Food and drug rewards share common neural substrates (Cardinal and Everitt, 2004; Kelley and Berridge, 2002; Volkow et al., 2008), and manipulation of food access (i.e., food-restriction) is commonly used in animal models of drug self-administration to increase responding (Carr, 2002; Carr, 2020). Food and drug (e.g., cocaine) rewards take effect in the mesolimbic dopamine pathway, and this commonality is reflected in the relationships between feeding and substance use behaviors (Carr, 2020). The impact of food-restriction on opioid taking and seeking behaviors has been well-established. For example, male rats undergoing food-restriction display heightened opioid self-administration relative to ad libitum- (i.e., as much or as often as wanted) feeding rodents (Carroll et al., 1979) with reduced body weight being necessary for this effect (Carroll and Meisch, 1980). Chronic food-restriction also increases opioid-seeking behavior in male rat models of relapse (D’Cunha et al., 2013, 2017, 2020). However, research studies examining the impact of food-restriction on cocaine self-administration and the reinstatement of cocaine-seeking are very limited and report mixed results. Bongiovanni & See reported no difference between food-restricted and ad libitum feeding male rats on cocaine self-administration, extinction, or reinstatement (Bongiovanni and See, 2008). On the contrary, food-restricted male rats met the criterion for acquisition of cocaine self-administration faster than their food-satiated counterparts (Campbell and Carroll, 2001). Therefore, further exploration is necessary to elucidate the impact of food-restriction on cocaine self-administration and reinstatement.

Importantly, the studies mentioned above only included male animals. However, (biological) sex differences in cocaine self-administration and reinstatement have been reliably observed (Becker and Hu, 2008; Becker and Koob, 2016; Lynch and Carroll, 1999; Lynch and Taylor, 2004; Nicolas et al., 2019; Nicolas et al., 2022; Roth and Carroll, 2004), and other studies have explored sex differences at the intersection of cocaine use and food intake. For example, more female rats choose cocaine over palatable pellets than male conspecifics (Perry et al., 2013), and gonadal sex hormones (e.g., estradiol) may mediate this effect (Kerstetter and Kippin, 2011). Carroll and Campbell found that food-restriction increased heroin self-administration similarly in both sexes; however, ketoconazole (a corticosterone synthesis inhibitor) attenuated this effect in females, but not in males (Carroll et al., 2001). Sedki et al. report an enhancement of opioid-seeking by food-restriction in ovariectomized female rats that is attenuated by estradiol replacement (Sedki et al., 2015). Despite these reports, it remains unclear how food-restriction impacts cocaine self-administration and reinstatement or how sex may modulate that relationship. Therefore, this study compared the behavior of female and male rats subjected either to ad libitum or restricted food access on cocaine self-administration and cocaine-seeking following a period of forced abstinence. We hypothesized that food-restriction would increase self-administration and cocaine-seeking in both sexes and that, within each food access condition, females would display greater cocaine self-administration and cocaine-seeking relative to males.

The present study assessed the impact of standard-chow food-restriction on the acquisition of cocaine self-administration, and cue-induced reinstatement of cocaine-seeking following a drug-free period, in female and male rats. Our findings reveal that food-restriction enhanced cocaine self-administration over time in females, but not in males. Reduced body weight is necessary for food-restriction’s enhancement of opioid self-administration (Carroll and Meisch, 1980). This relationship has not yet been assessed for cocaine self-administration, but our data suggest that the sex-dependent impact of food-restriction on body weight in our sample may have promoted its sex-dependent effect on self-administration. Our data do not show a decrease in body weight of food-restricted animals over time – rather an increase in body weight of ad libitum feeding rats, leading to group differences – but body weight may still impact cocaine use behaviors. Perhaps a more stringent food-restriction, such as the 15 g daily allotment for males employed by D’Cunha et al. (2013), may have produced more robust effects on body weight and/or cocaine-related behaviors.

Notably, sex differences in cocaine self-administration and reinstatement can be dose-dependent (Becker and Chartoff, 2019; Datta et al., 2018; Fuchs et al., 2005), and we therefore chose a moderate dose of 0.5 mg/kg/infusion that reportedly minimizes sex differences (Fuchs et al., 2005). In female rats, food-restriction decreases gonadal hormone (e.g., estradiol, follicular-stimulating hormone) levels (Ahmed et al., 2012). Leptin and ghrelin levels are also impacted by food restriction, though findings are mixed (Ahmed et al., 2012; D’Cunha et al., 2020). Thus, reproductive and orexigenic/metabolic hormone signaling may explain our sex-dependent results. Conversely, ad libitum food intake may have reduced vulnerability to cocaine use behaviors in females. To this point, blockade of appetite suppressant leptin signaling in the NAc enhances cocaine-conditioned reward (Shen et al., 2016).

We identified a main effect of lever in the cocaine-seeking session, suggesting that the rats in this study successfully reinstated cocaine-seeking. However, this effect was driven by food-restricted females, which was the only group that differed significantly in the post hoc analysis of active and inactive lever presses. This result suggests that food-restriction in females, but not males, may promote stronger learning that persists to the cocaine-seeking session. Indeed, food-restriction increases memory persistence in spatial learning tasks (Talhati et al., 2014) and increases reinstatement of reward-related learning (Carr, 2011; Zheng et al., 2012), though the intersection of food access and biological sex remains understudied.

While our study identified a sex-dependent impact of food-restriction on acquisition of cocaine-self-administration, we found little evidence to support an impact on reinstatement of cocaine-seeking. In contrast, food-restriction reliably potentiates opioid self-administration and opioid-seeking (Carroll and Meisch, 1980; Carroll et al., 1979; Carroll et al., 2001; D’Cunha et al., 2013, 2017, 2020). A number of differences between opioids and psychostimulants may contribute to these discordant findings and are reviewed in Badiani et al. (2011). As methodological considerations (e.g., dose, acquisition period, extinction vs. abstinence, etc.) can impact findings, additional research exploring the impact of food-restriction on models of cocaine use in both sexes is warranted to determine whether food-restriction differentially affects behavioral responses to these two drug classes.

Despite best efforts to the contrary, this study has some limitations. For example, acquisition criterion was relatively liberal, despite being more conservative than studies that employ lower or no discrimination threshold in their acquisition criteria (Lacy et al., 2018; Mojica et al., 2014). We chose this criterion to avoid artificially excluding differences that might naturally arise between groups while only including rats that performed better than chance. Additionally, ad libitum rats were pair-housed when possible, and food-restricted rats were single-housed by necessity to control food dosing. Consistent housing among groups would have been ideal, but Westenbroek and colleagues found that, relative to pair-housing, single housing does not impact cocaine self-administration on a fixed-ratio 1 schedule in males or females (Westenbroek et al., 2013). Additionally, single-housed rats in this study were given an extra piece of enrichment to minimize housing stress (Simpson and Kelly, 2011). Although we do not think that this limitation substantially impacted our results, male rats socially isolated through adolescence (post-natal days 22-51) showed greater amphetamine-seeking in adulthood (Garcia and Cain, 2021). While single-housing alone does not impact acquisition of cocaine self-administration, it is possible that it may have an effect in combination with food-restriction, and future studies could systematically address this question. Finally, we did not measure gonadal hormones in this study, despite evidence suggesting that gonadal hormones influence cocaine self-administration (Jackson et al., 2006) and are impacted by food-restriction (Ahmed et al., 2012).

In summary, we observed that food-restriction impacted the acquisition of cocaine self-administration in females, but not males, and that it had little impact on cocaine-seeking. We speculate that this impact derives from early improvements in learning relevant to the operant task and that this learning may persist to improve lever discrimination during cocaine-seeking for females, but not males. Though these results highlight the importance of assessing the effects of experimental conditions such as food access (and potentially the energy density of available food) in both sexes, the molecular mechanisms that impact this sex-dependent influence of food-restriction are poorly understood and should be further explored.