Elsevier

Hormones and Behavior

Volume 43, Issue 1, January 2003, Pages 67-82
Hormones and Behavior

Regular article
Are subordinates always stressed? a comparative analysis of rank differences in cortisol levels among primates

https://doi.org/10.1016/S0018-506X(02)00037-5Get rights and content

Abstract

Among primate species there is pronounced variation in the relationship between social status and measures of stress physiology. An informal meta-analysis was designed to investigate the basis of this diversity across different primate societies. Species were included only if a substantial amount of published information was available regarding both social behavior and rank-related differences in stress physiology. Four Old World and three New World species met these criteria, including societies varying from small-group, singular cooperative breeders (common marmoset and cotton top tamarin) to large-troop, multi-male, multi-female polygynous mating systems (rhesus, cynomolgus, talapoin, squirrel monkeys, and olive baboon). A questionnaire was formulated to obtain information necessary to characterize the stress milieu for individuals in particular primate societies. We standardized cortisol values within each species by calculating the ratio of basal cortisol concentrations of subordinates to those of dominants in stable dominance hierarchies and expressing the ratio as a percentage (relative cortisol levels). The meta-analysis identified two variables that significantly predicted relative cortisol levels: subordinates exhibited higher relative cortisol levels when they (1) were subjected to higher rates of stressors, and (2) experienced decreased opportunities for social (including close kin) support. These findings have important implications for understanding the different physiological consequences of dominant and subordinate social status across primate societies and how social rank may differ in its behavioral and physiological manifestations among primate societies.

Introduction

In mammals, including humans, challenges to homeostasis commonly evoke a series of endocrine and neural actions known as the stress response. While not as stereotyped as once thought, the stress response typically involves release of catecholamines (norepinephrine and epinephrine) from the sympathetic nervous system and adrenal medulla, release of adrenocortical steroid hormones (glucocorticoids), and initiation of a variety of other endocrine responses, including suppression of hormones related to anabolism, growth, and reproduction Orth and Kovacs 1998, Sapolsky et al., 2000, Sapolsky 2002. Collectively, the stress response aids in adapting an individual to an acute stressor by stimulating hepatic glucose release and visceral lipolysis, enhancing delivery (via enhanced cardiovascular tone) of glucose, fatty acids, and triglycerides to skeletal muscle and brain, triaging processes nonessential to immediate survival (such as growth, digestion, reproduction, and tissue repair), and constraining inflammatory and immune responses. Despite these adaptive functions, it has long been recognized that chronic or prolonged activation of the stress response can have deleterious physiological effects, including the induction or worsening of hypertension, insulin-resistant (type II) diabetes, ulceration in the gastrointestinal tract, anovulation, impotence, osteoporosis, psychogenic dwarfism, and immunosuppression (Sapolsky, 2002).

Because of the pathological consequences of a prolonged stress response, much attention has focused on individual differences in patterns of secretion of stress response hormones. These can include differences in blood hormone levels prior to and during the stress response, the speed of physiological recovery of homeostasis after the stress response abates, or the ease with which different types of stressors differentially provoke separate components of the stress response. As examples of such differences, a number of psychiatric disorders and personality types have been associated with what can arguably be termed a maladaptive and overactive stress response. These include major depression (Holsboer, 1999), repressive personality (Brandtstadter et al., 1991), and Type A personality (Williams, 1989). Differential fetal programming (Seckl, 2001) and endocrine disruption of pre- and perinatal environments (Clarke et al., 1994) can result in further exaggeration of adult stress responses.

Another branch of psychoneuroendocrine research has focused on ways in which individual differences in the stress response can reflect differences in dominance status among social animals. In some species, subordinance has been reported to be associated with a chronically overactive stress response (as assessed by hypersecretion of glucocorticoids or catecholamines, higher blood pressure, and greater incidences of stress-related pathologies). This pattern is thought to reflect the classical picture of dominance hierarchies as linear “pecking orders” in which resources are unevenly distributed, inequalities are maintained through aggression and intimidation, and subordinates are subject to the most severe resource limitations, the fewest opportunities for coping, and the greatest physical and social stressors (Albeck et al., 1997).

The relationship between social rank and patterns of the stress response is not consistent among primates in that there is no monolithic relationship between social status and any aspect of stress physiology Abbott et al., 1997, Clarke and Boinski 1995, Sapolsky 1993. This diversity appears to reflect, among other things, the extraordinary variety of primate social systems, which often constitute striking exceptions to hierarchical pecking orders Dunbar 1988, Strier 1996, de Waal 1989 and include complex, affiliative relationships that ameliorate responses to stressors, such as through reconciliation de Waal 1993, de Waal 2000, providing different forms of coping mechanisms to individuals (Bercovitch, 1991).

The purpose of the meta-analysis performed in this study is to identify the social variables that predict primate societies in which (1) subordinates have overactive stress responses, (2) subordinates have diminished stress responses, or (3) stress responses do not vary with social rank. More broadly, we are attempting to understand the social roles of individuals, in terms of their rank, sex, and species, and the relationship of such roles in different primate species to differences in the stress response.

In order to identify primate societies that give rise to differing relationships between rank and physiology, it is important to first review factors that stimulate or ameliorate the stress response.

One factor is the frequency at which an individual is exposed to physical stressors. As such, it is important to consider rank-related differences in (1) availability of food and the effort needed to acquire it, (2) exposure to pathogens, (3) likelihood of being threatened by a predator, and (4) the incidence of aggressive challenge and injury.

Of at least equal importance is the rate at which an individual is exposed to psychological stressors. An extensive and elegant literature has demonstrated that psychological variables can stimulate the stress response outright and can dramatically modulate the response to physical stressors Bakshi and Kalin 2000, Levine et al., 1989, Abbott et al., 1997, McEwen 2001, Sapolsky et al., 2000. Broadly, for the same physical insult, individuals are far more likely to have a stress response and a stress-related disease if they lack a sense of control or lack predictive information about the timing, severity, or duration of the stressor. It thus becomes important to understand whether there are rank-related differences in the extent to which individuals (1) can control access to resources, (2) are subject to aggression, and (3) can establish stable and predictable social relations.

Finally, the response to physical or social stressors is considerably blunted by the availability of coping responses, such as social outlets for frustration or social support (Sapolsky, 2002). Thus, rank-related differences in the availability of social contact, social grooming, sexual behavior, and frustration displacement of aggression become as relevant as low rank and the degree of accompanying stressors.

Section snippets

Questionnaire

Based upon the known physical and psychological modulators of the stress response, the first and last authors formulated a questionnaire containing key questions that were designed to obtain information necessary to characterize the stress milieu for individuals in particular primate societies. More broadly, we attempted to ask questions that would formalize the characteristics of subordinate and dominant individuals—what stressors and what sources of coping are available—for each primate

Results

Basal cortisol concentrations of subordinate individuals in stable dominance hierarchies, as compared to those of their dominant counterparts, show considerable variability among primate species, with relative cortisol levels ranging from 45% (subordinates lower than dominants) to 154% (subordinates higher than dominants; Table 1). Species in which cortisol concentrations in subordinates are lower than those of dominants (relative cortisol levels = 45–82%) include female common marmosets Abbott

Discussion

In some species, social subordinance is associated with a more pathologic, hypercortisolemic profile than in dominant animals (e.g., Blanchard et al., 1993, Fuchs and Flugge 2002. There is no consistent relationship, however, between social rank and stress response among primates Abbott et al., 1997, Sapolsky 1993. Results from the present study suggest that the different relationships between social rank and stress response among primate species may be explained by crucial differences in

Conclusion

We have identified two social variables in primate societies (Questions 2A and 2B; Table 2) that consistently predict whether subordinate or dominant animals exhibit elevated basal cortisol levels (hypercortisolism). Our findings have important implications for understanding the different physiological consequences of low social rank in different primate societies:

  • (1)

    “Rank” is not uniform in its social meaning across different primate species, and some investigators have questioned the utility of

Acknowledgements

This paper was written with support from the following: NSF Grants IBN-9604321 (D.H.A. and W.S.), DEB-0196384 (T.G.); NIH Grants RR00167 (D.H.A. and W.S.), MH60728 (W.S. and D.H.A.), MH53709 and HD07678 (W.S.), RR03640 (F.B.B.), MH35215 (C.T.S. and T.E.Z.); an MRC Programme Grant (E.B.K.); The Harry Frank Guggenheim Foundation (F.B.B. and R.M.S.); and The Templeton Foundation (R.M.S.). This is publication No. 42–007 of the Wisconsin Primate Research Center.

References (108)

  • C.L. Coe et al.

    Social status constrains the stress response in the squirrel monkey

    Physiol. Behav.

    (1979)
  • J.A. Eberhart et al.

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

    Physiol. Behav.

    (1983)
  • J.A. Eberhart et al.

    Subordination in male talapoin monkeys lowers sexual behaviour in the absence of dominants

    Physiol. Behav.

    (1985)
  • E. Fuchs et al.

    Social stress in tree shrewsEffects on physiology, brain function, and behavior of subordinate individuals

    Pharmacol. Biochem. Behav.

    (2002)
  • A.J. Ginther et al.

    Reproductive biology of captive male cottontop tamarin monkeys as a function of social environment

    Anim. Behav.

    (2001)
  • M.S. Golub et al.

    Plasma cortisol levels and dominance in peer groups of rhesus monkey weanlings

    Horm. Behav.

    (1979)
  • A.R. Hermus et al.

    Cytokines and the hypothalamic–pituitary–adrenal axis

    J. Steroid Biochem. Mol. Biol.

    (1990)
  • E.O. Johnson et al.

    The biobehavioral consequences of psychogenic stress in a small, social primate (Callithrix jacchus jacchus)

    Biol. Psychiatry

    (1996)
  • P.G. Judge et al.

    Rhesus monkey behaviour under diverse population densitiescoping with long-term crowding

    Anim. Behav.

    (1997)
  • E.B. Keverne et al.

    Social influences on sex differences in the behavior and endocrine state of talapoin monkeys

    Prog. Brain Res.

    (1984)
  • C. Lazaro-Perea

    Intergroup interactions in wild common marmosets Callithrix jacchusterritorial defence and assessment of neighbours

    Anim. Behav.

    (2001)
  • S.P. Mendoza et al.

    Breeding readiness in squirrel monkeysfemale-primed females are triggered by males

    Physiol. Behav.

    (1991)
  • M.D. Pagel

    A method for the analysis of comparative data

    J. Theor. Biol.

    (1992)
  • W. Saltzman et al.

    Sociophysiology of relationships in squirrel monkeys I. Formation of female dyads

    Physiol. Behav.

    (1991)
  • W. Saltzman et al.

    Social and reproductive influences on plasma cortisol in female marmoset monkeys

    Physiol. Behav.

    (1994)
  • W. Saltzman et al.

    Behavioural and endocrine predictors of dominance and tolerance among female common marmosets (Callithrix jacchus)

    Anim. Behav.

    (1996)
  • W. Saltzman et al.

    Suppression of cortisol levels in subordinate female marmosetsreproductive and social contributions

    Horm. Behav.

    (1998)
  • W. Saltzman et al.

    Reduced adrenocortical responsiveness to ACTH stimulation in female marmoset monkeysinfluences of social rank and reproductive hormones

    Psychoneuroendocrinology

    (2000)
  • R.M. Sapolsky

    The endocrine stress-response and social status in the wild baboon

    Horm. Behav.

    (1982)
  • R.M. Sapolsky

    Adrenocortical function, social rank and personality among wild baboons.

    Biol. Psychiatry

    (1990)
  • J.G. Scammell et al.

    Overexpression of the FK506-binding immunophilin FKBP51 is the common cause of glucocorticoid resistance in three New World primates

    Gen. Comp. Endocrinol.

    (2001)
  • J.R. Seckl

    Glucocorticoid programming of the fetus; adult phenotypes and molecular mechanisms

    Mol. Cell. Endocrinol.

    (2001)
  • C.A. Shively et al.

    Behavior and physiology of social stress and depression in female cynomolgus monkeys

    Biol. Psychiatry

    (1997)
  • C.A. Shively

    Social subordination stress, behavior, and central monoaminergic function in female cynomolgus monkeys

    Biol. Psychiatry

    (1998)
  • D.H. Abbott

    Behavioral and physiological suppression of fertility in subordinate marmoset monkeys

    Am. J. Primatol.

    (1984)
  • D.H. Abbott

    Behaviorally mediated suppression of reproduction in female primates

    J. Zool. (Lond.)

    (1987)
  • D.H. Abbott et al.

    Specific neuro-endocrine mechanisms not involving generalized stress mediate social regulation of female reproduction in cooperatively breeding marmoset monkeys

    Ann. NY Acad. Sci.

    (1997)
  • D.S. Albeck et al.

    Chronic social stress alters levels of corticotropin-releasing factor and arginine vasopressin mRNA in rat brain

    J. Neurosci.

    (1997)
  • A.J. Baker et al.

    Immigration in wild groups of golden lion tamarins (Leontopithecus rosalia)

    Am. J. Primatol.

    (1996)
  • Blomberg, S.P., Garland Jr., T., 2002. Tempo and mode in evolution: phylogenetic inertia, adaptation, and comparative...
  • S. Boinski

    Geographic variation in behavior of a primate taxonstress responses as a proximate mechanism in the evolution of social behavior

  • D.D. Brandon et al.

    Glucocorticoid resistance in humans and nonhuman primates

    Cancer Res.

    (1989)
  • A.S. Clarke et al.

    Temperament in nonhuman primates

    Am. J. Primatol.

    (1995)
  • A.S. Clarke et al.

    Long-term effects of prenatal stress on HPA axis activity in juvenile rhesus monkeys

    Dev. Psychobiol.

    (1994)
  • C.L. Coe et al.

    Rapid cortisol and corticosteroid-binding globulin responses during pregnancy and after estrogen administration in the squirrel monkey

    J. Endocrinol.

    (1986)
  • C.L. Coe et al.

    Phylogenetic influences on hormone levels across the Primate order

    Am. J. Primatol.

    (1992)
  • R.J. Colman et al.

    Absence of estrogen depletion bone loss in female common marmosets

    J. Bone Mineral Res.

    (1997)
  • W.B. Denny et al.

    Squirrel monkey immunophilin FKBP51 is a potent inhibitor of glucocorticoid receptor binding

    Endocrinology

    (2000)
  • L.J. Digby

    Sexual behavior and extragroup copulations in a wild population of common marmosets (Callithrix jacchus)

    Folia Primatol

    (1999)
  • R.I.M. Dunbar
    (1988)
  • Cited by (549)

    View all citing articles on Scopus
    View full text