This study investigated the modulatory effect of a selective CB2 receptor agonist, JWH-133, on the neurochemical and behavioural changes induced by cocaine in mice and explored the role of CX3CL1 in mediating microglial respiration and release of the cytokines, TNFα and IL-1β. JWH33 attenuated the reinstatement of cocaine-induced conditioned place preference (CPP) in mice without affecting their episodic memory. In addition, JWH-133 increased Δ-FOSB expression in the Nac, increased CX3CL1 levels in the VTA and PFC, and decreased IL-1β in the PFC and Nac in cocaine-treated animals. Furthermore, stimulation of primary microglial cells with CX3CL1 induced a biphasic effect; at a low concentration (50 nM) CX3CL1 enhanced glycolysis with no significant effect on ATP-linked respiration, decreased IL-1β release and had no effect on TNFα release. However, at a concentration of 100 nM, CX3CL1 increased mitochondrial oxidative metabolism without modifying glycolysis and increased both IL-1β and TNFα release.
Although CB2 receptors were initially thought to be expressed predominantly in peripheral tissues, recent studies demonstrate that they are also expressed in the central nervous system principally in the mesocorticolimbic dopamine pathway, in glia and neuronal cells, and functionally modulate motivational and motor behaviour (Jordan and Xi, 2019, Zhang et al., 2014). Thus, CB2 receptor agonists were suggested as a potential treatment for drug addiction, relapse and craving (Yang et al., 2012, Onaivi et al., 2008).
In our experiments cocaine produced a robust preference for the drug-paired compartment and, at a lower dose, evoked reinstatement of CPP after extinction, corroborating earlier studies (Ródenas-González et al., 2021, Shippenberg and Heidbreder, 1995). Cocaine-induced CPP reinstatement was inhibited by JWH-133 at a dose of 10 mg/kg which abolished both chamber preference and cocaine-induced locomotor activity. We should note that JWH-133 had no effect on locomotion in the absence of cocaine, which excludes the possibility of an unspecific inhibitory effect of JWH-133 on locomotion. These data extend previous studies that showed that activation of CB2 receptors by JWH-133 (Delis et al., 2017) or β-caryophyllene (Al Mansouri et al., 2014) reduces the reinforcing effect of cocaine in the CPP. Notably, it has been demonstrated that JWH-133 administered systemically or directly infused into the Nac decreases cocaine-elevated dopamine in rodent Nac (Xi et al., 2011; Zhang et al., 2017). These JWH-133 effects appear to depend on the activation of the CB2 receptors; as they were inhibited by pre-administration of AM630, a CB2 antagonist (Galaj et al., 2020). Accordingly, electrophysiological evidence of brain CB2 receptors modulation ex vivo and in vivo showed that CB2 receptor agonists may reduce mesocorticolimbic dopaminergic neuron firing (Zhang et al., 2019, Zhang et al., 2017). Thus, the effects of JWH-133 on the mesocorticolimbic dopaminergic function could explain the general effects of JWH-133 on the inhibition of CPP reinstatement with cocaine and on cocaine evoked locomotor activity and give further support to the presence of functional CB2 receptors in the brain.
Since CPP depends on; the integrity of neural circuits responsible for declarative memory and the endocannabinoid system is implicated in memory consolidation, contextual memory and spatial learning (Huston et al., 2013, Prus et al., 2009, Cheer et al., 2000, Lopes et al., 2020), we examined the cognitive effect of JWH-133. Our results showed that cocaine (10 mg/kg) conditioning followed by extinction and cocaine (5 mg/Kg) reinstatement did not alter the performance of mice in the novel object recognition test. In contrast, previous studies showed that prolonged cocaine consumption causes deficits in object recognition memory in rats (Briand et al., 2008). The difference in results may have been due to the cocaine conditioning period being shorter in our study, insufficient to cause significant long-term changes in memory. Alternatively, the long extinction period (10 days), may have negated any short-term effects of cocaine conditioning on episodic-like memory. Similarly, treatment with the JWH-133 had no effect on the episodic-like memory which indicates that the prevention of reinstatement of cocaine-induced CPP by the CB2 agonist was not due to an altered cognitive function. Also, the apparent lack of a JWH-133 cognitive effect does not depend on altered motor activity since it did not reduce spontaneous locomotion. Therefore, the effect of the CB2 agonist was likely specific to the pathways involved in the cocaine reinforcement.
On the other hand, JWH-133 prevented the increase in Δ-FOSB in the Nac mirroring its preventive effect of cocaine-induced reinstatement, as elevated expression of Δ-FOSB is necessary for plasticity changes in the Nac and subsequent cocaine-related behaviour.
Although this result does not necessarily imply that the Nac is the direct site of action of this drug, it strengthens the case for the CB2 agonist attenuating the specific neurochemical changes induced by cocaine, as Δ-FOSB is considered a “master control protein” in the development and maintenance of addiction (Robison and Nestler, 2011). Previous studies show that increased expression of Δ-FOSB in transgenic mice is associated with increased response to the rewarding and locomotor-activating effects of cocaine (Chen et al., 1997; Kelz et al., 1999). However, it is important to note that the effect of cannabinoid drugs on cocaine-naïve mice was not measured in our study, so it could be that CB2 activation decreases Δ-FOSB expression in the Nac regardless of prior cocaine exposure.
As CX3CL1 expression may have a neuroprotective role, its decrease in cocaine conditioning and reversal with JWH-133 treatment could be an underlying mechanism by which CB2 agonism had its behavioural effects. Relevant studies suggest that cocaine exerts its behavioural stimulant effects by eliciting neural autophagy in vitro and in vivo and showed that autophagy inhibition impairs cocaine-CPP in mice (Guo et al., 2015, Kim et al., 2021, Harraz et al., 2021a). Cocaine-induced autophagy degrades transporters for dopamine in the Nac (Harraz et al., 2021a). Moreover, it has been shown that autophagy could alter CX3CL1 expression by neurons and consequently change microglial inflammatory activity (Su et al., 2016, He et al., 2019). This suggests that the JWH-133 effect might also be attributed to its downregulation of autophagy (Angelina et al., 2022). Unfortunately, most studies on the implication of the endocannabinoid system in autophagy have been performed in tumoral models (Lee et al., 2021, Angelina et al., 2022). Thus, further studies are needed to establish how cannabinoidergic signalling is implicated in autophagy in models of cocaine-induced toxicity and behavioural changes.
CX3CL1 acts as a “find-me” signal released by neurons undergoing apoptosis and autophagy (Sokolowski et al., 2014). Modulation of CX3CL1 expression alters the effects of cocaine, as this chemokine increases microglial clearance of damaged neurons and upregulates antioxidant enzymes (such as heme oxygenase-1) within microglia (Noda et al., 2011). Montesinos et al. (2020) showed that repeated cocaine treatments in mice had a structure-related effect; it increases CX3CL1 expression in the hippocampus and decreases the expression in the PFC and striatum, hence, highlighting the complex effects of cocaine on CX3CL1 in different brain regions. CX3CL1 is expressed in two forms: a transmembrane-anchored protein and a soluble isoform. It is the soluble isoform of this protein that ameliorates microglial activation and proinflammatory cytokine release (Morganti et al., 2012). In the present study, a potential limitation was that the distinct isoforms of CX3CL1 were not separately assessed.
Cocaine conditioning also increased IL-1β expression, which was also reversed by CB2 agonism. Cocaine may be causing neuroinflammation by preventing clearance of apoptotic neurons due to down-regulation of CX3CL1, promoting the microglial release of pro-inflammatory cytokines and limiting antioxidant production. However, there is a lack of literature investigating the effects of cocaine on CX3CL1 signalling (Montesinos et al., 2020). Previous studies have shown that cocaine interacts with the Toll-like receptor 4 (TLR4), the immunosurveillance receptor complex located on microglia to induce neuroinflammatory response and IL-1β is suggested to serve as a final effector in cocaine rewarding effects (Periyasamy et al., 2018). It has also been shown that disruption of this interaction attenuates cocaine-induced increases in Nac extracellular dopamine (Northcutt et al., 2015). Thus, the reversal of the neuroinflammatory effects of cocaine by the CB2 agonist, particularly in the PFC, may explain its decrease in cocaine-seeking behaviour, as the PFC is particularly implicated in compulsive behaviours observed in addiction (Goldstein and Volkow, 2011). Complementarily, the rewarding effect of cocaine observed in the CPP was associated with altered CX3CL1 concentration, particularly in the VTA and the PFC which suggests that CX3CL1 signalling might be implicated in cocaine-induced CPP and its associated inflammatory response. This hypothesis is supported by previous reports on the role of CX3CL1 in the regulation of glial activation and subsequent release of inflammatory cytokines, such as TNF-α and IL-1β (Liu et al., 2015, Souza et al., 2013). On the other hand, the integrity of microglial status requires the maintaining of its metabolic reprogramming. It is well established that immune cells, including microglia, switch their metabolism from oxidative phosphorylation to glycolysis to produce metabolic resources necessary to satisfy the requirements of cell proliferation and activation. This metabolic programming plays a key role in the process of the innate inflammatory response. In particular, the proinflammatory state, which has been shown to be associated with a shift of energy production from oxidative phosphorylation to aerobic glycolysis, provides microglia to produce a faster rate of ATP production necessary for the generation of intermediates for cell growth and cytokine production (Borst et al., 2019, Lauro and Limatola, 2020).
Thus, in the present study, we also evaluated the effects of CX3CL1 on the metabolic profile of microglial that has been shown to be altered by cocaine. Our data showed that the stimulation of primary microglial cells with CX3CL1 induced a biphasic effect. At a low concentration (50 nM), CX3CL1 decreased oxidative phosphorylation without altering glycolytic parameters; decreased the release of the proinflammatory cytokine IL-1β and had no effect on TNFα. These results confirmed earlier findings showing that CX3CL1 reduces microglial activation and inhibits inflammatory cytokine release (Lyons et al., 2009, Zujovic et al., 2000), and suggest that low concentrations of CX3CL1 might be responsible for maintaining microglia in a steady state by reducing their oxidative phosphorylation. However, at the higher concentration of 100 nM, CX3CL1 increased mitochondrial oxidative metabolism, and both, IL-1β and TNFα release, which suggests that higher concentrations of CX3CL1 may switch microglial metabolism toward a proinflammatory phenotype. to the best of our knowledge, no study reported this biphasic effect of CX3CL1. Notwithstanding, this effect could be probable due to a differential modulation of GABAA receptors to increase their sensitivity to GABAA (Roseti et al., 2013). Also this biphasic effect of CX3CL1 might be linked to differences in the intracellular second messengers implicated in CX3CL1 receptors signaling. For example, in microglia, the activation of the CX3CL1 receptors was shown to decrease LPS-induced inflammation in a phosphatidylinositol-3 kinase pathway-dependent manner in vivo and in vitro (Lyons et al., 2009), but at higher doses and a time dependent manner, CX3CL1 can also increases the activation of Akt and/or ERK1/2 pathways and the transcription factor CREB (Deiva et al., 2004, Lyons et al., 2009, Chidambaram et al., 2020). Contrariwise to our hypotheses, these studies suggested that CX3CL1 at high levels leads to a tonic activation of CX3CL1 receptors to maintain microglia in a quiescent state. Moreover, these studies have demonstrated a significant decrease in inflammatory markers such as IL-1β and TNF-α following stimulation by LPS in a cell and animal model lacking CX3CL1 receptors, which accordingly suggested, that CX3CL1 is involved in the release of proinflammatory substances by activated microglia (Szepesi et al., 2018).
Recent studies have shown that microglia undergo aerobic glycolysis and release pro-inflammatory cytokines; and under chronic inflammatory conditions, acquire an anti-inflammatory phenotype (Kelly and O’Neill, 2015). In this view, cocaine-induced changes in CX3CL1 release, and CX3CL1 subsequent effects on microglia metabolic state would be associated with its specific functional activities and thus may influence its contribution to cocaine effects on behaviour, regulatory protein expression and inflammatory mediators.
In conclusion, the present study demonstrates that CB2 receptor modulation affects the behavioural and neurobiological changes induced by cocaine. CB2 receptor modulation should be investigated further as a potential therapeutic mechanism to attenuate cocaine reward-seeking behaviour. Future studies may investigate the effects of cocaine conditioning in CX3CR1 knockout mice, to determine the complex role played by CX3CL1 in cocaine-induced neuroinflammation. Also studying the effects of CB2 agonists in CX3CR1−/− cocaine-conditioned mice would provide insight into the importance of the role of CX3CL1 in mediating cannabinoid effects. Furthermore, specific CB2 agonists and antagonists may be used in CPP-reinstatement models, to gain an understanding of the precise roles of these receptors in the development of the reinforcing effects of cocaine.