30 September 2008. Following a scientific thread first picked up two years ago from an abnormal chromosome of a single individual with schizophrenia and mental retardation, Ben Pickard and colleagues at the University of Edinburgh have identified a common insertion/deletion, or "indel," variant in which the deletion allele appears to exert a protective effect against bipolar disorder. In a paper published online September 29 in PNAS, they report that the variant lies within an untranslated region of GRIK4, a gene that codes for a subunit of KA1, a kainate-type glutamate receptor expressed in cerebral cortex and in the dentate gyrus of the hippocampus.
Since the glutamate system first piqued the interest of researchers in biological psychiatry by way of the phencyclidine model of schizophrenia (see SRF Current Hypothesis), dysregulated glutamatergic neurotransmission has emerged as a possible culprit in other psychiatric disorders, especially bipolar disorder, anxiety disorders, and obsessive-compulsive disorder (see, e.g., Pittenger et al., 2008). Though the functional aspects of kainate glutamate receptors are rather poorly understood compared to their NMDA and AMPA cousins, postmortem studies have shown abnormal kainate receptor expression in prefrontal cortex of individuals with schizophrenia (Meador-Woodruff et al., 2001; Sokolov, 1998), and differential expression of kainate receptor subunits in anterior cingulate cortex in schizophrenia versus bipolar disorder patients (Woo et al., 2007). Recent work at the National Institute of Mental Health with knockout mice suggests that the GluR6 subunit encoded by GRIK2 plays a role in mania (Shaltiel et al., 2008).
The tail end
The indel variant described in the new study is the latest upshot of a fruitful study published by the Edinburgh group in 2006 (see SRF related news story), in which a karyotypic analysis of an individual with schizophrenia and mental retardation revealed a chromosomal breakpoint in GRIK4 on chromosome 11. A follow-up case-control association study by the same team identified two haplotypes within GRIK4, one that confers a higher risk of schizophrenia and another, with greater statistical significance, that is associated with a protective effect against bipolar disorder. The proximity of the two SNPs in the latter haplotype to GRIK4’s predicted mRNA polyadenylation signal prompted the researchers to examine more closely the 3’ end of the gene.
In the newest study, individuals from the previous association study who were homozygous for the protective haplotype were sequenced across the final exon and 3’ untranslated region of GRIK4, and a 14-bp deletion was identified 33 bp downstream of the stop codon. Subsequent genotyping of the original cohort and a replication cohort, in total comprising 630 individuals with bipolar disorder and 662 healthy controls, showed a strong negative association of this deletion with the disease.
A comparison of the odds ratios for the protective effect of the deletion in heterozygous (OR = 0.4718) and homozygous (OR = 0.3258) individuals demonstrated an additive effect. To better explain this effect, the authors quantified the endogenous expression of insertion- and deletion-bearing mRNA in experimental cell lines and in postmortem human brain tissue from healthy control individuals. In all cases, they found a preponderance of deletion-bearing mRNA.
Since one of the cell lines used, the SHSY-5Y neuroblastoma line, is naturally heterozygous, the greater abundance of mRNA sequences exhibiting the deletion could not be attributed to the relative representation of the insertion and deletion alleles or to increased transcription of the deletion allele. Instead, the authors suggest that the modification to the 3' untranslated region induced by the deletion somehow renders that allele more resistant to cytoplasmic degradation.
Though the researchers present no definitive mechanism for greater resilience of the deletion allele, they present consensus secondary structures of the insertion- and deletion-containing GRIK4 alleles, and point out that, in the insertion sequence, “several regions exist as single strands and thus are potential targets of specific protein factor or miRNA binding” that could affect the stability or translational efficiency of the sequence. “The GRIK4 findings presented here,” they write, “raise the possibility that pharmacologically mediated increases in kainate receptor activity, mirroring the protective deletion’s effect on expression, might present a therapeutic opportunity for bipolar disorder and other psychiatric illnesses.”
The authors note that GRIK4 was not a "hit" in the Wellcome Trust Case Control Consortium genomewide association study (GWAS; see SRF related news story), and this has been a theme for other susceptibility gene candidates in GWASs of bipolar disorder and schizophrenia (see SRF related news story). However, the SNP chips used in GWASs were not created specifically to test psychiatric gene candidates, and in the case of GRIK4 in the Wellcome Trust study, the authors write, "only one (rs2282586) of the two SNPs that give rise to the strongest original associated protective haplotype (and only one of the five SNPs that compose the extended haplotype) was present on the chip."—Peter Farley.
Pickard BS, Knight HM, Hamilton RS, Soares DC, Walker R, Boyd JFK, Machell J, Maclean A, McGhee KA, Condie A, Porteous DJ, St. Clair D, Davis I, Blackwood DHR, Muir WJ. A common variant in the 3’UTR of the GRIK4 glutamate receptor gene affects transcript abundance and protects against bipolar disorder. Proc Natl Acad Sci USA. Sept. 29, 2008.