Cannabis use alters DNA methylation, with implications beyond smoking effects

By Dr. Chinta SidharthanNov 8 2023Reviewed by Susha Cheriyedath, M.Sc.

In a recent study published in the journal Molecular Psychiatry, a team of researchers conducted a large-scale meta-analysis consisting of an epigenome-wide association study to understand whether lifetime use of cannabis was linked to deoxyribonucleic acid (DNA) methylation observed in peripheral blood.

Study: Trans-ancestry epigenome-wide association meta-analysis of DNA methylation with lifetime cannabis use. Image Credit: Juan Gaertner / Shutterstock

Background

With an increasing number of states in the United States (U.S.), as well as countries across the world legalizing the medicinal use of cannabis, cannabis usage has become exceedingly prevalent. However, while the therapeutic benefits of cannabis through medicinal use have been well-studied, its recreational use also raises numerous concerns, especially regarding problems associated with addiction, cognitive deficits, and mental health disorders such as depression, anxiety, psychosis, mania, and schizophrenia.

DNA methylation is an indicator of the impact of environmental factors on health, and some forms of DNA methylation due to environmental factors are long-lasting, while others are transient. It occurs when a methyl group gets added to the fifth carbon of cytosine in regions with cytosine and guanine (CpG) repeats. Studies have reported that cigarette smoking results in both persistent and transient DNA methylation at CpG sites across the genome. DNA methylation patterns in specific genes have also been observed in groups such as adolescents who frequently use cannabis and patients who are dependent on cannabis.

About the study

In the present study, the team built on the methods from their previous study, where they performed the first epigenome-wide association study using peripheral blood samples and investigated a large study population consisting of seven cohorts of individuals of different ancestries. They examined the association between lifetime cannabis use and DNA methylation patterns while adjusting for factors such as age, sex, technical covariates, blood cell proportions, and cigarette smoking behavior.

The data was obtained from seven cohorts that participated in the study, spanning diverse study groups such as twins, older adults, parents and children, and adult twins. The final study population comprised 4,190 individuals who reported using cannabis in their lifetime and 5,246 individuals who had never used cannabis in their lifetime, constituting a total of 9,436 participants. Cannabis use was characterized based on reports by the participants or parents, and an individual with a minimum of one cannabis use event before the collection of peripheral blood samples was defined as an ever-user.

DNA methylation was measured in the peripheral blood samples, and the β-values, which is the percentage of methylated DNA at the targeted CpG sites, were calculated. The association between lifetime cannabis use and DNA methylation levels was tested using linear models or a generalized estimating equations model in cohorts where the participants were related. The epigenome-wide association study analyses were stratified according to the European-American and African-American genetic ancestry groups, and the analyses were adjusted for sex, age, cigarette smoking, and blood cell type estimates.

The meta-analysis summarized the ancestry and cohort-specific results from the epigenome-wide association study, and statistical analyses were conducted to assess. The methylation score, which is the weighted sum of the CpG sites significantly linked to cannabis use, was also calculated. Additionally, DNA methylation correlations between whole blood and regions of the brain, such as the prefrontal cortex, cerebellum, superior temporal gyrus, and entorhinal cortex were examined.

Results

Four CpG sites showed significant associations with cannabis use, none of which were previously reported as being significant in cigarette smoking epigenomic analyses. These included CpG sites in the disintegrin and metalloprotease 12 (ADAM12) and alpha-actinin 1 (ACTN1) genes and near the adhesion G protein-coupled receptor F1 (ADGRF1) gene and long noncoding ribonucleic acid (RNA) LINC01132. In addition to these four CpG sites, upon analyzing the blood of lifetime cannabis users who had never smoked cigarettes, an additional CpG site of cg14237301 in the apolipoprotein B receptor (APOBR) gene was identified.

Importantly, the CpGs associated with cannabis use in this study showed significant overlap with CpGs previously associated with cigarette smoking. Other enriched traits that significantly overlapped with the results of this study included those observed in Crohn’s disease, alcohol consumption, body mass index (BMI) and multiple sclerosis.

To elucidate the role of an individual’s genetics in these DNA methylation changes, the researchers examined the methylation quantitative trait loci (meQTLs) for the five CpGs that were significantly associated with lifetime cannabis use. None of these meQTLs were significantly associated with cannabis use, thus indicating that the study findings are likely not influenced by genetic variants.

The five genes containing DNA-methylated CpG sites associated with cannabis use have significant roles in various health outcomes. For example, LINC01132 functions as an oncogene and is linked to malignancy in hepatocellular carcinoma and ovarian cancer, although cannabis use has been reported to lower the incidence of hepatocellular carcinomas.

Genetic variation in the ACTN1 gene, which encodes the cytoskeletal protein that binds actin fibers to cell membranes, has been linked to various diseases such as Bowen disease, Angelman syndrome, systemic lupus erythematosus, and congenital macrothrombocytopenia, as well as coronavirus disease 2019 (COVID-19). Genetic variants of APOBR have been associated with obesity, bladder cancer, pneumonia, allergies, and lifetime cannabis use.

Conclusions

Five significant cigarette smoking-independent CpGs were significantly associated with cannabis use. These CpG sites have also been implicated in cigarette smokers; therefore, both cannabis use and cigarette smoking may independently influence DNA methylation at these CpGs.

A key advantage of the current study is its large and diverse study cohort. However, this heterogeneity may have prevented the researchers from identifying ancestry-specific DNA methylation associations in cannabis users.

Overall, the study findings provide important insights into the shared DNA methylation profiles between cannabis users and cigarette smokers. Nevertheless, future studies are needed to better understand the potential health impacts of DNA methylation at these CpG sites.