Diabetes susceptibility genes, part 3: Genome-wide searches for type 1 genes 21/4/05. Fernando Gibson and Philippe Froguel In part 2 of their five-part, in-depth series exploring the search for diabetes susceptibility genes, Fernando Gibson and Philippe Froguel looked at studies of candidate genes. In part 3, they examine genome-wide searches for type 1 diabetes susceptibility genes.

As diabetes tends to run in families, the implication is that the members of a family share one or more common genetic risk factors. Finding such factors is not at all straightforward, however, as each genetic factor many have only a tiny effect on the overall risk of disease.

The approach used by researchers is therefore to examine lots of families with multiple offspring affected by diabetes, using a strategy called a genome-wide scan for linkage. By looking at short sequences of DNA that show inherited variations in length ('microsatellite' genetic markers), spaced evenly throughout the genome, researchers can trace the inheritance of regions of chromosomes in families.

If marker alleles are shared by affected sibling pairs more often than expected by random Mendelian segregation, then this marker is said to be 'linked' to (co-inherited with) a nearby genetic region that affects the risk of diabetes.

This strategy requires no presumptions as to the function of the susceptibility region; it attempts to map loci purely in relation to the chromosomal position of the variable genetic markers.

The first genome-wide scan for type 1 diabetes susceptibility regions was published in 1994; indeed, this was the first genome scan for a common human disease. Several more scans have been carried out since, each obtaining extremely strong evidence for linkage to the IDDM1 HLA region. This corroborates the HLA association data (see part 2 of this series).

In an attempt to increase the statistical power to detect linkage, an international consortium of investigators pooled all the available linkage data together in a study that was published in 2001. This 'meta-analysis' produced statistically significant evidence of linkage for only three regions (the previously identified IDDM1 and IDDM2, and a novel locus on chromosome 16q22-24) and 'suggestive' evidence for six regions (IDDM7, IDDM10, IDDM12, IDDM13 and IDDM15, and a novel locus on chromosome 1q42).

In total, more than 20 putative chromosomes regions that increase susceptibility to type 1 diabetes have been identified. (Table 1; see below)

Having obtained evidence for linkage, the next challenge is to go from a genomic region (often 20-40 million bases, encompassing hundreds of genes) to the one or a few genes in the region that harbour alleles conferring an increased risk of disease.

The one notable success story from this work has been the identification of the CTLA4 (cytotoxic T-lymphocyte-associated protein 4) gene, located in the IDDM12 region, a variant of which appears to confer a 1.3-fold increased risk of type 1 diabetes. Genetic variation at CTLA4 has also been implicated in other forms of autoimmune endocrinopathies, with robust association to Graves' disease and autoimmune hypothyroidism.

News: Genetic variation in 'molecular brake' for immune system increases risk of three serious common diseases.

Part 4 of the five-part series examines genome-wide searches for type 2 diabetes susceptibility genes.

Related links

Diabetes susceptibility genes, part 1: The search for genes

Diabetes susceptibility genes, part 2: Candidate genes

Diabetes susceptibility genes, part 5: Animal models of diabetes

Davies JL, et al. (1994) A genome-wide search for human type 1 diabetes susceptibility genes. Nature, 371, 130-136. Abstract