Science Editor

In the past decade, individual genomewide association studies have identified genetic contributions to most common and many uncommon diseases. But collectively, they have also shown something else: that the idea that common variants will account for most cases of common diseases is false.

Indeed, Shiv Pillai told BioWorld Today, one way of looking at it is that the common genetic variants that genomewide association studies are set up to detect are exactly the ones that are least likely to lead to health problems.

"Maybe the reason a variant is so common is because it doesn't affect protein function very much," he said.

The ideas for what sorts of genetic alterations might lead to disease include more complex genetic variations such as insertions and deletion, epigenetic contributors, and the idea that common diseases might be due to many rare genetic factors - each of which would account for only a small number of cases - instead of a few common ones.

The latter idea got experimental support this week from a study that senior author Pillai, who is at the Massachusetts General Hospital Center for Cancer Research, and his colleagues published online in Nature on June 16, 2010. In their paper, they showed that mutations in the gene for sialic acid acetylesterase, or SIAE, although rare overall, occurred far more often in individuals with autoimmune disease than in healthy controls.

The relationship between the two is indirect, but SIAE, through its effects on sialic acid, ultimately contributes to the negative regulation of antibody-secreting B cells. Pillai and his colleagues had previously shown that animals missing either the enzyme itself or the CD22 receptor on B cells whose signaling SIAE ultimately influences are prone to autoimmune disease.

Based on the animal studies, Pillai and his colleagues sequenced the entire genes of slightly more than 900 individuals with different autoimmune disorders including inflammatory bowel disease, Type I diabetes and rheumatoid arthritis, and about 650 controls. They found that individuals with autoimmune disorders were far more likely than controls to have rare variants that impaired SIAE functions than controls.

Pillai and his team then expressed each defective version of SIAE they had identified through sequencing in cell cultures to answer two questions: "Is the protein defective, and is it secreted."

Mutations in SIAE appear to prevent the protein from functioning in one of two ways. In some mutations, which are dominant, the enzyme produced by the defective copy prevents the normal enzyme that is still being made by the other copy from functioning; Pillai said that SIAE works as a dimer or multimer, and one defective subunit appears to be enough to gum up the enzyme as a whole.

The group also found one variation that occurs fairly frequently in controls, but is a problem only when it is homozygous; biochemical assays showed that particular variation is catalytically active, but useless nevertheless because it cannot be transported.

Mutations in SIAE itself is fairly rare even in patients with autoimmune disorders; Pillai said that only 2-3 percent of the autoimmune patients his team looked at had defective SIAE, and each individual variation is found in less than 1 percent of patients.

But the findings might nevertheless lead to clinical payoff: Like basically every enzyme, SIAE is part of a pathway, and preliminary studies suggested that pathway is defective "in a much higher percentage of people" with autoimmune disorders, suggesting that SIAE or another of its components could potentially be targeted to prevent autoimmune disorders.