Sex Differences in Escalated Methamphetamine Self-Administration and Altered Gene Expression Associated With Incubation of Methamphetamine Seeking

Abstract

Background

Methamphetamine (METH) use disorder is prevalent worldwide. There are reports of sex differences in quantities of drug used and relapses to drug use among individuals with METH use disorder. However, the molecular neurobiology of these potential sex differences remains unknown.

Methods

We trained rats to self-administer METH (0. 1 mg/kg/infusion, i.v.) on an fixed-ratio-1 schedule for 20 days using two 3-hour daily METH sessions separated by 30-minute breaks. At the end of self-administration training, rats underwent tests of cue-induced METH seeking on withdrawal days 3 and 30. Twenty-four hours later, nucleus accumbens was dissected and then used to measure neuropeptide mRNA levels.

Results

Behavioral results show that male rats increased the number of METH infusions earlier during self-administration training and took more METH than females. Both male and female rats could be further divided into 2 phenotypes labeled high and low takers based on the degree of escalation that they exhibited during the course of the METH self-administration experiment. Both males and females exhibited incubation of METH seeking after 30 days of forced withdrawal. Females had higher basal mRNA levels of dynorphin and hypocretin/orexin receptors than males, whereas males expressed higher vasopressin mRNA levels than females under saline and METH conditions. Unexpectedly, only males showed increased expression of nucleus accumbens dynorphin after METH self-administration. Moreover, there were significant correlations between nucleus accumbens Hcrtr1, Hcrtr2, Crhr2, and Avpr1b mRNA levels and cue-induced METH seeking only in female rats.

Conclusion

Our results identify some behavioral and molecular differences between male and female rats that had self-administered METH. Sexual dimorphism in responses to METH exposure should be considered when developing potential therapeutic agents against METH use disorder.

Significance Statement.

Methamphetamine (METH) abuse continues to be a public menace throughout the world. Sex differences reported in METH use disorder may be dependent on the expression of neuropeptides in the nucleus accumbens (NAc), an important structural and functional node in brain reward circuitries. Here we report that male rats took more METH than female rats. There were, however, no differences in cue-induced METH-seeking behavior, a potential marker for relapse. Importantly, we found that there were no specific sex differences in METH SA-induced alterations in the mRNA expression of several neuropeptides and their receptors including Hcrtr1, Hcrtr2, Crhr2, and Avpr1b in the NAc. Of note, changes in mRNA levels were significantly correlated with cue-induced METH-seeking behaviors in female, but not male, rats. Our data may have important implications for potential sex differences in METH-taking behaviors that may need to be considered when planning treatment for diverse groups of patients with METH use disorder.

Introduction

Methamphetamine use disorder (MUD) is a debilitating neuropsychiatric disorder in both men and women (Zhang et al., 2013; Radfar and Rawson, 2014; Kogachi et al., 2017). Humans who abuse METH experience substantial deficits in cognitive functions (Cadet and Bisagno, 2015; Proebstl et al., 2018) that are probably secondary to drug-induced neuroadaptations and/or neuropathological changes in the brain (Cadet et al., 2014a; Radfar and Rawson, 2014). Importantly, sex-related socio-demographic differences and psychiatric co-morbidities exist in humans diagnosed with MUD (Becker and Hu, 2008; Dluzen and Liu, 2008; He et al., 2013; Zhang et al., 2013). Although female and male users are usually reported to take similar quantities of METH, women appear to suffer greater dependence and experience more frequent relapses during periods of abstinence (Gonzales et al., 2010; Reichel et al., 2012; Radfar and Rawson, 2014; Ruda-Kucerova et al., 2015). Moreover, there appear to be sex-dependent differences in the severity of METH-induced cognitive deficits (Radfar and Rawson, 2014) and structural changes in the brain (Kogachi et al., 2017). Clinical studies have also suggested that women might show greater improvements in medical complications and family interactions while in treatment (Hser et al., 2005).

Animal studies have also documented some sex differences in models of METH abuse (Roth and Carroll, 2004; Cox et al., 2013; Lynch, 2018). For example, female rats have been reported to take more METH than males when somewhat smaller doses were used for females in self-administration (SA) studies (Reichel et al., 2012). Males were said to take more METH when higher doses were used (Ruda-Kucerova et al., 2015). Similarly, a human study reported that a low dose of d-amphetamine functioned as a better reinforcer in women whereas a higher dose was more reinforcing for men (Vansickel et al., 2010). Male rats were also reported to show more escalation of METH SA than females (Venniro et al., 2017). Cox et al. 2013 reported greater cue-induced METH seeking in female rats, whereas Venniro et al., 2017 did not find any sex differences in cue-induced incubation of METH craving. Of clinical relevance, oxytocin suppressed progressive ratio responding only in female rats (Cox et al., 2013).

Sex differences in the biochemical consequences to METH exposure have also been observed. For example, METH-induced activation of the hypothalamic-pituitary-adrenal axis is reportedly dependent on the sex of the animals (Zuloaga et al., 2014). METH SA also affects the glutamatergic system in the prefrontal cortex in a sex-dependent fashion (Pena-Bravo et al., 2019). Moreover, only male rats that underwent 7 days of METH SA showed increased hippocampal brain-derived neurotrophic factor (BDNF) levels (Johansen and McFadden, 2017).

Sex differences in METH taking and/or relapse to drug taking might be due, in part, to interactions of the drug with endogenous systems that might exhibit differences in basal expression and/or in responses to drug exposure (Bisagno and Cadet, 2014; Yang and Shah, 2014; Cadet et al., 2015; Becker and Chartoff, 2019). Indeed, biochemical and molecular aftermaths of drugs of abuse are based on interactions with various interconnected brain regions, including the nucleus accumbens (NAc), that constitute nodal points in reward circuitries (Volkow et al., 2012; Everitt, 2014; Cadet et al., 2016, 2017). The NAc is a major player that links different brain regions that can initiate cognitive processes during the development and maintenance of addiction-associated behaviors and/or relapse to drug taking during abstinence (Berendse et al., 1992; Cornish et al., 1999; Fuchs et al., 2004; Bossert et al., 2007; Walsh and Han, 2014). Transcriptional and epigenetic changes have indeed been reported in the rat NAc after METH SA (Cadet et al., 2016, 2017). We thus reasoned that any sex-associated differences in METH taking may be dependent on the expression of some sexually dimorphic neuropeptides in the NAc. Some neuropeptides have indeed been implicated in models of psychostimulant addiction (Bisagno and Cadet, 2014; James et al., 2017).

Methods

Subjects

Female and male Long Evans rats (24 each, NIDA OTTC breeding facility, USA) were group-housed in a humidity- and temperature-controlled (22.2°C ± 0.2°C) room with free access to food and water ad libitum. Female rats weighed 350–500 g and males were 450–600 g in the beginning of the experiments. All animal procedures were approved by the National Institute of Drug Abuse Animal Care and Use Committee and conducted according to the Guide for the Care and Use of Laboratory Animals (ISBN 0-309-05377-3).

Intravenous Surgery

Rats were anesthetized with ketamine and xylazine (50 and 5 mg/kg, i.p., respectively). A silastic catheter (SAI Infusion Technologies) was then implanted into the right jugular vein as described previously (Blackwood et al., 2019). We attached the other end of the catheter to modified 22-gauge cannulae that were mounted to the back of the rats. Rats were injected with buprenorphine (0.1 mg/kg, s.c.) once post-surgery to relieve pain and allowed to recover for 5 to 7 days before METH SA training. During the recovery and training phases, catheters were flushed every 24 to 48 hours with sterile saline and gentamicin (0.05 mg/kg).

Apparatus

Rats were trained in SA chambers (30 × 20 × 20 cm) located inside sound-attenuated cabinets (Med Associates, St Albans, VT). Each chamber was equipped with active and inactive levers located 8.5 cm above the grid floor. Pressing the active lever activated the infusion pump and tone-light cue. Presses on the inactive lever had no reinforced consequences. The catheters were connected to modified cannulae connected to a swivel via tubing that was protected by a metal spring. A 20-mL syringe was mounted outside the SA chambers on an infusion pump attached to the tubing and served as a reservoir for the METH infusion.

METH SA Training Phase

We performed training procedures for METH SA according to our previous publications (Cadet et al., 2016, 2017; Krasnova et al., 2017). Rats used in the study were drug-naive rats and were not food-trained before the start of the METH SA experiment. On the first day of training, rats were brought to SA rooms and were randomly assigned to either saline (n = 6) or METH (n = 18) groups. Rats were chronically housed in SA chambers and had free access to food and water throughout the experiment. Rats were trained to self-administer METH-HCl (NIDA Pharmacy) during two 3-hours sessions/d (separated by a 30-minute off interval) for 20 days under a fixed-ratio-1 schedule with 20-second timeouts. Presses on the retractable active lever activated the infusion pump to deliver METH (0.1 mg/kg/infusion) over 3.5 seconds (0.1 mL/infusion). Active lever presses were also accompanied by a 5-second compound tone-light cue. At the end of each 3-hour session, the house light was turned off and the active lever was retracted. We trained rats to self-administer METH for 5 days a week, with weekends off. During the 2 days off, rats remained housed in SA chambers but were disconnected from i.v. SA lines. Control rats self-administered saline under identical conditions. After 20 days of training phase, rats were returned to the animal vivarium and individually housed with no access to METH. They were then tested for cue-induced drug seeking on withdrawal days 3 (WD3) and 30 (WD30). During that time, i.v. catheters were covered using sterile dust caps and rats had access to home-cage food and water ad libitum.

Tests of Cue-Induced METH Seeking

To test cue-induced drug seeking during withdrawal, rats were brought back to their corresponding SA chambers on the morning of each test day. Each test consisted of a 3-hour session during which presses on the METH-associated lever resulted in no METH infusions but continued presentations of the tone and light cues previously paired with METH infusions. All rats tested on WD3 were also tested on WD30.

Collection of Tissues and Quantitative RT-PCR

We killed rats 24 hours after the WD30 by decapitation with guillotine. NAc was isolated from the brains using coordinates and snap-frozen on dry ice.

Total RNA was isolated from the NAc using Qiagen RNeasy Mini kit (Qiagen). Five hundred nanograms (500 ng) of RNA were reverse-transcribed with oligo dT primers using Advantage RT-for-PCR kit (Clontech). Real-time quantitative polymerase chain reaction (RT-qPCR) was performed with Roche LightCycler 480 II using iQ SYBR Green Supermix (Bio-Rad). Primer sequences used in the study are listed in supplementary Table 1. Gene-specific qPCR primers were purchased from the Synthesis and Sequencing Facility of Johns Hopkins University. The relative quantities of mRNAs were normalized to 18 seconds and beta-2-microglobulin.

Statistical Analyses

Behavioral data were analyzed with the statistical program GraphPad Prism 8 using 3-way ANOVA with repeated measures with variables sex (male vs female), group (METH vs saline), and training day (20 days). To further probe the nature of the interactions, a 2-way ANOVA with repeated-measures analysis was also undertaken to compare METH groups. Variables were numbers of METH infusions on training days, between-subject factors (saline, males, and females), and within-subject factor SA days (training days 1–20) and their interactions. METH-seeking data were also analyzed using 3-way ANOVA with repeated measures, with variables being sex (male vs female), group (METH vs saline), and withdrawal days (WD3 and WD30). Fishers protected least significance difference post-hoc tests were used when ANOVAs were significant.

PCR results of relative mRNA abundance were analyzed using 2-way ANOVA followed by Fisher's protected least significance difference post-hoc test using Stat View (version 4. 02). We used baseline expression in the analysis to account for potential sexual dimorphism in basal expression that might be missed if fold changes were used in these initial analyses. For the regression analysis, levels of expression of individual genes are reported as fold changes calculated as the ratios of normalized gene expression data of METH SA groups compared with data of the saline group. For all analyses, the null hypothesis was rejected at P ≤ .05.

Results

Male Rats Showed Greater METH Intake Than Female Rats

The timeline of our behavioral experiment is illustrated in Figure 1A. Figure 1B shows the patterns of METH SA for female and male rats. We analyzed behavioral data of female and male using repeated-measures 3-way ANOVA with factors including sex (female vs male), group (saline vs METH), and training days (20 days). The effects of training days [F(8, 415) = 4.577, P < .0001] and group (saline vs METH) [F(1, 53) = 135.4, P < .0001] were statistically significant, but effects of sex (female vs male) [F(1, 53) = 1.865, P = .1779] were not significant. In addition, there were statistically significant interactions between training days and sex [F(19, 1007) = 3.539, P < .0001], training days and group [F(19, 1007) = 19.73, P < .0001], and sex and group [F(1, 53) = 6.018, P = .0175]. An overall training days × sex × group [F(19, 1007) = 2.362, P = .0009] interaction was also significant. To further probe the nature of these interactions, a 2-way repeated-measures ANOVA was run; this consisted of the following factors: sex, training day, and their interaction. All 3 of these terms were significant: sex [F(1,31) = 5.813, P = .0220], training day [F(19,589) = 16.58, P < .0001], and sex × training day interaction [F(19,589) = 2.999, P < .0001], indicating that male rats took a significantly greater number of METH infusions than female rats (see statistical keys in Figure A). Figure 1C shows the total amount of METH taken by males and females during the 20 days of SA training.

Figure 1.

Male rats take high amount of methamphetamine (METH) than female rats. (A) Experimental timeline for METH self-administration (SA) experiments. (B) Patterns of METH or saline SA in female and male rats during 20 days of SA training. Data represent the number of daily infusions during 6 hours of access to METH or saline [3-way ANOVA showed significant interaction of training days × sex [F(19, 1007) = 3.539, P < .0001], training days × group [F(19, 1007) = 19.73, P < .0001], sex × group [F(1, 53) = 6.018, P = .0175], and training days × sex × group [F(19, 1007) = 2.362, P = .0009]. (C) Male rats took higher amounts of METH than females during the 20 days of the experiments. (D) Total METH intake for the first 9 days of the experiment [2-way ANOVA showed significant effects of sex [F(1,31) = 6.040, P = .0198], training day [F(8,248) = 17.20, P < .0001], and sex × training day interaction [F(8,248) = 3.762, P = .0004]. (E) Patterns of drug infusions for rats in low [(female Me LT, n = 9) and (male Me LT, n = 4)] and high [(female Me HT, n = 8) and (male Me HT, n = 11)] METH takers. Key to statistics: *P < .05, **P < .01, ***P < .001, female and male METH groups compared with respective saline groups; #P < .05, ##P < .01, comparison of male METH group with female METH group; $P < .05, $$P < .01 comparison of daily METH infusions of female and male rats with respect to METH infusion on the first day of training; !P < .05, !!P < .01, comparison between female and male high METH takers to respective low METH groups. All values represent means ± SEM of number of animals indicated in the figure.

Additional statistical analyses also revealed earlier escalation of METH intake in male rats during the first 9 days compared with female METH rats (see Figure 1D for METH intake × sex interaction during that time), with significant effects of sex [F(1,31) = 6.040, P = .0198], training day [F(8,248) = 17.20, P < .0001], and sex × training day interaction [F(8,248) = 3.762, P = .0004] (see Figure 1B,D).

Very close inspection of the behavioral data of individual rats revealed that some female and male rats escalated their METH intake, whereas others only moderately increased their intake from their baseline intake (see Figure 1E). We were thus able to separate both female and male METH groups, based on their escalation patterns, into 2 separate drug intake phenotypes that consisted of (1) low [(female Me LT, n = 9) and (male Me LT, n = 4)] and (2) high [(female Me HT, n = 8) and (male Me HT, n = 11)] METH takers (Figure 1E). Rats that significantly escalated their METH intake from days 4 to 20 compared with their first day of SA were termed high takers, whereas those that did not significantly increase their intake over time were termed as low takers. Patterns of METH intake for low and high METH rats are shown in Figure 1E. We think it is important to classify the animals according to their intake because biochemical and/or molecular responses might vary according to the amount of METH that rats had taken, as shown previously in experiments in which investigators had injected various doses of METH to male rats (Jayanthi et al., 2005; Beauvais et al., 2010). This separation is also clinically relevant because human subjects have been reported to show differences in the amount of METH they ingest (He et al., 2013). The ANOVA analysis for METH infusions received included the between-participant factors (low vs high) and within-participant factor day (training days 1–20) and their interactions. Repeated-measures 2-way ANOVA comparing low to high female METH rats showed significant effects of group [F(1,15) = 22.30, P = .0003] and training day [F(19,285) = 7.474, P < .0001] but no significant interaction [F(19,285) = 0.9571, P = .5125]. Similar comparisons for male METH rats also revealed significant effects of group [F(1,13) = 11.74. P = .0045] and training day [F(19 247) = 7.408, P < .0001] but no significant interactions [F(19,247) = 1.363, P = .1456].

Incubation of METH Seeking in Male and Female Rats After Prolonged Withdrawal

To test if there were differences in METH-seeking behaviors in female and male rats that had self-administered METH and then withdrawn from the drug, we compared cue-induced active lever responses on WD3 and WD30 (Figure 2A). Previous studies had reported that male rats withdrawn from METH SA increased active level pressing at various intervals after withdrawal (Scheyer et al., 2016; Cadet et al., 2017; Krasnova et al., 2017).

Figure 2.

Female and male rats show comparable incubation of methamphetamine (METH) seeking. (A) METH-seeking behaviors by female and male rats at withdrawal days 3 (WD3) and 30 (WD30) [3-way ANOVA showed significant effects of withdrawal days [F(1, 51) = 24.86, P < .0001] and group (saline vs METH) [F(1, 51) = 34.06, P < .0001] but no effect of sex [F(1, 51) = 1.454, P = .2334]. In addition, METH seeking at WD3 correlates positively with total METH intake for (B) female and (C) male rats. Drug-seeking behaviors at WD30 also correlate positively with total METH intake for both (D) female and (E) male rats. Key to statistics: *P < .05, **P < .01, comparison of female and male METH rats to saline rats; #P < .05, ##P < .001, within sex comparison of lever pressing at WD3 and WD30.

For the cue-induced METH-seeking tests, 3-way ANOVA revealed main effect of withdrawal days [F(1, 51) = 24.86, P < .0001] and group (saline vs METH) [F(1, 51) = 34.06, P < .0001] but no effect of sex [F(1, 51) = 1.454, P = .2334]. There were also no significant interactions between withdrawal days and sex [F(1, 51) = 0.5760, P = .4514], withdrawal days and group [F(1, 51) = 1.704, P = .1976], and sex and group [F(1, 51) = .7082, P = .4040]. Moreover, the overall withdrawal days × sex × group was not significant [F(1, 51) = 0.007915, P = .9295]. Post-hoc tests showed greater responses of active lever on WD3 and WD30 for both female and male METH rats compared with respective saline rats. Both female and male METH rats also showed greater cue-induced lever pressing on WD30 compared with WD3 (Figure 2A). No differences were observed in active lever pressing between female and male METH rats on WD3 and WD30, indicating no sex differences in drug-seeking behavior, a potential marker of relapse to drug-taking behaviors. Interestingly, regression analyses revealed significant correlations between total amount of METH intake per animal (mg/kg) and active lever presses for both sexes (Figure 2B–E). These results suggest that humans who use larger amounts of METH may be more prone to relapse during treatment.

Expression of Neuropeptide mRNA Levels in the Rat NAc After Prolonged Withdrawal from METH SA

Dynorphin mRNA Levels

Previous studies have reported increases in prodynorphin (Pdyn) mRNA (Cadet et al., 2016) and protein (Whitfield et al., 2015) levels in male rats after METH SA. We thus compared Pdyn mRNA levels in the NAc of drug naïve and METH-exposed female and male rats after the second drug-seeking test. Two-way ANOVA revealed significant effects of sex [F(1,29) = 225.2, P < .0001] and METH intake [F(2,29) = 6.704, P = .0040] but not their interaction [F(2,29) = 1.436, P = .2542]. Post-hoc tests showed that female rats had significantly higher basal Pdyn mRNA levels than males (Figure 3A), findings that are comparable with the observations of higher levels of DYN protein levels in the hippocampus (Van Kempen et al., 2013) and of striatal Pdyn mRNA levels in the striatum (Chen et al., 2009) of female mice.

Figure 3.

Prodynorphin (Pdyn) mRNA levels in the nucleus accumbens (NAc) shows increases in male, but not female, methamphetamine (METH) self-administration (SA) rats after 30 days of withdrawal: 2-way ANOVA revealed significant effects of sex [F(1,29) = 225.2, P < .0001] and METH intake [F(2,29) = 6.704, P = .0040] but no significant interaction [F(2,29) = 1.436, P = .2542]. Key to statistics: ***P < .001, compared with control rats; #P < .05 comparison between low and high METH takers; !!!P < .001, comparison between female and male rats. F Ct, female controls; F Me LT, female low METH takers; F Me HT, female high METH takers; M Ct, male controls; M Me LT, male low METH takers; M Me HT, male high METH takers.

As previously reported (Cadet et al., 2016; Whitfield et al., 2015), male high, but not low, METH takers showed significant increases in the expression of Pdyn mRNA in the NAc compared with control male rats (Figure 3A). These observations support the idea of separating the rats based on the amount of self-administered METH. However, METH had no significant effects on the Pdyn mRNA expression in female rats (Figure 3A). Even with the METH-associated increases in Pdyn mRNA levels in the high-METH takers, these values remained lower than those of female rats irrespective of conditions (Figure 3A).

To test if there were any relationships between drug-seeking behaviors and Pdyn mRNA expression, we ran regression analysis between fold-changes in Pdyn expression and active lever presses after a month of forced withdrawal. No significant correlation was observed in either female (r = 0.2458, P = .3255) or male (r = 0.3434, P = .1772) rats.

Orexin/Hypocretin Receptors

Because sexual dimorphism has also been reported in brain orexin/hypocretin (HCRT) signaling systems (Johren et al., 2002; Grafe et al., 2017; Loewen et al., 2017), we compared mRNA levels of orexin receptors in male and female rats. Indeed, Hcrtr1 (Figure 4A) and Hcrtr2 (Figure 4B) mRNA levels were higher in the NAc of female rats compared with males, findings that are consistent with a report of higher Hcrt expression in the hypothalamus of female compared with male rats (Johren et al., 2002).

Figure 4.

Expression of orexin receptors in the nucleus accumbens (NAc) of rats after 30 days of withdrawal from methamphetamine (METH) self-administration (SA). (A) Relative Hcrtr1 mRNA levels in female and male METH rats compared with controls. There were significant effects of sex [F(1,30) = 65.57, P < .0001] and METH intake [F(2,30) = 1.907, P = .1661] but no significant interaction [F(2, 30) = 0.3862, P = .6829]. (B) Relative Hcrtr2 mRNA levels in female and male METH SA rats compared with controls. Results of 2-way ANOVA: sex [F(1,30) = 36.48, P < .0001], METH intake [F(2,30) = 5.178, P = .0117], and interaction [F(2,30) = 1.250, P = .3009]. (C) Hcrtr1 mRNA levels are negatively correlated with active levels presses (withdrawal day 30 [WD30]) in female METH SA rats at WD30. (D) Hcrtr2 mRNA levels are negatively correlated active levels presses (WD30) in female METH SA rats. There were no significant correlations between active lever presses (WD30) and (E) Hcrtr1 or (F) Hcrtr2 mRNA levels in male METH SA rats. Key to statistics: *P < .05, **P < .01, ***P < .001, comparison control vs low and high METH taker groups; #P < .05, ##P < .01, ###P < .001, comparison between low and high METH taker groups; !P < .05, !!P < .01, !!!P < .001, comparison between female and male. F Ct, female controls; F Me LT, female low METH takers; F Me HT, female high METH takers; M Ct, male controls; M Me LT, male low METH takers; M Me HT, male high METH takers.

We also found significant effects of sex [F(1,30) = 65.57, P < .0001], but not METH, intake [F(2,30) = 1.907, P = .1661] or their interaction [F(2, 30) = 0.3862, P = .6829] on Hcrtr1 expression. Figure 4A shows that METH SA did not impact the expression of Hcrtr1 mRNA in either females or males. Regression analysis revealed a negative correlation between Hcrtr1 expression and active lever presses at WD30 in females (Figure 4C) but not in males (Figure 4E).

Analysis of Hcrtr2 mRNA expression revealed significant effects of sex [F(1,30) = 36.48, P < .0001] and METH intake [F(2,30) = 5.178, P = .0117] but no significant interaction [F(2,30) = 1.250, P = .3009]. A regression analysis between Hcrtr2 expression and active lever presses at WD30 also showed significant correlation in females (Figure 4D) but not males (Figure 4F).

CRH and Its Receptors

Sex differences in the expression of CRH and its receptors have also been reported in some brain regions (Weathington et al., 2014; Zohar et al., 2015; Lukkes et al., 2016; Bangasser and Wiersielis, 2018). In the present study, we found that levels of Crh (Figure 5A) and Crhr1Figure 5B) were comparable between males and females; however, Crhr2 (Figure 5C) expression was significantly higher in females than in males.

Figure 5.

The effects of withdrawal from methamphetamine (METH) self-administration (SA) on the expression of Crhr2 mRNA levels are sexually dimorphic. (A) Relative Crh mRNA expression in female and male rats exposed to METH. Results of 2-way ANOVA: sex [F(1,30) = 7.045, P = .0126], METH intake [F(2,30) = 6.947, P = .0033], and interaction [F(2,30) = 0.4342, P = .6518]. (B) Crhr1 mRNA levels show no significant METH-related changes in either female or male rats. (C) Effects of METH SA on Crhr2 mRNA levels in female and male METH rats. Results of 2-way ANOVA: sex [F(1,30) = 29.58, P < .0001], METH intake [F(2,30) = 3.067, P = .0614], and interaction [F(2,30) = 0.7693, P = . 4723]. (D) Crhr2 mRNA levels are positively correlated to active levels presses measured on withdrawal day 30 (WD30) in female rats but not (E) in male rats. Key to statistics: *P < .05, **P < .01, compared with controls; #P < .05, comparison between low METH takers; !P < .05, !!P < .01, !!!P < .001, comparison between females and males. F Ct, female controls; F Me LT, female low METH takers; F Me HT, female high METH takers; M Ct, male controls; M Me LT, male low METH takers; M Me HT, male high METH takers.

Analysis of Crh mRNA expression revealed significant effects of sex [F(1,30) = 7.045, P = .0126] and METH intake [F(2,30) = 6.947, P = .0033] but no significant interaction [F(2,30) = 0.4342, P = .6518] (Figure 5A). There were no significant correlations between active lever presses at WD30 in either females (r = −1881, P = .4549) or males (r = −0882, P = .7279).

Crhr1 expression showed no significant effects of METH intake [F(2,31) = 0.5599, P = .5770], significant effects of sex [F(1,31) = 4.173, P = .0496], but no sex by METH interaction [F(2,31) = 0.7514, P = .4801] (Figure 5B). There were also no significant correlations between drug seeking at WD30 and Crhr1 expression in females (r = 0.1834, P = .4663) or males (r = 0.0775, P = .7524).

In the case of Crhr2 expression, there were significant effects of sex [F(1,30) = 29.58, P < .0001] and a trend for METH intake [F(2,30) = 3.067, P = .0614] but no significant interaction [F(2,30) = 0.7693, P = .4723]. The expression of Crhr2 was slightly but not significantly higher in both low and high female METH takers compared with the control group (Figure 5C). In contrast, Crhr2 mRNA levels were significantly increased in male high-METH takers compared with control and low takers. Like the observations for dynorphin, the levels of Crhr2 mRNA were still lower than those of female rats even after METH withdrawal-associated increases in the males. Figure 5D and E illustrate the fact that drug-seeking behaviors correlated with increases in Crhr2 mRNA levels in females but not in males at WD30.

Vasopressin and Its Receptors

As in the case of other neuropeptides, sexually dimorphic expression and distribution of vasopressin signaling systems have been reported in the brain (Dumais and Veenema, 2016; Smith et al., 2017). Here, we found higher Avp mRNA levels in male rats (Figure 6A) but higher Avpr1a mRNA levels in females (Figure 6B). However, there were no sex differences in basal Avpr1b expression (Figure 6C).

Figure 6.

Changes in the expression of Avp and its receptors after prolonged withdrawal from methamphetamine (METH) self-administration (SA). (A) Relative Avp mRNA levels in female and male METH SA rats compared with controls: sex [F(1,29) = 50.40, P < .0001], METH intake [F(2,29) = 17.10, P < .0001], and interaction [F(2,29) = 11.84, P = .0002]. (B) Relative Avpr1a mRNA levels in female and male METH rats compared with controls. Results of 2-way ANOVA: sex [F(1,32) = 38.89, P < .0001], METH intake [F(2,32) = 11.3845, P = .0002], and interaction [F(2,32) = 1.757, P = .1888]. (C) Changes in Avpr1b mRNA levels in female and male METH SA rats compared with controls: METH intake [F(2,31) = 21.94, P < .0001], sex [F(1,31) = 0.5209, P = .4759], and interaction [F(2,31) = 0.5064, P = .6076]. Avpr1b mRNA levels are positively correlated to active levels presses (withdrawal day 30 [WD30]) measured in (D) female rats, but not (E) in male METH SA rats. Key to statistics: *P < .05, **P < .01, ***P < .001, comparison control vs low and high METH taker groups; #P < .05, ##P < .01, ###P < .001 comparison between low and high METH taker group; !P < .05, !!P < .01, !!!P < .001, comparison between female and male rats. F Ct, female controls; F Me LT, female low METH takers; F Me HT, female high METH takers; M Ct, male controls; M Me LT, male low METH takers; M Me HT, male high METH takers.

Changes in Avp expression showed significant effects of sex [F(1,29) = 50.40, P < .0001], METH intake [F(2,29) = 17.10, P < .0001], and their interaction [F(2,29) = 11.84, P = .0002] (Figure 6A). Regression analysis revealed no correlation in Avp expression to drug-seeking behaviors in either females (r = −3444, P = .1615) or males (r = −0.1004, P = .6918).

In the case of Avpr1a mRNA expression, there were significant effects of sex [F(1,32) = 38.89, P < .0001] and METH intake [F(2,32) = 11.3845, P = .0002], but no interaction [F(2,32) = 1.757, P = .1888], with METH SA being associated with decreased Avpr1a expression in both males and females (Figure 6B). There were, however, no correlations between mRNA levels of Avpr1a and drug seeking in females (r = −1955, P = .3958) or males (r = −2.2335, P = .3511).

In the case of Avpr1b mRNA levels, there were significant effects of METH intake [F(2,31) = 21.94, P < .0001] but no effect of sex [F(1,31) = 0.5209, P = .4759] or their interaction [F(2,31) = 0.5064, P = .6076] (Figure 6C). In addition, we found significant correlations between drug seeking at WD30 and decreases in Avpr1b mRNA levels in female (Figure 6D) but not in male (Figure 6E) rats.

Discussion

Sex differences have been reported in humans suffering from substance use disorders (SUDs) and in some animal models of SUDs (Dluzen and Liu, 2008; Becker and Chartoff, 2019). However, much remains to be done to test the generalizability of these behavioral differences and to identify potential neurobiological bases of any identified sexually dimorphic behaviors. Thus, the initial aim of our study was to assess potential sex-related behavioral differences in rats that were trained to self-administer METH. A secondary aim was to identify potential differences in molecular consequences between the 2 sexes. A third aim was to assess the existence of any relationship of these molecular indices to METH-seeking behaviors after a 30-day withdrawal. Interestingly, we found that male rats took more drug overall during the 20 days of SA training. We observed, in addition, that both sexes exhibited incubation of cue-induced METH seeking during 30 days of forced abstinence. In terms of gene expression, there were no sex differences in the expression of Crh, Crhr1, and Avpr1b. However, females expressed higher levels of Pdyn, Hcrtr1, Hcrtr2, Crhr2, and Avpr1a mRNAs relative to males, whereas male rats had higher levels of Avp mRNA expression under both saline and METH conditions. Importantly, METH SA impacted the expression of most genes in similar ways in both sexes, although Pdyn mRNA expression was significantly increased only in males after the month of withdrawal from METH SA.

Escalation of METH Intake and Incubation of METH Seeking

Our observations of sex differences in the behavioral responses to METH exposure are somewhat comparable with those of some previous reports dealing with METH SA (Ruda-Kucerova et al., 2015; Venniro et al., 2017). For example, Ruda-Kucerova et al. (2015) reported that male rats took more METH during the last 5 days of training during a 15-day experiment. We also found that males took more METH than female rats. These observations are of clinical interest because He et al. (2013) reported that hospitalized METH-addicted individuals consisted of about 81% men and 19% women on an inpatient unit, suggesting that more males might suffer from problematic METH use that necessitated hospital admission. Also like our results, Venniro et al., 2017 reported that males exhibited greater escalation of METH SA than females. However, they reported that male and female rats took similar amounts of METH (Venniro et al., 2017), whereas we found that males took more METH. These differences may be due to different strains of rats used in the 2 studies. We used Long Evans in the present study, whereas Venniro et al. 2017 used Sprague-Dawley rats. In addition, they food-trained their animals before the start of their experiments but we did not. Importantly, consistent with our report, Venniro et al., 2017 did not find any sex differences in cue-induced METH craving after extended abstinence. Time-dependent increases in cue-induced drug seeking have previously been reported for cocaine (Grimm et al., 2001, 2003), METH (Krasnova et al., 2014), and opioids (Blackwood et al., 2019). In our studies of METH, we have observed incubation of drug seeking using both within- and between-participant designs (Krasnova et al., 2014). We have also reported incubation of oxycodone seeking using a within-participant approach (Blackwood et al., 2019). However, it appears that observations of incubation of food seeking depend on the between-participant design (Krasnova et al., 2014). Enhanced responses using the within-participant approach were previously observed after exposure to aversive stimuli (McAllister and McAllister, 1963), phenomena that were later dubbed incubation of fear/anxiety (McMichael, 1966; Eysenck, 1968; Allen and Mitcham, 1970). The observations of incubation using both within- and between-participant paradigms support the robustness of this phenomenon in rodents. Nevertheless, more remains to be done to extend its utility in clinical situations.

It is noteworthy, nevertheless, that our results are quite different from those of Reichel et al., 2012, who had reported that female rats escalated METH intake more than males. The differences between our study and theirs may, in part, be due to the fact the 2 studies used different doses of METH for each drug infusion. Specifically, we used 0.1 mg/kg of METH for both sexes, whereas they used 0.02 and 0.0175 mg/kg of METH for males and females, respectively. Thus, our dose was about or more than 5 times the dose that Reichel et al. 2012 used for males and females, respectively. Thus, it is possible that our results would have been similar to those of that study (Reichel et al., 2012) had we used lower doses of METH. These issues are important and will need to be addressed in future experiments. In any case, taken together, our results and those of Reichel et al. 2012 are consistent with those of a previous study that had reported that lower doses of d-amphetamine functioned as better reinforcers in women, whereas higher doses served as better reinforcers for men (Vansickel et al., 2010). We chose to use METH (0.1 mg/kg) based on earlier findings that male rats trained to self-administer METH using that dose displayed reliable METH SA (Cadet et al., 2016; Krasnova et al., 2017; Torres et al., 2017) and incubation of METH seeking (Cadet et al., 2017; Krasnova et al., 2017). It is important to note that Venniro et al. 2017 used a similar dosing schedule in their study. Another difference between the 2 studies is the fact that Reichel et al. 2012 had trained their rats to self-administer METH using 7 days of 1-hour daily sessions followed by extended access (6 hours) to METH for another 7 days, whereas we gave 6 hours access to the rats from the beginning of our study.

It could also be argued that hormonal status might have impacted our results. However, although we did not measure hormonal status in our study using freely cycling female rats, Reichel et al., 2012 did not observe any significant effects of estrous cycle on METH-related behaviors. It is worth noting that, even though male animals are often used as “controls” in studies focusing on sexual dimorphisms, male hormones can also cycle with diurnal and seasonal changes reported for testosterone levels in rodents (Moeller et al., 1988) and men (Gupta et al., 2000). In any case, future studies will need to be more thoughtful when measuring differences between male and female rodents and should take into consideration using females and males as alternate control depending on the behavior of interest.

Sex Differences in Dynorphin Expression at Baseline and After METH Exposure

Our findings of higher basal Pdyn expression in the NAc of female rats are consistent with those of a previous study that had reported similar findings in the hippocampus (Van Kempen et al., 2013) and in the striatum (Chen et al., 2009) of female mice. Our observations of increases in Pdyn mRNA in male Long Evans rats exposed to METH SA are also consistent with the previous report of increased Pdyn mRNA in the NAc of male Sprague-Dawley rats killed within 2 hours of cessation of escalated METH intake (Cadet et al., 2016). Male Wistar rats killed at 24 hours after stopping METH SA also showed increased Pdyn immunoreactivity in their NAc (Whitfield et al., 2015). That study did not include female rats. Thus, the lack of changes in Pdyn mRNA levels after escalated METH intake in females is a novel finding consistent with previous reports that female and male animals respond differentially to administration of kappa opioid agonists (Russell et al., 2014; Chartoff and Mavrikaki, 2015). For example, female Sprague-Dawley rats were reported to be less sensitive to the depressive effects of the kappa opioid agonist U-50488 (Russell et al., 2014). It is also important to note that Pdyn mRNA levels in males did not reach the levels of expression observed in female rats even in male rats with METH-associated increases. These findings indicate that females may exhibit much higher dynorphinergic tone than males in the presence or absence of METH. In any case, when taken together with our present observations, the accumulated evidence supports the notion that activation of the dynorphin-KOR signaling pathway that is located in D1-containing neurons in the NAc (Al-Hasani et al., 2015) may show sexual dimorphic responses during drug SA. This dimorphism might depend on sex differences in KOR-dependent regulation of dopamine dynamics in reward circuitries (Yang et al., 2018; Tejeda and Bonci, 2019).

METH SA and Changes in Other Neuropeptide mRNAs in the NAc

Some neuropeptides, which include orexin/hypocretin, CRH, and vasopressin, are thought to participate in several aspects of SUDs (Chen et al., 2013; Bisagno and Cadet, 2014; James et al., 2017; Roberto et al., 2017; Koob and Schulkin, 2018). In the case of orexin/hypocretin, most of the studies have evaluated the role of that system in models of cocaine use disorder (James et al., 2017; Schmeichel et al., 2017, 2018). For example, systemic injections of SB334867, a HCRTR1 antagonist, dose-dependently decreased cocaine intake via HCRTR neurotransmission in the central amygdala (Schmeichel et al., 2017). In addition, escalated heroin SA was accompanied by increased Hcrtr2 mRNA levels in the central amygdala and heroin SA was blocked by the systemic administration of the HCRTR2 antagonist NBI-80713 (Schmeichel et al., 2015). In contrast to the data reported for heroin, we found significant decreases in Hcrtr2, but no significant changes in Hcrtr1 or mRNA levels in the NAc of females (see Figure 4C) or males (see Figure 4E).

We did, however, find significant correlations between METH-seeking behaviors and Hcrtr1 and Hcrtr2 mRNA levels in only female rats, suggesting the existence of sexual dimorphism between relapse and receptor sites post-synaptic to terminals of projections emanating from the hypothalamus (Peyron et al., 1998; Nambu et al., 1999). Thus, taken together, these observations suggest that changes in hypocretin/orexin signaling pathways might play a potential role in relapse to METH taking as suggested for other addictive drugs (Plaza-Zabala et al., 2012; James et al., 2017). This idea is consistent with the report of higher plasma hypocretin in recently abstinent METH abusers (Chen et al., 2016).

The potential role of CRF/CRH in addiction to drugs including METH (Nawata et al., 2012; Cadet et al., 2014b; Jacobskind et al., 2018; Jayanthi et al., 2018) has been discussed extensively (Roberto et al., 2017; Koob and Schulkin, 2018). CRH-regulated pathways have been implicated in compulsive drug taking, withdrawal-associated psychiatric pathologies, and relapses to drug taking after variable periods of abstinence (Nawata et al., 2012; Roberto et al., 2017). In this study, we observed significant decreases in Crh but increases in Crhr2 mRNA levels in both female and male rats that exhibited incubation of METH seeking. The decreases in CRH expression are consistent, in part, with the report of decreased plasma cortisol in a sample of METH polydrug users (Carson et al., 2012). It is important to point out that, when we assessed the relationships of the changes to lever pressing observed 1 day before tissue collection, only the changes in Crhr2 mRNA expression in the female rats showed positive correlation to drug-seeking behaviors at WD30. These observations support sexual dimorphism in the response of the CRH signaling system working through the CRHR2 receptor during withdrawal from METH SA. These findings further suggest that the role of these receptors in addiction models need to be further evaluated.

Unlike the other peptides that show higher basal mRNA levels in female rats, basal Avp mRNA levels were substantially higher in males. These observations are consistent with previous reports (see Dumais and Veenema, 2016 for a review). In contrast, as previously reported (Smith et al., 2017), female rats had higher Avpr1a mRNA levels in our study. Analysis of the relationships observed in mRNA levels revealed significant negative correlations between only Avpr1b expression and cue-induced relapse to drug seeking only in female rats after a long period of withdrawal. The results of these analyses suggest that the development and evaluation of agents that impact these receptors may be worthwhile endeavors.

In summary, we report that male rats show greater METH intake than females during METH SA experiments. Interestingly, only males show increased Pdyn mRNA levels after exposure to METH SA. We found, in addition, that there were significant correlations between cue-induced METH seeking and the levels of Hcrtr1, Hcrtr2, Crhr2, and Avpr1b mRNAs only in the female rats that showed incubation of craving. Altogether, these results suggest sexual dimorphism in behavioral responses during METH SA and molecular responses after a month of forced withdrawal. Further work is necessary to clarify the direct involvement of these molecular changes in METH SA. In any case, these results support the notion that sex-dependent changes need to be considered when assessing potential therapeutic agents for individuals who are suffering from MUD.

Interest Statement: None.