ANALYTICAL

BIOCHEMISTRY

1%!,96-103

(1991)

Detection of Possible DNA Repair Enzymes on Sodium Dodecyl Sulfate-Polyacrylamide Gels by Protein Blotting to Damaged DNA-Fixed Membranes Shogo Ikeda,l Shuji Seki,2 Sekiko Watanabe, Masao Hatsushika,

and Ken Tsutsui

Department of Biochemistry, CancerInstitute, Okayama University Medical School, Okayama 700,Japan

Received May 21, 1990

A novel method for detecting possible DNA repair enzymes on sodium dodecyl sulfate-polyacrylamide gels by blotting them onto a damaged DNA-fixed membrane is presented. To prepare the membrane, highly polymerized calf thymus DNA immobilized on a nylon membrane is damaged chemically. Enzymes, either homogeneous or crude, that are possibly involved in the priming step of DNA repair are fractionated by SDSpolyacrylamide gel electrophoresis (SDS-PAGE) and are renatured to active form by incubating the gel in an appropriate buffer. The renatured enzyme is then blotted onto the damaged DNA-fixed membrane, a process during which incision andfor excision are introduced to the damaged DNA by the enzymes. The incision and/or excision provide priming sites for repair DNA synthesis in the subsequent step in which the membrane is incubated with DNA polymerase in the presence of (r-32P-labeled substrate. The site of substrate incorporation on the membrane reflecting the molecular weight of the repair enzyme is finally visualized by autoradiography. The present technique is established using Escherichia coli exonuclease III and a DNA-fixed membrane treated with bleomycin or acid-depurinated. By application of this method, a priming factor (an exonuclease) involved in the initiation of bleomycin-induced DNA repair is detected in the extract of mouse ascites sarcoma cells, and thus the molecular weight of the enzyme is estimated. Some apuriniclapyrimidinic endonucleases of mammals are also detected by the present procedure. o lssl Academic

Press,

Inc.

damage and incision-excision reaction), repair DNA synthesis, and repair patch ligation. In mammalian cells, DNA polymerases and DNA ligases involved in DNA repair have been limited to a few candidates, but factors involved in the priming step are complicated because of the variety of DNA damage. Only several enzymes that catalyze the priming reaction for specific types of DNA damage have been purified and characterized (l-4). The authors have been studying the repair mechanism of bleomycin-damaged DNA in vitro in mammalian cells (5-11). Bleomycin is known to cause at least three types of DNA damage. The major type is a single-strand

DNA

excision

repair is known reactions: priming

to occur by the follow(recognition of DNA

’ Present address: Department of Biological Chemistry, Faculty of Science, Okayama University of Science, Okayama 700, Japan. ’ To whom correspondence should be addressed.

resulting

(apurinic/apyrimidinic

in 3’-phosphoglycolate

(12,13). Alkali-labile

sites)

and double-strand

sites breaks

are also produced (14). To repair the single-strand breaks a priming enzyme that can recognize and remove 3’-phosphoglycolate is required. In Escherichia coli, exonuclease III was reported to be involved in bleomycininduced DNA repair as some kind of a priming enzyme (15), although AP3 sites generated by bleomycin are not cleaved readily by the enzyme (15,161. Recently we found

a priming

enzyme

involved

in the

initiation

of

bleomycin-induced DNA repair in the extract of mouse ascites sarcoma cells (10,ll). In this paper we describe a novel method for detecting priming factors (priming enzymes) on SDS-polyacrylamide gels by protein blotting to a damaged DNA-fixed membrane. The method was developed using exonuclease III and a DNA-fixed membrane acid-depurinated or treated with bleomycin. The molecular weight of the mouse

ing sequential

break

and 5’-phosphate termini

priming

enzyme

was determined

by using

this

method. 3 Abbreviations used: AP, apurinic/apyrimidinic; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; BLM, bleomycin; CBB, Coomassie brilliant blue R-250; BSA, bovine serum albumin; TEMED, N,N,W,N’-tetramethylethylenediamine.

96 All

Copyright 0 1991 rights of reproduction

0003-2697/91$3.00 by Academic Press, Inc. in any form reserved.

DETECTION

MATERIALS

AND

METHODS

OF

REPAIR

ENZYMES

USING

DNA-FIXED

MEMBRANES

97

phenylmethylsulfonyl fluoride was added to all buffers used. After adjustment of the KP, concentraMaterials. Mouse ascites sarcoma (SR-C3H/He) tion to 0.1 M, the extract was mixed with packed phoscells were obtained and maintained as described phocellulose equilibrated with 0.1 M KP, and rocked previously (17). Klenow fragment (Klenow polymerase) gently at 4°C overnight. The phosphocellulose was of E. coli DNA polymerase I and E. coli exonuclease III washed three times with 0.1 M KP, and then transwere purchased from Takara Shuzo Co. Ltd., Kyoto, ferred onto a column. The enzymes were eluted with Japan. Bovine serum albumin (BSA) and pancreatic 0 3 M KP, DNase I were obtained from Sigma Chemical Co. Mouse . Preparckon of E. coli whole cell extract. E. coli strain DNA polymerase @was purified from SR-C3H/He cells HBlOl (19) was cultured at 37°C in 10 ml of Luria-Beraccording to the method of Chang with slight modificatani medium until late log phase. The cells were coltions (18). One unit of the enzyme is defined as the incorlected and resuspended in 1 ml of the loading buffer for poration of 1 nmol of total nucleotide into activated SDS-PAGE (20). The cell suspension was boiled for 5 DNA per hour at 37°C. Calf thymus DNA was purmin and then centrifuged at 10,OOOgfor 10 min. The chased from Pharmacia LKB Biotechnology. Nylon supernatant was used as E. coli whole cell extract. membrane (Hybond-N) was purchased from AmerPolyacrylamide gel electrophoresis. SDS-polyacrylsham. Reagents for electrophoresis were obtained from Nakarai Tesque, Inc., Kyoto, Japan. The other reagents amide gel electrophoresis was performed as described by Laemmli (20) using an 8 X 8.5 X O.l-cm slab gel. The used were obtained as described previously (10). Fixation of DNA on nylon membrane. The stock so- electrode buffer contained 0.025 M Tris-base, 0.192 M glycine, 0.1% SDS, and 2 mM EDTA. The stacking gel lution of highly polymerized calf thymus DNA (2.5 mgf contained 3% acrylamide, 0.08% N,N’-methylenebisml) was diluted with 2~ SSC (1X SSC is 0.15 M NaCl, acrylamide, 0.125 M Tris-HCl, pH 6.8,0.1% SDS, 2 mM 0.015 M sodium citrate, pH 7.0) to a final concentration EDTA, 0.1% N,N,N’,N’-tetramethylethylenediamine of 0.145 mg/ml. A nylon membrane (6 X 9 cm) prewetted in 2~ SSC was placed in a sealable plastic bag and (TEMED), 0.03% ammonium persulfate, and 10 pg/ml incubated in 2 ml of the diluted DNA solution for 1 h at BSA. The separation gel contained 12.5% acrylamide, 0.375 M Trisroom temperature. After incubation, the membrane was 0.33% N,N’-methylenebisacrylamide, HCl, pH 8.8, 0.1% SDS, 2 mM EDTA, 0.05% TEMED, rinsed three times in 2~ SSC and then air-dried over0.05% ammonium persulfate, and 10 pg/ml BSA. The night in the dark. It was wrapped loosely in aluminum samples were adjusted to 0.0625 M Tris-HCl, pH 6.8,2% foil and stored in a desiccator at room temperature unSDS, 2.5% 2-mercaptoethanol, 0.25 M sucrose, and til use. 0.01% bromphenol blue, incubated at 100°C for 2 min, Chemical modification of DNA fixed on the mem- and loaded onto the gel. Electrophoresis was conducted brune. To damage DNA on the membrane with bleomy- at room temperature at a constant current of 15 mA tin, native DNA-fixed membrane was treated with Triuntil the dye front reached the bottom of the gel (apton-buffer B (0.0175% Triton X-100, 0.25 M sucrose, 10 proximately 3 h). To determine molecular weights of mM Tris-HCl, 4 mM MgCl,, 1 mM EDTA, and 6 mM 2- proteins, the following standard proteins were used for mercaptoethanol, pH 8.0) supplemented with 5 pg/ml calibration: BSA (A4, 68,000), ovalbumin (A!, 43,000), bleomycin A, and 30 pM ferrous ammonium sulfate (10). carbonic anhydrase (M, 29,000), trypsinogen (M, To depurinate DNA, the native DNA-fixed membrane 24,000), and lysozyme (M, 14,300). was incubated in 37.5 mM sodium citrate (pH 3.5) at 60°C Renaturation of enzymes on SDS-polyacrylamide gel for 30 min. After the treatment with DNA-damaging and protein blotting to DNA-fired membrane. After agents, the membranes were rinsed three times in 2X electrophoresis the gel was rinsed briefly in a buffer SSC, rinsed once in a blotting buffer (0.04 M Tris-HCl, pH consisting of 0.04 M Tris-HCl, pH 8.0, 2 mM MgCl,, 8.0, 2 mM MgCl,, 0.02% sodium azide, and 7 mM 2-mer0.02% sodium azide, and 0.1% (W/V) Triton X-100 and captoethanol), and used immediately for protein blotting. shaken gently in the buffer at room temperature for 1 h Extraction and partial purification of mammalian re- with two changes of the buffer. The gel was left overpair enzymes. A priming enzyme that can recognize night at room temperature in the fresh buffer with genbleomycin-damaged sites of DNA and remove the tle shaking. The gel was transferred to the blotting damaged ends was extracted from SR-C3HHe cells buffer described above and shaken gently for 30 min. and fractionated by ion-exchange chromatographies The damaged DNA-fixed membrane was then placed as described previously (10). Enzymes having priming onto the gel and proteins were transferred to the memactivity on bleomycin-damaged DNA and acid-depurbrane with the blotting buffer at 28°C for 2 days, using inated DNA for DNA polymerase were extracted from the capillary transfer procedure described for DNA SR-C3H/He cells, bovine liver, and HeLa cells with transfer by Southern (21). After the blotting, the mem0.2 M potassium phosphate buffer (pH 7.5) (KP,) acbrane was washed briefly in TEN buffer (10 mM Triscording to the method of Chang (18), except that 0.2 HCl, pH 8.0,1 mM EDTA, 100 mM NaCl). mM

98

IKEDA

Native DNA Em--(units)

Qo1 (I1

1 2

ET

AL.

BLM-treated iii-@-j o.ol

34

itiz-a1

Exoll ’ (l”&+) ODl

a1

0

DNA

w106KTKTww

1

1

s

“56

FIG. 1. Detection of exonuclease III using DNA-fixed membranes. Purified exonuclease III (sp act, 94 units per microgram of protein) was electrophoresed on a SDS-polyacrylamide gel (0.01 unit in lanes 1, 3, and 5; 0.1 unit in lanes 2, 4, and 6). After renaturation, the enzyme was blotted onto native DNA-fixed membranes (lanes 1 and 2) or DNA-fixed membranes treated with bleomycin (lanes 3 and 4) or acid-depurinated (lanes 5 and 6). DNA polymerase /3 was used for repair DNA synthesis. The membranes were exposed to Fuji X-ray film at -60°C for 2 days.

Repair DNA synthesis and localization of the priming enzymes. The blotted membrane was incubated with 3% BSA in TEN buffer at 37°C for 1 h to saturate remaining protein binding sites and then washed briefly with a buffer (0.04 M Tris-HCl, pH 8.0,5 mM MgCl,, 50 InM NaCl). The membrane was placed in a sealable plastic bag and incubated at 37°C for 30 min in 2 ml of a DNA polymerase-substrate solution containing 0.04 M Tris-HCl, pH 8.0, 5 mM MgCl,, 50 mM NaCl, 5 mM 2-mercaptoethanol, 100 pM each dATP, dGTP, and dTTP, 1 PM unlabeled dCTP, 10 PCi of [a-32P]dCTP (3000 Ci/mmol, ICN Biochemicals Inc.), and 1.4 units of DNA polymerase p (or 0.7 unit of Klenow polymerase). Two milliliters of the reaction cocktail was used for about 50 cm2 of the blotted membrane. After the incubation for DNA synthesis, the membrane was washed in 5% (W/V) trichloroacetic acid with five changes for 30 to 60 min each to remove unincorporated nucleotides and air-dried after being absorbed with paper towels. The membrane was processed for autoradiography by using Fuji RX X-ray film and DuPont Cornex Lightning-plus screens at -70°C. Assay for AP endonuclease. Superhelical (form I) pUC18 DNA was depurinated by incubation with 3 vol of 50 mM sodium citrate (pH 3.5) at 60°C for 15 min as described previously (10,15). After the incubation the mixture was chilled and dialyzed at 0°C for 5 h against 50 mM Tris-HCl (pH 7.5) and then overnight against distilled water. The assay mixture for AP endonuclease, in a final volume of 15 ~1, contained 10 mM Tris-HCl

FIG. 2. Sensitivity of detection for exonuclease III blotted onto DNA-fixed membranes treated with bleomycin. Purified exonuclease III was serially diluted, electrophoresed, and blotted onto DNA-fixed membranes treated with bleomycin. The units of exonuclease III loaded per lane were as follows: lane l,O; lane 2, 10W6; lane 3, 10s5; lane 4, 10e4; lane 5, lo-%; lane 6, 10m2; lane ‘7,10-l; lane 8,1. DNA polymerase p was used for repair DNA synthesis. The exposure for autoradiography was conducted for 2 days.

(pH 8.0), 4 mM MgCl,, 1 mM EDTA (pH 8.0), 6 mM 2-mercaptoethanol, 0.5 pg of depurinated DNA, and O-7.5 ~1of an appropriate dilution of enzymes separated by SDS-PAGE, eluted from the sectioned gel, and renatured according to the method of Hager and Burgess

C8B stain ‘Ilk

Native DNA W”

12

W

34

E ”

Aciddeqwinated DNA

treated DNA W

56

E ”

W

E ’

78

FIG. 3. Detection of exonuclease III in Escherichia coli whole cell extract using DNA-fixed membranes. E. coli whole cell extract (W: 10 pl; lanes 2,3,5, and 7) and 10 mU of purified exonuclease III (E: lanes 4,6, and 8) were electrophoresed and blotted onto a native DNA-fixed membrane (lanes 3 and 4) or DNA-fixed membranes treated with bleomycin (lanes 5 and 6) or acid-depurinated (lane 7 and 8). DNA polymerase p was used for repair DNA synthesis. The exposure for autoradiography was conducted for 1 day. Lane 1 (Mr, molecular weight standard) and lane 2 were electrophoresed on a separate gel containing no BSA, and stained with Coomassie brilliant blue R-250 (CBB).

DETECTION

kDa

29, 24,

OF

REPAIR

ENZYMES

ii

* * *

FIG. 4. Detection of priming factors of E. coli whole cell extract for DNA synthesis on DNA-fixed membrane treated with bleomycin. Preparation of E. coli whole cell extract, SDS-PAGE, protein renaturation, blotting on DNA-fixed membrane treated with bleomycin, DNA synthesis, and autoradiography were conducted as described under Materials and Methods and in the legend to Fig. 3, except that DNA polymerase I replaced DNA polymerase /3. The exposure for autoradiography was conducted for 8 days. Asterisks indicate positive bands.

(22). The assay mixture was incubated at 37°C for 20 min. The reaction was stopped by chilling to 0°C and by adding 3 ~1 of 6-fold-concentrated gel loading buffer (0.25% bromphenol blue, 0.25% xylene cyanol, and 30% glycerol in H,O). The mixture was loaded into a slot of a submerged 0.8% agarose gel. Electrophoretic analyses of conformation of pUC18 DNA were conducted as described previously (23).

USING

DNA-FIXED

99

MEMBRANES

has been shown to possess 5’ AP endonuclease, DNA 3’-phosphatase, and ribonuclease H activities (2). After fractionation of proteins by SDS-PAGE, the gel was incubated in the buffer containing Triton X-100 to remove SDS and to allow renaturation of enzymes on the gel. The renaturation of enzymes after electrophoresis on SDS-polyacrylamide gels is thought to depend on many factors, including the composition of renaturation buffer, renaturation time, temperature, and so on. Successful detection of DNA polymerases and nucleases after electrophoresis on SDS-polyacrylamide gels containing appropriate template primers or substrate (socalled activity gel) has been reported (24-26). With some modification we successfully applied the conditions described in these reports to renature repair enzymes in SDS-PAGE. The conditions described under Materials and Methods were suitable for the renaturation of all enzymes tested in this study. After the renaturation, the proteins were blotted on damaged DNAfixed membrane by means of the capillary blotting procedure as originally described by Southern (21). Damaged sites of DNA on the membrane were incised and/or excised by the renatured repair enzyme during the blotting for 2 days at 28°C. A nylon membrane was chosen as the support to fix DNA because of its ability to bind double-stranded DNA and its high physical strength. In the blot hybridization technique, nucleic acids are usually fixed on membrane by baking in an oven or by ultraviolet irradiation. However, these physical treatments were avoided in the present experiment because they cause unnecessary DNA damage. DNA fixation by air-drying

BLM-treated

Native DNA RESULTS

AND

considerations. To detect DNA repair enzymes in SDS-PAGE, a useful method was developed. The method includes the following steps: (i) separation of proteins by SDS-PAGE, (ii) renaturation of proteins in gel, (iii) DNA immobilization to a membrane and DNA damage, (iv) protein blotting to the membrane and incision and/or excision of damaged DNA during the blotting process, (v) repair DNA synthesis on the membrane by DNA polymerase in the presence of s-32P-labeled nucleotide, and (vi) detection of the primed and DNA-synthesized sites by autoradiography. The use of DNA-fixed nylon membrane facilitates multiple sequential reactions, such as various modifications of DNA, blotting, DNA synthesis, and autoradiographic or immunologic detection of modified or repaired DNA. The method was developed using E. coli exonuclease III as a model enzyme. It was chosen because of its monomeric structure and well-known characteristics. In addition to 3’ + 5’ exonuclease activity, the enzyme GeneraE

DNA

DISCUSSION

D.&e

1 -lho1 aor

al

1

ID 100

(w)

3Dk-w

1

2345

6

789x)

FIG. 6. Detection of pancreatic DNase I using DNA-fixed membrane. Pancreatic DNase I was serially diluted, electrophoresed, and blotted onto a native DNA-fixed membrane (lanes 1 to 5) or DNAfixed membrane treated with bleomycin (lanes 6 to 10). The amount of DNase I loaded per lane was as follows: lanes 1 and 6,O.Ol ng; lanes 2 and 7,0.1 ng; lanes 3 and 8,l ng; lanes 4 and 9,10 ng; lanes 5 and 10, 100 ng. DNA polymerase (I was used for repair DNA synthesis. The exposure for autoradiography was conducted for 1 day.

100

IKEDA

overnight was sufficient for the present experiments. The DNA-fixed membrane was incubated with a chemical reagent to induce DNA damage, and the reaction was terminated by brief washings. Fixation of DNA on the membrane makes possible easy removal of excess reagents by washing after the treatment and allows multiple sequential reactions on the membrane. The free 3’-hydroxyl termini generated in the damaged DNA by the priming reaction during the blotting process were used in DNA synthesis by mouse DNA polymerase ,# or Klenow polymerase. DNA synthesis was demonstrated by autoradiography at the membrane site blotted with the priming enzyme. The DNA synthesis (32P-labeled nucleotide incorporation) by Klenow polymerase was higher than that by DNA polymerase ,8 under the present conditions. It may be due to the 3’ + 5’ exonuclease activity of Klenow polymerase. Therefore, the autoradiography required longer exposure with DNA polymerase p than with Klenow polymerase. Background, however, was relatively high when Klenow polymerase wasused. Detection of exonuclease III. Exonuclease III is thought to be involved in the initiation of repair of bleomycin-treated DNA in E. coli by removing 3’-blocking damages (15). To detect the enzyme electrophoresed on SDS-PAGE, native DNA-fixed membranes were modified by bleomycin treatment or acid-depurination, and the enzyme on SDS-polyacrylamide gel was blotted onto the membranes. After repair DNA synthesis and autoradiography, a band (28 kDa) corresponding to exonuclease III was observed on both types of damaged DNA-fixed membranes (Fig. 1). The result indicates that the exonuclease III electrophoresed on SDS-gel was renatured, and that the enzyme converted the damaged DNA on the nylon membrane to active template primers for DNA polymerase during the blotting process. The faint bands observed on the control (native DNA-fixed membrane) may arise from exonucleolytic attack of the enzyme to ends or nicks already present in the control DNA. In the present experiment, bleomytin-treated DNA and acid-depurinated DNA are thought to be converted to active template primers for DNA polymerase by 3’ --, 5’ exonuclease activity and the sequential reaction with 5’ AP endonuclease and 3’ + 5’ exonuclease activities of exonuclease III, respectively. As little as 10m3unit of exonuclease III could be detected using bleomycin-treated DNA-fixed membrane (Fig. 2). Exonuclease III, which is known to be a major AP endonuclease in E. coli (2,27), was detectable even in whole cell extract by using this technique (Fig. 3). In the present detection system using Klenow polymerase or DNA polymerase /3, the nucleotide incorporation depends largely on gaps generated by exonucleolytic activity of the priming enzymes. Therefore, the enzymes having both 5’ AP endonuclease and 3’ exonuclease activities,

ET

AL.

e-I----1 kkP

Bulctr&ed NativeDNA DNA E

PE

P

FIG. 6. Detection of a mouse priming enzyme for bleomycin-induced DNA synthesis in its partially purified preparation using bleomycin-treated DNA-fixed membrane. A partially purified mouse priming enzyme (P) for bleomycin-induced DNA synthesis (lanes 2, 4, and 6) and purified exonuclease III (E: 0.1 mU, lanes 3 and 5) were electrophoresed and blotted onto a native DNA-fixed membrane (lanes 3 and 4) or DNA-fixed membrane treated with bleomycin (lanes 5 and 6). Klenow polymerase was used for repair DNA synthesis. The exposure for autoradiography was conducted for 2 days. Lane 1 (molecular weight standard) and lane 2 were electrophoresed on a separate gel without BSA, and stained with CBB. The band of the priming enzyme electrophoresed at 34 kDa is indicated by an asterisk.

like exonuclease III, are easily detected by the present method using the acid-depurinated DNA-fixed membrane, whereas endonuclease IV, an important enzyme for repair of AP sites produced by bleomycin (16), was not clearly demonstrated in the present system. It is suggested that there are structural differences in the AP site produced by simple depurination and that produced by bleomycin. The AP site produced by bleomycin is resistant to exonuclease III but sensitive to endonuclease IV (15,16). In the present experiment exonuclease III was detected easily, because bleomycin-damaged DNA has single-stranded DNA breaks as the major lesion which are sensitive to exonuclease III, and also acid-depurinated (simple depurinated) DNA is cleaved easily by exonuclease III. Modification of the detection system is thought to be required to detect repair enzymes having no exonuclease activity, such as endonuclease IV. Figure 4 shows such an attempt. In the experiment DNA polymerase I, having properties of nick-translation and strand displacement DNA synthesis, was used in place of Klenow polymerase or DNA polymerase /3. To detect faint bands the autoradiography was overexposed. Therefore, each band became broad. In addition to the heavy band due to exonuclease III, at least four positive bands were detected, although bands adjacent to the heavy band of exonuclease III were difficult to document in the photographic print because of the halation of the heavy band. Enzymes corresponding to these light bands are not yet identified.

DETECTION

Priming

0

OF

Activity

REPAIR

ENZYMES

(cpm x 10e3) 1

2

kDa

66, 43,

-FIG. 7. Priming activity for bleomycin-induced DNA synthesis of the samples obtained from sections of a SDS-polyacrylamide gel. A crude preparation of the mouse priming enzyme was fractionated by SDS-PAGE. After staining with CBB and destaining, the gel was sliced into about 4-mm sections. Proteins were eluted from the sections and renatured according to the method of Hager and Burgess (22). The renatured proteins were concentrated 50-fold by ultrafree CL-TGC filter units (Millipore). The priming activities for DNA polymerase (3 of the concentrates were assayed as described previously (10). The gel stained with CBB is shown on the left of the figure (Mr, molecular weight standard, C, crude preparation of the priming enzyme). The band of the 34-kDa enzyme is indicated by an asterisk. The priming activities of each fraction measured in the presence or absence of bleomycin are shown on the right.

Detection of pancreatic DNase I, a nonrepair enzyme. After having been subjected to electrophoresis, pancreatic DNase I was blotted onto native DNAfixed membranes and DNA-fixed membranes treated with bleomycin. In the autoradiograms, bands with almost the same intensity appeared at an identical position for both types of membrane (Fig. 5). The result indicates that DNase I digested bleomycin-treated DNA and native DNA equally, providing priming sites for DNA polymerase. Therefore, when DNA repair enzymes are sought in crude extracts by this technique, bands appearing on both native and damaged DNA membranes should be excluded from consideration. Priming enzymes detected specifically on damaged DNA-fixed membranes using the present method are possible candidates for repair enzymes and should be examined further by other appropriate methods to determine whether the enzymes are true repair enzymes, because it is possible that nonrepair enzymes give false positive results. Detection of priming enzyme for repair of bleomycindamaged DNA in mouse cells. We found the enzyme activity in extracts of mouse ascites sarcoma cells that converts bleomycin-treated DNA to efficient template primers for DNA polymerase, and partially purified the enzyme (10,ll). Analogous to E. coEi exonuclease III, this enzyme was thought to be an exonuclease acting on

USING

DNA-FIXED

101

MEMBRANES

3’-blocking damages caused by bleomycin. To determine the molecular weight of the priming enzyme, its partially purified preparation was electrophoresed on SDS-polyacrylamide gels, and proteins in the gels were blotted onto DNA-fixed membranes treated with bleomycin or native DNA-fixed membranes. A band showing the priming activity appeared at 34 kDa on the bleomycin-damaged DNA-fixed membrane (Fig. 6). There were no bands on the native DNA-fixed membrane. The result suggested that in mouse cells the priming enzyme for bleomycin-induced repair DNA synthesis consists of a 34-kDa single polypeptide. To confirm this result, the proteins in a crude preparation of the enzyme were fractionated by SDS-PAGE, eluted from the sectioned gel, and renatured (22). Priming activity for bleomycin-induced repair DNA synthesis was assayed as described previously (10). The majority of the priming activity was present in the section corresponding to 34 kDa (Fig. 7). The result agrees well with the result obtained by the use of DNA-fixed membrane treated with bleomycin. Detection of some AP endonucleases in mammalian cells. The present method is useful not only in detection of repair enzymes in crude extracts but also in comparison of repair enzymes from a variety of organisms. Recently we purified extensively the mouse repair enzyme having exonucleolytic activity on bleomycin-damaged DNA (10). The highly purified enzyme electro-

I kDa

cB8 stain MrM -

B

Native DNA l-l

II

M

B

567

Acid-dep@nated DNA

H”M

B

89

H’

.

10

FIG. 8. Detection of priming factors for repair of acid-depurinated DNA in mouse, bovine, and human cell extracts using DNA-fixed membranes. The extraction and fractionation using phosphocellulose of priming enzymes from cells were described under Materials and Methods. The enzymes of mouse ascites sarcoma cells (M: lanes 2,5, and a), bovine liver tissue (B: lanes 3, 6, and 9), and HeLa cells (H: lanes 4, 7, and 10) were electrophoresed and blotted onto a native DNA-fixed membrane (lanes 5 to 7) or DNA-fixed membrane aciddepurinated (lanes 8 to 10). Klenow polymerase was used for repair DNA synthesis. The exposure for autoradiography was conducted for 4 days. Lanes 1 to 4 were electrophoresed on a separate gel without BSA, and stained with CBB. The band of the mouse 34-kDa enzyme is indicated by an asterisk.

102

IKEDA

(a) 12345678

(b)

k&depwinated

Native DNA

I-

12345678

FIG. 9. Assay of AP endonuclease activity of the 34-kDa proteins from mouse cells, bovine liver tissue, and HeLa cells. Mouse ascites sarcoma cells, bovine liver tissue, and HeLa cells were extracted with 0.2 M potassium phosphate buffer (pH 7.5), and the extracts were chromatographed using phosphocellulose. Fractions containing the 34-kDa proteins were electrophoresed on SDS-PAGE as described under Materials and Methods. The band of the 34-kDa protein was cut out and crushed, and the protein was eluted and renatured according to the method of Hager and Burgess (22) as described briefly in the legend to Fig. 7. Protein concentration was roughly estimated from the intensity of the stained band. The preparation of acid-depurinated pUC18 DNA and assay of AP endonuclease activity were performed as described under Materials and Methods. (a) Lane 1, acid-depurinated DNA (control); acid-depurinated DNAs incubated at 37°C for 20 min with 1 ng (lane 2) and 3 ng (lane 3) of the mouse 34-kDa protein (M), with 25 ng (lane 4) and 75 ng (lane 5) of the bovine 34-kDa protein (B) with 1 ng (lane 6) and 3 ng (lane 7) of the HeLa 34-kDa protein (H), and without the 34-kDa proteins (lane 8). (b) Lane 1, acid-depurinated DNA; lanes 2,3, and 4, acid-depurinated DNAs incubated with 0.3,0.7, and 1.5 ng of the HeLa 34-kDa protein, respectively; lane 5, native pUC18 DNA; lanes 6, 7, and 8, native pUC18 DNAs incubated with 0.3,0.7, and 1.5 ng of the HeLa 34-kDa protein, respectively. Abbreviations: OC, nicked, open circular pUC18 DNA; SC, supercoiled pUCl8 DNA.

phoresed at 34 kDa on SDS-PAGE and showed both exonuclease (mononucleotide release) and AP endonuclease activities (unpublished results). To detect enzymes similar to the mouse priming enzyme in some other mammals, 0.2 M KPi extracts were obtained from bovine liver tissue and HeLa cells. The extracts were adsorbed on phosphocellulose in 0.1 M KP, and then eluted with 0.3 M KP, as described under Materials and Methods. The eluant was subjected to SDS-PAGE. Proteins electrophoresed were blotted onto DNA-fixed membranes and acid-depurinated. Priming factors for DNA synthesis were detected at 34 kDa on the acid-depurinated DNA-fixed membranes blotted with proteins from not only mouse ascites sarcoma cells but also bovine liver tissue and HeLa cells (Fig. 8). The results suggested that the 34-kDa proteins of bovine liver tissue and HeLa cells possess AP endonuclease activity similar to that of the mouse 34-kDa protein. To confirm this,

ET

AL.

the band of the 34-kDa protein on SDS-PAGE was excised, and the protein was eluted from the slice and renatured (22). AP endonuclease activity of the renatured protein was measured using acid-depurinated pUC18 DNA. The 34-kDa proteins from these cells showed AP endonuclease activity (Fig. 9). AP endonucleases have been purified from various species, including mouse (2%30), rat (31), bovine (32), and human (33,34). The molecular weights reported differ from each other in the range of 30 to 82 kDa. Further characterizations of the present 34-kDa enzymes are necessary to compare with mammalian AP endonucleases reported previously. (2% 34). So far we could not detect any concrete bands other than the 34-kDa band on the damaged DNA-fixed membranes blotted with proteins from mouse ascites sarcoma cells, bovine liver tissue, and HeLa cells. Further improvements in enzyme preparation, renaturation, blotting, damaged DNA-fixed membranes, and detection procedures may increase the usefulness of the present method. ACKNOWLEDGMENTS The authors thank Mr. T. Nakamura and Ms. T. Yasui for their technical assistance and Nippon Kayaku Co. for providing copperfree bleomycin A,. This investigation was supported in part by a Grant-in-Aid for Scientific Research from the Japan Ministry of Education, Science and Culture.

REFERENCES 1. Lindahl,

T. (1982)

2. Friedberg, York. 3. Mosbaugh, 118.

Annu.

E. C. (1985)

D. W., and Linn,

Repair,

pp. 323-374,

S. (1983)

4. Goffin, C., and Verly, W. G. (1984) 5. Seki, S., and Oda, T. (1979) B&him. 6. Seki, 7. Seki,

51,61-87.

Reu. Biochem. DNA

J. Biol.

B&hem. Biopkys.

S., and Oda, T. (1984) Experientiu S., Ikeda, S., and Oda, T. (1985)

Freeman, Chem.

258,

New 108-

J. 220,133-137. Actu 521,520-528.

40,869-871. Carcinogenesis

6,853-858.

8. Seki, S., and Oda, T. (1986) Carcinogenesis 7, 77-82. S., and Oda, T. (1987) Acta Med. Okuyama 9. Seki, S., Mori, 195-199.

41,

10. Seki, S., and Oda, T. (1988) Curcinogenesis 9,2239-2244. 11. Seki, S., Arakaki, Y., and Oda, T. (1989) Acta Med. Oknyuma 43, 73-80. L., Takeshita, M., Johnson, F., Iden, C., and Grollman, 12. Giloni, A. P. (1981) J. Biol. Chem. 266,8608-8615. 13. Murugesan, N., Xu, C., Ehrenfeld, kie, R. E., Rodriguez, L. O., Chang, Biochemistry 24,5735-5744.

G. M., Sugiyama, L.-H., and Hecht,

H., KilkusS. M. (1985)

14. Sausville, E. A., and Horwitz, S. B. (1979) in Effects of Drugs on the Cell Nucleus (Busch, H., Crooke, S. T., and Daskal, Y., Eds.), pp. 181-205, Academic Press, New York. 20,238-244. 15. Niwa, O., and Moses, R. E. (1981) Biochemistry 16. Hagensee,

M. E., and Moses,

R. E. (1990)

Biochim.

1048,19-23. 17. Seki, S., and Oda, T. (1977) Cancer Res. 37,137-144. 18. Chang, L. M. S. (1973) J. Biol. Chem. 248,3789-3795.

Biophys.

Acta

DETECTION 19. Boyer,

OF

H. W., and Roulland-Dussoix,

REPAIR

D. (1969)

ENZYMES

J. Mol.

Biol.

41,

459-472. 20.

Laemmli,

21. Southern, 22. Hager, 86.

U. K. (1970)

Nature

(London)

E. M.

J. Mol.

Biol.

(1975)

D. A., and Burgess,

23. Seki, S., Ikeda, S., Tsutui, sis 11, 1213-1216.

227,680-685.

98,503-517.

R. R. (1980) K., andTeraoka,

Anal.

Biochem.

H. (1990)

109,76-

Curcinogene-

24. Rosenthal, A. L., and Lacks, S. A. (1977) Anal. Biochem. 90. 25. Spanos, A., Sedgwick, S. G., Yarranton, G. T., Hubscher, Banks, G. (1981) Nucleic Acids Res. 9, 1825-1839.

80,76-

26. Ramotar, D., Auchincloss, Chem. 262, 425-431.

J. Biol.

A. H., and

Fraser,

M.

(1987)

U., and

USING

DNA-FIXED

103

MEMBRANES

27. Rogers, S. G., and Weiss, B. (1980) in Methods in Enzymology (Grossman, L., and Moldave, K., Eds.), Vol. 65, pp. 201-211, Academic Press, New York. 28. Ludwig, G., and Thielmann, H. W. (1979) Nucleic 2901-2917. 29. Nes, I. F. (1980) Nucleic Acids Res. 8, 1575-1589.

Acids

Res. 6,

30. Tomkinson, A. E., Bonk, R. T., and Linn, S. (1988) 263, 12,532-12,537. 31. Cesar, R., andverly, W. G. (1983) Eur. J. Biochem.

J. Biol.

Chem.

129,509-517.

32. Henner, W. D., Kiker, N. P., Jorgensen, T. J., and Nunck, J.-N. (1987) Nucleic Acids Res. 15, 5529-5544. 33. Kane, C. M., and Linn, S. (1981) J. Biol. Chem. 256,3405-3414. 34. Shaper, N. L., Grafstrom, Chem. 257,13,455-13,458.

R. H., and Grossman,

L. (1982)

J. Biol.