Introduction

The current etiological classification defines type 1 diabetes as a chronic hyperglycemia due to a cellular mediated immune destruction of the insulin-secreting pancreatic beta-cells. This disease is characterized by the presence at the onset of antibodies against insular molecules (islet cell antibodies (ICA), anti-insulin (IAA), anti-glutamic acid decarboxylase (GADA) and anti-tyrosine phosphatase (IA-2) and by a susceptible genotype at the HLA class II DRB1 and DQB1 genes. There is no cure and diabetic subjects require lifetime daily multiple injections of insulin to maintain glucose homeostasis.

Type 1 diabetes occurs in both sexes, but a slight male excess has been found in some populations.¹ Incidence data are available for children up to 14 years of age in 50 countries of the five continents and show a striking variation among populations and within the population of the same country. The age-adjusted incidence rates range from 0.1/100,000 per year in China and Venezuela to ˜37/100,000 per year in Sardinia and Finland.¹

It is widely accepted that type 1 diabetes results from interaction of a polygenic trait with environmental risk factors. This hypothesis has received support from different observations. Approximately one-tenth of the cases occurs in families and identical twins (that share the same genome, except for mitochondrial DNA, post-zygotic and epigenetic DNA changes) are more often concordant for the disease than dizygotic twins (that share half genome) thus suggesting the existence of inherited factors.^2–4 On the other hand, concordance is below 100% in identical twins indicating that a susceptible genome is not sufficient to develop the disease. Moreover, it is also likely that identical twins share more of their environment than non-identical twins. The contribution of genetic susceptibility has been underlined by assessment of the incidence of type 1 diabetes in children of migrants from high incidence to low incidence regions. Two studies have demonstrated that children of Sardinian heritage born and resident in Lazio and Lombardy (two Italian regions with incidence ˜8/100,000 per year) have the same fourfold higher incidence as the population of origin. Moreover, children born in continental Italy of mixed couples with one Sardinian partner have an intermediate rate between those of Sardinia and of continental Italy.^5,6

It is known that the major genes responsible for approximately 40% of genetic susceptibility are within the HLA region on chromosome 6p21 (insulin-dependent diabetes mellitus 1, IDDM1). This region includes highly polymorphic genes in tight linkage disequilibrium that encode proteins processing and presenting antigens to T lymphocytes. Initially, by case-control studies, it was shown that HLA class I alleles were increased in type 1 diabetic subjects.⁷ Subsequently, the primary role of HLA class II loci has been assessed.⁸ Alleles associated with type 1 diabetes are different among populations and ethnic groups. Typically, most Caucasian type 1 diabetics have DRB1*03-DQB1*0201 and/or DRB1*04-DQB1*0302 haplotypes. Trans-racial comparison of disease associated alleles and haplotypes helps to dissect the complexity of HLA-diabetes association: a susceptible HLA genotype requires predisposing alleles at DRB1, DQB1 and DQA1 loci, there is a scale of allelic association from susceptibility through neutrality up to protection and the population frequencies of susceptible and protective haplotypes may explain the population-specific patterns of incidence and of HLA association.^9,10 Moreover, other loci within classes II and III of HLA seem to increase type 1 diabetes risk.^11–14

The existence of non HLA-genes was suspected by genetic analysis of animal models of type 1 diabetes. The search was encouraged by technological advances, availability of a dense map of polymorphic genetic markers and of a large numbers of families with multiple affected siblings. Some whole genome linkage analyses have been conducted in the past decade that have led to the localization of several chromosome regions linked to type 1 diabetes.^15–18 Some researchers have instead adopted a more selective approach or have directly tested genes theoretically implicated in the pathogenesis of the disease. A list of loci so far mapped is reported in Table 1. In this chapter we will describe in detail the discovery of the IDDM12 (CTLA-4) locus. Results obtained in the different studies are not always overlapping and not every linkage or association has been independently replicated. The reasons for inconsistent results are discussed below. A likely explanation is the genetic heterogeneity of the disease: combinations of different genes may result in an identical phenotype. Efforts to identify the etiologic gene variations mapping in the non-HLA IDDM loci are currently ongoing.^43–45

Table 1

Loci and genes linked to or associated with type 1 diabetes.

Exposure to environmental agents as factors initiating or precipitating beta-cell destruction has also been considered. The average daily energy intake of food items of animal origin (especially meat and diary products)⁴⁶ and the consumption in childhood⁴⁷ or by the mother during pregnancy⁴⁸ of foods with nitrosamine additives, that are toxic for beta-cells, are positively correlated with the disease incidence. Also the early introduction of cow's milk, due to short duration of breast feeding, increases the risk for childhood type 1 diabetes.⁴⁹ The underlying hypothesis is that the bovine beta-casein, whose sequence differs from the human protein, may pass through the immature mucosa of the gut and trigger a cellular and humoral immune response cross-reacting with a beta-cell antigen.⁵⁰

Table 3

CTLA-4 association with type 1 diabetes: case-control studies.

Viral infections (rubella, coxsackie, and cytomegalovirus), contracted during fetal life, have been associated with increased risk of type 1 diabetes.^51–53 The discovery of similarities in amino acid sequences of viral proteins and beta-cell components (insulin, glutamic acid decarboxilase and other unidentified proteins) has supported the hypothesis of molecular mimicry.⁵⁴ A mechanism of direct cytotoxicity may also involved.

The Discovery of the CTLA-4 Association with Type 1 Diabetes

It all began in the early '90s when we joined Roberto Tosi at the CNR in Rome who was planning to perform an identical-by-descent linkage analysis of type 1 diabetes in Italian affected sib-pair families. We chose to study a dozen chromosome regions that either contained genes involved in autoimmune response or were orthologous to murine or rat genome regions where susceptibility loci to autoimmune diabetes had been mapped. One region was the long arm of human chromosome 2 that corresponds to murine chromosome 1, where a locus for periinsulitis in non-obese diabetic (NOD) mouse, named Idd5, had been reported.^55,56

Initially, we tested microsatellite markers mapping in 2q12-q22 and in 2q33-35. The latter region gave evidence of increased allele sharing in our Italian sib-pair families and we presented these preliminary results in scientific meetings from 1994.^57–60 Detection of linkage indicates that a disease gene should lie in a rather extended region (several centiMorgans) around the positive markers. We were lucky because microsatellites linked in our data set included an (AT) repeat in the 3' untranslated (UTR) sequence of the CTLA-4 gene. CTLA-4 had been identified in 1987 by Brunet et al⁶¹ as a transcript of cytotoxic T cells and at the time of our observation its role in regulating T cell activation was still controversial. It was also known that CD28, another T cell co-stimulatory molecule structurally homologous to CTLA-4, mapped close to CTLA-4 on chromosome 2.^62,63 Thus, it was clear to us that CTLA-4 and CD28 deserved to be considered type 1 diabetes candidate genes.

Following a suggestion by John A. Todd, we tested the CTLA-4 microsatellite for association in the presence of linkage with the disease by the transmission disequilibrium test (TDT). The second most frequent allele, that we named 104 mobility units (mu), showed increased transmission, albeit not significant, to diabetic offspring in Italian multiplex families (20 transmissions versus 10 non transmissions, p=0.07). We felt encouraged and undertook the collection of Italian “simplex” families (one diabetic child, his/her parents and a non-affected child, if available), a more abundant source suitable for TDT.

Searching the Genbank we found that a CTLA-4 sequence submission (accession number L15006) reported three “conflicts”, i.e., potential single base polymorphisms, at positions 49, 272 and 439. Only the A49G variation, that codes for a threonine to alanine change in the signal peptide, turned out to be a true allelic difference (frequency of G allele in Italian population is approximately 30%). Preliminary analysis on 104 Italian type 1 diabetes families showed that the G49 allele, that is in tight linkage disequilibrium (LD) with the 104 mu allele, was significantly more transmitted to affected children (p<0.025).⁶⁴ We continued collecting and typing Italian families and established collaboration with foreign groups: 44 Spanish type 1 diabetic families were tested for the A49G single nucleotide polymorphism (SNP). The results (40 transmissions versus 18 non-transmissions to all affected siblings, p=0.004, 21 transmissions and 25 non-transmissions to non-affected children) replicated the data obtained in Italian families. In the meantime, an article came out that showed a significant increase of the second most frequent allele of the CTLA-4 microsatellite (the Authors named this allele 106 base-pairs (bp)) in Caucasian patients affected by Graves' disease.⁶⁵ This independent observation strengthened the hypothesis that a gene controlling a step in an autoimmune process was in LD with CTLA-4, if not CTLA-4 itself.

In collaboration with Todd's group, the A49G SNP was typed in 284 UK, 180 US and 123 Sardinian type 1 diabetic families and in a panel of Chinese patients affected by Graves' disease and controls. TDT of G49 allele in diabetic families was not statistically significant, although a trend toward increased transmissions was observed in the US data set (177 transmissions versus 145 non-transmissions, p=0.074). Instead, the Chinese case-control study was significant and in agreement with Italian and Spanish data. We typed 83 additional Italian families and TDT of the G49 allele in the whole Italian family panel (n=187) was significant (114 transmissions and 75 non-transmissions to all affected children, p=0.004). To obtain a valid test of association, we analyzed transmissions to Italian and Spanish probands only (one affected child in multiplex families) and obtained significant data (112 transmissions versus 65 non-transmissions, p=0.0004). Importantly, no transmission distortion was observed to non-diabetic sibs. The overall TDT to affected siblings of the five family data sets was significant (p=0.002).

The Belgian Diabetes Registry, which is a large population-based collection of type 1 diabetic patients, was also analyzed. Belgian patients (n=483) had significantly higher frequencies of G allele, G phenotype and GG genotype compared to controls (n=529). The overall data (three out of six diabetic panels studied were significant and consistent) provided evidence for an association of the CTLA-4 region with type 1 diabetes and the locus was named IDDM12. Results of this collaborative effort were joined and published in 1996.⁶⁶

Later on the UK, US, Italian and Sardinian family collections, previously typed for the A49G variation, plus 185 additional families, were also analyzed for the CTLA-4 microsatellite. The two most frequent alleles showed percentage of transmission significantly different from random expectation: alleles 262 and 280 (the latter corresponds to allele 104 mu or 106 bp) showed decreased and increased transmissions, respectively, to all diabetic children either in the overall data set (allele 262, p=0.009; allele 280, p=0.001) or in the UK and US families combined (allele 262, p=0.001; allele 280, p=0.02).³¹ These data indicate that the CTLA-4 microsatellite is a better disease risk marker than the A49G variation.

Genetic Studies on CTLA-4 and Human Type 1 Diabetes

Since our first observations, many papers testing the hypothesis of CTLA-4 association with different autoimmune diseases in diverse populations followed one another.⁶⁷ To our knowledge, sixteen articles that studied the association with type 1 diabetes have been published up to the year 2000 (Tables 1 and 2). Alternate conclusions have been drawn. There are some explanations for inconsistent results among distinct populations and among different data sets from the same population.

Table 2

CTLA-4 association with type 1 diabetes: family studies.

First, the study should have enough power (that is probability) to detect association at a statistically significant level. Power depends on the relative risk and on the frequency in the general population of the factor tested and on the size of the sample studied (either families or case-controls). Type 1 diabetes genes, apart from HLA, are estimated to confer low risk (i.e., 1.5), thus for an allele frequency ranging between 10% to 50%, 700 to 300 patients and controls should be analyzed (for an 80% power and a significance level of 0.05). Studying data set of inappropriate size reduces the chance of detecting true small effects (false negative).

There may be also population specific factors such as pattern and level of LD between the marker allele tested and the disease-predisposing allele. Population isolates have a higher degree of LD than cosmopolitan populations.⁸³

Finally, locus heterogeneity must be considered: a gene may account only for a fraction of the cases depending on genetic background and on environmental factors. In this respect, some researchers studied CTLA-4 association sub-grouping diabetic patients on the basis of genetic, immune and clinical features. Some conflicting results have been reported but these are probably due to the small number of cases included in the sub-classes and to non-stringent statistic thresholds (p values not corrected for number of comparisons performed). In the largest study of this type the Authors found no difference in the CTLA-4 G49 allele distribution with respect to age of disease onset, HLA-DR, -DQ and insulin genotypes and the presence of islet or thyrogastric autoantibodies.⁷³

Role of CTLA-4 in Human Type 1 Diabetes

There is no direct proof that CTLA-4 is a diabetes susceptibility gene. CTLA-4 gene itself may be primarily involved in the disease pathogenesis if one or more of its polymorphisms determine an effect on the protein function that may be relevant to the disease process. Alternatively, it is possible that CTLA-4 is in LD with the etiologic variation residing somewhere in the region.

Genetic studies on CTLA-4 have focused on three polymorphisms, so far: a C-T SNP in position -318 from the ATG start codon⁸⁴, the A49G in exon 1 and the dinucleotide repeat in the 3'-UTR. Recently, Marron et al⁶⁹ described another SNP in CTLA-4 intron 1 (C/T in position -819 from exon 2-start site) that showed more significant association with type 1 diabetes than the A49G and the microsatellite.

The biological significance of these polymorphisms is not known. The C-318T variation does not affect any known consensus sequence in the regulatory region of the promoter.⁸⁴

As far as the A49G variation is concerned, Kouki et al⁸⁵ demonstrated that proliferative response of T lymphocytes is increased in the presence of soluble blocking anti-human CTLA-4 monoclonal antibody. The augmentation is significantly higher in individuals A/A homozygous compared with G/G subjects regardless of the affection status. This is the first report that shows that CTLA-4 effects can be modulated by its genotype, however the phenotype observed cannot be directly related to exon 1 variation due to its strong LD with the (AT)n polymorphism at 3'-UTR, the C/T SNP described by Marron et al⁶⁹ and, possibly, with others yet unknown.

With regard to the microsatellite, it has been speculated that (AT)n length in 3'-UTR can affect mRNA stability, but this has not been proven for CTLA-4.

We and others^84,86 have searched CTLA-4 exons and part of introns for additional polymorphisms: nothing has been found in coding regions. We have identified two new C-T SNPs in 5' sequence and in intron 1, respectively, but they did not show significant transmission distortion in 200 Italian diabetic families (unpublished).

Sequencing of cDNA from human and rodents T and B lymphocytes revealed a CTLA-4 alternate transcript that lacks the transmembrane domain^87,88 and results in a soluble form (sCTLA-4) detectable in the sera. Oaks et al⁸⁹ have also demonstrated that serum levels of sCTLA-4 are higher in patients with autoimmune thyroid diseases compared to healthy subjects. It is not clear whether this phenomenon is due to DNA differences or to an indirect effect related to the affection status.

Finally, it cannot be ruled out that additional polymorphisms map in flanking and intron sequences of CTLA-4 locus and that these, taken singly or in combination, could be the etiologic variation.

Genetic Studies in Animals

A great deal of information on CTLA-4 function and on its possible role in the pathogenesis of autoimmune diabetes comes from genetic and functional studies in mice.

Animal models of human disorders constitute a powerful tool for investigation of the mechanisms involved in complex diseases since confounders acting in research on human populations are not present in strains. Being inbred, the genetic background is homogeneous and fixed among generations and sibship. Also the environment, which probably plays an important role in the unfolding of human diseases, is controlled and homogeneous for laboratory animals. Moreover, information can be drawn by manipulating the genetic background of the animal and producing extreme phenotypic consequences.

Non-obese diabetic (NOD) mouse is a strain that spontaneously develops autoimmune diabetes with many similarities to the human disorder. The Ctla-4 gene possibly controls insulitis and diabetes in NOD mouse since it maps within the Idd5 diabetes susceptibility locus. Idd5 had been identified in genomic screens conducted on NOD mouse as a large (34 cM) region on proximal chromosome 1 (orthologous to human chromosome 2q) that was linked to insulitis.^55,56 To identify the minimal region containing the susceptibility gene, Hill et al⁹⁰ have recently developed some congenic strains by backcrossing NOD mice (diabetes prone) to a B10 (diabetes resistant) strain and selecting the progeny that has NOD genetic background except for the region of interest on chromosome 1 that is derived from the B10 resistant animals. NOD.B10 Idd5 mice have a lower frequency of diabetes than NOD animals. Additional congenic strains with an even shorter segment of the B10-derived chromosome 1 have then been generated. Comparing diabetes incidence among different strains it was possible to narrow the region(s) containing the gene(s) of resistance to the disease. They have determined that Idd5 is actually a two-gene locus. Idd5.1 is a 1.5 cM region that includes four genes: Casp8, Cflar, Cd28 and Ctla-4 and corresponds to IDDM12 region identified in humans. Idd5.2 contains Nramp1 and Il-8ra (interleukin-8 receptor a). No coding variations have been found between NOD and B10 sequences of Cflar, Cd28 and Ctla-4. A conservative Ala-Val substitution has been found in Casp8, but its role in disease has not been explored. Thus, similarly to what has been seen in humans, it is possible that the etiologic variation resides in non-translated sequences around these genes.

Functional Studies in Mice

The controversy concerning the role of Ctla-4 in the regulation of T cell activation has been resolved by generating Ctla-4 knock-out mice, that are mice in which both copies of the Ctla-4 genes have been disrupted by homologous recombination. These animals have an extensive accumulation of activated lymphocytes in several tissues and organs. Infiltrating cells express activation markers and exhibit high proliferation rates in response to stimuli or spontaneously thus implying that Ctla-4 acts as a negative regulator of T cells.⁹¹

The NOD model of autoimmune diabetes has been exploited to understand how and when Ctla-4 exert its immunoregulatory role.

Luhder et al⁹² demonstrated in a TCR transgenic model of autoimmune diabetes that blockade with anti-Ctla-4 monoclonal antibody (mAb) accelerated the development of diabetes. They also established that Ctla-4 does not act on naive cells, but in a narrow time window, when activated T cells migrate to the pancreatic islets and re-encounter the antigen for a second time.⁹³ This is consistent with the observation that maximal Ctla-4 protein expression is seen 48–72 hours later T cell activation. A sub-population of CD4+ T lymphocytes, that is CD25+ and constitutively expresses Ctla-4, probably exerts the inhibitory role. This T cell subset is reduced in NOD mice compared to other mouse strains. Moreover, if NOD splenocytes depleted of CD4+ CD25+ Ctla-4+ T cells are transferred into NOD.SCID mice an increase in the frequency of diabetes is observed compared to animals that received total splenocytes.⁹⁴ Interestingly, presence of CD4+ CD25+ Ctla-4+ cells requires the integrity of the CD28/B7 system. Regulatory T cells positive for CD25, CD4 and Ctla-4 suppressively control other T cells by two possible mechanisms: they may compete, through Ctla-4, with other T cells for the costimulatory signal on antigen presenting cells. Alternatively, regulatory T cells deliver to other T cells a negative signal of proliferation or activation.⁹⁵

Conclusion

Much detailed information has been gained in the last few years on CTLA-4, and its function is compatible with a role in the pathogenesis of type 1 diabetes and other immune-mediated diseases. However, genetic data neither show nor rule out in a definite and conclusive manner that CTLA-4 and/or the flanking genes CD28 and ICOS are primarily responsible for the observed disease association. Sequencing of the murine and human CTLA-4 genes in a diabetes prone strain and in affected people has demonstrated that the extracellular binding domain and the cytoplasmic tail that delivers the signal are not modified in this disease, thus the explanation must be searched for elsewhere. Possibly, every SNP in the region has to be looked for and genetically tested. In tight conjunction, studies of functional genomics are needed to investigate the effect encoded by the sequence variations of the CTLA-4 locus that are found associated with the disease. Slight modifications of the transcribed or translated product are possibly involved and such evidence may be difficult to detect and discriminate from secondary effects. Thus very sensitive methods and artificial systems may be required to demonstrate a direct relationship between the genotype and the biological effect at the expression/protein level. Correlation of the protein phenotype with the disease outcome is a further step that should take into account the contribution of the other disease genes and of the environmental factors.

Acknowledgements

We are deeply grateful to Roberto Tosi. A special thank goes to John A. Todd, Lynn Pritchard and Francesco Cucca. We wish also to acknowledge the physicians of the IMDIAB group, Maria Teresa Martinez, Manuel Serrano-Rios, Frans Gorus, Bart Van der Auwera and Emanuele Bosi for sharing their samples and results and Richard H. Butler for revising this manuscript.

References

1.

Karvonen M, Viik-Kajander M, Moltchanova E. et al. Incidence of childhood type 1 diabetes worldwide. Diabetes care. 2000;23:1516–1526. [PubMed: 11023146]

2.

Barnett AH, Eff C, Leslie RDG. et al. Diabetes in identical twins: a study of 200 pairs. Diabetologia. 1981;20:87–93. [PubMed: 7193616]

3.

Olmos P, A'Hern R, Heaton DA. et al. The significance of the concordance rate for type 1 (insulin-dependent) diabetes in identical twins. Diabetologia. 1988;31:747–50. [PubMed: 3240835]

4.

Kyvik KO, Green A, Beck-Nielsen H. Concordance rates of insulin dependent diabetes mellitus: a population based study of young Danish twins. BMJ. 1995;311:913–917. [PMC free article: PMC2550917] [PubMed: 7580548]

5.

Muntoni SA, Fonte MT, Stoduto S. et al. Incidence of insulin dependent type 1 diabetes mellitus among Sardinian heritage children born in Lazio region (Italy). Lancet. 1997;349:160–162. [PubMed: 9111539]

6.

Calori G, Gallus G, Bognetti E. et al. Insulin-dependent diabetes mellitus in Sardinian heritage children living in Lombardy. Lancet. 1998;351:263–264. [PubMed: 9457099]

7.

Singal DP, Blajchman MA. Histocompatibility (HL-A) antigens, lymphocytotoxic antibodies and tissue antibodies in patients with diabetes mellitus. Diabetes. 1973;22:429–432. [PubMed: 4541338]

8.

Nerup J, Platz P, Andersen OO. et al. HL-A antigens and diabetes mellitus. Lancet. 1974;2:864–866. [PubMed: 4137711]

9.

Cucca F, Todd JA. HLA susceptibility to type 1 diabetes: methods and mechanisms In: Browning M, McMichael A, eds.HLA and MHC: genes, molecules and function Oxford: Bios Scientific Publishers Ltd,1996.

10.

Undlien DE, Lie BA, Thorsby E. HLA complex genes in type 1 diabetes and other autoimmune diseases. Which genes are involved? Trends Genet. 2001;17:93–100. [PubMed: 11173119]

11.

Erlich HA, Rotter JI, Chang JD. et al. Association of HLA-DPB1*0301 with IDDM in Mexican-Americans. Diabetes. 1996;45:610–614. [PubMed: 8621011]

12.

Noble JA, Valdes AM, Cook M. et al. The role of HLA class II genes in insulin-dependent diabetes mellitus: molecular analysis of 180 Caucasian, multiplex families. Am J Hum Genet. 1996;59:1134–1148. [PMC free article: PMC1914851] [PubMed: 8900244]

13.

Cucca F, Dudbridge F, Loddo M. et al. The HLA-DPB1-associated component of the IDDM1 and its relationship to the major loci HLA-DQB1, -DQA1, and -DRB1. Diabetes. 2001;50:1200–1205. [PubMed: 11334427]

14.

Zavattari P, Lampis R, Motzo C. et al. Conditional linkage disequilibrium analysis of a complex disease superlocus, IDDM1 in the HLA region, reveals the presence of independent modifying gene effects influencing the type 1 diabetes risk encoded by the major HLA-DQB1, -DRB1 disease loci. Hum Mol Genet. 2001;10:881–889. [PubMed: 11285254]

15.

Davies JL, Kawaguchi Y, Bennett ST. et al. A genome-wide search for human type 1 diabetes susceptibility genes. Nature. 1994;371:130–136. [PubMed: 8072542]

16.

Hashimoto L, Habita C, Beressi JP. et al. Genetic mapping of a susceptibility locus for insulin-dependent diabetes mellitus on chromosome 11q. Nature. 1994;371:161–164. [PubMed: 8072544]

17.

Concannon P, Gogolin-Ewens KJ, Hinds DA. et al. A second-generation screen of the human genome for susceptibility to insulin-dependent diabetes mellitus. Nat Genet. 1998;19:292–296. [PubMed: 9662408]

18.

Mein CA, Esposito L, Dunn MG. et al. A search for type 1 diabetes susceptibility genes in families from the United Kingdom. Nat Genet. 1998;19:297–300. [PubMed: 9662409]

19.

Bell GI, Horita S, Karam JH. A polymorphic locus near the human insulin gene is associated with insulin-dependent diabetes mellitus. Diabetes. 1984;33:176–183. [PubMed: 6363172]

20.

Catignani Kennedy G, German MS, Rutter WJ. The minisatellite in the diabetes susceptibility locus IDDM2 regulates insulin transcription. Nat Genet. 1995;9:293–298. [PubMed: 7773292]

21.

Pugliese A, Zeller M, Fernandez A Jr. et al. The insulin gene is transcribed in the human thymus and transcription levels correlated with allelic variation at the INS VNTR-IDDM2 susceptibility locus for type 1 diabetes. Nat Genet. 1997;15:293–297. [PubMed: 9054945]

22.

Vafiadis P, Bennett ST, Todd JA. et al. Insulin expression in human thymus is modulated by INS VNTR alleles at the IDDM2 locus. Nat Genet. 1997;15:289–292. [PubMed: 9054944]

23.

Field LL, Tobias R, Magnus T. A locus on chromosome 15q26 (IDDM3) produces susceptibility to insulin-dependent diabetes mellitus. Nat Genet. 1994;8:189–194. [PubMed: 7842018]

24.

Luo DF, Buzzetti R, Rotter J. et al. Confirmation of three susceptibility genes to insulin-dependent diabetes millitus: IDDM4, IDDM5 and IDDM8. Hum Mol Genet. 1996;5:693–698. [PubMed: 8733139]

25.

Davies JL, Cucca F, Goy JV. et al. Saturation multipoint linkage mapping of chromosome 6q in type 1 diabetes. Hum Molec Genet. 1996;5:1071–1074. [PubMed: 8817350]

26.

Delepine M, Pociot F, Habita C. et al. Evidence of a non-MHC susceptibility locus il type 1 diabetes linked to HLA on chromosome 6. Am J Hum Genet. 1997;60:174–187. [PMC free article: PMC1712534] [PubMed: 8981961]

27.

Merriman T, Twells R, Merriman M. et al. Evidence by allelic association-dependent methods for a type 1 diabetes polygene (IDDM6) on chromosome 18q21. Hum Mol Genet. 1997;6:1003–1010. [PubMed: 9215667]

28.

Merriman TR, Eaves IA, Twells RC. et al. Transmission of haplotypes of microsatellite markers rather than single marker alleles in the mapping of a putative type 1 diabetes susceptibility gene (IDDM6). Hum Mol Genet. 1998;7:517–524. [PubMed: 9467012]

29.

Copeman JB, Cucca F, Hearne CM. et al. Linkage disequilibrium mapping of a type 1 diabetes susceptibility gene (IDDM7) to chromosome 2q31-33. Nat Genet. 1995;9:80–85. [PubMed: 7704030]

30.

Luo DF, Maclaren NK, Huang HS. et al. Intrafamilial and case-control association analysis of D2S152 in insulin-dependent diabetes. Autoimmunity. 1995;21:143–147. [PubMed: 8679904]

31.

Esposito L, Hill NJ, Pritchard LE. et al. 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. 1998;47:1797–1799. [PubMed: 9792551]

32.

Luo DF, Bui MM, Muir A. et al. Affected-sib-pair mapping of a novel susceptibility gene to insulin-dependent diabetes mellitus (IDDM8) on chromosome 6q25-q27. Am J Hum Genet. 1995;57:911–919. [PMC free article: PMC1801491] [PubMed: 7573053]

33.

Reed P, Cucca F, Jenkins S. et al. Evidence for a type 1 diabetes susceptibility locus (IDDM10) on human chromosome 10p11-q11. Hum Mol Genet. 1997;6:1011–1016. [PubMed: 9215668]

34.

Field LL, Tobias R, Thomson G. et al. Susceptibility to insulin-dependent diabetes mellitus maps to a locus (IDDM11) on human chromosome 14q24.3-q31. Genomics. 1996;33:1–8. [PubMed: 8617492]

35.

Morahan G, Huang D, Tait BD. et al. Markers on distal chromosome 2q linked to insulin-dependent diabetes mellitus. Science. 1996;272:1811–1813. [PubMed: 8650584]

36.

Verge CF, Vardi P, Babu S. et al. Evidence for oligogenic inheritance of type 1 diabetes in a large Bedouin Arab family. J Clin Invest. 1998;102:1569–1575. [PMC free article: PMC509007] [PubMed: 9788970]

37.

Morahan G, Huang D, Ymer SI. et al. Linkage disequilibrium of a type 1 diabetes susceptibility locus with a regulatory IL12B allele. Nat Genet. 2001;27:218–221. [PubMed: 11175794]

38.

Rowe RE, Wapelhorst B, Bell GI. et al. Linkage and association between insulin-dependent diabetes mellitus (IDDM) susceptibility and markers near the glucokinase gene on chromosome 7. Nat Genet. 1995;10:240–245. [PubMed: 7663523]

39.

McDermott MF, Ramachandran A, Ogunkolade BW. et al. Allelic variation in the vitamin D receptor influences susceptibility to IDDM in Indian Asians. Diabetologia. 1997;40:971–975. [PubMed: 9267994]

40.

Pani Ma, Knapp M, Donner H. et al. Vitamin D receptor allele combinations influence genetic susceptibility to type 1 diabetes in Germans. Diabetes. 2000;49:504–507. [PubMed: 10868975]

41.

Chang TJ, Lei HH, Yeh JI. et al. Vitamin D receptor gene polymorphisms influence susceptibility to type 1 diabetes mellitus in the Taiwanese population. Clin Endocrinol. 2000;52:575–580. [PubMed: 10792336]

42.

Cucca F, Goy JV, Kawaguchi Y. et al. A male-female bias in type 1 diabetes and linkage to chromosome Xp in MHC HLA-DR3-positive patients. Nat Genet. 1998;19:301–302. [PubMed: 9662410]

43.

Nakagawa Y, Kawaguchi Y, Twells RCJ. et al. Fine mapping of the diabetes-susceptibility locus, IDDM4, on chromosome 11q13. Am J Hum Genet. 1998;63:547–556. [PMC free article: PMC1377315] [PubMed: 9683605]

44.

Eckenrode S, Marron MP, Nicholls R. et al. Fine-mapping of the type 1 diabetes locus (IDDM4) on chromosome11q and evaluation of two candidate genes (FADD and GALN) by affected sibpair and linkage-disequilibrium analyses. Hum Genet. 2000;106:14–18. [PubMed: 10982176]

45.

Owerbach D. Physical and genetic mapping of IDDM8 on chromosome 6q27. Diabetes. 2000;49:508–512. [PubMed: 10868976]

46.

Muntoni SA, Cocco P, Aru G. et al. Nutritional factors and worldwide incidence of childhood type 1 diabetes. Am J Clin Nutr. 2000;71:1525–1529. [PubMed: 10837294]

47.

Dahlquist GG, Blom LG, Persson LA. et al. Dietary factors and the risk of developing insulin dependent diabetes in childhood. BMJ. 1990;300:1302–1306. [PMC free article: PMC1663017] [PubMed: 2369660]

48.

Helgason T, Jonasson MR. Evidence for a food additive as a cause of ketosis-prone diabetes. Lancet. 1981;2:716–20. [PubMed: 6116858]

49.

Borch-Johnsen K, Joner G, Mandrup-Poulsen T. et al. Relation between breast-feeding and incidence rates of insulin-dependent diabetes mellitus. A hypothesis. Lancet. 1984;2:1083–6. [PubMed: 6150150]

50.

Cavallo MG, Fava D, Monetini L. et al. Cell-mediated immune response to beta casein in recent-onset insulin-dependent diabetes: implications for disease pathogenesis. Lancet. 1996;348:926–928. [PubMed: 8843812]

51.

Menser MA, Forrest JM, Bransby RD. Rubella infection and diabetes mellitus. Lancet. 1978;1:57–60. [PubMed: 74564]

52.

Dahlquist GG, Ivarsson S, Lindberg B. et al. Maternal enteroviral infection during pregnancy as a risk factor for childhood IDDM. A population-based case-control study. Diabetes. 1995;44:408–413. [PubMed: 7698508]

53.

Pak CY, Eun HM, McArthur RG. et al. Association of cytomegalovirus infection with autoimmune type 1 diabetes. Lancet. 1988;2:1–4. [PubMed: 2898620]

54.

Kukreja A, Maclaren NK. Current cases in which epitope mimicry is considered as a component cause of autoimmune disease: immune-mediated (type 1) diabetes. Cell Mol Life Sci. 2000;57:534–541. [PubMed: 11130453]

55.

Garchon H-J, Bedossa P, Eloy L. et al. Identification and mapping to chromosome 1 of a susceptibility locus for periinsulitis in non-obese diabetic mice. Nature. 1991;353:260–262. [PubMed: 1896073]

56.

Cornall RJ, Prins J-B, Todd JA. et al. Type 1 diabetes in mice is linked to the interleukin-1 receptor and Lsh/Ity/Bcg genes on chromosome 1. Nature. 1991;353:262–265. [PubMed: 1832743]

57.

Nisticò L, Buzzetti R, Latini L. et al. Linkage analysis in IDDM affected sib-pair from Continental Italy In: Abstract book of the 13th International Immunology and Diabetes Workshop.199482. [PMC free article: PMC225979]

58.

Buzzetti R, Nisticò L, Pozzilli P. et al. CTLA-4 microsatellite identifies a new region on chromosome 2 linked to IDDM. Diabetologia. 1995;38:A29.

59.

Nisticò L, Buzzetti R, Pozzilli P. et al. Identification of a new region linked to IDDM in affected multiplex families from the Mediterranean area. Diabetes. 1995;44(suppl1):17A (abstract 55).

60.

Buzzetti R, Nisticò L, Mesturino C. et al. A novel region of susceptibility to IDDM on chromosome 2. J Endocrinol Invest. 1995;18(suppl to no 5):3 (abstract 3).

61.

Brunet J-F, Denizot F, Luciani M-F. et al. A new member of the immunoglobulin superfamily- CTLA-4. Nature. 1987;328:267–270. [PubMed: 3496540]

62.

Harper K, Balzano C, Rouvier E. et al. CTLA-4 and CD28 activated lymphocyte molecules are closely related in both mouse and human as to sequence, message expression, gene structure, and chromosomal location. J Immunol. 1991;147:1037–1044. [PubMed: 1713603]

63.

Buonavista N, Balzano C, Pontarotti P. et al. Molecular linkage of the human CTLA4 and CD28 Ig-superfamily genes in yeast artificial chromosomes. Genomics. 1992;13:856–861. [PubMed: 1322357]

64.

Nisticò L, Buzzetti R, Mesturino C. et al. Linkage and association of CTLA-4 gene to IDDM in the Italian population. Autoimmunity. 1995;21(1):18 (abstract 070).

65.

Yanagawa T, Hidaka Y, Guimaraes V. et al. CTLA-4 gene polymorphism associated with Graves' disease in a Caucasian population. J Clin Endocrinol Metabol. 1995;80:41–45. [PubMed: 7829637]

66.

Nisticò L, Buzzetti R, Pritchard LA. et al. The CTLA-4 gene region of chromosome 2q33 is linked to, and associated with type 1 diabetes. Hum Molec Genet. 1996;5:1075–1080. [PubMed: 8817351]

67.

Kristiansen OP, Larsen ZM, Pociot F. CTLA-4 in autoimmune diseases-a general susceptibility gene to autoimmunity? Genes Immunity. 2000;1:170–184. [PubMed: 11196709]

68.

Marron MP, Raffel LJ, Garchon HJ. et al. Insulin-dependent diabetes mellitus (IDDM) is associated with CTLA4 polymorphisms in multiple ethnic groups. Hum Molec Genet. 1997;6:1275–1282. [PubMed: 9259273]

69.

Marron MP, Zeidler A, Raffel LJ. et al. Genetic and physical mapping of a type 1 diabetes susceptibility gene (IDDM12) to a 100kb PAC clone containing D2S72-CTLA4-D2S105 on chromosome 2q33. Diabetes. 2000;49:492–499. [PubMed: 10868973]

70.

Donner H, Seidl C, Braun J. et al. CTLA4 gene haplotypes cannot protect from IDDM in the presence of high risk HLA DQ8 or DQ2 alleles in German families. Diabetes. 1998;47:1158–1160. [PubMed: 9648843]

71.

Perez de Nanclarez G, Bilbao JR, Nisticò L. et al. No evidence of association of chromosome 2q with type 1 diabetes in the Basque population. Diabetologia. 1999;42:119–120. [PubMed: 10027592]

72.

Larsen ZM, Kristiansen OP, Mato E. et al. IDDM12 (CTLA4) on 2q33 and IDDM13 on 2q34 in genetic susceptibility to type 1 diabetes (insulin dependent). Autoimmunity. 1999;31:35–42. [PubMed: 10593567]

73.

Van der Auwera BJ, Vandewalle CL, Schuit FC. et al. CTLA-4 gene polymorphism confers susceptibility to insulin dependent diabetes mellitus (IDDM) independently from age and from other genetic or immune disease markers. Clin Exp Immunol. 1997;110:98–103. [PMC free article: PMC1904800] [PubMed: 9353155]

74.

Donner H, Rau H, Walfish PG. et al. CTLA4 alanine -17 confers susceptibility to Graves' disease and to type 1 diabetes. J Clin Endocrinol Metab. 1997;82:143–146. [PubMed: 8989248]

75.

Owerbach D, Naya FJ, Tsai M-J. et al. Analysis of candidate genes for susceptibility to type 1 diabetes. A case-control and family association study of genes on chromosome 2q31-35. Diabetes. 1997;46:1069–1074. [PubMed: 9166681]

76.

Krokowski M, Bodalski J, Machejko P. et al. CTLA-4 gene polymorphism is associated with predisposition to IDDM in a population from central Poland. Diabetes Metab. 1998;24:241–243. [PubMed: 9690057]

77.

Awata T, Kurihara S, Iitaka M. et al. Association of CTLA-4 gene A-G polymorphism (IDDM12 locus) with acute-onset and insulin-depleted IDDM as well as autoimmune thyroid disease (Graves' disease and Hashimoto's thyroiditis) in the Japanese population. Diabetes. 1997;47:128–129. [PubMed: 9421386]

78.

Abe T, Takino H, Yamasaki H. et al. CTLA-4 gene polymorphism correlates with the mode of onset and presence of ICA512 Ab in Japanese type 1 diabetes. Diabetes Res Clin Pract. 1999;46:169–175. [PubMed: 10724097]

79.

Hayashi H, Kusaka I, Nagasaka S. et al. Association of CTLA-4 polymorphism with positive anti-GAD antibody in Japanese subjects with type 1 diabetes mellitus. Clin Endocrinol. 1999;51:793–799. [PubMed: 10619986]

80.

Yanagawa T, Maruyama T, Gomi K. et al. Lack of association between CTLA-4 gene polymorphism and IDDM in Japanese subjects. Autoimmunity. 1999;29:53–56. [PubMed: 10052685]

81.

Lowe RM, Graham J, Sund G. et al. The length of the CTLA-4 microsatellite (AT)n-repeat affects the risk for type 1 diabetes. Autoimmunity. 2000;32:173–180. [PubMed: 11092697]

82.

Lee Y-J, Huang F-Y, Lo F-S. et al. Association of CTLA-4 gene A-G polymorphism with type 1 diabetes in Chinese children. Clin Endocrinol. 2000;52:153–157. [PubMed: 10671941]

83.

Zavattari P, Deidda E, Whalen M. et al. Major factor influencing linkage disequilibrium by analysis of different chromosome regions in distinct populations: demography, chromosome recombination frequency and selection. Hum Molec Genet. 2000;9:2947–2957. [PubMed: 11115838]

84.

Deichman K, Heinzmann A, Bruggenolte E. et al. An Mse I RFLP in the human CTLA4 promoter. Biochem Biophys Res Commun. 1996;225:817–818. [PubMed: 8780695]

85.

Kouki T, Sawai Y, Gardine C. et al. CTLA-4 gene polymorphism at position 49 in exon 1 reduces the inhibitory function of CTLA-4 and contributes to the pathogenesis of Graves' disease. J Immunol. 2000;165:6606–6611. [PubMed: 11086105]

86.

Ligers A, Xu C, Saarinen S. et al. The CTLA-4 gene is associated with multiple sclerosis. J Neuroimmunol. 1999;97:182–190. [PubMed: 10408973]

87.

Magistrelli G, Jeannin P, Herbault N. et al. A soluble form of CTLA-4 generated by alternative splicing is expressed by nonstimulated human T cells. Eur J Immunol. 1999;29:3596–3602. [PubMed: 10556814]

88.

Oaks M, Hallet KM, Todd Penwell R. et al. A native soluble form of CTLA-4. Cell Immunol. 2000;201:144–153. [PubMed: 10831323]

89.

Oaks MK, Hallet KM. Cutting edge: a soluble form of CTLA-4 in patients with autoimmune thyroid disease. J Immunol. 2000;164:5015–5018. [PubMed: 10799854]

90.

Hill N, Lyons PA, Armitage N. et al. NOD Idd5 locus controls insulitis and diabetes and overlaps the orthologous CTLA4/IDDM12 and NRAMP1 locus in humans. Diabetes. 2000;49:1744–1747. [PubMed: 11016460]

91.

Waterhouse P, Penninger JM, Timms E. et al. Lymphoproliferative disorders with early lethality in mice deficient in Ctla-4. Science. 1995;270:985–988. [PubMed: 7481803]

92.

Luhder F, Hoglund P, Allison JP. et al. Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) regulates the unfolding of autoimmune diabetes. J Exp Med. 1998;187:427–432. [PMC free article: PMC2212113] [PubMed: 9449722]

93.

Luhder F, Chambers C, Allison JP. et al. Pinpointing when T cell costimulatory receptor CTLA-4 must be engaged to dampen diabetogenic T cells. Proc Natl Acad Sci USA. 2000;97:12204–12209. [PMC free article: PMC17319] [PubMed: 11035773]

94.

Salomon B, Lenschow DJ, Rhee L. et al. B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes. Immunity. 2000;12:431–440. [PubMed: 10795741]

95.

Takahashi T, Tagami T, Yamazaki S. et al. Immunologic self-tolerance maintained by CD25+CD4+ regulatory T cells constitutively expressing cytotoxic T lymphocyte-associated antigen 4. J Exp Med. 2000;192:303–309. [PMC free article: PMC2193248] [PubMed: 10899917]