Chem.-Rio/. Interactions, 46 (1983) 109420 Elscvicr ScienLlfie Publishers Ireland Ltd.

109

NUCLEAR DNA ENDONUCLEASE ACTIVITIES ON PARTIALLY APURINICIAPYRIMIDINIC DNA IN NORMAL HUMAN AND XERODERMA PIGMENTOSUM LYMPHOBLASTOID AND MOUSE MELANOMA CELLS

MURIEL W. LAMBERT, W. CLARK LAMBERT and ANTHONY 0. OKOKODUDU

Department nf Pathalogy# UMDNJ- New Jersey Medical LSchool, 100 Bergen Street. Neulark. ,VJ 0710.1 (U.S.A.)

(Received November 18th. 1982) (Revision received January 24th. 1983) (Accepted January 28th. 1933)

SUMMARY

DNA endonuclease activities from nuclear proteins of normal human and xeroderma pigmentosum (XP), complementation group A, lymphoblastoid and Cloudman mouse melanoma cells were examined against partial11 apurinic/apyrimidinic (AP) DNA. Non-hi&one chromatin-associated and nucleoplasmic proteins, obtained from isolated nuclei. were subfractionated by isoelectric focusing and assayed for DNA endonuclease activity against linear, calf thymus DNA. All of the nine chromatin-associated and three of the nucleoplasmic fractions, which lacked DNA exonuclease activity, were tested for DNA endonuclease activity against both native and partially .r\P. circular, duplex, supercoiled PM2 DNA. In all three cell lines, four chroma- tin-associated, but none of the nucleoplasmic fractions, showed increased activity against DNA rendered AP by either heat/acid treatment or by alkylation with methyl methanesulfonate (MMS) followed by heat. One chromatin-associated activity, with p1 9.8, which was not active on native DNA, showed the greatest activity on AP DNA. AP activity was moderately decreased in XP cells and slightly decreased in mouse melanoma cells. as compared with normal cells, in the fraction at p1 9.8. Little or no increased activity was observed in any of the endonucleases from any of the cell lines on MMS alkylated DNA. -

Key words: Apurinic endonuclense - Xeroderma pigmentosum cells - Non-

histone proteins

-- ---

Abbreviations: AP. apurinic/apyrimidinic; MMS. mrthyl methanesulfonote: XI’. ?krodrrma pigmentoaum.

00042797/83/$03.00 $3 1983 Elsevier Scientific Publishers Ireland Ltd. Print4 and Published in Ireland

110

INTRODUCTION

DNA in cells may undergo AP damage by a number of different mechanisms. Base release can occur either spontaneously or by chemical or enzymatic reactions that occur after the bases are altered by specific alkylat- ing agents or by y-irradiation 11, 2; reviewed in Refs. 3, 41. AP damage is repaired most commonly by an excision repair pathway which is initiated by an endonuclease specific for AP sites that causes a single-strand break in the DNA in the vicinity of the AP site (reviewed in Refs. 3-6). Endonucleases which specifically recognize AP sites have been purified from a variety of sources [4,7,-141 and total repair of AP sites has been demonstrated with human enzymes 1151.

Cell lines derived from patients with XP, an autosomal recessive disease characterized by extreme sun sensitivity, have been reported to be able to repair AI? lesions in vivo and this repair has been observed as unscheduled DNA synthesis or DNA repair synthesis 116201. There are conflicting reports, however, concerning AP endonuclease activity in these cells. The work of Kuhnlein et al. 1211 and of Witte and Thielmann I221 indicates that there is a decrease in AP endonuclease activity in whole cell extracts from )c,P fibroblasts, complementation groups A (XPA), D (XPD), or both. On the other hand, Moses and Beaudet 1231 found normal levels of AP endonuclease activity in extracts of XPD fibroblasts.

We have recently reported eight clearly distinguishable chromatin-asso- ciated DNA endonuclease activities, containing no exonuclease activity, in the nuclei of normal human and XPA lymphoblastoid cells and of mouse melanoma cells, which are active on both native linear duplex calf thymus DNA and native circular, supercoiled, duplex PM2 bacteriophage DNA [24- 261. One of these chromatin-associated endonuclease activities, with p1 4.6, from normal human lymphoblastoid cells, but not from XP or mouse melanoma cells, was exceedingly more active on anthramycir treated PM2 DNA than on native DNA 1271. We now report that four other chromatin= associated DNA endonuclease activities, with I&values 8.2-9.8, show in= creased activity on AP DNA, with maximal activity in a fraction with pI 9.8, which is not active against either type of native DNA. AP activity was moderately decreased in XP cells and slightly decreased in mouse melanoma cells, as compared with normal cells, in the fraction at ~19.8.

?dATERlALS AND METHODS

Cell lines and culture conditions

Normal human (GM 1989) and xeroderma pigmentosum, complementation group A (GM 2345) lymphoblastoid cell lines (transformed with Epstein-Barr virus) were obtained from the Institute for Medical Research, Camden, New Jersey. The cells were grown in suspension culture in RPM1 1640 medium (Grand Island Biological Co.) as previously described I251 and harvested under conditions of maximal proliferation. Cultures were routinely tested for

111

mycoplasma and steps were taken to insure that the cells were not exposed to UV light and that other light exposure was minimal 1251.

Cloudman mouse melanoma cells (S91 NCTC 3960, CCL 53) were pro- pagated by serial subcutaneous injection into male &week-old DBA/W mice and were harvested after 3 weeks growth as previously described 124,261.

Enzyme extraction Cell nuclei were isolated as previously described [24,25]. Chromatin-

associated and nucleoplasmic proteins were extracted from the purified nuclei and electrophoresed on an isoelectric focusing column with a 5-509 sucrose gradient and ampholines, pH 3.0-10.0 1241. Fractions collected from the columns were assayed for DNA endonuclease and exonuclease activity against calf thymus DNA [281 and [Hlpoly(dA-dT) 1291 respectively. Peaks of endonuclease activity were pooled, dialyzed into 50 mM potassium phosphate (pH 7.1), 1 mM dithiothreitol, 1 mM NaPEDTA, 40% ethylene glycol, and stored unfrozen at -20°C.

DNase assay methods DNA endonuclease activity was measured using a DNA polymerase primer

assay and linear calf thymus DNA (Worthington Biochemical Corp) as substrate 1281. DNA endonuclease activity was also assayed using circular, duplex, supercoiled PM2 bacteriophage DNA as substrate. The enzymatically treated DNA was electrophoresed on 0.9% agarose gels which were sub- sequently stained with ethidium bromide, photographed under ultraviolet light and the negatives scanned using a Zeineh Soft Lasar Scanning Den- sitometer (LKB) 124, 271. Endonuclease activity, expressed as the number of enzyme-induced breaks per DNA molecule, was detected by the conversion of superhelical DNA (Form I) to nicked, relaxed circular DNA (Form II) and linear unit length DNA (Form III). Quantitation of the amount of cut versus superhelical DNA was made from integral analysis of densitomerric tracings of negatives of photographs of the gels, and the number of brea?s per DNA molecule was established by the Poisson formula, number of breaks = -In x. where x -_ fraction of superhelical molecules [30,311.

Alkylation and depurinationldepyrimidinntion of DNA PM2 DNA was rendered partially apurinic/apyrimidinic (AF’r by each of

two different methods. Partially AP DNA was prepared according to the method of Lindahl and Nyberg [32] by heating at 70°C for 20 min in 0.1 M NaCl, 0.01 M sodium citrate, 0.01 M Tris-HCl (pH 5.2). The number of AP sites per DNA molecule was determined by hydrolyzing the AP DNA for 4 h at 25°C in 1.0 M glycine-NaOH (pH 12.8) and analyzing the DNA on agarose gels (Table I). Under these conditions, all AP sites are hydrolyzed so that each break represents one hydrolyzed AP site 1321.

PM2 DNA was also alkylated with the monofunctional alkylating agent. MMS (12.5mM) (Aldrich Chemical Co.), at 37°C for 20 min and then heat depurinated at 70% for 5 min. Under these conditions, the heat-labile alkyl-

112

ated purineslpyrimidines are released to gi-e AP sites [31,321. MMS was removed by dialysis. The number of AP sites per molecule was determined by hydrolyzing the DNA in l.OM glycine_NaOH (pH 12.8) as described above. The number of AP sites per DNA molecule was then computed. Correction was made for the nicking that occurred during the preparation of AP DNA and exposure of DNA to hydrolytic conditions as outlined in Table II. The number of MMS alkylated sites per DNA molecule and the number of AP sites per DNA molecule induced by MMS treatment, alone, were computed as described in Table II.

TABLE I

DEPIJRINATION/DEPYRIMIDINATION OF PM2 DNA BY HEATING AT pH 5.2

Total number of Total number of Heating at 70°C Heating at 70°C Hydrolysisb breaks per PM2 AP sites per (pH ~2)~ (PH 7.5) DNA molecule PM2 DNA molecule

_ _ (a) 0.15 . . t (b) 0.15 0.0 (b-a) t _ (c) 0.16

+ (d) 0.37 0.21 (d-c) (e) 0.92

_ + (0 3.59 2.67 (f - e)

’ PM2 DNA was heated for 20 min in 0.1 M NaCl, 0.01 M sodium citrate, 0.01 M Tris-HCl (pH 5.2). b DNA was hydrolyzed for 4 h at 25°C in 1.0 M glycine_NaOH (pH 12.8).

TABLE II

ALKYLATION AND DEPURINATION/DEPYRIMIDINATION OF PM2 DNA BY MMS (125mM)

Total number of Total number of Treatment Heating breaks per PM2 AP sites per with MMS” at 70”Cb Hydrolysis’ DNA molecule PM2 DNA molecule

- (a) 0.14 _ (b) 0.28 _ (c) 0.16 + (d) 0.14 0.0 (d- a) _ (e) 0.35 + (f) 0.22 0.06 (f-c) + (g) 0.65 0.37 (g - b) + (h) 2.72 2.37 (II e)d -

a PM2 DNA was alkylated with 12.5 mM MMS at 37°C for 20 min. b DNA was heated at 70°C for 5 min. ’ Hydrolysis was for 4 h at 25°C in 1 M glycine-NaOH (pH 12.8). d Of these 2.37 AP sites per PM2 DNA molecule, 0.37 (g - b) were produced by MMS treatment

alone; 0.06 (f*) by heating alone. The number of MMS alkylated sites per PM2 DNA molecule was 9.9401-e-(g-b).-(f-c)].

113

Enzymatic actioity on damaged DNA The activity of each of the chromatin-associated and nucleoplasmic DNA

endonuclease fractions was tested on PM2 DNA which had been alkylated or rendered partially AP as described above. Each endonuclease was reacted with native and damaged DNA in 5mM MgC&, 2OrnM KC1 and 10mM Tris-maleate, pH 7.5. The concentration and incubation time of each enzyme activity was adjusted to produce ,-65% form I DNA or ~0.28 breaks per DNA molecule on untreated DNA. The reaction was terminated with 0.1 M EDTA and the sample prepared and electrophoresed as previously described [24,27].

All experiments were repeated 6-8 times using enzymes obtained from 2-3 different cell extractions.

RESULTS

Depurinationldepyrimidination of PM2 DNA by heating at 70°C for 20 min

at pH 5.2 produced 2.67 AP sites per DNA molecule (Table I). When each nuclear DNA endonucleaee fraction, from all three cell lines, was reacted with AP DNA, increased activity, producing single-strand breaks, was found to be associated only with the chromatin proteins. Maximal activity was found in the fraction with p1 9.8 (Fig. l), which produced 2.08-2.96 single- strand breaks per DNA molecule and which had little to no activity against either type of native DNA [25,26]. Progressively lesser AP activities were found in the fractions with pIs 9.2, 8.8 and 8.2, respectively (0.84-1.89 single-strand breaks per DNA molecule) with some very small activities in the other fractions (Fig. 1). AP activity was only slightly decreased in XP and mouse melanoma cells, except for the fraction at p1 9.8, where XP activity was 70% and mouse melanoma activity was 88% of normal human cell activity.

Alkylation of PM2 DNA with 12.5 mM MMS followed by heat depurination produced 2.37 AP sites per DNA molecule (Table II). Each endonuclease fraction was incubated with the partially AP DNA, and, as was found for the DNA depurinated by the first method, those chromatin fractions with basic pIs were the most active, producing single-strand breaks, with the fraction at p1 9.8 again having maximal activity (producing 1.93-2.71 breaks per DNA molecule) (Fig. 2). Activity was again decreased 11, XP and mouse melanoma cells, particularly in the fraction at p1 9.8, where actl<ity was 71% and 89%, respectively, of that of the normal human cell AP activitv.

These endonuclease fractions were also tested on PM2 DNA which had been alkylated with MMS, but not heat depurinated, to produce 1.94 alkyl- ated bases and 0.37 AP sites per DNA molecule (Table III). Little increased activity over that of control values was observed on the alkylated DNA by any of the fractions (~0.37 breaks/DNA molecule) (Table III) and this may represent activity on the AP sites genera’ied in the alkylation procedure (0.37 AP sites) (Table II). The enzymatically treated MMS alkylated DNA was then hydrolyzed in order to determiile whether any of the enzyme fractions had produced AP sites, possibly through the action of a glycosylase. In all

N kM

P?!O: Bfl’i)? Q?to:

pI OF ENDONUCLEASE FRACTIONS

PIE. 1. Action of ohromatin-associated DNA endonuclease fractions on heat/acid depurinatedl dcpynm&nated PM2 DNA. Vertical lines represent d3.E.M.

I:

k I ‘.. 7” I :’ (1 x M *” x pi I r,4 ‘* 1 H I< 1 v b, rv 1 1 ”

ii::: 4 6, (. 5 2 :srrJ2 f,f,l’.Z : f, - i, : y ,P ‘,(J ,i $4 ’ 1 ’ a.< *

p1 OF ENDONUCLEASE FRACTIONS

Fig. 2. Action of chromatin-associated DNA endonuclease fractions on MMS (12.5 mM) alkyl. ated and heat depurinated PM2 DNA. Vertical lines represent 93.E.M.

115

TABLE 111

LJNA ENDWWCLEASE ACTIVITIES ON MMS TREATED Hvl2 DNA OF NOHMAL HL!MAN INL). XP AND MOUSE MELANOMA (MM) CWKOMATIN-ASSOCIAT~;~~ PROTF:INS

Numbtbr of enzyme-induced sctsslons p’r PM2 DNA moh~culc (mcsn * S.E.)”

pl f + o.:I) NL” Xr(l MM’ -. ._

3.9 0 * 0 0.01 * 0 01 0 * 0 4.6 0.01 ’ 0.01 0 * 0 0 + 11 5.4 0.01 + 0.01 0.03 ’ 0.02 0 * 0 6.6 0.04 * 0.02 o.o:! 1 0.0 1 0 ’ 0 7.6 0.04 ’ 0.02 0.05 ’ 0.02 0.02 + 0.0 I 8.2 0.12 * 0.03 0.12 * 0.03 0.14 ’ 0.04 W.H 0. IN ’ 0.04 0.20 * 0.03 0.28 + o.n5 9.2 Il.27 ’ O.OH 0.24 * 0.03 0.22 * 0.04 9.n 0.37 + 0.06 0.24 + 0.03 0.26 * 0.05

II This value has had subtracted from it the number of breaks the enzyme produced on norn~l DNA and the number of breaks in non-enzyme treated DNA.

h 11 pg of fractions (PI 3.9-5.4) or 21 pg of fractions (pl 6.6-9.H) were incubattbd w th 0.18 +i:

12.5 n&l MMS treated DNA at 37°C for 30 min. r 18 pg of each fraction was incrrbated with 0.18 pg 12.5 mM MMS treated DNA at 37 (‘ for

30 min (PI 3.9-5.4) or 60 min (pl 7.6-9.8).

three cell lines, only chromatin fractions with pIs 8.2-9.8. produced a vrr~ small number of AP sites (0.2 .46/DNA molecule) above con trcl values (Table IV). Thus, e results indicate that neither a DNA endonuclease wit,h specificity for S alkylated sites nor DNA glycosylase activity specific

A were present in any fraction tested under these assay conditions.

NA endonuclease fractions, from all three cell lines;. had excess of control values on either type of AP DNA

ks/DNA molecule) or on MMS alkylated DNA (.- 0.133

XP fibroblasts in culture appear to be as proficient as normal human fibroblnsts in repair of A [l6--2XIl. Several types of mouse cells have also been sh mage [IO. 11,331 The question of

cont;Gn a defective 01

decreased AP e

ucleast.. fractions from XP ~~,rnpb~bl~~~to~~

116

EFFECT OF HYDROLYSIS ON ENZYMATICALLY-TREATED DNA

MMS ALKYLATED PM2

-- Number of enzyme-induced scissions per PM2 DNA molecule (mean 2 SE.)”

______ -

pl (- 0.3) NLh :trpb MMh --

x9 0-O 0 .? 0 o-to 4.6 Or0 oto o-e0 r,. 4 o*o o-0 O&O fi.6 0 -0 O?O o-+0 76 0.07 ‘f 0.03 0.04 t 0.02 0.05 t 0.01 ti.2 0.26 + 0.09 0.28 2 0.08 0.3350.11 r(H 0.27 r 0.08 0.38 : 0.08 0.39-to.12 9’L 0.32 t 0.01 0.41 + 0.12 0.44? 0.09 9 H 0.46r0.11 0.39-r 0.08 cr.40 lr 0.05

“this value has had subtracted from it the number of breaks the enzyme produced on MMS alkylated DNA and the number of breaks in nonenzyme treated Mh4S alkylated and hydrolyzed DNA.

r, PM:! DNA (0.18 ug) was alkylated witb 12.5mM MMS for 20 min at 37°C. then reacted with normal human (NL), XP or mouse melanoma (MM) chromatin-associated endonucleases, as described in Table III, and hydroiyzed for 4 h at 25°C.

and mouse melanoma cells on DNA which has been rendered partially AP by two diRerent methods, and have compared them with corresponding activities from normal human lymphoblastoid cells. We found, in all three cell lines, AP endonuclease activities on PM2 DNA rendered partially AP by either hea! ‘,qg at pH 5.2 or by depurination of MMS alkylated DNA. Activity was founl in the same basic chromatin-associated fractions in all three lines, with a moderate decrease in activity in XP and a slight decrease in activity in mouse melanoma cells, in particular in the fraction at ~19.8 which was not active on native DNA, but which had the highest AP activity. The decreased activity in XI? cells is in agreement with the work of Kuhnlein et al. I211 which showed that whole cell extracts from XPA and XPD fibroblasts had 60% and 17% of normal activity, respectively, on heat/acid depurinated phage DNA. It is also in agreement with the results of Witte and Thielmann I221 in which activity from whole cell extracts of XPA fibroblasts was 40% of normal activity on MMS treated phage DNA. On the other hand, Witte and Thielman I221 and Moses and Beaudet [23] found normal levels of AP endonuclease activity in XP.A and XPD fibroblasts, respectively, against heat/acid depurinated phage DNA. Since the major difference in activity was found, in the present work, in the fraction at ~19.8, it is possible that various whole cell extracts used in these other studies may or may not have retained this activity or that it may have been masked or altered. Further purification

117

TABLE V

DNA ENDONUCLEASE ACTIVITIES ON AP OR MMS ALKYLATED PM2 DNA OF NORMAL HUMAN (NL), XP AND MOUSE MELANOMA (MM) NUCLEOPLASMIC PROTEINS -lp--l_--

Number of enzymesinc, iced scissions per PM2 DNA molecule (mean t SE.)”

Trentment” pl ( l :0.2) NL’ XI’ MM” ~__I -. ~_ Heat/acid AP DNA 3.7 O?O ot_o oco

4.6 0.26 t 0.09 0.17 + 0.06 0.19 Z 0.09 5.6 0.30 ? 0.14 0.27 ? 0.12 0.28 -’ 0.11

MMS AP DNA 3.7 o-to 020 o-_o 4.6 0.09 -c 0.02 oto 0.01 Z 0 5.6 0.12 1’: 0.04 0.03 t 0.01 0.04 f 0.02

MMS alkylated DNA 3.7 0 .? 0 0.01 +o Or0 4.6 0.01 f 0 0.02 -c 0.01 0.01 f 0.01 5.6 0.07 I? 0.04 0.01 rJ 0.03 2 0.01

a This value has had subtracted from it the number of breaks the enzyme produced on normal DNA and the number of breaks in non-enzyme treated DNA.

b PM2 DNA was depurinatedldepyrimidinated (AI’ DNA) or alkylated as in Tables I and II. c LO pg of fractions (PI 3.7 and 5.6) or 2 kg of fraction (~14.6) were incubated with 0.18~pg treated

DNA for 30 min at 37°C. d 10 +g of each fraction was incubated with 0.18qg treated DN.1 for 45 min at 37-C.

of these fractions is needed in order to more clearly define the differences in

AP activity observed in XPA and mouse melanoma cells as compared with normal human cells.

The major AP endonuclease activity was found in four chroma*lin.asso- ciated fractions (which produced approx. l-3 breaks per AP DNA molecule under these assay conditions). Whether these are the same or different enzyme activities is unclear at present. Since these fractions were separated by isoelectric focusing, it is possible that some of them are modified versions of the same proteins. However, muliiple forms of AP endonuclease activity have been found in human placenta [34] and two species have been identified in normal human and XPA fibroblasts [35]. XPD fibroblasts have been shown to lack one of these two AP endonuclease activities 1351. The enzymes studied in the present work cut AP DNA to give single-strand scissions and may be similar to some of the AP endonucleases which have been isolated and purified

from various human and rodent cell sources [7-141. Further purification and characterization studies will be needed to determine this.

XP cell lines have been shown to be heterogeneous in their ability to repair certain types of alkylnted DNA [5, 16-201 and it has been further shown that transformation may induce a repair defect in these cell lines, in particular in repair of Os-alkylguanine residues [36-381. Although it is possible that the differences we observed between XP and normal lymphoblastoid cells are due

to transformation, we do not feel that this is likely, since MMS (unlike some other monofunctional alkylating agents, e.g., methyl nitrosourea) produces very few 06-methylguanine residues (0.1-0.3%) 139,401 and, in addition, repair of Q”-alkyl groups is thought not to involve an endonuclease but to proceed via other types of mechanisms [4,41-441. Also the differences between cell lines we observe using MMS alkylatedldepurinated DNA and DNA depurinated by heat/acid were very nearly identical.

Repair of alkylation damage to DNA in a number of different cell types has been shown to involve the action of a DNA glycosylase (reviewed in Refs. 3 and 4) as well as that of an AP endonuclease. Our results indirectly indicate that there is little DNA glycosylase activity in any of the nuclear fractions under the conditions used here. The studies comparing numbers of Pndonuclease-mediated strand breaks with the total number of alkylated sites on MMS treated DNA showed that only a few scissions were introduced by any fraction, and these may be explained on the basis of AP endonuclease activity against the few AP sites created by the MMS treatment, itself. The number of enzyme-induced breaks per AP DNA molecule is similar to the number of AP sites present, not appreciably higher as would be expected if a DNA glycosylase was present. This indicates that little or no glycosylase activity coupled with AP endonuclease was present in any fraction. Fur- thermore, subsequent hydrolysis of this MMS treated DNA following enzy- matic incubation failed to show a significant increase in numbers of AP sites for any endonuclease fraction. This suggests that little or no endonuclease- independent glycosylase activity was present m any fraction. Further purification steps and possibly different reaction conditions, however, will be necessary to conclusively determine whether glycosylase activity is asso- ciated with any of these fractions.

ACKNOWLEDGEMENTS

We would like to thank Douglas Fenkart for his technical assistance in this work. This research was supported in part by Basil O’Connor Grant #5-287 from the March of Dimes Birth Defects Foundation and the Foun- dation of UMDNJ.

REFERENCES

1 P.D. Lawley and P. Brookes, Further studies on the alkylation of nucleic acids and their constituent nucleotides, Biochem. J., 89 (1963) 127.

2 B. Strauss and T. Hill, An intermediate in the degradation of DNA alkylated with a monofunctional alkylating agent, B&him. Biophys. Acta, 213 (19701 14.

3 J. LavaI and F. LavaI, Enzymology of DNA repair, in: Montesano, Bartsch and Tomatis fEds.1, Molecular and Cellular Aspects of Carcinogen Screening Tests, IABC Scientific Publishers, Lyon, France, 1980. pp. 55-73.

4 T. Lindahl, DNA repair enzymes, Annu. Rev. Biochem.. 51 (1982) 61. 5 AR Lehmann and P. Ear-ran, DNA repair, Int. Rev. Cytol., 72 (1981) 101.

119

6 E.C. F&dherg. T.M. Carrie, ‘I’. Anderson. ‘I’. Bonura. R. Cone. E.H. Radany and R.J. ficlyno)d?i. k%cllt ~l~~f~)fJp~n~n~~ in the eneymoiogy of exeislon repatr of DNA, Prog. Nut). Acid, I&i. Mol. Bin]., zfi (t%l) 197.

7 W.G. Vcrly and V. Paquattc, An endonuclcase for depurinatcd DNA in rat liver. Can. J. Biochem.. 61 (1973) 1003.

8 S. Ljungquist and ‘1’. Lindahl, A mammalran encionuciea.se specific for aput-inic sites in (~l~ubl~~-~tra~ld~~ d~?oxyrihonu~i~!I~ acid. J. Biol. Chem.. 249 f 1974) Ifi30.

9 ‘l’.P. Brent, Purificwticm and characterLzatton of human endonucleases syzcific for damaged DNA, Biochim. Biqhys. Actn, 454 (1976) 172.

10 G. Ludwig and H.W. Ttielmann, Apunmc acid endonuclease activity in mouse epiderma) cells, Nucleic Acids RPS., 6 t 1979) 290 I

11 I.F. Nos. ~~~~ati~n and ch~ra~~eri~atlon ot an endonuciease specific for apurinir sites in DNA from R pcrm:montly estuhhshed mouse pinsmacytoma cell line, Nudeic Acids Res., (3 (1980) 1575.

12 N.L. Shaper and L. (Grossman, Purificuclon and properties of the human placenta) ~~purinic/~ipyrimidinic cndonuciease, Methods Enzyme)., 65 f19800f 216.

13 L. Thihodeau, S. Briet!*ux and W.G. Verly, Purification and properties of the major apuriniolapyrimidinic endodcoxyrlbonuciease ot rat-hver chromatin, Eur. J. Biochem.. 110 (1980) 379.

14 C.M. Kane and S. Linn, Puriiication and characterization of an apurinic~ap~rimidi~ic endonuclease from fieLa rells, J. Biol. Chem., 256 (1981) 3405.

15 K. Bose, P. Karran and B. Strauss. Repair of depurinated DNA in vitro by enzymes purified from human lymphoblasrs, Proc. lJat1. Acad. Sci. U.S.A., 75 (1978) 794.

16 J-E. Cleaver. Repair of alkylatton damage m uitravloiet-sensltlve (xeroderma pigmentosum) human tells, Mutat. Res., 12 (1971) 453.

17 H.F. Stick, R.H.C. Sau and Y. Kanajoe, increased sensiclvity of xeroderma pigmentosun cells to some chemical carcinogens and muragrns, Mutat. Res.. 17 (1973) 127.

18 R.D. Snyder and J.D. Regan, DNA repair m normal human and xeroderma plgmentosum group A fibroblasts followrng treatment with various m~~~hanesui~anote~ and the demon- stration of a long-patch {U.V.-like) repaIr component, Carcinogenesis. 3 (1982) 7.

19 E.C. Ffitdberg, U.K. Ehmann arid J.1. Williams. Human diseases associated with defertlve DNA repair, Adv. Kadiat. Biol., H (1979) 85.

20 K.H. Kraemer, Xeroderma plgmenrosum. in: D.J. Drmir. R.L. Dobsm ax!d J. Mclh~re tEds.). Clinical Dermatology, Vol. 4, Unit 19-7. Harper and Row, New York, 1980. pp. l-33.

21 U. Kuhnlein. E.E. Penboet and S. Linn, An altered apurinic DNA endonuciease activity in k~oup A and group D xeroderma plgmcntosum fibroblasts, Proc. Nat). Acad. Sci. C.S.A., 73 t 1976) 1169.

22 1, Witty and H.W. Thielmann, Extracts of xerodcrma pigmentosum group .A tibroblasts introduce less nicks into methyl methan~su)fL~na~e-errata DNA than extracts of normal fibrohiusts. Cancer Lett.. 6 (1979) 129.

23 R.E. Moses and A.L. Braudet , Apurinic DNA endonuciraat~ activities in repau-deticirnt human cell lines. Nucleic Attda Res.. 5 (1978) 463.

24 M.W. l,amb~,r~ antj G.P. Stu~lnski, DIG.4 endonu~le~~se actlvlties associated with melanoma tell chromatin, Biochem. B~ophys. Res. Commun.. 91 (1979) 1481.

2fi A.(.). Dkorodudu, W.C. Lamberr aud M.W. Lamberr, Nuclear deo.xyrlbonucllaasr activltles in normal and xt~r&rma ptgmt$nlosum tymphohlastoid ct~lla. Biochrm. Biophy,:. Res. Commun.. IOH (i!w) 576.

27 1j.E. &II, ~0. Okorwbh, WA!. Lambert and M.W. Lambert, Defective DNA rndonuclease ectivily on ~inthramy~In trtxat& DNA in xer~erma p~~en~osum and mouse meianoma cells, Biochem. Biophys. Rrs. Commu.1.. 107 (19*lf 395.

28 M.W. Lamberr and G.P. Ytudxinski. . Id&(& for assessment of DNase activity. in: G. Stein and J. Stein (I&.), M&ds in Cell Eiolob!, Academic F’rrss. New York. 19% pp. 4:H9.

s T. L~odahl, J.A. Gnliy and GM. Edelman, ~xy~~nuclea~. Dr. A new exonrnclease from mammalian tissues. Proc. Natl. Acad. Sci. U.S.A., 62 (1969) 697.

2) T. Lindahl and A. Anderson, Rate of chain breakage at apurinic sites in double-stranded deoxyribonucleic acid, Biochemistry, 11 (1972) 3618.

31 N.J. Duker and G.W. Teebor, Detection of different types of damage in alkylated DNA by means of human corrective endonuclease (~~ndonuclease), Proc. N&l, Acad. Sci. USA., 73

t 1976) 2629. 32 T. Lindahl and B. Nyberg, Rate of depurination of native deoxyribonucleic acid, Biochemis.

try. 11 (1972) 3610. 33 P. Karran, The repair of methyl methanesulfona~-induce lesions in the DNA of a murine

lymphoma cell, Chem.-Biol. Interact., 7 (1973) 389. 34 W.S. Linsley, E.E. Penhoet and S. Linn, Human endonuclease specific for apurinic/apy&

midinic sites in DNA, J. Biol. Chem., 252 (1977) 1235. 35 U. Kuhnlein, B. Ltw, E.E. Penhoet and S. Linn, Xeroderma pigmentosum fibroblasts of the D

group lack an apurinic endonuclea~ species with a low apparent K,, Nucleic Acids Res., 5 (1978) 951.

36 J.A. Heddle and C.F. Arlett, Untransformed xeroderma pigmentosum cells n:ti not hyper. sensitive to sister-chromatid exchange production by ethyl methanesulfonate -implications for the use of tran5fo~~ cell lines and for the mechanism by which SCE arise, Mutat. Res., 72 (1980) 119.

37 R.S. Day III, C.H.J. Ziolkowski, D.A. Scudiero, S.A. Meyer, A.S. Lubiniecki. A.J. Girardi, SM. Galloway and CD. Bynum, Defective repair of alkylated DNA by human tumor and SV40-transformed cell strains, Nature, 288 (1980) 724.

38 1R. Sklar and B. Strauss, Removal of G6-methy~~ine from DNA of normal and xeroderma pigmentosumderived 1ymphobIastoid lines, Nature, 289 (1981) 417.

39 ‘P.D. Lawley, D.J. Orr and M. Jarman, Isolation and identification of products from alkyh ation of nucleic acids: ethyl- and isopropyl-purines, Biochem. J., 145 (1975) 73.

40 D-T. Beranek, C.C. Weis and D.H. Swenson, A comprehensive quantitative analysis of methylated and ethylated DNA using high pressure liquid chromatography, Careinogenesis, 1 (1980) 595.

41 J. Bogden, A. Eastman and E. Bresnick, A system in mouse liver for the repair of @-methylguanine lesions in methylated DNA, Nucleic Acids Rea., 9 (1981) 3089.

42 M. Lemaitre, A. Renard and W.G. Verly, A common chromatin fact-. involved in the repair of 06-methylguanine and C%thylguanine lesions in DNA, FEBS Lett., 144 (1982) 242.

43 E.A. Waldstein, E.-H. Cao and R.B. Setlow, Adaptive resynthesis of 06.methylguanine. accepting protein can explain the differences between mammalian cells proficient and deficient in methyl excision repair, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 5117.

44 A.E. Pegg, M. Roberfroid. C. von Bahr, R.S. Foote, S. Mitra, H. B esil, A. Likhachev and R. Montesano, Removal of CP-methylguanine from DNA by human liver fractions, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 5162.