Abstract

Smoking cannabis daily doubles an individual’s risk of developing a psychotic disorder, yet indicators of specific vulnerability have proved largely elusive. Genetic variation is one potential risk modifier. Single-nucleotide polymorphisms in the AKT1 and catechol-O-methyltransferase (COMT) genes have been implicated in the interaction between cannabis, psychosis and cognition, but no studies have examined their impact on an individual’s acute response to smoked cannabis. A total 442 healthy young cannabis users were tested while intoxicated with their own cannabis—which was analysed for delta-9-tetrahydrocannbinol (THC) and cannabidiol content—and also ±7 days apart when drug-free. Psychotomimetic symptoms and working memory were assessed on both the sessions. Variation at the rs2494732 locus of the AKT1 gene predicted acute psychotic response to cannabis along with dependence on the drug and baseline schizotypal symptoms. Working memory following cannabis acutely was worse in females, with some suggestion of an impact of COMT polymorphism on working memory when drug-free. These findings are the first to demonstrate that AKT1 mediates the acute response to cannabis in otherwise healthy individuals and implicate the AKT1 pathway as a possible target for prevention and treatment of cannabis psychosis.

Discussion

To our knowledge, this study provides the first evidence that the acute psychotic effects of cannabis are predicted by variation at the rs2494732 locus of the AKT1 genotype. No evidence was found for an interaction of the COMT Val158Met genotypes with cannabis use, in producing psychotomimetic symptoms in this group of healthy cannabis users. Cannabis dependence predicted non-intoxicated schizotypal symptoms, but neither genotype had any impact on these. COMT Val158Met genotype had a marginal impact on performance on a working memory task when non-intoxicated and when memory load was low; however, at higher load, schizotypy was the only emerging predictor of performance. When intoxicated with cannabis, gender was the only predictor of working memory performance, with poorer performance in females at a high working memory load.

In the current study, which is the largest ever to be conducted on the acute response to cannabis, psychotomimetic symptoms while intoxicated were found to be predicted by variation at the rs2494732 locus of the AKT1 genotype in healthy young cannabis smokers, increasing with C allele dosage. These data are very important as acute psychotic response to cannabis is thought to be a marker of the risk of developing psychosis from smoking the drug.1 Two previous studies have implicated this polymorphism in the interaction with cannabis and psychosis,9, 18 but this work concentrated on individuals who were at familial risk of schizophrenia. This study is the first to demonstrate that the acute response to cannabis is modulated by AKT1 in otherwise healthy cannabis smokers. The mechanism for this modulation of acute effects may be through the interaction of AKT1 with dopamine.2, 9Our sensitivity analyses suggested that these effects may be confined to dependent cannabis smokers but further investigation of these data with larger samples is required.

AKT1 codes for a protein that is a serine/threonine kinase, which has a variety of functions, one of which is as a signalling molecule downstream of the dopamine D2 (DRD2) receptor. Decreased AKT1 functionality may result in enhanced responses to DRD2 receptor stimulation.19 THC has been found to acutely induce dopamine release in

rats20, 21 and in humans,22, 23 although not in all studies.24 Dopamine release is thought to occur via the blockade of cannabinoid 1 (CB1) receptors on GABAergic neurons that target pyramidal cells. These neurons normally exert an inhibitory effect on the firing of dopamine neurons that project back to the nucleus accumbens, so agonism of CB1 receptors by THC may produce increased dopamine release. This THC-mediated increase in dopamine release may be further exacerbated by decreased AKT1 functionality. Elevated levels of mesolimbic dopamine are known to have a role in the development of psychotic symptoms, potentially through disrupted salience attribution.25

In contrast to the role of variation at the rs2494732 locus of AKT1, this study found no support for the direct involvement of the functional polymorphism of the COMT gene in mediating acute psychotic response to cannabis. This is in contrast to one previous small-scale acute laboratory study giving acute THC to patients with schizophrenia,26 and other work that suggested that COMT may mediate the psychotomimetic risk of cannabis3 but in agreement with subsequent larger studies that failed to replicate these findings.4, 27 There was a marginal effect of COMT on working memory performance at a low load when not intoxicated. This polymorphism of COMT initially caused some excitement as several studies emerged demonstrating its association with working memory,28, 29 but this finding was not confirmed by meta-analyses,30 which suggested that this may be a case of publication bias.

Greater schizotypal symptoms predicted in poorer working memory performance on the more difficult section of the task among drug-free cannabis users. This echoes recent findings of poorer working memory in individuals high in schizotypy31 and indeed of the relationship between working memory performance and transition to psychosis.32 Working memory impairment is considered a central cognitive impairment in schizophrenia, and there is some evidence that such impairments are related to symptoms, particularly to negative symptoms.33, 34

Only gender predicted acute working memory impairment from cannabis, with greater impairment in females. Very few studies have examined gender differences in neurocognitive acute response to THC, with those that have using very small samples and in finding little evidence of gender differences.35 However, this study examined the acute effects of cannabis in over 400 cannabis smokers. There is an emerging preclinical literature that might explain this effect. CB1 density has been found to vary by gender, with animal studies reporting greater CB1 receptor density among males across several brain regions.36, 37 However, across their lifetime, adult female brains show increases in CB1 receptor density, with levels eventually surpassing those observed in males.38 Furthermore, greater CB1 de-sensitization after exposure to THC in the prefrontal cortex, hippocampus, striatum, amygdala and midbrain is seen in female adolescent rats.36, 37 Preclinical studies have also demonstrated that female rats preferentially metabolize THC to its most highly active metabolite, while male rats metabolize THC to multiple compounds.39 In combination, these findings may in part explain the finding of greater acute working memory impairment from cannabis in females. This also may partly be driven by gender differences in frequency of cannabis use. Users who smoked cannabis less frequently experienced stronger effects, and as there was a higher proportion of low frequency female cannabis users compared to males this may have contributed to the observed gender differences in working memory following the drug.

Strengths of this study include the large sample size for assessing acute cannabis effects. We also used independent verification of the cannabinoid content of the cannabis consumed and drug history. Further, the hypothesis-driven approach we took to genetic analysis was a strength, examining only loci implicated in previous studies and, therefore, circumventing some of the problems of type I error that have dogged earlier research. However, inevitably there are several limitations of the study. For the cannabis use data, while verifying past 3 months use with hair analysis, we inevitably relied on retrospective self-reports of drug use, which are particularly complicated as cannabis is known to acutely impair episodic memory. However, we opted to use years of cannabis use in this model as this was considered the most reliable to estimate. As we purposely recruited a young group of cannabis users, there was restricted variation in years used and future studies may investigate this further. We used a predominantly white Caucasian sample. However, it is unlikely that ethnic differences in allele frequency at rs2494732 biased the outcome of the study, as there was no difference between the frequency of rs2494732 alleles across the dichotomized ethnic groups. In addition, analyses with only Caucasian participants gave the same results to the analysis containing all ethnicities.

In summary, we found that the AKT1 rs2494732 C allele was associated with increased psychotomimetic symptoms after smoking cannabis. The other factor impacting on acute psychotomimetic response to cannabis was baseline schizotypy. Gender was the only factor to predict acute working memory impairment, with poorer performance in females. When drug free, cannabis dependence weakly predicted schizotypal symptoms and COMT genotype had a marginal impact on working memory, along with ethnicity. The findings of this study contribute to a recent and growing body of evidence suggesting that variation at the AKT1 locus confers details of the risk of cannabis smoking for schizophrenia. This is likely to be in the context of numerous other genetic variants, so the clinical utility at the moment is unclear. It is nonetheless encouraging that there is concordance between genetic influences on acute effects of cannabis and those mediating risk of psychosis. However, the fact that AKT1 is relevant to the biology of psychotic symptoms suggests that this might be a promising direction for novel therapeutics for cannabis-induced psychosis.

Source:  Citation: Translational Psychiatry (2016) 6, e738; doi:10.1038/tp.2015.219 Published online 16 February 2016 

For complete paper log on to: http://www.nature.com/tp/journal/v6/n2/full/tp2015219a.html