteins-The growth media and the methods used for labeled and unlabeled infections by phage T4 have been described (10). Infections by amHA9 amN82 wlere carried out for 30 min at 30 "C. To label the T4 proteins, 35S0,2was added 2 min after infection to a separate culture, equivalent in volume to about 0.1 of the unlabeled infected culture, at a final activity of 50-100 pCi/ml. Cells infected by amHA9 amN82 will incorporate approximately 50% of the into protein in 30 min under the conditions described (10).
Get Electrophoresis and Autoradiography-SDS-polyacrylamide gel electrophoresis was performed according to Laemmli (11) with the following modifications. The thickness of the slab gels was 1.5 mm and the heights of thle separating and stacking gels were 9.6 and 1.0 cm, respectively. The proteins were fixed in the gel in a solution of 10% acetic acid and 25% isopropanol, stained for 1-2 h in 0.05% Coomassie brilliant blue R, and dest,ained in 10% acetic acid. Eastman Kodak XAR-5 film was employed for autoradiography of the dried gels. Samples for separation by unidimensional electrophoresis were brought to a final concentration of 0.05 M Tris-HC1 (pH 6.8), 1% SDS, 1% mercaptoethanol, 4.7% glycerol, and 0.01% bromphenol blue, and the mixtures were heated for approximately 3 min in a boiling water bath. The method of sample preparation for two-dimensional electrophoresis and the conditions for nonequilibrium pH gradient electrophoresis in the first dimension have been previously described (10).
The proteins used as molecular weight markers and their assigned 1221 Analytical Methods-DNA topoisomerase was assayed as described (2), except that bovine serum albumin was substituted for human serum albumin and supercoiled pBR322 DNA was used as the substrate rather than supercoiled I'M2 DNA molecules. The protein concentrations were determined by the methods of Kalb and Bernlohr (12) for crude extracts and Sedmak and Grossberg (13) for the purified fractions using bovine serum albumin uncorrected for water as a standard.
Purification of T4 DNA Topoisomerase-Four 1-liter cultures of E. coli B grown to 5 X 10" cells/ml in a rotatory shaker bath were infected by T4 phage amHA9 amN82 at multiplicities of 8 a t 30 "C. and the infections were allowed to continue for 30 min. The resulting cell pellet was combined with that from 350 ml of cells infected in the same manner but labeled with ''"S0.I'' at 2 min after infection. The combined cells were resuspended in 20 ml of a solution of 40 mM Tris-HCI buffer (pH 7.8 at 25 "C), 2 mM EDTA, 24% (w/w) sucrose (n = 1.37051, and 0.63 mg of egg-white lysozyme/ml, and the suspension was incubated a t 0 "C for 1 h. An equal volume of a solution consisting of 1% Brij 58, 40 mM Tris-HCI at pH 7.8, and 20 mM P-mercaptoethanol was added, and the solution was stirred in an ice bath for 1.5-2 h. The lysed extract was centrifuged at 30,000 X g for 1 h a t 4 "C, and the supernatant fluid (33.0 ml) was loaded on a DEAE-cellulose column (0.7 X 13 cm) which had been equilibrated with a solution of 40 mM Tris-HCI buffer at pH 7.8, 10 mM P-mercaptoethanol, 1 mM EDTA, and 10% (v/v) glycerol. The unadsorbed fraction was applied directly to a hydroxylapatite column (0.7 X 5.5 cm) equilibrated with a solution of 10% glycerol, 10 mM P-mercaptoethanol and 20 mM potassium phosphate buffer at pH 7.2 (25 "C). After the column was rinsed with equilibration buffer and washed with a solution containing 10% glycerol, 10 mM P-mercaptoethanol, and 0.3 M potassium phosphate buffer, the enzyme was eluted with a mixture of 10% glycerol, 10 mM P-mercaptoethanol, and 0.5 M potassium phosphate buffer. The fraction containing the peak enzyme activity and radioactivity was brought to 50% in glycerol and stored a t -20 "C. This purification scheme is a modification of the procedure of Liu et al. (2) in which several of their steps are deleted.

RESULTS
Purification of T4 DNA Topoisomerase-All steps were carried out at 4 "C unless otherwise indicated. An outline of the results of the purification is presented in Table 11. The data in this table and in Fig. 2B are presented in terms of total protein while the remaining figures include only radiolabeled T4 early proteins. In two chromatographic steps, a yield of 5 X lo4 units of T4 DNA topoisomerase activity in 200 pg of protein was obtained from the 30,000 X g supernatant of the crude extract prepared from 4 liters of amHA9 amN82infected E. coli culture. If the 30,000 X g pellet is further extracted with a solution containing 2 M NaCl, 50 mM Tris-HCI, pH 7.8, and 0.2% Brij 58 and centrifuged at 100,000 X g for 30 min as described by Stetler et al. (3), an additional 1 X lo4 units of purified enzyme can be realized from the dialyzed supernatant by the same purification scheme starting at the DEAE-cellulose column step. In the purification presented in Table 11, the specific activity of the T4 DNA topoisomerase product was 2.5 X 10" units/mg of protein. The activities in Fractions 1-111 were not measured because of interference by nucleases.
When T4 DNA topoisomerase was purified from amN82or SP62 amN55-infected cells, an additional protein having a subunit molecular weight of 24,000 by SDS-polyacrylamide electrophoresis accounted for approximately 40% of the radiolabeled protein in the purified fraction. It could be separated from the topoisomerase complex by either a carboxymethyl-  Role of Gene 60 Protein in Phage T4 DNA Topoisomerase cellulose or a gel filtration column (data not shown). This very basic protein was identified as the protein product of the ipZZI gene by two-dimensional gel electrophoresis (10) using the purified protein product of ipZZI and amHA9, an amber mutant of ipZZZ, both from Dr. Lindsay Black, University of Maryland, Baltimore. The protein product of ipZZZ corresponds closely in size to a protein found in the purified topoisomerase enzyme described in an earlier study (2). This protein appeared to copurify with the enzyme complex through the two chromatographic steps in this procedure, although it was not established whether an interaction between pip111 and the enzyme complex occurs. A double mutant, amHA9 amN82, was constructed to obviate the removal of pipIII, and Fig. 1 compares the purified T4 topoisomerase fractions isolated after infections by ipZZZ+ dnaand ipZIZdnaphage. Fig. 2A presents an analysis of the "'S-labeled proteins found at each successive step in the purification as monitored by electrophoresis on a 12% polyacrylamide gel and subsequent autoradiography. Lane 6 shows 3 major protein bands corresponding to the products of genes 39, 52, and 60 and representing greater than 92% of the total radioactivity in the purified fraction. In  fraction (A, lane 6). are estimated to be 80% of the total stained protein in the gel. Two host proteins having relative molecular weights of 39,000 and 17,000 are present as minor contaminants in this topoisomerase preparation.
The molecular weights of p39, p52, and p60 have been determined by comparison with protein standards of accepted molecular weights on one-dimensional SDS-polyacrylamide gels. The subunit molecular weights of the p39 and p52 chains were determined to be 56,500 and 48,000, respectively, with a 10% polyacrylamide gel (Fig. 3). However, it should be noted that membrane proteins may give molecular weight estimates from SDS-gel electrphoresis which are not a direct measure of polypeptide molecular weight since hydrophobic proteins may bind greater than normal amounts of SDS (14,15). Using a 12% polyacrylamide gel, the protein product of gene 60 was shown to have a molecular weight of 18,000 (Fig. 4). The 16,000-dalton chain associated with the purified T4 topoisomerase described by Liu et al. (2) most probably was p60 based on the enzymatic activity of their preparation and our identification of p60 by two-dimensional gel electrophoresis (see below).
Studies with the purified T4 topoisomerase have suggested that it is a stable complex. The three subunits of the complex remained associated when sedimented on a linear sucrose gradient (5-20%, w/w). In this experiment, the native topoisomerase complex migrated between catalase (M, = 243,000 (16)) and aldolase (M, = 160,000) (data not shown). The three proteins also remained associated when extracted from the membrane with a solution of 2 M NaCl and 0.2% Brij 58. The complex did not dissociate in the pH range of 5.1 to 8.0 or in the presence of 5 mM CHAPS, a highly disaggregating detergent used for protein solubilization (17,18). These preliminary studies suggest that the three proteins are held together both by hydrophobic and ionic forces.
Identification of Components of T4 DNA Topoisomerase Using Two-dimensional Gel Electrophoresis-Two-dimen-sional separation by nonequilibrium pH gradient electropho-amHl7, respectively, and their locations are indicated on the resis followed by SDS-polyacrylamide electrophoresis has gel pattern (amN55 infection) shown in Fig. 5. Two-dimenbeen used to identify 17 bacteriophage T4 prereplicative pro-sional gel electrophoresis of the crude extract from E. coli B teins (10)

Role of Gene 60 Protein in
Phage T4 DNA Topoisomerase demonstrated that an 18,000-dalton protein chain was absent by comparison to the gel pattern obtained on infection with amN55 (Fig. 5A). When amE594 was used to infect the E. coli B amber suppressor strain, B40, a protein chain reappeared which had the same charge but migrated slightly slower than the wild type gene 60 product (Fig. 5B). Because of the nature of the suppressor mechanism, this slower mobility is undoubtedly the result of the amino acid substitution by the suppressor and not an alteration in chain size. An amino acid substitution can alter the hydrophobicity of a protein and thereby change the migration of the suppressed gene product (19,20). The identification of the gene 60 protein product was confirmed by two-dimensional gel analysis of crude extracts from two other gene 60 amber mutants, amEd00 and amE429. Both lacked the protein spot identified as p60, but contained wild type amounts of p39 and p52 (data not shown). Fig. 6 shows a two-dimensional separation of purified T4 DNA topoisomerase. When the autoradiograms from the purified enzyme and crude extracts are aligned, the three "Slabeled protein chains correspond exactly to the spots identi- fied as p39, p52, and p60. The two-dimensional gel patterns establish that gene 60 codes for a protein chain of 18,000 daltons and that this protein is present in the purified T4 DNA topiosomerase with p39 and p52.
p60 Plays a Structural/Functional Role in the T4 DNA Topoisomerase Complex-As mentioned previously, the T4 DNA topoisomerase complex is rather stable and does not readily dissociate into its subunits. However, when extracts of E. coli B cells infected by an amber mutant of gene 60 were carried through the purification scheme outlined in Table 11, p39 and p52 no longer eluted together from the hydroxylapatite column with 0.5 M potassium phosphate buffer. The autoradiogram in Fig. 7 shows the "'S-labeled proteins eluted from the hydroxylapatite column at varying potassium phosphate concentrations after infection by the gene 60 amber mutant, amE429 or amE594. In each case, p52 was eluted with 0.3 M potassium phosphate buffer, while p39 bound more tightly to the column and was eluted only after increasing the potassium phosphate concentration to 0.5 M. The identifications of p39 and p52 were confirmed by analysis of the fractions eluted by 0.3 and 0.5 M potassium phosphate using twodimensional gel electrophoresis. The "'S-labeled protein having a molecular weight of 44,500 and eluting with p39 in the 0.5 M potassium phosphate buffer has been shown to be a proteolytic fragment of p39 by the method of Cleveland et al. (21) (data not shown).
These experiments give clear evidence that p60 has a structural role in the T4 DNA topoisomerase complex, i.e. that this subunit is directly involved in the association of p39 and p52.

DISCUSSION
This work has characterized the protein product of gene 60 and established that p60 is part of T4 DNA topoisomerase and has a structural/functional role in the enzyme complex. The primary findings are as follows: 1) p60 has been identified by two-dimensional gel electrophoresis and shown to be an 18,000-dalton protein chain; 2) T4 DNA topoisomerase has been purified by a simplified procedure. The purified enzyme is a stable complex and was shown to contain p39, p52, and p60 by both one-and two-dimensional gel electrophoresis; 3) in the absence of p60, the other two proteins, p39 and p52, did not remain associated but were readily separated on a hydroxylapatite column.
The finding that p60 is required for the tight association of p39 and p52 suggests that this small protein chain serves as a structural link, perhaps providing flexibility to the enzyme while holding the larger chains in juxtaposition. Preliminary studies show that the enzyme complex sediments in sucrose gradients between catalase and aldolase. The coaxial ratio of the molecule will need to be considered (22), but on the basis of the precedence of an a& structure for the host enzyme (23)(24)(25), the observations are consistent with p60 (7) or its dimer acting as a linker between p39 (a) and p52 (p) in the T4 enzyme with a structure such as a2-y2-p2.
It is appropriate to review the differences between the phage-coded enzyme and its comparable type I1 DNA topoisomerase in the host. Liu et al. (2) and Stetler et al. (3) have shown that the phage enzyme relaxes positive or negative supercoils, whereas the host enzyme has the unique ability to introduce negative supercoils (26). The host enzyme dissociates readily into its a2 and p2 subunits (27), whereas the isolation procedures of Stetler et al. and of Liu et al. and our own attempts at dissociation demonstrate that the phage T4 enzyme is a tight complex. Whether the third protein chain in the T4 enzyme is related to these differences remains to be determined.
The studies presented have led to procedures to separate Phage T4 DNA Topoisomerase the three protein chains of T4 DNA topoisomerase in pure form so that each can be studied in terms of its interaction with synthetic phospholipid vesicles, binding to DNA, and activity as an ATPase. T4 DNA topoisomerase may have other functions in addition to its proposed role in replication or in initiation of replication (2, 3, 28-31). Not only has this enzyme been implicated in deoxyribonucleotide synthesis in vivo (5), but it has also been found in quantity in preparations of T4-induced deoxyribonucleotide synthetase complex (32). Recent experiments have shown that T4 topoisomerase may have a role in the synthesis of the , & subunit of T4 ribonucleoside diphosphate reductase (33), perhaps by mechanisms similar to those proposed for DNA topoisomerase I of E . coli in the transcription of several genes (34).