Nocturnal hyperthermia induced by social stress in male tree shrews: Relation to low testosterone and effects of age

Abstract

Stress is known to elevate core body temperature (CBT). We recorded CBT in a diurnal animal, the male tree shrew, during a one-week control period and a one-week period of social stress using a telemetric system. During the stress period, when animals were confronted with a dominant male for about 1 h daily, CBT was increased throughout the day. We analyzed CBT during the night when animals were left undisturbed and displayed no locomotor activity. To determine whether nocturnal hyperthermia may be related to stress-induced changes in hormonal status, we measured testosterone, noradrenalin and cortisol in the animals' morning urine.

The daily social stress increased the mean nocturnal temperature by 0.37 °C. Urinary testosterone was reduced during the stress period, and there was a significant negative correlation between testosterone and the area under the curve (AUC) of the nocturnal CBT. This means that stress-induced hyperthermia was strongest in the animals with the lowest testosterone concentrations. As expected, urinary noradrenalin was elevated during the stress week but a positive correlation with the AUC data was only found for animals younger than 12 months. Cortisol was also increased during the stress week but there were no correlations with nocturnal hyperthermia. However, the stress-induced increases in noradrenalin and cortisol correlated with each other. Furthermore, there were no correlations between the stress-induced increase in nocturnal CBT and body weight reduction or locomotor activity during the light phase.

Interestingly, the extent of nocturnal hyperthermia depended on the animals' ages: In animals younger than 12 months, stress increased the AUC by 48%, in animals aged between 12 and 24 months, stress increased the AUC by 36%, and older animals showed only a 7% increase. However, testosterone was not significantly reduced in the older animals.

The present data reveal an interrelation between the extent of stress-induced nocturnal hyperthermia, the animals' gonadal hormone status and their ages. The negative correlation between hyperthermia and testosterone indicates that this hormone in particular plays an important role in the regulation of body temperature in male tree shrews.

Introduction

Stress is known to induce a general activation of an individual's metabolism and to raise core body temperature (CBT). In rodents, a transient temperature rise of about 2 °C can even be observed after a minor challenge such as an injection of a vehicle solution [1], and in humans, psychological stress may evoke low-grade idiopathic fever [2]. Elevations in nocturnal temperature have been observed in humans during periods of depression [3]. Hyperthermia induced by emotional stress can be long-lasting as was e.g. shown in male rats where a single social defeat by a dominant male elevated body temperature for more than four days [4]. A recent study, also conducted on male rats, revealed that following three weeks with daily social defeat during the light phase, nocturnal body temperature was elevated even after a recovery period of one week [5]. We studied hyperthermia induced by daily social defeat in a day-active animal, the male tree shrew. Because CBT might be influenced by activity, we compared only nocturnal temperature to other stress-modulated parameters; tree shrews rest and display no relevant locomotor activity during the entire dark phase [6], [7].

It is generally accepted that a persistent CBT elevation evoked by emotional stress is related to activation of emotional brain centers which regulate the activity of autonomic systems [8], [9]. The sympathetic nervous system is strongly activated by social encounters that are regarded as strong stressors [10], [11], [12]. Among the stress-activated systems is the hypothalamus–pituitary–adrenal (HPA) axis [13], and stress-induced hyperthermia is accompanied by HPA axis hyperactivity [1]. Therefore, in the present study, we investigated whether there is a correlation between stress-induced nocturnal hyperthermia and the activation of the sympathetic nervous system and/or the HPA axis in male tree shrews. In this species, high levels of noradrenalin and cortisol in the morning urine reflect hyperactivity of those systems [14], [15].

Furthermore, in male tree shrews, social stress leads to testicular atrophy and thus to reduced urinary testosterone [15], [16]. It has been proposed that this gonadal steroid plays a role in the consolidation of circadian activity rhythms in male animals [17]. We therefore analyzed whether there might be a correlation between the stress-evoked nocturnal hyperthermia and changes in testosterone. In addition, we tested whether the elevated temperature correlated with basal locomotor activity, because it has been reported that general physical activity of an individual influences CBT [6].

The data presented here were derived from eight different experiments in which adult male tree shrews were submitted to daily social defeat according to an established and validated social stress paradigm [15]. We studied a total of 43 subordinate animals during a one-week control period and a subsequent stress week of daily confrontations with a dominant male. Nocturnal CBT was recorded by telemetry between 8.00 p.m. and 8.00 a.m. the following morning, and the degree of stress-induced hyperthermia was evaluated by analyzing the areas under the temperature curves. Moreover, the CBT decline between 8 p.m. and 10 p.m. and the time points of the lowest nocturnal temperature (nadir of the temperature curve) were determined. Individual levels of noradrenalin, cortisol and testosterone were determined in morning urine, and locomotor activity was recorded daily for 1 h in the late afternoon (between 4.30 p.m. and 5.30 p.m.), which is a time period when tree shrews are awake and active but not disturbed by ongoing activities in the animal facility. Correlation analyses were performed, and we grouped the animals according to their ages to detect potential age-related effects.