Male and female C57BL/6 mice display drug-induced aversion and reward in the combined conditioned taste avoidance/conditioned place preference procedure

Abstract

Background:

Drugs of abuse have rewarding and aversive effects that, in balance, impact abuse potential. Although such effects are generally examined in independent assays (e.g., CPP and CTA, respectively), a number of studies have examined these effects concurrently in rats in a combined CTA/CPP design. The present study assessed if similar effects can be produced in mice which would allow for determining how each is affected by subject and experiential factors relevant to drug use and abuse and the relationship between these affective properties.

Methods:

Male and female C57BL/6 mice were exposed to a novel saccharin solution, injected (IP) with saline or 5.6, 10 or 18 mg/kg of the synthetic cathinone, methylone, and placed on one side of the place conditioning apparatus. The following day, they were injected with saline, given access to water and placed on the other side of the apparatus. After four conditioning cycles, saccharin avoidance and place preferences were assessed in a final two-bottle CTA test and a CPP Post-Test, respectively.

Results:

In the combined CTA/CPP design, mice acquired a significant dose-dependent CTA (p = 0.003) and a significant CPP (p = 0.002). These effects were independent of sex (all ps > 0.05). Further, there was no significant relationship between the degree of taste avoidance and place preference (p > 0.05).

Conclusions:

Similar to rats, mice displayed significant CTA and CPP in the combined design. It will be important to extend this design in mice to other drugs and to examine the impact of different subject and experiential factors on these effects to facilitate predictions of abuse liability.

1. Introduction

Drugs of abuse are complex compounds with multiple stimulus properties (Koob and Moal, 2006; Wise et al., 1976). One such property is a drug’s rewarding effects which have been well documented in a variety of behavioral procedures (Bozarth, 1987; Wise, 1996), including the conditioned place preference design that assesses a drug’s rewarding effect by demonstrating shifts in preference for the drug-paired side (for reviews, see Tzschentke, 1998, 2007). Not all of these stimulus effects are rewarding, however, in that the vast majority of drugs of abuse also induce significant taste avoidance (see Cappell and LeBlanc, 1977; Hunt and Amit, 1987; Riley et al., 2022). In this procedure, animals come to avoid consuming fluids and foods paired with drug injection (presumably indicative of the aversive effect of the drug; for reviews, see Garcia and Ervin, 1968; Rozin and Kalat, 1971; for a history of taste avoidance learning, see Freeman and Riley, 2009; for assessments of conditioned place aversion see Cunningham and Henderson, 2000; Risinger and Oakes, 1995; for a direct comparison of taste and place avoidance procedures, see Gore-Langton et al., 2015). Such apparent paradoxical findings of rewarding and aversive effects of the same drug have been discussed in the context of the vulnerability to drug use and abuse, specifically, how these two effects interact to impact the intake of a drug, i.e., its self-administration. If a drug is highly rewarding (as assessed in a conditioned place preference design; CPP), but with minimal aversive effects (as assessed in a conditioned taste avoidance design; CTA), it would be readily consumed. Conversely, a drug that induces a strong taste avoidance, but no place preference, would not support IVSA. That is, the rewarding effects of a drug would support its intake, whereas its aversive effects would limit it (for a discussion of the potential interaction of these affective properties see Cunningham, 1979; Stolerman and D’Mello, 1981; Turenne et al., 1996). Importantly, these two effects of drugs of abuse appear dissociable as they are not correlated, i.e., animals that display a strong conditioned taste avoidance could show a weak or strong place preference (and vice versa; Verendeev and Riley, 2011; see also King et al., 2015; Turenne et al., 1996) and manipulations known to affect either CTA or CPP often have no effect on the other (see Sellings et al., 2008; Shram et al., 2006; Yu et al., 2021).

While assessments of the rewarding and aversive effects of drugs of abuse have generally been made in separate studies and often under different parametric conditions (for examples, see Acevedo et al., 2013; Kosten et al., 1994), work assessing aversion and reward have also examined these effects in a combined CTA/CPP design that allows for their concurrent assessment in the same animal and under identical parametric conditions. For example, Reicher and Holman (1977) gave female Sprague-Dawley rats injections of amphetamine and placed them on one side of a two-compartment shuttle box during which they had access to a novel-flavored solution (banana or almond). On the following day, animals were injected with the drug vehicle and placed on the opposite side of the shuttle box with access to the other novel solution (almond or banana). Under these conditions, amphetamine induced a significant taste avoidance and place preference, supporting the fact that drugs can concurrently produce aversive and rewarding effects, respectively. Subsequent to these initial studies, a variety of compounds have been reported to support both CTA and CPP in this design (Dannenhoffer and Spear, 2016; King et al., 2015; Mayer and Parker, 1993; Verendeev and Riley, 2011; for a discussion, see Riley et al., 2022). In addition to demonstrations of their concurrent acquisition, this design also allows investigators to explore the impact of a host of experimental manipulations (pharmacological, drug history, anatomical) and subject variables (sex, age, genetic background) on their display (Cunningham et al., 2009; Ettenberg et al., 2015; Stolerman and D’Mello, 1981; for a recent review, see Riley et al., 2022). Given that the balance of reward and aversion mediates drug intake, an awareness of the impact of various factors on either of these effects (and their balance) may be important to understanding use and abuse potential (see Cunningham et al., 2009; Ettenberg et al., 2015; Hunt and Amit, 1987; Riley et al., 2022).

Although the combined CTA/CPP design has been widely used, work with this procedure has been limited to rats. To date, assessments of aversion and reward in mice have been made in different groups run separately in taste avoidance and place preference designs (for excellent work in such assessments, see Cunningham, 2014; Risinger and Cunningham, 2000; see also Cunningham et al., 1991; Phillips et al., 2005). Establishing the combined design in mice may be important given their use in a number of genetic and pharmacological manipulations in relation to aversion and reward. For example, mice are popular among CTA and CPP studies using inbred (Broadbent et al., 2002; Cunningham, 2014), transgenic (Crooks et al., 2010; Janus et al., 2004), selectively bred lines (Phillips et al., 2005; Schmill et al., 2021), and KO/KI models (Risinger et al., 2001; Sora et al., 1998). In assessing the role of different neurotransmitter systems in CTA and CPP, mice have been commonly used in pharmacological assessments as well (Gommans et al., 2000; Jerlhag and Engel, 2011). Such an extension of the combined design also allows for the assessment of other subject and experimental factors important in the potential use and abuse of drugs (for examples see Dannenhoffer and Spear, 2016; Mayer and Parker, 1993; Wang et al., 2021; for reviews see Cunningham et al., 2009; Davis et al., 2009).

Interestingly, mice and rats have been reported to differ in taste avoidance and place preference conditioning. For instance, while ethanol reliably produces CPP in mice (Cunningham et al., 2000, 2003; Risinger et al., 2001), it is often reported to produce CPA in rats (Bedingfield et al., 1999; Cole et al., 2003). In addition, mice have been reported to show weaker ethanol-induced CTA than rats (Broadbent et al., 2002; Crabbe et al., 2019). The basis for these differences is not known, although mice have been shown to metabolize ethanol faster (Abel, 1982), have sharper and faster increases and declines in blood ethanol concentration (BEC; Livy et al., 2003) and faster gastric emptying which affects the rate of absorption for gavage (Bossoni et al., 1979). Although the majority of work comparing taste avoidance and place preference in mice have been done with alcohol, species differences in these behaviors have been explored with other drugs as well (Davis et al., 2009; Freet et al., 2018; Nachman and Ashe, 1973; Risinger and Cunningham, 2000). Thus, the extension of the CTA/CPP design to mice may allow for the opportunity to draw parallels between species (i.e., mice and rats) which is important for informing future studies which may employ other drugs or assess different subject and experiential factors.

The purpose of the present study was to examine drug-induced aversion and reward (and their relationship) in mice in the combined CTA/CPP procedure. Further, given that taste avoidance and place preference have been reported to be impacted by sex (see Becker et al., 2017; Riley et al., 2018) and the general importance of considering sex as a biological variable (see Miller et al., 2017), both male and female mice were assessed. Specifically, male and female C57BL/6 mice were given access to a novel saccharin, injected with either 0, 5.6, 10 or 18 mg/kg of methylone and then placed on one side (either their preferred or non-preferred side; see below) of a place preference apparatus. These two measures were co-examined to determine the relationship between the two affective responses. Given the reports of taste avoidance and place preference with rats in the combined design and past work with methylone in separate behavioral assays, it was expected that methylone would produce dose-dependent CTA and CPP when concurrently assessed and that the two endpoints would be unrelated. Sex differences were not expected given prior work with methylone.

2. General methods

2.1. Subjects

The subjects were male (n = 32) and female (n = 32) experimentally naïve C57BL/6 mice. All subjects were bred within the American University animal research facility and were allowed to mature undisturbed until the start of testing. Starting between post-natal days (PND) 56–84 (8–12 weeks of age), animals were weighed daily for a minimum of 7 days to index health status and reduce handling stress during the subsequent experimental procedures. Subjects weighed between 22.7-26.8 g (males) and 18.1-21.3 g (females) at the start of experimental procedures. They were run in two replicates, each of which contained an equal number of both sexes (n = 16 of each sex; n = 32 total/replication) with all groups represented. For each replicate, four groups of subjects (n = 8 per group) were assessed daily with each group consisting of same sex subjects to avoid any potential effects of cross-contamination due to scents from the other sex. All procedures adhered to the Guidelines for the Care and Use of Laboratory Animals (National Research Council US, 2011) and the Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research (National Research Council US, 2003) and were approved by the Institutional Animal Care and Use Committee at American University.

2.2. Apparatus

Subjects of the same sex were housed four per group in OptiMouse cages (13.5 × 11.5 × 6.1; 75 sq. in). The room in which the cages were located was maintained on a 12-h light/dark cycle (0800–2000 h) and at 23 °C. Training and testing occurred during the lights-on phase of the light cycle. Unless stated otherwise, food and water were available ad libitum. For fluid access during CTA training and testing (see below), animals were transferred to separate individual OptiMouse cages on the side of which graduated Nalgene tubes could be placed for fluid presentation. For CPP training and testing, subjects were transferred to one of eight identical three-chambered CPP systems (68.5 × 21 × 34.5 cm; San Diego Instruments Place Preference System, San Diego, CA). Each system was divided into three distinct areas each of which was equipped with a photo beam array to record time spent in specific locations in the apparatus. The left side chamber of the apparatus (28 × 21 × 34.5 cm) had white walls and white metal diamond-plate flooring, and the right side (28 × 21 × 34.5 cm) had black walls and black plastic haircell textured flooring. The middle connecting chamber (14 × 21 × 34.5 cm), which was not used for conditioning, had grey walls and metal grid flooring consisting of stainless-steel rods spaced approximately 1 cm apart. The place preference chambers (as well as the room in which the chambers were located) were unlit, and a white noise generator was used to mask background noise.

2.3. Drugs and solutions

Racemic methylone hydrochloride (synthesized and provided by the Drug Design and Synthesis section, MTMDB, NIDA and NIAAA) was dissolved in isotonic saline (0.9 %) and injected intraperitoneally (IP) at 5.6, 10 and 18 mg/kg. Concentration was held constant across dose groups (1 mg/ml), and animals were injected at a volume based on the dose of methylone they were assigned to receive and their body weight on each day of training. Isotonic saline (vehicle) was administered to controls at the same volume. Drug and vehicle solutions were prepared daily and passed through a 0.2-um filter prior to injection to remove any potential particulates. Saccharin (sodium saccharin, Acros Organics) was prepared as a 1 g/l (0.1 %) solution in tap water.

Methylone was used as the conditioning drug in the present assessment given that is has been reported to induce significant taste avoidance in rats (see Manke et al., 2021) and place preferences in mice (see Karlsson et al., 2014; Miyazawa et al., 2011) when examined separately, although the conditions under which these effects were assessed varied across studies, e.g., species, dose, deprivation conditions. Methylone is a first-generation synthetic cathinone (also known as a “bath salt”) that is a beta ketone analogue of 3,4-methylenedioxymethamphetamine (MDMA), and by virtue of acting as a nonselective substrate releaser at the monoamine transporters, it triggers their release, while also blocking their reuptake (López-Arnau et al., 2012; Simmler et al., 2013).

2.4. Procedure

2.4.1. Combined CTA/CPP design

2.4.1.1. Water habituation.

At the beginning of experimental procedures, subjects were deprived of water and 24 h later were given 20-min access to tap water in the individual plastic testing cages. Following this access, the animals were returned to their home cages and each cage was wiped down with a cleaning solution (Sani-Plex 128 M, one-step disinfectant germicidal detergent) between animals. This limited access procedure (used to induce water consumption) was repeated for 10 days to allow consumption to stabilize (approaching the drinking tube within 2 s with the average volume of water consumed not increasing or decreasing by >0.15 ml for 3 consecutive days). Water was presented in graduated 50-ml Nalgene tubes, and intake was evaluated by the difference between pre- and post-consumption volumes (see Fig. 1, Procedural Timeline).

Fig. 1.

Procedural Timeline for subjects undergoing a combined conditioned taste avoidance (CTA)/conditioned place preference (CPP) procedure. On conditioning days, animals were given saccharin followed by a drug injection and placed on the drug-paired side (DPS). On saline days, animals are given water access followed by a saline injection and placed on the non-drug-paired side (Non-DPS).

Created by Shihui Huang with BioRender.com.

2.4.1.2. CPP pre-test.

Subsequent to stabilization of water consumption, subjects were given 20-min access to water in the test cages before being placed in the middle grey chamber of the place preference apparatus and allowed to explore all chambers for 15 min. Following this, animals were returned to their home cages. A paired sample t-test on absolute time spent on the white side vs. absolute time spent on the black side during the 15-min pre-test indicated an unbiased apparatus for each replicate (replicate 1: t = 0.867, p = 0.393; replicate 2: t = 0.991, p = 0.329). Although the apparatus was statistically unbiased, four animals (two males and two females) spent >65 % of the 15-min testing time on one side of the conditioning chamber during the Pre-Test, indicative of a strong natural bias, and were excluded from the statistical analysis of place preference and taste avoidance conditioning (although still run in the behavioral assessments). Time spent in the middle chamber was not used for conditioning or in the calculation of side preferences as it was not paired with drug or saline during conditioning. Each apparatus was wiped down with Sani-Plex 128 M between animals.

2.4.1.3. CTA/CPP conditioning.

On Day 1 of conditioning, subjects were placed in their individual test cages and given 20-min access to a novel saccharin solution. Same-sex subjects were run in groups of 8 and were assigned to conditioning groups such that saccharin consumption among groups was comparable. Males and females were then assigned to one of four drug groups and injected IP with either the saline vehicle or 5.6, 10 or 18 mg/kg of racemic methylone. These doses were based on previous work with rats in our laboratory in which racemic methylone induced significant taste avoidance (Manke et al., 2021) and on other work demonstrating that methylone induces significant CPP at comparable doses (Karlsson et al., 2014; Miyazawa et al., 2011). This resulted in a total of eight groups, i.e., Groups F0, F5.6, F10, F18, M0, M5.6, M10 and M18, where the letter indicates female or male and the number indicates drug dose (n = 8 per dose group). Following the injections, subjects were taken to a separate room and placed on one side of the place preference apparatus in a counterbalanced fashion such that half the animals in a given group were placed on their preferred side (defined as the side on which each mouse spent the most time during the Pre-Test) and the remaining half were placed on their non-preferred side for 30 min. Subjects were then returned to their home cages, and the test cages and place preference chambers were sanitized prior to the next set of animals. On the next day (Day 2), they were given 20-min access to water in the test cages, injected with vehicle and placed on the opposite side of the place preference chamber for 30 min. This two-day cycle was repeated for a total of four cycles.

2.4.1.4. CPP post-test and CTA two-bottle test.

On Day 9 of CTA/CPP testing, animals were given 20-min access to tap water, placed in the center grey middle area of the place preference apparatus and allowed to explore freely all three chambers for 15 min. Time spent in the two conditioning chambers was recorded to determine the percentage of time spent on the drug-paired side (DPS). On the next day, animals were placed in the plastic test cages and given 20-min access to both saccharin and tap water in a two-bottle avoidance test with no subsequent injections. On this test, one bottle was offered (saccharin or water) and once sampled, it was removed, and the second bottle was presented. Once animals sampled both bottles, they were then placed simultaneously on their respective sides of the cage. The order of presentation and side placement were counterbalanced across subjects, and consumption of both saccharin and water was recorded after 20 min had elapsed. Animals were then returned to their home cages with ad libitum water access. To determine saccharin preference, the percentage of saccharin consumed was calculated by diving the volume of saccharin consumed by total fluid consumption (volume of saccharin + volume of water) and then multiplied by 100.

2.5. Statistical analysis

The data collected during CTA acquisition was analyzed using a mixed model ANOVA with the between-subject factors of Sex (male or female) and Dose (0, 5.6, 10, and 18 mg/kg) and the within-subjects factor of Trial (1–4). The data obtained during the Pre- and Post-Tests and on the two-bottle test were analyzed using a two-way ANOVA with the same between-subject factors. In the case of a significant interaction, univariate and multivariate analyses were assessed followed by Bonferroni-adjusted multiple comparisons.

The relationship between the percent of saccharin consumed on the two-bottle CTA test and percent of time spent on the DPS on the CPP Post-Test was determined using Pearson correlation coefficients. Included in this analysis were both male and female subjects injected with 5.6, 10 and 18 mg/kg methylone.

Statistical significance was set to p ≤ 0.05.

3. Results

3.1. Conditioned taste avoidance

The 2 × 4 × 4 mixed model ANOVA on saccharin consumption in male and female mice revealed a significant main effect of Dose [F(3, 52) = 12.665, p < 0.001] and Trial [F(3, 156) = 47.328, p < 0.001] and a significant Dose × Trial interaction [F(9, 156) = 2.958, p = 0.003] (see Fig. 2). There was no main effect of Sex [F(1, 52) = 2.318, p = 0.134] or significant Sex × Dose [F(3, 52) = 0.696, p = 0.559], Sex × Trial [F(3, 156) = 1.298, p = 0.277] or Sex × Dose × Trial [F(9, 156) = 0.827, p = 0.592] interactions. In relation to the significant Dose × Trial interaction (collapsed across Sex), all methylone-injected subjects significantly decreased their consumption across conditioning. Specifically, subjects conditioned with 10 and 18 mg/kg methylone significantly decreased their saccharin consumption from their own baselines by Trial 2 and maintained these differences for the remainder of conditioning (all p’s < 0.001). Subjects conditioned with 5.6 mg/kg drank less than their baseline by Trial 3 (p = 0.024) and maintained this difference on Trial 4. Although all methylone injected groups significantly decreased consumption across trials, only the 10 and 18 mg/kg groups significantly differed from vehicle. Specifically, on Trials 2–4 subjects injected with 10 or 18 mg/kg of methylone drank significantly less than those injected with vehicle (all p’s p ≤ 0.018). Subjects injected with 10 mg/kg of methylone drank significantly less than those injected with 5.6 mg/kg on Trials 2 and 3, but they no longer differed on the final trial. Subjects injected with 18 mg/kg drank significantly less than those injected with 5.6 mg/kg on Trials 2–4. This suppression of saccharin consumption during conditioning was not a function of any unconditioned effects of methylone that might carryover from one trial to the next given that methylone’s half-life is relatively short (1 h; see Elmore et al., 2016; López-Arnau et al., 2013). Further, between each conditioning trial, subjects were given water access and injected with saline. On these water-recovery days, consumption of subjects previously injected with methylone did not differ from vehicle-injected controls (data not shown), indicating that there was no residual effect of methylone on general consumption.

Fig. 2.

Mean (±SEM) saccharin consumption (ml) over Trials 1–4 for male (top) and female (bottom) subjects injected with vehicle or 5.6, 10 or 18 mg/kg of methylone. ⁺Subjects injected 10 and 18 mg/kg of methylone significantly differed from those injected with vehicle. *Subjects injected with 10 or 18 mg/kg of methylone differed significantly from those injected with 5.6 mg/kg and vehicle. ^$Subjects injected with 18 mg/kg significantly differed from those injected with 5.6 mg/kg of methylone. Within group differences across trials are noted in the text.

3.2. Two-bottle avoidance test

The two-way ANOVA on the percent of saccharin consumption during the two-bottle test revealed a significant main effect of Dose [F(3, 52) = 9.562, p < 0.001] but no main effect of Sex [F(1, 52) = 1.620, p = 0.209] (see Fig. 3) or significant interaction of Dose × Sex [F(3, 52) = 1.847, p = 0.150]. In relation to the significant Dose effect (collapsed across Sex), subjects injected with 10 or 18 mg/kg of methylone consumed a significantly lower percent of saccharin than subjects injected with vehicle (all p’s < 0.001). Subjects injected with 18 mg/kg also consumed a significantly lower percent of saccharin compared to subjects injected with 5.6 mg/kg (p = 0.014).

Fig. 3.

Mean (±SEM) percent saccharin consumed (ml) on the two-bottle test for male (top) and female (bottom) subjects injected with vehicle or 5.6, 10 or 18 mg/kg of methylone. ^{^}Subjects injected with 10 or 18 mg/kg of methylone differed significantly from those injected with vehicle. ⁺Subjects injected with 18 mg/kg differed significantly from those injected with 5.6 mg/kg of methylone.

3.3. Conditioned place preference

Methylone produced dose-dependent conditioned place preference in mice regardless of sex. While there were no differences among groups in the percent of time on the DPS on the CPP Pre-Test (all p’s > 0.05), the two-way ANOVA on the percent of time on the DPS during the CPP Post-Test revealed a significant main effect of Dose [F(3, 52) = 5.630, p = 0.002] but not of Sex [F(1, 52) = 0.325, p = 0.571], and no significant interaction of Dose × Sex [F(3, 52) = 0.274, p = 0.844]. In relation to the main effect of Dose, subjects injected with 10 or 18 mg/kg of methylone spent a significantly higher percent of time on the DPS compared to animals injected with vehicle (both p’s ≤ 0.026). There were no significant differences among methylone-injected subjects (see Fig. 4).

Fig. 4.

Mean (±SEM) percent of time spent on the drug-paired side (DPS) during the CPP Post-Test for male (top) and female (bottom) subjects injected with vehicle or 5.6, 10 or 18 mg/kg of methylone. *Percent time on the DPS for subjects injected with 10 or 18 mg/kg differed significantly from subjects injected with vehicle.

3.4. CTA/CPP relationship

Analysis of the relationship between the percent of saccharin consumed on the two-bottle CTA test and percent of time spent on the DPS during the CPP Post-Test for all subjects injected with methylone revealed no significant relationship (see Fig. 5 below).

Fig. 5.

Scatterplot (with best line of fit) displaying the relationship between the percent of saccharin consumed on the two-bottle CTA test and percent of time spent on the DPS during the CPP Post-Test. Males and females injected with 5.6, 10 and 18 mg/kg methylone were included in the analysis.

4. General discussion

The combined CTA/CPP procedure has been used to examine concurrently the aversive and rewarding effects of drugs; however, such assessments have been limited to rats. In this context, the present study examined mice in the combined design using the synthetic cathinone methylone. Both male and female mice given access to saccharin, injected with methylone, and then placed on one side of a place preference chamber decreased consumption of the methylone-paired taste and preferred the methylone-associated chamber, i.e., they displayed taste avoidance and place preference, respectively, with the same drug. There was no effect of sex for either measure. Further, there was no significant relationship between these two endpoints (see below). These results demonstrate that the combined design can be used in mice to assess concurrently both the aversive and rewarding effects of a drug in a manner consistent with previous reports in rats.

Although the present work demonstrated that the combined CTA/CPP design can effectively assess both the aversive and rewarding effects of methylone in mice, it is important to compare the results of these findings to other work with methylone in other assessments (albeit in separate studies and under different conditions). For example, Manke et al. (2021) recently demonstrated that methylone at 5.6, 10 and 18 mg/kg induced significant conditioned taste avoidance in rats with such effects comparable in males and females. In this earlier work with rats, 10 and 18 mg/kg methylone also resulted in near complete suppression of saccharin intake. In the present assessment with mice, 5.6 mg/kg resulted in a significant decrease relative to baseline, but consumption by this group never differed from control subjects. Further, at the two higher doses at which animals avoided consumption relative to controls, avoidance did not approach the levels reported with rats. Interestingly, assessments with several other compounds, e.g., ethanol, heroin, and LiCl, have shown similar results in which mice display relatively weaker taste avoidance than rats (for ethanol, see Broadbent et al., 2002; Crabbe et al., 2019; for heroin, see Davis et al., 2009; Freet et al., 2018; for LiCl, see Nachman and Ashe, 1973; Risinger and Cunningham, 2000). The place preference observed with 5.6, 10 and 18 mg/kg of methylone also parallels the findings of other studies examining methylone-induced place preference in mice in independent assessments of its rewarding effects. For example, Karlsson et al. (2014) demonstrated that male C57BL/6 mice injected with 5, 10 and 20 mg/kg of methylone displayed significant conditioned place preference (see also Miyazawa et al., 2011) and like the findings presented here there were no dose-dependent effects as methylone induced comparable preference at all doses (although in this earlier work, preference scores were higher compared to controls).

Although methylone-injected mice displayed CTA and CPP, they were independent of sex. It is important to note that reports on sex differences in taste avoidance and place preference conditioning (in independent assessments) are often drug- and dose-dependent. For example, females generally show stronger CTA with cocaine (van Harren and Hughes, 1990), but for other drugs such as alcohol, reports of sex differences are mixed (for F > M, see Morales et al., 2014; for M > F, see Cailhol and Mormede, 2002; for F = M, see Acevedo et al., 2013; see Riley et al., 2018 for a review). Similarly, females tend to display stronger cocaine-induced CPP (Russo et al., 2003), while cannabinoids (Hempel et al., 2017) and oxycodone (Collins et al., 2016) do not produce differences between males and females. Sex-dependent effects are also mixed with the synthetic cathinones. MDPV, for example, has been reported to induce weaker CTA in female rats, but to produce comparable CPP in males and females (King et al., 2015). Interestingly, α-PVP also produces weaker avoidance in females, but did not condition a place preference in females at any dose (Nelson et al., 2019). While methylone has not been analyzed for sex differences in CPP, it produces comparable CTA in male and female rats (Manke et al., 2021; see above). The underlying mechanisms for such differences have not been explored, although work with sex differences in other designs suggest that drugs of abuse impact males and females differently based on differential susceptibility to tolerance or adaptation to drug effects (see Goldsmith et al., 2019; Marusich et al., 2016), impact of estradiol on general toxicity (Chambers, 1980; Meitzen et al., 2018; although sex differences in body weight and consumption over conditioning were not seen in the present assessments; data not shown) or pharmacokinetic differences in drug distribution or clearance (Hambuchen et al., 2017; McClenahan et al., 2019). While these are certainly possible explanations of differences (or lack thereof) between males and females, these endpoints do not necessarily correlate to the rewarding and aversive effects of drugs.

Although methylone induced taste avoidance and place preference, there were no significant relationships between the two indices (r = 0.1853; p = 0.2230), suggesting that these behavioral effects co-occur and are independent. Work examining the relationship between taste avoidance and place preference is somewhat limited, and the data that have been generated are somewhat mixed. For example, although Turenne et al. (1996) did not report a relationship between morphine-induced taste avoidance and place preference in a serial taste/place conditioning procedure (e.g., CTA was conducted followed by CPP conditioning), they did find a significant positive relationship between amphetamine-induced CTA and CPP at the highest dose administered, i.e., the high dose group that displayed the strongest amphetamine-induced CTA also showed the strongest amphetamine-induced CPP. Verendeev and Riley (2011) using a combined CTA/CPP design also found a relationship with animals conditioned with the highest dose of amphetamine (but not lower doses), although under their conditions the relationship was opposite to that reported by Turenne et al. Specifically, subjects that showed greater decreases in saccharin consumption were less likely to display a place preference. Like Turenne et al., they report no relationship with morphine at any dose tested. In the only study comparing such a relationship for the synthetic cathinones, King et al. (2015) reported a significant inverse relationship for females conditioned with 1.8 mg/kg of MDPV but not at any other dose (and no relationships with males at any dose). Given that these relationships appeared to be dose dependent, correlational analyses were also examined by dose for methylone in the present experiment (data not shown). Interestingly, for 5.6 mg/kg there was an inverse relationship, i.e., as CPP increased, CTAs decreased (r = 0.6324; p = 0.0086), an effect similar to King et al. and Verendeev and Riley. The general effect seen in these relationship assessments across studies is that for the majority of comparisons (roughly 75–80 %), there are no significant correlations, supporting the general conclusion that taste avoidance and place preference are independent, co-occurring stimulus properties of drugs. Significant correlations that are seen (infrequent and often in opposite directions) may reflect chance occurrences that are evident with multiple comparisons. Notably, more work needs to be done to determine such relationships for other compounds and under what conditions significant relationships occur in that it may provide insight into the nature of reward and aversion in drugs of abuse (for a discussion, see Verendeev and Riley, 2011).

5. Implications

The current work supports the use of the combined design to assess the rewarding and aversive effects concurrently in mice which are often the subject of choice in a variety of genetic, neurochemical, and molecular assays on these effects (see above). The findings described above provide a baseline in mice from which future studies may extend these findings to other drugs and to investigate the effects of subject (e.g., sex, age, genetics) and experiential (e.g., drug history) factors (and their balance) that have been previously demonstrated to impact these effects for other drugs of abuse (for reviews of different factors, see Cunningham et al., 2009; Randich and Lolordo, 1979; Riley et al., 2018). Additionally, the utility of the combined procedure is that one can assess both effects concurrently in the same experiment which reduces the number of subjects by 50 % and can ensure that the demonstration of drug-induced CTA and CPP are made under identical parametric conditions (which are often factors in their display in independent assessments). Extending the present analyses to other drugs and manipulations will increase our understanding of how these factors (and their balance) impact abuse liability for other compounds.

The results of the present study support the fact that drugs have multiple stimulus properties, and such effects are co-occurring and independent. Such findings are consistent with prior work with rats (see above), and more importantly, parallel what has been reported in clinical anecdotal reports indicating that drugs have both rewarding and aversive effects in humans (with their balance driving the likelihood of use/abuse; see de Wit and Phillips, 2012 for a comprehensive review). For example, DiFranza et al. (2007) found that in first-time cigarette users, throat irritation with the first puff was a predictor of reduced use whereas relaxation, dizziness, and nausea predicted subsequent cigarette use disorder. Similarly, heroin has been reported to have multiple stimulus properties with initial use resulting in an orgasmic rush often accompanied by nausea, retching, and vomiting (Lasagna et al., 1955). The fact that these properties are displayed in both clinical and preclinical populations support the importance of understanding these effects, their balance and factors affecting their interaction (for excellent analyses of these issues in relation to alcohol use, see Baker and Cannon, 1982; Logue et al., 1981).

6. Limitations

Although the present study supports the use of the combined design in mice and the extension for such an assessment (i.e., application to other drugs, ability to use in the context of pharmacological/genetic manipulations and investigations of subject/experiential factors), the mechanisms mediating the stimulus properties supporting these effects was not examined. It will be important to determine the neurochemical basis for these effects as well as their neuroanatomical substrates using pharmacological (i.e., agonists/antagonists) or genetic manipulations (i.e., KO/KI animals) to understand how reward and aversion are mediated and impacted by factors known to affect drug use and abuse. Investigating underlying mechanisms of these properties is certainly necessary; however, it is important to note that the precise neurochemical and neuroanatomical substrates of such effects is not known, although we and others have speculated a major role for DA in both effects (at least for psychostimulants; for CTA, see Manke et al., 2022, 2023; Serafine and Riley, 2010; for CPP, see Chen et al., 2006; Ritz et al., 1987). Although work has suggested the role of DA, any speculations of the neurochemistry underlying these effects must be made cautiously as few studies have addressed the bases of these properties directly. Instead, they have mostly related the neurochemical actions of the agents used for conditioning to what is observed in CTA and CPP.

While the basis for the dual properties of drugs in general is not well characterized, what is clear is that drugs from a wide range of classes with very different neurochemical actions, e.g., opiates, sedatives, anxiolytics, psychostimulants, all have both stimulus effects, and often do so in the combined design (see above; for a discussion, see Riley et al., 2022). Elucidating the basis of these effects may provide insight into therapeutic targets for a host of behavioral and pharmacological interventions that could be utilized in the prevention and/or treatment of drug use and abuse. Additionally, while the present study examined whether methylone’s effects were sex-dependent, there are a host of other subject (e.g., age, strain) and experiential (e.g., drug history, drug combinations) factors that may impact the affective properties of drugs of abuse and future assessments must extend the present analysis to examine the contribution of other factors to abuse vulnerability, e.g., drug self-administration and its escalation.

Data availability

Data will be made available on request.