Ž .Mutation Research 401 1998 27–32

Caffeine does not potentiate g-radiation induced DNA damage inataxia telangiectasia lymphoblastoid cells

D. Gwyn Bebb a,b, Patricia P. Steele a, Pamela J. Warrington a, Joyce A. Moffat a,Barry W. Glickman a,)

a Centre for EnÕironmental Health, Department of Biology, UniÕersity of Victoria, P.O. Box 3020, Victoria, BC, Canada V8W 3N5b Department of AdÕanced Therapeutics, British Columbia Cancer Research Centre, 601 W. 10th AÕenue, VancouÕer, B.C. V5Z 1L3, and

Department of Pathology and Laboratory Medicine, UniÕersity of British Columbia, 2211 Westbrook Mall, VancouÕer, B.C. V6T 2B5

Received 5 May 1997; revised 26 August 1997; accepted 5 September 1997

Abstract

Ž .Ataxia telangiectasia AT cells display a profound sensitivity to ionizing radiation, exhibiting more frequent chromoso-mal breaks, increased micronuclei formation and abnormal DNA repair kinetics following exposure. Despite the recentcloning of the ATM gene there remains a need for a simple and rapid means of discriminating AT heterozygotes from

Ž .normal individuals. Caffeine 1,3,7-trimethyl xanthine , known to inhibit the repair of double-strand DNA breaks followingionizing radiation, increases the frequency of radiation induced chromosomal breaks in normal cells. Here we report thatcaffeine potentiates the induction of chromosomal breaks in G arrested AT heterozygote and normal lymphoblastoid cells,2

but not in homozygous AT lymphoblastoid cells. This observation parallels the findings reported by others that caffeine failsto potentiate the effect of ionizing radiation in radiation-sensitive yeast strains and radiation sensitive CHO cells. It alsosuggests that caffeine may somehow mimic the effect of the ATM gene product in normal cells. We also report that caffeineis unlikely to be useful in helping to discriminate AT heterozygotes from normal individuals. q 1998 Elsevier Science B.V.All rights reserved.

Keywords: Caffeine; Ionizing radiation; Chromatid breaks; Ataxia telangiectasia

1. Introduction

Ž .Caffeine 1,3,7-trimethyl xanthine is a chemicalwell known for its DNA repair inhibiting activityw x1,2 . Caffeine has been reported to enhance thelethality of numerous cytotoxic agents, including

w xchemical 3 as well as ionizing and non-ionizing

) Ž .Corresponding author. Tel.: q1 250 472-4067; Fax: q1Ž .250 472-4075; E-mail: bwglick@uvic.ca

w xradiation 4,5 . Treatment with caffeine has also beenassociated with loss of the G arrest that normally2

w xfollows exposure to specific genomic insults 6,7suggesting that caffeine somehow inhibits cellularresponses to DNA damage.

Ž .Ataxia telangiectasia AT is an autosomal reces-sive, heritable human condition that is characterized

w xby a profound sensitivity to ionizing radiation 8 .When AT patients receive what is usually a thera-peutic dose of radiation, severe, sometimes fataltissue necrosis occurs. In vitro, cells from AT pa-

0027-5107r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.Ž .PII S0027-5107 97 00214-5

( )D.G. Bebb et al.rMutation Research 401 1998 27–3228

tients also respond abnormally to radiation display-ing more frequent micronuclei, increased chromoso-

w xmal aberrations and excess cell death 9,10 . Theunderlying defect involves an inability to respondappropriately to ionizing radiation by p53 inductionw x w x11 or apoptosis initiation 12 , and is characterized

w xby radio-resistant DNA synthesis 13 . It has alsobeen suggested that an ill-defined defect in chro-matin structure and organization increases the ten-dency for the conversion of double-strand DNA

w xbreaks into chromosomal breaks 14 . Recently, theŽ .gene for AT ATM , located to chromosome 11q

22–23, was cloned and sequenced and shown toshare homology with the PI-3 kinase and the yeast

w xradiation induced protein, rad-3 15 .The similarity of caffeine-induced effects and the

w xAT cellular phenotype 16,17 has led several re-searchers to investigate whether caffeine potentiatesthe effects of ionizing radiation in AT and ATheterozygote cells or finds its effect masked by apre-existing defect. In addition, the reported associa-tion between AT heterozygosity and breast cancerw x18–20 has led to attempts to develop a convenientand rapid means of discriminating AT heterozygotesfrom normal individuals to confirm this association.There are, to our knowledge, no reports of the effects

Ž .of caffeine on AT heterozygote AT" cells afterradiation and we wondered whether caffeine wouldhelp form the basis of a simple and inexpensivediscriminating test. Here we report on the results ofour investigation of the effect of caffeine on AT

Ž .presumed normal ATqrq , AT heterozygousŽ . Ž .AT" and AT homozygous ATyry cell linesusing radiation induced chromosomal breaks in G2

arrested cells as our assay.

2. Materials and methods

2.1. Cell lines

Lymphoblastoid cell lines from three AT familieswere obtained from the Coriell Cell RepositoriesŽ .Camden, NJ . The cell lines for each family arelisted in the following order; the proband, the femaleobligate heterozygote and the male obligate het-erozygote and include; FAMILY 1: GMO3332C,3334A, and 3382A; FAMILY 2: GMO3189C, 3188A,

and 3187; FAMILY 3: GMO8436A, 8388, and 8390.Three AT unaffected control cell lines, GMO3714,GM13068, and GM13068 were also investigated.

2.2. Cell culture, irradiation, and harÕesting

Initiation of cell cultures was performed as per theinstructions from the Coriell Cell Repositories, withthe only modification being that the level of FBSŽ .Gibco was increased to 20% for all cell lines.Cultures were grown for no more than one monthŽ .about 20 passages before a new culture was startedfrom frozen cultures.

At 72 h, a cell suspension of log phase cells foreach sample was irradiated with 50 cGy using a137 w xCs g cell with a dose rate of 100 cGyrmin 21 .For caffeine treatment samples, 1 mM caffeine wasadded immediately after irradiation and all sampleswere maintained in a 378C water bath for 30 min.

Ž .Ž .Colcemid Gibco 0.2 mgrml was added to thesamples 30 min after irradiation, and the samplesincubated for a further 60 min. Metaphase prepara-tion slides were made by conventional protocol and

Ž . w xstained with 10% Giemsa Gurr for analysis 21,22 .

2.3. Cytogenetic scoring and statistical analysis

Chromatid breaks, being defined as a space greaterthan the width of the chromatid, were scored by oneobserver from coded slides. Unpaired Student’s t-tests and descriptive statistical analyses were per-formed using Microsoft Excel.

3. Results

Ž .The lymphoblastoid cell lines LCLs from knownAT homozygotes, AT heterozygotes and AT unaf-fected controls were examined for chromatid breaksfollowing 50 cGy irradiation in G . Although more2

Žbreaks were observed at higher radiation doses data.not shown , the 50 cGy dose maximized the number

of cells proceeding into mitosis and permitted effec-tive analysis. After irradiation and without caffeinetreatment, controls showed an average of 0.97Ž ."0.30 breaksrcell, while AT heterozygotes and

Ž .AT homozygotes demonstrated 1.58 "0.37 andŽ . Ž .3.24 "0.41 breaksrcell respectively Table 1 . AT

( )D.G. Bebb et al.rMutation Research 401 1998 27–32 29

Table 1Chromosome aberrations in lymphoblastoid cell lines exposed to 50 cGy from AT-families and controls showing chromatid breaksrcellwith and without 1 mM caffeine

Sample Phenotype Chromosome aberrations Chromosome aberrationsa aWithout caffeine Metaphase spreads With 1 mM caffeine Metaphase spreads

GMO3332C AT homozygote 3.60 24.5 3.75 30GMO3189C AT homozygote 3.33 37 3.17 30GMO8436A AT homozygote 2.80 30 2.50 30

Ž . Ž . Ž .mean "SD 3.24 "0.41 30.5 3.14 "0.63 30

GMO3334A AT het 1.95 25 3.30 30GMO3382A AT het 2.06 30.5 2.60 30GMO3188A AT het 1.57 26 2.37 30GMO3187 AT het 1.23 27 3.03 30GMO8388 AT het 1.15 23 2.35 30GMO8390 AT het 1.50 30 2.20 30

Ž . Ž . Ž .mean "SD 1.58 "0.37 26.9 2.64 "0.43 30

GMO3714 AT unaffected 0.72 26 2.09 27GM13068 AT unaffected 0.90 24 1.97 25GM13079B AT unaffected 1.30 15 1.80 15

Ž . Ž . Ž .mean "SD 0.97 "0.30 21.6 1.95 "0.15 22.3

aAverage of scored metaphase spreads.

Ž .homozygous cells ATyry could always be dis-Ž .tinguished from the heterozygotes AT" which in

turn were significantly more sensitive to radiation-in-duced chromatid breakage than the unaffected con-

Ž .trols Fig. 1 .The effect of caffeine was examined by adding 1

mM of the drug immediately following irradiation.We note that this concentration of caffeine did notsignificantly increase the number of breaks in non-

Ž .irradiated cells data not shown . However, caffeineenhanced the number of radiation induced chromoso-

Fig. 1. Effect of caffeine on chromosome breaks after g-irradiationŽ .50 cGy in G2.

mal breaks per cell in both the unaffected controls,Ž Ž . .1.95 "0.15 breaksrcell , and AT heterozygotes,Ž Ž . .2.64 "0.43 breaksrcell , but not in AT homozy-

Ž Ž . .gotes, 3.14 "0.63 breaksrcell . The level of radia-tion induced chromatid breaks in the control and theAT heterozygote cell lines were still significantly

Ž .different pF0.005 but this difference was reducedcompared to results in a parallel experiment carried

Ž . Ž .out in the absence of caffeine pF0.0005 Fig. 1 .

4. Discussion

As predicted, caffeine significantly increased thenumber of radiation induced chromatid breaks in theAT-unaffected controls and AT heterozygote cells.However, caffeine did not alter the number of radia-tion induced chromatid breaks in the AT homozy-gote cells. In fact, in the presence of caffeine, thenumber of breaks in the controls and AT heterozy-gote cells approached that of the AT homozygotecells. This suggests that there is a ceiling to caffeineinduced effects which has already been attained inAT homozygote cells. Although this is the firstdescription of the effect of caffeine post irradiationon AT heterozygote cells, this observation is in

( )D.G. Bebb et al.rMutation Research 401 1998 27–3230

agreement with most work done using caffeine onAT homozygous cells. It suggests that caffeine wouldnot be a useful agent for screening for AT hetero-zygotes, since rather than improving discrimination,it masks any pre-existing difference.

Attempts to reliably discriminate AT hetero-zygotes on the basis of g-radiation in G have been2

w xinconclusive 10,21–24 . Although this may, in part,reflect the fact that the AT defect includes several

w xmutations with heterogeneous expression 15 it islikely that small, but significant, variations in experi-mental protocol in carrying out the G chromosomal2

radiosensitivity assay also contribute to this. Suchprocedural differences may include the nature of

Ž . w xirradiation X-rays vs. g-rays , the dose rate 23,25 ,the cell concentration at dose administration, cell

w x w xpellet 21 vs. suspension 26 , and the time follow-w xing exposure that damage is assessed 21 . Another

source of variation may be the very definition ofŽ .chromosomal damage gaps vs. breaks , since chro-

matid gaps that are less than the width of the chro-matid have a linear dose response whereas breaks do

w xnot 27 . The Chatham Barrs Inn Conference recom-mended the use of the length of the lesion as acriterion for distinguishing gaps and breaks; namely,when the length of the achromatic lesion is equal toor longer than the width of a chromatid, it is called achromatid break, whereas, when the length of thelesion is shorter than the width of a chromatid, it iscalled a gap. Although this is the definition weselected for our purposes, it is not universally usedw x21 .

Caffeine has been shown to synergize ionizingradiation induced chromosomal damage in a range of

w xcell types 16,17,28–31 . However, very little ispublished on inhibitors of DNA synthesisrrepair,including caffeine, in G -irradiated AT cells. Potenti-2

ation of chromatid breaks by caffeine in normallymphoblasts and fibroblasts but not in ATMyrylymphoblasts or fibroblasts has previously been de-

w xscribed by Hansson et al. 17 . However, increasedŽ .chromatid aberrations including gaps in both AT

and normal lymphoblasts have also been reportedw xwith caffeine 28 . Other investigators have treated

Ž .ATMyry, ATM" and control ATMqrqlymphoblastoid cells with different inhibitors of DNA

Žsynthesisrrepair cytosine arabinoside, aphidicolin,. w xbutyl–phenylen–guanine post irradiation 32 . In

these studies, no enhancing effect on chromosomalŽaberrations was observed in AT homozygous ATM

.yry cells, but a significant increase was seen inŽ . ŽAT heterozygous ATM" and normal ATMq

.rq cells. Little information is available on therelative effect of caffeine post radiation on ATMqrq, ATM" and ATMyry lymphoblastoid orfibroblast cells.

In studies of DNA repair deficiencies other thanAT, a correlation between radiosensitivity and lackof potentiation of G ionizing radiation induced DNA2

damage by caffeine has been reported. Darroudi andw xNatarajan 33 showed that treatment with 1 mM

Ž .caffeine after G irradiation 70, 100 and 140 cGy2

potentiated damage in normal CHO cells but onlyminimally in xrs-5 and not at all in xrs-6 cells, bothof which are radiation sensitive. Similarly, Parshad

w xet al. 34 demonstrated that irradiation and subse-quent treatment with caffeine increased chromosomedamage in a normal human fibroblast line, but not ina radiosensitive, malignant, derivative. These resultsare reminiscent of those seen in the yeast rad-3radiosensitive mutant which are insensitive to furtherpotentiation of g-radiation induced damage by caf-

w xfeine 35 . Interestingly, the ATM gene shares ho-mology with the yeast rad-3 gene sequence, which

w xis implicated in DNA repair 15,36 .Although an epidemiological association between

AT carrier status and breast cancer proposed byw x w xSwift 37 has been verified by others 38,39 , it is

yet to be confirmed at the molecular level. Despitethe recent cloning and sequencing of the AT gene, itssheer size and the lack of mutational specificitymakes it unlikely that a rapid, convenient and inex-pensive screening test for carrier status will be avail-

w xable in the near future 40,41 . Nevertheless, theclinical implications of confirming this epidemiolog-

w xical association 42 makes it still desirable to de-velop such a phenotypic assay. Unfortunately, ourwork indicates that caffeine is not a useful agent forfacilitating the discrimination of AT heterozygoteŽ . Ž .AT" from normal ATqrq individuals. How-ever, the observation that caffeine increases chromo-

Ž .somal breaks in controls ATqrq and AT het-Ž .erozygote AT" cell lines but not in AT homozy-

Ž .gote ATyry cell lines suggests that caffeineinterferes with a regulatory pathway in which theATM gene product plays a central role.

( )D.G. Bebb et al.rMutation Research 401 1998 27–32 31

References

w x1 O. Jelmert, I.L. Hansteen, S. Langard, Enhanced cytogeneticdetection of previous in vivo exposure to mutagens in humanlymphocytes after treatment with inhibitors of DNA synthe-

Ž .sis and DNA repair in vitro, Mutation Res. 271 1992289–298.

w x2 T.T. Puck, H. Morse, R. Johnson, C.A. Waldren, Caffeineenhanced measurements of mutagenesis by low levels ofg-irradiation in human lymphocytes, Somatic Cell Mol.

Ž . Ž .Genet. 19 5 1993 423–429.w x3 I.G. Walker, B.D. Reid, Caffeine potentiation of the lethal

action of alkylating agents on L cells, Mutation Res. 12Ž .1971 101–104.

w x4 C.P. Selby, A. Sancer, Molecular mechanisms of DNA inhi-Ž .bition by caffeine, Proc. Natl. Acad. Sci. USA 87 1990

3522–3525.w x5 P.M. Busse, S.K. Bose, R.W. Jones, L.J. Tlmach, The action

of caffeine on X-irradiated HeLA cells: II. Synergistic lethal-Ž .ity, Radiat. Res. 71 1977 666–667.

w x6 J.H. Hong, R.A. Gatti, Y.K. Huo, C.S. Chiang, W.H.McBride, G rM-phase arrest and release in ataxia telang-2

iectasia and normal cells after exposure to ionizing radiation,Ž .Radiat. Res. 140 1994 17–23.

w x7 J.P. Murnane, Cell cycle regulation in response to DNAdamage in mammalian cells: a historical perspective, Cancer

Ž .Metastasis Rev. 14 1995 17–29.w x8 R.A. Gatti, E. Boder, H.V. Vinters, R.S. Sparkes, A. Nor-

man, K. Lange, Ataxia telangiectasia: an interdisciplinaryŽ . Ž .approach to pathogenesis, Medicine 70 2 1991 99–117.

w x9 M.C. Paterson, A.K. Anderson, B.P. Smith, P.J. Smith,Enhanced radiosensitivity of cultured fibroblasts from ataxiatelangiectasia heterozygotes manifested by defectivecolony-forming ability and reduced DNA repair replication

Ž .after hypoxic g-irradiation, Cancer Res. 39 1979 3725–3734.

w x10 P. Chen, A. Farrell, K. Hobson, A. Girjes, M. Lavin, Com-parative Study of radiation-induced G phase delay and2

chromatid damage in families with ataxia telangiectasia,Ž .Cancer Genet. Cytogenet. 76 1994 43–46.

w x11 G.W. Birrell, J.R. Ramsey, Induction of p53 protein bygamma radiation in lymphocyte lines from breast cancer and

Ž .ataxia telangiectasia patients, Br. J. Cancer 72 1995 1096–1101.

w x12 E. Duchaud, A. Ridet, Y. Delic, E. Cunardi, E. Moustacchi,F. Rosselli, Changes in the radiation induced apoptotic re-sponse in homozygotes and heterozygotes for the ataxia

Ž .telangiectasia gene, C. R. Acad. Sci. 317 1994 983–989.w x13 M.A. Hannan, M. Kunhi, M. Einspenner, B.A. Khan, S.

Al-Sedairy, Post-irradiation DNA synthesis inhibition and G2

phase delay in radiosensitive body cells from non-Hodgkin’slymphoma patients: an indication of cell cycle defects, Muta-

Ž .tion Res. 311 1994 265–276.w x14 T.K. Pandita, W.N. Hittelman, Increased initial levels of

chromosome damage and heterogeneous chromosome repairin ataxia telangiectasia heterozygote cells, Mutation Res.

Ž .310 1994 1–13.

w x15 K. Savitsky, A. Bar-Shira, S. Gilad, G. Rotman, Y. Ziv, L.Vanagaite, D.A. Tagle, S. Smith, T. Uziel, S. Sfez, M.Ashkenazi, I. Pecker, M. Frydman, R. Harnik, S.R. Patanjali,A. Simmons, G.A. Clines, A. Sartiel, R.A. Gatti, L. Chessa,O. Sanal, M.F. Lavin, N.G.J. Jaspers, A.M.R. Taylor, C.F.Arlett, T. Miki, S.M. Weissman, M. Lovett, F.S. Collins, Y.Shiloh, A single ataxia telangiectasia gene with a product

Ž .similar to PI-3 kinase, Science 268 1995 1749–1753.w x16 P.R. Bates, F.P. Imray, M.F. Lavin, Effect of caffeine on

gamma-ray-induced G delay in ataxia telangiectasia, Int. J.2Ž .Radiat. Biol. 47 1985 713–722.

w x17 K. Hansson, A.T. Natarajan, B.A. Kihlman, Effect of caf-feine in G on X-ray-induced chromosomal aberrations and2

mitotic inhibition in ataxia telangiectasia fibroblast andŽ .lymphoblastoid cells, Hum. Genet. 67 1984 329–335.

w x18 M. Swift, L. Sholman, M. Perry, C. Chase, Malignant neo-plasms in the families of patients with ataxia telangiectasia,

Ž .Cancer Res. 36 1976 209–215.w x19 M. Swift, P.J. Reitnauer, D. Morrell, C.L. Chase, Breast and

other cancers in families with ataxia telangiectasia, N. Eng.Ž .J. Med. 316 1987 1289–1294.

w x20 M. Swift, D. Morrell, R.B. Massey, C.L. Chase, Incidence ofcancer in 161 families affected by ataxia telangiectasia, N.

Ž .Eng. J. Med. 325 1991 1831–1836.w x21 D. Scott, A.R. Spreadborough, S.A. Roberts, Radiation-in-

duced G delay and spontaneous chromosome aberrations in2

ataxia telangiectasia homozygotes and heterozygotes, Int. J.Ž .Radiat. Biol. 66 1994 S157–163.

w x22 K.K. Sanford, R. Parshad, F.M. Price, G.M. Jones, R.E.Tarone, L. Eierman, P. Hal, T.A. Waldmann, Enhancedchromatid damage in blood lymphocytes after G phase2

X-irradiation, a marker for the ataxia telangiectasia gene, J.Ž .Natl. Cancer Inst. 82 1990 1050–1054.

w x23 M. Waghray, S. Al-Sedairy, P.T. Ozand, M.A. Hannan,Ž .Cytogenetic characterization of ataxia telangiectasia AT

heterozygotes using lymphoblastoid cell lines and chronicŽ .g-irradiation, Hum. Genet. 84 1990 532–534.

w x24 Y. Shiloh, R. Parshad, M. Frydman, K.K. Sanford, S. Port-noi, Y. Ziv, G.M. Jones, G chromosomal radiosensitivity in2

Ž .families with ataxia telangiectasia, Hum. Genet. 84 19898–15.

w x25 L.A. Jones, D. Scott, R. Cowan, S.A. Roberts, Abnormalradiosensitivity of lymphocytes from breast cancer patientswith excessive normal tissue damage after radiotherapy:chromosome aberrations after low dose-rate irradiation, Int.

Ž . Ž .J. Radiat. Biol. 67 5 1995 519–528.w x26 D. Scot, L.A. Jones, S.A.G. Elyan, A. Spreadborough, R.

Ž .Cowan, G. Ribiero, in: R.A. Gatti, R.B. Palmer Eds. ,Identification of AT-heterozygotes in ataxia telangiectasia,NATO ASI Series, 1993, Vol. H77, pp. 101–116.

w x27 M.N. Cornforth, J.S. Bedford, Ionizing radiation damage andits early development in chromosomes, in: Advances inRadiation Biology, vol. 17, 1993.

w x28 I. Lopez, G. Tapia, C. Oyarzun, J. Pincheira, Evaluation ofchromosome damage induced by X-rays in human lympho-

Ž .cytes, Rev. Med. Chile 121 1993 1240–1244.w x29 A.T. Natarajan, G. Obe, F.N. Dulout, The effect of caffeine

( )D.G. Bebb et al.rMutation Research 401 1998 27–3232

post-treatment on X-ray-induced chromosomal aberrations inŽ .human blood lymphocytes in vitro, Hum. Genet. 54 1980

183–189.w x30 F. Zampetti-Bosseler, D. Scott, The response of normal and

ataxia telangiectasia cells to bleomycin: relationships be-tween chromosome damage, cell cycle delay and cell killing,

Ž .Mutation Res. 151 1985 89–94.w x31 J. Pincheira, M. Rodriguez, M. Bravo, M.H. Navarrete, J.F.

Lopez-Saez, Defective G repair in Down syndrome: effect2

of caffeine, adenosine and niacinamide in control and X-rayŽ .irradiated lymphocytes, Clin. Genet. 45 1994 25–31.

w x32 A. Antoccia, F. Palitti, T. Raggi, C. Catenas, C. Tanzarella,Lack of effect of inhibitors of DNA synthesisrrepair on theionizing radiation-induced chromosomal damage in G stage2

Ž .of ataxia telangiectasia cells, Int. J. Radiat. Biol. 66 3Ž .1994 309–317.

w x33 F. Darroudi, A.T. Natarajan, Cytological characterization ofChinese hamster ovary X-ray-sensitive mutant cells xrs 5 and

Ž .xrs 6, Mutation Res. 177 1987 133–148.w x34 R. Parshad, R. Gantt, K.K. Sanford, G.M. Jones, R.E. Tarone,

Repair of chromosome damage induced by X-irradiationduring G phase in a line of normal human fibroblasts and its2

Ž .malignant derivative, J. Natl. Cancer Inst. 69 1982 409–414.w x35 G. Jimenez, J. Yucel, R. Rowley, S. Subramani, The rad3q

gene of Shizosaccharomyces pombe is involved in multiplecheckpoints and in DNA repair, Proc. Natl. Acad. Sci. USA

Ž .89 1992 4952–4956.

w x36 V.A. Zakian, ATM-related genes: what do they tell us aboutŽ .functions of the human gene?, Cell 82 1995 685–687.

w x37 M. Swift, Ionizing radiation, breast cancer, and ataxiaŽ . Ž .telangiectasia, J. Natl. Cancer Inst. 86 21 1994 1571–

1572.w x38 R.A. Eeles, M.R. Stratton, D.E. Goldgar, D.F. Easton, The

genetics of familial breast cancer and their practical implica-Ž .tions, Eur. J. Cancer 30 1994 1383–1390.

w x39 D.F. Easton, Cancer risks in AT heterozygotes, Int. J. Radiat.Ž .Biol. 66 1994 S177–182.

w x40 S. Gilad, R. Khosravi, D. Shkedy, T. Uziel, Y. Ziv, K.Savitsky, G. Rotman, S. Smith, L. Chessa, T.J. Jorgensen, R.Harnik, M. Frydman, O. Sanal, S. Portnoi, Z. Goldwicz,N.G.J. Jaspers, R. Gatti, G. Lenoir, M. Lavin, K. Tatsumi,R.D. Wegner, Y. Shiloh, A. Bar-Shira, Predominance of null

Ž .mutations in ataxia telangiectasia, Human Mol. Genet. 5 4Ž .1996 433–439.

w x41 M. Telatar, Z. Wang, N. Udar, T. Liang, E. Bernatowska-Matuszkewicz, M. Lavin, Y. Shiloh, P. Concannon, R.A.Good, R.A. Gatti, Ataxia telangiectasia: mutations in ATMcDNA detected by protein truncation screening, Am. J. Hu-

Ž .man Genet. 59 1996 40–44.w x42 G. Bebb, K. Gelmon, B. Glickman, R. Gatti, AT risk for

breast cancer, Lancet, 21st June, 1997.