Abused substances feature different origins, legal status, abuse liability, exert different pharmacological effect, harm on the user and are responsible for different socioeconomic burden. One of the drugs usually accepted by society is nicotine - an active ingredient of tobacco leaves. Nicotine acts mainly on nicotinic acetylcholine receptors; increasing alertness, memory function, blood pressure and heart rate (Prochaska and Benowitz, 2019). The most common way to consume tobacco is through inhalation via cigarettes, but recently vaping devices, nicotine pouches and lozenges are gaining popularity due to heavy advertising and the mistaken belief that they provide a healthier way to consume nicotine (Dinardo and Rome, 2019; Jenssen and Wilson, 2019; Walley et al., 2019; Unger et al., 2022). Given that nicotine is readily available to the general public (Cwalina et al., 2021), its addictive properties pose a significant health hazard that may continue to grow in the future (Dinardo and Rome, 2019). Although there are some approved pharmacotherapies for nicotine addiction (Tobacco Use and Dependence Guideline Panel, 2008), they have limitations (Polosa and Benowitz, 2011). Thus, searching for new pharmacotherapies for nicotine addiction is still essential.

In contrast, there is methamphetamine, a synthetic substance that is widely deemed illegal (Courtney and Ray, 2014). It is highly addictive due to its ability to target multiple neural pathways, all of which lead to a robust increase of monoamines in the reward pathway. Methamphetamine is typically administered intravenously, orally, sniffed or smoked, resulting in euphoric stimulant effects, anorexia, and sleep deprivation (Cruickshank and Dyer, 2009). Methamphetamine addiction, despite its extensive history (Grobler et al., 2011) and its severe impact on both physical and mental health (Darke et al., 2008), continues to rise worldwide (Han et al., 2021; Lewis et al., 2021). Importantly, there is no approved pharmacological treatment (Chan et al., 2019), which makes it a challenging issue to research and resolve.

This study focuses on both drugs – nicotine and methamphetamine – with very different pharmacodynamics (Tiwari et al., 2020; Yasaei and Saadabadi, 2022). Despite their differences, both substances eventually lead to increase dopamine release in the reward pathway (Volkow et al., 2011), as well as increased glutamatergic signaling in the cortex, nucleus accumbens, hippocampus and striatum (Gass and Olive, 2008). Although accumbal dopaminergic mechanisms are necessary for reinforcing effects of drugs, their chronic use leads to predominantly cortical neuroadaptations mediated by glutamate that contribute to the development of addiction in human (Volkow et al., 2019; Picciotto and Kenny, 2021).

The glutamate homeostasis hypothesis connects addiction to an imbalance in glutamate levels, causing alterations in brain circuits and diminishing control over drug-seeking behavior, thereby increasing vulnerability to relapse (Kalivas, 2009; Fischer et al., 2021). Notably, elevated glutamate release in the nucleus accumbens is a prominent feature in the reinstatement of extinguished drug-seeking, observed across various substances, including nicotine and methamphetamine (Scofield et al., 2016). The underlying neurotransmission changes involve differential subunit composition of the glutamatergic receptors (Gass and Olive, 2008; Stone, 2021), colocalization of ionotropic glutamatergic receptors with other receptor types (Volkow et al., 2019) and involvement of glial cells and astrocytes in the neuroadaptations following substance use (Fouyssac and Belin, 2019).

There seems to be a particular yet inconclusively described role of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors in the human substance abusers pointing to a single relatively consistent outcome – an increase in the hippocampal expression of AMPA receptor subunits (Ueno et al., 2019). In preclinical research, multiple drugs were shown to exert alterations of AMPA receptor subunits’ expression in reward-related brain areas, typically nucleus accumbens or ventral tegmental area, e.g., cocaine (Sutton et al., 2003), amphetamine (Cruz et al., 2008), nicotine (Wang et al., 2007) or methamphetamine (Scheyer et al., 2016; Murray et al., 2019). Hence, the potential of pharmacological manipulation of AMPA receptors has been repeatedly suggested as a treatment target for SUD (Lee et al., 2016; Cooper et al., 2017; Ueno et al., 2019). Specifically, it has been shown that injecting AMPA receptors agonist into NAc decreased cocaine intake – suggesting that glutamate transmission augments the threshold of the reinforcing effect of cocaine intake (Cornish et al., 1999). On the other hand, microinjections of AMPA agonist reinstated cocaine-seeking behavior (Cornish et al., 1999; Cornish and Kalivas, 2000; Kruzich and Xi, 2006), and conversely, intra-NAc injections of AMPA antagonists were able to attenuate it (Cornish and Kalivas, 2000; Di Ciano and Everitt, 2001; Park et al., 2002; Ciano and Everitt, 2004; Bäckström and Hyytiä, 2007). Moreover, during the withdrawal of the drugs, expression of AMPA receptors tends to be increased in NAc (Boudreau, 2005; Ferrario et al., 2011; Moretti et al., 2020), which offers a possible explanation for the known phenomenon of drug craving and relapse (Cornish and Kalivas, 2000).

However, there is currently limited evidence supporting the use of AMPA/kainate antagonists as a potential treatment for nicotine and methamphetamine dependence behavior. Therefore, in this study, we used an AMPA/kainate antagonist, CNQX [which also has an affinity for the NMDA receptor’s glycine site (Davies, 2007)], to evaluate its effect on the operant drug-taking and drug-seeking in two pharmacologically distinct model drugs: nicotine and methamphetamine.

In both studies, we observed the expected acquisition of drug intake, stable responses and drug self-administration over 2 weeks of the maintenance period. Data from the maintenance training of both studies are shown in Figure 2.

Followed parameters during maintenance sessions were responding to the drug by active nose-pokes, number of self-administered infusions, and drug intake in mg/kg. In the context-induced relapse-like session, we analyzed responses for the drug by active nose-pokes, inactive nose-pokes and active operandum preference (% of previously drug-paired responses over the total number of responses).

The findings from this research show a decrease in nicotine consumption after a single administration of a higher dose (6 mg/kg) of CNQX. However, CNQX did not have any impact on drug-seeking behavior during a relapse-like session following the forced abstinence. In the methamphetamine study, no effect of CNQX was observed in any phase of the operant protocol.

Quinoxalinediones, such as CNQX, are a group of ligands that are specifically designed to target AMPA receptors. However, many AMPA receptor antagonists also block kainate receptors, making the molecular targets broader and attenuating glutamate in a more complex manner (Gass and Olive, 2008). CNQX, for example, is a competitive antagonist on AMPA/kainate receptors, but it also has an affinity towards the glycine site of NMDA receptors (Lester et al., 1989), which may contribute to its effect on drug-related behaviors. Studies have shown that NMDA/glycine site antagonists alone can reduce cue-induced cocaine reinstatement (Bäckström and Hyytiä, 2006), prevent the expression of amphetamine-induced conditioned place preference (Mead and Stephens, 1999), and decrease ethanol-seeking behavior (Bäckström and Hyytiä, 2004).

Only a few behavioral studies have explored the impact of AMPA/kainate antagonists on the nicotine dependence model, with even fewer dedicated specifically to the self-administration. The results of these studies indicate that MPQX, a competitive AMPA antagonist, can decrease nicotine-induced dopamine release in the nucleus accumbens core and reduce the increase in locomotor activity caused by nicotine. However, NBQX, a substance that also has an affinity to kainate receptors, did not impact either of these processes (Kosowski et al., 2004). This is consistent with previous studies showing that NBQX had no effect on nicotine intake (Kenny et al., 2009; Ruda-Kucerova et al., 2021). On the other hand, antagonism on NMDA receptors blocked nicotine-induced lowering of intracranial self-stimulation thresholds, attenuated nicotine self-administration (Kenny et al., 2009), decreased nicotine-induced dopamine release and locomotor activity (Kosowski et al., 2004). These findings and our study suggest that the mechanism of CNQX responsible for reduction of nicotine intake is rather NMDA than AMPA or kainate receptors antagonism.

Conversely, we demonstrated no effect of CNQX on nicotine-seeking after two weeks of forced abstinence, while in our previous study, NBQX was able to reduce it (Ruda-Kucerova et al., 2021). This may indicate that AMPA/kainate receptors are more involved in the post-abstinence relapse-like phase of nicotine seeking behavior. However, currently there is no more studies testing quinoxalinediones in models of nicotine dependence to support this notion.

In the model of operant methamphetamine self-administration, we observed no impact of CNQX on methamphetamine maintenance or its seeking in the relapse-like session. Behavioral research of AMPA/kainate receptors and dependence to stimulants includes a number of studies with cocaine, where both CNQX and NBQX were able to attenuate drug seeking (Bäckström and Hyytiä, 2003, 2006; Zavala et al., 2008). However, NBQX showed a difference in its effect depending on how the abstinence was induced: the effect was shown only in the group in which abstinence was forced, not when animals underwent extinction protocol (Zavala et al., 2008). What seems more consistent are data from cocaine studies where AMPA/kainate antagonists were injected directly into a specific part of the brain, resulting in a higher concentration and increased bioavailability at the site of action. E.g., when these ligands were infused into the nucleus accumbens core, cocaine intake was decreased (Suto et al., 2009) as well as cocaine-seeking behavior induced by a variety of factors (Cornish and Kalivas, 2000; Di Ciano and Everitt, 2001; Park et al., 2002; Ciano and Everitt, 2004; Bäckström and Hyytiä, 2007; Xie et al., 2012). The effect was absent when infused into the nucleus accumbens shell (Di Ciano and Everitt, 2001; Suto et al., 2009). Based on these studies, AMPA/kainate antagonists may play a role in cocaine-seeking behavior through their effects on the nucleus accumbens core. However, there is limited knowledge regarding the influence of AMPA/kainate antagonists on behavior associated with other stimulant drugs, particularly methamphetamine, and results are overall contradictory. For instance, NBQX did not attenuate behavioral sensitivity induced by amphetamine (Li et al., 1997) but did attenuate amphetamine (Dalia and Wallace, 1995; Mead and Stephens, 1998; Vanover, 1998) and methamphetamine-induced locomotor activity (Witkin, 1993). Yet in another study, several days of pretreatment with NBQX did not attenuate methamphetamine-induced hyperactivity (Akiyama et al., 1998).

In amphetamine-conditioned place preference protocol, CNQX or DNQX was able to block its acquisition (Layer et al., 1993; Mead and Stephens, 1999) and also its expression (Miyatake et al., 2005; Banasikowski et al., 2012). In stark contrast, NBQX was ineffective, did not block the amphetamine-conditioned place preference (Mead and Stephens, 1998, 1999) and caused conditioned place aversion at a 30 mg/kg dose (Mead and Stephens, 1999). Moreover, in operant self-administration studies, NBQX did not reduce the intake of methamphetamine nor its seeking in relapse (Ruda-Kucerova et al., 2021). Similarly, CNQX did not affect the stimulant effect of methamphetamine on self-administration of palatable pellets (Wooters et al., 2011). Based on these findings, it seems unlikely that CNQX or AMPA/kainate antagonists in general, represent a promising treatment for methamphetamine dependence. However, the effect of administering antagonists directly to specific brain regions has not yet been studied in depth. Given the variations in the mechanisms of action among different stimulants, the effectiveness of AMPA/kainate antagonist treatment may differ depending on the specific drug. For example, when comparing the behavioral effects and subsequent neuroadaptations related to incubation of craving for cocaine and methamphetamine, numerous distinctions in glutamate synaptic plasticity in the nucleus accumbens core were reported (Scheyer et al., 2016; Murray et al., 2019; Funke et al., 2023). Hence, it is important to note that cocaine and amphetamine should not be grouped together when discussing the role of glutamate-related plasticity. Specifically, the results from studies on AMPA receptors for one drug cannot be applied to the other (Wolf and Ferrario, 2010).

The effectiveness of CNQX in reducing self-administration of nicotine but not methamphetamine may be due to the difference in dopamine release in the reward pathway caused by these drugs and the effect of glutamate on dopamine release. The amount of dopamine released after nicotine injection is significantly lower than the amount which is released after stimulants (Di Chiara and Imperato, 1988; Scherma et al., 2008; Ruda-Kucerova et al., 2015b). Since glutamate release modifies dopamine release (Tzschentke and Schmidt 2003), this difference in dopamine release may explain why a dose of 6 mg/kg of CNQX effectively reduced self-administration of nicotine but not methamphetamine. However, further increasing the dose of CNQX could negatively impact spontaneous locomotor activity, which was reported to be partially affected already by 3 mg/kg of CNQX (Bäckström and Hyytiä, 2003). Nevertheless, in our study, the 6 mg/kg dosage did not affect locomotor activity, as evidenced by the lack of a significant effect on both inactive and active nose pokes. Another aspect to consider is the different mechanisms of action between these two drugs. However, both of these drugs share the discriminative stimulus effect as shown in a model of operant food self-administration (Gatch et al., 2008). Hence, despite their pharmacological differences, they are likely to induce craving.

Notably, use of CNQX in animal research has some methodological limitations, such as its pH-sensitive binding to the AMPA receptor GluA2 LBD (Dudić and Reiner, 2019), poor solubility, and limited ability to penetrate the central nervous system (Bigge and Nikam, 1997), which can affect its bioavailability and dose-dependent efficacy. As of our present knowledge, a comprehensive pharmacokinetic investigation following different routes of CNQX administration is lacking. Nevertheless, the route of administration can significantly impact effects of any drug. The prevailing method of CNQX administration predominantly employs intraperitoneal administration. Our use of IV route can be expected to have earlier onset of effect, 100% bioavailability and higher peak plasma concentration. We reflect these pharmacokinetic aspects in the experimental design by administering CNQX only10 minutes before the self-administration session. Another aspect to consider is the behavioral protocol, as the impact of short-term versus extended drug exposure has been established. Research has shown that CNQX, when administered systemically or directly to the nucleus accumbens, only decreased cocaine seeking in the group with prolonged drug exposure (Doyle et al., 2014; Lynch et al., 2021). Hence, further research is needed to investigate the impact of CNQX in studies allowing extended drug access. Additionally, it would be interesting to explore the potential of CNQX in altering the affective state. There is evidence of CNQX affecting the emotional aspect of pain (Spuz and Borszcz, 2012; Yan et al., 2012). Tracking this effect could be particularly interesting in the context of craving, as reported as the case of nicotine-degrading enzyme which normalized hyperalgesic reaction following nicotine withdrawal from extended access self-administration protocol (Kallupi et al., 2018).

Taken together, in this study, we found that the use of CNQX at a dose of 6 mg/kg significantly reduced nicotine response during the maintenance phase but not nicotine-seeking in the relapse-like session. We previously showed that NBQX reduces nicotine seeking in the relapse-like session but not nicotine intake (Ruda-Kucerova et al., 2021). This suggests a different role for glutamate receptors in nicotine dependence, as CNQX is not selective for AMPA or kainate receptors alone. Additionally, no effect of CNQX was observed on either methamphetamine intake or its seeking. We propose that the effects of CNQX on influencing drug-related behavior are drug-specific.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by Expert committee for ensuring the welfare of experimental animals of the Faculty of Medicine, Masaryk University. The study was conducted in accordance with the local legislation and institutional requirements.

MH: Conceptualization, Data curation, Investigation, Methodology, Visualization, Writing – original draft, Formal analysis. PA: Data curation, Writing – review & editing, Methodology. JR-K: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – review & editing.