CBLB variants in type 1 diabetes and their genetic ... - CiteSeerX

CBLB variants in type 1 diabetes and their genetic interaction with CTLA4 Regine Bergholdt, Camilla Taxvig, Stefanie Eising, Jørn Nerup, and Flemming Pociot1 Steno Diabetes Center, Gentofte, Denmark

Abstract: Type 1 diabetes (T1D) is a multifactorial disease with genetic and environmental components involved. Recent studies of an animal model of T1D, the Komeda diabetes-prone rat, have demonstrated that the Casitas-B-lineage lymphoma b (cblb) gene is a major susceptibility gene in the development of diabetes and other autoimmune features of this rat. As a result of the inhibitory role of Cbl-b in T cell costimulation, dysregulation of Cbl-b may also contribute to autoimmune diseases in man. Different isoforms of Cbl-b exist; we evaluated expression levels of two known transcript variants. Constitutive expression of both isoforms was demonstrated, as well as an increased expression, after cytokine exposure, of an isoform lacking exon 16, suggesting a possible role of this variant in the pathogenesis of autoimmunity. We screened coding regions of the human CBLB gene for mutations in a panel of individuals affected with several autoimmune diseases. Eight single nucleotide polymorphisms (SNPs) were detected. One SNP in exon 12 of the CBLB gene was significantly demonstrated to be associated to T1D in a large Danish T1D family material of 480 families. Evidence for common genetic factors underlying several autoimmune diseases has come from studies of cytotoxic T lymphocyte antigen 4 (CTLA4), which encodes another negatively regulatory molecule in the immune system. Gene-gene interactions probably play substantial roles in T1D susceptibility. We performed stratification of CBLB exon 12 SNP data, according to an established CTLA4 marker, CT60, and evidence for a genetic interaction between the CTLA4 and CBLB genes, involved in the same biological pathway of T cell receptor signaling, was observed. J. Leukoc. Biol. 77: 579 –585; 2005. Key Words: T cell regulation 䡠 mutational screening 䡠 gene-gene interaction

involved. The humal leukocyte antigen region is the major genetic susceptibility region, but in addition, several minor susceptibility loci have been suggested [1]. Knowledge about these genetic and environmental factors is important in understanding mechanisms of disease pathogenesis. Recent studies of a spontaneous animal model of T1D, the Komeda diabetes-prone (KDP) rat, have shown that the major part of the genetic disease susceptibility in the KDP rat is accounted for by two loci, the major histocompatibility complex region on rat chromosome 20 as well as a locus on rat chromosome 11 (kdp1) containing the Casitas-B-lineage lymphoma b (Cblb) gene (also known as Cas-Br-M murine ecotropic retroviral-transforming sequence B) [2, 3]. The KDP rat is characterized by autoimmune destruction of pancreatic ␤ cells, rapid onset of diabetes with no sex difference, and no significant T cell lymphopenia. Lymphocyte infiltration is furthermore seen in other tissues of the rat, such as thyroid gland, kidney, adrenal gland, and pituitary, indicating general autoimmunity [3]. A nonsense mutation in the Cblb gene in the KDP rat (arginine to a stop codon at codon position 455) truncates the protein by removing 484 amino acids, including a proline-rich region and leucine-zipper domain [3]. Transgenic rescue experiments have confirmed that this mutation, which seems specific to the KDP rat, is pathogenic [3]. Cbl-b is a member of a family of ubiquitin-protein ligases, and functional studies in knockout mice have indicated an important role in T cell costimulation. In Cblb-deficient mice, infiltration of lymphocytes in endocrine tissue has been shown to be caused by enhanced T cell activation, and transgenic complementation with wild-type Cblb significantly suppressed development of the KDP phenotype [3, 4]. Thus, Cbl-b is believed to be a negative regulator of autoimmunity [4, 5]. The Cblb gene, therefore, is a major susceptibility gene for T1D in the rat. As a result of the importance of Cbl-b in regulation of autoimmunity, dysregulation of Cbl-b may also contribute to autoimmune diseases in humans. In human T1D, an increased frequency of other autoimmune diseases (especially thyroid diseases) is often seen [6 – 8], thereby resembling the phenotype of the KDP rat and making the human CBLB gene a candidate for conferring susceptibility to human T1D and autoimmunity in general. The CBLB gene maps to human chromosome 3q11-13.1 [9]. No evidence for linkage to the

INTRODUCTION Type 1 diabetes mellitus (T1D; MIM 222100) is caused by an immune-mediated destruction of the insulin-producing ␤ cells in the pancreas. T1D is characterized as a complex genetic disease with multiple genetic and environmental components 0741-5400/05/0077-579 © Society for Leukocyte Biology

1

Correspondence: Steno Diabetes Center, 2, Niels Steensensvej, DK-2820 Gentofte, Denmark. E-mail: [email protected] Received September 17, 2004; accepted November 29, 2004; doi: 10.1189/ jlb.0904524.

Journal of Leukocyte Biology Volume 77, April 2005 579

3q11-13.1 region on chromosome 3 has been found in any of the T1D genome scans [10 –15]. A locus termed IDDM9, in two of the earlier genome scans, although with quite low maximum logarithm of the odds score (LOD; maximum LOD score), has been assigned to chromosome 3q24-25 [11, 12], ⬃50 Mb distal to the CBLB region, not supporting the CBLB gene as an important T1D locus, and the finding was not replicated in the two most recent and larger genome scans [14, 15]. In support of common genetic factors underlying several autoimmune diseases, there is accumulating genetic evidence of contribution of the cytotoxic T lymphocyte antigen 4 (CTLA4) locus on human chromosome 2q33 to several autoimmune diseases (reviewed in refs. [16, 17]). CTLA4 encodes an important negative regulatory molecule in the immune system. Following T cell activation, inhibitory receptors such as CTLA-4 become expressed and may promote inhibitory effects on the immune response. CTLA-4 may counter-regulate the CD28 T cell antigen receptor activation of T cells and has an even stronger affinity for binding to ligands on the antigen-presenting cells, although mechanisms are not fully understood [18]. The aim of the present study was to screen the human CBLB gene for mutations of importance in human autoimmune disease and test these by testing for association to T1D and interaction with CTLA4.

Semiquantitative reverse transcriptase-polymerase chain reactions (RT-PCRs) Total RNA was isolated using a TRIzol method, according to the manufacturer’s instructions of the RNAzol method (Gibco, Invitrogen) with minor modifications [21]. cDNA synthesis was undertaken by oligo-dT-primed RT of total RNA, as described by the manufacturer (TaqMan RT reagents, Applied Biosystems, Foster City, CA). Quantitative PCR, measuring CBLB mRNAs with and without cytokine stimulation, was carried out using the following primers: 5⬘ atgctgaatggaacacatgg 3⬘ and 5⬘ actatgccttgcaggaggtg 3⬘, located in exon 15 and 17, respectively, to detect transcripts with and without exon 16. The housekeeping gene tata-binding protein (TBP) was used as an internal standard for quantification, comparing transcript expression, as it was stable under the cell-stimulation conditions used. The TBP primers were 5⬘ gccagcttcggagagttctg 3⬘ and 5⬘ tgaaaatcagtgccgtggtt 3⬘. These primers were included in the same PCR mixture, which contained 7.5 ng cDNA as template and 1⫻ polymerase buffer, 50 ␮M deoxy-unspecified nucleoside 5⬘-triphosphate, 1.5 mM MgCl2, 0.6 ␮M (CBLB primers), and 0.2 ␮M (TBP primers) of each primer, 1.0 unit Taq DNA polymerase (Gibco, Invitrogen), 0.11 Mbeq ␣33P cytidine 5⬘-triphosphate, and water in a total volume of 25 ␮l. The following protocol was used: initial denaturing for 10 min at 95°, followed by denaturing, annealing, and extension temperatures of 95°, 63°, and 72°, each for 30 s and repeated in 33 cycles, followed by a 20-min final extension at 72°. Samples were diluted in Stop buffer and denatured for 3 min at 95° before being loaded on a preheated 6% polyacrylamide gel and run for 2 h at 65 W in 1⫻ Tris-boric acid-EDTA buffer. After drying and exposure of the gel to a storage phosphor screen (Molecular Dynamics, Sunnyvale, CA), the transcription products were scanned and quantified on a Taiphoon 8600 variable mode imager (Molecular Dynamics) and expressed relative to the internal standard (TBP) coamplified in each PCR reaction.

DNA extraction

MATERIALS AND METHODS Subjects Peripheral blood mononuclear cells (PBMC), for cell stimulation and RNA isolation, were obtained from 12 donors, six T1D subjects (mean diabetes duration, 8 years; range, 3–12 years) from a cohort of 253 Danish T1D families and six control subjects without any autoimmune disease. For mutational scanning by sequencing, DNA from a panel of 24 individuals with T1D and at least one other autoimmune disease was identified from a cohort of 253 Danish T1D families. Of the 24 T1D subjects, 17 also suffered from thyroid disease, four from celiac disease, two from ulcerative colitis, and one from pernicious anemia. All diagnoses were obtained and/or confirmed from medical records. Additionally, five controls without any autoimmune disease were included in the mutational scanning. For evaluation of allele frequencies, DNA from a panel of 96 unrelated T1D patients was genotyped for all identified single nucleotide polymorphisms (SNPs). For genotyping of SNPs with a minor allele frequency (MAF) above 1%, DNA from a collection of 253 Danish T1D families (1097 individuals), comprising 155 sib-pair families and 98 simplex families, was used [14, 19]. All probands were aged below 30 years at onset of T1D. For replication of positive association, DNA from an independent Danish T1D family material, comprising 227 simplex families, age at T1D onset for probands below 15 years, was used. A local ethics committee approved the study.

Methods Cell stimulation PBMC were isolated from 12 individuals; 1 ⫻ 106 cells/ml were stimulated with a cytokine mixture of interleukin (IL)-1 (300 pg/ml, BD Biosciences PharMingen, San Diego, CA), tumor necrosis factor ␣ (TNF-␣; 400 U/ml, Endogen, Pierce Biotechnology, Rockford, IL), and interferon-␥ (IFN-␥; 400 U/ml, BD Biosciences PharMingen) and incubated for 6 h (37°, 5% CO2). This cytokine mixture is toxic to pancreatic ␤-cells [20] but does not impair T cell function. The media contained RPMI 1640 with glutamax and 10% fetal calf serum (Gibco, Invitrogen, Carlsbad, CA). After incubation, cells were harvested, washed in phosphate-buffered saline, and pelleted by centrifugation.

580

Journal of Leukocyte Biology Volume 77, April 2005

DNA was extracted from leukocytes using standard procedures [22] or prepared as primer extension preamplification DNA [23].

CBLB sequencing and genotyping CBLB sequencing primers were based on available sequence, obtained from the National Center for Biotechnology Information (NCBI; Bethesda, MD, http://www.ncbi.nlm.nih.gov) and University of California Santa Cruz Genome Browser (http://www.genome.ucsc.edu). Primers were designed to give PCR products covering exons, exon-intron boundaries, and 5⬘ and 3⬘ untranslated regions (UTRs). PCR products of 300 –500 nucleotides in length were amplified and sequenced on automated DNA sequencing equipment (ABI3100, Applied Biosystems). Data were analyzed by SeqScape software, version 2.0 (Applied Biosystems). Identified mutations were verified by genotyping, and genotyping of the CBLB SNPs were performed by PCR-based restriction fragment length polymorphism assay, mutagenically separated PCR, or primer extension method (SnapShot, Applied Biosystems).

CTLA4-CT60 genotyping The CT60 SNP was genotyped using a specifically designed 5⬘ nuclease PCR assay, using TaqMan chemistry, designed on request (Assay-by-Design) by Applied Biosystems. The allelic discrimination, based on fluorescence, was performed by use of ABI Prism 7900HT sequence detection system equipment (Applied Biosystems), according to the manufacturer’s instructions.

Statistics For comparing relative expression levels of mRNA transcripts among groups, Student’s t-test was used. Allele transmission patterns from heterozygous parents to all affected offspring in the multiplex and simplex families were tested for linkage in the presence of linkage disequilibrium (LD) by the combined transmission disequilibrium test (TDT)/Sib-TDT [24]. Transmission from heterozygous parents to unaffected offspring was analyzed by ␹2 statistics. Parental haplotypes were estimated by maximum likelihood estimates in GeneHunter, version 2.1 (UK Human Genome Mapping Project Resource Centre, Cambridge). D⬘ values were calculated in HaploXT (http://archimedes. well.ox.ac.uk/pise/haploxt-simple.html). Two-sided P values of ⬍0.05 were set as level of statistical significance. P values were not corrected for number of comparisons, as correction factors are not simple to determine as a result of an

http://www.jleukbio.org

observed high degree of LD along the gene, suggesting that the polymorphisms are not independent.

RESULTS Several alternatively spliced human mRNAs have been reported, yielding shorter predicted proteins [9, 25]. These include a form of CBLB lacking exon 16 [9]. We evaluated the expression of CBLB transcripts, including this particular form, by testing cDNA isolated from PBMC from 12 donors in a semiquantitative RT-PCR assay. We demonstrated that the isoform lacking exon 16 as well as the isoform containing exon 16 were expressed in all 12 individuals tested, indicating constitutive expression of both isoforms in lymphocytes (Fig. 1). In addition, the constitutive expression in unstimulated PBMC was compared with the expression in cytokine-stimulated PBMC from the same individuals. No significant difference in expression levels of the transcript containing exon 16 was observed between unstimulated and stimulated cells, whereas the isoform without exon 16 was up-regulated significantly by cytokine exposure (Fig. 1). Of the 12 individuals tested, six were T1D subjects, and six were controls. As we were not able to demonstrate any significant difference between these two groups (data not shown), the two groups were not separated for analyses of expression levels. We screened the human CBLB gene for mutations in 24 individuals with T1D and at least one other autoimmune disease (resembling the KDP phenotype) as well as in five controls without any autoimmune disease. Screening was performed by resequencing coding regions, exon-intron boundaries, and 5⬘ and 3⬘ UTR regions of the gene. The CBLB gene maps to human chromosome 3q11-13.1, spans 206 kb, and comprises 19 exons [9]. Several CBLB mRNAs have been identified, among them, three forms containing three different first exons,

Fig. 1. Semiquantitative real-time PCR assay. Relative expression levels of two CBLB isoforms, related to expression of the housekeeping gene TBP. Lined columns, Transcripts from unstimulated PBMC; solid columns, transcripts from PBMC stimulated with a cytokine mixture containing IL-1, TNF-␣, and IFN-␥. Twelve individuals were tested for all four conditions.

1A, 1B, and 1C [9, 25]. Exons 1A and 1B are almost completely overlapping; the only difference is that exon 1A starts 8 base pairs (bp) farther upstream compared with exon 1B. These were identified from published sequence, but we were only able to resequence the complete exon 1C and the major part of exon 1A/1B. (We were not able to sequence the initial 119 bp of exon 1A, corresponding to the initial 111 bp of exon 1B.) Additionally, we sequenced the complete exons 2–19. We identified eight mutations by sequencing, located in exons 1A/1B (5⬘ UTR), 6, 10, 11, 12, 18, and 19 (3⬘ UTR), respectively (Table 1). All eight mutations were verified by genotyping, and five of them existed in the dbSNP (NCBI), whereas the three remaining were novel. Two additional coding SNPs in CBLB were identified from dbSNP. We were not able to demonstrate these in our screening panel, probably because one (rs1503922) is very rare (MAF, 0.0053), whereas the other (rs1503921) is located just upstream of the region screened. We initially tested these two additional SNPs, as well as the eight we have identified in this study, in 96 unrelated T1D patients. Only SNPs with a MAF above 1% were examined further in the complete T1D family material consisting of 253 T1D families. We demonstrated eight SNPs with a MAF above 1%; these were genotyped in the 253 T1D families and tested for T1D association by means of the combined TDT/Sib-TDT. One silent SNP in exon 12 (rs3772534) was demonstrated associated to T1D. Exon 12 encodes a proline-rich region of Cbl-b [9]. Sib-TDT revealed 26 transmissions of allele G versus 11 nontransmissions (P⫽0.03). Equal transmission (13 transmissions vs. 13 nontransmissions of allele G) to unaffected offspring was demonstrated. We attempted to replicate this observation in another Danish T1D family material comprising 227 families. In this sample, we were not able to demonstrate T1D association of this SNP, as few as 15 transmissions versus 12 nontransmissions were observed. The distribution of the transmissions in the two family materials, however, was not significantly different (P⫽0.29), and the two datasets were therefore pooled. Combined, 480 families were evaluated, and 41 (64%) transmissions of allele G versus 23 (36%) transmissions of allele A were observed (P⫽0.0459). The MAF in the 96 T1D patients initially tested was only 0.021, whereas the MAF (allele A) reported in dbSNP was 0.126. The MAF of the probands in all 480 families examined was 0.024 (0.025 in the material of 253 families and 0.022 in the second material of 227 families), thus quite different from the allele frequency stated in dbSNP. The T1D family material comprising 253 families was furthermore genotyped for the CTLA4 marker CT60, a SNP in the 3⬘ UTR region of CTLA4. We identified 164 (54%) transmissions of allele G versus 140 (46%) transmissions of allele A (P⫽0.18) to affected offspring in our material. Equal transmission to unaffected offspring (78 transmissions of the G allele and 78 transmissions of the A allele) was demonstrated. Genotyping data of the CBLB exon 12 SNP (rs3772534) was stratified according to this CT60 marker of CTLA4. Data were stratified depending on CT60 genotype in affected individuals, homozygous for the susceptibility allele G/G in one group and heterozygous A/G and homozygous A/A in the other group. Sib-TDT was used for analysis of the two subgroups (Table 2), and a further increased transmission distortion of the exon 12

Bergholdt et al. CBLB variants in T1D and interaction with CTLA4 581

TABLE 1.

Identified and Genotyped Variants of the Human CBLB Gene Affected

Exon Exon 1

dbSNP rs1503921 rs1503922 New

Exon 6 New Exon 10 rs2305035 rs2305036 Exon 11 rs2305037 Exon 12 rs3772534 rs3772534 rs3772534 Exon 18 New Exon 19 rs1042852

SNP

Material

C/T Family material A (253 fam.) T/G 96 T1D patients, Not detected C/T Family material A (253 fam.) T/C 96 T1D patients C/T Family material A (253 fam.) A/C Family material A (253 fam.) A/G Family material A (253 fam.) A/G Family material A (253 fam.) A/G Family material B (227 fam.) A/G Material A ⫹ B (480 fam.) C/T Family material A (253 fam.) A/G Family material A (253 fam.)

MAF

Transmissions Nontransmissions

Unaffected P

Transmissions Nontransmissions

P

0.199

101 (55%)

81 (45%)

NS

42 (55%)

34 (45%)

NS

0.164

119 (60%)

80 (40%)

NS

59 (49%)

61 (51%)

NS

0.0053 0.218

120 (57%)

89 (43%)

NS

51 (49%)

54 (51%)

NS

0.225

143 (59%)

101 (41%)

NS

51 (54%)

44 (46%)

NS

0.255

153 (57%)

116 (43%)

NS

66 (47%)

74 (53%)

NS

0.025

26 (70%)

11 (30%)

0.03

13 (50%)

13 (50%)

NS

0.022

15 (56%)

12 (44%)

NS

11 (61%)

7 (39%)

NS

0.024

41 (64%)

23 (36%)

0.046

24 (55%)

20 (45%)

NS

0.188

144 (59%)

99 (41%)

NS

55 (49%)

58 (51%)

NS

0.279

156 (55%)

128 (45%)

NS

68 (47%)

78 (53%)

NS

Identified and genotyped variants of the human CBLB gene. Genotyping data were analyzed by Sib-TDT and ␹2 statistics. Corresponding P values are shown. NS, Nonsignificant. Exact positions of the three newly identified variants in NCBI, Chromosome 3 genomic contig NT_005612 (Aug. 23, 2004), exon 1 (New): position 11,872,422; exon 6 (New): position 11,995,326; and exon 18 (New): position 12,070,939. dbSNP, SNP database.

SNP (79% transmissions of the CBLB, exon 12 G-allele) was seen in the susceptible CT60-G/G subgroup, comprising 104 T1D families (P⫽0.02), whereas no significantly distorted transmission was observed in the CT60-A/G and A/A subgroup of 136 T1D families (P⫽0.64; Table 2). The pattern of transmissions between the two groups was significantly different (P⬍0.0001). Furthermore, to evaluate the degree of LD in the region, we evaluated the pair-wise LD between the identified SNPs in CBLB. Parental haplotypes were estimated by maximum likelihood estimates in GeneHunter versus 1.2, and D⬘ values were calculated in HaploXT (Table 3). A high degree of LD in the region was demonstrated, and the presence of one large block and two smaller LD blocks with D⬘ values above 0.8, spanning ⬃140 kb, 2 kb, and 11 kb, respectively, was identified. These

TABLE 2.

three LD blocks include most of the gene from exons 1–10, exons 11 and 12, and exons 18 and 19, respectively (Table 3).

DISCUSSION Based on reports in the KDP rat as well as Cblb knockout mice, the Cblb gene has been suggested as a candidate gene for autoimmunity, strongly supported by functional studies [3, 4]. Dysregulation of signaling pathways modulated by Cbl-b may contribute to human autoimmune diseases as well [5]. Cbl-b is a key regulator of activation thresholds in mature lymphocytes and immunological tolerance and autoimmunity. Cbl-b has been shown to be a negative regulator of T and B cell activation, and it furthermore seems to be

Sib-TDT Results for CT60 as well as CBLB Exon 12 after CT60 Stratification Affected

CTLA4-CT60 Allele G CBLB-exon12 Stratified for: CT60 (G/G) CT60 (A/G⫹A/A)

Unaffected

Transmissions

Nontransmissions

P value

Transmissions

Nontransmissions

P value

253 families

164 (54%)

140 (46%)

0.176

78 (50%)

78 (50%)

1

104 families 136 families

15 (79%) 10 (62%)

4 (21%) 6 (38%)

0.022 0.763

6 (55%) 5 (42%)

5 (45%) 7 (58%)

0.643 0.564

Sib-TDT results of CTLA4 CT60 SNP genotyping as well as results of CBLB exon 12 data stratified according to the CTLA4 CT60 marker. Corresponding P values are shown.

582



TABLE 3.

Pair-Wise LD Measures

Pair-wise LD of consecutive markers Marker 1 Exon 1, rs1503921 Exon 1, New Exon 10, rs2305035 Exon 10, rs2305036 Exon 11, rs2305037 Exon 12, rs3772534 Exon 18, New

Marker 2

D⬘ value

Exon 1, New Exon 10, rs2305035 Exon 10, rs2305036 Exon 11, rs2305037 Exon 12, rs3772534 Exon 18, New Exon 19, rs1042852

0.800 0.833 0.933 0.700 0.868 0.494 0.967

Pair-wise LD measures, D⬘ values, among the eight genotyped CBLB SNPs, illustrating the extent of LD among the markers. A LD block is represented by consecutive D⬘ values above 0.8 and is indicated in bold. The location of the SNPs ranges from exon 1 to exon 19, spanning almost the entire length of the CBLB gene, approximately 206 kb.

able to regulate the triggering threshold of antigen receptors in T and B cells in mice [4]. Cbl-b may influence the CD28 dependence of T cell activation by selectively suppressing T cell receptor-mediated Vav activation [5]. Thus, CBLB can also be considered a functional candidate gene in man. We evaluated expression levels of two known isoforms of CBLB and identified constitutive expression of both isoforms in lymphocytes. In addition, the effects of cytokine exposure on the expression levels of the two isoforms were analyzed, and it is interesting that significant up-regulation of the isoform lacking exon 16 was demonstrated upon stimulation with cytokines. This may suggest a possible role for this isoform in the pathogenesis of T1D and perhaps other autoimmune diseases, although mechanisms, involving several forms of CBLB, are unknown and probably complex. The identified, although not disease-specific, effect of cytokines on the expression level of CBLB isoforms substantiates the probable importance of Cbl-b in autoimmunity. As the KDP phenotype includes several autoimmune tissues and in addition to T1D, often demonstrates infiltration in thyroid gland and other endocrine organs, we hypothesized that individuals with more than one autoimmune disease are more likely to harbor polymorphisms in CBLB, conferring susceptibility. We therefore screened the human CBLB gene in a panel of individuals with at least two autoimmune diseases, as well as five control subjects. We identified eight SNPs, of which five were known already (from dbSNP), whereas three were new. These eight SNPs as well as two additional SNPs identified from dbSNP were tested. We demonstrated no variants in exon 16 or splice sites flanking this exon, and as none of the known functional domains of Cbl-b are encoded for by exon 16 [9], the effect of the splice variant lacking exon 16 is not known. We observed a significant T1D association for an exon 12 (rs3772534) SNP, an exon encoding a proline-rich region of the protein. We attempted to replicate this finding in an independent T1D family material comprising 227 simplex families but found no independent association in this material. The distributions of transmissions in the two family materials, however, were not significantly different, and the two datasets were

therefore pooled. Combined, 480 families were evaluated for this exon 12 SNP, and a significant association to T1D was still observed, although quite few informative families were available as a result of a low MAF. The exon 12 SNP was also identified in a recent study by Payne et al. [26]. They determined a MAF based on their sequencing panel of 32 T1D subjects, and as the SNP was not identified as a haplotype-tagged SNP in their study, it was not genotyped. Transmitted and nontransmitted alleles were estimated from the regression equations computed, based on their sequencing panel. In their panel of 32 T1D subjects, they found a MAF of 0.043, comparable with the one observed in the present study. Their estimations showed equal transmission of alleles (P⫽0.19) [26]. So, although the genotyping burden in their study has been reduced significantly, they might have missed important information regarding this SNP. In another recent study of CBLB, this SNP was also detected with a MAF in 16 individuals of 0.031 but was not further analyzed [27]. Unlike the present study, both of these previous studies have used heterogeneous populations, which might confer possible biases as a result of potential population stratification. Recent studies suggest that the genome is organized into blocks of haplotypes, and efforts to create a genome-wide haplotype map of SNPs are underway, however are not finished yet (HapMap project, www.hapmap.org). As ancestral haplotypes propagate through a population, their physical length is reduced by recombination events, and markers belonging to the same haplotype are expected to be tightly linked and thereby being in LD with each other, which means they are inherited together more often than expected by chance. Traditionally, LD is described by pair-wise measures, such as Lewontin’s standardized disequilibrium coefficient D⬘ [28, 29], and the so-called LD-based method proposed by Gabriel et al. [29] involves calculating D⬘ values for all pairs of SNPs in the region. Pair-wise LD analyses of the CBLB SNPs revealed one large LD block, as well as two smaller blocks with D⬘ values exceeding a threshold of D⬘ ⬎ 0.8, which is a generally accepted threshold for reporting LD [30]. These blocks include most of the gene and suggest that genotyping of few SNPs only will provide near-complete information [30]. The demonstration of a D⬘ value of 0.868 for the identified exons 11 and 12 SNPs indicates a strong but not absolute LD between the two SNPs. It was surprising that we were not able to demonstrate T1D association of the exon 11 SNP as well. The exon 11 SNP was more informative, and a distorted transmission was demonstrated [153 (57%) vs. 116 (43%) transmissions], however was not statistically significant (P⫽0.31). A possible explanation for the difference in association between the two SNPs could be that the degree of LD was not absolute. In T1D, minor genes individually only confer small risks for susceptibility, whereas it is likely that gene-gene interactions may play substantial roles in the genetic susceptibility. Therefore, examining interactions of genes is a valuable approach, and genes involved in the same biological pathways are obvious candidates for such analyses. The T cell regulatory gene CTLA4 is an established candidate


gene/protein in T1D and other autoimmune diseases, and CTLA4 acts as a coreceptor in T cell activation, believed to be able to counter-regulate the action of another coreceptor, CD28. Cbl-b is believed to exert inhibitory effects on the CD28-dependent T cell activation. A recent report demonstrated association of polymorphisms of CTLA4 with several autoimmune diseases, including Graves’ disease, autoimmune hypothyroidism, and T1D, and identified a new marker, CT60, in the 3⬘ UTR region of the CTLA4 gene as the best marker for this association [31]. Ueda et al. [31] demonstrated increased transmission of the G allele to T1D offspring at approximately the same level (53.3%) as in the present study. As CTLA4 and CBLB are involved in biological pathways regulating T cell activation, it is relevant to examine the CBLB gene for genetic interaction with CTLA4. We therefore performed a combined genetic analysis of the two genes, combining the CTLA4 marker CT60 with the CBLB exon 12 SNP, demonstrating significant association to T1D in the current study. Stratifying CBLB exon 12 data according to a high-risk (G/G) CTLA4 CT60 genotype and lower risk (A/G and A/A) CT60 genotype revealed a further increased distortion of transmissions of the exon 12 SNP, suggesting genetic interaction between the two genes. The interaction observed, although based on few transmissions, supports genetic control of T cell activation in autoimmunity. The CBLB SNPs identified in the current study would be interesting to evaluate in materials comprising individuals with several autoimmune diseases as well as materials of thyroid patients to clarify the contribution of CBLB to autoimmunity. Functional studies of the significance of the CBLB exon 12 SNP are important as well. Furthermore, replication of the suggested genetic interaction between the CBLB and CTLA4 genes should be attempted, and the possible mechanism for this should be elucidated further. In summary, in a large Danish T1D family material of 480 families, we demonstrated significant association to T1D of a SNP in exon 12 of the human CBLB gene. In addition, we found indication of genetic interaction between the human CBLB and CTLA4 genes, both involved in the same pathway of T cell activation and supporting the theory of common genetic factors underlying several autoimmune diseases.

6.

7. 8.

9. 10.

11.

12.

13.

14.

15.

16. 17. 18. 19.

20.

21.

REFERENCES 22. 1. Pociot, F., McDermott, M. F. (2002) Genetics of type 1 diabetes mellitus. Genes Immun. 3, 235–249. 2. Yokoi, N., Kanazawa, M., Kitada, K., Tanaka, A., Kanazawa, Y., Suda, S., Ito, H., Serikawa, T., Komeda, K. (1997) A non-MHC locus essential for autoimmune type I diabetes in the Komeda diabetes-prone rat. J. Clin. Invest. 100, 2015–2021. 3. Yokoi, N., Komeda, K., Wang, H., Yano, H., Kitada, K., Saitoh, Y., Seino, Y., Yasuda, K., Serikawa, T., Seino, S. (2002) Cblb is a major susceptibility gene for rat type 1 diabetes mellitus. Nat. Genet. 31, 391–394. 4. Bachmaier, K., Krawczyk, C., Kozieradzki, I., Kong, Y. Y., Sasaki, T., Oliveira-Dos-Santos, A., Mariathasan, S., Bouchard, D., Wakeham, A., Itie, A., Le, J., Ohashi, P. S., Sarosi, I., Nishina, H., Lipkowitz, S., Penninger, J. M. (2000) Negative regulation of lymphocyte activation and autoimmunity by the molecular adaptor Cbl-B. Nature 403, 211–216. 5. Chiang, Y. J., Kole, H. K., Brown, K., Naramura, M., Fukuhara, S., Hu, R. J., Jang, I. K., Gutkind, J. S., Shevach, E., Gu, H. (2000) Cbl-B

584


23.

24. 25.

26.

regulates the Cd28 dependence of T-cell activation. Nature 403, 216 – 220. Chikuba, N., Akazawa, S., Yamaguchi, Y., Kawasaki, E., Takino, H., Takao, Y., Maeda, Y., Okuno, S., Yamamoto, H., Yokota, A., et al. (1992) Type 1 (insulin-dependent) diabetes mellitus with coexisting autoimmune thyroid disease in Japan. Intern. Med. 31, 1076 –1080. Payami, H., Joe, S., Thomson, G. (1989) Autoimmune thyroid disease in type I diabetic families. Genet. Epidemiol. 6, 137–141. Tait, K. F., Marshall, T., Berman, J., Carr-Smith, J., Rowe, B., Todd, J. A., Bain, S. C., Barnett, A. H., Gough, S. C. L. (2004) Clustering of autoimmune disease in parents of siblings from the type 1 diabetes Warren Repository. Diabet. Med. 21, 358 –362. Nau, M. M., Lipkowitz, S. (2003) Comparative genomic organization of the Cbl genes. Gene 308, 103–113. Concannon, P., Gogolin-Ewens, K. J., Hinds, D. A., Wapelhorst, B., Morrison, V. A., Stirling, B., Mitra, M., Farmer, J., Williams, S. R., Cox, N. J., Bell, G. I., Risch, N., Spielman, R. S. (1998) A second-generation screen of the human genome for susceptibility to insulin-dependent diabetes mellitus. Nat. Genet. 19, 292–296. Mein, C. A., Esposito, L., Dunn, M. G., Johnson, G. C., Timms, A. E., Goy, J. V., Smith, A. N., Sebag-Montefiore, L., Merriman, M. E., Wilson, A. J., Pritchard, L. E., Cucca, F., Barnett, A. H., Bain, S. C., Todd, J. A. (1998) A search for type 1 diabetes susceptibility genes in families from the United Kingdom. Nat. Genet. 19, 297–300. Davies, J. L., Kawaguchi, Y., Bennett, S. T., Copeman, J. B., Cordell, H. J., Pritchard, L. E., Reed, P. W., Gough, S. C. L., Jenkins, S. C., Palmer, S. M., et al. (1994) A genome-wide search for human type-1 diabetes susceptibility genes. Nature 371, 130 –136. Hashimoto, L., Habita, C., Beressi, J. P., Delepine, M., Besse, C., Cambon-Thomsen, A., Deschamps, I., Rotter, J. I., Djoulah, S., James, M. R. (1994) Genetic mapping of a susceptibility locus for insulin-dependent diabetes mellitus on chromosome 11q. Nature 371, 161–164. Nerup, J., Pociot, F., European Consortium for Iddm Studies (2001) A genomewide scan for type 1-diabetes susceptibility in Scandinavian families: identification of new loci with evidence of interactions. Am. J. Hum. Genet. 69, 1301–1313. Cox, N. J., Wapelhorst, B., Morrison, V. A., Johnson, L., Pinchuk, L., Spielman, R. S., Todd, J. A., Concannon, P. (2001) Seven regions of the genome show evidence of linkage to type 1 diabetes in a consensus analysis of 767 multiplex families. Am. J. Hum. Genet. 69, 820 – 830. Pociot, F. (2002) Ctla4 gene and autoimmune endocrinopathies: a new marker? J. Endocrinol. Invest. 25, 1001–1005. Kristiansen, O. P., Larsen, Z. M., Pociot, F. (2000) Ctla-4 in autoimmune diseases—a general susceptibility gene to autoimmunity? Genes Immun. 1, 170 –184. Rudd, C. E., Schneider, H. (2003) Unifying concepts in Cd28, Icos and Ctla4 co-receptor signalling. Nat. Rev. Immunol. 3, 544 –556. Lorenzen, T., Pociot, F., Stilgren, L., Kristiansen, O. P., Johannesen, J., Olsen, P. B., Walmar, A., Larsen, A., Albrechtsen, N. C., Eskildsen, P. C., Andersen, O. O., Nerup, J. (1998) Predictors of Iddm recurrence risk in offspring of Danish Iddm patients. Diabetologia 41, 666 – 673. Karlsen, A., Pavlovic, D., Nielsen, K., Jensen, J., Andersen, H., Pociot, F., Mandrup-Poulsen, T., Eizirik, D., Nerup, J. (2000) Interferon-␥ induces interleukin-1 converting enzyme expression in pancreatic islets by an interferon regulatory factor-1-dependent mechanism. J. Clin. Endocrinol. Metab. 85, 830 – 836. Karlsen, A. E., Ronn, S. G., Lindberg, K., Johannesen, J., Galsgaard, E. D., Pociot, F., Nielsen, J. H., Mandrup-Poulsen, T., Nerup, J., Billestrup, N. (2001) Suppressor of cytokine signaling 3 (Socs-3) protects ␤-cells against interleukin-1␤- and interferon-␥-mediated toxicity. Proc. Natl. Acad. Sci. USA 98, 12191–12196. Miller, S. A., Dykes, D. D., Polesky, H. F. (1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 16, 1215. Bergholdt, R., Larsen, Z. M., Andersen, N. A., Johannesen, J., Kristiansen, O. P., Mandrup-Poulsen, T., Nerup, J., Pociot, F. (2000) Characterization of new polymorphisms in the 5⬘ UTR of the human interleukin-1 receptor type 1 (IL1R1) gene: linkage to type 1 diabetes and correlation to IL-1RI plasma level. Genes Immun. 1, 495–500. Spielman, R. S., Ewens, W. J. (1998) A Sibship test for linkage in the presence of association: the Sib transmission/disequilibrium test. Am. J. Hum. Genet. 62, 450 – 458. Keane, M. M., Rivero-Lezcano, O. M., Mitchell, J. A., Robbins, K. C., Lipkowitz, S. (1995) Cloning and characterization of Cbl-b: a Sh3-binding protein with homology to the C-Cbl proto-oncogene. Oncogene 10, 2367– 2377. Payne, F., Smyth, D. J., Pask, R., Barratt, B. J., Cooper, J. D., Twells, R. C. J., Walker, N. M., Lam, A. C., Smink, L. J., Nutland, S., Rance,


H. E., Todd, J. A. (2004) Haplotype tag single nucleotide polymorphism analysis of the human orthologues of the rat type 1 diabetes genes Ian4 (Lyp/Iddm1) and Cblb. Diabetes 53, 505–509. 27. Kosoy, R., Yokoi, N., Seino, S., Concannon, P. (2004) Polymorphic variation in the Cblb gene in human type 1 diabetes. Genes Immun. 5, 232–235. 28. Daly, M. J., Rioux, J., Schaffner, S. F., Hudson, T. J., Lander, E. S. (2001) High-resolution haplotype structure in the human genome. Nat. Genet. 29, 229 –232. 29. Gabriel, S. B., Schaffner, S. F., Nguyen, H., Moore, J. M., Roy, J., Blumenstiel, B., Higgins, J., Defelice, M., Lochner, A., Faggart, M., Liu-Cordero, S. N., Rotimi, C., Adeyemo, A., Cooper, R., Ward, R.,

Lander, E. S., Daly, M. J., Altshuler, D. (2002) The structure of haplotype blocks in the human genome. Science 296, 2225–2229. 30. Carlson, C. S., Eberle, M. A., Rieder, M. J., Yi, Q., Kruglyak, L., Nickerson, D. A. (2004) Selecting a maximally informative set of singlenucleotide polymorphisms for association analyses using linkage disequilibrium. Am. J. Hum. Genet. 74, 106 –120. 31. Ueda, H., Howson, J. M. M., Esposito, L., Heward, J., Snook, H., Chamberlain, G., Rainbow, D. B., Hunter, K. M. D., Smith, A. N., Di-Genova, G., Herr, M. H., Dahlman, I., Payne, F., Smyth, D., Lowe, C., Twells, R. C., Howlett, S., Healy, B., Nutland, S., Rance, H. E., et al. (2003) Association of the T-cell regulatory gene Ctla4 with susceptibility to autoimmune disease. Nature 423, 506 –511.
