How and why genetic linkage has not solved the problem of psychosis: review and hypothesis.

OBJECTIVE: The author examined the chromosomal linkage method as an approach to the genetic basis of schizophrenia and bipolar disorder. METHOD: Comparisons were conducted of recent meta-analyses of genome scans of schizophrenia and bipolar disorder and of the three largest (N>300) sibling pair studies of schizophrenia and schizoaffective disorder and a comparable study of bipolar illness. RESULTS: Recent meta-analyses have not identified consistent sites of linkage. The three largest studies of schizophrenia fail to agree on a single locus, no commonality with bipolar illness has been demonstrated, and there is no replicable support for any of the current candidate genes. DISCUSSION: An alternative to the concept that DNA sequence variation lies in "multiple genes of small effect" is the hypothesis that the variation is epigenetic but related to the genetic transition ("the speciation event") that separated Homo sapiens from a prior hominid species. This hypothesis draws attention to the chromosomal rearrangement (the Xq21.3/Yp translocation) that occurred some 6 million years ago in the hominid lineage and subsequent rearrangements, including a paracentric inversion, that have taken place within the translocated segment. Here it is argued that the most recent of these events is relevant to specifically human characteristics, including language. The gene pair protocadherin X and Y within this region is under new selective pressure and is in a novel (epigenetic) situation with respect to X inactivation. CONCLUSIONS: Epigenetic variation associated with chromosomal rearrangements that occurred in the hominid lineage and that relates to the evolution of language could account for predisposition to schizophrenia and schizoaffective and bipolar disorder and failure to detect such variation by standard linkage approaches.