Endocrine effects of grouping and dominance rank in squirrel monkeys

Social setting, social rank and HPA axis response in cynomolgus monkeys

RATIONALE

Hypothalamic-pituitary-adrenal (HPA) axis activity under different social settings in non-human primates is understudied.

OBJECTIVE

The aim of this study is to evaluate the response of pituitary-adrenal hormones (adrenocorticotropic hormone (ACTH) and cortisol) to pharmacological challenges of the HPA axis in male cynomolgus macaques under different social settings.

METHODS

Male cynomolgus macaques (Macaca fascicularis, n = 11) were individually (A) and socially housed (B) in alternation, over consecutive months, in an ABA design. During each experimental phase, plasma ACTH and cortisol were measured in response to low- and mild-intensity psychological stressors and following administration of saline, naloxone, ovine-corticotropin-releasing factor (oCRF), and dexamethasone.

RESULTS

These data demonstrate that cortisol measured under low stress conditions is sensitive to social rank (dominance hierarchy) and distinguishes dominant from non-dominant animals during both individual and social settings. Administration of naloxone resulted in elevated circulating ACTH and cortisol, while oCRF only increased circulating cortisol. During social housing, the cortisol response to naloxone and oCRF was increased, whereas dexamethasone suppression of ACTH and cortisol remained consistent across all social settings.

CONCLUSIONS

Circulating ACTH and cortisol are differentially sensitive to changes in social settings in non-human primates. Cortisol response increased during social housing and could be stimulated by both naloxone and oCRF, whereas ACTH response was generally not influenced by social setting or oCRF but was increased by naloxone. These data show differential adrenal and pituitary response to changes in social settings and a small, but consistent, effect of social dominance.

Dominance: The baby and the bathwater

The concept of dominance is used in the behavioral and biological sciences to describe outcomes in a variety of competitive interactions. In some taxa, a history of agonistic encounters among individuals modifies the course of future agonistic encounters such that the existence of a certain type of relationship can be inferred. If one is to characterize such relationships as dominance, however, then they must be distinguished from other kinds of interaction patterns for which the term tends to be used, as well as from factors such as territoriality and "trained" winners and losers, which may also influence the expression of agonistic behavior. Operational definitions based on causal, functional, evolutionary, and ontogenetic considerations have been proposed. Reliability and validity problems have been discussed, but the dominance concept has proved useful despite methodological difficulties. The confusion of dominance relationships (which involve two or more individuals) with dominance ranks (which are assigned to a single individual) has obscured the possible evolutionary basis of dominance relationships. If benefits accrue to dominant members of pairs, then those attributes which allow an animal to establish dominance can be selected. Dominance per se and dominance ranks, on the other hand, cannot be genetically transmitted since they constitute relationships with other individuals rather than absolute attributes. Dominance rankings in particular may be useful for describing behavioral patterns within a group, but they may reflect our own ability to count rather than any important variable in social organization.

Relationship between social rank and cortisol and testosterone concentrations in male cynomolgus monkeys (Macaca fascicularis)

In nonhuman primate social groups, biological differences related to social status have proven useful for investigating the mechanisms of sensitivity to various disease states. Physiological and neurobiological differences between dominant and subordinate monkeys have been interpreted in the context of chronic social stress. The present experiments were designed to investigate the relationships between basal cortisol and testosterone concentrations and the establishment and maintenance of the social hierarchy in male cynomolgus monkeys. Cortisol concentrations were measured at baseline and following suppression with dexamethasone (DEX) and subsequent administration of adrenocorticotrophic hormone (ACTH) while monkeys were individually housed (n = 20) and after 3 months of social housing (n = 4/group), by which time dominance hierarchies had stabilised. Cortisol was also measured during the initial 3 days of social housing. Neither pre-social housing hormone concentrations, nor hypothalamic-pituitary-adrenal (HPA) axis sensitivity predicted eventual social rank. During initial social housing, cortisol concentrations were significantly higher in monkeys that eventually became subordinate; this effect dissipated within 3 days. During the 12 weeks of social housing, aggressive and submissive behaviours were observed consistently, forming the basis for assignment of social ranks. At this time, basal testosterone and cortisol concentrations were significantly higher in dominant monkeys and, after DEX suppression, cortisol release in response to a challenge injection of ACTH was significantly greater in subordinates. These results indicate that basal cortisol and testosterone concentrations and HPA axis function are state variables that differentially reflect position in the dominance hierarchy, rather than trait variables that predict future social status.

Behavioural and physiological aspects of stress and aggression in nonhuman primates

There is considerable interest in the study of stress and aggression in primates as a model for their interpretation in humans. Despite methodological and interpretational problems associated with behavioural and physiological measurement and definition, a considerable body of literature exists on these phenomena in primates. In the course of reviewing this literature we examine examples of many of the sources of variation in stress and aggression, including species identity, sex, age, breeding and social status, individual temperament, background, learning and resource distribution. This is followed by an examination of the interaction between stress and aggression before reviewing the most important areas in which changes in both stress and aggression are measured. In particular we examine those studies covering social aspects of an animal's life, specifically relating to social isolation, crowding as well as group formation, composition and instability. This review reveals the complex and often contradictory nature of relationships, not just between an animal's physiology and its behaviour, but between its stress status and display or receipt of aggression.

Predictors of social status in cynomolgus monkeys (Macaca fascicularis) after group formation

The purpose of the present study was to determine whether various behavioral and hormonal markers obtained in individually housed monkeys would be predictive of social rank following group housing. Body weight, serum cortisol and testosterone levels, and locomotor activity in an open-field apparatus were examined in 20 experimentally naive male cynomolgus monkeys (Macaca fascicularis) while they were individually housed. It was hypothesized that eventual subordinate monkeys would have higher cortisol levels and increased locomotor activity scores. These monkeys were then placed in social groups of four (five pens of four monkeys), and social rank was determined based on outcomes of dyadic agonistic encounters. Body weight correlated significantly with eventual social rank. In general, the heavier the monkey the higher the social rank. Locomotor activity in an open-field apparatus following administration of a low dose of cocaine (0.01 mg/kg, i.v.), which has been shown to increase CNS dopamine, correlated with eventual social rank such that individually housed monkeys with high levels of locomotion were more likely to become subordinate. Serum cortisol and testosterone levels failed to correlate with eventual social rank. Hypothalamic-pituitary feedback sensitivity and adrenal responsiveness were examined by measuring cortisol levels after administration of dexamethasone and following ACTH challenge. Cortisol responses in these tests were not associated with eventual social rank. These results suggest that, in addition to body weight, the level of reactivity in a novel environment after administration of a low dose of cocaine is a potential trait marker for social rank. This trait is apparently not associated with hormone levels, but may involve other CNS mechanisms.

Behavioural, physiological and immunological consequences of social status and aggression in chronically coexisting resident-intruder dyads of male rats

The behavioural and physiological consequences of social status and reciprocal fighting in resident-intruder dyads of Long Evans male rats were evaluated. Before a chronic cohabitation of 10 days, residents and intruders were individually housed for one month to increase their aggressiveness. Control animals included isolates, i.e., animals kept individually housed throughout the experiment and pair-housed rats, i.e., pairs of rats housed together from their rats in the laboratory. In 19 out of 20 dyads, a clear dominance relationship developed with an advantage to the resident in 68% of the cases. Dominants showed more exploratory activity than subordinates in a open-field test at the end of the cohabitation period; subordinates groomed longer than animals from other experimental groups. Dominants had lower pain thresholds than individually and pair-housed animals. Both dominants and subordinates had higher tyrosine hydroxylase enzymatic activities in the left adrenal than isolated and pair-housed rats. Subordinates lost body weight and had higher plasma corticosteroid concentrations than animals from the other experimental groups. In addition, they had smaller thymus glands and reduced spleen lymphocyte responses to mitogenic stimulation in vitro, in comparison to dominant animals. These results show that subordination in the dyadic resident-intruder paradigm leads to a complex syndrome of behavioural and physiological changes, some of which may be modulated by the intensity of aggressive interactions.

Social influences on circulating levels of cortisol and prolactin in male talapoin monkeys

Three captive groups of adult talapoin monkeys (Miopithecus talapoin), each with four males and four or five ovariectomized, estrogen-treated females, were observed. Behavioral interactions were recorded and levels of cortisol and prolactin were measured. In each group, males formed a linear dominance order, determined by the direction of spontaneous aggression; highest male rank was associated with frequent socio-sexual interactions and lowest rank with infrequent socio-sexual behavior. The first study compared males' cortisol and prolactin titers when all males were either housed with estrogen-treated females, or singly caged. The two lowest-ranking males of each group had elevated cortisol levels when group-housed; prolactin levels did not reflect these changes. In the second study, males of one group interacted with females when only the lowest-ranking, or the highest-ranking, female was made attractive (i.e., received estradiol), while the other females were present, but not estradiol-treated. Across these conditions (1) frequencies of some socio-sexual and aggressive behaviors changed, (2) cortisol levels increased in all males, and (3) prolactin levels decreased in three of four males. Dissociation of changes in cortisol and prolactin titers suggests that these hormones may be differentially responsive to social modifications, not simply reflecting a single intervening variable, such as "stress."

Effects of rank and novel females on behaviour and hormones in male talapoin monkeys

Two social groups of captive talapoin monkeys (Miopithecus talapoin), each with three intact adult males and three or four ovariectomized, estrogen-treated adult females, were observed. Socio-sexual and aggressive interactions were recorded, and levels of testosterone, cortisol, and prolactin were measured in male serum. Cortisol and prolactin titers did not reflect male rank, nor did changes in one of these hormones parallel changes in the other. In both groups males formed a linear dominance hierarchy, defined in terms of the direction of aggression among animals. Highest male rank was associated with frequent socio-sexual interaction and elevated testosterone levels, even in the absence of ejaculations. Males in one group copulated with females, but no male in the other group copulated. All males of each group were then housed with all females of the other group which resulted in marked changes in males' behaviour; socio-sexual interactions increased in C group males and decreased in B group males. This illustrates the influence that females may have on male behaviour in general, and in particular outlines the potential for female dominance and the consequences this may have especially on reproductive behaviour.

Cortisol responses under different housing conditions in female squirrel monkeys

Adult female squirrel monkeys were housed in a group, in pairs or individually. Plasma levels of cortisol obtained under basal and stress conditions once weekly for four weeks were significantly lower in pair-housed females than in those living in a social group or individually. The increment in cortisol levels after stress (induced by handling and ether anesthesia) also was smaller in females housed in pairs. The cortisol values of the pair-housed females were positively correlated with those of their partners. Basal cortisol levels in the group-living females showed a significant rank-order correlation with dominance status. This indicated that social interactions in group-living animals can influence cortisol levels in a complex manner, either increasing or decreasing them. The relatively lower pituitary--adrenal activation when a single partner was present also indicated that the social environment can affect an individual's general level of arousal and subsequently alter the response to stressful stimuli.

Pairwise Dominance and Social Preference in Squirrel Monkeys

5 male and 5 female squirrel monkeys, previously dominance tested, were observed in a noncompetitive social preference situation. Following nonsocial adaptation trials, all possible pairs of monkeys were observed, one pairmate serving as S and the other as a stimulus animal in a free-choice preference test. Ss' latencies to enter the apparatus and activity levels were greatly reduced by the addition of social partners. While social choices were largely uninfluenced by the dominance of the stimulus animal, the sex of the stimulus monkey proved to be an effective variable in that females elicited more approach behaviors than males. These results are consistent with previous studies of squirrel monkeys' preferences measured in noncompetitive social situations.