Family-based analysis of genetic variation underlying psychosis-inducing effects of cannabis: sibling analysis and proband follow-up

Arch Gen Psychiatry. 2011 Feb;68(2):148-57. doi: 10.1001/archgenpsychiatry.2010.152. Epub 2010 Nov 1.


Context: Individual differences exist in sensitivity to the psychotomimetic effect of cannabis; the molecular genetic basis underlying differential sensitivity remains elusive.

Objective: To investigate whether selected schizophrenia candidate single-nucleotide polymorphisms (SNPs) moderate effects of cannabis use.

Design: Interactions between recent cannabis use, determined by urinalysis results, and 152 SNPs in 42 candidate genes were examined in 740 unaffected siblings of 801 patients with psychosis to examine genetic moderation of the association between Structured Interview for Schizotypy-Revised positive schizotypy and recent cannabis use (at-risk paradigm). The SNPs showing Bonferroni-adjusted association in the at-risk paradigm were used in a case-only analysis in the 801 patients, as well as in a case-sibling and case-control analysis (using 419 controls) focusing on genetic moderation of developmental effects of cannabis on later psychotic disorder.

Setting: The Netherlands and Flanders, Belgium.

Participants: Eight hundred one patients with psychosis and their 740 unaffected siblings.

Main outcome measure: Significant interaction between any of the selected SNPs and cannabis in the at-risk paradigm, followed by selective case-only, case-sibling, and case-control analyses.

Results: In the unaffected siblings, 16 SNPs in 12 genes showed significant interaction at P < .05, 3 of which survived correction for multiple testing (P < .0003), situated in AKT1 (rs2494732 and rs1130233) and LRRTM1 (rs673871). Follow-up analysis supported AKT1 rs2494732 × cannabis interaction in the case-only (β = 0.20; P = .007), case-sibling (interaction P = .040), and case-control (interaction P = .057) analyses, with individuals with C/C genotypes having an approximately 2-fold odds of being diagnosed with a psychotic disorder when having used cannabis. In the unaffected siblings, the AKT1 × cannabis interaction explained 2.2% additional variance in schizotypy in the whole sample and 19.0% additional variance in the exposed siblings with recent cannabis use.

Conclusions: Genetic variation in AKT1 may mediate both short-term as well as longer-term effects on psychosis expression associated with use of cannabis, possibly through a mechanism of cannabinoid-regulated AKT1/GSK-3 signaling downstream of the dopamine D(2) receptor.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Belgium
  • Case-Control Studies
  • Dronabinol / toxicity*
  • Female
  • Follow-Up Studies
  • Gene Expression Regulation / genetics
  • Genetic Association Studies
  • Genetic Predisposition to Disease / genetics*
  • Genetic Variation / genetics*
  • Genotype
  • Humans
  • Male
  • Marijuana Abuse / genetics*
  • Middle Aged
  • Netherlands
  • Polymorphism, Single Nucleotide / genetics
  • Psychiatric Status Rating Scales
  • Psychoses, Substance-Induced / genetics*
  • Risk Factors
  • Schizotypal Personality Disorder / genetics
  • Siblings
  • Young Adult


  • Dronabinol