ronald r. swaisgood, megan a. owen, nancy m. czekala, nathalie mauroo, KathY hawk, jason c. l. tang introduction
Giant pandas are being maintained in captivity largely for the purpose of creating a reproductively viable population that will support conservation of the species in nature. Toward this end, researchers and managers have targeted many aspects of husbandry for improvement through scientific investigations. Among the many priorities is the ability to measure 'well-being' and possibly alleviate 'stress' imposed by a captive environment. Stress research has been increasingly incorporated into captive wildlife breeding programmes, in part because it is widely believed that small enclosures may not allow animals to execute normal escape and avoidance responses to aversive stimuli. Coping mechanisms may be constrained, thus resulting in stress that can compromise psychological and physiological health, including reproduction (Carlstead & Shepherdson, 2000). Among the many deleterious consequences, stress compromises immune function, reproduction, pregnancy sustainability and maternal care (Munck et al., 1984; Baker et al., 1996; Carlstead, 1996; Moberg & Mench, 2000).
How susceptible is the giant panda to stress imposed by ex situ environments? The charisma of this species causes it to attract large
Giant Pandas: Biology, Veterinary Medicine and Management, ed. David E. Wildt, Anju Zhang, Hemin Zhang, Donald L. Janssen and Susie Ellis. Published by Cambridge University Press. # Cambridge University Press 2006.
and noisy crowds. Also, giant pandas are commonly held at major institutions that often undertake large construction projects. This chapter deals with the sensitivity of the giant panda to its captive environment. Stress, more than other biological concepts, has limited utility at the population level. In a single species, however, individual animals seem to vary remarkably in response to environmental change. Thus, we have become advocates for assessing stress on the basis of individuals, and not only for minimising it but to identify those frequently occurring or chronic factors to which an animal fails to habituate. Whether due to suboptimal enclosure design, husbandry practices or disturbances, these variables need to be thoroughly understood and then modified so that an optimal environment can be created for that individual. Our 'individual assessment approach' across many giant pandas living in diverse institutions has been useful for developing a list of idiosyncratic stress responses that potentially may well impact on the species or the individual. We suggest that searching out common threads among these factors post hoc will provide a more comprehensive understanding of stress responsiveness in this species while helping to manage this phenomenon to promote well-being, health and reproduction.
WHAT is STRESS AND HOW IS IT MEASURED?
'Stress' is not a well-defined concept (Hofer & East, 1998; Moberg & Mench, 2000; Sapolsky et al., 2000). Most contemporary definitions involve the animal's perception of a threat that challenges internal homeostasis (both motivational and physiological 'set points'), and behavioural and physiological adjustments that the animal undergoes to avoid or adapt to a 'stressor' and return to homeostasis.
Most stress is acute, short-lived and harmless. However, if a perceived threat persists and the animal fails to adapt, chronic stress may develop, homeostasis is not achieved, and hyperadrenal activity results in sustained and problematically elevated glucocorticoid concentrations (Mendoza et al, 2000). Hereafter, we use the term 'stress' loosely as a descriptive concept involving the behavioural and physiological reactions to external threats to homeostasis (stressors) (Selye, 1956). It is generally inferred that an animal's subjective experience is negative during protracted stress. In this light, avoidance or minimisation of excessive stress may have implications for psychological and physiological well-being, including health and reproduction.
To understand stress and its implications for individuals, both the stressor and the type of reactive response must be studied. Environmental challenges can be recorded systematically or manipulated experimentally. If their occurrence is associated with a measurable stress response, then one can conclude that they are biologically salient stressors. However, responses can be remarkably variable. As Hofer & East (1998) point out:
if individuals vary in the extent to which hormonal, resource allocation, immunological and behavioural systems are triggered. . ., and if these systems follow different time courses. . ., then measuring only part of the response may provide incomplete information about the magnitude and consequences of the full response.
Thus, stress studies must examine multiple indices (Cook et al.,
Hypothalamic-pituitary-adrenal (HPA) activity is one of the most useful measures of stress, in part because it can be monitored non-invasively via metabolites excreted in urine or faeces (Wasser et al., 2000). When an animal perceives a threatening stimulus, a relatively predictable series of physiological adjustments, known as the General Adaptation Syndrome, are activated (Selye, 1946). The sympathetic-adrenal response involves epinephrine and norepinephrine secretion, which increases heart rate and mobilises energy reserves for use by the muscles. This fast-acting response is followed by more long-term HPA activation, resulting in glucocorticoid release by the adrenal cortex, which ultimately can be quantified by increased amounts of metabolites in urine or faeces (Sapolsky et al., 2000; see also Chapter 8). A primary function of glucocorticoids is to shut down components of the stress response in a negative feedback loop to keep it from escalating out of control. Short-term HPA activation is adaptive, but prolonged glucocorticoid production can have pathological effects, including immunological and reproductive suppression, disease and even death (Carlstead & Shepherdson, 1994; 2000).
Stressors do not always activate the HPA system, and changes in glucocorticoids can occur for non-stress-related reasons, for example, due to seasonality, in support of reproductive activity or to mobilise energy reserves (Walker et al, 1992; Altemus et al, 1995; Hofer & East 1998; Romero & Wikelski, 2002). Indeed, our data show that glucocorticoid concentration fluctuate seasonally in the great panda, potentially obfuscating or confounding HPA activity related to stress (Owen et al., in press). Thus, there is not always a clean cause and effect. For example, one recent study of cheetahs demonstrated that paired females experienced evidence of behavioural stress and suppressed ovarian activity, but no elevated corticoid excretion (Wielebnowski et al., 2002). In cattle, individuals that appeared to habituate physiologically (in terms of HPA activity) still showed behavioural distress (Cook et al., 2000). And, in nonhuman primates, chronic stress - or failure to restore homeostasis - has been associated with reduced corticoids (Mendoza et al., 2000). Therefore the accurate measurement of a response to stress must go beyond simply tracking fluctuations in adrenal hormone activity (Cook et al., 2000).
One other commonly used and viable index is behaviour, an easily measured, sensitive indicator of an animal's perception of environmental change (Weary & Fraser, 1995; Baker & Aureli, 1997; Boinski et al, 1999; Swaisgood et al., 2001; Wielebnowski et al., 2002). Altered behaviours to a stressor may include stimulus avoidance, formation of stereotypies and changed activity levels associated with feeding, exploratory and sexual behaviours (Mason, 1991; Carlstead & Shepherd-son, 2000). However, behavioural signs of stress may also be unreliable because inappropriate behaviours may be selected for measurement, they may be subtle or correlate poorly with physiological indicators, or they may reflect a permanent adaptive response to an environmental change (Hofer & East, 1998; Cook et al., 2000). One frequently cited argument for the unreliability of behaviour is the dichotomous nature of the behavioural component to the stress response - the animal may show heightened activity suggestive of agitation and escape motivation (fight-flight response) or, alternatively, withdraw, hide and become sluggish (conserve-withdraw response) (Moberg, 1985). These differing response modes do not invalidate each other. On the contrary, they highlight the importance of examining species-specific or even individual responses to stressors rather than expecting universal monotypy.
We have found that behaviour is a valuable window into the perceptual processes of the animal, useful for inferring if its state has changed to address a specific environmental challenge. This deviation from 'behavioural homeostasis' can offer insight into the animal's well-being, especially given that a suite of behavioural variables is evaluated in concert. Cook et al. (2000) have suggested that behavioural indices illustrative of a stress response include measures of the startle or defence response, the time required to resume normal activity after a stress, aggression, stereotypy, lack of responsiveness or apathy and decreased complexity of behaviour. Assessments can also be carried out more experimentally, for example, by giving an animal a choice between two environments, one presumably less evocative of stress than the other. Better yet, one could determine how much effort an animal is prepared to expend to gain access to a less stressful environment.
Was this article helpful?