- Email: [email protected]
The effects of naloxone administered into the periaqueductal gray on shock-elicited freezing behavior in the rat
BEHAVIORAL AND NEURAL BIOLOGY 46, 189--195 (1986)
The Effects of Naloxone Administered into the Periaqueductal Gray on Shock-Elicited Freezing Behavior in the Rat G A R Y D . HAMMER AND BRUCE S . K A P P 1
Department of Psychology, The University of Vermont, Burlington, Vermont 05405
Freezing behavior that occurs following f o o t s h o c k was found to increase in rats in which naloxone was injected into the ventrolateral region of the mesencephalic periaqueductal gray (PAG) area of the brain prior to f o o t s h o c k administration. Since naloxone administered into the ventrolateral region of the PAG induced minimal freezing in rats which did not receive footshock, the results suggest that the effect of n a l o x o n e on shock-induced freezing is not due to a nonspecific decrease in m o t o r activity. N a l o x o n e had no effect on freezing when injected into the dorsolateral region of the PAG. The data are consistent with the theory that conditioned fear induces opioid mediated analgesia, and that the ventrolateral region of the PAG is an important component of a pain-inhibitory system involved in this analgesia. © 1986Academic Press, Inc.
Microinjections of morphine into and electrical stimulation of the periaqueductal gray (PAG) produce profound analgesia, suggesting the existence of an endogenous opioid-mediated analgesia system (Akil, Mayer, & Liebeskind, 1976; Gebhart, Sandkuhler, Thalhammer, & Zimmerman, 1984; Yaksh, 1979). Recent research has been concerned with determining conditions which activate this system. For example, Bolles and Fanselow (1980) have presented a model of fear and pain in which they suggest that analgesia is produced during fear. Fear, triggered by either innate or conditioned threatening stimuli, inhibits pain via release of endogenous opioids. Without such an analgesia system, if injured and confronted by a predator an animal might engage in pain-induced recuperative behaviors rather than fear-induced defensive behaviors, thereby rendering it vulnerable to further injury and death. Evidence in support of fear-induced analgesia has been presented by Fanslow and Bolles (1979). Rats were administered three footshocks (20 sec ITI) after which freezing, an index of conditioned fear to apparatus cues, was measured. If fear, conditioned to apparatus cues following the Address correspondence and requests for reprints to Dr. Bruce S. Kapp. 189 0163-1047/86 $3.00
Copyright© 1986by AcademicPress, Inc. All rights of reproductionin any form reserved.
190
HAMMER AND KAPP
first shock, induces opioid-mediated analgesia, then administration of the opiate antagonist, naloxone, should render the subsequent two shocks more painful. Increased pain rendered by these shocks should strengthen the association between apparatus cues and footshocks. A stronger association would increase conditioned fear, manifested in increased freezing following shock. The results reported by Fanselow and Bolles (1979) support this interpretation. When compared to rats injected with saline, rats injected with naloxone prior to footshock showed a significant increase in freezing following either two or three shocks. Since Fanselow and Bolles (1979) administered naloxone intraperitoneaUy, no inference can be made concerning the brain region(s) contributing to fear-induced analgesia. However, it is reasonable to propose that the PAG may play an important role. In the present study it was hypothesized that injections of naloxone into the PAG prior to footshock, when compared with saline injections, would produce increased freezing in rats following the third of three footshocks. This result would be consistent with those of Fanselow and Bolles (1979), who suggest that a PAG opioid system is a component of a pain-inhibitory system activated during fear, and provide insights into the brain circuitry contributing to fear-induced analgesia. Sixty-two male Sprague-Dawley rats weighing 200-250 g were used. They were maintained on food and water ad libitum and a 12-hr lightdark cycle (lights on, 7:00 AM to 7:00 PM). Behavioral observations were performed between 1:00 and 4:00 PM. A 25.3 × 25.3 × 39 cm Plexiglas box with grid floor and located within an illuminated, sound attenuating chamber was used for behavioral testing. A bath of Joy detergent was placed beneath the grid floor as an odor deterrent. The box was cleaned prior to introduction of each rat. Rats were anesthetized with Nembutal (30-50 mg/kg) and a 23-gauge cannula was implanted unilaterally within either the dorsolateral or ventrolateral quadrant of the mesencephalic PAG. Coordinates for the ventrolateral quadrant were 6.4 mm posterior to bregma, 2.3 mm lateral to the midline, and 5.7 mm ventral to dura. Those for the dorsolateral quadrant were 6.4 mm posterior to bregma, 2.3 mm lateral to the midline, and 4.4 mm ventral (Pellegrino & Cushman, 1982). Rats were assigned to one of five groups (n = 12). Following a 1- to 2-week postoperative period, rats in three groups with a cannula in the ventrolateral PAG received either an injection of naloxone prior to footshock (group V N + ) , an injection of saline prior to footshock (group VS +), or an injection of naloxone in the absence of footshock (group VN - ) . Group VN - was included to determine if any apparent increase in freezing in group VN + might be attributed to a nonspecific depression of motor activity by naloxone. Rats in two additional groups with a cannula in the dorsolateral PAG received either an injection of naloxone
NALOXONE AND RAT FREEZING BEHAVIOR
191
(group DN + ) or saline (group DS +) prior to footshock. Naloxone was administered in a 5.0-/zg dose dissolved in 0.5/zl of isotonic saline. Saline control injections were of a 0.5-/zl volume. The rationale for cannula implantation into the ventrolateral PAG was based on the observation that this region is the most sensitive PAG region for the production of analgesia following microinjections of morphine in the rat (Yaksh, 1979). The dorsolateral region of the PAG, when compared to the ventrolateral PAG, is relatively insensitive to morphine administration in the induction of analgesia. Since implantation sites in the dorsolateral and ventrolateral regions were equidistant from the cerebral aqueduct, injections into the dorsolateral region were made to determine if any effects of ventrolateral naloxone injections might be due to its diffusion to other brain areas via spread into the aqueduct. Immediately following microinjection each rat was placed into the Plexiglas box. Two minutes later, all rats with the exception of those in Group V N - received three footshocks (0.75 sec; 0.5 mA; 20 sec ITI). Immediately following the third footshock (or after 2 rain and 40 see in Group V N - ) , they were observed for freezing behavior, defined as the absence of all skeletal movement, every 5 sec for 8 min. The data were converted to the percentage of the 12 total observations in each minute of the 8-rain period during which freezing occurred. Following behavioral testing, each rat was perfused, and frozen sections (90/zm) were taken through the PAG and stained with thionin. Cannula tip locations were analyzed microscopically with reference to the Pellegrino and Cushman (1982) atlas. A number of animals were rejected from the subsequent analysis because cannula tips were located either adjacent to the PAG or in the aqueduct. Hence, the number of animals in each group was as follows: Group DS +, n = 6; Group VS +, n = 5; Group D N + , n = 7; Group V N + , n = 6; and Group V N - , n = 7. Cannula tip locations for these groups were plotted in Fig. 1. The mean percentage freezing scores are plotted in Fig. 2. A twofactor (Groups × Minutes) analysis of variance with repeated measures on Minutes demonstrated a significant Group effect [F(4, 26) = 12.41, p < .025]. There was no significant effect of Minutes [F(7, 182) = 1.37, p > .05], nor was there a significant Group × Minutes interaction [F(28, 182) < 1, p > .05]. Since there was no effect of Minutes, a single mean was calculated for each animal's percentage freezing score across the 8-min period. This resulted in the following group means: Group DS+ = 34.9%; Group V S + = 21.9%; Group D N + = 43.3%; Group V N + = 82.8%; Group V N - = 8.0%. A Duncan's multiple range test revealed that Group VN + demonstrated a significantly greater percentage freezing score than Group V S + (p < .01), Group V N - (p < .01), and Group D N + (p < .01). Group DN + did not differ significantly from Group DS +.
192
HAMMER AND KAPP
-- 6.2
( FLM
_
.
l LL
FI~. 1. Location of injection stylet tips within the mesencephalic periaqueductal gray of the rat. A, Group VN + ; ©, Group DN + ; @, Group DS + ; A, Group VS + ; A, Group V N - . See text for detailed description of various treatment groups. Histological plates taken from Pellegrino and Cushman (1982).
NALOXONE AND RAT FREEZING BEHAVIOR
193
90"
80
70'
~
60'
~
50"
40"
30-
20-
,o-
/ MINUTES
F~G. 2. Mean percentage freezing scores for the various treatment groups for each minute following the third footshock. Refer to Fig. 1 caption for group designation.
The results demonstrated that when compared with saline injections, naloxone injections into the ventrolateral region of the PAG produced a significant increase in freezing following footshock. Naloxone injected into the ventrolateral region of the PAG resulted in minimal freezing in the group that did not receive footshock, suggesting that naloxone did not simply decrease motor activity and thereby produce an apparent increase in freezing. Only the group receiving injections into the ventrolateral PAG demonstrated a significant increase in freezing when compared with its respective saline control group. This group also demonstrated a significant increase in freezing when compared with the group receiving naloxone injections into the dorsolateral PAG. Since sites located within the dorsolateral and ventrolateral PAG were generally equidistant from the cerebral aqueduct, diffusion of the drug to distant tissue via the aqueduct cannot account for the increased freezing following injections into the ventrolateral PAG. Furthermore, the site specificity observed in this experiment is consistent with the findings of Yaksh (1979) who demonstrated that injections of morphine into the ventrolateral PAG produced a more potent
194
HAMMER AND KAPP
analgesic effect, as measured by either paw lick or tail flick latency in response to heat, than injections into the dorsolateral PAG. The increase in shock-elicited freezing produced by intracranial naloxone in this study is similar to that reported by Fanselow and Bolles (1979) following intraperitoneal injections of naloxone given prior to footshock. The results of the present study are also consistent with their interpretation that conditioned fear activates an opioid-mediated analgesia system. Of additional importance, these results, taken together with the evidence that the PAG is a source of a descending system which contributes to opiate-mediated analgesia (Fields & Basbaum, 1979), suggest that the ventrolateral PAG is an important component of a circuit which contributes to conditioned fear-induced opioid-mediated analgesia. A particularly interesting question for future research arises concerning the neural circuitry by which conditioned emotional states such as fear engage the PAG descending system to induce analgesia. In this context, it is of interest that the ventrolateral PAG is recipient of direct projections from the amygdaloid central nucleus (Hopkins & Holstege, 1978). Furthermore, evidence suggests that the amygdala, and in particular the amygdaloid central nucleus, contributes to the arousal of fear, as well as to the acquisition of a variety of responses during Pavlovian fear conditioning (Kapp, Pascoe, & Bixler, 1984; Werka, Sk/ir, & Ursin, 1978). Investigations into the interactions of the amygdala and the PAG would appear to be an important area for future investigations designed to delineate the neural substrates of fear-induced analgesia. REFERENCES Akil, E., Mayer, D. J., & Liebeskind, J. C. (1976). Antagonism of stimulation-produced analgesia by naloxone, a narcotic antagonist. Science, 191, 961-962. Bolles, R. C., & Fanselow, M. S. (1980). A perceptual-defensive-recuperative model of fear and pain. Behavioral and Brain Sciences, 3, 121-131. Fanselow, M. S., & Bolles, R. C. (1979). Naloxone and shock-elicited freezing in the rat. Journal of Comparative and Physiological Psychology, 93, 736-744. Fields, H. L., & Basbaum, A. I. (1979). Anatomy and physiology of a descending pain control system. In J. J. Bonica, J. C. Liebeskind, & D. G. Albe-Fessard (Eds.), Advances in pain research and therapy (Vol. 3, pp. 427-440). New York: Raven Press. Gebhart, G. F., Sandkuhler, J., Thalhammer, J. G., & Zimmerman, M. (1984). Inhibition in spinal cord of nociceptive information by electrical stimulation and morphine microinjection at identical sites in midbrain of the cat. Journal of Neurophysiology, 51, 75-89. Hopkins, D. A., & Holstege, G. (1978). Amygdaloid projections to the mesencephalon, pons, and medulla oblongata in the cat. Experimental Brain Research, 32, 529-547. Kapp, B. S., Pascoe, J. P., & Bixler, M. A. (1984). The amygdala: A neuroanatomical systems approach to its contribution to aversive conditioning. In L. R. Squire & N. Butters (Eds.), The neuropsychology of memory (pp. 473-488). New York: Guilford Press. Pellegrino, L. J., & Cushman, A. J. (1982). A Stereotaxic atlas of the rat brain. New York: Appleton-Century-Crofts.
NALOXONE AND RAT FREEZING BEHAVIOR
195
Werka, T., Sk/ir, J., & Ursin, H. (1978). Exploration and avoidance in rats with lesions in the amygdala and piriform cortex. Journal of Comparative and Physiological Psychology, 92, 672-681. Yaksh, T. L. (1979). Central nervous system sites mediating opiate analgesia. In J. J. Bonica, J. C. Leibeskind, & D. C. Albe-Fessard (Eds.), Advances in pain research and therapy (Vol. 3, pp. 411-426). New York: Raven Press.
The Effects of Naloxone Administered into the Periaqueductal Gray on Shock-Elicited Freezing Behavior in the Rat G A R Y D . HAMMER AND BRUCE S . K A P P 1
Department of Psychology, The University of Vermont, Burlington, Vermont 05405
Freezing behavior that occurs following f o o t s h o c k was found to increase in rats in which naloxone was injected into the ventrolateral region of the mesencephalic periaqueductal gray (PAG) area of the brain prior to f o o t s h o c k administration. Since naloxone administered into the ventrolateral region of the PAG induced minimal freezing in rats which did not receive footshock, the results suggest that the effect of n a l o x o n e on shock-induced freezing is not due to a nonspecific decrease in m o t o r activity. N a l o x o n e had no effect on freezing when injected into the dorsolateral region of the PAG. The data are consistent with the theory that conditioned fear induces opioid mediated analgesia, and that the ventrolateral region of the PAG is an important component of a pain-inhibitory system involved in this analgesia. © 1986Academic Press, Inc.
Microinjections of morphine into and electrical stimulation of the periaqueductal gray (PAG) produce profound analgesia, suggesting the existence of an endogenous opioid-mediated analgesia system (Akil, Mayer, & Liebeskind, 1976; Gebhart, Sandkuhler, Thalhammer, & Zimmerman, 1984; Yaksh, 1979). Recent research has been concerned with determining conditions which activate this system. For example, Bolles and Fanselow (1980) have presented a model of fear and pain in which they suggest that analgesia is produced during fear. Fear, triggered by either innate or conditioned threatening stimuli, inhibits pain via release of endogenous opioids. Without such an analgesia system, if injured and confronted by a predator an animal might engage in pain-induced recuperative behaviors rather than fear-induced defensive behaviors, thereby rendering it vulnerable to further injury and death. Evidence in support of fear-induced analgesia has been presented by Fanslow and Bolles (1979). Rats were administered three footshocks (20 sec ITI) after which freezing, an index of conditioned fear to apparatus cues, was measured. If fear, conditioned to apparatus cues following the Address correspondence and requests for reprints to Dr. Bruce S. Kapp. 189 0163-1047/86 $3.00
Copyright© 1986by AcademicPress, Inc. All rights of reproductionin any form reserved.
190
HAMMER AND KAPP
first shock, induces opioid-mediated analgesia, then administration of the opiate antagonist, naloxone, should render the subsequent two shocks more painful. Increased pain rendered by these shocks should strengthen the association between apparatus cues and footshocks. A stronger association would increase conditioned fear, manifested in increased freezing following shock. The results reported by Fanselow and Bolles (1979) support this interpretation. When compared to rats injected with saline, rats injected with naloxone prior to footshock showed a significant increase in freezing following either two or three shocks. Since Fanselow and Bolles (1979) administered naloxone intraperitoneaUy, no inference can be made concerning the brain region(s) contributing to fear-induced analgesia. However, it is reasonable to propose that the PAG may play an important role. In the present study it was hypothesized that injections of naloxone into the PAG prior to footshock, when compared with saline injections, would produce increased freezing in rats following the third of three footshocks. This result would be consistent with those of Fanselow and Bolles (1979), who suggest that a PAG opioid system is a component of a pain-inhibitory system activated during fear, and provide insights into the brain circuitry contributing to fear-induced analgesia. Sixty-two male Sprague-Dawley rats weighing 200-250 g were used. They were maintained on food and water ad libitum and a 12-hr lightdark cycle (lights on, 7:00 AM to 7:00 PM). Behavioral observations were performed between 1:00 and 4:00 PM. A 25.3 × 25.3 × 39 cm Plexiglas box with grid floor and located within an illuminated, sound attenuating chamber was used for behavioral testing. A bath of Joy detergent was placed beneath the grid floor as an odor deterrent. The box was cleaned prior to introduction of each rat. Rats were anesthetized with Nembutal (30-50 mg/kg) and a 23-gauge cannula was implanted unilaterally within either the dorsolateral or ventrolateral quadrant of the mesencephalic PAG. Coordinates for the ventrolateral quadrant were 6.4 mm posterior to bregma, 2.3 mm lateral to the midline, and 5.7 mm ventral to dura. Those for the dorsolateral quadrant were 6.4 mm posterior to bregma, 2.3 mm lateral to the midline, and 4.4 mm ventral (Pellegrino & Cushman, 1982). Rats were assigned to one of five groups (n = 12). Following a 1- to 2-week postoperative period, rats in three groups with a cannula in the ventrolateral PAG received either an injection of naloxone prior to footshock (group V N + ) , an injection of saline prior to footshock (group VS +), or an injection of naloxone in the absence of footshock (group VN - ) . Group VN - was included to determine if any apparent increase in freezing in group VN + might be attributed to a nonspecific depression of motor activity by naloxone. Rats in two additional groups with a cannula in the dorsolateral PAG received either an injection of naloxone
NALOXONE AND RAT FREEZING BEHAVIOR
191
(group DN + ) or saline (group DS +) prior to footshock. Naloxone was administered in a 5.0-/zg dose dissolved in 0.5/zl of isotonic saline. Saline control injections were of a 0.5-/zl volume. The rationale for cannula implantation into the ventrolateral PAG was based on the observation that this region is the most sensitive PAG region for the production of analgesia following microinjections of morphine in the rat (Yaksh, 1979). The dorsolateral region of the PAG, when compared to the ventrolateral PAG, is relatively insensitive to morphine administration in the induction of analgesia. Since implantation sites in the dorsolateral and ventrolateral regions were equidistant from the cerebral aqueduct, injections into the dorsolateral region were made to determine if any effects of ventrolateral naloxone injections might be due to its diffusion to other brain areas via spread into the aqueduct. Immediately following microinjection each rat was placed into the Plexiglas box. Two minutes later, all rats with the exception of those in Group V N - received three footshocks (0.75 sec; 0.5 mA; 20 sec ITI). Immediately following the third footshock (or after 2 rain and 40 see in Group V N - ) , they were observed for freezing behavior, defined as the absence of all skeletal movement, every 5 sec for 8 min. The data were converted to the percentage of the 12 total observations in each minute of the 8-rain period during which freezing occurred. Following behavioral testing, each rat was perfused, and frozen sections (90/zm) were taken through the PAG and stained with thionin. Cannula tip locations were analyzed microscopically with reference to the Pellegrino and Cushman (1982) atlas. A number of animals were rejected from the subsequent analysis because cannula tips were located either adjacent to the PAG or in the aqueduct. Hence, the number of animals in each group was as follows: Group DS +, n = 6; Group VS +, n = 5; Group D N + , n = 7; Group V N + , n = 6; and Group V N - , n = 7. Cannula tip locations for these groups were plotted in Fig. 1. The mean percentage freezing scores are plotted in Fig. 2. A twofactor (Groups × Minutes) analysis of variance with repeated measures on Minutes demonstrated a significant Group effect [F(4, 26) = 12.41, p < .025]. There was no significant effect of Minutes [F(7, 182) = 1.37, p > .05], nor was there a significant Group × Minutes interaction [F(28, 182) < 1, p > .05]. Since there was no effect of Minutes, a single mean was calculated for each animal's percentage freezing score across the 8-min period. This resulted in the following group means: Group DS+ = 34.9%; Group V S + = 21.9%; Group D N + = 43.3%; Group V N + = 82.8%; Group V N - = 8.0%. A Duncan's multiple range test revealed that Group VN + demonstrated a significantly greater percentage freezing score than Group V S + (p < .01), Group V N - (p < .01), and Group D N + (p < .01). Group DN + did not differ significantly from Group DS +.
192
HAMMER AND KAPP
-- 6.2
( FLM
_
.
l LL
FI~. 1. Location of injection stylet tips within the mesencephalic periaqueductal gray of the rat. A, Group VN + ; ©, Group DN + ; @, Group DS + ; A, Group VS + ; A, Group V N - . See text for detailed description of various treatment groups. Histological plates taken from Pellegrino and Cushman (1982).
NALOXONE AND RAT FREEZING BEHAVIOR
193
90"
80
70'
~
60'
~
50"
40"
30-
20-
,o-
/ MINUTES
F~G. 2. Mean percentage freezing scores for the various treatment groups for each minute following the third footshock. Refer to Fig. 1 caption for group designation.
The results demonstrated that when compared with saline injections, naloxone injections into the ventrolateral region of the PAG produced a significant increase in freezing following footshock. Naloxone injected into the ventrolateral region of the PAG resulted in minimal freezing in the group that did not receive footshock, suggesting that naloxone did not simply decrease motor activity and thereby produce an apparent increase in freezing. Only the group receiving injections into the ventrolateral PAG demonstrated a significant increase in freezing when compared with its respective saline control group. This group also demonstrated a significant increase in freezing when compared with the group receiving naloxone injections into the dorsolateral PAG. Since sites located within the dorsolateral and ventrolateral PAG were generally equidistant from the cerebral aqueduct, diffusion of the drug to distant tissue via the aqueduct cannot account for the increased freezing following injections into the ventrolateral PAG. Furthermore, the site specificity observed in this experiment is consistent with the findings of Yaksh (1979) who demonstrated that injections of morphine into the ventrolateral PAG produced a more potent
194
HAMMER AND KAPP
analgesic effect, as measured by either paw lick or tail flick latency in response to heat, than injections into the dorsolateral PAG. The increase in shock-elicited freezing produced by intracranial naloxone in this study is similar to that reported by Fanselow and Bolles (1979) following intraperitoneal injections of naloxone given prior to footshock. The results of the present study are also consistent with their interpretation that conditioned fear activates an opioid-mediated analgesia system. Of additional importance, these results, taken together with the evidence that the PAG is a source of a descending system which contributes to opiate-mediated analgesia (Fields & Basbaum, 1979), suggest that the ventrolateral PAG is an important component of a circuit which contributes to conditioned fear-induced opioid-mediated analgesia. A particularly interesting question for future research arises concerning the neural circuitry by which conditioned emotional states such as fear engage the PAG descending system to induce analgesia. In this context, it is of interest that the ventrolateral PAG is recipient of direct projections from the amygdaloid central nucleus (Hopkins & Holstege, 1978). Furthermore, evidence suggests that the amygdala, and in particular the amygdaloid central nucleus, contributes to the arousal of fear, as well as to the acquisition of a variety of responses during Pavlovian fear conditioning (Kapp, Pascoe, & Bixler, 1984; Werka, Sk/ir, & Ursin, 1978). Investigations into the interactions of the amygdala and the PAG would appear to be an important area for future investigations designed to delineate the neural substrates of fear-induced analgesia. REFERENCES Akil, E., Mayer, D. J., & Liebeskind, J. C. (1976). Antagonism of stimulation-produced analgesia by naloxone, a narcotic antagonist. Science, 191, 961-962. Bolles, R. C., & Fanselow, M. S. (1980). A perceptual-defensive-recuperative model of fear and pain. Behavioral and Brain Sciences, 3, 121-131. Fanselow, M. S., & Bolles, R. C. (1979). Naloxone and shock-elicited freezing in the rat. Journal of Comparative and Physiological Psychology, 93, 736-744. Fields, H. L., & Basbaum, A. I. (1979). Anatomy and physiology of a descending pain control system. In J. J. Bonica, J. C. Liebeskind, & D. G. Albe-Fessard (Eds.), Advances in pain research and therapy (Vol. 3, pp. 427-440). New York: Raven Press. Gebhart, G. F., Sandkuhler, J., Thalhammer, J. G., & Zimmerman, M. (1984). Inhibition in spinal cord of nociceptive information by electrical stimulation and morphine microinjection at identical sites in midbrain of the cat. Journal of Neurophysiology, 51, 75-89. Hopkins, D. A., & Holstege, G. (1978). Amygdaloid projections to the mesencephalon, pons, and medulla oblongata in the cat. Experimental Brain Research, 32, 529-547. Kapp, B. S., Pascoe, J. P., & Bixler, M. A. (1984). The amygdala: A neuroanatomical systems approach to its contribution to aversive conditioning. In L. R. Squire & N. Butters (Eds.), The neuropsychology of memory (pp. 473-488). New York: Guilford Press. Pellegrino, L. J., & Cushman, A. J. (1982). A Stereotaxic atlas of the rat brain. New York: Appleton-Century-Crofts.
NALOXONE AND RAT FREEZING BEHAVIOR
195
Werka, T., Sk/ir, J., & Ursin, H. (1978). Exploration and avoidance in rats with lesions in the amygdala and piriform cortex. Journal of Comparative and Physiological Psychology, 92, 672-681. Yaksh, T. L. (1979). Central nervous system sites mediating opiate analgesia. In J. J. Bonica, J. C. Leibeskind, & D. C. Albe-Fessard (Eds.), Advances in pain research and therapy (Vol. 3, pp. 411-426). New York: Raven Press.