New vistas on cannabis use disorder

Abstract

Cannabis sativa preparations are the most consumed illicit drugs for recreational purposes worldwide, and the number of people seeking treatment for cannabis use disorder has dramatically increased in the last decades. Due to the recent decriminalization or legalization of cannabis use in the Western Countries, we may predict that the number of people suffering from cannabis use disorder will increase. Despite the increasing number of cannabis studies over the past two decades, we have gaps of scientific knowledge pertaining to the neurobiological consequences of long-term cannabis use. Moreover, no specific treatments for cannabis use disorders are currently available. In this review, we explore new research that may help fill these gaps. We discuss and provide a solution to the experimental limitation of a lack of rodent models of THC self-administration, and the importance this model can play in understanding the neurobiology of relapse and in providing a biological rationale for potential therapeutic targets. We also focus our attention on glial cells, commenting on recent preclinical evidence suggesting that alterations in microglia and astrocytes might contribute to the detrimental effects associated with cannabis abuse. Finally, due to the worrisome prevalence rates of cannabis use during pregnancy, we highlight the associations between cannabis use disorders during pregnancy and congenital disorders, describing the possible neuronal basis of vulnerability at molecular and circuit level. This article is part of the Special Issue entitled "A New Dawn in Cannabinoid Neurobiology".

Keywords: Cannabis use disorder; Glia cells; Perinatal cannabis; Reward; THC self-administration.

Figure 1. Sample data from the rat model of THC + CBD self-administration and cue- or context-induced drug-seeking

Rats were exposed to 5 days of THC + CBD vapor prior to initiating intravenous THC + CDB daily self-administration sessions. Additionally, rats were food-trained in the absence of cues for a 1 hr prior to beginning self-administration. A 20 sec time out period was used and each infusion was associated with a 2 sec light and tone combined cue. A) Responses on the active and inactive levers, and number of drug infusions during the self-administration protocol. B) N=8 rats that underwent 10 days of abstinence prior to initiating context extinction training. The data shown are the active and inactive lever presses initiated during the first day of extinction training. Data are shown as mean ± sem and were analyzed using a Mann-Whitney U test, p= 0.010. C) Active and inactive lever pressing during context extinction training. D) N=6 rats went directly into extinction training for 10 days. These extinction and reinstatement data were pooled with the data from the animals shown in Panel C that were extinguished and cue-reinstated after the context-induced drug seeking session. Data are shown as mean ± sem and statistically analyzed using a two-way repeated measures ANOVA. Lever F_(1,13)= 8.63, p= 0.012; Ext vs Cue F_(1,13)= 9.28, p= 0.009; Interaction F_(1,13)= 5.82, p= 0.031. *p< 0.05, comparing all groups to active lever pressing induced by context (panel B) or cue (panel D).

Figure 2. Schematic hypothesis of glial activations following chronic THC treatment at the “quadpartite” synapse

Overactivation of CB1 receptors on synaptic terminals might lead to CB1 receptor downregulation and/or desensitization that is compensated during the treatment but is unmasked when treatment is discontinued. In this specific withdrawal window, the decreased control exerted by CB1 receptors on the release activity at the level of glutamatergic terminals might produce an extended presence of glutamate in the synaptic cleft. Increased glutamate release activates mGluRs associated with the postsynaptic membrane, leading to formation and release of endocannabinoids from the postsynaptic terminal (Hashimotodani et al. 2008). Endocannabinoids released from pyramidal neurons, acting through CB1 receptors on astrocytes, can increase astrocyte Ca2+ levels stimulating the release of glutamate from these cells (Navarrete et al. 2014), thus contributing/sustaining excitotoxicity. In addition, increased synaptic glutamate activates glutamate (AMPA) receptors on microglial cells, promoting microglia activation and IL-1β/TNF-α release (Domercq et al. 2013) that might contribute to the learning disabilities associated with chronic THC intake (Cutando et al. 2013; Zamberletti et al. 2015). Remarkably, activated microglial cells rapidly overexpress CB2 receptors (Carlisle et al. 2002; Stella, 2010; Walter et al. 2003), whose stimulation has been shown to modulate microglial reactivity, chemotaxis, proliferation, phagocytosis, migration, promoting neuroprotection (Carrier et al. 2004; Dirikoc et al. 2007; Eljaschewitsch et al. 2006; Ramirez et al. 2005; Walter et al. 2003). Thus, the overall effect of stimulating CB2 receptors via enhancement of endocannabinoid signaling or exogenous agonists would dampen microglia activation or skew microglial cells towards a neuroprotective phenotype (Navarro et al. 2016).

Figure 3. Diagram illustrating the risk factors leading to an at-risk endophenotype for SUD

The interaction among biological make up of the individual, environment (e.g. one or both parental SUD, parental neglect, peer influence, etc) and age-related effects of indirect (i.e. pre/peri-natal exposure to drugs of abuse such as cannabis derivatives) and direct (early onset SUD, including CUD) results in epigenetic modifications and changes at cellular and synaptic level that contribute to the development of an at-risk endophenotype for SUD.