NEW YORK (GenomeWeb News) – Researchers from the Wellcome Trust Sanger Institute and elsewhere reported online in the American Journal of Human Genetics yesterday that they have tracked down a new risk gene for osteoarthritis.
The team did not initially find the association through a genome-wide association study involving 3,177 individuals with osteoarthritis and 4,894 unaffected controls genotyped at around 600,000 sites in the genome. But by using an imputation method based on data from the 1000 Genomes Project, they were able to look at millions more sites, which allowed them to track down a previously undetected risk locus for osteoarthritis in the chromosome 13 gene MCF2L — findings that they subsequently verified in tens of thousands more individuals with or without the disease.
"By using the 1000 Genomes Project data to add value to our original genome-wide association scan for osteoarthritis, we have uncovered a disease-associated gene that had previously remained hidden," Sanger researcher Eleftheria Zeggini, the study's senior author, said in a statement. "We were able to analyze our results in greater detail and zoom in on variants that we hadn’t been able to identify before."
Previous studies had identified two loci — one on chromosome 7 and another in the GDF5 gene on chromosome 20 — that are linked to osteoarthritis in individuals of European descent, though much of the heritability of the degenerative joint disease remains unexplained.
"Osteoarthritis is a complex disease with many genetic causes," Medical Director of Arthritis Research UK Alan Silman, who was not involved in the new study, said in a statement. "Yet it has proved very difficult to find the genes involved and help us to identify potential areas of treatment."
In an effort to find more risk factors for the disease, members of the arcOGEN consortium genotyped 3,177 individuals of European descent who had osteoarthritis at roughly 600,000 sites in the genome. These genetic patterns were compared with those for 4,894 unaffected controls enrolled through the 1958 British Birth Cohort in the UK and the UK National Blood Donor Service and tested through the Wellcome Trust Case Control Consortium.
To broaden their search for osteoarthritis-associated variants, the team turned to imputation strategies that relied on information from the 1000 Genomes Project, using data on 60 individuals of European ancestry who were sampled through the pilot phase of that project to impute information at more than 7.2 variant sites genome-wide in the osteoarthritis GWAS cases and controls.
"Imputation based on the 1000 Genomes Project has been proposed as an approach that will increase power and resolution in genetic association studies, and researchers have already applied the technique to fine map known association signals," the study authors explained. "In this work, we applied a [1000 Genomes Project]-based imputation and identify a genome-wide significant locus for [osteoarthritis] within a gene previously unlinked to the disease."
From this analysis, the team found eight suspicious SNPs at six loci that they subsequently genotyped in 5,165 osteoarthritis cases and 6,155 controls from the UK population.
One chromosome 13 locus in an intron of MCF2L remained associated with the disease after this stage of the study. Genotyping experiments on thousands more cases and controls from sample sets collected in the UK, Iceland, Estonia, and the Netherlands verified this association, while the team's meta-analyses point to an especially pronounced association with osteoarthritis affecting the knee.
The MCF2L gene is believed to help regulate nerve growth factor-related cell migration, the researchers noted, and there is some evidence suggesting antibodies that inhibit members of the nerve growth family can help alleviate osteoarthritis related joint pain and other symptoms of the disease.