Recent studies have suggested that the alpha 7-nicotinic receptor gene (CHRNA7) may play a role in the pathogenesis of schizophrenia. The alpha 7-nicotinic receptor gene (CHRNA7) is involved in P50 auditory sensory gating deficits, and the genomic locus for this gene lies in the chromosome 15q13-14 regions. The human gene is partially duplicated (exons 5-10) with four novel upstream exons. The marker D15S1360 has been shown to be significantly linked with the phenotype of abnormal P50 suppression in schizophrenia families. The marker L76630 is 3 kb in the 3' direction from the last exon of the CHRNA7 gene and is located in the duplicated region. The function of the two L76630 copies is unknown. We genotyped three polymorphic markers D15S1360, D15S165, and L76630 that are localized in a genomic fragment containing the CHRNA7 in 31 Azorean schizophrenia families/trios (including 41 schizophrenia individuals and 97 unaffected families members). An overall analysis utilizing the family-based association test revealed significant linkage disequilibrium between L76630 and schizophrenia (P = 0.0004). Using the extended transmission disequilibrium test and limiting the analysis to one triad per family, transmission disequilibrium of D15S1360 was near significance (P = 0.078). The 15q13 region overlaps with the location of two well-known genomically imprinted disorders: Angelman syndrome and Prader-Willi syndrome. Therefore, we investigated maternal and paternal meioses. We found significant transmission disequilibrium for D15S1360 through paternal transmission (P = 0.0006) in our schizophrenia families. The L76630 marker showed a significant disequilibrium in maternal transmissions (P = 0.028). No parent-of-origin effect was found in D15S165. Overall, our results suggest that the CHRNA7 may play a role in schizophrenia in these families. A parent of origin effect may be present and requires further study.

Many psychiatric disorders are influenced by genetic factors, but the genetic components of complex diseases may not follow clear inheritance patterns. Although the patients may share a common clinical phenotype, the cause of the syndrome may consist of a heterogeneous collection of both genetic and/or environmental components. One method to minimize genetic heterogeneity in studies of complex disorders is to select a very homogenous study population. The average number of families with the same last name, when corrected for population size, is an excellent marker for the degree of homogeneity. We used surname analysis to evaluate the homogeneity of the Portuguese population of Madeira, comparing it with previous data on the homogeneity of populations of mainland Portugal, the Azores, and both rural and urban US populations. The average number of families with the same last name corrected for population size was 33.84 in Madeira, 30.88 in the Azores, and 21.42 in Coimbra (mainland Portugal) compared with 1.13 in rural and 0.38 in urban United States. This surname analysis supports the premise that the Portuguese population is a highly homogenous population, with the highest homogeneity in Madeira and the Azores, making it a good study population for molecular genetic analyses.