Abstract

A recent report detected association between GPR50, an orphan G protein-coupled receptor, and bipolar disorder (BD) in the Scottish population [29]. We sought to replicate this study in a second sample from the same population, consisting of 338 patients with BD, 359 patients with major depressive disorder (MDD) and 913 control individuals. In addition, the effect of GPR50 genotype on clinical phenotype and treatment response was assessed in a subset of 56 patients with early onset MDD (eoMDD). We identified an association with BD in women with an intronic SNP, rs1202874, that withstood correction for multiple testing (p = 0.0035, permuted p = 0.037, OR = 1.9, 95%CI 1.2–3.0). However, we failed to find an association with the previously associated Δ502-505 polymorphism (p = 0.2). Combined analysis of this and the original samples did detect association between the deletion and susceptibility to BD in females, but with a reduced effect size (p = 0.0006, permuted p = 0.0024, OR = 1.41, 95%CI 1.16–1.71). In the highly phenotyped eoMDD subgroup, we found an association between the Δ502-505 deletion polymorphism and age of onset (p = 0.049), number of episodes (p = 0.044), hypomanic symptoms (p = 0.019), and initial thinking time (p = 0.027), in women; and in family history of depression in men (p = 0.038), uncorrected for multiple testing. No association was seen between Δ502-505 genotype and treatment response at 3 months. To our knowledge this is the first association of rs1202874 with BD and is the second positive association at the GPR50 locus.