Role of ITPR3 as a candidate gene for Type 1 diabetes

Introduction
This page provides a summary of the following paper, as well as access to download the results and supplementary materials:
Genetic mapping at 3-kilobase resolution reveals inositol 1,4,5-triphosphate receptor 3 as a risk factor for type 1 diabetes in Sweden.
Roach JC, Deutsch K, Li S, Siegel AF, Bekris LM, Einhaus DC, Sheridan CM, Glusman G, Hood L, Lernmark A, Janer M, Swedish Childhood Diabetes Study Group, Diabetes Incidence in Sweden Study Group
Am J Hum Genet. 2006
PubMed ID: 16960798
 
About ITPR3
The human gene ITPR3 is located next to the MHC locus on chromosome 6. During a dense SNP mapping of the MHC and its immediate surroundings, an association peak was found through exons 3-33 of ITPR3, with the center of the peak at exon 8 (rs2296336). Although it is close in proximity, the contribution of ITPR3 is separate from the MHC. There is a known recombination hotspot between ITPR3 and the MHC. In addition, the combination of an ITPR3 genotype with the HLA-DQ genotype found in the MHC are more predictive than either genotype alone, showing that they contribute independently (Roach et al. 2006).
ITPR3 is part of the gene family, ITPR. The ITPR family contains 3 genes, which all show strong conservation with each other. Their expression is enriched in islets and the thyroid. These genes contain the IP3 binding domain, which mediates second messenger Ca++ signaling in the cell.
ITPR3 could contribute to the pathogenesis of T1D in one of two ways. First, because ITPRs are associated with Ca++ signaling, any possible change in behavior could cause Ca++ to be off-balance, which would make the cell vulnerable. Second, ITPR3 is localized to insulin and somatostatin secretory granules, which could mean that it is secreted by the cell, and may contribute to an immune response.
A separate mouse experiment where ITPR2 and ITPR3 were knocked out showed that these genes were important for metabolism and development (Futatsugi, et al. 2005). Animals that were double-knockouts were hypoglycemic and lean, despite ingesting the same amount of food as the control group. They showed signs of exocrine dysfunction and decreased nutritional digestion due to severely impaired Ca++ signaling in acinar cells in the salivary glands and the pancreas.
Dense SNP mapping of the MHC
To save the following files to your computer, right-click on the link and choose "Save Target As..."
Download Data:
  • All SNPs
  • SNPs in ITPR3 Gene
    expander How were these files created? Use T1DMart to query a subset of the data.
  • expander Haploview Files
  • Genotypes for each anonymized individual
Supplemental Materials:
  • Color version of Figure 1. PostScript PDF This is a version of Figure 1 from the paper with some elements in color.
  • Supplemental Figure 1. An alternative display of linkage disequilibrium in our population. Shades of red portray the D’/LOD default schema of Haploview 3.2. The SNPs rs2296336 and rs2229634 are at the center of the green oval; associations of blocks containing these SNPs are bounded by green lines. The haplotype blocks determined by Haploview’s default settings are indicated by bold outlines. SNPs are ordered consecutively with equal spacing regardless of intervening physical distance; actual physical spacings are portrayed along the bar at the top of the figure. To avoid crowding, only SNPs with MAF >= 0.3 are displayed.
  • Supplemental Table 1. List of SNPs, allele key (which nucleotide corresponds to the Illumina code-letter designations “A” and “B”), scores, positions, and sequences used to design SNP assays. Positions are relative to the axis-origin of Figure 1; 28,969,648 bp should be added to each position to obtain the position relative to the canonical UCSC/NCBI build. Sequences are derived from dbSNP, and they do not necessarily map directly to the canonical UCSC genome sequence, and when they do, they may be in a non-canonical (reverse) orientation. The name of the SNP is the dbSNP accession number. The score is -log10(survey p-value). The positions column lists all positions to which the SNP maps (usually a single position; multiple positions are comma separated). Illumina uses only bialleleic SNPS, and codes each allele as either “A” or “B”; the keys to the Illumina code for each SNP are in the columns “A-coded Allele” and “B-coded Allele”. The sequence describes the sequence flanking the SNP (indicated in brackets within the sequence).
  • Supplemental Table 2. Reference sequences to which SNPs were mapped. After mapping by sequence-identity matching to these reference sequences, all SNPs not mapping directly to the UCSC/NCBI canonical MHC were mapped using indirect alignment information based on the alignment between the haplotype to which they mapped and the UCSC/NCBI canonical MHC. IMGT/HLA is described by Marsh (2003).
  • Supplemental Table 3. Allele counts for each SNP in the survey study, including p-values and odds ratios. These are computed with Haploview which implements allele-based rather than genotype-based contingency tables for computations.
  • Supplemental Table 4. Association of HLA Class II types and subtypes with T1D and two-locus logistic regression. This is an expanded version of Table 6 from the main text. The expansion includes full counts for each contingency table, conditioning on rs2296636 (the reverse of the conditioning in the table in the main text), and chi square statistics. The presence of some MHC alleles, such as DQA-01, is correlated with reduced risk, while the presence of others, such as DQB-0302, is correlated with increased risk. The table also includes analysis of selected DQA/DQB haplotypes.
  • Illumina GenCall Scores and Call Rates for Survey SNPs. The Locus_Summary_Report file contains information on the quality and success of typing at each survey SNP.
Suggested Views
ITPR3 in GBrowse  
ITPR3 Gene Page The gene page includes gene models for ITPR3.
ITPR Gene Family in Gene Atlas  
Genes in regions of LD
Markers were selected from the ITPR3 SNP data which had a reasonably small p-value, and which did not occur in the HLA region (here defined as chr6: 32,000,000-33,500,000). LD analysis was carried out on these markers, and regions of strong LD were identified, using different r2 cut-off values. These regions were then examined for genes, and the presence of markers within genes was determined.
Genes in LD regions
Parameter Tabular Data
(Opens in new window)		 Raw Data File
r2 > 0.4 View Download
r2 > 0.8 View Download
r2 = 1 View Download
expander Analysis Parameters
Folders
Related Papers
IP3 receptor types 2 and 3 mediate exocrine secretion underlying energy metabolism.
Futatsugi A, Nakamura T, Yamada MK, Ebisui E, Nakamura K, Uchida K, Kitaguchi T, Takahashi-Iwanaga H, Noda T, Aruga J, Mikoshiba K
Science. 2005
PubMed ID: 16195467
 
Send List to Tool
Download to Screen
Gene Dossier
Send List to Tool
Download to Screen