Hill and Bray report mRNA expression analysis of the schizophrenia risk gene ZNF804A in adult and fetal brain. Importantly, they used an allele-specific expression assay that allows them to isolate cis-acting effects influencing ZNF804A expression (cis effects on expression are those which are restricted to the particular copy of a gene carrying the feature that influences expression, for example, a regulatory polymorphism at the gene locus) from artifacts related to RNA quality and more general trans-effects on gene expression (trans-effects influence expression of both copies and may arise as a result of environmental exposures such as drugs, or variable levels of a transcription factor). As a result, they were able to sensitively search for possible relationships between gene expression and genotype at rs1344706, the ZNF804A single-nucleotide polymorphism (SNP) that is currently the most strongly schizophrenia-associated variant (see Williams et al., 2011). Previous work by Williams and colleagues using similar methodology to...  Read more

Hill and Bray report mRNA expression analysis of the schizophrenia risk gene ZNF804A in adult and fetal brain. Importantly, they used an allele-specific expression assay that allows them to isolate cis-acting effects influencing ZNF804A expression (cis effects on expression are those which are restricted to the particular copy of a gene carrying the feature that influences expression, for example, a regulatory polymorphism at the gene locus) from artifacts related to RNA quality and more general trans-effects on gene expression (trans-effects influence expression of both copies and may arise as a result of environmental exposures such as drugs, or variable levels of a transcription factor). As a result, they were able to sensitively search for possible relationships between gene expression and genotype at rs1344706, the ZNF804A single-nucleotide polymorphism (SNP) that is currently the most strongly schizophrenia-associated variant (see Williams et al., 2011). Previous work by Williams and colleagues using similar methodology to that of Hill and Bray had shown that in adult brain, chromosomes carrying the schizophrenia risk allele at rs1344706 express ZNF804A at a higher level than those which do not carry the risk allele. However, the relationship between higher expression and genotype at rs1344706 was not perfect. Moreover, since another polymorphism at the ZNF804A locus was much better correlated with gene expression than rs1344706, but was only weakly associated with schizophrenia risk, Williams and colleagues concluded that the observed overexpression was not likely to be the mechanism underpinning the association between ZNF804A and schizophrenia.

In the present study, Bray and Hill reach the same conclusion with respect to adult brain. However, when they extended their work into fetal brain, a very different picture emerged. Specifically, in second-trimester fetal brain, chromosomes carrying rs1344706 were underexpressed. This is in stark contrast to the findings of higher expression in adult brain. Moreover, they also found that genotype at rs1344706 was strongly correlated with cis-measures of ZNF804A expression to an extent that is consistent with, although does not prove, the hypothesis that rs1344706 is, per se, the functional variant. The finding of a strong correlation between a schizophrenia risk allele and a gene function that is observed only in the second trimester of fetal development but not in adult brain supports the widely held hypothesis that the origins of schizophrenia, at least in part, begin during brain development. That conclusion is necessarily somewhat tentative since it is not established that the observed changes in expression are relevant to disease. However, the absence of protein-coding variants to explain the ZNF804A association, and the evidence from earlier work by the same authors (Hill and Bray, 2011) showing that binding of as yet unknown nuclear proteins to ZNF804A DNA is sensitive to genotype at rs1344706, make a good circumstantial case that it may be. Regardless, the demonstration of developmental specificity in the functional associations of a strongly implicated schizophrenia risk variant emphasizes the importance of studying regulatory effects of risk alleles at a range of developmental time points, and point to the need to augment genetic studies of gene expression with larger-scale studies using tissues from a range of developmental time points.

References:

Hill MJ, Bray NJ. Allelic differences in nuclear protein binding at a genome-wide significant risk variant for schizophrenia in ZNF804A. Mol Psychiatry . 2011 Aug 1 ; 16(8):787-9. Abstract

Williams HJ, Norton N, Dwyer S, Moskvina V, Nikolov I, Carroll L, Georgieva L, Williams NM, Morris DW, Quinn EM, Giegling I, Ikeda M, Wood J, Lencz T, Hultman C, Lichtenstein P, Thiselton D, Maher BS, , Malhotra AK, Riley B, Kendler KS, Gill M, Sullivan P, Sklar P, Purcell S, Nimgaonkar VL, Kirov G, Holmans P, Corvin A, Rujescu D, Craddock N, Owen MJ, O'Donovan MC. Fine mapping of ZNF804A and genome-wide significant evidence for its involvement in schizophrenia and bipolar disorder. Mol Psychiatry . 2011 Apr ; 16(4):429-41. Abstract