Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

Contents lists available at ScienceDirect

Pharmacology, Biochemistry and Behavior journal homepage: www.elsevier.com/locate/pharmbiochembeh

Endocannabinoid influence on partner preference in female rats Nicoletta K. Memos, Rebekah Vela, Courtney Tabone, Fay A. Guarraci ⁎ Department of Psychology, Southwestern University, 1001 East University Ave, Georgetown, TX 78626, USA

a r t i c l e

i n f o

Article history: Received 25 March 2014 Received in revised form 9 July 2014 Accepted 12 July 2014 Available online 18 July 2014 Keywords: Paced mating Sexual motivation THC Anandamide Locomotion Rimonabant AEA SR141716A

a b s t r a c t The present study investigated the role of the endocannabinoid system on sexual motivation in the female rat. In Experiment 1, gonadally intact female rats were first tested for partner preference after a vehicle injection. Approximately 2 weeks later, all rats were tested again after an injection of the endocannabinoid antagonist, SR141716 (SR; also known as Rimonabant; 1.0 mg/kg). During the first 10 min of each partner preference test, subjects could spend time near either a male or female stimulus animal that was placed behind a wire mesh (No-Contact). During the second 10 min of each partner preference test, subjects had unrestricted access to both stimulus animals (Contact). When the female subjects were treated with SR, they made fewer visits to either stimulus animal during the no-contact phase of the partner preference test compared to when they were treated with vehicle. In Experiment 2, ovariectomized (OVX) subjects primed with estrogen were administered SR or vehicle and tested for partner preference (Experiment 2A). Approximately 2 weeks later, the subjects from the control group were tested again after an injection of SR (Experiment 2B). In contrast to Experiment 1, treatment with SR reduced the number of visits specifically to the male stimulus during the contact phase of the test in Experiment 2. Experiment 3 tested the effects of SR on general locomotion and found no effect of SR on line crossings in an open field. Finally, in Experiment 4, OVX estrogen- and progesterone-primed subjects were administered the endocannabinoid agonist anandamide (AEA: 1.0 mg/kg) or vehicle and tested for partner preference. AEA-treated subjects made more visits to the male stimulus than vehicle-treated subjects during the contact phase of the test. The results of the present study suggest that the endocannabinoid system may contribute to sexual motivation in female rats by specifically altering approach behavior. © 2014 Elsevier Inc. All rights reserved.

Marijuana (Cannabis sativa) is the most commonly used illicit drug in the United States, with a majority of users reporting that marijuana was the first illicit drug they had ever used (SAMHSA, 2009, 2011). People report that one of the main reasons for using marijuana is to experience alterations in mood (e.g., euphoria, relaxation, and intensification of ordinary sensory experiences; (Hall et al., 2001)). The main psychoactive ingredient in C. sativa is delta 9-tetrahydrocannabinol (THC), which acts on endogenous CB1 cannabinoid receptors. Of the many endogenous cannabinoids that have been discovered, anandamide (AEA) and 2arachioldonylglycerol ether (2-AG) were the first two endocannabinoids identified and found to act as neuromodulators in the brain (Grotenhermen, 2006). In marijuana users, activation of CB1 receptors has been found to produce various ‘aphrodisiac-like’ effects (Halikas et al., 1982; Koff, 1974). However, these effects follow an inverted-u shaped dose– response on sexual behavior (e.g., performance, arousal, and desire; (Halikas et al., 1982; Koff, 1974)). Specifically, 61% of individuals who smoked approximately one joint reported an increase in sexual desire, whereas individuals who smoked two or more joints reported a

⁎ Corresponding author. Tel.: +1 512 863 1747; fax: +1 512 863 1846. E-mail address: [email protected] (F.A. Guarraci).

http://dx.doi.org/10.1016/j.pbb.2014.07.010 0091-3057/© 2014 Elsevier Inc. All rights reserved.

decrease in sexual desire (Halikas et al., 1982; Koff, 1974). Nevertheless, there is evidence that marijuana affects men and women differently (Halikas et al., 1982; Koff, 1974). For example, after using marijuana, men report increased quality of orgasm, increased duration of intercourse, greater enjoyment of sexual activity, increased perception of partners' satisfaction of sexual activity, and greater attraction towards an unfamiliar partner as compared to when they were not using marijuana (Halikas et al., 1982; Koff, 1974). However, women report fewer of these increases in sexual motivation and if increases are reported, they are reported to a lesser degree (Halikas et al., 1982; Koff, 1974). Furthermore, in women, increases in sexual arousal elicited from viewing erotic stimuli are associated with decreases in endocannabinoid levels (i.e., AEA and 2-AG) (Klein et al., 2012). Taken together, these results suggest that activation of the endocannabinoid system facilitates sexual responses in men, but may inhibit or have limited effects on sexual responses in women. A possible explanation for these gender differences could be an interaction between estrogen and the endocannabinoid system. Estrogen regulates the fatty acid, amide hydrolase (FAAH), which is the enzyme that degrades AEA (Deutsch and Chin, 1993; Hill et al., 2007). Downregulation of FAAH by estrogen increases AEA signaling. Although this connection between estrogen and the endocannabinoid system has not yet been explored in sexual behavior, the regulation of FAAH by estrogen

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

could contribute to the inconsistent effects of cannabinoids on sexual responses in women, due to variable levels of circulating estrogen in women. Similar to the results from human studies, studies using animal models have revealed differences in how endocannabinoids affect male and female sexual behaviors. For example, administration of AEA facilitates sexual behavior in male rats. Specifically, Canseco-Alba and Rodriguez-Manzo (2013) reported that AEA induced sexual behavior (e.g., mounts, intromissions, and successive ejaculatory series) in 50% of the male rats that previously would not copulate; an effect that persisted for at least 14 days after AEA administration (Canseco-Alba and Rodriguez-Manzo, 2013). However, Gorzalka et al. (2008) reported that antagonism of endocannabinoid signaling also facilitated sexual behavior in male rats. Specifically, administration of the CB1 receptor antagonist AM251 produced a dose-dependent facilitation of ejaculation, such that the number of intromissions necessary to achieve ejaculation and ejaculation latency were reduced compared to rats that received vehicle (Gorzalka et al., 2008). These inconsistent results may be a function of the different populations tested in each of these two studies: non-copulating males vs. sexually vigorous males, respectively. Administration of endocannabinoids to female rats has produced varied effects on sexual behavior. Mani et al. (2001) concluded that THC facilitated lordosis responses in ovariectomized (OVX), hormoneprimed (estradiol benzoate [EB] and progesterone [P]) rats. Furthermore, intracerebroventricular infusions of the CB1 antagonist/inverse agonist, SR141716 (SR), inhibited progesterone-, dopamine- and THCfacilitated sexual receptivity (i.e., lordosis responses) in female rats (Mani et al., 2001). In contrast, Lopez et al. (2009) found that treatment with the endocannabinoid antagonist/inverse agonist, AM251, significantly enhanced sexual motivation, as measured by increases in lordosis ratings and proceptive behaviors (e.g., hops and darts) in EB- and P-primed OVX rats. These contradictory results prompted Zavatti et al. (2011) to examine the effects of SR on sexual motivation, receptivity, and proceptivity in female rats using the partner preference test. In their study, female rats treated with SR displayed reduced interest in both social (female) and sexual (male) stimulus animals when subjects were allowed to spend time in the vicinity of the stimulus animals confined to an area in an open field (Zavatti et al., 2011). However, SR treatment also decreased lordosis responses during a test for sexual receptivity (Zavatti et al., 2011). Although Zavatti et al. (2011) measured some aspects of sexual motivation in a complex paradigm (i.e., partner preference), the authors did not measure all of the typical sexual behaviors displayed by the female rat. Specifically, they did not measure paced mating behavior. Although Lopez et al. (2009) tested female rats in a paced mating test, they did not record measures of paced mating behavior (e.g., latency to return to the male after receiving sexual stimulation and likelihood of leaving the male after the receipt of sexual stimulation). Furthermore, alterations in sexual receptivity confound any interpretation of alterations in the partner preference paradigm, because animals that are not sexually receptive typically do not prefer a sexual partner (Clark et al., 2004). Therefore, the present study was designed to investigate the acute influence of the endocannabinoid antagonist, SR141716 (1 mg/kg), as well as the agonist AEA (1 mg/kg) on a full range of sexual behaviors in the female rat. The partner preference paradigm used in the present study allowed us to test sexual motivation with and without physical contacts, as well as measure all typical patterns of sexual behavior in the female rat (i.e., lordosis, proceptive behaviors) and the active pacing of sexual contact by the female (i.e., paced mating behavior). 1. Method 1.1. Subjects Fifty-two female Long-Evans rats (Rattus norvegicus; 200–400 g) were used as experimental subjects (Experiment 1: n = 11 sexually

381

naïve; Experiment 2: n = 15 sexually naïve; Experiment 3: n = 14 sexually experienced; Experiment 4: n = 12 sexually experienced). Sexually experienced female Long-Evans rats (200–400 g), as well as, sexually experienced male Long-Evans rats (400–600 g) were used as stimulus animals. All rats were purchased from Harlan Sprague–Dawley (Indianapolis, IN). Rats were housed in plastic hanging cages with Aspen wood shavings used for bedding. Food and water were available ad libitum. Female rats were housed three to a cage and male rats were housed two to a cage. All rats were weighed weekly. Temperature and humidity in the animal colony were monitored daily. The lights in the colony were maintained on a reversed 12:12 h light–dark cycle (with lights off at 10:00 a.m.). All of the behavioral procedures took place during the dark phase of the cycle, under dim red light. At least one week before any tests took place, the female subjects in Experiments 2, 3 and 4 were bilaterally ovariectomized (OVX) under Nembutal (sodium pentobarbital; 50.0 mg/kg i.p.; Henry Schein, Indianapolis, IN) anesthesia after pretreatment with atropine sulfate (2.5 mg), which reduces respiratory distress. 1.2. Hormones & drugs For Experiments 1–3, the experimental female rats received either an intraperotineal (i.p.) injection of vehicle or SR (1.0 mg/kg). SR141716 was dissolved in a vehicle of 0.9% saline, TWEEN-80, and DMSO. A stock solution of 5.0 mg of SR in 500 μl DMSO and 500 μl of TWEEN-80 was sonicated for 30 min producing 5.0 mg/ml of SR. Next, the solution was diluted to 1.0 mg/ml by adding 250 μl of stock to 4.75 ml saline, producing a ratio of 1:1:8 of vehicle solutions (DMSO: TWEEN-80:Saline). For Experiment 4, the female subjects received either an injection of vehicle or AEA (1.0 mg/kg i.p.). AEA was dissolved in the same vehicle, following identical procedures used for SR. All drug injections were made 20 min prior to partner preference tests. These doses of SR and AEA were used previously and found to affect sexual behavior in rats (Canseco-Alba and Rodriguez-Manzo, 2013; Zavatti et al., 2011). In Experiments 2 and 3, female subjects received 2.0 μg of estradiol benzoate (EB) daily at 1 p.m. for 6 days with the last dose administered 24 h prior to each behavioral test. All of the female stimulus animals and the female subjects in Experiment 4 received 10.0 μg of EB 48 h and 1.0 mg of progesterone (P) 4 h prior to each mating test. All hormone injections were administered subcutaneously in the flank. Each hormone was delivered in a sesame seed oil vehicle. The doses of EB and/or P used in the present study produce high levels of sexual receptivity in OVX rats (Blasberg and Clark, 1997). Hormones, atropine sulfate and vehicle solvents were purchased from the Sigma-Aldrich Chemical Company (St. Louis, MO). SR141716 and AEA were purchased from Tocris Bioscience (Minneapolis, MN). 1.3. Estrous cyclicity The female subjects in Experiment 1 were monitored for one month using vaginal cytology to ensure normal estrous cyclicity. Vaginal cytology was examined once daily at 9:00 a.m., by collecting vaginal secretions using a sterile plastic pipette filled with saline (Marcondes et al., 2002). Vaginal fluid was placed onto glass slides and examined under a microscope. Female rats were recorded as being in proestrus, estrus, metestrus, or diestrus based on the proportion of cell types. Proestrus vaginal secretions consisted mainly of nucleated epithelial cells; estrus secretions consisted mainly of cornified, non-nucleated cells; metestrus secretions consisted of equal proportions of round leukocytes, cornified, and nucleated epithelial cells; and diestrus secretions consisted mostly of round leukocytes (Marcondes et al., 2002). After approximately one month of monitoring, the female subjects were tested for partner preference in the afternoon (~ 1:00 p.m.) of behavioral estrus. (i.e., they were in proestrus in the morning based on vaginal secretions) (Zipse et al., 2000). Immediately before drug

382

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

treatment (vehicle or SR) was administered, we confirmed that female subjects were sexually receptive by screening each subject with a male stimulus until at least one mount occurred and the subject displayed the lordosis posture. On occasion, female subjects were not receptive even though vaginal secretions observed in the morning indicated that they would be in behavioral estrus in the afternoon. If the female subject was not receptive, she was returned to her home cage and estrous cyclicity was monitored daily until she was in proestrus again. To ensure that no female subjects became pregnant during any mating test, 50.0 μg of EB was administered immediately after mating tests (Snoeren et al., 2011). Normal estrous cyclicity resumed within approximately two weeks. 1.4. Behavioral test procedure 1.4.1. Apparatus and acclimation All rats were acclimated to the mating chambers on two separate occasions for 15 min each, prior to any mating tests. Each mating chamber consisted of a Plexiglas arena (101.0 cm long × 32.0 cm high × 37.0 cm wide) divided into three equal compartments using clear Plexiglas dividers. Each of the dividers had a 5.0 cm hole in each of the two bottom corners. Wood shavings covered the floor of each compartment. During acclimation sessions for the stimulus animals, a single stimulus animal was placed in each of the outer compartments and was tapped lightly on the nose if he or she attempted to exit through the holes in the partition. During acclimation sessions for the female subjects, a single female rat was placed alone in the arena and allowed to move freely between the three compartments. 1.4.2. Partner preference test Approximately one week after acclimation, female subjects were given an opportunity to spend time with either a male or a female stimulus animal in a test for partner preference. Prior to the start of each mating test, a female subject was placed into the center compartment of the mating chamber, confined with solid opaque dividers, and allowed to acclimate for 5 min. The dividers prevented the female subject from entering either of the two side compartments, each of which held one male stimulus animal or one female stimulus animal. The location of each stimulus animal (on the left or right) was randomly assigned. During the first 10 min of the partner preference test (Phase 1: No-Contact), wire mesh partitions (37 cm wide × 32 cm high) further divided the side compartments of the chamber in half (16.8 cm in the outer chamber). The wire mesh partitions allowed for the transmission of visual, auditory, and olfactory cues, but prohibited physical contact (No-Contact). The No-Contact phase of the test began when both opaque partitions were removed, allowing the female subject to come into the vicinity of either stimulus animal through the clear Plexiglas partitions. After 10 min, both opaque partitions were replaced, confining the female subject to the center compartment. At that point, the wire mesh partitions were removed. The Contact phase of the test began when both opaque dividers were again removed, allowing the female subject unrestricted access to either stimulus animal through the clear Plexiglas partitions. Finally, after 10 min, the opaque dividers were once again replaced, and all rats were returned to their home cages. Between each successive partner preference test, the mating chamber and the dividers (i.e., mesh dividers, clear dividers, opaque dividers) were cleaned with Windex and fresh bedding was added to the mating chamber. If a female subject did not receive any intromissions during the Contact phase of the test, a test for sexual receptivity was performed. The female subject was placed into a chamber (36 cm long × 32.0 cm high × 37.0 cm wide) with a male stimulus animal. The sexual receptivity test ended when the female subject received 10 sexual stimulations (e.g., mounts, intromissions, ejaculations) from the male stimulus animal. Stimulus animals were always unfamiliar to female subjects.

Behaviors observed throughout the No-Contact phase of the test included the number and timing of exits and entries into each compartment, as well as proceptive behaviors (i.e., hops, darts, ear wiggles). Entering a compartment was scored when all four paws of the female subject passed through the holes in the clear divider and into the adjacent compartment. The time spent in each compartment and the percentage of time the female subject spent near each of the stimulus animals were calculated. Additional behaviors were observed throughout the Contact phase of the test including the type and timing of sexual stimulations (i.e., mounts, intromissions, ejaculations) and sexual receptivity, as measured by lordosis responses (LR) to each sexual stimulation. Each LR was rated on a 4-point scale (0–3), with 0 being no lordosis and scores of 2 or 3 considered indications of sexual receptivity (Hardy and DeBold, 1972). A number of measures were calculated based on these behaviors. The lordosis quotient (LQ) was calculated as the percentage of lordosis responses of 2 or 3 that the female subject displayed in response to each sexual stimulation (Hardy and DeBold, 1972). The time spent in each compartment and percentages of time spent with each of the stimulus animals were calculated. Percentage of exits (PE) was calculated as the likelihood that the female subject left the male's compartment following the receipt of each type of sexual stimulation. Contact-return latency was calculated as the time (in seconds) that elapsed between receiving a type of sexual stimulation, leaving the male's compartment, and re-entering the male's compartment. All mating tests were recorded with digital video cameras (Sony DCRHC65) for posttest analysis of behaviors by observers who were blind to the assignment of drug condition for each subject. 1.5. Experiment 1: SR and partner preference test in naturally cycling female rats 1.5.1. Test 1: vehicle partner preference test Gonadally intact, female subjects were monitored for estrous cyclicity for approximately 4 weeks. In the afternoon of estrus, each sexually naïve female subject was first screened for sexual receptivity to confirm behavioral estrus. Any female subject in behavioral estrus was administered a vehicle injection and tested for partner preference 20 min later. 1.5.2. Test 2: SR partner preference test After estrous cyclicity resumed (approximately 2 weeks after Test 1), any female subject in behavioral estrus was administered SR, 20 min prior to a second Partner Preference test. Female subjects were tested with stimulus animals that they had not been tested with previously. 1.6. Experiment 2: SR and partner preference test in EB-primed OVX rats In Experiment 2A, sexually naïve, OVX, hormone-primed (EB for 6 days), subjects were randomly assigned to receive either an injection of vehicle or SR. Twenty minutes after injections, female subjects were tested for partner preference (No-Contact followed by Contact). Because SR produced more robust effects on motivation in OVX hormone-primed subjects (Experiment 2A) than in naturally cycling subjects (Experiment 1), the vehicle-treated subjects of Experiment 2A were tested again, but this time the subjects were given SR 20 min prior to a second test of partner preference for Experiment 2B. Because this protocol is similar to Experiment 1, differences in results between Experiment 1 and Experiment 2B could be attributed to differences in hormonal condition (naturally cycling vs. OVX + EB, respectively). 1.7. Experiment 3: SR and open field test in EB-primed OVX rats To confirm that the effects of SR in Experiments 1 and 2 were not merely a consequence of alterations in locomotor behavior, OVX hormone-primed (EB 6 days) subjects were administered SR or vehicle, 20 min prior to an Open Field Test. Line crossings were recorded during

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

a 10 min-period in a clear Plexiglas arena (101.0 cm × 3 7.0 cm wide × 3 32.0 cm high) with lines marking the floor of the arena every 5.0 cm. Line crossings were counted when all four legs of the subject crossed any line. 1.8. Experiment 4: AEA and partner preference test in EB and P-primed OVX rats To evaluate the effect of activating the endocannabinoid system, OVX, sexually experienced rats were administered AEA or vehicle, 20 min prior to a partner preference test (No-Contact followed Contact). Hormone priming was used to more closely mimic naturally cycling behavioral estrus (EB + P), but at supraphysiological levels. 1.9. Data analysis 1.9.1. Experiment 1 and 2B For the No-Contact phase of the Partner Preference Test, a 2 × 2 repeated measures analysis of variance (ANOVA) was calculated to examine the within subjects effect of drug treatment (Vehicle vs. SR) on female behaviors (e.g., time spent and visits made) directed towards the stimulus animals (male vs. female). For the Contact phase of the Partner Preference Test, a separate identical 2 × 2 repeated measures ANOVA was calculated to examine the effect of drug treatment (Vehicle vs. SR) on female behaviors (e.g., time spent and visits made) directed towards the stimulus animals (male vs. female). Additional mating behaviors were analyzed with paired t-tests to examine the effect of treatment (Vehicle vs. SR) on LR, LQ, proceptive behaviors (hops and ear wiggles), percentage of exits, and contact-return latency. 1.9.2. Experiment 2A and 4 For the No-Contact phase of the Partner Preference Test, a 2 × 2 mixed design repeated measures ANOVA was calculated to examine the between subjects effect of drug treatment (Vehicle vs. SR) on female

383

behaviors (e.g., time spent and visits made) directed towards the stimulus animals (male vs. female). For the Contact phase of the Partner Preference Test, a separate, identical 2 × 2 mixed design repeated measures ANOVA was calculated to examine the between subjects effect of drug treatment (Vehicle vs. SR) on female behaviors (e.g., time spent and visits made) directed towards the stimulus animals (male vs. female). Additional mating behaviors were analyzed with independent t-tests to examine the effect of treatment (Vehicle vs. SR) on LR, LQ, proceptive behaviors (hops and ear wiggles), percentage of exits, and contact-return latency. All significant interactions between drug treatment and sex of the stimulus animals were followed up with Tukey's post-hoc tests. The alpha level was set at p ≤ .05. 2. Results 2.1. Experiment 1: SR and partner preference test in naturally cycling female rats 2.1.1. No-contact A 2 × 2 repeated measures ANOVA with two within subjects factors (sex of the stimulus animal [male vs. female] and drug treatment [vehicle vs. SR]) was calculated for each dependent variable (i.e., time spent near the stimulus animals and visits made in the vicinity of the stimulus animals). Ten of the 11 female subjects were sexually receptive during both tests (Vehicle and SR tests). The data from only receptive female subjects were included in the statistical analysis. There was a significant main effect of sex on time spent with the stimulus animals (F(1,9) = 75.33, p = .0001). Independent of drug treatment, the female subjects from both treatment groups spent more time with the male stimulus than the female stimulus. However, there was no significant main effect of drug treatment or interaction between sex and drug treatment on time spent with the stimulus animals. See Fig. 1 top left. Similarly, there was a significant main effect of sex on visits made to the stimulus animals (F(1,9) = 20.30, p = .001). Independent of drug

Fig. 1. During the tests for partner preference, naturally cycling female rats (when treated with vehicle or with SR141716: 1.0 mg/kg; N = 10) spent more time with the male stimulus than the female stimulus when physical contact was limited (No-Contact) (top left) ##. However, when treated with SR, the female rats made fewer visits to either stimulus animal (bottom left) #. When physical contact was not limited (Contact), female rats, independent of drug treatment, spent more time with the male stimulus (top right) ## and visited the male stimulus more frequently (bottom right) ## than the female stimulus. For all figures, data are expressed as means ± S.E.M. Note: n's = number of rats in each group. N's = total number of rats in each experiment. A # indicates a significant main effect of drug treatment (i.e., vehicle vs. SR), independent of sex of the stimulus. A ## indicates a significant main effect of sex of stimulus animal (male vs. female), independent of drug treatment. An * indicates a significant post hoc comparison between drug treatments (vehicle vs. SR) for one stimulus. A ** indicates a significant post-hoc comparison between sex of the stimulus (male vs. female) for one drug treatment. Alpha level is set at p ≤ .05.

384

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

treatment, female subjects visited the male stimulus more frequently than they visited the female stimulus. However, there was also a significant main effect of drug treatment on visits made to the stimulus animals (F(1,9) = 6.0, p = .04). Specifically, female subjects treated with SR visited either stimulus less frequently than female subjects treated with vehicle. Furthermore, there was no significant interaction between sex and drug treatment on visits made to the stimulus animals. See Fig. 1 bottom left. 2.1.2. Contact There was a significant main effect of sex (F(1,9) = 16.02, p = .004) on time spent with the stimulus animals. Specifically, female rats treated with either SR or vehicle spent a greater amount of time with the male stimulus than with the female stimulus. However, there was no main effect of drug treatment on time spent with the stimulus animals nor was there a significant interaction between sex and drug treatment on time spent with the stimulus animals. See Fig. 1 top right. Similarly, there was a significant main effect of sex on visits made to the stimulus animals (F(1,9) = 12.57, p = .008). Independent of drug treatment, female subjects visited the male stimulus more frequently than they visited the female stimulus. There was no significant main effect of drug treatment on visits made to the stimulus animals, nor was there a significant interaction between sex of the stimulus and drug treatment on visits made to the stimulus animals. See Fig. 1 bottom left. Paired t-tests were calculated comparing drug treatment (SR vs. Vehicle) on all other mating behaviors: LQ, proceptive behaviors (hops, ear wiggles), contact-return latency (for mounts, intromissions, ejaculations), and percentage of exits (for mounts, intromissions, ejaculations). There were no significant effects of drug treatment on any of the other mating behaviors (data not shown). 2.2. Experiment 2A: SR and partner preference test in EB-primed OVX rats 2.2.1. No-contact A 2 × 2 mixed design repeated measures ANOVA with one within subjects factor (sex of the stimulus animal: male vs. female) and one between subjects factor (drug treatment: Vehicle vs. SR) was calculated

for each dependent variable (i.e., time spent near the stimulus animals and visits made in the vicinity of the stimulus animals). Eleven of the 15 female subjects were sexually receptive during the test (Vehicle: n = 5: SR: n = 6). Two subjects from each group were not sexually receptive during the test. The data from only receptive females were included in the statistical analysis. There was a significant main effect of sex on time spent with the stimulus animals (F(1,9) = 6.05, p = .036). Independent of drug treatment, the female subjects from both treatment groups spent more time with the male stimulus than the female stimulus. However, there was no significant main effect of drug treatment or interaction between sex and drug treatment on time spent with the stimulus animals. See Fig. 2 top left. In contrast, there was a significant main effect of drug treatment on visits made to the stimulus animals (F(1,9) = 5.51, p = .043). Thus, independent of the sex of the stimulus, the female subjects treated with SR made significantly fewer visits to either stimulus animal than the female subjects treated with vehicle. There was no significant main effect of sex or interaction between sex and drug treatment on visits made to the stimulus animals. See Fig. 2 bottom left.

2.2.2. Contact There was no significant main effect of sex or interaction between sex and drug treatment on time spent with the stimulus animals. Nor was there a significant main effect of drug treatment on time spent with the stimulus animals. See Fig. 2 top right. However, there was a significant main effect of sex (F(1,9) = 18.70, p = .002) and a significant main effect of drug treatment (F(1,9) = 7.66, p = .02) on visits made to the stimulus animals. But most importantly, there was a significant interaction between sex and drug treatment on visits made to the stimulus animals (F(1,9) = 14.73, p = .004). Specifically, female subjects treated with vehicle visited the male stimulus more frequently than they visited the female stimulus (Tukey's Q = 11.59, p b .05), whereas the SR-treated subjects visited the male and female stimulus animals at about the same rate. Furthermore, female subjects treated with vehicle made more visits to

Fig. 2. Both groups of OVX EB-primed rats (vehicle n = 5; SR141716 1.0 mg/kg n = 6) spent more time with the male stimulus than the female stimulus when physical contact was limited (No-Contact) during the partner preference test (top left) ##. However, SR treated rats made fewer visits to either stimulus animal (bottom left) #. When physical contact was not limited (Contact), neither group spent significantly more time with the male stimulus than the female stimulus (top right). However, the vehicle treated female rats made more visits to the male stimulus than to the female stimulus **, whereas the SR-treated rats visited the male and female stimulus animals at about the same rate. Furthermore, the SR treated female rats made significantly fewer visits to the male stimulus than the vehicle treated rats (bottom right) *.

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

the male stimulus than female subjects treated with SR (Tukey's Q = 11.66, p b .05). See Fig. 2 bottom right. Independent t-tests were calculated comparing drug treatment (SR vs. Vehicle) as the independent factor on all other mating behaviors: LQ, proceptive behaviors (hops, ear wiggles), contact-return latency (for mounts, intromissions, ejaculations), and percentage of exits (for mounts, intromissions, ejaculations). There were no significant effects of drug treatment on any mating behaviors (data not shown). During this experiment, most of the SR-treated subjects (5 of the 6) received no intromissions or ejaculations. Nevertheless, they were all receptive (LQ N 75%) during the sexual receptivity test immediately after the partner preference test (LQ: M = 87.5%, SEM = 4.79%). 2.3. Experiment 2B: SR and partner preference test in EB-primed OVX within subjects 2.3.1. No-contact A 2 × 2 repeated measures ANOVA with two within subjects factors (sex of the stimulus animal [male vs. female] and drug treatment [Vehicle vs. SR]) was calculated for each dependent variable (i.e., time spent near the stimulus animals and visits made in the vicinity of the stimulus animals). All female subjects were sexually receptive and included in all analyses. There was no significant main effect of sex on time spent with the stimulus animals. There was no significant main effect of drug treatment or interaction between sex and drug treatment on time spent with the stimulus animals. See Fig. 3 top left. There was a significant main effect of sex on visits made to the stimulus animals (F(1,4) = 14.24, p = .02). Independent of drug treatment, female subjects visited the male stimulus more frequently than they visited the female stimulus. There was no significant main effect of drug treatment or interaction between sex and drug treatment on visits made to the stimulus animals. See Fig. 3 bottom left.

385

2.3.2. Contact There was no significant main effect of sex or interaction between sex and drug treatment on time spent with the stimulus animals. Nor was there a main effect of drug treatment on time spent with the stimulus animals. See Fig. 3 top right. However, there was a significant main effect of sex (F(1,4) = 27.00, p = .007) and a significant main effect of drug treatment (F(1,4) = 14.69, p = .02) on visits made to the stimulus animals. But most importantly, there was a significant interaction between sex and drug treatment on visits made to the stimulus animals (F(1,4) = 7.60, p = .05). Specifically, when the female subjects were treated with vehicle, they visited the male stimulus more frequently than they visited the female stimulus (Tukey's Q = 7.71, p b .05). However, when they were then treated with SR, they visited the male and female stimulus animals at about the same rate, which was less than when they were treated with vehicle (Tukey's Q = 6.10, p b .05). See Fig. 3 bottom left. Paired t-tests were calculated comparing drug treatment (SR vs. Vehicle) on all other mating behaviors: LQ, proceptive behaviors (hops, ear wiggles), contact-return latency (for mounts, intromissions, ejaculations), and percentage of exits (for mounts, intromissions, ejaculations). There were no significant effects of drug treatment on any of the other mating behaviors (data not shown). During this experiment, all of the female subjects received intromissions during both tests, except for one rat after she received SR. Nevertheless, during the test for sexual receptivity, she was receptive with an LQ of 90%. 2.4. Experiment 3: SR and open field test in EB-primed OVX rats There was a significant main effect of consecutive, 2-min bins on the number of lines crossed during the open field test (F(4,52) = 12.89, p = .001). Independent of drug treatment, female subjects made fewer line crossings as the test progressed. Furthermore, there was no

Fig. 3. Ovariectomized EB-primed female rats treated first with vehicle spent slightly more time with the male stimulus than the female stimulus, which was not significantly different from when they were treated with the endocannabinoid antagonist, SR141716 (1.0 mg/kg), during the No-Contact phase of the partner preference test (top left; N = 5). Furthermore, SR treatment had no effect on visits to the either stimulus animal (bottom left). When physical contact was not limited (Contact), SR treatment had no effect on time spent with either stimulus animal (top right). However, when female rats were treated with vehicle, they made more visits to the male stimulus than to the female stimulus **, whereas when female rats were treated with SR, they visited the male and female stimulus animals at about the same rate. Furthermore, when female rats were treated with SR, they made significantly fewer visits to the male stimulus than when treated with vehicle (bottom right)*.

386

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

significant main effect of drug treatment (Mean Totals ± SEM Vehicle: 187.86 ± 23.60 and SR: 147.27 ± 20.57) or interaction between drug treatment and consecutive, 2-min bins on line crossings during the open field test. 2.5. Experiment 4: AEA and partner preference test in EB and P-primed OVX rats 2.5.1. No-contact A 2 × 2 mixed design repeated measures ANOVA with one within subjects factor (sex of the stimulus animal: male vs. female) and one between subjects factor (drug treatment: Vehicle vs. AEA) was calculated for each dependent variable (i.e., time spent near the stimulus animals and visits made in the vicinity of the stimulus animals). Ten of the 12 female subjects were sexually receptive during the partner preference tests (Vehicle: n = 6; AEA: n = 4). Two subjects from the AEA group were not sexually receptive during the test. The data from only receptive female subjects were included in the statistical analysis. There was a significant main effect of sex on time spent with the stimulus animals (F(1,8) = 47.50, p = .001). Independent of drug treatment, the female subjects from both treatment groups spent more time with the male than the female stimulus animal. There was no significant main effect of drug treatment or significant interaction between sex and drug treatment on time spent with the stimulus animals. See Fig. 4 top left. Similarly, there was a significant main effect of sex on visits made to the stimulus animals (F(1,8) = 22.06, p = .002). Independent of drug treatment, the female subjects from both treatment groups visited the male stimulus more frequently then they visited the female stimulus. There was no significant main effect of drug treatment or significant interaction between sex and drug treatment on visits made to the stimulus animals. See Fig. 4 bottom left. 2.5.2. Contact There was a significant main effect of sex (F(1,8) = 44.47, p = .001) and a significant main effect of drug treatment (F(1,8) = 6.445,

p = .035) on time spent with the stimulus animals. But most importantly, there was a significant interaction between sex and drug treatment on time spent with the stimulus animals (F(1,8) = 7.154, p = .028). The female subjects treated with AEA spent significantly less time with the male stimulus than the female subjects treated with the vehicle (Tukey's Q = 4.76, p b .05). In contrast, both groups spent a similar amount of time with the female stimulus. See Fig. 4 top right. There was a significant main effect of sex (F(1,8) = 101.49, p =.001) and a significant main effect of drug treatment (F(1,8) = 6.88, p = .030) on visits to the stimulus animals. But most importantly, there was a significant interaction between sex and drug treatment on visits to the stimulus animals (F(1,8) = 21.90, p = .002). Specifically, female subjects treated with AEA made significantly more visits to the male stimulus than female subjects treated with the vehicle (Tukey's Q = 13.42, p b .05). In contrast, the female subjects from both treatment groups made a similar number of visits to the female stimulus. See Fig. 4 bottom right. Independent t-tests were calculated comparing drug treatment (AEA vs. Vehicle) as the independent factor on all other mating behaviors: LQ, proceptive behaviors (hops, ear wiggles), contact-return latency (for mounts, intromissions, ejaculations), and percentage of exits (for mounts, intromissions, ejaculations). There were no significant effects of drug treatment on any mating behaviors (data not shown).

3. Discussion In the present study, we found that acute administration of the endocannabinoid antagonist, SR141716 (Rimonabant), 20 min prior to a test for partner preference disrupted social motivation (i.e., the motivation to approach either a sexual or a social conspecific). For example, female rats treated with SR visited the stimulus animals less frequently in Experiments 1 and 2 when contact was limited between the stimulus animals and female subjects (i.e., during the No-Contact phase). This reduction in general social motivation when physical contact was limited is consistent with observations made by Zavatti et al. (2011). In their

Fig. 4. Both groups of OVX EB + P-primed female rats (vehicle n = 6; and anandamide (AEA) 1.0 mg/kg n = 4) spent more time with the male stimulus than the female stimulus (No-Contact; top left) ##. Both groups visited the male stimulus more frequently than the female stimulus (bottom left) ##. Although both groups spent more time with the male stimulus than the female stimulus when physical contact was not limited (Contact) ##, AEA treated female rats spent less time with the male stimulus than the vehicle treated rats (top right) *. Although both groups made more visits to the male stimulus than the female stimulus ##, AEA treated female rats made significantly more visits to the male stimulus than the vehicle treated rats (bottom right) *.

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

experiment, female rats treated with SR spent less time near either stimulus animal (one male and one female rat) when the stimulus animals were placed inside wire cages on either side of an open field. A more selective disruption of sexual motivation was observed in the present study when contact between the female subjects and the stimulus animals was unrestricted and mating was possible (i.e., during the Contact phase). For example, female subjects treated with SR made significantly fewer visits to the male stimulus animal when compared to vehicle-treated subjects, whereas, SR treatment did not alter visits to the female stimulus animal when mating was possible. Unlike the disruption in social preference, which was observed in Experiments 1 and 2, the disruption of motivation to mate with the male (as measured by the number of visits to the male stimulus), was only observed in OVX EB-primed subjects and not evident when subjects were naturally cycling and tested during behavioral estrus. Because the disruption was observed in sexually naïve OVX EB-primed subjects (Experiment 2A) and sexually experienced OVX-EB-primed subjects (Experiment 2B), the protection from disruptions in sexual motivation in naturally cycling subjects is likely due to differences in hormonal state and not a consequence of our experimental procedure (e.g., within subjects test of Vehicle first followed by the test of SR, or sexual experience). Because Zavatti et al. (2011) did not measure mating behavior during their partner preference test, nor did they evaluate different hormone conditions, the results of the present study are new and add to what has been previously reported (Zavatti et al., 2011). The results of Experiment 3 confirm that the effects of SR on visits to both stimulus animals in Experiments 1 and 2 when contact was limited (i.e., No-Contact phase) are not likely the consequence of disruptions in general locomotion. Across a 10 min test in an open field, OVX EB-primed subjects treated with SR made a similar number of line crossings as female subjects treated with vehicle. Given that Zavatti et al. (2011) failed to find any effect of a similar dose of SR on general locomotor behavior in their OVX EB + P-primed subjects, and others have found that only relatively high doses of SR (e.g., 5.6 mg/kg) suppress locomotion (Jarbe et al., 2006), we feel it is unlikely that the effects of SR on partner preference are due to reduced general locomotion. Furthermore, we found specific effects on visits to only one stimulus animal (i.e., the male), but not to the female stimulus animal in Experiments 2 and 4, which also suggests that the low dose of endocannabinoid drugs tested in the present study do not merely alter general locomotor activity. To confirm the specific role of the endocannabinoid system, we tested the effects of the endocannabinoid agonist AEA on partner preference. Consistent with what the results of Experiments 1 and 2 would predict, activation of the endocannabinoid system produced a selective increase in sexual motivation during the phase of the partner preference test that permitted mating. Specifically, female subjects treated with AEA visited the male stimulus animal more frequently than the vehicletreated subjects. However, unlike the results from Experiments 1 and 2, AEA did not produce a general increase in social behavior as measured by an increase in visits to both stimulus animals during the phase of the partner preference test with limited physical access (i.e., the No-Contact phase). Unexpectedly, AEA-treated subjects spent significantly less time with the male stimulus animal when compared to vehicle-treated subjects. One possible explanation for the decrease in time spent with the male stimulus animal is that the increased number of visits to the male artificially decreased total time spent with the male. Visits and time spent with the female stimulus animal were unaffected by AEA, suggesting that the effect of AEA is specific. Unlike Zavatti et al. (2011), we did not find an effect of SR on lordosis responses in any of our experiments. Although OVX EB-primed, SR-treated subjects failed to receive many sexual stimulations with intromissions during the contact phase of the partner preference test, we were able to confirm that they were sexually receptive when female subjects were confined with a male stimulus animal in a small chamber

387

and we measured lordosis behavior immediately after the partner preference test. Therefore, we can conclude that the effects of SR on sexual motivation during the partner preference test are not artifacts of alterations of lordosis. The partner preference test is not a valid measure of sexual motivation when animals are not sexually receptive (Clark et al., 2004), because you cannot separate the effects on lordosis from the effects on sexual motivation when lordosis is also affected. Unfortunately, because the OVX EB-primed, SR-treated subjects did not receive many intromissions during the partner preference test, we were unable to thoroughly assess whether SR disrupted measures of paced mating behavior in Experiment 2. Tests with much longer time limits may be needed to assess the likely disruptions of SR on measures of paced mating behavior, such as contact-return latencies and percentage of exits, following intromissions and ejaculations in OVX EB-primed subjects. We are unsure why there are so many conflicting reports of the effects of antagonists of the endocannabinoid system on lordosis. For example, Zavatti et al. (2011) found disruptive effects on LQ following administration of SR, whereas Lopez et al. (2009) found slight enhancements in LQ following administration of AM251. Yet, we found no robust alterations in LQ in any of our experiments following administration of SR or AEA. It is possible that one factor that contributed to these different results is the conditions under which sexual receptivity was assessed. Although Zavatti et al., used a non-paced mating protocol, their chamber was twice as large as our chamber; therefore, the female subjects could run away from the male even though they could not escape. In our smaller chambers, it may have been easier for a male stimulus animal to catch and mount a female subject, ensuring proper positioning for intromission. When we tested partner preference using a protocol that allowed for mating and for escape, which is similar to paced mating paradigm used by Lopez and colleagues, OVX EB-primed female subjects treated with SR escaped into their protected compartment so frequently that the males were unsuccessful at achieving any intromissions or ejaculations. Interestingly, when the male stimulus animals and female subjects were confined to the small chamber without an escape, more typical of what is used to assess sexual receptivity, we observed robust lordosis responses. In addition, we used different hormone conditions in our studies when compared to Lopez and colleagues and Zavatti and colleagues. Specifically, we tested naturally cycling subjects during behavioral estrus, as well as, an EB alone condition. It is possible that SR interacted differently with these different types and levels of hormones. Furthermore, pharmacological differences between SR and AM251 (McMahon and Koek, 2007) could be responsible for the striking differences between the effects produced by these two antagonists in previous studies. Because we found that sexual motivation was specifically disrupted when SR was administered to OVX subjects primed with EB alone, it is possible that ovarian hormones may protect against the effects of SR on sexual motivation but not on social motivation. Our results are consistent with the findings of Mani et al. (2001), who found that THC only facilitates lordosis when subjects were OVX and given EB + P. This facilitation of lordosis by THC was greatly suppressed when subjects were OVX and given EB alone. Furthermore, when they blocked progestin receptors in OVX EB + P-primed-subjects, THC-facilitated lordosis was also greatly suppressed. Previous findings confirm the presence of progestin receptors (Parsons et al., 1982) and CB1 receptors (Mailleux and Vanderhaeghen, 1992) in a number of areas of the brain, including the hypothalamus. Furthermore, progesterone has been shown to increase the density of cannabinoid receptors in the hypothalamus of OVX rats primed with E2 (Rodriguez de Fonseca et al., 1993). Although the present study did not test the role of progesterone specifically, previous results suggest that there is crosstalk between endocannabinoid signaling and progesterone signaling. Progesterone priming and natural levels of progesterone in intact subjects may protect against the effects of low doses of SR by increasing the density of CB1 receptors, thereby minimizing the effects of SR. Interestingly, because SR disrupted social

388

N.K. Memos et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 380–388

motivation in female subjects with and without circulating progesterone, it suggests that the neural mechanism underlying SR effects on social motivation is hormone independent and distinct from the disruption of sexual motivation by SR. However, future studies investigating the effects of progestin antagonists are needed to confirm this protective role of progesterone. Taken together the results of the present study add to our understanding of motivation and the endocannabinoid system. This is the first study to use a comprehensive method to assess multiple aspects of sexual motivation in female rats administered either an endocannabinoid agonist or antagonist. Because SR or AEA did not affect lordosis, conclusions about the impact of drug exposure on sexual motivation can be made more confidently. The disruptive effects of SR are likely attributed to alterations of the endocannabinoid system, because we found an opposite effect of the endocannabinoid agonist AEA. Furthermore, we found that blockade of the endocannabinoid system may alter motivation to be social (independent of gonadal hormones) as well as sexual motivation (dependent on gonadal hormones). Finally, we found evidence that an interaction between the endocannabinoid system and ovarian hormones may contribute to effects observed on partner preference. We can conclude that antagonism of the endocannabinoid system disrupts sexual motivation, whereas activation of the endocannabinoid system enhances sexual motivation by specifically altering approach behavior. Acknowledgments Financial support for this work was generously provided by the Howard Hughes Medical Institute (#52007558) through the Undergraduate Science Education Program. References Blasberg ME, Clark AS. Anabolic–androgenic steroid effects on sexual receptivity in ovariectomized rats. Horm Behav 1997;32:201–8. Canseco-Alba A, Rodriguez-Manzo G. Anandamide transforms noncopulating rats into sexually active animals. J Sex Med 2013;10:686–93. Clark AS, Kelton MC, Guarraci FA, Clyons EQ. Hormonal status and test condition, but not sexual experience, modulate partner preference in female rats. Horm Behav 2004;45: 314–23. Deutsch DG, Chin SA. Enzymatic synthesis and degradation of anandamide, a cannabinoid receptor agonist. Biochem Pharmacol 1993;46:791–6. Gorzalka BB, Morrish AC, Hill MN. Endocannabinoid modulation of male rat sexual behavior. Psychopharmacology (Berl) 2008;198:479–86.

Grotenhermen F. Cannabinoids and the endocannabinoid system. Cannabinoids 2006;1: 10–4. Halikas J, Weller R, Morse C. Effects of regular marijuana use on sexual performance. J Psychoactive Drugs 1982;14:59–70. Hall W, Degenhardt L, Lynskey M. The health and psychological effects of cannabis use. Australia: Commonwealth of Australia; 2001. Hardy DF, DeBold JF. Effects of coital stimulation upon behavior of the female rat. J Comp Physiol Psychol 1972;78:400–8. Hill MN, Karacabeyli ES, Gorzalka BB. Estrogen recruits the endocannabinoid system to modulate emotionality. Psychoneuroendocrinology 2007;32:350–7. Jarbe TU, Ross T, DiPatrizio NV, Pandarinathan L, Makriyannis A. Effects of the CB1R agonist WIN-55,212-2 and the CB1R antagonists SR-141716 and AM-1387: openfield examination in rats. Pharmacol Biochem Behav 2006;85:243–52. Klein C, Hill MN, Chang SC, Hillard CJ, Gorzalka BB. Circulating endocannabinoid concentrations and sexual arousal in women. J Sex Med 2012;9:1588–601. Koff WC. Marijuana and sexual activity. J Sex Res 1974;10:194–204. Lopez HH, Webb SA, Nash S. Cannabinoid receptor antagonism increases female sexual motivation. Pharmacol Biochem Behav 2009;92:17–24. Mailleux P, Vanderhaeghen JJ. Distribution of neuronal cannabinoid receptor in the adult rat brain: a comparative receptor binding radioautography and in situ hybridization histochemistry. Neuroscience 1992;48:655–68. Mani SK, Mitchell A, O'Malley BW. Progesterone receptor and dopamine receptors are required in delta 9-tetrahydrocannabinol modulation of sexual receptivity in female rats. Proc Natl Acad Sci U S A 2001;98:1249–54. Marcondes FK, Bianchi FJ, Tanno AP. Determination of the estrous cycle phases of rats: some helpful considerations. Braz J Biol 2002;62:609–14. McMahon LR, Koek W. Differences in the relative potency of SR 141716A and AM 251 as antagonists of various in vivo effects of cannabinoid agonists in C57BL/6 J mice. Eur J Pharmacol 2007;569:70–6. Parsons B, Rainbow TC, MacLusky NJ, McEwen BS. Progestin receptor levels in rat hypothalamic and limbic nuclei. J Neurosci 1982;2:1446–52. Rodriguez de Fonseca F, Cebeira M, Ramos JA, Martin M, Fernandez-Ruiz JJ. Cannabinoid receptors in rat brain areas: sexual differences, fluctuations during estrous cycle and changes after gonadectomy and sex steroid replacement. Life Sci 1993;54: 159–70. Substance Abuse and Mental Health Services Administration. Results from the 2008 National Survey on Drug Use and Health: National Findings. (Office of Applied Studies, NSDUH Series H-36, HHS Publication No SMA 09–4434). MD: Rockville; 2009. Substance Abuse and Mental Health Services Administration. Results from the 2010 National Survey on Drug Use and Health: National Findings. (Office of Applied Studies, NSDUH Series H-36, HHS Publication NoSMA 09–4434). MD: Rockville; 2011. Snoeren EM, Chan JS, de Jong TR, Waldinger MD, Olivier B, Oosting RS. A new female rat animal model for hypoactive sexual desire disorder; behavioral and pharmacological evidence. J Sex Med 2011;8:44–56. Zavatti M, Carnevale G, Benelli A, Zanoli P. Effects of the cannabinoid antagonist SR 141716 on sexual and motor behaviour in receptive female rats. Clin Exp Pharmacol Physiol 2011;38:771–5. Zipse LR, Brandling-Bennett EM, Clark AS. Paced mating behavior in the naturally cycling and the hormone-treated female rat. Physiol Behav 2000;70:205–9.