Regulatory Peptides, 38 (1992) 55-69
55
© 1992 Elsevier Science Publishers B.V. All rights reserved. 0167-0115/92/$05.00 REGPEP 01150
Thyrotropin-releasing hormone action in the preoptic/anterior hypothalamus decreases thermoregulatory set point in ground squirrels Julie H. H e n d r i k s e n a, Patricia A. B a c h e l o r b, R o b e r t J. N e w m a n b and T o n i L. S t a n t o n a aDepartment of Physiology and bDepartment of Psychology, California State University at Long Beach, Long Beach, CA (U.S.A.)
(Received 13 September 1991; revised version received 14 November 1991; accepted 20 November 1991) Key words: T R H ; Neuropeptide; Brain; Behavior; Heat escape
Summary Earlier work has shown that thyrotropin releasing hormone (TRH) produces dosedependent decreases in body temperature (Tb) and metabolic rate when microinjected into the dorsal hippocampus (HPC) or preoptic/anterior hypothalamus (PO/AH) of awake ground squirrels. This study employed a behavioral paradigm to investigate the possibility that TRH-induced hypothermia is associated with a decrease in thermoregulatory set point. Six animals were successfully trained to press a bar for radiant heat escape and cool air reinforcement in order to obtain a cooler ambient temperature (Ta). During experimental testing, the animals were microinjected remotely with T R H (10-1000 ng//~l) or a control solution (sterile saline or T R H - O H ) into the PO/AH. The micro-injections were delivered via bilateral injection cannulae inserted through chronic bilateral cannula guides that had been stereotaxically implanted under pentobarbital anesthesia. Cumulative and time-integrated bar presses were obtained on a computer generated display. Tb, measured in the brain via a bead-type thermistor, and chamber Ta were recorded continuously. Following T R H administration, a significant increase in mean bar-press rate was observed during the period in which To was falling, when compared to a comparable time period just prior to the microinjection. These findings complement results obtained from four animals that were trained to press a bar for heat reinforcement in a cold ( - 10 ° C) environment. In this alternative behavioral paradigm,
Correspondence: T.L. Stanton, Department of Physiology,California State University at Long Beach,
1250 BellflowerBlvd., Long Beach, CA 90840-3701, U.S.A.
56 microinjection of TRH into the PO/AH or HPC induced a decrease in mean bar-press rate as Tb was falling. The results support the hypothesis that TRH-induced hypothermia in golden-mantled ground squirrels is achieved by lowering thermoregulatory set point.
Introduction
Thyrotropin releasing hormone (TRH), a tripeptide identified and characterized in 1969 [ 1,2], appears to mediate a variety of processes within the central nervous system (CNS) that are independent of its role in the regulation of thyroid function [3,4]. A substantial body of evidence suggests that it acts as a neurotransmitter or neuromodulator within the CNS [4,5]. Similar to results obtained from many other species [4,5], studies in our laboratory using the golden-mantled ground squirrel have demonstrated that TRH and high affinity receptors for TRH are widely distributed throughout the brain [6,7]. Given the wide-spread distribution of TRH and its receptors, it is not surprising that a wide range of centrally mediated behavioral and physiological effects are stimulated by TRH administration [3-5]. In particular, TRH is one of the most extensively studied peptides that have been implicated in the mediation of thermoregulatory responses in mammals [8]. The results of initial studies demonstrated the ability of TRH to antagonize drug-induced hypothermia and narcosis [9-11]. Subsequent data from a variety of mammalian species showed that TRH administration alone generally elicits rapid hyperthermic responses [8,12,13]. However, it is evident in rats [14-17] and rabbits [ 18 ] that TRH produces a complex pattern of temperature effects involving both heat loss and heat production responses that are dependent upon ambient temperature (Ta) and route of administration. Cats, on the other hand, consistently respond to TRH with a fall in body temperature (Tb) whether delivered intraventricularly [ 19] or intracerebrally [20]. In our earlier work in ground squirrels [21,22] we demonstrated complex statedependent effects of TRH when microinjected into the dorsal hippocampus (HPC). That is, TRH administered to animals in a state of deep hibernation [21 ] or slow wave sleep [22] evoked a rise in Tb and metabolic rate, whereas, in the awake animal [22], TRH produced dose-dependent hypothermia and decreased metabolic rate. These studies, however, did not address the question as to whether the effects of TRH on Tb are thermoregulatory in nature (i.e., acting to change the set point for Tb) or merely the result of the activation of thermal effector mechanisms. Thus, we designed experiments using a behavioral paradigm in order to evaluate the ability of TRH to lower thermoregulatory set point (Tsp) in awake ground squirrels. In the initial experiments, ground squirrels were trained to press a bar for radiant heat reinforcement in a - 10 °C environment. The results of these experiments, some of which were reported in abstract form [23 ], demonstrated that TRH microinjection into the HPC or preoptic/anterior hypothalamus (PO/AH) of ground squirrels that were
57 pressing a bar to activate a radiant heat source produced a decrease in mean bar-press rate during the period in which Tb was falling. Thus, the animals' behavioral and physiological responses were complementary and, therefore, consistent with the interpretation that T R H is involved in thermoregulatory control processes by lowering Tsp. However, because T R H also elicits sedative effects and a decrease in electromyographic activity in this species [22], it is possible that the TRH-induced decrease in bar-press rate could have been the result of muscle relaxation and sedation, rather than representing complementary thermoregulatory behavior associated with a lowering of Tsp. In order to test this possibility, we have continued our investigation by examining the effects of T R H administration into the PO/AH of animals that were trained to press a bar in order to reduce an elevated Ta. If, as hypothesized, T R H lowers Tsp in the awake ground squirrel, then an increase in bar-press rate would be expected to accompany a TRH-induced fall in Tb in this alternate behavioral paradigm.
Materials and Methods
Animals Male and female golden-mantled ground squirrels were obtained by a licensed trapper in the mountains of northern California during late spring and early summer. The animals were individually housed in stainless steel cages in a colony room maintained at 21 + 2 ° C. Room illumination was regulated on an adjustable schedule that approximated the natural light:dark cycle. Food (lab chow, sunflower seeds, fresh fruit) and water were available ad libitum. Behavioral training and criteria Behavioral heat escape paradigm. The behavioral test chamber consisted of a wooden enclosure (26 cm long × 24 cm wide × 30 cm high) covered by a removable wire screen. A quiet, low velocity fan under rheostat control was mounted in one wall of the chamber adjacent to a stainless steel bar situated approx. 1 cm above the floor that extended 1 cm into the chamber. The bar was bent at a right angle to extend alongside the wall in front of the fan. This placement enabled the animal to position itself to receive maximum cool air reinforcement without unnecessary movement. Two infrared heat lamps (250 W each) under rheostat control were mounted 30 cm overhead. Depression of the bar closed a microswitch that, via computer control, activated the fan and simultaneously turned off the heat lamps for three seconds. In addition, an incandescent lamp (100 W) was mounted 43 cm above the test chamber. This lamp remained illuminated throughout training and testing sessions in order to keep the chamber adequately illuminated when heat lamps were turned off by the animal. Six ground squirrels were successfully trained to escape the heat by pressing a bar to turn off the heat lamps and were subsequently tested for T R H dose response characteristics. Training was initiated by placing a naive animal into the preheated (38 ° C) chamber. Its back was pre-shaved to expose the skin directly to the infrared heat stimulus. Shaving the animal's back decreased the latency for it to respond to the stimulus, as evidenced by the animal's skin twitch response and earlier exploration of
58 the chamber. Any movement toward the bar was immediately reinforced by manual activation of the bar by the trainer. Once the animal discovered the bar, the trainer ceased to activate the bar. Reinforcement time (i.e., the duration of bar-activated heat lamp off/fan on time) was initially set for 5 s, and bar press responses that were made by the animal while heat lamps were off did not prolong the reinforcement time. Any sign of serious stress, such as jumping or wetness around the mouth and neck, prompted the immediate removal of the animal from the chamber. Behavioral heat reinforcement paradigm. Four ground squirrels were trained to press a bar to turn on the heat lamps. The effects of T R H were subsequently tested in the P O / A H of two of these animals and in the H P C of the other two. The behavioral test chamber and the approach to behavioral training were the same as that employed in the heat escape paradigm, with the following differences: the test chamber was situated at the bottom of a chest-type freezer maintained under rheostat control at - 10 °C; depression of the bar resulted only in the activation of the heat lamps; the fan was inactivated throughout the training and test sessions. Criteria. Training sessions were spaced by at least 3 days and conducted for approx. 6 weeks. An animal was considered well trained if it depressed the bar in a regular, non-agitated manner with little or no extraneous movement. In addition, the bar-press rate of the animal had to be of sufficient regularity in order to allow a discernable increase or decrease in response rate following a change in the stimulus-reinforcement parameters, such as changes in heat lamp or fan intensity or duration. Once the animal became familiar with the test environment, the reinforcement duration was reduced to three seconds for all remaining training and experimental test sessions. Bar-press responding for a well-trained animal in the heat escape paradigm reliably maintained T a at an average of 29 + 1.5 °C.
Surgical procedures Following completion of training, each animal was anesthetized with sodium pentobarbital (75 mg/kg, intraperioneaily), given an intramuscular injection of procaine penicillin (30,000 units), and stereotaxically implanted under aseptic conditions with a bilateral guide apparatus (21-gauge stainless steel tubes). The guide tubes were lowered to a level 1 mm above the P O / A H (AP 8.3, L + 1.5, H 5.0) or, alternatively in two animals, 1 m m above the H P C (AP 5.3, L + 2.5, H 10.0) according to the atlas of Joseph et al. [24]. A calibrated bead-type miniature thermistor (VECO 32A7) was implanted 2 to 3 mm into the parietal cortex through a third cranial opening. The implants and Amphenol connector containing the thermistor leads were secured to the animal's skull with self-curing dental acrylic and miniature stainless steel screws (size 0-80). Removable stainless steel inserts (26-gauge) were used to seal the guide tubes. The wound was cleansed with 3 % hydrogen peroxide, dusted with furacin powder, and closed with a 9 mm stainless steel wound clip. At least 1 week was permitted for post-operative recovery.
Temperature measurements and recording Tb was sensed in the brain and measured continuously during each test session by connecting a flexible cable between the skull-mounted Amphenol connector and a
59 Wheatstone bridge, the output of which was displayed on a Honeywell recording potentiometer. Although Ta was not recorded in the initial heat reinforcement experiments conducted in the cold, Ta was measured in the heat escape paradigm by placing a YSI telethermometer probe in the corner of the test chamber, opposite the bar and at a height and distance from the fan that approximated the distance between the animal's head and the fan. Ta was calibrated through the telethermometer and recorded continuously on the same potentiometer used for the recording of Tb.
Behavioral data recording Bar-press responses (both cumulated and integrated over 5-min periods) and cumulated reinforcements (heat lamp off/fan on events) were obtained on a computergenerated display (y-axis: responses and reinforcements; x-axis: time). Each time the animal pressed the bar, the computer added a point to the y-axis display, creating two lines whose slopes designated the animal's bar-press response and reinforcement rate. (The complete absence of bar-pressing resulted in the formation of a horizontal line.) 5-min integrations of heat escape responses were displayed as a histogram. The percent change in the heat escape response rate was determined from the difference between equal-length periods of the integrated record before and after microinjection.
Experimental protocol Prior to experimental testing, empty injection cannulae were inserted into each animal to expose the site for microinjection to the presence of the cannulae. This was done in order to avoid possible artifacts due to mechanical and lesioning effects at the time of the first microinjection. In addition, each animal was observed in the behavioral paradigm to determine any change in bar-press behavior compared to pre-surgical training sessions. No change was observed in any of the animals. Animals that were trained in the heat escape paradigm maintained chamber Ta at the same level that was observed during pre-surgical training trials. Prior to each test session, a bilateral injection apparatus (26-gauge stainless steel tubing connected via lengths of PE20 polyethylene tubing to two Hamilton microliter syringes) was filled with fresh solutions of T R H (10 to 1000 ng/#l (Bachem, Torrance, CA)), T R H - O H (deaminated TRH, 100 ng/#l), or sterile saline (0.9~o NaC1). A small air bubble was introduced in the end of the polyethylene tubing proximal to the syringes in order to monitor the flow of the drug at the time of microinjection. The filled polyethylene tubes were then either prewarmed or precooled, depending on the behavioral paradigm, by placing them over the side of the preheated or precooled chamber for a few minutes in order to accommodate temperature-induced changes in tubing volume. A recording cable was attached to the skull-mounted Amphenol connector, and the prefilled injection apparatus was inserted into the chronic guide tubes. When fully inserted, the injection cannulae extended 1 mm beyond the tips of the guide tubes and into the PO/AH. Prior to drug delivery during a test session, a preinjection period of 30 to 60 min was used to obtain baseline records. Microinjections were delivered remotely over a 2 min period in a volume of 1 #1. Slow delivery of the drug was used to avoid unnecessary ancillary effects of the microinjection procedure. Following the microinjection, the test
60
session continued until Tb returned to a stabilized level. An animal was tested no more than twice weekly, each session lasting 1.5 to 2 h. Except for one animal that dislodged its implant before receiving the last dose of T R H , all animals were tested at three dose levels of T R H plus a control test, selected on a randomized schedule.
Histology Animals were deeply anesthetized with sodium pentobarbital and killed by perfusion through the ascending aorta with 50 ml of saline followed by 50 ml of 10~o formalin. The brains were removed and stored in 10~'0 formalin until sectioned on a freezing microtome. Fixed sections (80 #m) were examined by light microscopy to verify cannulae placement.
Results The results obtained from four ground squirrels that were microinjected with T R H or saline into either the P O / A H or H P C while pressing a bar for heat reinforcement in the cold are displayed in Table I. T R H action at both sites induced a dose-related fall in Tb that was associated with a dose-dependent decrease in the animal's rate of responding. A compilation of the data obtained from five animals that received microinjections of T R H and control solutions into the P O / A H while behaviorally responding for heat escape is shown in Figs. 1 and 2. The results show that T R H produced a fall in Tb that was associated with a statistically significant increase in bar-press response rate (i.e., increase in heat escape response) between 0 and 100 ng of T R H . Ta was typically lowered 1 to 5 °C following T R H administration. By contrast, the response to control microinjections of either sterile saline or T R H - O H was characterized by a decline in the rate of bar-press behavior, a rise in Ta, and a mean increase in Tb of 0.13 °C. Although differences between dose levels for TRH-induced hypothermia did not reach statistical significance, regression analysis demonstrated a significant linear dose-
TABLE I Mean change in Tb and bar-press rate following microinjection of TRH or 0.9% saline into the PO/AH or the HPC of ground squirrels that were bar-pressing for heat reinforcement in the cold TRH dose (#g//21)
0 (S) 25 50 100
PO/AH
HPC
Tb (C)
B-P Rate
Tb (C)
B-P Rate
,L0.2 ,L1.2 ~,1.8 ,t2.3
~ 1~o ~24% ,t37% +44%
NC ~ 1.4 ~,1.8 ~,2.4
NC ~30~o ~32% +35~,o
Ta = - 10°C; PO/AH = preoptic/anterior hypothalamus; HPC = dorsal hippocampus; (S) = saline; NC = no change; N = 2 PO/AH + 2 HPC
61 0.25-
v
,~
0.25.
E
050-
c 0.75. J3
c
& -lo. c
E~ J~ U
-1.25.
-1.50 J
10 C
100 TRH
1000
ng/pl
Fig. 1. Change in T b following microinjection of T R H and control solution (saline or T R H - O H ) into the PO/AH of ground squirrels. Each bar represents the mean _+ S.E.M. Number within bars represents number of animals. C = control tests.
60] 50
40-
3O c~ _a
20
X m e
10
© c
g o
-10
U
-20
-30' 4 10
C
100
TRH
1000
ng/pl
Fig. 2. Percent change in heat escape response rate following microinjection of TRH and control solution (saline or T R H - O H ) into the PO/AH of ground squirrels. Each bar represents the mean + S.E.M. Number within bars represents number of animals. C = control tests. C vs. 10 ng (t = 2.61, P < 0.05); C vs. 100 ng (t = 2.673, P < 0.05); C vs. 1000 ng (t = 2.13, P < 0.10).
62
421 "O 414
k, CL
4o
E c
'~L
39 '
m
~ L
38-
:~
36.
~o c~
34.
~
32. 30'
-z,
,
,
28
E <
26.
Fig. 3. Baseline record of Tb (top) and chamber Ta (bottom) of a well-trained animal during a behavioral training session. A
42.
0v
e
41
9 ~.,
40
E
.E
39.
P
38.
3if: 36 34 32' 30 28
E
26.
<
400]
J
3001 TRN lOOng
.E E
8 i t.
g
i
lOmin
Fig. 4. Experimental record from same animal depicted in Fig. 3 following microinjection of 100 ng/#l of TRH into the PO/AH. Changes in Tb and chamber Ta (A); cumulative (upper B) and integrated (lower B) bar-press (heat escape) responses.
63 response trend between 0 and 100 ng o f T R H (F = 5.02; P < 0.02). At the highest dose of T R H (1000 ng) tested, other unexpected effects became apparent, such as chirping (a characteristic sound of alarm, but infrequently heard in captive animals) or grooming behavior. In addition, periodic or complete interruptions in the animals' heat escape behavior occurred, resulting in great variability in the Tb response. Although well trained, the animals did not immediately begin to bar-press when first placed in the preheated test chamber. Rather, each animal would sit quietly for several minutes as Ta continued to rise and Tb increased to 40.5 + 1.5 °C. In no instance was any sign of heat stress or apparent ill effect evident in the animals as a result of allowing their Tb to rise to these levels. Once an animal initially activated the heat escape response, a regular response rate immediately ensued, causing Ta to drop precipitiously. Subsequently, the animal's Tb fell steadily and usually resulted in a declining Tb baseline.
42-
'S ¢, L
:~
41
E
40
+,
~
A
c
P m
39T R H - - 01.4 l O O n g
3432
~
30
E
28
~
26 i
E 40O-
g
I
10min B
3oo-
Q_
200loo-
lOOng
1.
c
E
U3
C O
L O if)
'10rain
Fig. 5. Experimental record from same animal depicted in Figs. 3 and 4 followingcontrol microinjection of 100 ng/#l of TRH-OH into the PO/AH. Changes in Tb and chamber Ta (A); cumulative (upper B) and integrated (lower B) bar-press (heat escape) responses.
64
Fig. 3 represents a training record from one animal and illustrates the typical pattern of response. T R H microinjection into the PO/AH precipitated an initial disruption of bar-pressing (lasting 3-4 rain), accompanied by behavioral quieting and a transient, small rise in Th ( < 0.3 °C) before an increase in heat escape response rate and a fall in Tb occurred (Fig. 4). Although a decline in response rate was observed following control microinjections of saline or TRH-OH, complete interruption of heat escape responding did not occur. The record following T R H - O H administration (Fig. 5) in the same animal depicted in Figs. 3 and 4 is quite similar to that observed during the animal's training session (Fig. 3). In one animal, the guide cannulae placement missed the intended target site by 0.6 mm in the anterior/posterior plane (A/P coordinate: 8.9) and by 1 mm in the horizontal plane (H coordinate: 4.0). The injection site in this animal included the most anterior border of the preoptic nucleus, the supraoptic nucleus, the diagonal band of Broca and the medial forebrain bundle. In this animal, TRH at all three dose levels evoked a prolonged disruption of heat escape behavior that lasted approx. 10 min. Microinjection of T R H - O H in this animal produced no change in Tb and a 22~o decline in bar-press rate that was comparable to the mean change observed in PO/AH animals following control microinjections. Histological examination of the cannulae placements in the animals included in
~
11 1C
CC
' r11 .t
F
~jN'"PM '°, a i/v.~,|le
A 8.6 -2
0 1 2 3 4
A83
• °"
I
A8.0
~ i 2 3 4 5 6 ~ 8 ~
Fig. 6. Schematic representation ofmicroinjection sites for the PO/AH animals represented in Table I and Figs. 1 and 2 as determined by histological analysis. Encircled regions represent the sites of injection cannulae tip placement within the PO/AH (AP 8.0-8.6; L + 1.5; and H 3.5-4.0). Structures identified according to Joseph et al. [24]: corpus callosum (CC); fornix (F); anterior commissure (CA); optic chiasm (CO); optic tectum (TO); medial and lateral preoptic nuclei (NPM and NPL); paraventricular nucleus (NPH); anterior hypothalamic nucleus (NAH); lateral ventricle (VL).
65 Table I and Figs. 1 and 2 indicated that injection sites were well within the designated site. A schematic representation of the PO/AH injection sites are shown in Fig. 6.
Discussion
The results of this study have demonstrated that microinjection of T R H into the PO/AH of the awake ground squirrel elicits a fall in Tb that is accompanied by a behavioral response that is consistent with a Tsp-lowering action of TRH. In the heat escape paradigm, in which animals pressed a bar to turn off a radiant heat source and activate cool air reinforcement, T R H induced an increase in bar-press rate during the period in which Tb was falling. These findings complement the results we obtained from animals in a heat reinforcement paradigm, in which TRH microinjection into the PO/AH and HPC led to a fall in Tb and a decrease in behavioral heat gain bar-press rate. These data are also consistent with the results of our earlier study [22] in which T R H produced dose-dependent decreases in Tb and metabolic rate in awake ground squirrels. Taken together, these data support the hypothesis that T R H plays an important role in thermoregulation in C. lateralis by lowering Tsp. Thermoregulatory set point is a concept that is used to account for the fact that internal temperature may be regulated at different levels under different physiological conditions by the coordinated activation of behavioral and autonomic thermal responses. It is, therefore, adjustable according to the nature of the neurotransmitter/neuromodulator-mediated inputs to the CNS thermal set point mechanism. Whether or not neuroactive compounds that alter body temperature do so via a direct action on the set point mechanism or by a direct action on downstream thermal effector mechanisms (e.g., neurons that control individual heat production or heat loss responses, such as cutaneous vasoconstriction or vasodilation) can be determined by assessing the nature of motivated behavioral thermal responses activated during the drug-induced autonomically-mediated change in Tb [25,26]. Thus, if the behavioral response complements the autonomically-induced change in Tb (i.e., both serve to raise Tb or both serve to lower Tb), then the set point has been changed. Alternatively, if behavioral and autonomic thermal responses diverge, then drug action has occurred at the sites of thermal effector control neurons. Accordingly, the coordinated actions of T R H in the PO/AH (the principal CNS site for coordinating and adjusting the activity of thermoregulatory systems [27,28]) to lower Tb, decrease metabolic rate [22], and evoke a complementary behavioral response provide strong evidence that T R H is acting to decrease Tsp in the awake ground squirrel. In the heat escape paradigm, a significant linear dose response trend was evident for the changes in Tb and bar-press rate between 0 and 100 ng of TRH, but was lost at the 1000 ng dose level at which other effects became apparent. The intense grooming behavior observed in one animal that resulted in complete cessation of bar pressing was the major source of the increased variation in heat escape behavior at 1000 ng. Although grooming may be interpreted as an appropriate thermoregulatory response [28] by providing another means of heat loss through saliva spreading, such behavior disrupts bar press responding for heat escape and cool air reinforcement. Furthermore, we have
66 observed in other experimental paradigms that spontaneous active grooming is typically associated with a rise in Tb in these animals. Thus, it is likely that the increase in metabolic activity associated with grooming behavior would have countered the effect of any mechanisms that might have been activated to lower Tb. The excessive chirping behavior that was evoked by 1000 ng of TRH in the other three animals may indicate the activation of other neural systems that control an 'alarm' response. Whether this effect represents high-dose pharmacological non-specific actions of T R H or the specific action of T R H that has diffused to neurons outside the PO/AH thermoregulatory network, where TRH receptors are also present [7], is not clear. Nevertheless, neural networks that drive an alarm response may utilize similar effector systems (respiratory and cardiovascular mechanisms, for example). Thus, the increased variability that occurred in Tb at the highest dose could also be due to conflicting priorities and competition for effector mechanisms. We considered the possibility that bar-press responding might have become erratic or irregular and, thereby, account for the ineffectiveness of heat escape behavior to significantly lower Tb in two of the three animals. This, however, did not occur. As noted earlier, T R H microinjection typically resulted in a sudden but brief cessation of heat escape behavior and a small, transient increase in Tb ( < 0.3 °C) before hypothermia was produced. Given the ability of T R H to cause behavioral sedation and reduced resting electromyographic activity in the awake ground squirrel [22], it is possible that such an effect in the present behavioral paradigm was responsible for the initial disruption in bar pressing activity, thus allowing the animal's T b to increase. It is also possible that, despite the slow rate of injection, ancillary mechanical effects could have led to the transient interruption in bar press behavior. A decrease in heat escape behavior also occurred following control injections; however, in these tests, the decrease was more gradual and was not characterized by a complete interruption of bar press behavior that was observed with TRH. Perhaps both factors played a role following TRH. Although Tb was typically quite high in the initial stages of most training and test sessions, it did not alarm us because the animals showed no signs of heat stress. Compared to rats, thermoregulatory control in ground squirrels is more labile, and the animals characteristically exhibit a 2 to 3 °C rise in Tb during the few minutes of handling required to prepare them for the test session. Also, the animals were welltrained to respond for cool air reinforcement and heat escape, and they performed the bar-press response after a latency that was consistent for each animal. The notion that TRH plays an important role in mammalian thermoregulation, in general, is also supported by other lines of evidence. For example, cold stress or a pyrogenic substance has been shown to elevate TRH levels in the rat hypothalamus [29]. Additional support from biochemical studies includes the identification of TRHcontaining fibers and cell bodies in the PO/AH of the rat [30-33], the presence of TRH receptors in the PO/AH of several species [34,35] including C. lateralis [7], and the presence of pro-TRH messenger RNA in the preoptic area of the rat [36]. In the rat, microiontophoresis of T R H in the PO/AH, wherein thermosensitive neurons are most highly concentrated [27], has been shown to alter neuronal firing rate [37,38]. Moreover, the effect of T R H on the firing rate of specifically identified thermosensitive
67 n e u r o n s was shown to be consistent with the characteristic T R H - i n d u c e d h y p e r t h e r m i a in the rat at n o r m a l r o o m t e m p e r a t u r e s [38]. F u r t h e r evidence that T R H plays a p r o m i n e n t role in thermoregulation is p r o v i d e d by our finding that the density o f T R H receptors in the medial preoptic area is significantly d e c r e a s e d in hibernating g r o u n d squirrels c o m p a r e d to winter euthermic (i.e., not hibernating) g r o u n d squirrels [7]. It is particularly n o t e w o r t h y that the preoptic area was one o f only a few areas o f the brain that u n d e r w e n t significant s t a t e - d e p e n d e n t (i.e., hibernation-related) changes in receptor density. G i v e n that a lowering o f Tsp is one o f the hallmarks o f hibernation, changes in T R H receptor density in this area o f primary i m p o r t a n c e to thermoregulation further u n d e r s c o r e s the potential significance o f T R H in thermoregulatory processes.
Acknowledgments This w o r k was s u p p o r t e d by California State University, Long Beach a w a r d s to T.L.S. and by the R i c h a r d B. L o o m i s A w a r d to J . H . H . The authors wish to t h a n k Dr. A l e x a n d e r L. B e c k m a n for his valuable assistance a n d advice in establishing the p a r a m e t e r s for training the g r o u n d squirrels in the h e a t - e s c a p e paradigm.
References 1 Boler, R.M., Enzman, F., Folkers, K., Bowers, C.Y. and Schally, A.V., The identity of chemical and hormonal properties of the thyrotropin releasing hormone and pyroglutamyl-histidyl-proline amide, Biochem. Biophys. Res. Commun., 37 (1969) 705-710. 2 Burgus, R., Dunn, T. F., Desiderio, D. and Guillemin, R.G., Structure moleculaire du facteur hypothalamique hypophysiotrope TRH origine ovine: Mise en evidence por spectrometric de masse de la sequence PCA-His-Pro-NH2, C.R. Acad. Sci., 269 (1969) 1870-1873. 3 Griftiths, E.C., Thyrotropin releasing hormone: endocrine and central effects, Psychoneuroendocrinology, 10 (1985) 225-235. 4 Horita, A., Kalivas, P.W. and Simasko, S. M., Thyrotropin releasing hormone (TRH): possible physiological functions not related to the neuroendocrine system, Rev. Pure Appl. Pharmacol. Sci., 4 (1983) 111-137. 5 Jackson, I.M.D., Thyrotropin releasing hormone, N. Engl. J. Med., 306 (1982) 145-155. 6 Stanton, T. L., Winokur, A. and Beckman, A.L., Seasonal variation in TRH content of different brain regions and the pineal in the mammalian hibernator, Citellus lateralis, Regul. Pept., 3 (1982) 135-144. 7 Caine, S.B., Stanton, T. L. and Winokur, A., Changes in TRH receptors in the brain of hibernating and euthermic ground squirrels, Soc. Neurosci. Abstr., 14 (1988) 980. 8 Clark, W.G. and Lipton, J.M., Brain and pituitary peptides in thermoregulation, Pharmac. Ther., 22 (1983) 249-297. 9 Breese, G. R., Cott, J. M., Cooper, B. R., Prange, A. J. Jr. and Lipton, M., Antagonism of ethanol narcosis by thyrotropin releasing hormone, Life Sci., 14 (1974) 1053-1063. 10 Breese, G.R, Cott, J.M., Cooper, B.R., Prange, A.J., Jr., Lipton, M. and Plotinoff, N.P., Effects of thyrotropin releasing hormone (TRH) on the actions of pentobarbital and other centrally acting drugs, J. Pharm. Exp. Ther., 193 (1975) 11-22. 11 Prange, A.J., Jr., Breese, G.R., Cott, J.M., Martin, B.R., Cooper, B.R., Wilson, I.C. and Plotnikoff, N.P., Thyrotropin releasing hormone: Antagonism of pentobarbital in rodents, Life Sci., 14 (1974) 447-455. 12 Chi, M.L. and Lin, M.T., Involvement of adrenergic receptor mechanisms within hypothalamus in the
68
13 14 15 16 17
18
19 20
21 22 23 24
25 26
27 28 29
30
31 32 33
34
fever induced by amphetamine and thyrotropin releasing hormone in the rat, J. Neural Transm., 58 (1983) 213-222. Lin, M.T. and Yang, C.P., Effects of thyrotropin releasing hormone thermoregulatory responses and hypothalamic neuronal activity in rats, Ann. N.Y. Acad. Sci., 553 (1989) 579-581. Griffiths, E.C., Rothwell, N.J. and Stock, M.J., Thermogenic effects of thyrotropin releasing hormone and its analogues in the rat, Experientia, 44 (1988) 40-42. Lin, M.T., Chandra, A., Chern, Y.F. and Tsay, B. L., Effects of thyrotropin releasing hormone (TRH) on thermoregulation in the rat, Experientia, 36 (1980) 177-181. Paakkari, I., Vonhop, S. and Feurerstein, G., Respiratory effects of TRH in conscious unrestrained rats, Ann. N.Y. Acad. Sci., 553 (1989) 514-516. Prasad, C., Matsui, T., Williams, J. and Peterkofsky, A., Thermoregulation in rats: opposing effects of thyrotropin releasing hormone and its metabolite histidyl-proline diketopiperazine, Biochem. Biophys. Res. Commun., 85 (1978) 1582-1587. Metcalf, G., Dettmar, P.W. and Watson, T., The role ofneuropeptides in thermoregulation, In B. Cox, P. Lomax, A.S. Milton, and E. Schonbaum (Eds.), Thermoregulatory Mechanisms and Their Therapeutic Implications, 4th Int. Symp. on the Pharmacology ofThermoregulation, Karger, Basel, 1980, pp. 175-179. Metcalf, G., TRH: a possible mediator of thermoregulation, Nature, 252 (1974) 310-311. Myers, R.D., Metcalf, G. and Rice, J. C., Identification by microinjection of TRH-sensitive sites in the cat's brain stem that mediate respiratory, temperature and other autonomic changes, Brain Res., 126 (1977) 105-115. Stanton, T.L., Winokur, A. and Beckman, A.L., Reversal of natural CNS depression by TRH action in the hippocampus, Brain Res., 181 (1980) 470-475. Stanton, T.L., Beckman, A. L. and Winokur, A., Thyrotropin releasing hormone effects in the central nervous system: dependence on arousal state, Science, 214 (1981) 678-681. Stanton, T. L. and Beckman, A. L., Behavioral and physiological thermoregulatory responses following central administration of thyrotropin releasing hormone in ground squirrels, Physiologist, 25 (1982) 305. Joseph, A.A., Knigge, K.A., Kalejs, L.M., Hoffman, R.A. and Reid, P. A stereotaxic atlas of the brain of the 13-lined ground squirrel (Citellus tridecemlineatus), U.S. Army Edgewood Arsenal Spec. Publ. (1966) 100-112. Satinoff, E., Drugs and thermoregulatory behavior, In P. Lomax and E. Schonbaum (Eds.), Body temperature, drug effects and therapeutic implications, Marcel Dekker, New York, 1979, pp. 151-t58. Corbit, J. D., Behavioral regulation of body temperature, In J. Hardy, A. P. Gagge and J. Stolwijk (Eds.), Physiological and Behavioral Temperature Regulation, Charles C. Thomas, Springfield, 1970, pp. 777-801. Boulant, J.A. and Dean, J.B.,Temperature receptors in the central nervous system, Annu. Rev. Physiol., 48 (1986) 639-654. Satinoff, E., Neural organization and evolution of thermal regulation in mammals, Science, 201 (1978) 16-21. Lin, M. T., Wang, P. S., Chuang, J., Fan, L.J. and Won, S.J., Cold stress or a pyrogenic substance elevates thyrotropin releasing hormone levels in the rat hypothalamus and induces thermogenic reactions, Neuroendocrinology, 50 (1989) 177-181. Hokfelt, T., Fuxe, K., Johannson, S., Jeffcoate, S. and White, N., Distribution of thyrotropin releasing hormone (TRH) in central nervous system as revealed with immunohistochemistry, Eur. J. Pharmacol., 34 (1975) 389-392. Kreider, M. S., Engber, T.M., Nilaver, G., Zimmerman, E.A. and Winokur, A., Immunohistochemical localization of TRH in rat CNS: comparison with RIA studies, Peptides, 6 (1985) 997-1000. Lechan, R.M. and Jackson, I.M.D., Immunohistochemical localization of thyrotropin releasing hormone in the rat hypothalamus, Endocrinology, 112 (1982) 55-65. Simerly, R. B., Gorski, R.A. and Swanson, L.W., Neurotransmitter specificity of cells and fibers in the medial preoptic nucleus: an immunohistochemical study in the rat, J. Comp. Neurol., 246 (1986) 343-363. Manaker, S., Rainbow, T.C. and Winokur, A., Thyrotropin releasing hormone (TRH) receptors: localization in rat and human central nervous system, In Quantitative Receptor Autoradiography, Alan R. Liss, 1986, pp. 103-135.
69 35 Taylor, R.L. and Burt, D.R., Species differences in the brain regional distribution of receptor binding for thyrotropin releasing hormone, J. Neurochem., 38 (1982) 1649-1656. 36 Lechan, R.M. and Segerson, T.D., Pro-TRH gene expression and precursor peptides in rat brain, Ann. N.Y. Acad. Sci., 553 (1989) 29-59. 37 Winokur, A. and Beckman, A.L., Effects of thyrotropin releasing hormone, norepinephrine and acetylcholine on the activity of neurons in the hypothalamus, septum and cerebral cortex of the rat, Brain Res., 150 (1978) 205-209. 38 Salzman, S.K. and Beckman, A.L., Effects ofthyrotropin releasing hormone on hypothalamic thermosensitive neurons of the rat, Brain Res. Bull., 7 (1981) 325-332.