Role of oxytocin in the ventral tegmental area in social reinforcement

Abstract

The rewarding properties of social interactions play a critical role in the development and maintenance of social relationships, and deficits in social reward are associated with various psychiatric disorders. In the present study, we used a novel Operant Social Preference (OSP) task to investigate the reinforcing properties of social interactions under conditions of high or low reward value, and high or low behavioral effort in male Syrian hamsters. Further, we investigated the role of oxytocin (OT) in a key structure of the mesolimbic reward system, the ventral tegmental area (VTA), in mediating the reinforcing properties of social interaction. Adult male hamsters were placed in a three-chambered apparatus, and allowed access to either a social chamber containing an unrestrained conspecific or a non-social chamber, by pushing through a one-way entry, vertical-swing door. Increasing the duration of social interaction (reward value) decreased the frequency of entering the social interaction chambers, whereas decreasing the duration of social interaction conversely increased the frequency of entries. Moreover, increasing behavioral effort required to access social interaction decreased the frequency of entries, especially under conditions when the duration of social interaction was only 5 s. OT injected into the VTA decreased the frequency of entering social interaction chambers in a manner similar to that observed when duration was increased, whereas injection of an OT receptor antagonist in the VTA increased the frequency of seeking social interaction. Taken together, these data support the hypothesis that activation of OT receptors in the VTA are critical for the reinforcing properties of social interactions. Furthermore, social interactions may exhibit duration and cost dependent reinforcing effects on behavior similar to those observed with food and drugs of abuse.

Keywords: Dopamine; Mesolimbic dopamine system; Neuropeptides; Operant; Social behavior; Social cognition; Social interaction; Social motivation; Social reward; Social salience; Vasopressin.

Fig. 1.

Operant Social Preference Apparatus: schematic (left) and vertical view (right). The main chamber contains an experimental subject whereas, one of the side chambers is occupied by a stimulus hamster; the subject and stimulus hamster are separated by a vertical-swing door. Visual, auditory and olfactory cues are shared between the chambers through the open top and holes in the door. An acrylic bucket positioned on the stimulus hamster side of the doors serves to hold weights.

Fig. 2.

Effect of Time Allowed in Chambers Containing a Stimulus Hamster (Social Chamber) and Time Allowed in Non-Social Chambers on Social Preference. a) More time allowed in chambers decreased number of entries into social chambers, but had no effect on entries into non-social chambers (n = 11; * p ≤ 0.050; *** p ≤ 0.001). b) There were no differences in the latency to the first entry into chambers in groups assigned to different durations of time in the chambers containing a social stimulus or to groups assigned to different durations of time in non-social chambers. There was, however, a significantly shorter latency to the first entry into chambers containing a stimulus hamster compared to the latency to the first entry into non-social chambers (* p = 0.002; inset). c) More time in chambers decreased social preference score (** p ≤ 0.010). d) More time in chambers decreased the rate of social chamber entries as a proportion of time in the main chamber.

Fig. 3.

Effect of Time Allowed in Social and Non-Social Chambers on the Post Reinforcement Pause and Social Behavior. a) More time in the chambers increased the latency to re-enter the chambers containing stimulus hamsters, but had no effect on the latency to re-enter the non-social chambers, an effect observed on both the first and second post reinforcement interval (n = 11; * p ≤ 0.050; ** p ≤ 0.010). b) More time allowed in the chambers decreased the proportion of time spent socially investigating stimulus hamsters when in the social interaction chambers, but an increase in the proportion of time spent grooming when in the main chamber (*** p ≤ 0.001). c) More time allowed in the social interaction chambers also increased the rate of flank marking when in the main chamber.

Fig. 4.

Effect of the Duration of Social Interaction on Social Preference under Conditions Requiring Increasing Effort. a) While all hamsters (n = 9) entered the chambers containing stimulus hamsters when the door weighed 113 g or 227 g, significantly fewer hamsters entered the social chambers at least once as the door weights increased (n = 9; * p < 0.05). Significantly more hamsters in the 60 s group entered the social chamber at least once in the session compared to the 5 s duration group (p = 0.05). b) All door weights (compared to 113 g) decreased the number of entries into chambers containing stimulus hamsters for subjects that were allowed 5 s of social interaction. Only the heaviest three door weights decreased the number of entries into chambers containing stimulus hamsters for subjects that were allowed 60 s of social interaction. c) Compared to a door weight of 113 g, all door weights decreased the social preference score for subjects that were allowed 5 s of social interaction. Only the heaviest door weight decreased social preference score for subjects that were allowed 60 s of social interaction. d) Compared to a door weight of 113 g, all door weights decreased the number of entries per minute for subjects allowed 5 s of social interaction. Only the heaviest 3 door weights decreased social entries per minute in the main chamber for subjects that were allowed 60 s of social interaction. e) Correlation between door weights and the number of entries into chambers containing stimulus hamsters revealed that both 5 s and 60 s treatment conditions show a linear decrease in number of entries with increasing door weights. The slope is greater in the 5 s condition than the 60 s condition. Asterisks in graphs (b, c, d) indicate a significant difference from the door weight of 113 g, with asterisks and lines color-coded (gray 5 s Condition; black 60 s Condition) (* p ≤ 0.050; ** p ≤ 0.010; *** p ≤ 0.001).

Fig. 5.

Effect of injection of OT and an OTR Antagonist in the VTA on Social Preference. a) OT (9 μM n = 18; 90 μM n = 14) injected in the VTA 5 min before the test decreased the number of entries into chambers containing a stimulus hamsters (i.e., Social Chamber), but had no effect on the number of entries into non-social chambers (i.e., non-social chamber; * p ≤ 0.050; *** p ≤ 0.001). b) Injection of the OTR antagonist at 0.9 μM (n = 14) but not 9 μM (n = 14) increased the number of entries into the Social Chamber, but neither dose affected the number of entries into the non-social chamber (** p ≤ 0.010).

Fig. 6.

Effect of Injections of OT and an OTR Antagonist in the VTA on the latency to reenter chambers containing stimulus hamsters (post reinforcement pause). a) OT (90 μM) injected into the VTA increased the latency to reenter chambers containing stimulus hamsters trended towards significance (p = 0.132). b) Likewise, injections of the OTR antagonist (9 μM) decreased the latency to re-enter chambers containing stimulus hamsters also just missed significance (p = 0.096).

Fig. 7.

Effect of injection of OT and an OTR Antagonist in the VTA on Social Investigation, Aggression, Grooming and Flank Marking, and Locomotor Activity. a) OT (9 μM n = 18; 90 μM n = 12) injected into the VTA had no effect on social investigation, aggression or c) attacks per minute. b) Injection of the OTR antagonist (0.9 μM n = 14) had no effect on social investigation, aggression or e) attacks. d) A trend for the 90 μM concentration of OT to increase the duration of grooming that just missed significance (p = 0.066) was observed. f) Injections of the OTR antagonist had no effect on grooming. g) The 90 μM concentration of OT increased the number of flank marks (* p ≤ 0.050). i) The OTR antagonist injected into the VTA had no effect on the number of flank marks observed. h) Injection of OT had no effect on the amount of locomotor activity. j) Injection of 0.9 μM OTR antagonist increased locomotor activity (p ≤ 0.050) and a trend for 9 μM OTR antagonist to increase locomotor activity just missed significance (p = 0.061).

Fig. 8.

Localization of Drug Injections. a) Representative injection site in the caudal VTA. b) Localization of the sites of injection of oxytocin (OT) and OT receptor antagonist. Subjects with ink found within the caudal VTA were classified as hits (black circle), while subjects with ink found outside the caudal VTA were classified as misses (gray circle). IF: interfascicular nucleus IPC: interpeduncular nucleus caudal IPDM: interpeduncular nucleus dorsomedial ml: medial lemniscus MM: medial mammillary nucleus MnR: median raphe nucleus Pn: pontine nucleus PnO: pontine reticular nucleus oral RMC: red nucleus rs: rubrospinal tract SNC: substantia nigra compact SNR: substantia nigra reticular VTA: ventral tegmental area xscp: decussation of the superior cerebellar penduncle (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).