Genetics of emotional reactivity in bipolar disorders

- First genome wide association study in bipolar disorder using measurement of affect intensity.
- Use of a quantitative intermediate phenotype can facilitate the identification of genetic susceptibility variants.
- General regression model is powerful to detect association, even in samples of moderate size.
- Four genetic variants accounts for around 25% of the affect intensity variance.
- Two main genes were identified, both involved in the modulation of immuno-inflammation processes.

BackgroundEmotional reactivity has been proposed as a relevant intermediate phenotype of bipolar disorder (BD). Our goal was to identify genetic factors underlying emotional reactivity in a sample of bipolar patients.MethodsAffect intensity (a proxy measure of emotional reactivity) was measured in a sample of 281 euthymic patients meeting DSM-IV criteria for BD. We use a validated dimensional tool, the 40-item self-report Affect Intensity Measure scale developed by Larsen and Diener. Patients with BD were genotyped for 475. 740 SNPs (using Illumina HumanHap550 Beadchips or HumanHap610 Quad chip). Association was investigated with a general mixed regression model of the continuous trait against genotypes, including gender as covariate.ResultsFour regions (1p31.3, 3q13.11, 11p15.1 and 11q14.4) with a p-value lower or equal to 5×10−6 were identified. In these regions, the joint effect of the four variants accounted for 24.5% of the variance of AIM score. Epistasis analysis did not detect interaction between these variants. In the 11p15.1 region, the rs10766743 located in the intron of the NELL1 gene remained significant after correction for multiple testing (p=2×10−7).ConclusionsThese findings illustrate that focusing on quantitative intermediate phenotypes can facilitate the identification of genetic susceptibility variants in BD.