Ambient temperature and the pituitary hormone responses to exercise in humans

Ambient temperature and the pituitary hormone responses to exercise in humans Translation and Integration M. W. Bridge *, A. S. Weller †, M. Rayson ‡ and D. A. Jones School of Sport and Exercise Sciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, † Centre for Human Sciences, QinetiQ Ltd, A50 Building, Cody Technology Park, Ively Road, Farnborough, Hampshire GU14 0LX and ‡ Optimal Performance Ltd, Old Chambers, 93/94 West Street, Farnham, Surrey GU9 7EB, UK (Manuscript resubmitted 25 April 2003; accepted 16 June 2003) Pituitary hormones have an important role during exercise yet relatively little is known about the stimulus for their release. Body temperature progressively increases during prolonged steady-state exercise in the heat and we have investigated the role that this may play in the release of prolactin, growth hormone and cortisol (as an indicator of adrenocorticotropic hormone) into the circulation. Fit young male subjects exercised at 73 % ◊J,max until volitional fatigue at 20 °C and at 35 °C (30 % relative humidity at both temperatures). Rectal temperature and mean skin temperature were monitored and blood samples analysed for lactate, glucose, cortisol, growth hormone and prolactin concentrations. During the first 20 min, core temperature rose continuously and to a similar extent at both temperatures, while mean skin temperature was approximately 4 °C lower during exercise in the cool. Blood glucose concentration was essentially constant throughout the period of exercise while lactate concentration increased in the first 10 min and then remained constant with very similar changes in the two exercise conditions. Prolactin and growth hormone concentrations both increased during the exercise period while the concentration of cortisol declined slightly before rising slightly over the 40 min period. Prolactin release was significantly greater when exercise was carried out in the heat while there was no difference in the release of growth hormone or cortisol in the two conditions. When plotted as a function of rectal temperature, growth hormone concentration showed a linear relationship which was the same at ambient temperatures of 35 °C and 20 °C. Prolactin concentration had a curvilinear relationship with rectal temperature and this differed markedly at the two ambient temperatures. Cortisol concentration showed no dependence on any measure of body temperature. Our results are consistent with some aspect of body temperature being a stimulus for growth hormone and prolactin secretion; however, the precise mechanism clearly differs between the two hormones and we suggest that skin temperature modulates prolactin release, but does not affect the release of growth hormone. Experimental Physiology (2003) 88.5, 627–635. Experimental Physiology : Exercise presents a serious challenge to the homeostasis of the body, and pituitary hormones have an important part to play in regulating the supply of fuel, modulating inflammatory reactions and promoting repair of damaged tissues. Although much is known about the time course and extent of hormone release in response to exercise, the nature of the releasing signal is far from clear. Obvious candidates, such as blood lactate and glucose concentrations, often show little change or remain constant during prolonged exercise while hormone levels increase with time (Piacentini et al. 2002). The pituitary hormone response to such exercise is often described as a ‘stress response’ and, as such, implies a common mechanism for stimulating the release of several different hormones. Most hormones released during exercise are also increased as the result of passive heat exposure (Vigas et al. 2000) and the idea of a role for body temperature in the control of hormone release during exercise has been supported by a number of studies (Christensen et al. 1984; Brisson et al. 1986, 1987, 1991). Prolactin release from the anterior pituitary gland occurs in response to raised core body temperature both as a result of exercise and passive heating (Brisson et al. 1991). However, core temperature, at least as measured by rectal temperature, does not provide a complete explanation since the relationship between rectal temperature and prolactin release is modified by ambient temperature, being much reduced for the same rectal temperature when Publication of The Physiological Society 2593 * Corresponding author: m.w.bridge@bham.ac.uk Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 628 M. W. Bridge and others Exp Physiol 88.5 the skin is relatively cool (Brisson et al. 1986, 1991; Bridge et al. 1999). Growth hormone concentrations increase as a result of exercise and the extent is governed by exercise intensity (Pritzlaff et al. 1999) and has been reported to be linearly related to core temperature during exercise (Buckler, 1972; Karagiorgos et al. 1979). It has been suggested that this increase is the result of a direct stimulation of core thermoreceptors rather than cutaneous thermoreceptors (Christensen et al. 1984). Exercise in the heat often results in greater acidosis and higher blood lactate concentrations and it is possible that acidosis may play some part in the exercise-induced release of growth hormone (Luger et al. 1992). However, other studies have found no role for lactate in the growth hormone response to exercise (Karagiorgos et al. 1979). It is also possible that increased sympathetic nervous activity may be a mediator of the growth hormone response to exercise (Pritzlaff et al. 1999). The mechanisms behind the increased release of growth hormone in hot environments are therefore not clear. It has been shown that cortisol concentrations are elevated both during exercise in the heat (Galbo et al. 1979; Hargreaves et al. 1996) and by passive heat exposure (Collins & Few, 1979; Moller et al. 1989), and it has been reported that thermal clamping of body temperature abolishes the exercise-induced increase in cortisol concentrations (Cross et al. 1996). It has also been shown that cortisol increases during swimming in water of different temperatures only if body temperature increases (Galbo et al. 1979). The possible stimulus for this release is an increased core temperature; however an effect of increased skin temperature and cardiovascular demands during exercise in the heat cannot be ruled out. Additionally there is an element of psychological stress in the plasma cortisol response to exercise in the heat (Brenner et al. 1997). Cortisol concentrations can increase in the heat possibly as the result of increases in subjective feelings of discomfort (Follenius et al. 1982). It is therefore not clear whether the augmented cortisol response to exercise in the heat is due to increases in core temperature, an increased skin temperature or increased psychological stress. Our preliminary work, looking at prolactin release during exercise (Bridge et al. 1999), has shown differences in the relationship between prolactin concentration and rectal temperature in hot and cool conditions. The purpose of the present work was to determine whether growth hormone and cortisol release are affected in a similar way, suggesting a common stress hormone release mechanism in response to increased body temperature. skin temperature were measured every 5 min. Tests were randomly assigned and balanced for order. Subjects Thirteen recreationally active subjects participated in the study. Their mean age, body mass and maximal oxygen consumption (◊J,max) were 23.4 ± 3.2 years (mean ± S.D.), 71.4 ± 5.1 kg and 4.15 ± 0.15 l min_1, respectively. The study was performed according to the Declaration of Helsinki and was approved by the South Birmingham Local Research Ethics Committee. Subjects gave their written informed consent to the procedures. Experimental design Visit 1: maximal exercise test. Subjects completed an incremental exercise test to exhaustion on an electrically braked cycle ergometer (Lode Excalibur Sport, Groningen, The Netherlands) to determine maximal aerobic power output (Wmax) and ◊J,max. Workload was increased by 35 W every 3 min until volitional exhaustion. Expiratory gases were collected and averaged over a 10 s period, using a computerised on-line system (Oxycon Alpha, Jaeger, Bunnik, The Netherlands). Wmax was estimated using the equation of Kuipers et al. (1985): Wmax = Wfinal + (t w W)/T, where Wfinal (W) is the power of the last completed stage, t (s) is the time completed in the final stage, T (s) is the duration of each stage and W (W) is the workload increment. Visits 2 and 3: constant load exercise tests. On the day before visit 2, subjects recorded their diet and were then asked to adhere to the same diet on the day before visit 3. On the day of a test, subjects arrived at the laboratory at 08.00 h having fasted from midnight. A cannula was inserted into an antecubital vein to obtain blood samples. To ensure that subjects began each trial euhydrated they were given a bolus of water (8 ml (kg body weight)_1) to consume during the 45 min rest period between cannulation and the start of exercise. A resting blood sample was taken and subjects then began to exercise on a stationary electrically braked cycle ergometer. Exercise continued at a constant work rate of 65 % of Wmax and an ambient temperature of 35 °C or 20 °C, and 30 % relative humidity, until volitional fatigue. Subjects were asked to drink a minimum of 3 ml (kg body weight)_1 water every 15 min to maintain hydration during the exercise. Heart rate was continuously recorded (Polar Vantage NV, Polar OY, Finland) and expiratory minute ventilation (◊E), O2 consumption rate (◊J) and rate of CO2 production (◊CO2) were measured every 15 min. Venous blood samples (8 ml) were taken every 10 min during exercise for determination of haematocrit and haemoglobin, lactate, glucose, prolactin, growth hormone and cortisol concentrations. Body temperature measurements Rectal and mean skin temperatures, the latter using the four site formula of Ramanathan (1964), were recorded every 5 min (Squirrel Meter Logger, Grant Instruments, Cambridge, UK) Sweat rate Subjects were weighed nude immediately before the start of exercise and at the point of fatigue after having first dried themselves with a towel. Allowances were made for fluid ingested, respiratory water loss (Snellen, 1966), metabolic water (Mitchell et al. 1972) and quantity of blood drawn, to arrive at an overall sweat loss which was divided by time to give an average sweat rate over the entire period of exercise. Blood analysis Haematocrit was measured in triplicate by centrifugation. Blood glucose and lactate concentrations were measured using enzyme- Experimental Physiology : Translation and Integration METHODS General design Subjects performed two exercise tests to volitional fatigue on a cycle ergometer at ambient temperatures of 20 °C or 35 °C during which blood samples were collected every 10 min and rectal and Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 Exp Physiol 88.5 Temperature and hormone responses to exercise 629 Translation and Integration linked assays (Sigma Diagnostics, Poole, UK); haemoglobin concentration was measured using the cyanomethaemoglobin method (Sigma Diagnostics, Poole, UK). Prolactin and growth hormone concentrations were measured using radioimmunoassays (Skybio Ltd, UK). Cortisol concentration was measured by enzyme-linked immunosorbent assay (ELISA; DRG Instruments GmbH, Germany). All hormone analysis from a single subject was carried out in the same assay batch. Statistical analysis Changes in plasma volume were calculated from haemoglobin concentrations and haematocrit values using the equations of Dill & Costill (1974). Data were tested for approximation to a normal distribution. Exercise data were analysed up to 40 min to include the maximum number of subjects and were analysed using repeated measures ANOVA (SPSS 10) where data were normally distributed or, otherwise, a Friedman test was used. P values from ANOVA were corrected for sphericity using the Huynh-Feldt method, and significant differences between time points within trials were identified using Tukey’s post hoc test, and between trials with Student’s paired t tests. Significance was determined using Wilcoxon non-parametric tests if the data were found not to be normally distributed. Total hormone release was measured from the area under the curve of hormone level with time (AUC) calculated using the trapezoid method and corrected for basal values. Data are reported as mean ± S.E.M. unless otherwise stated. RESULTS Exercise duration Subjects exercised at the same level of absolute oxygen uptake in both trials (20 °C, 3.04 ± 0.13 l min_1; 35 °C, 2.99 ± 0.13 l min_1), which constituted a relative value of 73 ± 2 % ◊J,max. Exercise times at 20 °C ranged from 45 min to just over 2 h (mean, 81.2 ± 8.2 min) while at 35 °C they ranged from 36 to 98 min (mean, 56.8 ± 5.4 min). The time to fatigue in the heat was significantly shorter at 73 ± 5 % of the time to fatigue in the cool condition (P = 0.002). Experimental Physiology : Figure 1 Rectal temperature (A) and mean skin temperature (B) during exercise at 35 °C (•) and 20 °C (1). Rightmost data points indicate values at fatigue. * Significant difference between trials (P < 0.05); ^ significant difference from time point 0 min within a trial (P < 0.05). At all time points mean skin temperature was higher in the hot trial. Data are mean ± S.E.M. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 630 M. W. Bridge and others Exp Physiol 88.5 Translation and Integration Body temperature and sweat rate Rectal temperature increased in a similar fashion for the first 20 min when exercising at both 35 °C and 20 °C. Thereafter, the temperature continued to rise in the hot condition while rectal temperature was significantly lower in the cool condition (Fig. 1A). Mean rectal temperature at the time of volitional fatigue was 38.4 ± 0.2 °C at 20 °C and 38.8 ± 0.2 °C at 35 °C (P = 0.008). Mean skin temperature at 35 °C was higher than at 20 °C, by approximately 3–4 °C, throughout the exercise period (Fig. 1B, P = 0.001). Mean skin temperature in the heat rose significantly in the first 25 min of exercise (P < 0.05) and then remained constant. At 20 °C there was an initial small drop followed by a slight rise during the course of the exercise which resulted in a value that was significantly higher after 30 min than at the start; values at fatigue were 32.0 ± 0.4 °C at 20 °C and 35.2 ± 0.2 °C at 35 °C (Fig. 1B). Sweat rate was significantly higher during exercise at 35 °C than at 20 °C (20 °C, 1.32 ± 0.08 l h_1; 35 °C, 1.58 ± 0.12 l h_1, P = 0.03). Metabolic parameters Blood lactate concentration increased to steady-state values after the first 10 min of exercise in both conditions (Fig. 2A) and there were no further significant changes from 10 min to the time of volitional fatigue. There was a tendency for steady-state values to be slightly higher during exercise at 35 °C, which was reflected in blood lactate levels at fatigue of 4.3 ± 0.9 mM at 20 °C and 5.2 ± 0.8 mM at 35 °C; however, this was not statistically significant (P = 0.059). Blood glucose concentration (Fig. 2B) did not differ between trials or over time in the first 40 min of exercise, or between 10 min and fatigue. There was a tendency for Experimental Physiology : Figure 2 Lactate (A) and glucose (B) concentrations during exercise at 35 °C (•) and 20 °C (1). Rightmost data points indicate values at fatigue. * Significant difference between trials (P < 0.05); ^ significant difference from time point 0 min in both trials (P < 0.05). Data are mean ± S.E.M. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 Exp Physiol 88.5 Temperature and hormone responses to exercise 631 blood glucose concentrations to fall over the more prolonged periods of exercise at 20 °C and this resulted in a significantly lower blood glucose level at fatigue at 20 °C, in comparison to exercise at 35 °C (20 °C, 4.97 ± 0.30 mM; 35 °C, 5.97 ± 0.40 mM, Fig. 2B, P = 0.046). Translation and Integration Experimental Physiology : Cardiovascular and respiratory parameters Heart rate increased during exercise in both trials and was significantly higher at all times during exercise at 35 °C (Fig. 3A). At 20 °C, heart rate was 147 ± 3 beats min_1 after 10 min, increasing to 159 ± 3 beats min_1 after 40 min. The corresponding figures for exercise at 35 °C were 154 ± 3 and 169 ± 4 beats min_1, respectively. Heart rate at fatigue was significantly different between trials (20 °C, 169 ± 3 beats min_1; 35 °C, 174 ± 3 beats min_1, P = 0.006) but was less than maximal heart rate determined during the maximal exercise test. No differences in haematocrit were found between trials, and haemoglobin concentration and plasma volume changes were only significantly different between trials at the 40 min time point. At fatigue no differences in the three variables were found between trials (haematocrit: 20 °C, 44.1 ± 1.1 %; 35 °C, 45.0 ± 0.9 %; haemoglobin concentration: 20 °C, 14.8 ± 0.6 g dl_1; 35 °C, 15.2 ± 0.7 g dl_1; change in plasma volume: 20 °C, _2.2 ± 1.8 %; 35 °C, _3.9 ± 1.0 %). Minute ventilation increased significantly during exercise at 35 °C (15 min, 79.2 ± 4.9 l min_1; 30 min, 83.2 ± 5.6 l min_1, P < 0.05, Fig. 3B) but this was not significant at 20 °C. Initial values did not differ significantly between trials but by 30 min, values at 35 °C were significantly higher than at 20 °C (20 °C, 77.9 ± 4.6 l min_1; 35 °C, 83.2 ± 5.6 l min_1, P < 0.05, Fig. 3B). No changes were seen in oxygen uptake or respiratory exchange ratio (RER) during the course of exercise in either condition and there were no differences between trials (◊J,max: 20 °C, 3.04 ± 0.13 l min_1; 35 °C, 2.99 ± 0.13 l min_1; RER: 20 °C, 0.97 ± 0.02; 35 °C, 0.98 ± 0.02. Blood hormone levels Prolactin. Changes in circulating prolactin concentration are shown in Fig. 4A. At 20 °C, there was a small increase in prolactin concentration by 30 min of exercise with the values at 40 min and fatigue being significantly higher than at the resting level. At 35 °C, prolactin concentration rose progressively and was significantly greater than the resting level after 20 min and greater than at 20 °C from 10 min onwards. The differences in plasma prolactin release were also reflected in the calculated area under the curve, which was greater after 40 min of exercise in the hot condition (Fig. 4A, P < 0.03). Figure 3 Heart rate (A) and ventilatory (B) responses to exercise at 20 °C (1 and 5), and 35 °C (• and 4). Rightmost data points in A indicate values at fatigue. * Significant difference between trials (P < 0.05); ^ significant difference from other time points within trials (P < 0.05). Data are mean ± S.E.M. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 632 M. W. Bridge and others Exp Physiol 88.5 Translation and Integration Growth hormone. The concentration of circulating growth hormone (Fig. 4B) increased in a linear fashion with duration of exercise up to 30 min (35 °C) or 40 min (20 °C) after which the concentration began to plateau. There was a suggestion of differences between temperature conditions at 30 and 40 min of exercise but this was not statistically significant. There were no differences between the two conditions in calculated area under the release curves. Cortisol. Plasma cortisol concentration (Fig. 4C) changed relatively little during exercise with no differences between 35 °C and 20 °C. There was a slight (non-significant) fall in the first 10 min followed by a steady, but slow, increase. However, this increase was not significant in the first 40 min of exercise. Experimental Physiology : DISCUSSION The present results are consistent with previous observations concerning the release of prolactin when exercising in hot and cool environments, indicating that ambient temperature has a major effect in modulating the response (Bridge et al. 1999). The new observations are that while growth hormone release during the first 30–40 min of exercise shows a close relationship with rising rectal temperature, unlike prolactin, this relationship was similar in hot and cool conditions. In total contrast, cortisol release showed very little dependence on either core or ambient temperature. Prolonged exercise above 70 % ◊J,max is demanding and this is especially true when carried out in warm, humid conditions when core temperatures can rise, sometimes to dangerous levels. The pituitary hormone response to such exercise is often described as a ‘stress response’ and, as such, implies a common mechanism causing the release of several different hormones. Possible stimuli arising during exercise include hypoglycaemia, acidosis and increasing body temperature. In the present study it appears most unlikely that there is a direct causal relationship between blood lactate or glucose concentration and release of prolactin or growth hormone. Blood glucose concentrations hardly changed with exercise duration and lactate concentration, although elevated, remained constant after 10 min in both conditions while hormone levels continued to rise. Glucose feeding studies have shown that plasma cortisol concentration is unlikely to be affected by changes in blood glucose concentration during exercise (Bishop et al. 1999, 2001) and is probably also unaffected by blood lactate levels. There was only a modest cortisol response to exercise (Fig. 4C) and no suggestion that the release was in any way initiated or modulated by body temperature (Fig. 5C). This is at odds with previous findings that have shown an augmented release of cortisol in the heat (Galbo et al. 1979; Cross et al. 1996; Hargreaves et al. 1996). The difference might be due to (1) the fact that rectal temperature during the first 40 min of exercise in this study was similar in both conditions whilst others have found greater differences after 40 min of exercise (38.4 ± 0.1 °C after 40 min at 35 °C Figure 4 Prolactin (A), growth hormone (B) and cortisol (C) concentrations in response to exercise at 35 °C (8) and 20 °C (1). Bars to the right (A) in each graph indicate the area under the curve for hormone release up to 40 min of exercise, corrected for baseline values. * Significant difference between trials (P < 0.05); ^ significant difference from 0 min time point within trial (P < 0.05). Data are mean ± S.E.M. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 Exp Physiol 88.5 Temperature and hormone responses to exercise 633 Translation and Integration vs. 39.1 ± 0.2 °C after 40 min at 40 °C; Hargreaves et al. 1996) or (2) the different ambient temperatures used. Cortisol is only an indirect indicator of hypothalamic function since adrenocorticotropic hormone is the pituitary hormone that causes the release of cortisol from the adrenal medulla. The cortisol response is bound to lag behind that of adrenocorticotropic hormone and it is possible therefore that a temperature effect might have been seen at times after 40 min. However, rectal temperature at fatigue was different between conditions (Fig. 1A) yet no differences in cortisol concentration were found and a number of subjects exercised for around 2 h in the cool and 90 min in the heat and no effect of ambient temperature was evident in their cortisol responses. There have been a number of studies investigating the release of prolactin both during exercise (Brisson et al. 1986, 1991) and at rest (Brisson et al. 1991), and it is clear that changes in blood pH and osmolality are not responsible for stimulating release (Brisson et al. 1986). Likewise, heart rate and ventilation were higher in the heat but it seems unlikely that an increase of between 7 and 10 beats min_1 should make much difference when, in the cool condition, heart rate had increased by 2.5 times above the resting level in the first 30 min with very little prolactin production. However, there is considerable evidence that rising body temperature, as a result of exercise or passive warming, is an important stimulus and a linear relationship between circulating prolactin concentration and core temperature during low-intensity exercise has been reported (Brisson et al. 1986; Melin et al. 1988). Our own results tend to show a more biphasic relationship between rectal temperature and prolactin release (Fig. 5A). Ambient temperature is known to influence prolactin release and the present results confirm our previous observations (Bridge et al. 1999). Growth hormone release from the pituitary is under complex and multiple controls. Release associated with growth and the maintenance of bone and muscle is thought to occur mainly during rapid eye movement (REM) sleep in a characteristic pulsatile fashion (Van Cauter et al. 1998). There is evidence that during exercise such pulsatile release still occurs and the quantity of growth hormone secreted per pulse increases (Pritzlaff et al. 1999). As with prolactin, growth hormone release can be stimulated by the serotonergic 1A receptor agonist buspirone, implicating serotonergic and/or dopaminergic pathways in the release mechanism (Anderson & Cowen, 1992). Our results show that lactate and glucose concentrations during exercise are not the stimulus for growth hormone release. The results shown in Fig. 5B indicate a very clear relationship between growth hormone release and rectal temperature, at least up to 30 or 40 min of exercise, as has been shown before (Buckler, 1972; Karagiorgos et al. 1979; Christensen et al. 1984). However, different ambient conditions had no effect on growth hormone release, suggesting that if central thermoreceptors are triggering growth hormone release in response to heat exposure (Christensen et al. 1984), the Experimental Physiology : Figure 5 Prolactin (A), growth hormone (B) and cortisol (C) concentrations in response to exercise in relation to rectal temperature at 35 °C (8) and 20 °C (1). Data are mean ± S.E.M. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 634 M. W. Bridge and others Exp Physiol 88.5 Translation and Integration receptors, or subsequent pathways involved, differ from those controlling prolactin since environmental temperature played no part in modulating the growth hormone response. This is in line with the suggestion of Christensen et al. (1984) that it is mainly the central thermoreceptors that govern the augmented growth hormone release in the heat and that peripheral skin thermoreceptors play little or no part. Heat loss mechanisms are responsive to core temperature, sensed by receptors in the brainstem and preoptic area, but are also modulated by skin temperature with the integration of these signals thought to occur in the preoptic area of the hypothalamus (Boulant, 1981). Serotonergic and dopaminergic pathways in the hypothalamus are involved in temperature regulation (Yamawaki et al. 1983) and it is known that prolactin secretion is stimulated by serotonergic, and inhibited by dopaminergic, activity in the hypothalamus. Therefore it appears that the activation of prolactin release during exercise is, in a large part, driven by the rise in core temperature but that central sensitivity to such changes is altered by signals coming from peripheral thermoreceptors. Our results suggest that central core temperature is the major stimulus for growth hormone release whilst in the case of prolactin it is a combination of peripheral and central temperature signals. However, there is an alternative explanation. It is likely that rectal temperature is not a true representation of the temperature of blood perfusing the brain. There is no one measure of core temperature and it is known that rectal temperature is generally slightly lower than that measured in the oesophagus. Oesophageal temperature is measured close to the heart and aorta and thus is closer than rectal temperature to the temperature of blood perfusing the brain. At the start of exercise in cool conditions, oesophageal temperature will be higher than rectal temperature but as exercise continues and the body approaches thermal equilibrium the two will approach one another with the rectal temperature becoming higher than the oesophageal temperature (Nielsen, 1962). However in the heat, where thermal equilibrium is not achieved, oesophageal temperature may remain higher than rectal temperature throughout the exercise (Nielsen, 1976). If this were the case in the present experiments, the rectal temperature measured in the cool condition would have overestimated the temperature of blood perfusing the brain, whilst in the heat, it would have been an underestimate. The effect of this would be to draw the two lines in Fig. 5A together suggesting that the temperature of blood perfusing the brain is the sole stimulus for prolactin release. However, if this is the case it would also change the relationship between growth hormone and rectal temperature shown in Fig. 5B. Here the effect would be to separate the two lines in such a way as to suggest that growth hormone release is greater when cooler blood perfuses the brain. This seems unlikely. Our original hypothesis was that there is a common stimulus for the release of pituitary hormones during exercise and that consequently all vary in the same way with changes in body temperature and ambient conditions. This was evidently not the case. Our results indicate that body temperature plays little or no role in cortisol release while temperature has an important role in the release of both prolactin and growth hormone. However the different responses under different ambient temperatures clearly shows that the details of the release mechanisms differ and this may involve a modulating role for peripheral temperature receptors in the case of prolactin, but not growth hormone. Experimental Physiology : Anderson IM & Cowen PJ (1992). Effect of pindolol on endocrine and temperature responses to buspirone in healthy volunteers. Psychopharmacology 106, 428–432. Bishop NC, Blannin AK, Robson PJ, Walsh NP & Gleeson M (1999). The effects of carbohydrate supplementation on immune responses to a soccer-specific exercise protocol. J Sports Sci 17, 787–796. Bishop NC, Blannin AK, Walsh NP & Gleeson M (2001). Carbohydrate beverage ingestion and neutrophil degranulation responses following cycling to fatigue at 75 % VO2 max. Int J Sports Med 22, 226–231. Boulant JA (1981). Hypothalamic mechanisms in thermoregulation. Fed Proc 40, 2843–2850. Brenner IKM, Zamecnik J, Shek PN & Shephard RJ (1997). The impact of heat exposure and repeated exercise on circulating stress hormones. Eur J Appl Physiol 76, 445–454. Bridge MW, Allen T, Sharma A & Jones DA (1999). Effect of ambient temperature on exercise-induced prolactinaemia. J Physiol 521.P, 103P. Brisson GR, Audet A, Ledoux M, Matton P, Pellerin-Massicotte J & Peronnet F (1986). Exercise-induced blood prolactin variations in trained adult males: a thermic stress more than an osmotic stress. Horm Res 23, 200–206. Brisson GR, Bouchard J, Peronnet F, Boisvert P & Garceau F (1987). Evidence for an interference of selective face ventilation on hyperprolactinemia induced by hyperthermic treadmill running. Int J Sports Med 8, 387–391. Brisson GR, Peronnet F, Perrault H, Boisvert P, Massicotte D & Gareau R (1991). Prolactinotrophic effect of endogenous and exogenous heat loads in human male adults. J Appl Physiol 70, 1351–1355. Buckler JM (1972). Exercise as a screening test for growth hormone release. Acta Endocrinol (Copenh) 69, 219–229. Christensen SE, Jorgensen OL, Moller N & Orskov H (1984). Characterization of growth hormone release in response to external heating. Comparison to exercise induced release. Acta Endocrinol (Copenh)107, 295–301. Collins KJ & Few JD (1979). Secretion and metabolism of cortisol and aldosterone during controlled hyperthermia. J Physiol 292, 1–14. Cross MC, Radomski MW, VanHelder WP, Rhind SG & Shephard RJ (1996). Endurance exercise with and without a thermal clamp: effects on leukocytes and leukocyte subsets. J Appl Physiol 81, 822–829. Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011 Exp Physiol 88.5 Temperature and hormone responses to exercise 635 Dill DB & Costill DL (1974). Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol 37, 247–248. Follenius M, Brandenberger G, Oyono S & Candas V (1982). Cortisol as a sensitive index of heat-intolerance. Physiol Behav 29, 509–513. Galbo H, Houston ME, Christensen NJ, Holst JJ, Nielsen B, Nygaard E & Suzuki J (1979). The effect of water temperature on the hormonal response to prolonged swimming. Acta Physiol Scand 105, 326–337. Hargreaves M, Angus D, Howlett K, Conus NM & Febbraio M (1996). Effect of heat stress on glucose kinetics during exercise. J Appl Physiol 81, 1594–1597. Karagiorgos A, Garcia JF & Brooks GA (1979). Growth hormone response to continuous and intermittent exercise. Med Sci Sports 11, 302–307. Kuipers H, Verstappen FT, Keizer HA, Geurten P & van Kranenburg G (1985). Variability of aerobic performance in the laboratory and its physiologic correlates. Int J Sports Med 6, 197–201. Luger A, Watschinger B, Deuster P, Svoboda T, Clodi M & Chrousos GP (1992). Plasma growth hormone and prolactin responses to graded levels of acute exercise and to a lactate infusion. Neuroendocrinology 56, 112–117. Melin B, Cure M, Pequignot JM & Bittel J (1988). Body temperature and plasma prolactin and norepinephrine relationships during exercise in a warm environment: effect of dehydration. Eur J Appl Physiol Occup Physiol 58, 146–151. Mitchell JW, Nadel ER & Stolwijk JA (1972). Respiratory weight losses during exercise. J Appl Physiol 32, 474–476. Moller N, Beckwith R, Butler PC, Christensen NJ, Orskov H & Alberti KG (1989). Metabolic and hormonal responses to exogenous hyperthermia in man. Clin Endocrinol (Oxf) 30, 651–660. Nielsen B (1962). Body temperature during work at different environmental temperatures. Acta Physiol Scand 56, 120–129. Nielsen B (1976). Metabolic reactions to changes in core and skin temperature in man. Acta Physiol Scand 97, 129–138. Piacentini MF, Meeusen R, Buyse L, De Schutter G, Kempenaers F, Van Nijvel J & De Meirleir K (2002). No effect of a noradrenergic reuptake inhibitor on performance in trained cyclists. Med Sci Sports Exerc 34, 1189–1193. Pritzlaff CJ, Wideman L, Weltman JY, Abbott RD, Gutgesell ME, Hartman ML, Veldhuis JD & Weltman A (1999). Impact of acute exercise intensity on pulsatile growth hormone release in men. J Appl Physiol 87, 498–504. Ramanathan NL (1964). A new weighting system for mean surface temperature of the human body. J Appl Physiol 19, 531–533. Snellen JW (1966). Mean body temperature and the control of thermal sweating. Acta Physiol Pharmacol Neerl 14, 99–174. Van Cauter E, Plat L & Copinschi G (1998). Interrelations between sleep and the somatotropic axis. Sleep 21, 553–566. Vigas M, Celko J & Koska J (2000). Role of body temperature in exercise-induced growth hormone and prolactin release in nontrained and physically fit subjects. Endocr Regul 34, 175–180. Yamawaki S, Lai H & Horita A (1983). Dopaminergic and serotonergic mechanisms of thermoregulation: mediation of thermal effects of apomorphine and dopamine. J Pharmacol Exp Ther 227, 383–388. Acknowledgements This study was funded by Technology Group 5 (Human Sciences and Synthetic Environments) of the UK Ministry of Defence Corporate Research Programme and was supported by the Centre for Human Sciences, QinetiQ Ltd, Farnborough, UK. Experimental Physiology : Translation and Integration Downloaded from Exp Physiol (ep.physoc.org) by guest on July 27, 2011