An analysis of the interaction between MAPK14, CNR1 and marijuana misuse on white matter brain volume abnormalities in schizophrenia

Abstract

Background: Longitudinal follow-up of birth cohorts indicates that adolescent marijuana users are at heightened risk for developing schizophrenia compared to non-users. Current theories regarding substance dependence view drug misuse as a means to normalize a fundamental dopaminergic deficit within the reward circuitry. Predisposition for marijuana misuse and for schizophrenia may therefore share common genetic origins. We previously reported that heavy marijuana use in conjunction with specific cannabinoid receptor 1 (CNR1) gene variants (rs12720071-G-allele carriers) contributed to greater white matter brain volume deficits and cognitive impairment among schizophrenia probands. In the current study, we investigate the influence of another endocannabinoid-related gene (mitogenactivated protein kinase 14 (MAPK14)) on MRI brain morphometry in schizophrenia probands, and its inter-relationships with marijuana misuse and CNR1. MAPK14 encodes a member of the MAP kinases involved in diverse cellular processes, including cannabinoid receptorinduced apoptosis. Methods: We genotyped 235 schizophrenia patients on 9 tagged SNPs (tSNPs) accounting for a substantial proportion of MAPK14 genetic variability. Each patient also underwent a high-resolution anatomic brain MRI scan that provided measurements for lobar gray matter (GM) and white matter (WM) volumes. Almost half the sample had no lifetime marijuana use; one-third had marijuana exposure not meeting DSM abuse or dependence criteria, ~15% had marijuana abuse and ~8% marijuana dependence. The effects of single-marker MAPK14 tSNPs on brain volumes were assessed using ANCOVA (covariates: intracranial volume, gender, age, alcohol/non-marijuana illicit drug use and antipsychotic treatment). A genotype-by-marijuana misuse interaction term was included in the statistical models so as to assess differential MAPK14 influence on brain volumes across patients with versus without marijuana abuse/dependence. Additional haplotype analyses tested for potential MAPK14-CNR1 epistasis in conferring brain volume abnormalities among schizophrenia probands with marijuana abuse/dependence. Results: Among the 9 MAPK14 tSNPs, only rs12199654 had significant genotype effects on frontal (p=.003), temporal (p=.01) and parietal (p=.01) WM volumes. A-homozygotes had smaller WM volumes than rs12199654-G-allele carriers. There were also statistically significant rs12199654-by-marijuana misuse interaction effects on these lobar WM volumes (p=.003, .03 and .01 respectively). A-homozygotes with marijuana abuse/dependence had significantly smaller WM volumes than the other comparison groups. There were no significant rs12199654 genotype or genotype-by-marijuana misuse effects on GM volumes (p>.08). Between the four MAPK14-CNR1 (rs12199654- rs12720071) diplotypes, A-G and G-A haplotypes were significantly associated with WM brain volumes. The A-G haplotype, which comprises MAPK14 and CNR1 alleles associated with WM volume deficits, correlated with smaller frontal, temporal and parietal WM volumes (p