Impaired social interaction (SI) is a common feature of drug-dependent individuals. In substance use disorders, impaired SI incites addicts to increase their drug consumption and restricts recovering addicts to adhere to treatment based on dyadic SI with healthcare providers (Zernig and Pinheiro, 2015). Thus, shifting the preference of drug-dependent individuals toward natural reward such as positive SI represents a challenge in drug therapy.

When SI is made available as an alternative choice to drugs, operant social reward has been shown to prevent drug self-administration in rats (Venniro et al., 2018, 2021). Moreover, pairing one compartment with cocaine and the opposite to a social stimulus during the conditioning sessions of the conditioned place preference (CPP) paradigm yielded to a similar preference (Fritz et al., 2011). These findings suggest that dyadic SI and cocaine have the same conditioned reward value and that the presence of a social partner in an opposite context to drugs abolishes drugs’ preference. Furthermore, after establishment of cocaine conditioning, mice that had the option to choose between a cocaine-paired compartment and a compartment where an unfamiliar juvenile mouse was placed during the CPP test in the non-cocaine paired compartment spent more time in the social-paired compartment (Sampedro-Piquero et al., 2019).

Even though both natural and drug reward are modulated by the nucleus accumbens (NAc) (Kelley, 2004), different intracellular cascades might be involved in the acquisition and expression of learning produced by either type of stimuli (Gerdjikov et al., 2007). It has been reported that protein kinase A (PKA) in the NAc is involved in the expression of drug but not natural reward-associated behaviors (Self et al., 1998; Baldwin et al., 2002; Misra and Pandey, 2006; Gerdjikov et al., 2007; Amaral et al., 2021). In addition, it was found that a bilateral infusion of Rp-cAMPS, a specific PKA inhibitor, directly into the NAc, shifted rats’ preference toward cocaine when a social partner was made available in the alternative compartment of the CPP (Amaral et al., 2021).

Many studies have focused on the involvement of the extracellular signal-regulated kinase (ERK), a member of the mitogen-activated protein kinase family (MAPK), in cocaine-rewarding properties (Valjent et al., 2000, 2006; Miller and Marshall, 2005); however, the effects of natural reward on the subsequently activated ERK expression are underinvestigated. In this study, we compared the levels of phosphorylated ERK after drug reward, particularly cocaine, and after SI natural reward. We hypothesized that after cocaine CPP, levels of activated ERK would be enhanced as compared to those after SI CPP. In another set of experiments, and based on findings reporting that inhibition of ERK before the CPP test impaired the expression of cocaine CPP (Miller and Marshall, 2005; Valjent et al., 2006), we wanted to explore whether the protective effects of SI when available as an alternative to drugs are enhanced by an additional ERK inhibition. For that, rats received a bilateral intra-NAc infusion of U0126, a MAPK/ERK kinase (MEK) inhibitor before the CPP test of a concurrent CPP paradigm, in which a social partner was made available in the alternative compartment to the one associated with cocaine. We proposed that rats would shift their preference toward the social partner-associated compartment in comparison with intra-NAc vehicle-infused rats that would express similar preference to the cocaine and the social partner-paired compartments.

We found that natural reward does not affect ERK phosphorylation in the NAc. Indeed, our results show that ERK expression was not altered in the NAc after CPP to natural reward such as SI nor after CPP to cocaine at a dose of 15 mg/kg. Moreover, intra-NAc infusion of U0126 did not alter the expression of CPP in a concurrent paradigm for SI vs. cocaine, which suggests that the protective effects of SI were not enhanced by ERK inhibition.

Many studies have reported an increase in ERK activation in the NAc after cocaine CPP. For example, one day after the CPP test to cocaine, a reconditioning with cocaine in the cocaine-paired compartment has been shown to increase phospho-ERK2 levels in the NAc (Valjent et al., 2006; Marion-Poll et al., 2019). These findings suggest that a reexposure to both cocaine and a drug-paired context is required to activate ERK in the NAc. Yet, reexposure only to the cocaine-paired compartment yielded to activated ERK in the NAc when the reexposure occurred one day after the CPP test (Nygard et al., 2015) but not two days after the CPP test (Tropea et al., 2008). When evaluated directly after the CPP test without reexposure to cocaine or to the cocaine-associated context, ERK was activated in the NAc of animals conditioned with the drug at the dose of 20 mg/kg (Valjent et al., 2006; Nygard et al., 2013, 2017). On the other hand, phospho-ERK was reported to increase only in the core but not the shell subdivision of the NAc after the CPP test to cocaine at a dose of 10 mg/kg (Miller and Marshall, 2005). In our study, rats were trained with cocaine at a dose of 15 mg/kg. Potentially, this dose may not be sufficient to induce ERK phosphorylation in the NAc directly after the CPP test without reexposure to cocaine or to the cocaine-associated context. In addition, our data reflect ERK phosphorylation levels examined in the rats’ both core and shell subdivisions of the NAc, which could have masked any slight subregion-specific alteration.

Conditioned place preference to natural reward such as SI did not affect ERK phosphorylation in the NAc, thereby suggesting that natural reward preference does not involve ERK activation. In line with these findings, ERK2 activation in the NAc has previously been found to be increased in rats trained with a tone as a conditioned stimulus predicting food delivery in comparison with rats trained to the context only (Shiflett et al., 2008). Moreover, sexual behavior alone did not induce ERK phosphorylation in the NAc core or shell in a study by Frohmader et al. (2010).

Consistent with the ERK intracellular signaling pathway role in learning and memory, pretreatment with SL 327, a pharmacological inhibitor of MEK, during the conditioning phase, completely abolished cocaine CPP in a study by Valjent et al. (2000). In addition, the deletion of the ERK1 isoform, which leads to increased ERK2 stimulus-dependent signaling, has been shown to facilitate the acquisition of cocaine CPP (Ferguson et al., 2006). These findings emphasized the implication of ERK signaling in the acquisition of cocaine preference. Interestingly, pretreatment with SL 327 before cocaine reexposure in the drug-paired environment has been demonstrated to be sufficient to abolish the previously acquired cocaine CPP (Valjent et al., 2006). We investigated the effects of U0126 infusion on the expression of a concurrent CPP paradigm in which one compartment was paired with cocaine and the opposite compartment was paired with a social partner. It has previously been shown that a pretreatment with a peripherally administered SL 327 (Valjent et al., 2006) or PD325901 (Papale et al., 2016) decreases the expression of cocaine CPP. Furthermore, bilateral intra-NAc infusion of U0126 before the CPP test impaired the expression of cocaine CPP in a study by Miller and Marshall (2005). Therefore, as we followed the procedure described by Miller and Marshall (2005), we expected that rats infused by U0126 in the NAc would shift their preference toward SI if the protective effects of SI are cumulative to the effects associated with ERK inhibition before the CPP test. However, we found that both vehicle- and U0126-infused groups expressed similar preference to cocaine and SI. The lack of effects in our study might be due to our design as we used a concurrent protocol, whereas in the study of Miller and Marshall (2005), a normal CPP procedure was performed. It appears that NAc core ERK mediates the expression of cocaine reward only when saline is present as an alternative to drugs. Yet, when an alternative reward of the same strength is made available during the conditioning sessions, other kinases, rather than ERK, seem to get involved in cocaine reward expression (Amaral et al., 2021). Furthermore, when SI is available in an alternative context, ERK inhibition effects are not summed up with SI protective effects but rather, SI masks the effects of ERK inhibition on the expression of drug reward learning. Another discrepancy between our protocol and the one in the study of Miller and Marshall (2005) lays in the length of the conditioning protocols. In the latter study, the conditioning consisted of 30-min sessions over a period of 6 days in an alternate day design (three cocaine pairings) compared to our conditioning sessions (15 min over a period of 4 days with morning and afternoon sessions and a total of four cocaine pairings). Although unlikely, it is not impossible that ERK inhibition efficiency might be determined by how the drug conditioning was performed.

One limitation of this study is that ERK inhibition was performed only before the CPP test, thereby focusing on the expression of reward-related learning. Interestingly, infusion of U0126 into the ventral tegmental area (VTA) during the acquisition phase of the CPP paradigm, but not before the CPP test, has been shown to attenuate cocaine-induced CPP, thereby suggesting that ERK activity is required for the acquisition, but not the expression of CPP to cocaine (Pan et al., 2011). Indeed, ERK activity in the VTA, which is most likely required for the establishment of an association between environmental cues and rewarding effects of cocaine, is not necessary for maintenance of the expression of cue-associated memories once such an association is established (Pan et al., 2011). Possibly, other brain regions such as the NAc may be recruited to maintain the expression of CPP to cocaine (Miller and Marshall, 2005; Pan et al., 2011). The previous research has shown that both cocaine and SI CPP induce an activation of the NAc (Rawas et al., 2012). Hence, in a concurrent paradigm, both stimuli would engage the NAc to maintain an equal expression of preference for SI and cocaine reward, thereby supporting a lack of ERK inhibition in reward expression when both stimuli are offered.

Pharmacological manipulations that decrease the extent to which cocaine and cocaine cues induce ERK activity have been suggested as potential treatments for cocaine addiction (Lu et al., 2006; Papale et al., 2016). Yet, it seems that these manipulations should not be combined with other treatments based on dyadic SI between treatment-seeking individuals and various healthcare providers such as psychiatrists, psychotherapists, or social workers if applied after that drug reward learning already occurred.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by the Austrian National Animal Experiment Ethics Committee (Permit numbers BMWF-66.011/0131-WF/V/3b/2016 and BMWF-66.011/0040-WF/V/3b/2019).

RE designed the research. IA performed the research. IA and RE analyzed the data. RE, IA, and AH wrote and edited the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.