Electrophysiological studies have demonstrated that food and cocaine both activate neurons in the NAc. However, the two reinforcers do not activate the same cells within the NAc (Carelli et al., 2000, Carelli and Wondolowski, 2003). Moreover, food and sucrose self-administration do not cause long term alterations of electrophysiological properties as are induced by cocaine (Chen et al., 2008). In contrast, a collection of evidence suggests that male sexual behavior and drugs of abuse might indeed act on the same mesolimbic neurons. Psychostimulants and opioids alter the expression of sexual behavior in male rats (Mitchell and Stewart, 1990, Fiorino and Phillips, 1999a, Fiorino and Phillips, 1999b). Recent data from our lab showed that sexual experience alters the responsiveness to psychostimulants as evidenced by a sensitized locomotor responses and sensitized reward perception to d-amphetamine in sexually experienced animals (Pitchers et al., 2009). A similar response has previously been observed with repeated exposure to amphetamine or other drugs of abuse (Lett, 1989, Shippenberg and Heidbreder, 1995, Shippenberg et al., 1996, Vanderschuren and Kalivas, 2000). Together, these findings suggest that sexual behavior and responses to drugs of abuse are mediated by the same neurons in the mesolimbic system. Hence, the first objective of the present study is to investigate neural activation of the mesolimbic system by sexual behavior and drug administration in the same animal. In particular, we tested the hypothesis that the psychostimulant, methamphetamine (Meth), acts directly on neurons that normally mediate sexual behavior.

Meth is one of the most abused illicit drugs in the World (NIDA, 2006, Ellkashef et al., 2008) and it has been frequently linked to altered sexual behavior. Interestingly, Meth users report heightened sexual desire and arousal, as well as enhanced sexual pleasure (Semple et al., 2002, Schilder et al., 2005). Moreover, Meth abuse is commonly associated with sexually compulsive behavior (Rawson et al., 2002). Users often report having numerous sexual partners and are less likely to use protection than other drug abusers (Somlai et al., 2003, Springer et al., 2007). Unfortunately, studies indicating Meth use as a predictor of sexual risk behavior are limited as they rely on unconfirmed self-reports (Elifson et al., 2006). Therefore, an investigation into the cellular basis of Meth-induced changes in sexual behavior in an animal model is required for understanding this complex drug-sex nexus.

In view of the above outlined evidence suggesting that drugs of abuse, and particularly Meth, may act upon neurons normally involved in mediating sexual behavior, the objective of the present study was to investigate neural activation by sexual behavior and administration of the psychostimulant Meth. This study implemented a neuroanatomical technique, utilizing immunohistochemical visualization of the immediate early genes Fos and phosphorylated Map Kinase (pERK) to detect concurrent neural activation by sexual behavior and Meth respectively. Fos is only expressed within the nucleus of cells, with a maximal expression level 30–90 minutes after activation of the neuron. There is ample evidence that sexual activity induces Fos expression in the brain (Pfaus and Heeb, 1997, Veening and Coolen, 1998), including the mesocorticolimbic system (Robertson et al., 1991, Balfour et al., 2004). There is also evidence that drugs of abuse induce pERK expression within the mesocorticolimbic system (Valjent et al., 2000, Valjent et al., 2004, Valjent et al., 2005). In contrast to the expression of Fos, phosphorylation of ERK is a highly dynamic process and only occurs 5–20 minutes after neuronal activation. The distinct temporal profiles of Fos and pERK makes them an ideal set of markers for subsequent neuronal activation by two different stimuli.

The current finding that sexual behavior and administration of Meth activate overlapping populations of neurons in the NAc, BLA, and ACA is in contrast to findings from other studies showing that different populations of NAc neurons encode drug and natural reward. Specifically, electrophysiological studies that compared neural activation during self-administration of natural rewards (food and water) and intravenous cocaine have indicated that cocaine self-administration activated a differential, non-overlapping population of neurons that was generally not responsive during operant responding for water and food reinforcement (92%). Only 8% of accumbal neurons showed activation by both cocaine and natural reward (Carelli et al., 2000). In contrast, a majority (65%) of cell in the NAc showed activation by different natural rewards (food and water), even if one reinforcer was more palatable (sucrose) (Roop et al., 2002). Several factors may have contributed to the discrepancy with the current results. First, different technical approaches were used to investigate neural activity. The current study utilized a neuroanatomical method for detection of concurrent neural activation by two different stimuli using dual fluorescencent immunocytochemisty for Fos and pERK, allowing for investigation of single cell activation over large spans of brain areas. In contrast, the studies by Carelli and co-workers used electrophysiological recordings restricted to the NAc of behaving animals to address whether self-administration of drugs of abuse activate the same neural circuitry used by natural rewards. Second, the current study investigated a different natural reward i.e. sexual behavior compared to previous studies, which used food and water in restricted rats (Carelli, 2000). Food and water might have lesser rewarding value than mating. Sexual behavior is highly rewarding and rats readily form CPP to copulation (Agmo and Berenfeld, 1990, Martinez and Paredes, 2001, Tenk, 2008). Although, diet restricted rats do form CPP for water (Agmo et al., 1993, Perks and Clifton, 1997) and food (Perks and Clifton, 1997), diet unrestricted rats preferably consume and form CPP for more palatable foods (Jarosz et al., 2006, Jarosz et al., 2007). Third, our studies included different drugs of abuse compared to previous studies, utilizing methamphetamine and amphetamine instead of cocaine. The present results demonstrate that specifically Meth, and to a lesser extent amphetamine, resulted in activation of neurons also activated by sexual behavior. Drug experience may have also played a factor in our findings. The current studies utilized animals that were sexually experienced, but drug naïve. In contrast, the electrophysiological studies of Carelli and co-workers used “well-trained” animals that received repeated exposures to cocaine.

Hence, it is possible that the Meth-induced activation of neurons activated by sexual behavior is altered in drug experienced rats. However, preliminary studies from our lab suggest that drug experience is unlikely to be a major factor as sexual behavior and Meth treatment in males chronically treated with Meth co-activated similar percentages of drug-activated neurons as reported in the current study (20.3 ± 2.5 % in NAc core and 27.8 ± 1.3 % in NAc shell; Frohmader and Coolen, unpublished observations). Finally, the current study investigated the “direct” action of drugs utilizing passive administration. Therefore, the current analysis does not reveal information regarding neural circuits involved in drug seeking or cues associated to drug reward, but rather reveals neural activity caused by the pharmacological action of the drug. In the previous electrophysiological studies, NAc neural activity occurring within seconds of reinforced responses are not the result of the pharmacological action of cocaine, but is greatly dependent on associative factors within the self-administration paradigm (Carelli, 2000, Carelli, 2002). Specifically, NAc neural activity is influenced by response-independent presentations of stimuli associated with intravenous cocaine delivery as well as by instrumental contingencies (i.e., lever pressing) inherent in this behavioral paradigm (Carelli, 2000, Carelli and Ijames, 2001, Carelli, 2002, Carelli and Wightman, 2004). In summary, our findings of co-activation by natural and drug reward may be specific for activation by sexual behavior and passively administered Meth and Amph.

Meth and sex activated overlapping populations of neurons in the NAc core and shell in a dose-dependent manner. The co-activated neurons in the NAc may mediate potential effects of Meth on the motivation and rewarding properties of sexual behavior as lesions of the NAc disrupt sexual behavior (Liu et al., 1998, Kippin et al., 2004). In addition, these neurons are potentially a locus for dose-dependent drug effects on mating, since the lower Meth dose (1 mg/kg) reduced the number of dual labeled cells by 50% compared to the higher dose of Meth (4 mg/kg). Although this study does not identify the chemical phenotype of co-activated neurons, previous studies have shown that drug-induced pERK and Fos expression in the NAc is dependent on both dopamine (DA) and glutamate receptors (Valjent et al., 2000, Ferguson et al., 2003, Valjent et al., 2005, Sun et al., 2008). Although it is not clear if mating-induced neural activation in the NAc is dependent on these receptors, this has been demonstrated on other brain regions, particularly in the medial preoptic area (Lumley and Hull, 1999, Dominguez et al., 2007). Thus, Meth may act on neurons also activated during sexual behavior via activation of dopamine and glutamate receptors. The role of NAc glutamate in sexual behavior is currently unclear, but it is well established that DA plays a critical role in the motivation for sexual behavior (Hull et al., 2002, Hull et al., 2004, Pfaus, 2009). Microdialysis studies reported increases in NAc DA efflux during appetitive and consummatory phases of male sexual behavior (Fiorino and Phillips, 1999a, Lorrain et al., 1999) and mesolimbic DA efflux has been correlated to facilitation of the initiation and maintenance of rat sexual behavior (Pfaus and Everitt, 1995). Furthermore, DA manipulation studies show DA antagonists in the NAc inhibit sexual behavior, while agonists facilitate the initiation of sexual behavior (Everitt et al., 1989, Pfaus and Phillips, 1989). Thus, Meth may affect motivation for sexual behavior via activation of DA receptors.

In contrast to the NAc, the number of dual labeled cells in the BLA and ACA remained relatively unchanged regardless of the Meth dose. The BLA is critical for discrete associative learning and is strongly involved in conditioned reinforcement and reward evaluation during instrumental responding (Everitt et al., 1999, Cardinal et al., 2002, See, 2002). BLA lesioned rats display decreased lever pressing for conditioned stimuli paired with food (Everitt et al., 1989) or sexual reinforcement (Everitt et al., 1989, Everitt, 1990). In contrast, this manipulation does not affect the consummatory phase of feeding and sexual behavior (Cardinal et al., 2002). The BLA also plays a key role in memory of conditioned stimuli associated with drug stimuli (Grace and Rosenkranz, 2002, Laviolette and Grace, 2006). Lesions or pharmacological inactivations of the BLA block the acquisition (Whitelaw et al., 1996) and expression (Grimm and See, 2000) conditioned-cued cocaine reinstatement, while not affecting the process of drug administration. Furthermore, Amph infused directly into the BLA results in a potentiated drug reinstatement in the presence of the conditioned cues (See et al., 2003). Therefore, it is possible that psychostimulant-enhanced DA transmission in the BLA results in potentiated emotional salience and seeking (Ledford et al., 2003) of sexual reward, thus contributing to the enhanced sexual drive and desire reported by Meth abusers (Semple et al., 2002, Green and Halkitis, 2006).

In the ACA, neural activation of sex-activated neurons was dosage-independent and specific for Meth, as it was not observed with Amph. Although Meth and Amph have similar structural and pharmacological properties, Meth is a more potent psychostimulant than Amph with longer lasting effects (NIDA, 2006). Studies by Goodwin et al. showed that Meth generates a greater DA efflux and inhibits the clearance of locally applied DA more effectively in the rat NAc than Amph. These characteristics could contribute to the addictive properties of Meth compared to Amph (Goodwin et al., 2009) and perhaps the neural activation differences observed between the two drugs. However, it is not clear whether the different patterns of results are due to efficacy differences between the drugs or potency issues related to the doses employed and further investigation is required.

Co-activation by Meth and sex was not observed in other subregions of the mPFC (IL and PL). In the rat, the ACA has been extensively studied using appetitive tasks, supporting a role in stimulus–reinforcer associations (Everitt et al., 1999, See, 2002, Cardinal et al., 2003). There is ample evidence that the mPFC is involved in drug craving and relapse to drug-seeking and drug-taking behavior in both humans and rats (Grant et al., 1996, Childress et al., 1999, Capriles et al., 2003, McLaughlin and See, 2003, Shaham et al., 2003, Kalivas and Volkow, 2005). In line with this, it has been proposed that mPFC dysfunctioning caused by repeated exposure to drugs of abuse might be responsible for reduced impulse control and increased drug-directed behavior as observed in many addicts (Jentsch and Taylor, 1999). Recent data from our laboratory demonstrated that mPFC lesions result in continued seeking of sexual behavior when this was associated with an aversive stimulus (Davis et al., 2003). Even though this study did not investigate the ACA, it supports the hypothesis that the mPFC (and the ACA specifically) mediates the effects of Meth on a loss of inhibitory control over sexual behavior as reported by Meth abusers (Salo et al., 2007).