The extent to which concordance rates for schizophrenia in monozygotic twins (between 36 and 58%) fall short of 100% is often taken as an index of the role of an environmental factor in schizophrenia, but such an agent remains elusive. That schizophrenic symptoms are observed in some viral illnesses suggests that schizophrenia might be due to a gene-virus interaction, but analysis of age of onset in pairs of siblings with the disease rules out horizontal transmission. An alternative hypothesis is proposed that onset of disease is due to the expression of a ‘provirus', which is integrated in the genome, having been acquired either by prenatal infection or in the germ-line from an affected parent; this could explain why the season of birth effect is accentuated in, and perhaps confined to the group of patients without a family history of the disease.
Germ-line integration is known to occur following infection with agents of the retrovirus class.∗ Such agents can integrate at many sites in the host genome, but their interactions with proto-oncogenes (cellular genes which may act as growth factors) identify one type of integration site, and are. associated with some of their pathogenic effects. Some characteristics of schizophrenic illness, particularly their selectivity for the dominant hemisphere, can be understood on the assumption that the virus (perhaps a retrovirus) responsible for the disease interacts with a proto-oncogene, which induces the asymmetrical brain growth responsible, for laterality and cerebral dominance. The aetiologies of manic-depressive illness and schizophrenia may be related (the season of birth and onset effects are the same for the two conditions) and there is some evidence that the former transmutes into the latter in succeeding generations. The persistence of the functional psychoses may be due to the ability of the psychosis gene (or ‘provirus’) to induce change in the genetic mechanisms responsible for the development of laterality. (∗ See postscript pp 253).