Methylomic signature of current cannabis use in two first-episode psychosis cohorts

Abstract

The rising prevalence and legalisation of cannabis worldwide have underscored the need for a comprehensive understanding of its biological impact, particularly on mental health. Epigenetic mechanisms, specifically DNA methylation, have gained increasing recognition as vital factors in the interplay between risk factors and mental health. This study aimed to explore the effects of current cannabis use and high-potency cannabis on DNA methylation in two independent cohorts of individuals experiencing first-episode psychosis (FEP) compared to control subjects. The combined sample consisted of 682 participants (188 current cannabis users and 494 never users). DNA methylation profiles were generated on blood-derived DNA samples using the Illumina DNA methylation array platform. A meta-analysis across cohorts identified one CpG site (cg11669285) in the CAVIN1 gene that showed differential methylation with current cannabis use, surpassing the array-wide significance threshold, and independent of the tobacco-related epigenetic signature. Furthermore, a CpG site localised in the MCU gene (cg11669285) achieved array-wide significance in an analysis of the effect of high-potency (THC = > 10%) current cannabis use. Pathway and regional analyses identified cannabis-related epigenetic variation proximal to genes linked to immune and mitochondrial function, both of which are known to be influenced by cannabinoids. Interestingly, a model including an interaction term between cannabis use and FEP status identified two sites that were significantly associated with current cannabis use with a nominally significant interaction suggesting that FEP status might moderate how cannabis use affects DNA methylation. Overall, these findings contribute to our understanding of the epigenetic impact of current cannabis use and highlight potential molecular pathways affected by cannabis exposure.

Fig. 1. Differentially methylated positions (DMPs) identified in a meta-analysis of current frequent cannabis use and current frequent high-potency cannabis across two cohorts.

a Manhattan plot highlighting significant DMPs associated with current cannabis use from a comprehensive EWAS meta-analysis of two datasets (total n = 682 individuals). In total, 2 DMPs associated with cannabis use were identified at an experiment-wide significance level. The x-axis depicts chromosomes 1–22 and the y-axis the significance level (−log10(p)) for each DNA methylation site tested. The horizontal red line represents the experiment-wide significance level (p < 2.4E-07). Probe annotations are given for the top-ranked DMPs and the list of results is in Table 2 & Table S1. b Manhattan plot highlighting significant DMPs associated with current high-potency cannabis use (total n = 622 individuals). In total 2 DMPs associated with cannabis use were identified at an experiment-wide significance level. A list of results is given in Table 3 & Table S4.

Fig. 2. Pathway analyses revealed significant enrichment of biological pathways related to immune processes associated with current cannabis use.

Pathway analyses were performed on DMPs from the meta-analysis using the methylGSA package (see Methods). A significant enrichment of biological pathways involved in immune processes such as lymphocyte differentiation (padj=0.032) and B cell receptor signalling (padj=0.012) was observed.

Fig. 3. Differential DNA methylation in the promoter region of TOP1MT is associated with current frequent cannabis use.

Using comb-p [38] we identified regions of differential DNA methylation associated with current cannabis use including a differentially methylated region (DMR) comprising four probes located upstream of the TOP1MT gene (Sidak corrected p value = 3.88E-07). The figure depicts the specific probes exhibiting differential methylation patterns in both cohorts, with the meta-analysis p value plotted in the bottom panel. The beta value corresponds to the DNA methylation level ranging from 0 (unmethylated) to 1 (methylated). Groups are labelled as either 0 (control) or 1 (current cannabis user).