1 Department of Psyehopharmaeology, Utrecht Institute of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16, 3584CA Utrecht, The Netherlands 2Rudolf Magnus Institute of Neuroscience, University Medical Centre Utrecht, Utrecht, The Netherlands 3Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA
Abstract: When animals are confronted with a stressor, they respond by an extensive stress response, amongst which a rise in body temperature is prominent. This stress-induced hyperthermia (SIH) is a rapid response reaching a maximum within 10-15 min after the start of the stress-inducing stimulus. In mice, the amplitude and duration of SIH seem to depend on the intensity of the stressor: a short-lasting stress like measuring the rectal body temperature or injection of a drug leads to an SIH of maximally 1-1.5°C and lasts about 45 min, whereas a novel cage stress enhances body temperature 2-2.5°C and has a longer duration. However, large strain differences exist and intrinsic differences in basal core body temperature over the day are present between strains (C57BL/6J, 129SvEv, Swiss-Webster). There is a dispute in the literature whether SIH is a (emotional) fever or a hyperthermia. The absence of effects of an antipyretic dose of acetylacylic acid (aspirin) on both a rectal procedure and a novel cage stress supports the hypothesis that SIH is a real hyperthermia.
The standard SIH paradigm in singly housed mice is sensitive to anxiolytic-like effects of various psychoactive drugs (GABAA-benzodiazepine receptor agonists, alcohol, and 5-HT1A receptor agonists). An advantage of the method is that it simultaneously measures intrinsic effects of drugs on the core body temperature and that these effects are independent from the effects on SIH. The SIH procedure seems unable to find anxiogenic-like effects of drugs, presumably due to a ceiling effect in the enhanced temperature. The SIH procedure is very suitable to measure the effect of a genetic manipulation and also to study effects of drugs in mutants, as illustrated in 5-HT1A and 5-HT1B receptor knockout and CRF-overexpressing mice.
In rats, a similar procedure as in mice (T=— 60 min injection; 7 =0 min first rectal temperature measurement and r=+ 15 min, second rectal measurement) was developed that generated a reliable stress-induced hyperthermia, that is again sensitive to anxiolytic-like effects of various anxiolytics (benzodiazepines, alcohol, 5-HTIA receptor agonists).
Stress-induced hyperthermia is a simple, reproducible and species-, and strain-independent phenomenon associated with encountering stressful stimuli. The SIH procedure in mice is optimally suited to measure the putative modulating effects of drugs, but also of mutations on a physiological parameter reflecting anxiety. Moreover, intrinsic effects on body temperature are also measured, thereby creating an animal paradigm of anxiety that is, in contrast to almost all other anxiety tests, independent of locomotion. The latter is one of the biggest confounds in animal models of anxiety because of interference due to, e.g., sedation or psychostimulation.
Applying a stressor to an animal is a frequently used method to induce anxiety and is inherently present
Corresponding author. Tel.: +31 302533529; Fax: +31 302537900; E-mail: b.olivier(§ipharm.uu.nl in basically all anxiety paradigms presently used in scientific studies with an extensive range of applications, including drug and target finding, research into the mechanism of action of anxiolytic drugs, and brain mechanisms involved in anxiety and stress.
Animal anxiety paradigms are, for a large part, based on recording of one or more parameters of the behavior of the (stressed) animal. Often these paradigms reflect unconditioned responses of animals. In the case of the elevated plus maze, for example, investigators record the number of entries into, and time spent on the open and closed arms of the maze. Additionally, some groups (Rodgers and Cole, 1994) use extensive ethological parameters to profile the behavior. In this case it appears that most animals (rats, mice) avoid the open arms remaining largely in the closed spacing. The stressor in this case is the fear for the open and the elevation of the maze from the floor. Anxiolytics (e.g., benzodiazepines) apparently decrease the anxiety of the animals: they visit more frequently the open arms and consequently spend more time there. Many of such paradigms are used, including open field (Prut and Belzung, 2003), light-dark box (Bourin and Hascoet, 2003), defensive burying (De Boer and Koolhaas, 2003), social interaction (File and Seth, 2003), and variations of them. Several other paradigms are used in anxiety research based on conditioned responses such as Geller-Seifter conflict procedure (Geller and Seifter, 1960), Vogel-conflict test (Millan and Brocco, 2003), fear-potentiated startle (Davis et al, 1993), and contextual and cued fear conditioning (Fendt and Fanselow, 1999). In all cases the fear associated with a previous aversive stimulus (often an electrical shock) induced stress and anxiety. It is well known that anxiety, fear, and stress are associated with endocrine and autonomic phenomena (Carrasco and Van de Kar, 2003).
In the present contribution we focus on one particular autonomic phenomenon associated with fear, anxiety, and stress, the rise in core body temperature seen after a physical and/or psychological stressor. It is already known for a long time that certain stressful events lead to an increase in body temperature. In addition to the rise in body temperature, stressful stimuli induce various other physiological responses, including increases in blood pressure, heart rate, and plasma concentrations of ACTH and Cortisol or corticosterone. All these stress responses can be attributed to the stress-induced activation of the sympathetic nervous system. Several neurotransmitters/neuropeptides affect the activity of the sympathetic nervous system, including CRF, vasopressin, angiotensin II (Watanabe et al, 1999), and many others.
Is the rise in body temperature associated with a psychological or physical stressor a hyperthermia or a fever?
One of the big issues in the area of stress research, in particularly after psychological stressors, is the question whether the rise in body temperature associated with the stressor is a real fever or a hyperthermia (Oka et al, 2001). Fever is a centrally regulated rise in core body temperature (Tco) and is due to a raised "setpoint" temperature in the brain, toward which the thermoregulatory systems work to modulate Tco. To arrive at this new set point, the brain orchestrates changes in autonomic, neuroendocrine, and behavioral thermoregulatory responses by increasing heat-production responses (shivering, nonshivering thermogenesis, and heat-seeking behavior) and decreasing heat-loss responses (sweating, cutaneous vasodilatation, and cool-seeking behavior). When the new Tco has been reached, the body temperature will be regulated at this new setpoint, actively defended by the thermoregulatory centers in the brain, in particular the medial preoptic area of the hypothalamus (Hori and Katafuchi, 1998). In contrast, a hyperthermia is a rise in Tco above the setpoint temperature and is not actively defended by the brain thermoregulatory mechanisms. In fact, the unchanged setpoint of Tco leads to active regulatory processes trying to bring the temperature back to the wished setpoint. Therefore, a hyperthermia is expected to be associated with cutaneous vasodilatation, increased evaporative heat-loss responses (panting, wallowing), and cool-seeking behavior. These response patterns are opposite to those seen in the organism during the rising phase of the fever. Fever can be induced by exogeneous (microorganisms) and endogeneous pyrogens like IL-lot, IL-ip, IL-6, TNF-oc, IFN-oc, and MIP-1 (Kluger, 1991). The brain receives information about these processes by active transport of cytokines into the brain, signal transduction at the circumventricular organs, production of brain-permeable paracrine substances at endothelial cells in the cerebral microvessels, and stimulation of somatic and visceral (vagal) afferent nerves (Watkins et al, 1995). These signals ultimately alter the activity of thermosensitive neurons in the preoptic area (POA), the thermoregulatory center in the brain (Jessen, 2001). There are two types of thermosensitive neurons, cold- and warm-sensitive ones, and it has been shown that endogenous pyrogens increase the firing rate of cold-sensitive and decrease that of warm-sensitive neurons (Hori and Katafuchi, 1998). It is also found that a considerably portion of thermosensitive neurons in the POA respond to nonthermal emotional stimuli (Hori et al., 1986), thereby creating a mechanism for emotional or stress-induced hyperthermic effects.
In many species, including humans, psychological stress induces an acute rise in the core body temperature. Handling stress or exposure to a new environment rises the temperature in rats, mice, and rabbits (Yokoi, 1966; Snow and Horita, 1982). It is believed that fever during infection is mediated by prostaglandin release (primarily PGE2) in the POA (Blatteis and Sehic, 1997). Cyclooxygenase inhibitors, blocking prostaglandin synthesis, suppress pyrogen-induced fever. These antipyretics also attenuate the rise in body temperature induced by various forms of stress in rats, e.g., open field (Singer et al., 1986; Kluger et al., 1987), handling (Briese and Cabanac, 1980), or novel environment (Morimoto et al., 1991). These data suggest that, like pyrogen-induced fever, a major portion of emotional or psychological stress is also mediated by PGE2-induced activation of thermosensitive neurons in the POA. However, some stressors, including the procedure used in group-housed (Borsini et al., 1989;
Zethof et al., 1994) and singly housed (Van der Heyden et al., 1997; Olivier et al., 2003) stress-induced hyperthermia in mice are not antagonized by cyclooxygenase inhibitors, but instead by anxiolytic compounds, including benzodiazepines and 5-HT,A receptor agonists (Lecci et al., 1990a, b, 1991; Zethof et al., 1994). This suggests that SIH could be a real psychological hyperthermia and not a psychological fever. Watanabe et al. (1999) applied two kinds of stressors to mice, one was the injection of nonimmunological saline and the other the injection of IL-ip, acting as an immunological stressor. They found that saline gave one peak in the body temperature that vanished after approximately 60min, whereas the injection of IL-ip led to a dual picture; first the injection-induced peak, followed by a second increase in BT which was also long lasting (approximately 4h). Similar findings were reported by Gatti et al. (2002) on IL-ip and another pyrogenic compound, LPS (lipopolysaccharide) in C57BL/6 mice. This strongly suggests that the hyperthermia induced by a psychological stress (injection) and an immunological stress (IL-ip) are clearly different. In an attempt to further unravel this we studied the effect of 300 mg/kg of acetylsalicylic acid (aspirin) on two forms of stress in mice equipped with a telemetric device (Fig. 1). This antipyretic dose (Briese and Cabanac, 1980) was neither able to antagonize the SIH after a rectal procedure nor after the stress of
Acetylsalicylic acid (Aspirin)
Was this article helpful?