Background: Several investigations suggest that abnormalities in gamma-amino butyric acid (GABA) neurotransmission systems may be related to the pathophysiology of schizophrenia. A GABA(A) receptor gene cluster on 5q31-35 (beta2 [GABRB2], alpha6 [GABRA6], alpha1 [GABRA1], and gamma2 [GABRG2] subunit genes) is one of the most attractive candidate regions for schizophrenia susceptibility.
Methods: We performed 1) systematic polymorphism search of GABRB2, GABRA6, and GABRA1, in addition to our colleague's study of GABRG2; 2) evaluation of linkage disequilibrium (LD) within this cluster with 35 single nucleotide polymorphisms (SNPs); 3) "selection of haplotype-tagging (ht) SNPs"; and 4) two-stage association analysis that comprised first-set screening analysis of all htSNPs (288 Japanese schizophrenia patients and 288 control subjects) and second-set replication analysis of the positive htSNPs (901 schizophrenic patients and 806 control subjects).
Results: In the first-set scan, we found a significant association of two htSNPs in GABRA1, but no association of GABRB2, GABRA6, and GABRG2. In the following second-set analysis, however, we could not confirm these significant associations.
Conclusions: These results indicate that this gene cluster may not play a major role in Japanese schizophrenia. They also raised an alert with regard to preliminary genetic association analysis using a small sample size.