RESEARCH ARTICLE

Similar Mutant Frequencies Observed BetweenPairs of Monozygotic Twins

John Curry,1 Gwyn Bebb,1 Joyce Moffat,1 Daniel Young,1 Magomed Khaidakov,1 Alan Mortimer,2

and Barry W. Glickman1*1Centre for Environmental Health and the Department of Biology, University of Victoria, Victoria, British Columbia V8W 3N5,Canada; Fax: (250)472-4075

2Space Life Sciences, Canadian Space Agency, Ottawa, Ontario K1L 8E3, Canada

Communicated by Steve S. Sommer

The relative contribution of both genetic and environmental factors to spontaneous mutationfrequency in humans is unknown. We have investigated the contribution of genetic factors to thisphenomenon by determining the in vivo mutant frequency at the hypoxanthine-guaninephosphoribosyltransferase (hprt) locus in circulating T-lymphocytes obtained from pairs of monozy-gotic twins. hprt mutant frequencies were determined three times over fourteen days in six sets ofmonozygotic male twins (mean age 30) taking part in a Russian Space program inclined bed rest experi-ment. Blood samples were obtained prior to, during, and immediately following the experiment. Mono-nuclear cells were separated, frozen, and flown to Canada for analysis using the hprt T-lymphocyteclonal assay. There is no evidence within this data set to demonstrate that the period of inclined bed restto simulate the effects of weightlessness had any effect on the observed mutant frequency. However, theaverage mutant frequency for the six sets of Russian twins was found to be three times higher than thatof Western counterparts. More surprisingly, the spontaneous mutant frequency of monozygotic twinswas found to be much more similar within pairs than between pairs of twins. These data suggest thatthe contribution of genetics in the determination of mutation frequency is substantial. However, whetherhigh concordance within twin pairs reflects shared environmental experience as well as common ge-netic factors is not entirely clear. More data will be required to distinguish genetic from environmentalfactors and to determine the degree to which mutant frequency is genetically determined. Hum Mutat9:445–451, 1997. © 1997 Wiley-Liss, Inc.

KEY WORDS: monozygotic twins, Russian, T-cell clonal assay, hprt

HUMAN MUTATION 9:445�451 (1997)

© 1997 WILEY-LISS, INC.

INTRODUCTION

There is substantial evidence that spontaneousmutation rates are genetically controlled (Drake,1970, 1991). However, the relative contribution ofboth genes and the environment to the generationof background mutation is unclear. The extent towhich mutations are accumulated by individualsmust in part be determined by that individual’s ge-nome. Accordingly the inherent plasticity in the ge-nome must allow environmental factors to influencespontaneous mutation as well. A previous study em-ploying the classic twin study design with the sisterchromatid exchange (SCE) assay demonstrated thatafter accounting for smoking and coffee drinking sta-tus, genetic factors accounted for ~30% of the varia-tion in SCE rates (Hirsh et al., 1992). In this study,we have determined the spontaneous in vivo mu-tant frequency for six sets of monozygotic twins in

an attempt to examine the potential genetic and en-vironmental influences on this process.

Interindividual variation in spontaneous mutantfrequency has been well documented (Cole et al.,1988; Tates et al., 1991; Branda et al., 1993; Coleand Skopek, 1994). This variation at the cellular levelmight be expected since considerable interindividualvariation in sensitivity has been observed in vivo toknown mutagens such as radiation and cigarettesmoke (Paterson et al., 1985a; Caporaso and Landi,1994). Many factors including polymorphisms inmetabolizing enzymes, differences in DNA repaircapability, and differences in chromatin structurecould contribute to this variation (Spitz, 1995;

Received 4 December 1995; accepted 9 July 1996.

*Correspondence to Barry W. Glickman.

HUMU pg 445

446 CURRY ET AL.

Hittelman and Pandita, 1994). Such individual varia-tion in sensitivity has relevance in the clinical set-ting. A wide variation in radiosensitivity has beenobserved in normal individuals (Little et al., 1989)as well as in individuals with known DNA repair dis-orders such as ataxia telangiectasia, which are asso-ciated with an increased susceptibility for cancer(Paterson et al., 1984). For example, it is not un-common for a patient undergoing radiotherapy tohave the treatment terminated due to adverse ef-fects stemming from their intrinsic radiation sensi-tivity (Paterson et al., 1985b; Arlett and Harcourt,1980; Chen et al., 1978; Little et al., 1988; Nagasawaand Little, 1988). In an extreme example of radia-tion sensitivity, Badie et al. (1995) described a leu-kemia patient who died during radiation therapy. Thepatient was later realized to be defective in the re-pair of DNA double-strand breaks that was possiblya causative factor in the development of the initialdisease (Badie et al., 1995).

Clearly, the philosophy of “equal exposure =equal risk” is not valid. Yet, generally accepted safetystandards of exposure to hazardous agents are basedon this assumption and fail to take into account thebroad spectrum of potential human responses. Pre-cisely identifying factors which influence individualvariation in the spontaneous mutation frequency andsusceptibility to mutagens has far-reaching implica-tions. Appropriate protection or restriction from ex-posure to radiation or chemicals may be required ifan individual’s sensitivity to these agents can be es-tablished. Variation in individual response could beused as the basis of a screening program with thepossibility that life style counseling be considered forparticularly susceptible individuals. Before that ispossible however, a clearer understanding of the ge-netic basis for mutation and the influences of theenvironment on mutagenesis will be required.

The ability to grow hprt mutant T cells in vitrohas allowed researchers to establish in vivo mutantfrequencies1 (Albertini et al., 1982; Cole et al., 1988;Tates et al., 1991; Branda et al., 1993) and then char-acterize the nature of mutation at the molecular level(Rossi et al., 1992; Recio et al., 1990; Gibbs et al.,1990; Fuscoe et al., 1991; Steingrimsdottir et al.,1992, 1993; Cole and Skopek, 1994; Curry et al.,1993, 1995). Thus, in populations exposed to envi-

ronmental mutagens, hprt provides a molecularmarker for mutation.

This study involves a set of six male monozygotictwins participating in a Russian Space Agency studydesigned to simulate the physiological effects ofweightlessness in space. The simulation involves bedrest at an inclination (head-down) of 7°, for a pe-riod of 2 weeks. There was no expectation that thisregimen would affect mutant frequencies, but thestudy did provide access to healthy, well-monitored,monozygotic twins. The individual sets of twins hadalways lived together; and the entire group washoused and fed together, 2 weeks before and through-out the bed rest protocol. Thus, each of the twinsets had experienced similar environments for mostof their lives and throughout the course of the ex-periment. The question posed here is whethermonozygotic twins have similar mutant frequencies.The finding of similar mutant frequencies would sug-gest that the environmental conditions and/or ge-netic factors affect each member of a twin pair in asimilar manner.

METHODSSample Collection

The Russian Space Agency identified seven setsof monozygotic twins to participate in an inclinedbed rest experiment that simulates the physiologicaleffects of weightlessness. These sets of twins wereidentified from a national database established to fa-cilitate a variety of experimental applications. Thesesets of twins were demonstrated genetically to be truemonozygotic twins. The twins were not exposed toany other experimental factors except the bed rest,and only one set of twins smoked. Mononuclear cells(MNCs) were recovered from the volunteers (meanage 30 ± 2 years) using Leucoprep tubes (Becton-Dickinson, Franklin Lakes, NJ) following themanufacturer’s protocols. Blood was drawn fromeach set of twins at the same time of day. Codedsamples were collected prior to the bed rest experi-ment (t = 0 days), during the bed rest (t = 14 days),and following the conclusion (t = 21 days) of thesimulation. MNCs were counted and diluted to 107

cells/ml in 50% calf serum (CBS), 10% DMSO, and40% RPMI 1640. The cells were frozen in a styrofoambox placed into a –80°C freezer (Cole et al., 1988),transported to Canada on dry ice, and stored underliquid nitrogen until required.

T-Cell Clonal Assay

Frozen MNCs were thawed quickly in a 37°Cwaterbath, washed twice in RPMI 1640 supple-mented with 10% CBS, and pre-incubated in growth

1The correction of mutant frequency to mutation frequenciesrequires correction for clonally related mutants (Curry et al.,1995). This correction does not generally exceed 20�30% of theMF with the possible exception of subjects whose MF is greaterthan 40 × 10�6 (O�Neill et al., 1994). Thus while this study in-volves mutant frequencies, they can be considered as reasonableapproximations of mutation frequencies.

SIMILAR MUTANT FREQUENCIES IN TWINS 447

medium with no phytohaemagglutinin (PHA) orinterleukin 2 (IL-2) overnight (12–16 hr) (Cole etal., 1988). MNCs were plated in 200 ãl of selective(6-thioguanine) and nonselective media (2 plates)in microtiter plates (96-well, flat-bottom) at 104 andthree cells per well, respectively, along with 104 le-thally irradiated feeder cells (RJK 853 cells, alymphoblastoid cell line, derived from a Lesch-Nyhanpatient found to have a complete deletion of the hprtgene (Yang et al., 1984)). Our growth medium con-sisted of RPMI 1640, 20% HL-1, 5% CBS, 5% hu-man AB serum (Table 1), 5 U/ml IL-2, 0.25 ãg/mlPHA, 2 mM L-glutamine, 2 mM pyruvic acid, 100U/ml penicillin, 100 ãg/ml streptomycin and 4%

Fungizone. Selection plates contained 10–5 M 6-thioguanine. Plates were incubated on a sloped shelf(5°) in a 5% CO2, 37°C, humidified incubator for aperiod of 14 days. Twenty-four hr prior to scoring,the plates were rotated 180° on the sloped shelf thenscored visually using an inverted phase-contrast mi-croscope for wells that contained expanding colo-nies. All plates were re-scored 3 days later to allowsmall colonies that may have been missed during thefirst scoring to be counted. Individual mutants cloneswere collected from the selective plates and furtherexpanded when possible and stored to permit thecharacterization of the hprt and Ö-TCR sequences ata later date.

TABLE 1. Calculation of Plating and Mutant Frequenciesa

Non Mutant 95% 95%Subject selection 6-TG Plating frequency Upper Lowersample wells selection wells efficiency (× 10�5) (× 10�5) (× 10�5)

R2A1* 167 852/864 0.05 3.0 6.0 1.5R2B1* 163 946/960 0.05 2.7 5.1 1.4R2A2* 178 473/480 0.03 5.8 14.4 2.3R2B2* 182 666/672 0.02 5.0 13.8 1.8R2A3* 175 761/768 0.03 3.0 7.1 1.2R2B3* 166 760/768 0.05 2.2 4.8 1.0R3A1* 141 404/408 0.10 1.0 2.6 0.3R3B1* 139 397/400 0.11 0.7 2.2 0.2R3A2* 150 119/120 0.08 1.0 7.4 0.1R3B2* 157 451/456 0.07 1.6 4.2 0.6R3A3* 178 288/288 0.03 0.0 � �R3B3* 174 96/96 0.03 0.0 � �R4A1* 188 1048/1056 0.01 10.8 36.0 3.3R4B1* 186 355/360 0.01 13.2 43.3 4.0R4A2 85 398/414 0.27 1.5 2.5 0.9R4B2 106 366/376 0.20 1.4 2.6 0.7R4A3 113 386/400 0.18 2.0 3.6 1.1R4B3 124 209/216 0.15 2.3 4.9 1.0R5A1 91 165/168 0.25 0.7 2.3 0.2R5B1 99 104/104 0.22 0.0 � �R5A2 125 142/144 0.14 1.0 4.0 0.2R5B2 123 142/144 0.15 0.9 3.8 0.2F5A3 88 302/312 0.26 1.3 2.4 0.7R5B3 71 348/360 0.33 1.0 1.9 0.6R6A1 76 196/208 0.31 1.9 3.5 1.1R6B1 53 177/192 0.43 1.9 3.2 1.1R6A2 100 22/24 0.22 4.0 16.3 1.0R6B2 105 107/112 0.20 2.3 5.6 0.9R6A3 95 126/136 0.23 3.3 6.3 1.7R6B3 96 229/240 0.23 2.0 3.8 1.1R7A1 117 128/136 0.17 3.7 7.6 1.8R7B1 84 209/224 0.28 2.5 4.3 1.5R7A2 118 117/120 0.16 1.6 5.0 0.5R7B2 143 259/264 0.10 1.9 4.9 0.8R7A3 106 516/536 0.20 1.9 3.1 1.2R7B3 111 380/392 0.18 1.7 3.1 0.9

aSubject samples are codeed such that R# refers to a set of twins (# = 2�7) where A or B indicates the individual twin. The next numberfollowing refers to the sample collection period: (1) before (t = 0), (2) during (t = 14), and (3) after (t = 21) the bed rest experiment. Theage of the twins is as follows: R2 = 31, R3 = 35, R4 = 35; R5 = 24, R6 = 24, R7 = 31 years. The numbers given for the Non-Selection(192 wells plated) and 6-TG selection wells are those wells in which no colony was observed. The 95% lower and upper confidenceintervals for the mutant frequency are calculated following the methods of Branda et al. (1993). Subject codes identified (*) have lowerplating efficiencies than those not identified. These samples were plated in our laboratories in Toronto, Canada. At that time we had beenusing RPMI 1640 and HL-1 prepared in our laboratory and were not using human serum. The other samples, those with higher efficien-cies, were plated in Victoria, Canada with premade RPMI 1640, HL-1, and human serum was used. We attribute the difference observedin plating efficiencies to these factors.

448 CURRY ET AL.

RESULTS

Mutant frequencies (MF) were determined foreach of the three samples taken from the subjects(Table 1). From the 36 blood samples subjected tothe T-cell clonal assay, 33 yielded usable MF dataand those that were zero were excluded from fur-ther analysis. The mean plating efficiency (PE) andMF for all subjects was 15.3 ± 1.7% and 27.5 ± 4.7× 10–6, respectively. The average MF for each sub-ject, was determined using the three independentdeterminations (where possible). These averageswere plotted along with standard error and devia-tions for each subject (Fig. 1). In addition the mu-tant frequencies for each of the twins were plottedagainst each other, and the 95% confidence bandsfor the calculated linear regression line (R2 = .92, P< 10–6) (Fig. 2). An identical plot of the PE valuesobtained for the twin sets reveals a similar statisti-cally significant result (PETwin A = 1.24 + 0.74 PETwin

B, R2 = 0.933, P < 10–6). The twin A MF determi-nations were then randomly matched with MF de-terminations obtained from the twin B group andthe matches plotted as done for Figure 2. After sev-eral independent randomly assigned matches, nonedemonstrated any statistical significance nor theslope of the calculated linear regression line shownin Figure 2.

Using the calculations that relate MF and PE withage, given by both Branda et al. (1993) and Tates etal. (1991), the mean MF for the subjects presentedhere should be 10.0 × 10–6 and 11.4 × 10–6, respec-tively. The mean MF determined for these subjectsis approximately three times higher than the calcu-lated means obtained from these two formulas indi-cating that the Russian subjects have a higherbackground MF than their Western counterparts.The relationship between PE and 1nMF has beenpreviously demonstrated by various laboratories(Cole et al., 1991; Tates et al., 1991; Branda et al.,1993) to be negatively related, such that increasedPE result in decreased 1nMF values as demonstratedby Branda et al. (1993) (1nMF = 2.91 – 1.13PE, R2

= 0.109, P <0.001) using a rather large collectionof data. A similar relationship for the subjects pre-sented here was observed (1nMF = 3.49 – 2.8PE,R2 = 0.18, P <0.014).

DISCUSSION

The observation that the Russian MF were threetimes higher than that of Western counterparts isstriking. The reasons for this increase can only bespeculative at this time. One possibility could be re-lated to an increased mutagen burden placed on resi-dents of this area. Another possibility may reflect

FIGURE 1. Comparison of the mean 1n MF determined foreach subject. All three MF determinations were obtainedand used to provide the means plotted here with standarderror (box) and standard deviations (line). Two subjects (R3A

and R5B) yielded a zero for one of the three MF determina-tions and thus only two MF determinations were used toprovide the mean 1n MF.

SIMILAR MUTANT FREQUENCIES IN TWINS 449

differences in diet and lifestyle. Others have alsoobserved Russian subjects demonstrating elevatedMFs (Jones et al., 1995); however, these subjects hadpotential to receive low doses of radiation during thecleanup of the Chernobyl nuclear accident.

Throughout this study, mutant frequencies foreach monozygotic twin pair are remarkably similar(Fig. 2). This suggests that background mutation atthe hprt locus could have a significant genetic basis.However, since the early life environmentals arenearly identical for each set of twins, these factorsmay also contribute to the within pair similarity ofmutant frequencies.

A contribution of genotype in background mu-tant frequency is not unexpected. Factors that con-tribute to the sensitivity of mutagens, such as theactivity of enzymes involved in the metabolism ofxenobiotics (Daly et al., 1994) and DNA repair, anddeterminants of chromatin structure, are in part ge-netically defined. Moreover, these factors are sub-ject to considerable individual variation. The familialcomponent of cancer is well recognized, althoughthe relationship of these factors to carcinogenesis ismuch more complex. What remains unclear fromthis study is the identity of these genetic componentsand their relative contribution to the frequency ofmutation compared to environmental factors. Itshould be understood that neither the genetic northe environmental factors can be partitioned fromthis study. As these twins were reared together and

have thus for their entire lives been exposed to iden-tical environments and share identical genetic make-ups, such questions are beyond the scope of thepresent study. Studies with dizygotic twins couldhelp partition the relative contributions of envi-ronmental and genetic factors, but the most cru-cial study would involve monozygotic twins whohave been brought up under different environmen-tal circumstances.

One strong suggestion of the interaction betweengenetic and of environmental factors is revealed inother data. An environmental effect on mutationfrequency would be expected to increase with age(Branda et al., 1993; Cole et al., 1994), so that thegenetic effect would be most apparent in youngertwin pairs. Analysis of additional data kindly pro-vided by Branda et al. (1993) confirms this. The twineffect is apparent in twin pairs younger than 40 yearsof age (Fig. 3) though less so than in that presentedhere, but the effect has all but disappeared in setsolder than 40 years (Fig. 4). It is thus clear that dif-ferences in background mutant frequencies increasewith age and must result from a complex interactionbetween genotype and environment. While thisstudy of monozygotic twins may be one of the firstto demonstrate a genetic role in the determinationof mutation frequency in people, the interaction be-tween genetic and environmental factors in the gen-eration of mutation is complex, and factors such asdiet and lifestyle must play a considerable role. The

FIGURE 2. The mutant frequency of one-half of the subjectsversus the mutant frequency of their twin. Twin sets wereordered into groups A or B such that the subject with the

highest MF falls into group A. The calculated linear regres-sion line (1n MF twin B = �0.07 + 0.8 × 1n MF twin A, P<10�6) with 95% confidence bands is shown.

450 CURRY ET AL.

actual partitioning of genetic and environmental fac-tors in determining mutation frequency will requiremuch more sophisticated experimental designs us-

ing a large number of subjects, including bothmonozygotic and dizygotic twins, preferably some ofwhich have been reared separately.

FIGURE 3. The mutant frequency of monozygotic twins (n =7 sets of twins), with age less than 40 years versus that oftheir twin (Branda et al., 1993). Mean age 32.9 ± SD 0.86;mean PE 0.46 ± SD 0.14; mean MF 9.8 ± SD 6.0 × 10�6.

Twin sets were ordered into groups A or B such that thesubject with the highest MF falls into group A. The linearregression line (R2 = 0.71, P = 0.017) with 95% confidencebands is shown.

FIGURE 4. The mutant frequency of monozygotic twins (n =29 sets of twins), with ages greater than 40 years versusthat of their twin (Branda et al., 1993). Mean age 66.8 ± SD

3.8; mean PE 0.36 ± SD 0.17; mean MF 16.0 ± SD 9.1 ×10�6. The linear regression line (R2 = 0.08, P = 0.13) with95% confidence bands is shown.

SIMILAR MUTANT FREQUENCIES IN TWINS 451

ACKNOWLEDGMENTS

This work was supported by a grant from theNatural Sciences and Engineering Research Coun-cil of Canada and the Medical Research Council ofCanada. The authors thank Dr. J.A. Nicklas and thelaboratory of Dr. R.J. Albertini for the additional datarequired to decode the twin pairs from their expan-sive paper (Branda et al., 1993).

REFERENCESAlbertini RJ, Castle KL, Borcherding WR (1982) T-cell cloning to

detect the mutant 6-thioguanine-resistant lymphocytes presentin human peripheral blood. Proc Natl Acad Sci USA 79:6617–6621.

Arlett CF, Harcourt SA (1980) Survey of radiosensitivity in a vari-ety of human cell strains. Cancer Res 40:926–932.

Badie C, Iliakis G, Foray N, Alsbeih G, Pantellias GE, Okayasu R,Cheong N, Russell NS, Begg AC, Arlett CF, Malaise EP (1995)Defective repair of DNA double-strand breaks and chromosomedamage in fibroblasts from a radiosensitive leukemia patient.Cancer Res 55:1232–1234.

Branda RF, Sullivan LM, O’Neill JP, Falta MT, Nicklas JA, HirschB, Vacek PM, Albertini RJ (1993) Measurement of HPRT mu-tant frequencies in T-lymphocytes from healthy human popula-tions. Mutat Res 285:267–279.

Caporaso NE, Landi MT (1994) Molecular epidemiology: A newperspective for the study of toxic exposures in man. A consider-ation of the influence of genetic susceptibility factors on risk indifferent lung cancer histologies. Med Lav 85:68–77.

Chen PC, Lavin MF, Kidson C, Moss D (1978) Identification ofataxia telangiectasia heterozygotes, a cancer prone population.Nature 274:484–486.

Cole J, Skopek TR (1994) Somatic mutant frequency, mutation ratesand mutational spectra in the human population in vivo. MutatRes 304:33–105.

Cole J, Green MHL, James SE, Henderson L, Cole H (1988) Afurther assessment of factors influencing measurements ofthioguanine-resistant mutant frequency in circulating T-lympho-cytes. Mutat Res 204:493–507.

Curry J, Skandalis A, Holcroft J, de Boer J, Glickman BW (1993)Coamplification of hprt cDNA and Ö T-cell receptor sequencesfrom 6-thioguanine resistant human T-lymphocytes. Mutat Res288:269–275.

Curry J, Rowley GT, Saddi V, Beare D, Cole J, and Glickman BW(1995) Determination of hprt mutant and mutation frequen-cies and the molecular characterization of human derived invivo T-lymphocyte mutants. Environ Mol Mutagen 25:169–179.

Daly AK, Cholerton S, Armstrong M, Idle JR (1994) Genotypingfor polymorphisms in xenobiotic metabolism as a predictor ofdisease susceptibility. Env Health Perspect 102:55–61.

Drake JW (1970) The Molecular Basis of Mutation. San Francisco:Holden-Day.

Drake JW (1991) Spontaneous mutation. Annu Rev Genet 25:125–146.

Fuscoe JC, Zimmerman LJ, Lippert MJ, Nicklas JA, O’Neill JP,Albertini RJ (1991) V(D)J recombinase-like activity mediateshprt gene deletion in human fetal T-lymphocytes. Cancer Res51:6001–6005.

Gibbs RA, Nguyen P, Edwards A, Civitello AB, Caskey CT (1990)Multiplex DNA deletion detection and exon sequencing of the

hypoxanthine phosphoribosyltransferase gene in Lesch-Nyhanfamilies. Genomics 7:235–244.

Hirsch BA, Sentz KK, McCue M (1992) Genetic and environmen-tal influences on baseline SCE. Environ Mol Mutagen 20:2–11.

Hittelman WN, Pandita TK (1994) Possible role of chromatin al-teration in the radiosensitivity of ataxia-telangiectasia. Int JRadiat Biol 66:S109–S113.

Jones IM, Thomas CB, Tucker B, Thompson CL, Pleshanov P,Vorobtsova I, Moore DH II (1995) Impact of age and environ-ment on somatic mutation at the hprt gene of T-lymphocytes inhuman. Mutat Res 338:129–139.

Little JB, Nove J, Strong LC, and Nichols WW (1988) Survival ofhuman diploid skin fibroblasts from normal individuals after X-irradiation. Int J Radiat Biol 54:899–910.

Little JB, Nichols WW, Troilo P, Nagasawa H, Strong LC (1989)Radiation sensitivity of cell strains from families with geneticdisorders predisposed to radiation induced cancers. Cancer Res49:4705–4714.

Nagasawa H, Little JB (1988) Radiosensitivities of ten apparentlynormal human diploid fibroblast strains to cell killing, G2-phasechromosomal aberrations, and cell cycle delay. Cancer Res48:4535–4538.

Paterson MC, Gentner NE, Middlestadt MV, Wienfeld M (1984)Cancer prediposition, carcinogen hypersensitivity, and aberrantDNA metabolism. J Cell Physiol Suppl 3:45–62.

Paterson MC, MacFarlane SJ, Gentner NE, Smith BP (1985a) Cel-lular hypersensitivity to chronic gamma radiation in culturedfibroblasts from AT heterozygotes. In Gatti RA, Swift M (eds):Ataxia Telangiectasia: Genetics, Neuropathology and Immunol-ogy of a Degenerative Disease of Childhood. Krog FoundationSeries 19:73–87.

Paterson MC, Gentner NE, Middlesdadt MV, Mizayans R, WeinfeldM (1985b) Hereditary and familial disorders linking cancerproneness with abnormal carcinogen response and faulty DNAmetabolism. In Castellani A (ed): Epidemiology and Quan-titation of Environmental Risks in Humans from Radiation andOther Agents: Potential and Limitations. NATO AdvancedStudy Institute, New York: Plenum, pp. 235–265.

Recio L, Cochrane J, Simpson D, Skopek TR, O’Neill JP, NicklasJA, Albertini RJ (1990) DNA sequence analysis of in vivo hprtmutation in human T-lymphocytes. Mutagenesis 5:505–510.

Rossi AM, Tates AD, van Zeeland AA, Vrieling H (1992) Molecu-lar analysis of mutations affecting hprt mRNA splicing in hu-man T-lymphocytes in vivo. Env Mol Mutat 19:7–13.

Spitz MR (1995) Risk factors and genetic susceptibility. Cancer TreatRes 74:73–87.

Steingrimsdottir H, Rowley G, Dorado G, Cole J, Lehmann AR(1992) Mutations which alter splicing in the human hypoxan-thine-guanine phosphoribosyltransferase gene. Nucleic AcidsRes 20:1201–1208.

Steingrimsdottir H, Rowley G, Waugh A, Beare D, Cole J, LehmannAR (1993) Molecular analysis of mutations in the hprt gene incirculating lymphocytes from normal and DNA-repair-deficientdonors. Mutat Res 294:29–41.

Tates AD, van Dam FJ, van Mossel H, Schoemaker H, ThijssenJCP, Woldring VM, Zwinderman AH, Natarajan AT (1991) Useof the clonal assay for the measurement of frequencies of HPRTmutants in T-lymphocytes from five control populations. MutatRes 253:199–213.

Yang TP, Patel PI, Chinault AC, Stout JT, Jackson LG, HildebrandBM, Caskey CT (1984) Molecular evidence for new muta-tion at the hprt locus in Lesch-Nyhan patients. Nature310:412–414.