See Q&A with Michael O'Donovan
July 21, 2014. A highly anticipated schizophrenia genome-wide association study published July 21 in Nature identifies 83 novel risk markers and replicates 25 others. The authors—the Schizophrenia Working Group of the Psychiatric Genomics Consortium (PGC)—note that among the loci found are markers in or near genes involved in dopamine, glutamate, and calcium signaling, some already identified in prior GWAS studies and some the targets of current or experimental drugs.
"Detecting biological risk factors on this scale shows that schizophrenia can be tackled by the same approaches that have already transformed outcomes for people with other diseases," said senior author Michael O’Donovan of Cardiff University in the United Kingdom in a press release.
In an editorial accompanying the article, the UK’s Jonathan Flint at University of Oxford and Marcus Munafò at University of Bristol, call the study "a tremendous advance, of the sort that rewrites text books."
While noting that the genetic contribution to schizophrenia is already accepted, David Goldman and Colin Hodgkinson, of the National Institute on Alcohol Abuse and Alcoholism, in Rockville, MD, echo these sentiments. "[T]his genetic study, because of its large magnitude and with >100 genome wide significant loci, is a watershed event in the discovery of the genetic variation that must be responsible for this inheritance," they write in a comment for SRF (see full text below).
Prior to the new study, about 30 loci had previously been associated with schizophrenia using GWAS (see SRF related news report), a technique that scans the genome for single nucleotide polymorphisms (SNPs) that are "common," present in more than 10 percent of the population. A global view of the current findings has been announced at meetings (see SRF related meeting report; SRF related meeting report), but researchers have eagerly awaited a chance to take a peek at the individual risk markers.
Arriving at Manhattan
In the Nature study, first author Stephan Ripke, of Massachusetts General Hospital in Boston, and the Broad Institute in Cambridge, and colleagues ran every available GWAS sample through a common analysis pipeline, yielding a final sample size of 36,989 schizophrenia cases and 113,075 controls, derived from 55 different datasets.
A total of 128 SNPs were significant. When plotted across the genome, these variants exceed the high bar set for genome-wide significance (p ≤ 5 x10-8) to resemble skyscrapers amid the lower, non-significant "buildings." In contrast to prior studies that identified fewer SNPs, the so-called Manhattan plot of the current data is finally starting to resemble the skyline of its namesake.
In an attempt to identify candidate risk genes, the researchers combined correlated SNPs into loci, yielding a total of 108 physically distinct spots associated with schizophrenia. The current data set largely replicates past studies – all but five loci previously identified in large GWAS were significant in the new study. Three quarters of the 108 loci fall within protein-coding genes, and an additional 8 percent are located less than 20 kb from a gene.
Old and new
"In just a few short years, by analyzing tens of thousands of samples, our consortium has moved from identifying only a handful of loci associated with schizophrenia, to finding so many that we can see patterns among them," Ripke says in a press release from the Broad Institute.
It is noteworthy that variants in favorite schizophrenia genes such as the dopamine D2 receptor (DRD2) and several voltage-gated calcium channel subunits (e.g. CACNA1C, CACNB2) were again identified in this study. A number of genes linked to glutamatergic signaling and synaptic plasticity (GRM3, GRIN2A) also were tagged by the markers.
One especially exciting aspect of the study, National Institute of Mental Health Director Tom Insel writes on his blog, is that "many of the common variants overlap with rare mutations identified in previous reports as associated with risk of schizophrenia, confirming that some regions are worth a deeper dive."
A whopping 83 of the 108 loci, however, have not been previously implicated, and thus may represent novel biological pathways in the illness. The authors have made an explicit statement with their title, "Biological insights from 108 schizophrenia-associated genetic loci," and spend some time characterizing those regions. Because many markers were located outside exomic regions, and outside genes altogether, they suggest that regulatory regions for genes, as opposed to protein-coding variants, may be particularly important in schizophrenia.
Brain-expressed genes were enriched among the variants, as were genes expressed in tissues that play important immunological roles. The latter supports a role for immune system molecules in schizophrenia, an idea that is increasingly gaining traction in the research community (see SRF related meeting report; SRF related news report; SRF related news report).
Computation of a risk profile score (RPS) that accounts for all risk variants carried by an individual revealed that the odds of having schizophrenia increased with the number of risk alleles, though the authors caution that current sensitivity and specificity limitations preclude the use of RPS as a predictive test.
Beyond the list of hits, the loci that did not make the cut are also important, suggests SRF advisor Irving Gottesman of the University of Minnesota, in Minneapolis, who termed the study a "blockbuster."
"We no longer need to chase what now appear to be dead ends," Gottesman opined. "One of the most valuable contributions of the paper is that it tells you not only where to look next . . . but also where not to look."
Goldman and Hodgkinson caution that there is still a lot more work to be done to establish a definitive role for the identified variants in schizophrenia. "The conclusion that the variants are regulatory in nature will have to await a more complete accounting of the genes and loci involved, and the actual identification of the loci responsible," they write.
Insel also cautions the casual reader that the study has not found "schizophrenia genes," per se. "Most of these [variants] are not in or even near genes. And any one of these 108 regions contributes only a tiny fraction of risk in the population. Nevertheless, this is a major step forward in describing the genetic risk for schizophrenia," he writes.
Even larger sample sizes are currently being collected for the next generation of analyses. For Gottesman, the current paper evokes a popular Winston Churchill quote: "Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning."—Allison A. Curley and Hakon Heimer.
See Q&A with Michael O'Donovan
Ripke S et al. Biological insights from 108 schizophrenia-associated loci. Nat. 2014 July 21.
Flint J, Munafò M. Genesis of a complex disease. Nat. 2014 July 21.