NEW YORK (GenomeWeb) – Low-frequency coding variants contribute to the heritability of multiple sclerosis, according to a new analysis.
The International Multiple Sclerosis Genetics Consortium previously identified 233 genetic risk variants for the condition through genome-wide association studies. However, these variants account for about 20 percent of disease heritability, leaving the rest unexplained. It has been suggested, however, that common variants act together to increase risk or that families harbor their own private mutations.
Consortium researchers have now conducted a meta-analysis in nearly 70,000 cases and controls of nearly 121,000 variants, with a particular focus on those predicted to have a phenotypic effect. As they reported in Cell yesterday, the researchers identified four novel genes associated with MS disease risk and calculated that low-frequency variants contribute to disease risk.
"We simply would not have found them by continuing to look at common genetic variants; we had to look for rarer events, which means looking at many, many more people," corresponding author Chris Cotsapas, associate professor of neurology and genetics at Yale University, said in a statement. "These variants explain an extra 5 percent of risk."
In their meta-analysis, he and his colleagues examined 120,991 low-frequency coding variants, including 104,218 non-synonymous and 2,276 nonsense variants. For this analysis, they relied on 32,367 individuals with MS and 36,012 controls from Australia, Europe, and the US who had undergone genotyping on either the lllumina HumanExome Beadchip or a custom MS array.
While the researchers confirmed previously reported associations in this way, they also reported four novel ones.
Using a restricted maximum-likelihood approach, the researchers modeled the contribution of the different types of variants to disease. They found that low-frequency variants explained about 11 percent of the difference between cases and controls, which they noted is a mean 4.1 percent on the liability scale. When they separated low-frequency variants into intermediate and rare classifications based on their minor allele frequency values, they estimated that rare variants alone explain about 9 percent of the difference between cases and controls, and represents about 3.2 percent on the liability scale.
This suggested to the researchers that there could be many more rare variants that contribute to disease risk that have yet to be identified, even in large studies.
The loci the researchers tied in this study to MS further underscore the role of immune dysfunction in the disease. As the SNPs they identified were not in linkage disequilibrium with other variants, the researchers suspected that the genes harboring these variant are likely to be disease related.
One affected gene is PRF1, which encodes perforin, a part of the granzyme-mediated cytotoxicity pathway that is used by various types of lymphocytes as well as natural killer cells and certain T regulatory cells. Another is HDAC7, which encodes a class II histone deacetylase that affects T cell survival via its effects on FOXP3. In addition, PRKRA encodes a protein kinase interferon-inducible double-stranded RNA-depended activator, while NLRP8 is an intracellular cytosolic receptor that is involved in the innate immune response.
These findings, the researchers added, suggest an avenue for further teasing out the pathogenesis and biology behind MS.