Associations between serotonin transporter and behavioral traits and diagnoses related to anxiety

Abstract

The role of the serotonin transporter promoter-linked polymorphism (5-HTTLPR) in psychiatric disease remains unclear. Behavioral traits could serve as alternative outcomes that are stable, precede psychopathology, and capture more sub-clinical variation. We test associations between 5-HTTLPR and (1) behavioral traits and (2) clinical diagnoses of anxiety and depression. Second and third generation participants (N=203, 34.2±13.8 years, 54% female) at high- or low- familial risk for depression (where risk was defined by the presence of major depression in the 1st generation) were assessed longitudinally using the Schedule for Affective Disorders and Schizophrenia-lifetime interview, Barratt Impulsiveness Scale-11, Buss-Perry Aggression Questionnaire, and the NEO-Five Factor Inventory. High (but not low)-risk offspring with two risk (short, s) alleles had higher impulsivity (+13%), hostility (+31%) and neuroticism (+23%). SS was associated higher rates of panic (OR=7.05 [2.44, 20.38], p=0.0003) and phobic (OR=2.68[1.04, 6.93], p=0.04), but not other disorders. Impulsivity accounted for 16% of associations between 5-HTTLPR and panic, and 52% of association between 5-HTTLPR and phobias. We show that 5-HTTLPR predicts higher impulsivity, hostility, and neuroticism, and that impulsivity could serve as a useful independent outcome or intermediary phenotype in genetic studies of anxiety.

Figure 1

Direct and Indirect Associations between 5-HTTLPR genotype and anxiety disorders The figure illustrates a potential path between the serotonin transporter S/S genotype (the predictor) and two groups of DSM anxiety disorders: panic disorder (upper panel) and phobic disorders (bottom). The path model tests whether the relationship between genotype and diagnoses is explained by impulsivity. The ‘A’ path represents the independent effect of genotype on impulsivity; the B path represents the independent effect of impulsivity on diagnosis; the ‘C’ path reflects the relationship between gene and diagnosis. To test the mediation hypothesis, the C path is ‘broken’ into two portions: a direct path which reflects the association between gene and diagnosis that is independent of impulsivity, and the indirect path, which reflects the relationship between gene and outcome that is dependent upon (i.e., explained by) impulsivity. The sum of the direct and indirect effect makes up the total effect (100%). And the percent of the total effect that is contributed to by the indirect path reflects the degree to which the association between gene and diagnosis is mediated by impulsivity. As shown in the C paths, approximately half (52%) of the association between 5-HTTLPR and phobias, and 16% of that between 5-HTTLPR and panic disorder, were explained by impulsivity. A note of interest is that even though the magnitude of the direct and indirect effects are similar for phobias, the confidence intervals for the direct effect are much larger, suggesting that the association between genotype and clinical outcome is more heterogeneous than that between genotype and trait.