COMT haplotypes suggest P2 promoter region relevance for schizophrenia

Abstract

A recent study found, in a large sample of Ashkenazi Jews, a highly significant association between schizophrenia and a particular haplotype of three polymorphic sites in the catechol-O-methyl transferase, COMT, gene: an IVS 1 SNP (dbSNP rs737865), the exon 4 functional SNP (Val158Met, dbSNP rs165688), and a downstream SNP (dbSNP rs165599). Subsequently, this haplotype was shown to be associated with lower levels of COMT cDNA derived from normal cortical brain tissue, most likely due to cis-acting element(s). As a first step toward evaluating whether this haplotype may be relevant to schizophrenia in populations other than Ashkenazi Jews, we have studied this haplotype in 38 populations representing all major regions of the world. Adding to our previous data on four polymorphic sites in the COMT gene, including the Val158Met polymorphism, we have typed the IVS 1 rs737865 and 3 0 rs615599 sites and also included a novel IVS 1 indel polymorphism, yielding seven-site haplotype frequencies for normal individuals in the 38 globally distributed populations, including a sample of Ashkenazi Jews. We report that the schizophrenia-associated haplotype is significantly heterogeneous in populations worldwide. The three-site, schizophrenia-associated haplotype frequencies range from 0% in South America to 37.1% in Southwest Asia, despite the fact that schizophrenia occurs at roughly equal frequency around the world. Assuming that the published associations found between the exon 4 Val158Met SNP and schizophrenia are due to linkage disequilibrium, these new haplotype data support the hypothesis of a relevant cis variant linked to the rs737865 site, possibly just upstream in the P2 promoter driving transcription of the predominant form of COMT in the brain. The previously described HindIII restriction site polymorphism, located within the P2 promoter, varies within all populations and may provide essential information in future studies of schizophrenia.

Impact of complex genetic variation in COMT on human brain function

“…However, we suggest that this association also accommodates a functional interpretation, since it has been previously noted 10,20 that variation in this haplotype may be linked to the P2 promoter by asymmetric association of alleles: chromosomes with the intron one G allele almost always show the A allele at the promoter site, whereas the A allele at rs737865 was associated with both the A and G allele. Although there is incomplete LD between the intron 1 and P2 promoter SNP in our data and previously, 20 to the degree that this association holds in the population, the overtransmitted haplotype will be consistently linked to the same promoter allele. This is in agreement with the assumption that genetic variation at the P2 promoter contributes to the risk associated with this haplotype.…”

“…This is what was observed here (Supplementary Figure 3). Our data thus indicate interacting effects of these SNPs on prefrontal function by both coding and cisacting mechanisms, 20 whose combined effects are not linear. These results are also consistent with a recent study on schizophrenia risk indicating significant separate but interacting effects at two sites in strong LD with these loci.…”

“…11,36 In addition, we propose that our results provide a mechanistic explanation for inconsistencies in the clinical genetics literature of association with the val/met polymorphism. Owing to apparent balancing effects of functional variations in putative cis elements in COMT, which, like val/met, show considerable population variation in allele frequencies, 18,20 all val individuals do not show the same pattern of brain functional associations. Our data demonstrate that to the extent that the association of COMT with schizophrenia (or other psychoses) is related to prefrontal functional effects, only a subpopulation of val containing chromosomes will have inefficient prefrontal activity, and some met containing chromosomes will also have a similar functional profile.…”

“…Estimated allele frequencies were similar to previously reported results in Caucasians (Table 2). 11,16,17,20,29 Age, gender and task performance were balanced between genotype groups (Supplementary Table). With one exception, pairwise LD was significant and extensive, but not complete, underscoring the uncertainty of haplotype information (Table 3).…”

“…In support additional genetic of this hypothesis, a haplotype combining rs4680 with two common SNPs, one upstream in intron 1 (rs737865) and the other in the 3 0 untranslated region (rs165599), was highly associated with schizophrenia in a large sample of Israelis of Ashkenazi descent, 18 and differentially affected expression of rs4680 alleles in human brain tissue, 19 suggesting the presence of a cis acting functional locus within COMT interacting with val/met. Another population study found that this three marker haplotype is markedly heterogeneous in populations worldwide 20 despite the relatively constant prevalence of schizophrenia and suggested the relevance of another possible cis functional variant (rs2097603) linked upstream in the P2 promoter driving transcription of the predominant form of COMT in the brain (MB-COMT) (Figure 1). This variant also affects COMT activity in lymphocytes and post-mortem brain tissue.…”