Sex of Affected Sibpairs and Genetic Linkage to Type 1 Diabetes

Transcription

1 American Journal of Medical Genetics 84:15 19 (1999) Sex of Affected Sibpairs and Genetic Linkage to Type 1 Diabetes Andrew D. Paterson* and Arturas Petronis Neurogenetics Section, Clarke Division, Centre for Addiction and Mental Health, Toronto, Ontario, Canada A mouse model of diabetes shows gender dimorphism in the cumulative incidence of diabetes. Based on this, evidence for genetic linkage to IDDM13 on chromosome arm 2q was reported to be greater in type 1 diabetes families where there was a predominance of affected female offspring compared with families with a predominance of affected male offspring. Our objective was to investigate whether the sex of affected offspring affects evidence for linkage to susceptibility loci. Data from a genome scan of 356 affected sibpair families with type 1 diabetes were analysed to determine if there is differential evidence for linkage in families with affected children of a particular sex. At markers on chromosomes 3, 5, 7, 9, 11, and 19, we found a number of regions where the evidence for linkage is greater in families with affected sibpairs of a particular sex. Thus, evidence for linkage in families with affected sibpairs of the same gender suggests the presence of additional susceptibility loci. Several biological explanations are possible for these findings, including X and Y linkage, effects of sex hormones on gene expression, and quasi-linkage between sex chromosomes and autosomes. Am. J. Med. Genet. 84:15 19, Wiley-Liss, Inc. KEY WORDS: Type 1 diabetes; linkage; genome scan; sex effects; quasilinkage INTRODUCTION A recent worldwide survey of epidemiological studies has shown that there are quite marked differences in *Correspondence to: Andrew Paterson, Neurogenetics Section, Clarke Division, Centre for Addiction and Mental Health, 250 College Street, Toronto, Ontario, M5T 1R8, Canada. andrew.paterson@utoronto.ca Received 25 June 1998; Accepted 28 December 1998 the incidence of type I diabetes (IDDM) between males and females in various populations [Karvonen et al., 1997]. Up to age 14 years, 55% of European populations had a male excess in the incidence of IDDM, compared with a female excess in 38% of populations, suggesting that there may be susceptibility loci that confer risk differentially in males and females. Genetic linkage of IDDM to chromosome region 2q31-q33, termed IDDM13, has been found [Morahan et al., 1996]. In the latter study, stronger evidence for linkage to IDDM13 was observed in those families with an excess of affected females compared with those families with an excess of affected males. This finding was observed in the families where the sibpairs did not share HLA haplotypes. In other human studies of the same region of chromosome 2q [Copeman et al., 1995; Nistico et al., 1996], the families were not split on the basis of the sex of affected offspring, leaving it unclear whether the reported sex effect at this region is a true phenomenon or results from a type I error. In another study, no detectable influence of sex of affected siblings at IDDM13 has been observed [Esposito et al., 1998]. In the study by Morahan et al., the rationale for subdividing the families in such a way was based on sex differences in the incidence of diabetes in the nonobese diabetic (NOD) mouse [Makino et al., 1980]. In the inbred parental NOD strain of these mice, 90% of females are affected, compared with 50% of males [Risch et al., 1993]. Similarly, from the backcross F 1 NOD, diabetes developed in 13% of the females compared with 2.9% of males. To study whether additional loci show evidence for linkage in families with affected offspring of a particular sex, we analysed data from 356 affected sibpair families from the United Kingdom. It is worth noting that in the primary analysis of these data [Mein et al., 1998], none of the loci, except HLA, IDDM10, and a locus on chromosome 16q, met the suggested significant linkage criteria of Lander and Kruglyak [1995]. It may be that for many loci for complex traits, the sample sizes required to reach such stringent levels of significance are not practically feasible. Therefore, the analysis of biologically based genetic linkage heterogeneity may be particularly valuable in revealing loci of susceptibility Wiley-Liss, Inc.

2 16 Paterson and Petronis METHODS Pedigree and genotyping data for 351 markers from the autosomes for 356 affected sibpair families were used. These data were derived from 93 families from the original UK96 genome scan [Davies et al., 1994] as well as 263 UK families [Mein et al., 1998] and were obtained from data made available by Dr. Todd ( diesel.cimr.cam.ac.uk/todd/). Pedigrees were divided into those where the affected sibs were either both male (MM) or both female (FF). In the total sample there were 103 MM and 82 FF families. Multipoint nonparametric linkage (NPLall) analysis was performed after the families were grouped into MM and FF using GENEHUNTER v1.2 [Kruglyak et al., 1996; Kruglyak and Lander 1998]. Alleles had been scored uniquely in each family using a four-allele system, numbering the alleles observed in each family 1 4. Since estimates of allele frequencies are required for multipoint linkage analysis, we used 1/n for the frequency of each allele at a marker with n alleles. The marker order and intermarker recombination fractions provided by Dr Todd s Web site were used. Also, for the markers that produced multipoint NPL P < 0.01 in either sex, we assessed whether the results might be due to misspecification of allele frequencies. For this assessment, allele-sharing identity by descent (IBD) was obtained using the sib_ibd option of ASPEX v1.17 [Hinds and Risch, 1996] in the MM and FF families, and differences in %IBD shared between MM and FF families was compared using a 2 2 contingency table. RESULTS Multipoint NPL scores for selected chromosomes are plotted in Fig. 1. The exact values for the peak markers from these chromosomes are presented in Table I, and it can be seen that on chromosomes 3, 5, 7, 11, and 19 there are differences in the MM and FF families. Similar linkage of MM and FF families to the HLA at chromosome region 6p21 was observed and is not discussed further. D3S1279 produced mild excess sharing in the previous analysis of this data (MLS 1.1) [Mein et al., 1994], but here we demonstrate that the predominant evidence for linkage derives from FF families. Similarly, near D5S406, previous analysis provided weak evidence for linkage (MLS 1.2) [Mein et al., 1998], while here it appears that most of the evidence for linkage derives from MM families. No previous linkage to the region near D7S527 has been reported, but here we observed evidence for linkage in MM pairs. At the TH/ INS locus, it is interesting that the evidence for linkage arising from MM pedigrees is more than twice that from FF pedigrees. This region has been studied in detail regarding parental origin of alleles [Bennett et al., 1995; 1997], however, no sex effect has previously been investigated. It would be interesting to study transmission disequilibrium at the INS locus, to determine if there is a difference in transmission of maternal alleles to affected males and females; however, this is not possible using the current data, since the alleles are not coded consistently across families. The region flanking D19S221 has previously produced an MLS 1.7 in these data, and here we show that linkage in this region derives more from FF than MM families. The IBD results at these markers (Table II) provide support that the multipoint results are not due to misspecification of allele frequencies. In particular, at D3S1279 significant greater sharing in FF families compared with MM families was observed ( , 1 df, P ). At D19S221, only a small number of meioses were informative, and thus results at this marker have to be treated with caution. DISCUSSION A posteriori tests of linkage heterogeneity have been particularly helpful in defining subgroups of families with genetic disorders that demonstrate the strongest evidence for linkage: early-onset breast cancer and BRCA1 and early-onset familial Alzheimer disease and PS1. The approach used here is based on the a priori hypothesis that linkage to certain loci may be predominant in families with affected offspring of a particular sex. Clearly, none of these loci reach the stringent significance criteria proposed elsewhere for linkage [Lander and Kruglyak, 1995]; however, the linkage of FF pairs to D3S1279 with P reaches the criterion for suggestive linkage. It is important to note that the earlier study of chromosome arm 2q [Morahan et al., 1996] included 22 ASP families from the British Diabetes Association Warren Repositry, which may overlap with families included in the UK96. This previous study [Morahan et al., 1996] found stronger evidence for linkage to IDDM13, a region flanked by D2S143 D2S164, in sibpairs that did not share HLA markers and in which there were more affected females. Our analysis does not reveal a significant sex effect on chromosome 2, and this is concordant with the results of other researchers [Esposito et al., 1998], who have also failed to find a sex effect, suggesting that the finding by Morahan et al. was due to type 1 error. Since this is an exploratory analysis, we have not corrected for multiple statistical tests. It is noteworthy that heterogeneity analyses are already commonplace in many studies of IDDM, with the division of the sample into those families where the affected sib pair share and do not share HLA markers. This division is usually performed without correcting for multiple statistical tests. It is quite possible that the evidence we report for linkage in families with affected offspring of the same sex results from type I errors. It must be remembered, however, that the effect size of the evidence for linkage at some of the loci highlighted in Results is of a magnitude similar to that seen at the insulin gene locus (IDDM2). Only one study has provided weak evidence of linkage to IDDM2 [Davies et al., 1994], although evidence for linkage disequilibrium with IDDM is very strong [Bennett et al., 1997]. Nonetheless, until additional families have been analysed in the same fashion as undertaken here, no firm conclusions can be drawn. It is important to note that at the regions that we highlighted as showing some differences between the MM and FF families, the information content in the families showing no evidence for

3 Sex and Genetic Linkage of Type 1 Diabetes 17 Fig. 1. Nonparametric linkage scores plotted for IDDM families with two affected brothers (thick line) and two affected sisters (thin line) against distance in centimeters from the most pter marker for chromosomes 3 (a), 5 (b), 7 (c), 11 (d), and 19 (e). linkage was very similar to that in the families that show some evidence for linkage. Genetic linkage studies of the NOD mice [Todd et al., 1991; Risch et al., 1993; Ghosh et al., 1994] have generally concentrated on female mice. Male NOD mice have usually not been studied, although there is one exception to our knowledge, when males were used for the fine mapping of a chromosome 3 locus in the NOD mouse [Prins et al., 1993]. The cumulative incidence in NOD.D3Nds7 D3Nds8 males was 27% of the male NOD rate, whereas in the NOD.D3Nds7 D3Nds8 females it was 45% of the female NOD value. Although linkage analysis of this region was not performed for males and females separately, this locus may account

4 18 Paterson and Petronis TABLE I. Peak NPL Scores for Chromosomal Regions Showing Evidence for Linkage to IDDM in Either MM or FF Families Distance for pter MM* FF* C# Marker marker (cm) NPL P Info NPL P Info 3 D3S D5S D7S TH/INS D19S NPL, nonparametric linkage; IDDM, type I diabetes; MM, both male siblings; FF, both female siblings; C#, chromosome number; Info, Information content as assessed by GENEHUNTER. *P < 0.01 in either MM or FF families. for a greater proportion of the susceptibility in females than males. Several molecular findings support sexual dimorphism of susceptibility to diabetes in the NOD mouse. Gender-specific expression of the 65-kDa isoform of glutamic acid decarboxylase (GAD65) has been observed in the NOD mouse, and this molecule has been implicated as an autoantigen in type 1 diabetes [Pleau et al., 1996]. Furthermore, the serum level of tumor necrosis factor alpha in mice after stimulation with a streptococcal preparation is significantly lower in female NOD mice than in males and is correlated with risk for diabetes [Setoguchi et al., 1992]. Also, sexual dimorphism has been demonstrated in the stress response of NOD mice, with males responding less than females after exposure to repeated immobilization stress [Fitzpatrick et al., 1992] Studies in experimental animal models of a number of other complex traits have identified loci that demonstrate evidence for linkage to affected offspring of only one gender. Examples include hypertension [Clark et al., 1996], stroke [Rubattu et al., 1996], alcohol preference [Melo et al., 1996], obesity [Taylor and Phillips; 1997; Kovács et al., 1998], sensitivity to hotplate nociception [Mogil et al., 1997a], and non-opioid stress-induced analgesia [Mogil et al., 1997b]. The observed differences in the evidence for linkage of loci between males and females would argue that this effect has a genetic basis, although this effect may be mediated by differential effects of hormones. No evidence for linkage (MLS > 1.5) to the X chromosome was detected in an analysis that included the families studied as well as additional families [Cucca et al., 1998]. In a subgroup of IDDM families, where the sibpairs share DR3 haplotypes, evidence for linkage to Xp13-p11 was provided; however, this sample contains an excess of males, and the region of linkage includes a male-specific transmission distorter [Naumova et al., 1998]. Thus, exclusion of transmission ratio distortion is required in regions that exhibit linkage in sibpairs of one sex but not in the other sex. Moreover, we cannot exclude that the Y chromosome harbors a susceptibility locus. Gender effects of sex-specific hormones on gene transcription have been reported [e.g., Lauber et al., 1991; Roselli et al., 1997]. One more mechanism by which sex-specific genetic effects might be mediated is the phenomenon of quasilinkage, which refers to the nonrandom assortment of genes located on different chromosomes. It is possible that quasi-linkage exists between some autosomes and the sex chromosomes and results in different evidence for linkage in males and females. Epistatic interaction of loci that produce transmission ratio distortion on the sex chromosomes and autosomes may be responsible for the phenomenon of quasi-linkage. Although quasilinkage has been studied in Drosophila [Douglas et al., 1968], viruses [Stockert et al., 1976], and plants [reviewed in Korol et al., 1994], the phenomenon of quasilinkage has not been investigated in humans [Miké, 1977]. ACKNOWLEDGEMENTS We thank Cathy Spegg for assistance and Dr. John Todd for making his data available. A.D.P. is an MRC Canada Fellow and A.P. is an OMHF New Investigator and NARSAD Young Investigator. TABLE II. Comparison of IBD sharing in MM and FF families at markers with apparent sex-differences in linkage* MM FF MM vs. FF C# Marker IBD1 IBD0 %IBD IBD1 IBD0 %IBD 2 P 3 D3S D5S D7S TH/INS D19S *IBD, identity by descent; MM, both male siblings; FF, both female siblings; C#, chromosome number; IBD1, number of alleles shared identical by descent, IBD0, number of alleles not shared identical by descent; %IBD, proportion of alleles shared identical by descent. MM vs. FF a2 2contingency table to test for significant differences in MM and FF sharing. When %IBD was less than 50% in one group, it was made equal to 50% for this analysis ( 2,1df).

5 Sex and Genetic Linkage of Type 1 Diabetes 19 REFERENCES Bennett ST, Lucassen AM, Gough SC, Powell EE, Undlien DE, Pritchard LE, Merriman ME, Kawaguchi Y, Dronsfield MJ, Pociot F, Nerup J, Bouzekri N, Cambon-Thomsen, Ronningen KS, Barnett AH, Bain SC, Todd JA Susceptibility to human type 1 diabetes at IDDM2 is determined by tandem repeat variation at the insulin gene minisatellite locus. Nat Genet 9: Bennett ST, Wilson AJ, Esposito L, Bouzekri N, Undlien DE, Cucca F, Nistico L, Buzzetti R, the IMDIAB Group, Bosi E, Pociot F, Nerup J, Cambon-Thomsen A, Pugliese A, Shield JP, McKinney PA, Bain SC, Polychronakos C, Todd JA Insulin VNTR allele-specific effect in type 1 diabetes depends on identity of untransmitted paternal allele. Nat Genet 17: Clark JS, Jeffs B, Davidson AO, Lee WK, Anderson NH, Bihoreau MT, Brosnan MJ, Devlin AM, Kelman AW, Lindpaintner K, Dominiczak AF Quantitative trait loci in genetically hypertensive rats: possible sex specificity. Hypertension 28: Copeman JB, Cucca F, Hearne CM, Cornall RJ, Reed PW, Ronningen KS, Undlien DE, Nistico L, Buzzetti R, Tosi R, Pociot F, Nerup J, Cornelis F, Barnett AH, Bain SC, Todd JA Linkage disequilibrium mapping of a type 1 diabetes susceptibility gene (IDDM7) to chromosome 2q31-q33. Nat Genet 9: Cucca F, Goy JV, Kawaguchi Y, Esposito L, Merriman ME, Wilson AJ, Cordell HJ, Bain SC, Todd JA A male-female bias in type 1 diabetes and linkage to chromosome Xp in MHC HLA-DR3-positive patients. Nat Genet 19: Davies JL, Kawaguchi Y, Bennett ST, Copeman JB, Cordell HJ, Pritchard LE, Reed PW, Gough SC, Jenkins SC, Palmer SM, Balfour KM, Rowe BR, Farrall M, Barnett AH, Bain SC, Todd JA A genome-wide search for human type 1 diabetes susceptibility genes. Nature 371: Douglas LT, Geerts SJ, Oomen N A stock of Drosophila melanogaster evidently showing drive and quasi-linkage in females. Genetica 39: Esposito L, Hill NJ, Pritchard LE, Cucca F, Muxworthy C, Merriman ME, Wilson A, Julier C, Delepine M, Tuomilehto J, Tuomilehto-Wolf E, Ionesco-Tirgoviste C, Nistico L, the IMDIAB Group, Buzzetti R, Pozzilli P, Ferrari M, Bosi E, Pociot F, Nerup J, Bain SC, Todd JA Genetic analysis of chromosome 2 in type 1 diabetes: analysis of putative loci IDDM7, IDDM12, and IDDM13 and candidate genes NRAMP1 and IA-2 and the interleukin-1 gene cluster. Diabetes 47: Fitzpatrick F, Christeff N, Durant S, Dardenne M, Nunez EA, Homo- Delarche F Glucocorticoids in the nonobese diabetic (NOD) mouse: basal serum levels, effect of endocrine manipulation and immobilization stress. Life Sci 50: Ghosh S, Palmer SM, Rodrigues NR, Cordell HJ, Hearne CM, Cornall RJ, Prins JB, McShane P, Lathrop GM, Peterson LB, Wicker LS, Todd JA Polygenic control of autoimmune diabetes in nonobese diabetic mice. Nat Genet 4: Hinds DA, Risch N The ASPEX package: affected sib-pair exclusion mapping. ftp://lahmed.stanford.edu/aspex. Karvonen M, Pitkäniemi M, Pitkäniemi J, Kohtamäki K, Tajima N, Tuomilehto J, and the World Health Organization DIAMOND Project Group Sex difference in the incidence of insulin-dependent diabetes mellitus: an analysis of the recent epidemiological data. Diabetes Metab Rev 13: Korol AB, Preygel IA, Preygel SI Recombination variability and evolution: algorithms of estimation and population-genetic models. London: Chapman & Hall. p Kovács P, Voigt B, Klöting I Congenic strain confirms putative quantitative trait locus for body weight in the rat. Mamm Genome 9: Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet 58: Kruglyak L, Lander ES Faster multipoint linkage analysis using Fourier transforms. J Comput Biol 5:1 7. Lander E, Kruglyak L Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 11: Lauber AH, Romano GJ, Pfaff DW Sex difference in estradiol regulation of progestin receptor mrna in rat mediobasal hypothalamus as demonstrated by in situ hybridization. Neuroendocrinology 53: Makino S, Kunimoto K, Muraoka Y, Mizushima Y, Katagiri K, Tochino Y Breeding of a non-obese, diabetic strain of mice. Jikken Dobutsu 29:1 13. Mein CA, Esposito L, Dunn MG, Johnson GC, Timms AE, Goy JV, Smith AN, Sebag-Montefiore L, Merriman ME, Wilson AJ, Pritchard LE, Cucca F, Barnett AH, Bain SC, Todd JA A search for type 1 diabetes susceptibility genes in families from the United Kingdom. Nat Genet 19: Melo JA, Shendure J, Pociask K, Silver LM Identification of sexspecific quantitative trait loci controlling alcohol preference in C57BL/ 6 mice. Nat Genet 13: Miké V Theories of quasi-linkage and affinity : some implications for population structure. Proc Natl Acad Sci U S A 74: Mogil JS, Richards SP, O Toole LA, Helms ML, Mitchell SR, Belknap JK. 1997a. Genetic sensitivity to hot-plate nociception in DBA/2J and C57BL/6J inbred mouse strains: possible sex-specific mediation by delta2-opioid receptors. Pain 70: Mogil JS, Richards SP, O Toole LA, Helms ML, Mitchell SR, Kest B, Belknap JK.1997b. Identification of a sex-specific quantitative trait locus mediating nonopioid stress-induced analgesia in female mice. J Neurosci 17: Morahan G, Huang D, Tait BD, Colman PG, Harrison LC Markers on distal chromosome 2q linked to insulin-dependent diabetes mellitus. Science 272: Naumova AK, Leppert M, Barker DF, Morgan K, Sapienza C Parental origin-dependent, male offspring specific transmission-ratio distortion at loci on the human X chromosome. Am J Hum Genet 62: Nistico L, Buzzetti R, Pritchard LE, Van der Auwera B, Giovannini C, Bosi E, Larrad MT, Rios MS, Chow CC, Cockram CS, Jacobs K, Mijovic C, Bain SC, Barnett AH, Vandewalle CL, Schuit F, Gorus FK, Belgian Diabetes Registry, Tosi R, Pozzilli P, Todd JA The CTLA-4 gene region of chromosome 2q33 is linked to, and associated with, type 1 diabetes. Hum Mol Genet 5: Pleau JM, Esling A, Bach JF, Dardenne M Gene expression of pancreatic glutamic acid decarboxylase in the nonobese diabetic mouse. Biochem Biophys Res Commun 220: Prins JB, Todd JA, Rodrigues NR, Ghosh S, Hogarth PM, Wicker LS, Gaffney E, Podolin PL, Fischer PA, Sirotina A, Peterson LB Linkage on chromosome 3 of autoimmune diabetes and defective Fc receptor for IgG in NOD mice. Science 260: Risch N, Ghosh S, Todd JA Statistical evaluation of multiple-locus linkage data in experimental species and its relevance to human studies: application to nonobese diabetic (NOD) mouse and human insulindependent diabetes mellitus (IDDM). Am J Hum Genet 53: Roselli CE, Abdelgadir SE, Resko JA Regulation of aromatase gene expression in the adult rat brain. Brain Res Bull 44: Rubattu S, Volpe M, Kreutz R, Ganten U, Ganten D, Lindpaintner K Chromosomal mapping of quantitative trait loci contributing to stroke in a rat model of complex human disease. Nat Genet 13: Setoguchi J, Hashiramoto K, Nakano K, Kitagawa Y, Nakamura N, Kanatsuna T, Kondo M, Uno K Prediction of insulin dependent diabetes mellitus in non-obese diabetic mice by the endogeneous tumor necrosis factor-alpha level. Diabetes Res 19: Stockert E, Boyse EA, Sato H, Itakura K Heredity of the GIX thymocyte antigen associated with murine leukemia virus: segregation data simulating genetic linkage. Proc Natl Acad Sci U S A 73: Taylor BA, Phillips SJ Obesity QTLs on mouse chromosomes 2 and 17. Genomics 43: Todd JA, Aitman TJ, Cornall RJ, Ghosh S, Hall JR, Hearne CM, Knight AM, Love JM, McAleer MA, Prins JB, Rodrigues N, Lathrop M, Pressey A, DeLarato NH, Peterson LB, Wicker LS Genetic analysis of autoimmune type 1 diabetes mellitus in mice. Nature 351: