INTRODUCTION

Oxytocin (OT) and vasopressin (AVP) are neuropeptides of considerable current interest because of their demonstrated capacity to modulate social behavior in humans and other animals (Caldwell et al, 2008; Lee et al, 2009). Owing to poor blood–brain barrier (BBB) permeability, they are usually given intranasally to humans (Domes et al, 2007; Kosfeld et al, 2005) or via direct intracranial injection or subcutaneous administration in animal studies (Insel and Winslow, 1991; Lukas et al, 2011; Witt et al, 1992). Intranasal OT in humans can increase trust, social risk taking, and social perception (Baumgartner et al, 2008; Domes et al, 2007; Kosfeld et al, 2005), while repeated dosing may benefit patients with schizophrenia (Feifel et al, 2012), autism (Hollander et al, 2007; Tachibana et al, 2013), and social anxiety disorder (Guastella et al, 2009). AVP on the other hand, improves memory (Perras et al, 1997; Pietrowsky et al, 1996), influences aggression (Thompson et al, 2004), and increases anxiety (Thompson et al, 2006) in humans.

In rodent studies, OT delivered repeatedly through an intracerebroventricular (ICV) route increases social interaction in male rats (Witt et al, 1992), and single-ICV administration reverses the social avoidance seen in rats subjected to social defeat or other stressors (Lukas et al, 2011). Despite the association between AVP and aggression and anxiety, there is also evidence for positive social effects of this peptide. Studies in prairie voles show that ICV infusion of AVP in males promotes a partner preference before meeting, and that OT administration has the same effect in females (Lukas and Neumann, 2012). Region-specific central administration of AVP also enhances social memory (Dantzer et al, 1988) and improves social recognition in rats (Veenema et al, 2011) and mice (Bielsky et al, 2005). Furthermore, intraseptal infusion of arginine vasotocin (the non-mammalian equivalent of AVP) promotes gregariousness and larger flock sizes in Zebra Finches (Kelly et al, 2011).

However, there is relatively little evidence of acute prosocial effects of peripherally administered OT (Hicks et al, 2012) or AVP (Appenrodt et al, 1998). In a recent study from our own laboratory, peripheral OT given repeatedly during adolescence produced lasting increases in sociability and decreased anxiety measured in adulthood (Bowen et al, 2011). Therefore, we decided to further examine here, whether acute prosocial effects of peripheral OT, and perhaps AVP, might also be observed in rats.

The drug 3,4-methylenedioxymethamphetamine (MDMA, ‘Ecstasy’) is a popular entactogen with well-documented prosocial effects in humans and laboratory animals. People commonly describe MDMA as causing an increased feeling of closeness to others, accompanied by a desire to interact socially, as well as euphoria, extraversion, enhanced tactile pleasure and a sense of peacefulness (Dumont and Verkes, 2006). MDMA increases social interaction in rats meeting for the first time, eliciting a characteristic behavior called ‘adjacent lying’, where rats lie together in close contact (Ando et al, 2006; Thompson et al, 2009). A similar behavior to adjacent lying, passive contact, has been previously described by File (1980) and others (File and Pope, 1974; Humphreys and Einon, 1981; Renner and Rosenzweig, 2004) and is believed to reflect an innate motivation towards tactile contact in rats (File and Pope, 1974; Latane et al, 1970), a motivation that MDMA likely increases. MDMA also decreases anogenital sniffing during social interaction. Rodents typically sniff the anogenital regions of conspecifics to decode identity, and the reduction of this behavior by MDMA might imply increased comfort and familiarity in the presence of strangers. Indeed, mice and fish demonstrate decreased aggression towards novel conspecifics when given MDMA (Capurro et al, 1997; Maldonado and Navarro, 2001).

The prosocial effects of MDMA have been linked to central OT release through both rat and human studies. MDMA acutely increases plasma levels of OT (Dumont et al, 2009; Thompson et al, 2007), and this increase correlates with increased subjective feelings of sociability in humans (Dumont et al, 2009). Furthermore, MDMA activates hypothalamic OT-containing neurons via an action on 5-HT1A receptors (Hunt et al, 2011; Thompson et al, 2007). In one study, the prosocial effects of MDMA were partly prevented by intracerebroventricular administration of the relatively non-selective OT antagonist tocinoic acid (Thompson et al, 2007).

It is known that MDMA also stimulates hypothalamic AVP release (Henry et al, 1998), and increases plasma levels of AVP and copeptin, a marker of AVP secretion (Simmler et al, 2011). However, the possible role of the AVP system in MDMA-induced social effects has not been addressed. AVP systems have an increasingly well-documented role in regulating sociability in many species (Engelmann et al, 2006; Lukas and Neumann, 2012), and OT itself has a moderate to strong affinity for vasopressin 1A (V1AR) receptors (Hicks et al, 2012; Manning et al, 2008). Importantly, mice lacking OT receptors (OTR) display ‘autistic-like’ social deficits and impaired cognitive flexibility, and these are rescued via treatment with intracranial OT, an effect likely mediated by the V1AR (Sala et al, 2011). On the other hand, V1AR knockout mice show impaired social interaction (Egashira et al, 2007) and social recognition (Bielsky et al, 2004). Moreover, peripheral tissue contractions, analgesia and proconvulsive effects induced by OT are prevented by pretreatment with SR49059 (Loyens et al, 2011; Schorscher-Petcu et al, 2010), a potent and selective V1AR antagonist with efficacy at human and rat receptors (Manning et al, 2012; Schorscher-Petcu et al, 2010).

The present study examined whether peripherally administered OT and AVP could produce acute prosocial effects in laboratory rats, and the extent to which these resembled those seen with MDMA. A social interaction test was used, which assessed the interaction of two unfamiliar male rats meeting for the first time (File, 1980). There can be substantial differences in sociability between different rat strains (Rex et al, 1996), and our initial pilot work suggested that Long-Evans strain rats were particularly likely to display OT-induced social effects. To explore a mechanism for OT and AVP-facilitated social behavior, we investigated the antagonistic effects of compound 25 (C25), a non-peptide OTR antagonist (Brown et al, 2010), and SR49059, a non-peptide V1AR antagonist, on observed prosocial effects. Finally, to gain further evidence of a common mechanism of sociability, we examined whether low doses of a combination of these drugs (OT and MDMA, or AVP and MDMA), that were ineffective in stimulating social behavior when administered alone, had prosocial effects when given in combination.

MATERIALS AND METHODS

Animals

Tests were conducted on experimentally naïve adult male Long-Evans rats (250–300 g) purchased from Adelaide University (Adelaide, SA, Australia). Rats were housed in groups of eight in large plastic tubs (640 × 400 × 220 mm3) in a temperature (21±1 °C) controlled colony room and maintained under a reverse 12 : 12 h light–dark cycle (lights off at 0900 hours). Animals had ad libitum access to food and water except during testing. Rats were handled daily for 5 min for 7 days before testing. All experiments were conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (7th Edn, 2004) under the approval of The University of Sydney Animal Ethics Committee.

Drugs and Drug Preparation

OT and AVP were purchased from AusPep Ltd (Parkville, VIC, Australia) and dissolved in physiological saline (0.9%) to produce various doses (OT: 0.1, 0.25, 0.5, and 1 mg/kg; AVP: 0.001, 0.0025, 0.005, 0.01, and 0.1 mg/kg). MDMA was obtained from the Australian Government Analytical Laboratories (Pymble, NSW, Australia) and dissolved in saline to a dose of 2.5 and 5 mg/kg. The OTR antagonist C25 was synthesized in accordance with the methods of Brown et al, (2010) and was considered to be of >95% purity based on proton, carbon nuclear magnetic resonance spectroscopy and mass spectrometry. It was dissolved in a 15% dimethyl sulfoxide (DMSO), 2% Tween-80 and 83% saline vehicle at a dose of 5 mg/kg. The V1AR antagonist SR49059 (Axon MedChem BV, The Netherlands) was dissolved to a dose of 1 mg/kg in the same vehicle used for C25. All drugs were administered to rats via intraperitoneal (IP) injection at a volume of 1 ml/kg.

Social Interaction Test

Rats were tested in black, rectangular arenas (780 × 520 × 470 mm3) dimly lit with red light (80 W) and lined with recycled paper bedding pellets (Breeders Choice, Toowoomba, QLD, Australia). Each arena was equipped with a miniature infrared video camera placed vertically over the center of the arena, which sent images to a computer outside the room that allowed test sessions to be recorded. Test sessions were manually scored by an observer blind to treatment conditions using the program ODLog (Macropod Software; www.macropodsoftware.com).

The behaviors of interest were (1) adjacent lying: side-by-side contact, but not including climbing over or under the conspecific; (2) general investigation: sniffing of the conspecific, but not of the anogenital area; (3) anogenital sniffing: sniffing of the anogenital region of the conspecific; and (4) rearing: standing on hind legs, including leaning against the wall of the arena. Arenas were cleaned of fecal boli and the walls were sprayed with a 50% ethanol solution between each test session. Experiments lasted for 30 min, unless otherwise noted, and were performed in the dark phase at an ambient temperature of 23 °C (±0.5).

Experimental Procedures

For all experiments, rats were individually habituated to the arena in which they would be tested for 60 min on the day before testing. Subjects were weight paired (±10 g) with an unfamiliar conspecific given the same drug treatment that was randomly allocated. On test days, animals were individually placed in plastic holding cages (470 × 280 × 180 mm3 lined with the same recycled paper bedding used in the test arenas) in the testing room before, and between injections. Experiments involving MDMA used a between-groups design where MDMA-treated rats only received the drug once. This was done to avoid possible downregulation of MDMA social effects with repeated exposure to the drug (Thompson et al, 2008).

Prosocial Effects of OT (Experiment 1)

The first experiment characterized the dose-dependent effects of OT on social interaction in male Long-Evans rats. Rats (N=8 pairs) were tested with each of three doses of OT (0.1, 0.5, and 1 mg/kg) and saline in a counterbalanced order with two washout days between each of the four treatments. For each of the four tests, each rat was paired with a new partner relative to the previous sessions. Rats were placed in the test arenas for 30 min, 10 min after injection of OT.

Effects of C25 on OT-Induced Social Behavior (Experiment 2)

As a result of pilot work in our laboratory, we chose to give C25 at a dose of 5 mg/kg (IP) rather than the 2 mg/kg intravenous dose used to measure bioavailability by Brown et al (2010). In a between-groups design, pairs of rats (N=6 per condition) were administered either (1) vehicle+vehicle, (2) C25 (5 mg/kg)+vehicle, (3) vehicle+OT (0.5 mg/kg), or (4) C25 (5 mg/kg)+OT (0.5 mg/kg). Injections were given 15 min apart and rats were tested for social interaction for 30 min, 10 min after the second injection.

Effects of SR49059 on OT-Induced Social Behavior (Experiment 3)

In a between-groups design, pairs of rats (N=4 per condition) were administered either (1) vehicle+vehicle, (2) SR49059 (1 mg/kg)+vehicle, (3) vehicle+OT (0.5 mg/kg), or (4) SR49059 (1 mg/kg)+OT (0.5 mg/kg). Antagonist/vehicle injections were given 15 min before OT/vehicle treatment, and rats were tested for social interaction for 30 min, 10 min after the second injection.

Prosocial Effects of AVP (Experiment 4)

Rats (N=10 pairs) were tested with each of four doses of AVP (0.001, 0.005, 0.01 and 0.1 mg/kg) and saline in a counterbalanced order with two washout days between each of the five treatments. For each of the tests, each rat was paired with a different partner relative to the previous sessions. Ten min after injection of AVP, rats were placed in the test arenas for 30 min.

Effects of SR49059 on AVP-Induced Social Behavior (Experiment 5)

Using a between-groups design, pairs of rats (N=4 per condition) were administered either (1) vehicle+vehicle, (2) SR49059 (1 mg/kg)+vehicle, (3) vehicle+AVP (0.01 mg/kg), or (4) SR49059 (1 mg/kg)+AVP (0.01 mg/kg). Injections were given 15 min apart and rats were tested for social interaction for 30 min, 10 min after the second injection.

Prosocial Effects of MDMA (Experiment 6)

In a between-groups design, pairs of rats (N=8 per condition) were given either MDMA (5 mg/kg) or saline, and 20 min later placed in the test arenas. Following on from our earlier studies with MDMA (Thompson et al, 2007; Thompson et al, 2009), this social interaction test was only of 10 min duration and was intended to show that our earlier results demonstrating prosocial effects of MDMA in Wistar rats could be replicated in Long-Evans rats.

Effects of SR49059 on MDMA-Induced Social Behavior (Experiment 7)

Using a between-groups design, pairs of rats (N=8 per condition) received injections of vehicle or SR49059 (1 mg/kg) followed 15 min later by MDMA (5 mg/kg). Testing commenced 20 min after the MDMA treatment and lasted for 30 min. To minimize the number of rats used, this experiment did not include the vehicle+vehicle or SR49059+vehicle conditions that had already been studied in Experiment 2.

Prosocial Effects of Combined Sub-Threshold Doses of OT and MDMA (Experiment 8)

In a between-groups design, pairs of rats (N=4 per condition) were randomly allocated to one of four conditions: (1) saline+saline, (2) saline+OT, (3) saline+MDMA, and (4) OT+MDMA. A 0.25 mg/kg dose of OT and 2.5 mg/kg dose of MDMA were used as they are below the threshold of affecting social interaction. Drug injections were separated by 10 min, and the 30 min test commenced 10 min after the second injection.

Prosocial Effects of Combined Sub-threshold Doses of AVP and MDMA (Experiment 9)

Using a between-groups design, pairs of rats (N=4 per condition) were randomly assigned to one of four conditions: (1) saline+saline, (2) saline+AVP, (3) saline+MDMA, and (4) AVP+MDMA. Sub-threshold doses of AVP (0.0025 mg/kg) and MDMA (2.5 mg/kg) were used. Injections were administered 10 min apart, and the 30 min test began 10 min after the second injection.

Statistical Analysis

Experiments 1 and 4 used a counterbalanced design where each rat received all possible doses of OT and AVP, respectively. As each data point represented a novel pairing of two rats, a between-subjects analysis was used to compare doses. In the remaining experiments, each rat received only a single treatment, so again a between-subjects analysis was used. With the exception of Experiments 6 and 7, data were analyzed using the one-way ANOVA procedure followed by post-hoc tests (Tukey’s) to enable specific group comparisons. Normality and homogeneity of variance (Levene’s test) were checked and where these requirements were not satisfied, a log10(X+1) or square-root transformation was applied to the data and the most effective correction was used for analysis. In the case where these assumptions could not be satisfied, the Kruskal–Wallis non-parametric test was used followed by the Games–Howell multiple comparisons test. Data from Experiments 6 and 7 were analyzed using the independent-samples t-test owing to their two group design. All analyses were conducted using SPSS version 19 (SPSS., IBM, Chicago, IL), with significance set at P<0.05 and only untransformed data are presented in the figures.

RESULTS

Dose-Dependent Effects of OT on Social Interaction

The effects of OT in the social interaction test is presented in Figure 1. OT had a significant overall effect on adjacent lying (F(3,28)=15.62, P<0.001; Figure 1a), with post-hoc tests indicating that the 0.5 mg/kg dose significantly increased adjacent lying compared with all other conditions (all P⩽0.001) (see Supplementary Video). There was a significant effect of OT on general investigation (F(3,28)=5.38, P<0.01; Figure 1b), with 1 mg/kg reducing the amount of investigation time relative to the 0.1 (P<0.01) and 0.5 mg/kg doses (P<0.05). Anogenital sniffing was also affected by OT (F(3,28)=5.14, P<0.01; Figure 1c), with 0.5 mg/kg reducing sniffing compared with saline treatment (P<0.05) and 0.1 mg/kg (P<0.05), while 1 mg/kg OT also reduced this behavior relative to saline (P<0.05). OT also affected rearing behavior (F(3,28)=12.28, P<0.001; Figure 1d), with post-hoc tests indicating that 0.5 and 1 mg/kg OT attenuated rearing in comparison with saline (P<0.05 and 0.001, respectively) and 0.1 mg/kg OT (P<0.01 and 0.001, respectively).

Figure 1
figure 1

Dose-dependent effects of oxytocin (OT) on social interaction. (a) Rats given 0.5 mg/kg OT displayed increased adjacent lying. (b) A high dose of OT (1 mg/kg) resulted in decreased general investigation when compared with 0.1 and 0.5 mg/kg. (c) OT given at 0.5 and 1 mg/kg reduced anogenital sniffing compared with saline (SAL). The non-social behavior of rearing (d) was significantly reduced in rats given OT at 0.5 and 1 mg/kg compared with saline. Data are means+SEM, *P<0.05 vs saline, †P<0.05 vs 0.5 mg/kg OT, #P<.05 vs 0.1 mg/kg OT, ***P⩽0.001 vs all other conditions.

PowerPoint slide

Full size image

Effects of C25 on OT-Induced Social Behavior

The effects of C25 on OT-induced social interaction are presented in Figure 2. There was a significant overall effect of treatment on adjacent lying (F(3,20)=18.68, P<0.001; Figure 2a), anogenital sniffing (F(3,20)=18.57, P<0.001; Figure 2c), and rearing (F(3, 20)=3.60, P<0.05; Figure 2d). Post-hoc tests showed that OT given alone, and following pretreatment with C25, significantly increased adjacent lying and reduced anogenital sniffing and rearing, when compared with vehicle treatment and C25 alone (adjacent lying: all P<0.01; anogenital sniffing: all P<0.001; rearing: P<0.05 compared with C25+vehicle). There was no significant effect of treatment on general investigation (P>0.05; Figure 2b).

Figure 2
figure 2

The effects of compound 25 (C25) on oxytocin (OT)-induced social behaviors. The pretreatment with C25 did not affect OT-induced adjacent lying (a), or the decreased anogenital sniffing (c) and rearing (d) induced by OT. General investigation (b) was unaffected. Data are means+SEM, VEH, vehicle, #P<0.05 vs C25+VEH, **P<0.01 vs VEH+VEH and C25+VEH, and ***P<0.001 vs VEH+VEH and C25+VEH.

PowerPoint slide

Full size image

Effects of SR49059 on OT-Induced Social Behavior

The effects of SR49059 on OT-induced social interaction are presented in Figure 3. There was a significant overall effect of treatment on adjacent lying (F(3,12)=67.04, P<0.001; Figure 3a) and anogenital sniffing (F(3,12)=5.68, P<0.05; Figure 3c). Post-hoc tests showed that adjacent lying was significantly increased, whereas anogenital sniffing was reduced, by 0.5 mg/kg OT relative to all other conditions (adjacent lying: all P<0.001; anogenital sniffing: all P<0.05). There was no significant effect of treatment on general investigation (P>0.05; Figure 3b) or rearing behavior (P>0.05; Figure 3d).

Figure 3
figure 3

The effects of SR49059 (SR) on oxytocin (OT)-induced social behaviors. Increased adjacent lying (a) caused by OT was completely prevented by pretreatment with SR, while suppression of anogenital sniffing (c) by OT was also prevented. There was no significant effect of treatment on general investigation (b) or rearing (d). Data are means+SEM, VEH, vehicle, *P<0.05 vs all other conditions, ***P<0.001 vs all other conditions.

PowerPoint slide

Full size image

Dose-Dependent Effects of AVP on Social Interaction

The dose-response profile of AVP in the social interaction test is presented in Figure 4. AVP had a significant overall effect on adjacent lying (F(4,45)=19.22, P<0.001; Figure 4a), with post-hoc tests indicating that the 0.01 and 0.005 mg/kg doses significantly increased adjacent lying compared with all other conditions (all P⩽0.001), but did not differ significantly from each other (P>0.05) (see Supplementary Video). There was a significant effect of AVP on general investigation (F(4,45)=29.97, P<0.001; Figure 4b), with the 0.1 mg/kg dose reducing the amount of investigation time relative to all other conditions (all P<0.001). Moreover, AVP had a significant effect on anogenital sniffing (Kruskal–Wallis test; χ2=33.85, P<0.001, df=4; Figure 4c), with 0.005, 0.01, and 0.1 mg/kg AVP significantly reducing the time spent engaged in this behavior relative to saline treatment (all P<0.001) and 0.001 mg/kg AVP (all P<0.01). Rearing behavior was also significantly reduced by AVP (Kruskal–Wallis test; χ2=28.86, P<0.01, df=4; Figure 4d) at a dose of 0.005, 0.01, and 0.1 mg/kg relative to saline treatment (all P<0.01) and 0.001 mg/kg AVP (all P<0.01).

Figure 4
figure 4

Dose-dependent effects of vasopressin (AVP) on social interaction. (a) Adjacent lying increased in rats given 0.005 and 0.01 mg/kg AVP. (b) A high dose of AVP (0.1 mg/kg) resulted in decreased general investigation compared with all other conditions. AVP given at 0.005, 0.01, and 0.1 mg/kg reduced anogenital sniffing (c) and rearing (d) compared with saline (SAL) and 0.001 mg/kg AVP. Data are means+SEM, #P<0.01 vs saline and 0.001 mg/kg AVP, ***P⩽0.001 vs saline, 0.001 and 0.1 mg/kg AVP, †P⩽0.001 vs all other conditions.

PowerPoint slide

Full size image

Effects of SR49059 on AVP-Induced Social Behavior

The effects of SR49059 on AVP-induced social interaction are presented in Figure 5. There was a significant overall effect of drug treatment on adjacent lying (F(3,12)=22.45, P<0.001; Figure 5a), anogenital sniffing (F(3,12)=22.55, P<0.001; Figure 5c), and rearing (F(3, 12)=5.25, P<0.05; Figure 5d). Post-hoc tests showed that adjacent lying was significantly increased, whereas anogenital sniffing was reduced, by 0.01 mg/kg AVP relative to all other conditions (adjacent lying: all P⩽0.001; anogenital sniffing: all P<0.001). Pretreatment with SR49059 prevented the reduction in rearing caused by AVP (P<0.05). There was no significant effect of drug treatment on general investigation (P>0.05; Figure 5b).

Figure 5
figure 5

The effects of SR49059 (SR) on vasopressin (AVP)-induced social behaviors. SR prevented AVP-induced adjacent lying (a) and the decrease in anogenital sniffing (c) and rearing (d) caused by AVP. There was no significant effect of drug treatment on general investigation (b). Data are means+SEM, VEH, vehicle, *P<0.05 vs SR+AVP condition, ***P⩽0.001 vs all other conditions.

PowerPoint slide

Full size image

Prosocial Effects of MDMA

Rats given MDMA (5 mg/kg) spent significantly more time engaged in adjacent lying behavior in a 10-min test than saline-treated rats, (t(14)=9.42, P<0.001; see Figure 6 and Supplementary Video), and displayed less anogenital sniffing (t(14)=5.25, P<0.001) and rearing (t(14)=12.22, P<0.001). MDMA had no significant effect on general investigation (P>0.05).

Figure 6
figure 6

The effects of 3,4-methylenedioxymethamphetamine (MDMA) on social interaction. Rats given MDMA showed a significant increase in adjacent lying as well as a decrease in anogenital sniffing and rearing behavior. Data are means+SEM, ***P<0.001 vs saline (SAL) animals.

PowerPoint slide

Full size image

Effects of SR49059 on MDMA-Induced Social Behavior

The interaction of SR49059 with MDMA is shown in Figure 7. Pretreatment with SR49059 prevented the increase in adjacent lying caused by MDMA, (t(14)=3.01, P<0.01), and the MDMA-induced reduction in anogenital sniffing (t(14)=2.44, P<0.05). However, it did not prevent the MDMA effects on rearing behavior (P>0.05). Rats given SR49059 before MDMA engaged in more general investigation than those given MDMA alone, (t(14)=5.50, P<0.001).

Figure 7
figure 7

The effects of SR49059 (SR) on 3,4-methylenedioxymethamphetamine (MDMA)-induced social behaviors. Pretreatment with SR reduced MDMA-induced adjacent lying, and attenuated MDMA effects on general investigation and anogenital sniffing. Data are means+SEM, VEH, vehicle, *P<0.05 vs VEH+MDMA, **P<0.01 vs VEH+MDMA, ***P<0.001 vs VEH+MDMA.

PowerPoint slide

Full size image

Prosocial Effects of Combined Sub-Threshold Doses of OT and MDMA

Results are shown in Figure 8. Adjacent lying behavior was significantly affected by drug treatment (F(3,12)=12.53, P⩽0.001; Figure 8a), with post-hoc tests indicating that rats given combined OT and MDMA showed more adjacent lying than those in all other conditions (all P<0.01). Drug treatment significantly affected anogenital sniffing (F(3,12)=8.42, P<0.01; Figure 8c) and rearing behavior (F(3,12)=6.15, P<0.01; Figure 8d), with post-hoc tests showing that rats administered combined OT and MDMA spent significantly less time engaged in these behaviors relative to all other groups (anogenital: P<0.05 vs OT, P<0.01 vs saline and MDMA; rearing: all P<0.05). There was no significant effect of drug treatment on general investigation (P>0.05; Figure 8b).

Figure 8
figure 8

Effects of combined sub-threshold doses of oxytocin (OT) and 3,4-methylenedioxymethamphetamine (MDMA) on social behavior. (a) Rats given OT (0.25 mg/kg) and MDMA (2.5 mg/kg) spent significantly longer lying adjacent to each other, and significantly less time engaged in anogenital sniffing (c) and rearing (d) relative to all other conditions. General investigation was unaffected by drug treatment (d). Data are means+SEM, SAL, saline, *P<0.05 vs all other conditions, **P<0.01 vs all other conditions.

PowerPoint slide

Full size image

Prosocial Effects of Combined Sub-Threshold Doses of AVP and MDMA

Results are shown in Figure 9. Adjacent lying behavior was significantly affected by drug treatment (F(3,12)=16.32, P⩽0.001; Figure 9a), with post-hoc tests revealing that combined AVP and MDMA significantly increased adjacent lying relative to all other treatment conditions (all P⩽0.001). Drug treatment significantly affected anogenital sniffing (F(3,12)=20.15, P<0.001; Figure 9c), with post-hoc tests showing that rats administered AVP, MDMA or the combination of AVP and MDMA, spent significantly less time engaged in this behavior relative to saline (all P⩽0.001). There was no significant treatment effect on general investigation (P>0.05; Figure 9b) or rearing (P>0.05; Figure 9d).

Figure 9
figure 9

Effects of combined sub-threshold doses of vasopressin (AVP) and 3,4-methylenedioxymethamphetamine (MDMA) on social behavior. (a) Rats administered AVP (0.025 mg/kg) and MDMA (2.5 mg/kg) spent significantly longer lying adjacent to each other, and significantly less time engaged in anogenital sniffing (c). There was no significant effect of treatment on general investigation (b) or rearing (d). Data are means+SEM, ***P⩽0.001 vs all other conditions, ###P⩽0.001 vs saline (SAL).

PowerPoint slide

Full size image

DISCUSSION

The present study examined the prosocial effects of OT, AVP and MDMA and their combinations in the social interaction paradigm in Long-Evans rats. A principal aim was to assess whether robust prosocial effects of peripherally administered OT and AVP could be obtained, and a further aim was to compare this with previously demonstrated MDMA prosocial effects in rats. We also studied the possible involvement of the OTR and V1AR in mediating the prosocial effects of OT, AVP and MDMA.

Up until this point, prosocial effects of OT in rodent models have only been demonstrated with chronic intracranial administration (Insel and Winslow, 1991), or as a lasting residual effect of repeated peripheral (Bowen et al, 2011) or intracranial (Witt et al, 1992) OT administration, or in the ‘rescue’ of stressed animals that are socially phobic (Slattery and Neumann, 2010), socially defeated (Lukas et al, 2011), or socially stressed due to isolation (Grippo et al, 2009). Here, we show that acute peripheral injection of OT robustly increased the prosocial behavior of adjacent lying in Long-Evans rats under normal basal conditions. A narrow dose window was evident in which these effects were observed, with 0.5 mg/kg OT but not 0.1, 0.25 or 1 mg/kg increasing adjacent lying. The behavior of adjacent lying has been previously described in Wistar rats administered MDMA, or the 5-HT1A receptor agonist 8-OH-DPAT (Thompson et al, 2007; Thompson et al, 2009). This MDMA effect was replicated here in Long-Evans rats.

We also show here for the first time that AVP, like OT, induces adjacent lying in rats. A robust increase was observed at 0.01 mg/kg AVP and 0.005 mg/kg AVP in comparison with the other doses, which had no such effects. OT and AVP are sometimes conceptualized as having opposing influences, with mutually inhibitory effects evident in the amygdala (Huber et al, 2005), and OT is generally linked to passive prosocial behaviors while AVP is more linked to fear and aggression (Debiec, 2005). It is somewhat surprising then that AVP should have such clear prosocial effects in rats, mimicking those of OT. Up until this point, few studies have examined the behavioral effects of IP AVP administration. One study found that rats showed a decrease in anxiety-like behavior on the elevated plus maze (Appenrodt et al, 1998), while several studies focused on avoidance behavior and AVP’s ability to delay extinction (Bohus et al, 1972; Koob et al, 1981). Intracranial administration of AVP facilitates social memory and recognition in rats (Le Moal et al, 1987), and it helps form partner preference as well as mediates selective aggression in prairie voles (Winslow, 1993).

The commonality of effects of OT and AVP was also evident not only in adjacent lying, but also in both OT (0.5 mg/kg) and AVP (0.01 mg/kg) significantly reducing anogenital sniffing. Anogenital sniffing is an important behavior for conspecific recognition, and its reduction by OT and AVP may relate to a modulation of normal social recognition processes, whereby unfamiliar rats are in some way more rapidly accepted as familiar. Indeed, the infusion of AVP or OT into the olfactory bulb can improve social recognition in male rats, while administration of a V1AR antagonist disrupts social recognition processes (Dluzen et al, 1998; Tobin et al, 2010). Similarly, in mice, site-specific administration of an OTR antagonist into the medial amygdala results in diminished social recognition (Ferguson et al, 2001). It is possible, then, that an enhancement of social recognition by OT and AVP may be related to the reduced anogenital sniffing observed in the current study.

It is important to consider whether the apparently prosocial effects of OT and AVP reflected in increased adjacent lying and decreased anogenital sniffing may reflect a non-specific sedative effect. Higher peripheral doses of OT (Hicks et al, 2012) and AVP (Andrews et al, 1983; Ettenberg et al, 1983) in rats can reduce locomotor activity. Several factors are worth noting here. First, it was clear that the highest doses of AVP (0.1 mg/kg) and OT (1 mg/kg) used had inhibitory effects on ongoing activity, yet these doses did not increase adjacent lying. Rather it was at intermediate doses, where general investigation was not affected, that increased adjacent lying was seen with both OT and AVP. As the summary in Table 1 shows the critical doses that increased adjacent lying and decreased anogenital sniffing were without effect on general investigation and had only marginally inhibitory effects on rearing, it is also important to note that MDMA given to individual rats tends to produce hyperactivity, rather than sedation, yet in the current study, and our previous studies MDMA also increases adjacent lying and decreases anogenital sniffing and rearing (Thompson et al, 2007; Thompson et al, 2009).

Table 1 Drug Effects on Behavior in the Social Interaction Test (Values are Means)
Full size table

The striking similarity between OT, AVP and MDMA in these effects suggests a commonality of action, and this was confirmed in Experiments 8 and 9 where two drugs (OT and MDMA or AVP and MDMA) combined in low doses, produced the characteristic effects of either drug given alone (Table 1). Interestingly, the characteristic changes in adjacent lying and anogenital sniffing produced by MDMA, AVP and OT were reversed by pretreatment with SR49059, which by itself had no significant effects on social behavior. Moreover, the selective OTR antagonist C25 (Brown et al, 2010) had no effects on OT-induced social behaviors, suggesting that the effects of OT, AVP and MDMA are not mediated by the OTR. It is important to note that we have previously found intracranial administration of the non-selective OTR antagonist tocinoic acid reduced the prosocial effects of MDMA (Thompson et al, 2007). However, this may be due to non-specific blockade of AVP receptors given the high dose used (ie, 20 μg/μl) and the functional activity of this ligand at both OT and AVP receptors (Maggi et al, 1987). Therefore, our current findings indicate that OT and AVP may directly act on the V1AR to produce prosocial effects, while MDMA may indirectly stimulate the V1AR through serotonin-induced OT and/or AVP release in the hypothalamus (Jorgensen et al, 2003).

SR49059 has only weak affinity for the OTR (Manning et al, 2012) and when given intravenously at the current dose of 1 mg/kg inhibits AVP-induced hypertension, while exhibiting no agonist or antagonist effects at other peptide ligand receptors (Manning et al, 2012). Thus, the increase in sociability seen with AVP, OT and MDMA may be independent of direct OTR activation. However, close inspection of Figures 3, 5 and 7 suggests that SR49059 completely reverses OT- and AVP-induced increases in adjacent lying, but only partially blocks the increase caused by MDMA. Therefore, the possibility of MDMA acting to promote sociability through other mechanisms cannot be ruled out. Experiments using OTR or V1A receptor null mice might be particularly instructive.

One limitation of the current study is the extent to which the antagonists C25 and SR49059 can cross the BBB. Limited in vivo findings with C25 (Brown et al, 2010) have been reported; however, the compound has excellent bioavailability and aqueous solubility. A previous report suggests that SR49059 administered intravenously is not able to penetrate the BBB when dissolved in a 0.6% methylcellulose and saline vehicle (Tribollet et al, 1999). However, other studies suggest that a DMSO vehicle may increase the permeability of the BBB and carry along other dissolved substances (Jacob and Wood, 1967). Certainly, the fact that the social effects of MDMA were largely prevented by SR49059 in the present study strongly argues for its effectiveness at central sites.

The present results also add to mounting evidence that peripherally administered OT and/or AVP can have important behavioral effects in rodent models. Although it might be argued that these effects occur indirectly via vagal afferents, the evidence for direct penetration of the central nervous system is increasingly convincing. Peripherally administered OT has a wide variety of behavioral effects in rodents including decreased anxiety in adulthood in rats (Bowen et al, 2011), and increased social behavior in male prairie voles (Bales et al, 2012). Pretreatment with peripheral OT dose-dependently blocks disruption of pre-pulse inhibition seen in rats administered amphetamine and dizolcilpine, suggesting that OT can directly modulate dopamine and glutamate systems in the brain (Feifel and Reza, 1999). Peripheral OT also increases c-Fos expression in a region-specific manner, with activated areas including the paraventricular nucleus of the hypothalamus and the supraoptic nucleus, both of which contain high levels of OT neurons (Hicks et al, 2012). Moreover, the pattern of c-Fos expression seen with peripheral OT injections strongly resembles that seen with ICV OT, or with non-peptide OT agonists such as WAY 267,464 that readily penetrate the brain (Hicks et al, 2012). Furthermore, behavioral changes, such as, increased punished crossings in the four-plate anxiolytic activity test caused by peripheral OT, can be blocked with central administration of an OTR antagonist, suggesting central actions of peripheral OT (Ring et al, 2006).

Overall, our results with SR49059 imply that the prosocial effects induced by OT, AVP, and MDMA are mediated by brain AVP systems. This is consistent with previous observations that exogenous AVP prolongs social memory (Le Moal et al, 1987), improves social recognition in rats (Engelmann and Landgraf, 1994), and facilitates pair bonding in male Prairie voles (Insel et al, 1998). As noted above, MDMA has been reported to increase AVP plasma levels in humans (Henry et al, 1998) and stimulate AVP release in isolated rat hypothalamus (Fallon et al, 2002).

Perhaps the most striking finding, however, in the current study, was that the combination of sub-threshold doses of these drugs (doses that by themselves did not affect social behavior) induced a robust increase in adjacent lying. The combination of these drugs also resulted in reduced anogenital sniffing and decreased rearing. This finding is consistent with a recent study showing that OT partially substitutes for MDMA in a drug discrimination paradigm, indicating an overlap of the subjective effects of the two drugs (Broadbear et al, 2011). There is current interest in the utility of MDMA in treating post-traumatic stress disorder (PTSD) (Mithoefer et al, 2011). However, animal models suggest that repeated MDMA can perturb brain neurochemistry and increase vulnerability to depression and anxiety-like behaviors, even after prolonged cessation of use (McGregor et al, 2003). Based on our findings that the combination of sub-threshold doses of OT and MDMA, and AVP and MDMA, have positive effects on social behavior, treatment for PTSD could potentially be made safer by lowering the dose of MDMA and adding OT or AVP, thereby reducing the likelihood that these patients develop adverse effects with long-term treatment.

In summary, our study provides the first demonstration of acute prosocial effects of peripherally administered OT and AVP in rats, similar to that previously reported with MDMA. The adjacent lying induced by OT is unaffected by C25, however, SR49059 reversed the prosocial effects induced by OT, AVP, and MDMA providing support for a common mechanism of action involving the V1AR. These findings may be useful in the development of therapeutics for social disorders by allowing them to target the V1A receptor. This study also shows that co-administration of sub-threshold doses of these drugs stimulates prosocial behavior that further indicates a commonality of action between them. This observation may be of utility in clinical settings where use of MDMA is being contemplated for therapeutic purposes.

FUNDING AND DISCLOSURE

The authors declare no conflict of interest.