Oct 24, 1990 - The catalytic activity of HeLa DNA polymerase e, an enzyme consisting ... properties of the HeLa enzyme, its trypsin-digested form, and the calf ...
THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc
Vol. 266, No. 10, Issue of April 5,pp. 6336-6341,1991 Printed in CJ S.A.
Identification and Tryptic Cleavage of the Catalytic Core of HeLa and Calf ThymusDNA Polymerase E* (Received for publication, October 24,1990)
Tapio Kesti and Juhani E. SyvaojaS From the Department of Biochemistry, University of Oulu, Linnanmaa, SF-90570 Oulu, Finland
DNA polymerase t, formerly known as a proliferating cell nuclear antigen-independent form of DNA polymerase 6, has been shown elsewhere to be catalytically and structurally distinct from DNA polymerase 6. The catalytic activity of HeLa DNApolymerase e, an enzyme consisting of >200- and 55-kDa polypeptides, was assigned to the larger polypeptide by polymerase trap reaction. This catalytic polypeptide was cleaved by incubation with trypsin into two polypeptide fragments with molecular masses of 122 and 136 kDa, the former of which was relatively resistant to further proteolysis and possessed the polymerase activity. The cleavage increased the polymerase and exonuclease activities of the enzyme some 2-3-fold. DNA polymerase t was also purified in a smaller 140-kDa form from calf thymus. The digestion of this form of the enzyme by trypsin also generated a 122-kDa polypeptide. These results suggest that the catalytic core of DNA polymerase t is a 258-kDa polypeptide that is composed oftwo segments linked with a protease-sensitive area. One of the segments harbors both DNA polymerase and 3‘ + 5‘ exonuclease activities. In spite of the different polypeptide structures, the catalytic properties of the HeLa enzyme, its trypsin-digested form, and the calf thymus enzyme remained essentially the same.
many properties with DNApolymerase 6 , including a readily detectable exonuclease activityandresponsetonumerous inhibitors (reviewed in Burgers (1989), Downey et al. (1990), and Syvaoja (1990)) and has therefore, until recently, been regarded as a form of DNA polymerase 6 . It differs, however, from DNA polymerase 6 by its inherent processivity in the absence of PCNA (Sabatino et al., 1988; Focher et al., 1989; Syvaoja and Linn,1989). Unlike DNApolymerase 6 , it isalso inhibited by dimethyl sulfoxide and is relatively resistant to inhibition by carbonyldiphosphonate (Syvaoja et al., 1990). An antibody against DNA polymerase 6 is not able to recognize DNA polymerase t inimmunoblotting(Wong et al., 1989). Furthermore, DNApolymerases 6 and t can bepurified simultaneously from the same HeLa extract, and the large subunits of the two enzymes are structurally distinct when being compared by partial peptide mapping (Syvaoja et al., 1990). Similarresults were also obtained when the large polypeptides of HeLa DNA polymerase e and calf thymus DNA polymerase 6 were compared by peptide mapping (Wong et al., 1989). DNA polymerases a and 6 are probably both required for the replication of chromosomal DNA (Wong et al., 1988; Lee et al., 1989; Tsurimoto and Stillman, 1989; Weinberg and Kelly, 1989). A model, according to which DNA polymerase a carries outlagging strand synthesis and DNA polymerase 6 is required for leading strand synthesis, has been presented (Downey et al., 1988). This model is consistentwiththe properties of the purified enzymes and is further supported DNA polymerase t has been defined as a mammalian DNA by evidencefrom the studies of SV40 DNA replication in uitro polymerase that has a readily detectable, tightly associated (Prelich and Stillman, 1988; Tsurimoto et al., 1990). DNA exonuclease activity and is highly processive in the absence polymerase t has been implicated in DNA repair (Nishida et of proliferating cell nuclear antigen (PCNA)’ (Burgerset al., al., 1988). However, its yeast counterpart is required for cell 1990; Syvaoja et al., 1990). Since DNApolymerases are highly viability, which suggests that it is involved in the replication conserved in eukaryotic cells, it has beenproposed that a of chromosomal DNA (Morrison et al., 1990). Greek letter nomenclature system shouldbe adopted also for The structures of DNA polymerases a and 6 are well deyeast DNA polymerases (Burgers et al., 1990). In this system fined, and they arehighly conserved in eukaryotic cells from the yeast enzyme, formerly called DNA polymerase 11, would yeast to humans (reviewed in Burgers (1989) and Downey et be called DNA polymerase t like its mammalian counterpart. al. (1990)). The structure of DNA polymerase t, however, has PCNA is a replicationfactor(Prelich et al., 1987a) that not been well defined, and the polypeptide compositions of increases theprocessivity of DNA polymerase 6 in vitro (Tan of DNA et al., 1986; Prelich et al., 1987b). DNA polymerase t shares purifiedenzyme preparations vary.Alargeform polymerase t from HeLa cells is composed of >200- and 55* This research was supported by a grant from the Academy of kDa subunits (Syvaoja and Linn, 1989). This study was carof this HeLa enzyme, to Finland. The costsof publication of this article were defrayed in part ried out to identify the catalytic core by the paymentof page charges.This article must therefore he hereby study its domain structure, and to elucidate the relationship marked “aduertisement” in accordance with 18 U.S.C. Section 1734 between this large form and a smaller form that was purified solely to indicate this fact. from calf thymus. $ To whom correspondence should be addressed. ’ The abbreviations usedare: PCNA,proliferating cell nuclear antigen; BrdUTP, 5-bromo-Z’-deoxyuridine 5”triphosphate; BrdUMP, 5-bromo-2’-deoxyuridine 5’-monophosphate;BuAdATP, 2-(p-n-butylanilino)-dATP; BuPdGTP, W-(p-n-butylphenyl)dGTP;EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid; HEPES, N-2-hydroxyethylpiperazine-N’-2-ethanesulfonicacid NEM, N-ethylmaleimide;SDS, sodium dodecyl sulfate.
EXPERIMENTALPROCEDURES
Materials-Unlabeled nucleotides and calf thymus DNA were purchasedfromBoehringerMannheimand labelednucleotidesfrom Amersham Corp. Synthetic deoxypolymers, BrdUTP, aphidicolin, Nethylmaleimide (NEM), and trypsin were from Sigma, and soybean trypsin inhibitor from Fluka. BuAdATP and BuPdGTPwere kindly
6336
Catalytic Core of DNA Polymerase t donated by Dr. G. E. Wright (University of Massachusetts Medical School). DNA primer/templates were annealed as earlier described (Syvaoja et al., 1990). Neutralizing monoclonal antibody SJK 132-20 IgG againsthuman DNApolymerase a waspurified asdescribed (Tanaka et al., 1982). HeLa DNA polymerase t was purified from 75 liters of late logarithmic phase culture (Nishida et al., 1988; Syvaoja and Linn, 1989). Calf thymus DNA polymerase a was purified from a phosphocellulose side fraction of DNA polymerase t purification (see "Results")as described forthe purification of HeLa DNA polymerase (Y (Syvaoja et al.,1990), except that the heparin-agarosecolumn was included as an additional step before the glycerol gradient centrifugation. The polymerase a preparation obtainedhad nodetectable exonuclease activity. Enzyme and Protein Assays-Unlessotherwise indicated, DNA polymerases a and z were assayed as described (Syvaoja and Linn, 1989; Syvaoja et al., 1990), except that the concentration of activated DNA in the assay mixture for DNA polymerase a was 1.3 mM.One unit of DNA polymerase catalyzes the incorporation of 1 nmol of nucleotide in 60 min. Exonuclease activity wasassayedin a total volume of 30 pl, containing 50 mM HEPES-KOH (pH 7.5), 167 pg/ ml bovine serum albumin, 5 mM MgClz, and 4 p M (dT)15[3H](dT)40. The reaction mixture was incubated at 37 "C for 30 min after which it was treated as was described earlier (Nishida et al.,1988). Protein concentrations were determined according to Bradford(1976). Digestion ofDNA Polymerase t with Trypsin-For trypsin digestion, 2 pl of trypsin solution were added to 15-pl polymerase samples each containing approximately 50 ng of the large polypeptide of a givenenzyme.Sampleswere incubated for 15 min at 37 "C after which 7-pl aliquots were immediately prepared for SDS-polyacrylamidegelelectrophoresis.Soybean trypsininhibitor (25 pg) was added to the rest of the samples to be assayed for polymerase and exonuclease activities and to be analyzed later with the polymerase trap reaction. Control samples were treated similarly in the absence of trypsin. Polymerase Trap Reactions-The polymerase trap technique was employed essentially as described (Insdorf and Bogenhagen, 1989). This included incorporation of BrdUMP and [cY-~'P]~AMP from the corresponding triphosphatesby a polymerase to a special oligonucleotideprimer/template,cross-linking of incorporatedBrdUMP by ultraviolet light to the catalytic polypeptide, digestion of the excess oligonucleotide by DNases, and localization of the radioactivity on an SDS-polyacrylamide gel after electrophoresis of the precipitated protein. Polymerase reaction was carried out in a volume of 20 pl in a mixture containing 0.5-1.5 units of DNA polymerase, 300 pmol of t h e oligonucleotidein 50 mM HEPES-KOH (pH 7.5),200 pg/ml bovine serum albumin, 5 mM dithiothreitol, 0.05% Triton X-100, 15 mM MgClz, 0.2 mM BrdUTP, 1.0 p M [a-"PIdATP (1000 Ci/mol), and 70 mM KCl. A relatively high concentration of oligonucleotide was required for maximal cross-linking of DNA polymerase e . SDS-PolyacrylamideGel Electrophoresis-SDS-polyacrylamide gel electrophoresis wasperformed on slabgels of 10 X 17 X 0.08 cm according to Laemmli(1970) with constant 25-mA current. The silver staining protocol was that of Wray et al. (1981). RESULTS
Purification of DNA Polymerase t from Calf Thymus-The procedure for the purificationof DNA polymerase t from calf thymus (Table I) was modified from that of Nishida et al. (1988) for the purification of the enzyme from HeLa cells. Fresh nonfrozen calf thymus (1.2 kg) from 0.5-1.0-year-old calves was trimmed of fat and cut into small pieces with scissors. The tissue was homogenized in 1 liter of 10 mM HEPES-KOH (pH 7.7), 75 mM sucrose, 0.25% dextran (M, 500,000), 4 mM MgC12, 0.25 mM EGTA, 0.5 mM dithiothreitol with a Sorvall Omni-Mixerfor 1 min (3x 20 s), and then 750 ml of 80 mM HEPES-KOH (pH 7.7), 600 mM sucrose, 2% dextran, 32 mM McG12, 2 mM EGTA, 4 mM dithiothreitol, 933 mM KC1 was added. After 15 min on ice the suspension was centrifuged in a Beckman SW 27 rotor for 60min (21,000 rpm, 2 "C). The supernatantwas dialyzed overnight against 9 liters of 50 mM Tris-HC1 (pH8.0), 1mM dithiothreitol, 1 mM EDTA, 10%(v/v) glycerol (TDEG), changing the buffer once. The precipitate from 30-50% saturation of (NH4)$04 was dissolvedin TDEG and purified after dialysis on DEAE-
6337
TABLE I Purification of DNA polymerase t from calf thymus Fraction
ml
Protein
Total activity
Specific activity
mg
units
unitsfmg
Crude extract 22,000 1,710 ND" ND 30-50% (NH4)2S04 1941.7' 10,000b 6,000 42 86 600 25,000 DEAE-Sephacel 11 50 1,360 15,000 Phosphocellulose 1.6 2 3,400 5,320 Mono Q 0.076 1.5 1,900 25,000 Hydroxylapatite 0.033 4.0 1,100 33,000 Phenyl-Sepharose 0.001' 110 110,000' 0.5 Glycerol/ethylene glycol a ND, not determined. The activity could not be assayed reliably at this step. Estimated value. Sephacel (4.1 X 40 cm) and phosphocellulose (3.2 X 15 cm) columns as described (Nishida et al., 1988). DNA polymerase t eluted from the columns at 170 and 320 mM NaCl, respectively. The phosphocellulose fractions containing peak activity were diluted to a conductivity equivalent to that of 160 mM NaCl andloaded on a Mono Q HR 5 / 5 fast proteinliquid chromatography columnwhich had been equilibrated with 160 mM NaCl in TDEG supplemented with 0.05% Triton X100. The column was washed with 15 ml of 160 mM NaCl in the same buffer and eluted with a 20-ml linear gradient of NaCl from 160 to 450 mM. DNA polymerase t , eluting as a sharp peak at 260 mM NaCl, was diluted 1:4 in 20 mM potassium phosphate (pH 7.5), 20% glycerol, 5 mM dithiothreitol, 0.05% Triton X-100, and purified on a hydroxylapatite column (0.5 X 5 cm) as described(Syvaoja and Linn, 1989). DNA polymerase t eluted a t 70 mM potassium phosphate. The peak fractionswere brought to a final concentration of 1 M (NH4)*S04by adding 3.5 M solution and applied on a phenyl-Sepharose column (Pharmacia LKB Biotechnology Inc., 0.5 X 3 cm) equilibrated with 1 M (NH4)2S04in 20 mM potassium phosphate (pH7.5), 20% glycerol, 5 mM dithiothreitol, 0.005% Triton X-100. The column was washed with 25 ml of the same solution, eluted with a 20-ml descending gradient of (NH4)$04 from 1 to 0 M, and finally washed with the buffer containing no (NH4),S04. The majority of DNA polymerase t elutedat 0.1-0 M salt.Peakfractions from phenyl-Sepharose column chromatography were concentrated with a Centricon-30 microconcentrator(Amicon) and applied on a 5-ml glycerol gradient (0-10%) in 50% ethylene glycol, 385 mM potassium phosphate (pH 7.5), 5 mM dithiothreitol, and 0.05% Triton X-100, and centrifugedfor 61 hina Beckman SW 50.1 rotor (47,000 rpm, 2 "C). Fractions were collected from the bottom of the tube. DNA polymerase and exonuclease activities sedimented together in this gradient (Fig. 1).One major polypeptide with a molecular mass of 140 kDa was detected when the purified enzyme was analyzed on SDS-polyacrylamide gels (Fig. 2 A ) . Identification of the Catalytic Core of HeLa DNA Polymerase t-The DNA polymerase trap technique was used to localize the polymerase activity of HeLa DNA polymerase t. DNA polymerase activity was detected inpolypeptides of >200 and 108 kDa (Fig. 2 A ) . The molecular mass of the larger polypeptide corresponded precisely to the molecular weight of the >200-kDa polypeptide found in silver staining of HeLa DNA polymerase t. No polypeptide corresponding to the 108-kDa activity was detectable in silver staining. The activity at108 kDa could be due to a minor amountof a contaminating DNA polymerase that mightbe very active with theoligonucleotide primer/template used in polymerase trap reaction. An alternative explanation would be partial breakdown of the >200-
Catalytic Core of DNA Polymerase c
6338
10
0
20
Fraction number
FIG.1. Cosedimentation of the DNA polymerase (W) and exonuclease (0)activities of calf thymus DNA polymerase e. Centrifugation was carried out in 0-10% (v/v) glycerol gradient in the presence of 50% (v/v) ethyleneglycol as descibed under “Results.” U, units.
A
* Calfthymus
TryP*l” 0
(ng)
200
-
Sllver stain
I
2000
I
HeLa pol
0
Polvmerase trap
E
HeLa pol E 0
10 2000 200
10
200
”
C C
29
t
I
.
im
-
kDa polypeptide. No activity was detected in the 55-kDa polypeptide of the enzyme. These results indicate that the DNA polymerase activity of HeLa DNA polymerase t is located in the larger >POO-kDa subunit. Tryptic Cleavage of the Catalytic Core of DNA Polymerase €-The large subunit of HeLa DNA polymerase e was cleaved by 10 ng of trypsin into 136- and 122-kDa polypeptides, the smaller of which contained the polymerase activity when measured by the polymerase trap technique (Fig. 2 4 ) . Contrary to the situation before trypsin treatment, no activity was found a t 108 kDa, indicatingthat thisform of the polymerase was inactivated. When the amount of trypsin was increased to 200 ng, the 136-kDa polypeptide was hydrolyzed further, and neither it nor its breakdown products could be detected by silver staining. The 122-kDa fragment and the small 55-kDa polypeptide of the enzyme were relatively resistant to hydrolysis by tryspin, andthey were the only peptides detected. The activity was still located in the 122kDa fragment. Interestingly, the enzyme from calf thymus that was purified as a 140-kDa polypeptide was also digested by trypsin intoa 122-kDa polypeptide, but compared with the HeLa enzyme, a higher concentration of tryspin (2000 ng) was required. This may be due to the presence of ethylene glycol in the calf thymus enzyme samples, which could be inhibitory to trypsin. Both the polymerase and exonuclease activities of the tryptic fragment of the HeLa cell enzyme were 2-3-fold those of theintact enzyme (Fig. 2B). The activity of the calf thymus enzyme, however, was not increased when cleaved by trypsin. These resultsindicated that thecatalytic polypeptide of the mammalian DNA polymerase e is a 258-kDa polypeptide, which is composed of two segments linked by a proteasesensitive region. The results further indicated that both the DNA polymerase and exonuclease activities are located in the 122-kDa segmentof the polypeptide. The smaller form of the enzyme from calf thymus may also be a proteolytic fragment of the larger polypeptide. The increase in both the polymerase and exonuclease activities as a result of the cleavage of the HeLa enzyme suggested that the catalytic domains of the intact enzyme are partially masked in the 258-kDa polypeptide and are exposed by the removal of the noncatalytic segment. Comparison of the 258-kDa Polypeptide of HeLa and the 140-kDa Polypeptide of Calf Thymus DNA Polymerase e by Partial Peptide Mapping-To confirm the structuralrelation between the 258-kDa polypeptide of the HeLa enzyme and the 140-kDa polypeptide of the calf thymus enzyme, these two polypeptides were compared by partial peptide mapping using N-chlorosuccinimide (Fig. 3). Numerous putative common peptide fragments canbe detected onthe partialpeptide maps of the two enzymes, suggesting that thetwo polypeptides are structurally related and that the smaller 140-kDa polypeptide could be derived from the 258-kDa polypeptide.
Comparison of the Properties of theHeLaEnzyme,Its Tryptic Fragment, and the Calf Thymus Enzyme-The three Trypsin (ng) structurally different forms of DNA polymerase t provide an FIG.2. Identification of the catalytic polypeptide of HeLa opportunity to analyze the roles of different parts of the D N A polymerase (pol)e and trypticcleavage of the HeLa and enzyme protein in catalytic processes. Therefore, the recalf thymus enzymes. Apparently homogenous enzymes from HeLa cells and calf thymus were digested by varying amounts of trypsin sponses of the three forms of the enzyme to numerous com( A ) .Arrowheads indicate the intact and trypsin-digested form of the pounds and theirprimer/template preferences were compared putative catalytic polypeptide of calf thymus DNA polymerasee. The (Table 11). In studies on the inhibitor/activator effects, the polypeptides with polymerase activities were identified by DNA po- same template/primer, poly(dA) .oligo(dT), was used for all lymerase trap technique. The bands of varying molecular mass in the forms of the enzyme and for DNA polymerase a,which was range of 20-35 kDa are also present in nonirradiated samples (not shown) and hence represent nonspecific binding of dATP. All samples included for the sake of comparison. The HeLa enzyme, its trypsin-digested form, and thecalf thymus enzyme as well as were assayed for polymerase (W) and exonuclease (Ed) activities in standard assay conditions(R). DNA polymerase CY were equally sensitive to inhibition by 0
2000
0
10
200 2000
Core Catalytic
of DNA Polymerase
NEM. Aphidicolin was equally effective in inhibiting HeLa and calf thymus DNA polymerase t and DNA polymerase a. Both HeLa and calf thymus DNA polymerase t were more resistant thanDNA polymerase a to inhibition by BuAdATP and BuPdGTP. The HeLa and calf thymus enzymes were also inhibited by dimethyl sulfoxide, while DNA polymerase a was activated by this compound. Potassium chloride also
FIG.3. Partial peptide mapping of HeLa and c a l f t h y m u s DNA polymerase z with N-chlorosuccinimide. The samples for partial peptide mapping were the large polypeptides (1 pg) of HeLa and calf thymus enzymes. They were obtained from an SDS-polyacrylamide gel after staining with Coomassie Brilliant Blue R-250 (Fluka). The bandswere cut out from thegel, digested by N-chlorosuccinimide, equilibrated, and electrophoresed on a 15% SDS-polyacrylamide gel, which was stained with silver (Lischwe and Ochs, 1982). Putative common polypeptides are indicated on the right.
6339
E
inhibited both forms of the HeLa enzyme as well as the calf thymus enzyme when assayed with poly(dA) .oligo(dT). The effects of the compounds tested were consistent with the data published for both HeLa and calf thymus DNA polymerase e (Crute et al., 1986; Wahl et al., 1986; Focher et al., 1988; Nishida et al., 1988; Focher et al., 1989; Syvaoja and Linn, 1989; Syvaoja et al., 1990). Under the conditions employed, neither HeLa nor calf thymus DNA polymerase e enzyme was able to use other primer/templatesefficiently. All three forms were, however, relatively active with DNase-activated DNA when KC1 was included a t a concentration of 140 mM. The ratio of polymerase to exonuclease activities of the HeLa and calf thymus DNA polymerase t was 0.40 and decreased to 0.27 upon treatment of the HeLa enzyme with trypsin. These results together with the polypeptide structures andsedimentation coefficients are summarized in Table 111. Except for the increase of both polymerase and exonuclease activities and theslight decrease of the ratio of polymerase to exonuclease activities, the digestion of the HeLaDNA polymerase t into a smaller form by trypsin did not significantly affect the catalytic properties of the enzyme. This suggests that the 136-kDa noncatalytic segment of the catalytic core of the HeLa enzyme does not interact with DNA but may be involved in protein-protein interactions instead. It could also be required for binding the 55-kDa polypeptide to thecatalytic core of the enzyme. The catalytic properties of the two HeLa enzyme forms also appeared the same as the catalytic properties of the calf thymus enzyme, providing further evidence for the calf thymus enzyme being a proteolytic fragment from a larger form of the enzyme. DISCUSSION
DNA polymerase t was purified from calf thymus first as DNA polymerase 61,(Crute et al., 1986). This preparationalso possessed primase activity and consisted of several polypeptides ranging in size from 45 to 245 kDa. The polymerase activity was not assigned to any particular polypeptide. A second preparation from calf thymus was called PCNA-independent DNA polymerase 6 and consisted of polypeptides of
TABLE I1 Primerltemplate preferencesand effects of various compounds on the polymerase (pol) activities of HeLa DNA polymerase t, its tryptic fragment,and calf thymus DNA polymerase c DNA polymerase cr is shown for comparison. Except for primer/template, standard assay conditions were used for each enzyme.However, the concentration of MgC12 for DNA polymerase (Y was decreased to 1 mM. For measuring theeffect of NEM, the concentrationof MgClr ws 3 mM and the concentrationof dithiothreitol was 0.5 mM for all enzymes studied. SJK 132-20 IgG is a monoclonal antibody raised against human DNA polymerase CY (Tanaka et al., 1982). The trypsin-digested form of the HeLa DNA polymerase t was the one digested by 200 ng (Fig. 2A). The activity of each enzyme with poly(dA).oligo(dT) was taken as 1, and 0.1 unit was used in each assay. Primer/ template
Added factor
Quantity
SJK 132-20 Aphidicolin NEM BuAdATP BuPdGTP Dimethyl sulfoxide KC1 KC1
2 rg 10 p d m l 2 mM 20 p M 20 p M 10% 60 mM 140 mM
KC1
140 mM
Poly(dA). oligo(dT)
Activated DNA Poly(dA-dT) Denatured DNA Native DNA
HeLa pol 6
1.00 1.19 0.24 0.04 0.85 0.72 0.25 0.80 0.02 0.22 0.53 0.07 0.04
