RESUMEN
BACKGROUND: The link between cannabis use and schizophrenia is well-established in epidemiological studies, especially among adolescents with early-onset use. However, this association in rodent models is less clear. This meta-analysis examined the effects of adolescent cannabinoid exposure on distinct schizophrenia-like behaviours in rodents and how experimental variations influence outcomes. METHODS: Following a pre-registered protocol (CRD42022338761), we searched PubMed, Ovid Medline, Embse and APA PsychInfo for English-language original studies until May 2024. We synthesised data from experiments on schizophrenia-like behaviour in rats and mice after repeated peri-pubertal (onset between P23-P45) cannabinoid exposure. Risk of bias was assessed using the SYRCLE's tool. RESULTS: We included 359 experiments from 108 articles across 9 behavioural tests. We found meta-analytic evidence supporting that CB1R agonists, both natural and synthetic, elicited broad schizophrenia-like behavioural alterations, including impaired working memory [g = -0.56; (CI: -0.93, -0.18)], novel object recognition [g = -0.66; (CI: -0.97, -0.35)], novel object location recognition [g = -0.70; (CI: -1.07, -0.33]), social novelty preference [g = -0.52; (CI: -0.93, -0.11)], social motivation [g = -0.21; (CI: -0.42, -0.00)], pre-pulse inhibition [g = -0.43; (CI: -0.76, -0.10)], and sucrose preference [g = -0.87; (CI: -1.46, -0.27)]. By contrast, effects on novelty-induced locomotion were negligible. Subgroup analyses revealed similar effects across sexes and species. Substantial variance in the protocols and moderate-to-high heterogeneity in behavioural outcomes were observed. We found CBD may enhance fear memory recall, but data was limited. DISCUSSION: This is the first meta-analysis to comprehensively assess the link between cannabinoids and schizophrenia-like behaviours in rodents. Our results support epidemiological links between early cannabis use and schizophrenia-like phenotypes, confirming the utility of animal models. Standardising protocols will optimise models to strengthen reproducibility and comparisons, our work provides a framework for refining rodent models to elucidate biological pathways linking cannabis and schizophrenia.
RESUMEN
Individuals with schizophrenia frequently experience co-occurring substance use, including tobacco smoking and heavy cannabis use, and substance use disorders. There is interest in understanding the extent to which these relationships are causal, and to what extent shared genetic factors play a role. We explored the relationships between schizophrenia (Scz), cannabis use disorder (CanUD), and ever-regular tobacco smoking (Smk) using the largest available genome-wide studies of these phenotypes in individuals of African and European ancestries. All three phenotypes were positively genetically correlated (rgs = 0.17 - 0.62). Causal inference analyses suggested the presence of horizontal pleiotropy, but evidence for bidirectional causal relationships was also found between all three phenotypes even after correcting for horizontal pleiotropy. We identified 439 pleiotropic loci in the European ancestry data, 150 of which were novel (i.e., not genome-wide significant in the original studies). Of these pleiotropic loci, 202 had lead variants which showed convergent effects (i.e., same direction of effect) on Scz, CanUD, and Smk. Genetic variants convergent across all three phenotypes showed strong genetic correlations with risk-taking, executive function, and several mental health conditions. Our results suggest that both horizontal pleiotropy and causal mechanisms may play a role in the relationship between CanUD, Smk, and Scz, but longitudinal, prospective studies are needed to confirm a causal relationship.
RESUMEN
Individuals with schizophrenia frequently experience co-occurring substance use, including tobacco smoking and heavy cannabis use, and substance use disorders. There is interest in understanding the extent to which these relationships are causal, and to what extent shared genetic factors play a role. We explored the relationships between schizophrenia (Scz; European ancestry N = 161,405; African ancestry N = 15,846), cannabis use disorder (CanUD; European ancestry N = 886,025; African ancestry N = 120,208), and ever-regular tobacco smoking (Smk; European ancestry N = 805,431; African ancestry N = 24,278) using the largest available genome-wide studies of these phenotypes in individuals of African and European ancestries. All three phenotypes were positively genetically correlated (rgs = 0.17-0.62). Genetic instrumental variable analyses suggested the presence of shared heritable factors, but evidence for bidirectional causal relationships was also found between all three phenotypes even after correcting for these shared genetic factors. We identified 327 pleiotropic loci with 439 lead SNPs in the European ancestry data, 150 of which were novel (i.e., not genome-wide significant in the original studies). Of these pleiotropic loci, 202 had lead variants which showed convergent effects (i.e., same direction of effect) on Scz, CanUD, and Smk. Genetic variants convergent across all three phenotypes showed strong genetic correlations with risk-taking, executive function, and several mental health conditions. Our results suggest that both shared genetic factors and causal mechanisms may play a role in the relationship between CanUD, Smk, and Scz, but longitudinal, prospective studies are needed to confirm a causal relationship.