- Email: [email protected]
Pain threshold changes induced by acute exposure to altered ambient temperatures
Peptides, Vol. 6, Suppl. 1, pp. 19-22, 1985.©AnkhoInternationalInc. Printedin the U.S.A.
0196-9781/85$3.00 + .00
Pain Threshold Changes Induced by Acute Exposure to Altered Ambient Temperatures A. D. SCHOENFELD, C. D. LOX, C. H. C H E N A N D L. O. L U T H E R E R 1 D e p a r t m e n t s o f Physiology, Obstetrics and Gynecology, and Internal Medicine Texas Tech University Health Sciences Center, Lubbock, T X 79430
SCHOENFELD, A. D., C. D. LOX, C. H. CHEN AND L. O. LUTHERER. Pain threshold changes induced by acute exposure to altered ambient temperatures. PEPTIDES 6: Suppl. l, 19-22, 1985.--Our previous findings that the degree of endotoxin-induced hypotension in the dog is inversely related to ambient temperature (19° through 30°C) and that only increased doses of naloxone are effective at 19°C suggested that opioid activity is also influenced by ambient temperature, increasing in the cold and decreasing in the warm. Others have reported increases in plasma/3-endorphin in rats with acute exposure to both 5° and 36°C. In this study we measured changes in pain thresholds after both acute and chronic exposures to lesser alterations in ambient temperature as a potentially more sensitive index of changes in central opioid activity. Compared to 24°C there was a marked increase in pain threshold with acute exposure to 10°C and marked decreases at 30° and 35°C. A slight decrease occurred after 30 minutes but not 60 or 120 minutes at 19°C. All acute changes disappeared three hours after the animals had been returned from the altered ambient temperature to 24°C. No changes were observed after six days chronic exposure to 10° or 30°C. These findings suggest that moderate, acute changes in ambient temperature can produce inversely related, adaptable alterations in central opioid activity. Pain threshold
Altered ambient temperature
Cold
Warm
Opioids
Acclimatization
similar exposure to 30°C. However, a recent study done in rats [3] showed that plasma/3-endorphin levels increased with acute exposure to both 5° and 36°C. Possible explanations for the apparent discrepancy between this observation and our findings thus far include: (1) exposure to 40°C represents a stress which overrides the usual decrease seen with moderate increases in ambient temperature; (2) a reversal in secretion pattern occurs somewhere between 30° and 40°C; (3) changes in enkephalins are more important than changes in fl-endorphins; (4) changes in central opioid activity are not reflected by peripheral fl-endorphin levels; or (5) the response is different in dogs than in rats. The present study was designed to partially test some of these possibilities by (l) evaluating pain threshold as an index of central opioid activity and (2) testing the effect of an intermediate range of temperatures. From the temperature regulation perspective, we also examined the effect of chronic exposure to altered ambient temperatures.
STUDIES in our laboratory and by other investigators have shown that the competitive opioid antagonist naloxone administered either peripherally [5] or by ventriculocisternal perfusion [8] can markedly offset the hypotension seen in many shock states. This has been interpreted as evidence that a primary shock-inducing insult such as endotoxin, hemorrhage, etc., leads to an increased secretion of endogenous opioids which in turn produce further cardiovascular depression. We also observed that the protective effect of naloxone during endotoxin shock in dogs is eliminated if the ambient temperature is reduced minimally from 24° to 19°C [9]. Further studies [1] showed that the degree of hypotension induced by endotoxin is inversely related to the ambient temperature between 19° and 30°C, and that the effectiveness of naloxone at 19°C could be restored by increasing the dosage. This suggested to us that changes in ambient temperature and resultant changes in thermoreceptor activation also might lead to changes in endogenous opioid secretion. Thus, in a cool ambient environment, the summation of opioid activity induced by both the cold stimulus and the endotoxin is sufficient to override blockade by the low dose of naloxone. Conversely, a reduction in opioid activity induced by acute exposure to a warm environment would tend to offset the increase induced by endotoxin and thereby lessen the degree of hypotension observed. This hypothesis was strengthened by our preliminary findings in the dog (unpublished observations) that plasma fl-endorphin levels doubled after a 30minute exposure to 19°C while tending to decrease after a
METHOD Sixty male Sprague Dawley rats weighing 241"+6 g were housed three to a cage in an environmental chamber illuminated from 8 a.m. to 6 p.m. and maintained at 24.0-+0.5°C. Purina Laboratory Chow and tap water were available ad lib. For a minimum of seven days, each rat was transferred daily to a single cage in a second environment chamber (maintained at 24°C during this period) and then removed from the
~Requests for reprints should be addressed to L. O. Lutherer, MD. PhD., Department of Physiology, Texas Tech University Health Sciences Center, Lubbock, TX 79430.
19
20
SCHOENFELD, LOX, CHEN AND L U T H E R E R
50
? i
i
25
.= ~" 15 w (n Z 0
(1. I0 (n b.I (z:
I',
"m
±
,. I ; H ~.
z::::
5
~N ;:,6a
50' 60' 2C
I0"
19"
0 24*
50*
35 °
AMBIENT TEMPERATURE FIG. 1. Effect of varying lengths of exposure to altered ambient temperatures on pain threshold as reflected by the response time to the hot plate test. Mean response times_+one standard error of the mean are indicated by the bars and vertical arrows. Exposure temperatures (°C) are given below the bars and the length of exposure within the bar (' =minutes, d=days). Significant differences in mean response times from that at 24°C are indicated by asterisks over the respective bars (*,o<0.05, **p<0.01, ***p<0.001).
chamber and placed for one minute in the inactive hot plate apparatus to accustom it to the handling procedures. The rats were assigned randomly to twelve groups of five each for the following exposure conditions: (a) 24°C (control); (b) 10°C for 30 min, 60 min, or 6 days; (c) 19°C for 30 min, 60 min, or 120 min; (d) 30°C for 30 min, 60 min, or 6 days; and (e) 35°C for 30 or 60 min. In order to prevent huddling during exposure to the cold environments and to eliminate interanimal influences, all rats were placed in individual cages. All exposures were initiated in the early morning with rats being placed in their cages at 5-minute intervals in order to allow for immediate testing of each animal at the end of the exposure period. The animals were removed from the chamber in the same staggered order and transferred to a chamber at 24°C where pain threshold assessment was performed using a commercially available hot plate apparatus (Techni Lab Instruments, Inc.). The temperature of the hot plate was set at 60°C on the basis of preliminary testing with another group of rats in which consistent results, compatible with those described in the literature, were obtained. The end point (response time) accepted for the test was the time at which the rat lifted a hind leg to either lick or blow on the paw. Timing was done by the same observer for all rats and was confirmed by one or more additional observers. During the preliminary testing with these additional rats it was also found that if the rats were subsequently retested once three hours later, the response times were the same but decreased
with further testing. Thus, since one retest appeared valid, animals in the control group as well as those which appeared to have a change in response time following the acute exposure to an altered ambient temperature were tested a second time three hours after they had been returned to the 24°C environment. Analysis of variance was performed on all data. If the F-value was significant, differences between individual group means were tested using the least significant difference method. The 5% level of confidence was accepted as significant. RESULTS
The mean response times for the initial test (immediately after removal from the experimental ambient temperature) for all groups are presented in Fig. 1. In general, the pain threshold as reflected by the measured response times was increased with acute exposure to lowered ambient temperature and decreased with acute exposure to elevated ambient temperatures. For animals exposed to 10°C, no change was evident after 30 minutes, but a highly significant increase (p<0.001) was observed in those animals exposed for 60 minutes. Thirty minutes after exposure to 19°C, there was a slight reduction (p<0.05), but this was not evident in the groups exposed for 60 or 120 minutes. At 30°C, no change was observed after 30 minutes, but a highly significant de-
P A I N T H R E S H O L D S AT A L T E R E D T E M P E R A T U R E S TABLE 1 EFFECT OF RETURN TO CONTROL ENVIRONMENT (24"C) ON RESPONSE TIME CHANGES OBSERVED WITH ACUTE EXPOSURE TO ALTERED AMBIENT TEMPERATURE
Exposure Conditions
Immediately After Exposure
Three Hours After Exposure
(24°C : control) 10°C : 60 rain 30°C : 60 rain 35°C : 30 min 35°C : 60 min
(17.8 32.7 8.9 6.4 7.7
(15.8 15.7 13.2 12.2 12.5
± ± ± ± ±
2.7)* 7.4~t 1.2t 0.4~t 1.4~t
± 3.2) --- 7.0 ± 1.4 ± 0.4 ± 0.7
*Mean response time (seconds) to hot plate test -- 1 SEM. tSignificantly different from control (immediate) (p<0.01). ~:Significantly different from control (immediate) (/7<0.001).
crease (p<0.01) occurred after the 60-minute exposure. A marked decrease (p<0.001) was observed after both 30 and 60 minutes at 35°C. As in the preliminary studies, when the response times of the control group were retested three hours later, there were no differences from the original values (Table 1). However, when each of the acute exposure groups that originally showed a difference were retested three hours after being returned to the 24°C environment, the response times were no longer different from control (Table 1). Retesting of response times was not done in the animals in the chronic exposure groups. The initial measurements revealed no changes in pain threshold after the six-day exposure to either 10° or 30°C (Fig. 1). DISCUSSION
During the past several years, a number of observations have led to the suggestion that endogenous opioids may contribute to thermoregulatory responses. The direction of change in colonic temperature in response to morphine was shown to be dose-related and originally considered by many to be nonspecific. The hyperthermia observed with low doses of morphine was thought to be secondary to increased heat production resulting from increased muscle activity independent o f any specific thermoregulatory pattern. The studies by Cox et al. [2], using intrahypothalamic injections of low doses of morphine, demonstrated a specific peripheral vasoconstrictor response. Other aspects of their study led them to propose that the low dose of morphine was affecting the " s e t - p o i n t " of a central thermostat. Subsequently, Holaday et al. [6] demonstrated that similar changes in colonic temperature could be induced with intraventricular injections of fl-endorphin. Holaday also showed [7] that /3-endorphin administered by the same route produced marked behavioral responses, wet-dog shakes and salivation, which are associated with thermoregulatory responses to cold and warm environments, respectively. Both responses were augmented by increases in ambient temperature, and again the suggestion was advanced that in this case fl-endorphin influenced a "central hypothalamic thermostat." Wei et al. [l 1] showed that wet-dog shakes and escape attempts, which are withdrawal signs from morphine de-
21 pendence, are influenced markedly by the ambient temperature at the time of the induced (naloxone) withdrawal. This suggested to them that central thermoregulatory mechanisms might not be functionally intact during withdrawal. Thornhill et al. [10] found that administration of naloxone led to a lower colonic temperature in rats acutely exposed to 4°C and a higher temperature in those exposed to 38°C as compared with untreated controls at the same temperatures. More recently, several studies have been done in which plasma fl-endorphin levels have been measured after acute exposures to altered environmental temperatures. Giagnoni et al. [4], using rats, found a significant increase after two hours at 4°C. In agreement with this, Deeter and Mueller [3] obtained an increase in rats transferred from 22° to 4°C, but they also obtained an increase with transfer to 36°C. In contrast, our preliminary measurements in dogs (unpublished observations) showed no change in plasma /3-endorphins after 30 minutes in animals transferred from 24° to 3&C, while there was a twofold increase with transfer to 19OC. All of the studies cited above have used severe rather than moderate temperature exposures. The question must be raised whether the afferent pathways stimulated by these exposures represent pure thermoreceptor information or include modalities analagous to pain which produce an increase in endogenous opioid activity not related to thermoregulatory mechanisms. In other words, do these represent increased opioid secretion as seen with stress-induced analgesia rather than a specific response to a thermal stimulus? Our previous studies relating to the effect of altered ambient temperatures on the degree of hypotension induced by endotoxin [ 1] and the effectiveness of naloxone in ameliorating this hypotension [9] were done using moderate changes in ambient temperature (between 19° and 30°C), and significant effects were observed. However, while our results were consistent with an increased endogenous opioid activity in the cold (19OC), they suggested a decrease with warm exposure (30°C). The results of the present study in rats, using a series of more moderate changes in ambient temperature and measurements of pain threshold as an index of central opioid activity, would tend to confirm our observations in the dog and suggest that while opioid activity is increased with exposure to cold, it actually falls with acute exposure to moderately increased temperatures. Plasma fl-endorphin levels were also measured in the present study, but the absolute values were low, and in response to our inquiries, the manufacturer has admitted retrospectively that there are some problems with the kit when used with rat plasma. F o r this reason the values are not reported. However, it is worth noting that the direction of change was for the most part consistent with the response-time data. A decrease of approximately 60 percent was observed with acute exposure to both 30° and 35°C. Up to this time, attention has focused on/3-endorphin as the opioid involved with thermoregnlatory responses. A possible role of enkephalins must be considered as well as the possibility that plasma/3-endorphin levels do not always reflect central opioid activity. The use of response-time measurements and a range of moderate temperature changes in the present study make the fmdings more convincing in terms o f demonstrating a role for endogenous opioids in temperature regulation. The fact that the changes observed as a result of acute exposure are transient and not observed with chronic exposure suggests that the opioids may participate in acclimatization but are uninvolved once acclimatiza-
22
S C H O E N F E L D , LOX, C H E N A N D L U T H E R E R
tion has occurred. Since the return to the 24°C environment also represented an acute change in ambient temperature (from the experimental temperature), and since follow-up response times reflected changes opposite in direction from
the initial changes, the possibility exists that endogenous opioid secretion may be altered in accordance with the direction and extent of change in environmental temperature independently from the original temperature.
REFERENCES 1. Chen, C. H., E. L. O'Leary, H. F. Janssen and L. O. Lutherer. Changes in ambient temperature alter the blood pressure response to endotoxin and the effectiveness of naloxone. Circ Shock. In press. 2. Cox, B., M. Ary, W. Chesarek and P. Lomax. Morphine hyperthermia in the rat: An action on the central thermostats. Eur J Pharmacol 35: 33-39, 1976. 3. Deeter, W. T. and G. P. Mueller. Differential effects of warmand cold-ambient temperature on blood levels of B-endorphin and prolactin in the rat. Proc Soc Exp Biol Med 168: 369-372, 1981. 4. Giagnoni, G., A. Santagostino, R. Senini, P. Fumagalli and E. Coil. Cold stress in the rat induces parallel changes in plasma and pituitary levels of endorphin and ACTH. Pharmacol Res Commun 15: 15-21, 1983. 5. Holaday, J. W. and A. I. Faden. Endorphin involvement in the pathophysiology of shock and trauma: therapeutic effects of naloxone. In: Adv. Physiol Sci, vol 26, Homeostasis in Injury and Shock, edited by Zs. Biro, A. G. B. Kovach, J. J. Spitzer and H. B. Stoner. Budapest: Pergammon Press, 1980, pp. 131140. 6. Holaday, J. W., P. Y. Law, L. F. Tseng, H. H. Loh and C. H. Li. B-endorphin: Pituitary and adrenal glands modulate its action. Proc Natl Acad Sci USA 74" 4628-4632, 1977.
7. Holaday, J. W., H. H. Loh and C. H. Li. Unique behavioral effects of B-endorphin and their relationship to thermoregulation and hypothalamic function. Life Sci 22: 1525-1536, 1978. 8. Janssen, H. F. and L. O. Lutherer. Ventriculocisternal administration of naloxone protects against severe hypotension during endotoxin shock. Brain Res 194: 608-612, 1980. 9. Janssen, H. F., J. L. Pugh and L. O. Lutherer. Reduced ambient temperature blocks the ability of naloxone to prevent endotoxin-induced hypotension. Adv Shock Res 7: 117-124, 1982. 10. Thornhill, J. A., K. E. Cooper and W. L. Veale. Core temperature changes following administration of naloxone and naltrexone to rats exposed to hot and cold ambient temperatures. Evidence for the physiological role of endorphins in hot and cold acclimatization. Commun J Pharm Pharmaco132: 427-430, 1980. 11. Wei, E., L. F. Tseng, H. Loh and E. L. Way. Similarity of morphine abstinence signs to thermoregulatory behaviour. Nature 247: 398--400, 1974. 12. Woolfe, G. and A. D. MacDonald. The evaluation of the analgesic action of pethidine hydrochloride (Demerol). J Pharmacol Exp Ther 80: 300-301, 1944.
0196-9781/85$3.00 + .00
Pain Threshold Changes Induced by Acute Exposure to Altered Ambient Temperatures A. D. SCHOENFELD, C. D. LOX, C. H. C H E N A N D L. O. L U T H E R E R 1 D e p a r t m e n t s o f Physiology, Obstetrics and Gynecology, and Internal Medicine Texas Tech University Health Sciences Center, Lubbock, T X 79430
SCHOENFELD, A. D., C. D. LOX, C. H. CHEN AND L. O. LUTHERER. Pain threshold changes induced by acute exposure to altered ambient temperatures. PEPTIDES 6: Suppl. l, 19-22, 1985.--Our previous findings that the degree of endotoxin-induced hypotension in the dog is inversely related to ambient temperature (19° through 30°C) and that only increased doses of naloxone are effective at 19°C suggested that opioid activity is also influenced by ambient temperature, increasing in the cold and decreasing in the warm. Others have reported increases in plasma/3-endorphin in rats with acute exposure to both 5° and 36°C. In this study we measured changes in pain thresholds after both acute and chronic exposures to lesser alterations in ambient temperature as a potentially more sensitive index of changes in central opioid activity. Compared to 24°C there was a marked increase in pain threshold with acute exposure to 10°C and marked decreases at 30° and 35°C. A slight decrease occurred after 30 minutes but not 60 or 120 minutes at 19°C. All acute changes disappeared three hours after the animals had been returned from the altered ambient temperature to 24°C. No changes were observed after six days chronic exposure to 10° or 30°C. These findings suggest that moderate, acute changes in ambient temperature can produce inversely related, adaptable alterations in central opioid activity. Pain threshold
Altered ambient temperature
Cold
Warm
Opioids
Acclimatization
similar exposure to 30°C. However, a recent study done in rats [3] showed that plasma/3-endorphin levels increased with acute exposure to both 5° and 36°C. Possible explanations for the apparent discrepancy between this observation and our findings thus far include: (1) exposure to 40°C represents a stress which overrides the usual decrease seen with moderate increases in ambient temperature; (2) a reversal in secretion pattern occurs somewhere between 30° and 40°C; (3) changes in enkephalins are more important than changes in fl-endorphins; (4) changes in central opioid activity are not reflected by peripheral fl-endorphin levels; or (5) the response is different in dogs than in rats. The present study was designed to partially test some of these possibilities by (l) evaluating pain threshold as an index of central opioid activity and (2) testing the effect of an intermediate range of temperatures. From the temperature regulation perspective, we also examined the effect of chronic exposure to altered ambient temperatures.
STUDIES in our laboratory and by other investigators have shown that the competitive opioid antagonist naloxone administered either peripherally [5] or by ventriculocisternal perfusion [8] can markedly offset the hypotension seen in many shock states. This has been interpreted as evidence that a primary shock-inducing insult such as endotoxin, hemorrhage, etc., leads to an increased secretion of endogenous opioids which in turn produce further cardiovascular depression. We also observed that the protective effect of naloxone during endotoxin shock in dogs is eliminated if the ambient temperature is reduced minimally from 24° to 19°C [9]. Further studies [1] showed that the degree of hypotension induced by endotoxin is inversely related to the ambient temperature between 19° and 30°C, and that the effectiveness of naloxone at 19°C could be restored by increasing the dosage. This suggested to us that changes in ambient temperature and resultant changes in thermoreceptor activation also might lead to changes in endogenous opioid secretion. Thus, in a cool ambient environment, the summation of opioid activity induced by both the cold stimulus and the endotoxin is sufficient to override blockade by the low dose of naloxone. Conversely, a reduction in opioid activity induced by acute exposure to a warm environment would tend to offset the increase induced by endotoxin and thereby lessen the degree of hypotension observed. This hypothesis was strengthened by our preliminary findings in the dog (unpublished observations) that plasma fl-endorphin levels doubled after a 30minute exposure to 19°C while tending to decrease after a
METHOD Sixty male Sprague Dawley rats weighing 241"+6 g were housed three to a cage in an environmental chamber illuminated from 8 a.m. to 6 p.m. and maintained at 24.0-+0.5°C. Purina Laboratory Chow and tap water were available ad lib. For a minimum of seven days, each rat was transferred daily to a single cage in a second environment chamber (maintained at 24°C during this period) and then removed from the
~Requests for reprints should be addressed to L. O. Lutherer, MD. PhD., Department of Physiology, Texas Tech University Health Sciences Center, Lubbock, TX 79430.
19
20
SCHOENFELD, LOX, CHEN AND L U T H E R E R
50
? i
i
25
.= ~" 15 w (n Z 0
(1. I0 (n b.I (z:
I',
"m
±
,. I ; H ~.
z::::
5
~N ;:,6a
50' 60' 2C
I0"
19"
0 24*
50*
35 °
AMBIENT TEMPERATURE FIG. 1. Effect of varying lengths of exposure to altered ambient temperatures on pain threshold as reflected by the response time to the hot plate test. Mean response times_+one standard error of the mean are indicated by the bars and vertical arrows. Exposure temperatures (°C) are given below the bars and the length of exposure within the bar (' =minutes, d=days). Significant differences in mean response times from that at 24°C are indicated by asterisks over the respective bars (*,o<0.05, **p<0.01, ***p<0.001).
chamber and placed for one minute in the inactive hot plate apparatus to accustom it to the handling procedures. The rats were assigned randomly to twelve groups of five each for the following exposure conditions: (a) 24°C (control); (b) 10°C for 30 min, 60 min, or 6 days; (c) 19°C for 30 min, 60 min, or 120 min; (d) 30°C for 30 min, 60 min, or 6 days; and (e) 35°C for 30 or 60 min. In order to prevent huddling during exposure to the cold environments and to eliminate interanimal influences, all rats were placed in individual cages. All exposures were initiated in the early morning with rats being placed in their cages at 5-minute intervals in order to allow for immediate testing of each animal at the end of the exposure period. The animals were removed from the chamber in the same staggered order and transferred to a chamber at 24°C where pain threshold assessment was performed using a commercially available hot plate apparatus (Techni Lab Instruments, Inc.). The temperature of the hot plate was set at 60°C on the basis of preliminary testing with another group of rats in which consistent results, compatible with those described in the literature, were obtained. The end point (response time) accepted for the test was the time at which the rat lifted a hind leg to either lick or blow on the paw. Timing was done by the same observer for all rats and was confirmed by one or more additional observers. During the preliminary testing with these additional rats it was also found that if the rats were subsequently retested once three hours later, the response times were the same but decreased
with further testing. Thus, since one retest appeared valid, animals in the control group as well as those which appeared to have a change in response time following the acute exposure to an altered ambient temperature were tested a second time three hours after they had been returned to the 24°C environment. Analysis of variance was performed on all data. If the F-value was significant, differences between individual group means were tested using the least significant difference method. The 5% level of confidence was accepted as significant. RESULTS
The mean response times for the initial test (immediately after removal from the experimental ambient temperature) for all groups are presented in Fig. 1. In general, the pain threshold as reflected by the measured response times was increased with acute exposure to lowered ambient temperature and decreased with acute exposure to elevated ambient temperatures. For animals exposed to 10°C, no change was evident after 30 minutes, but a highly significant increase (p<0.001) was observed in those animals exposed for 60 minutes. Thirty minutes after exposure to 19°C, there was a slight reduction (p<0.05), but this was not evident in the groups exposed for 60 or 120 minutes. At 30°C, no change was observed after 30 minutes, but a highly significant de-
P A I N T H R E S H O L D S AT A L T E R E D T E M P E R A T U R E S TABLE 1 EFFECT OF RETURN TO CONTROL ENVIRONMENT (24"C) ON RESPONSE TIME CHANGES OBSERVED WITH ACUTE EXPOSURE TO ALTERED AMBIENT TEMPERATURE
Exposure Conditions
Immediately After Exposure
Three Hours After Exposure
(24°C : control) 10°C : 60 rain 30°C : 60 rain 35°C : 30 min 35°C : 60 min
(17.8 32.7 8.9 6.4 7.7
(15.8 15.7 13.2 12.2 12.5
± ± ± ± ±
2.7)* 7.4~t 1.2t 0.4~t 1.4~t
± 3.2) --- 7.0 ± 1.4 ± 0.4 ± 0.7
*Mean response time (seconds) to hot plate test -- 1 SEM. tSignificantly different from control (immediate) (p<0.01). ~:Significantly different from control (immediate) (/7<0.001).
crease (p<0.01) occurred after the 60-minute exposure. A marked decrease (p<0.001) was observed after both 30 and 60 minutes at 35°C. As in the preliminary studies, when the response times of the control group were retested three hours later, there were no differences from the original values (Table 1). However, when each of the acute exposure groups that originally showed a difference were retested three hours after being returned to the 24°C environment, the response times were no longer different from control (Table 1). Retesting of response times was not done in the animals in the chronic exposure groups. The initial measurements revealed no changes in pain threshold after the six-day exposure to either 10° or 30°C (Fig. 1). DISCUSSION
During the past several years, a number of observations have led to the suggestion that endogenous opioids may contribute to thermoregulatory responses. The direction of change in colonic temperature in response to morphine was shown to be dose-related and originally considered by many to be nonspecific. The hyperthermia observed with low doses of morphine was thought to be secondary to increased heat production resulting from increased muscle activity independent o f any specific thermoregulatory pattern. The studies by Cox et al. [2], using intrahypothalamic injections of low doses of morphine, demonstrated a specific peripheral vasoconstrictor response. Other aspects of their study led them to propose that the low dose of morphine was affecting the " s e t - p o i n t " of a central thermostat. Subsequently, Holaday et al. [6] demonstrated that similar changes in colonic temperature could be induced with intraventricular injections of fl-endorphin. Holaday also showed [7] that /3-endorphin administered by the same route produced marked behavioral responses, wet-dog shakes and salivation, which are associated with thermoregulatory responses to cold and warm environments, respectively. Both responses were augmented by increases in ambient temperature, and again the suggestion was advanced that in this case fl-endorphin influenced a "central hypothalamic thermostat." Wei et al. [l 1] showed that wet-dog shakes and escape attempts, which are withdrawal signs from morphine de-
21 pendence, are influenced markedly by the ambient temperature at the time of the induced (naloxone) withdrawal. This suggested to them that central thermoregulatory mechanisms might not be functionally intact during withdrawal. Thornhill et al. [10] found that administration of naloxone led to a lower colonic temperature in rats acutely exposed to 4°C and a higher temperature in those exposed to 38°C as compared with untreated controls at the same temperatures. More recently, several studies have been done in which plasma fl-endorphin levels have been measured after acute exposures to altered environmental temperatures. Giagnoni et al. [4], using rats, found a significant increase after two hours at 4°C. In agreement with this, Deeter and Mueller [3] obtained an increase in rats transferred from 22° to 4°C, but they also obtained an increase with transfer to 36°C. In contrast, our preliminary measurements in dogs (unpublished observations) showed no change in plasma /3-endorphins after 30 minutes in animals transferred from 24° to 3&C, while there was a twofold increase with transfer to 19OC. All of the studies cited above have used severe rather than moderate temperature exposures. The question must be raised whether the afferent pathways stimulated by these exposures represent pure thermoreceptor information or include modalities analagous to pain which produce an increase in endogenous opioid activity not related to thermoregulatory mechanisms. In other words, do these represent increased opioid secretion as seen with stress-induced analgesia rather than a specific response to a thermal stimulus? Our previous studies relating to the effect of altered ambient temperatures on the degree of hypotension induced by endotoxin [ 1] and the effectiveness of naloxone in ameliorating this hypotension [9] were done using moderate changes in ambient temperature (between 19° and 30°C), and significant effects were observed. However, while our results were consistent with an increased endogenous opioid activity in the cold (19OC), they suggested a decrease with warm exposure (30°C). The results of the present study in rats, using a series of more moderate changes in ambient temperature and measurements of pain threshold as an index of central opioid activity, would tend to confirm our observations in the dog and suggest that while opioid activity is increased with exposure to cold, it actually falls with acute exposure to moderately increased temperatures. Plasma fl-endorphin levels were also measured in the present study, but the absolute values were low, and in response to our inquiries, the manufacturer has admitted retrospectively that there are some problems with the kit when used with rat plasma. F o r this reason the values are not reported. However, it is worth noting that the direction of change was for the most part consistent with the response-time data. A decrease of approximately 60 percent was observed with acute exposure to both 30° and 35°C. Up to this time, attention has focused on/3-endorphin as the opioid involved with thermoregnlatory responses. A possible role of enkephalins must be considered as well as the possibility that plasma/3-endorphin levels do not always reflect central opioid activity. The use of response-time measurements and a range of moderate temperature changes in the present study make the fmdings more convincing in terms o f demonstrating a role for endogenous opioids in temperature regulation. The fact that the changes observed as a result of acute exposure are transient and not observed with chronic exposure suggests that the opioids may participate in acclimatization but are uninvolved once acclimatiza-
22
S C H O E N F E L D , LOX, C H E N A N D L U T H E R E R
tion has occurred. Since the return to the 24°C environment also represented an acute change in ambient temperature (from the experimental temperature), and since follow-up response times reflected changes opposite in direction from
the initial changes, the possibility exists that endogenous opioid secretion may be altered in accordance with the direction and extent of change in environmental temperature independently from the original temperature.
REFERENCES 1. Chen, C. H., E. L. O'Leary, H. F. Janssen and L. O. Lutherer. Changes in ambient temperature alter the blood pressure response to endotoxin and the effectiveness of naloxone. Circ Shock. In press. 2. Cox, B., M. Ary, W. Chesarek and P. Lomax. Morphine hyperthermia in the rat: An action on the central thermostats. Eur J Pharmacol 35: 33-39, 1976. 3. Deeter, W. T. and G. P. Mueller. Differential effects of warmand cold-ambient temperature on blood levels of B-endorphin and prolactin in the rat. Proc Soc Exp Biol Med 168: 369-372, 1981. 4. Giagnoni, G., A. Santagostino, R. Senini, P. Fumagalli and E. Coil. Cold stress in the rat induces parallel changes in plasma and pituitary levels of endorphin and ACTH. Pharmacol Res Commun 15: 15-21, 1983. 5. Holaday, J. W. and A. I. Faden. Endorphin involvement in the pathophysiology of shock and trauma: therapeutic effects of naloxone. In: Adv. Physiol Sci, vol 26, Homeostasis in Injury and Shock, edited by Zs. Biro, A. G. B. Kovach, J. J. Spitzer and H. B. Stoner. Budapest: Pergammon Press, 1980, pp. 131140. 6. Holaday, J. W., P. Y. Law, L. F. Tseng, H. H. Loh and C. H. Li. B-endorphin: Pituitary and adrenal glands modulate its action. Proc Natl Acad Sci USA 74" 4628-4632, 1977.
7. Holaday, J. W., H. H. Loh and C. H. Li. Unique behavioral effects of B-endorphin and their relationship to thermoregulation and hypothalamic function. Life Sci 22: 1525-1536, 1978. 8. Janssen, H. F. and L. O. Lutherer. Ventriculocisternal administration of naloxone protects against severe hypotension during endotoxin shock. Brain Res 194: 608-612, 1980. 9. Janssen, H. F., J. L. Pugh and L. O. Lutherer. Reduced ambient temperature blocks the ability of naloxone to prevent endotoxin-induced hypotension. Adv Shock Res 7: 117-124, 1982. 10. Thornhill, J. A., K. E. Cooper and W. L. Veale. Core temperature changes following administration of naloxone and naltrexone to rats exposed to hot and cold ambient temperatures. Evidence for the physiological role of endorphins in hot and cold acclimatization. Commun J Pharm Pharmaco132: 427-430, 1980. 11. Wei, E., L. F. Tseng, H. Loh and E. L. Way. Similarity of morphine abstinence signs to thermoregulatory behaviour. Nature 247: 398--400, 1974. 12. Woolfe, G. and A. D. MacDonald. The evaluation of the analgesic action of pethidine hydrochloride (Demerol). J Pharmacol Exp Ther 80: 300-301, 1944.