The approach taken by Rachel Craddock and Sabine Bahn’s group is highly laudable. The work represents a beautiful marriage of highly technical cutting-edge proteomics and classical genetic epidemiology, which still shows its worth. Only by including discordant monozygotic twins is it possible to see that the increase in α-defensins is associated not only with having schizophrenia, but with the risk for the disorder. This goes a long way to establish these as possible early risk indicators. Obvious future studies would include longitudinal profiling of α-defensin levels in individuals who are clinically or genetically at risk for schizophrenia, as well as analyses of the specificity (vis-à-vis bipolar disorder, for example). In addition, while this type of high(er)-throughput proteomic research is likely to identify some potentially distinguishing biomarkers, it is likely that the effects of some risk-conferring genetic polymorphisms are not easily reflected in alterations in protein levels or even in their configuration, so a multi-level analysis and integration of DNA, RNA, and protein biomarkers may be necessary. It should be a priority to determine if genetic polymorphisms in genes coding for α-defensins are associated with risk for the disorder.

View all comments by Stephen J. Glatt