NEW YORK (GenomeWeb News) – A study appearing online yesterday in Nature Genetics is offering an expanded view of the genetic factors contributing to schizophrenia.
Through a multi-stage genome-wide association study involving tens of thousands of individuals with or without schizophrenia, a large international team led by investigators at the University of North Carolina at Chapel Hill and Sweden's Karolinska Institute uncovered almost two-dozen loci with ties to the disease. That set included 13 loci not linked to schizophrenia or related conditions in the past, along with one new site previously detected in a bipolar disorder study.
An analysis of the new and known risk sites suggested schizophrenia risk may involve as many as 6,300 to 10,200 SNPs — many of them common variants found at relatively high frequency in human populations. As a greater number of such SNPs are identified, though, they are expected to reveal a far more limited number of genes and biological processes contributing to the condition.
"We suggest that a relatively thorough enumeration of the genomic loci conferring risk for schizophrenia (the 'parts list') should be a priority for the field," the study's corresponding author Patrick Sullivan, a researcher affiliated with UNC at Chapel Hill and the Karolinska Institute, and his colleagues wrote.
"Identifying all loci would surely be an exercise in diminishing returns," they added. "However, we propose a goal for the field: the identification of the top 2,000 loci (for example) might be sufficient to confidently and clearly identify the biological processes that mediate risk and protection for schizophrenia."
For the current analysis, the team started by directly genotyping 5,001 Swedish individuals with schizophrenia and 6,243 unaffected individuals from the same population, using data from the 1000 Genomes Project to impute additional SNPs in these cases and controls.
From that data, researchers detected more than 300 SNPs showing genome-wide significant associations with schizophrenia. These variants centered around two sites in the genome, they noted, a new locus on chromosome 2 and a major histocompatibility complex region on chromosome 6 that's been implicated in several prior schizophrenia studies.
The same MHC locus turned up when the team analyzed data for another 8,832 cases and 12,067 controls sampled by members of the Psychiatric Genomics Consortium, again using 1000 Genomes data to impute additional variants that weren't directly assessed in these individuals. So, too, did variants at four other loci on chromosomes 7, 8, 10, and 18.
By performing a meta-analysis using data for both the Swedish cases and controls and the Psychiatric Genomics Consortium samples, meanwhile, researchers detected genome-wide significantly associated variants at a dozen loci.
Data from the two cohorts also made it possible to look at how well a large set of apparent schizophrenia risk variants from the Psychiatric Genomics Consortium could distinguish Swedish cases from controls.
"Although the mean risk profiles showed highly significant differences between cases and controls, the distributions overlapped substantially and are insufficient for diagnostic purposes," the study authors noted. "However, these results strongly support the comparability of the Swedish and [Psychiatric Genomics Consortium] samples and the validity of the meta-analysis results."
To verify associations identified in the first stages of the schizophrenia study, the researchers went on to test the top SNPs in these loci in another 7,413 individuals with schizophrenia and 19,762 unaffected controls.
Though 24 loci showed significant ties to schizophrenia after that validation step, two spots contained overlapping pairs of loci, leaving the team with 22 sites showing independent schizophrenia associations.
Amongst the candidate genes found in and around these apparent risk loci, the group saw enrichment for calcium channel genes, which code for proteins involved in signaling processes that contribute to neuron function and other processes.
Based on findings from this and other studies, the researchers also highlighted the potential importance of the microRNA miR-137 in schizophrenia, as well as long intergenic non-coding RNAs, or lncRNAs, which are among the sequences found at more than half of the 22 associated loci.
They noted that additional research is needed to uncover new risk loci and to untangle the consistently detected association between the mental conditions and the complicated and gene-rich MHC region on chromosome 6.
"As indicated by our findings, greater knowledge of the genetic basis of schizophrenia can converge on increasingly specific neurobiological hypotheses that can be prioritized for subsequent investigation," they concluded.