Symmetry and Disinhibition are Heritable Endophenotypes For Tourette Syndrome

Abstract

ObjectivePhenotypic heterogeneity in Tourette syndrome (TS) is partly due to complex genetic relationships between TS, Obsessive Compulsive Disorder(OCD) and attention deficit/hyperactivity disorder (ADHD). Part of this genetic overlap may be due to specific symptom clusters that are shared across disorders. Identification of such symptom-based cross-disorder phenotypes may aid gene-finding efforts. MethodSymptom-level factor and latent class analyses were conducted in 3494 individuals recruited for genetic studies through the Tourette Association of America International Consortium for Genetics (TAAICG) and replicated in an independent sample. Disease-specific analyses (TS, OCD, and ADHD) were performed first, followed by cross-disorder analyses. Heritability was estimated for each phenotype. TS-, OCD-, and ADHD-associated polygenic load were estimated for the cross-disorders phenotypes of interest. ResultsThe tic-only exploratory factor analysis identified six factors that followed the somatotopic map of the basal ganglia (F1: eye tics, F2: facial and head tics, F3: body/trunk tics, F4: complex vocal and socially inappropriate/disinhibited tics, F5: touching (tapping) tics, and F6: simple vocal tics. The ADHD analysis identified two factors: F1: inattentive and F2: hyperactive/impulsive symptoms, paralleling the DSM-5. The OCD factor analysis identified eight factors: F1: doubts/scrupulosity, F2: symmetry/exactness, F3: contamination/cleaning, F4: aggressive urges, F5: fear of harm, F6: need for sameness, F7: superstitions, and F8: hoarding.We identified two cross-disorder symptom-based phenotypes: symmetry (symmetry, evening up, checking; ordering, arranging, counting, -rewriting, repetitive writing tics) and disinhibition (uttering syllables/words, echolalia/palilalia, coprolalia/copropraxia and obsessive urges to offend/mutilate/be destructive), in addition to tic-only and OCD-only factors. Although all phenotypes were heritable across analyses, symmetry and disinhibition showed the most promise, with heritability estimates of 53% (p=6.3 x 10-10, p=1.7 x 10-18 respectively) comparable to a 58% heritability for DSM-5 based TS.Polygenic risk scores derived from a TS GWAS were associated with symmetry (p= 0.02), while risk scores derived from OCD and ADHD GWAS were not (negative correlations). In contrast, disinhibition was correlated with OCD PRS (p = 0.02) and showed a trend for association with TS and ADHD PRS (p = 0.11 and p = 0.10, respectively). CNV carriers had higher scores for both symmetry and disinhibition. Higher symmetry scores were seen in cases harboring large (≥1Mb), rare singleton duplications (p=0.005). Disinhibition scores were higher in cases who carried CNTN6 duplications (p=0.008), but not those who carried large singleton deletions. ConclusionsWe identified two heritable TS-related endophenotypes that cross traditional diagnostic boundaries. The symmetry phenotype correlated with TS polygenic load and CNV carrier status, and was present in otherwise “TS-unaffected” mothers, suggesting that this phenotype may reflect additional TS (rather than OCD) genetic liability that is not captured by traditional DSM-based diagnoses. The disinhibition phenotype, in contrast, may reflect more global top-down cognitive control deficits that are present in all three related disorders (TS, OCD, and ADHD).