- Email: [email protected]
benzodiazepine receptor complex in the central amygdalar nucleus and stress ulcers in rats
BEHAVIORAL AND NEURAL BIOLOGY
51, 262-269 (1989)
The GABA/Benzodiazepine Receptor Complex in the Central Amygdalar Nucleus and Stress Ulcers in Rats R. M. SULLIVAN, P. G. HENKE, A. RAY, M. A. HEBERT, AND J. M. TRIMPER1 St. Francis Xavier University, Antigonish, Nova Scotia, Canada The effects of bilateral microinjections of chlordiazepoxide and GABA into the central amygdalar nucleus on gastric ulcer formation induced by cold-restraint were examined in chronically implanted Wistar rats. Higher doses of chlordiazepoxide (20 and 30/zg/amygdala) significantly reduced stress ulcer development, whereas a lower dose (2.5/xg) produced a nonsignificant increase in ulcer severity. A similar dose/response pattern was observed following GABA administration. The benzodiazepine receptor antagonist Ro15-1788, applied to the amygdala, abolished the protective effects of both chlordiazepoxide and GABA. In addition, when Ro15-1788 (10/zg) was injected into the amygdala by itself, it aggravated the gastric stress pathology. However, a lower dose (5/xg) had an attenuating effect, opposite to the pattern of effects produced by chlordiazepoxide and GABA. The role of the amygdalar GABA-benzodiazepine receptor complex in stressful conditions is discussed. © 1989AcademicPress, Inc.
Benzodiazepines are among the most widely prescribed drugs for the treatment of anxiety states (Shephard, 1986), and they demonstrate anticonflict activity in a variety of animal models of anxiety (Pellow & File, 1984; Quintero, Henney, Lawson, Mellanby, & Gray, 1985). It has been suggested that benzodiazepine receptors, widespread in the CNS (Young & Kuhar, 1980), may not only mediate the anxiolytic actions of benzodiazepines, but may be crucial in the pathogenesis of anxiety and its associated pathophysiological states in humans (Ninan et al., 1982). Benzodiazepines are reported to decrease stress-induced elevations of plasma corticosteroids (Lahti & Barsuhn, 1974), and they are known to reduce gastric ulcers induced by stress in rats (File & Pearce, 1981; Henke, 1987; Kunchandy & Kulkarni, 1987). Supported by the Natural Sciences and Engineering Research Council of Canada. The authors thank Hoffman-LaRoche Co., Ltd., for the gift of chlordiazepoxide and Ro151788, and also thank Ms. Jane Aker for the secretarial assistance in the preparation of the manuscript. Correspondence and reprint requests should be addressed to Peter G. Henke, Department of Psychology, St. Francis Xavier University, Antigonish, Nova Scotia, Canada, B2G 1C0. 262 0163-1047/89 $3.00 Copyright © 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
AMYGDALA AND STRESS
263
Benzodiazepine receptors are thought to interact with GABA (y-aminobutyric acid) receptors, as part of a supramolecular complex in neuronal cell membranes. Agonists at the benzodiazepine recognition site are believed to enhance GABAergic activity by facilitating the coupling of the GABA recognition site to the associated chloride ion channel, whereas inverse agonists (/~-carboline derivatives) are assumed to hinder this coupling (Costa, Corda, Epstein, Forcfietti, & Guidotti, 1983). GABA agonists, not surprisingly, share a number of behavioral effects with benzodiazepines. In addition to their anticonvulsant and sedative properties, centrally administered GABA agonists have been shown to produce anticonflict activity (Cananzi, Costa, & Guidotti, 1980; ScheelKr~iger & Petersen, 1982). GABA has also been reported to reduce stress ulcers in rats (Bhargava, Gupta, & Gupta, 1985; Ushijima et al., 1986). Furthermore, central GABAergic mechanisms have been suggested to be involved in both the ulcer attenuation and anticonflict activity produced by benzodiazepines (Kunchandy & Kulkarni, 1987; Scheel-Krfiger & Petersen, 1982). Several lines of evidence have implicated the central nucleus of the amygdala (CEA) as an important site in the response of the organism to stressful situations. For example, lesions in this area were found to attenuate stress-induced gastric pathology in rats (Henke, 1980), whereas low-level electrical stimulation of the amygdala produced gastric erosions only when localized in the CEA (Henke, 1985). Changes in multiple-unit activity of the amygdala have been observed in response to restraint stress and the behavior of some of these units was modified by ip injections of chlordiazepoxide (Henke, 1985). Additional recording studies in the CEA revealed that two multiple-unit activity profiles, recorded during restraint stress, were associated with stress ulceration, one with increased and one with decreased stomach pathology. The same activity profiles were differentially distributed in genetically selected Roman high- and low-avoidance rats, as well as Wistar rats differentiated on the basis of emotionality characteristics in an open-field test (Henke, 1988a). Interestingly, stimulation of the amygdala in humans has been reported to result in fear experiences (Gloor, Olivier, & Quesney, 1981). A recent study demonstrated that the increase in glucose utilization rate, seen in a number of brain areas following restraint and waterimmersion stress in rats, was predominant in the CEA, and this elevation was significantly reduced by bromazepam, at a non-muscle relaxant dose (Nakamura, Hayashi, & Nakamura, 1984). Furthermore, microinjections of benzodiazepines into the CEA of rats significantly increased punished responding in a conflict paradigm, suggestive of anxiolytic activity (Shibata, Kataoka, Gomita, & Ueki, 1982), but other amygdalar nuclei are also reportedly involved in this effect (Scheel-Krfiger & Petersen, 1982). The CEA has been shown to contain the highest concentration of
264
SULLIVAN ET AL.
GABA in the amygdalar complex (Ben-Ari, 1981), as well as a high density of benzodiazepine receptors (Young & Kuhar, 1980). We, therefore, investigated the effects of intraamygdalar applications of chlordiazepoxide, Ro15-1788, and GABA on stress ulcer formation in rats. METHOD Male Wistar rats, 90-120 days old (Charles River, Quebec) were individually housed, with free access to food and water. They were maintained at 22 ___ 2°C, on a 12-h light/dark cycle (lights on at 8 AM). For surgery, rats were anesthetized with sodium pentobarbital (50 mg/kg, ip) and secured in a stereotaxic apparatus. Twenty-three-gauge stainless-steel guide cannulae (Plastic Products) were lowered bilaterally to the dorsal edge of the central nucleus of the amygdala, using the following coordinates: 2.0 P (to bregma), 4.1 lateral (to midline), and 7.5 V (to dura), skull horizontal. Stainless-steel screws and dental acrylic secured the position of the cannulae. Following a 1-week recovery period, the rats were deprived of food (but not water) for 24 h prior to testing. The animals then received intraamygdalar injections of drug or vehicle in a volume of 2 ~1, slowly over a period of 1 min. The injection cannula, which did not protrude beyond the tip of the guide cannula, was attached to a 5/xl Hamilton syringe with microbore tubing and was left in position an additional 2 min to allow for diffusion of the injection solution. Following injections, the rats were immediately placed in Plexiglas restrainers (Fisher Scientific). After 3 h of cold-immobilization (4°C), the animals were sacrificed with an overdose of sodium pentobarbital (ip). The stomachs were dissected out and cut along the greater curvature, washed with cold water and examined microscopically for gastric mucosal lesions. The number of lesions, as well as their severity (cumulative length to the nearest 0.1 mm) were recorded. After removal of the stomachs, the rats were perfused intracardially with 0.9% saline and 10% formalin. The brains were removed, embedded in paraffin, sectioned at 15 /~m, and stained with thionin to determine the location of the cannula tips. The drugs used were chlordiazepoxide HCI, Ro15-1788 (HoffmanLaRoche), and GABA (from Sigma). Chlordiazepoxide (CDP) and GABA were dissolved in distilled water, whereas Ro15-1788 was suspended in distilled water with a drop of Tween 20. All injections were intraamygdalar and bilateral. Control animals received injections of either distilled water or distilled water with Tween 20. In the interaction groups, Ro15-1788 was injected 20 min prior to either CDP or GABA. All groups reported in the study consisted of six rats, except for the CDP (1.0/zg/am) group (n = 5). The results were analyzed using the Mann-Whitney U test (two-tailed) for multiple comparisons between groups. A p value of at least .05 was used as the level of significance for the statistical tests.
265
A M Y G D A L A A N D STRESS
TABLE 1 Effects of Intraamygdalar CDP, GABA, and Ro15-1788 on Stress-Induced Gastric Pathology in Rats (Means ± SEM)
Treatment (p-g/am)
n
Mean ulcer frequency
Controls
6
6.0 - 0.45
2.0 --- 0.31
CDP CDP CDP CDP CDP
6 6 6 6 5
1.2 1.3 4.5 7.2 6.0
+-+ -
0.31 a 0.21" 0.67 1.08 0.89
0.2 0.3 1.4 4.2 1.7
+± ± -+ _
0.06 a 0.04 a 0.31 1.21 0.22
6 6 6 6
1.8 2.3 4.3 5.5
-.+ 0.48 a -+ 0.33" -+ 0.49 -+ 0.72
0.4 0.7 3.0 1.7
+--. --±
0.10 ° 0.06 b 0.99 0.30
Ro15-1788 (10) Ro15-1788 (5) Ro15-1788 (2.5)
6 6 6
8.7 ± 1.48 3.8 - 0.48 b 5.8 - 0.70
5.7 -+ 1.80 ~ 1.1 - 0.26 2.0 -+- 0.56
Ro(10) + CDP (20) Ro(10) + G A B A (10)
6 6
6.8 -+ 0.60 5.2 _+ 0.79
2.1 -+ 0.20 2.3 --- 0.64
(30) (20) (10) (2.5) (1)
GABA GABA GABA GABA
(20) (10) (2.5) (1)
Mean severity (ram)
Note. All comparisons were by the M a n n - W h i t n e y U test (two-tailed). a p < .002. b p < .01 (compared to controls).
RESULTS
The control group consisted of three rats receiving distilled water injections, and three additional animals which were also injected with Tween 20. Both types of solutions produced similar pathology measures (mean ulcer frequency was six in each case, and mean severity was 1.8 and 2.2 ram, respectively). Consequently, the data were combined for statistical tests. Table 1 shows that intraamygdalar CDP (both 30 and 20/zg/am) resulted in significant reductions in both ulcer number and severity, compared to controls (19 < .002, Mann-Whitney U test). However, neither the apparent attenuation in gastric stress pathology produced by 10 /xg of CDP nor the apparent facilitation of ulcer development following 2.5/zg of CDP was statistically significant. Similarly, higher doses of GABA (20 and 10/zg/am) significantly inhibited ulcer formation, whereas the lower dose of GABA (2.5/~g/am) produced a slight but nonsignificant increase in ulcer severity. A significant increase in ulcer severity was observed following the largest dose of Ro15-1788 (p < .01 relative to controls). Conversely, a lower dose of Ro15-1788 (5/xg) was found to reduce the number of ulcers
266
SULLIVAN ET AL.
(p < .01). In the interaction groups, Ro15-1788 completely abolished the protective effect of both CDP and GABA, resulting in gastric pathology data almost identical to controls. Histological examination revealed that the majority of the cannula tips were located in the dorsal region of the CEA, or immediately dorsal to the nucleus. When the cannula tip was located outside the central nucleus, or destroyed a major portion of it, the data were not included in the statistical analysis. In general, unilateral injections resulted in gastric pathology similar to that found in controls. Similarly, animals found with accumulated blood near the cannula tips were also eliminated from the data analysis. Virtually in all these latter cases, abnormally severe ulcer development was seen, regardless of the drug treatment. DISCUSSION
The present results implicate the CEA as an important site in the protective effects of CDP and GABA on stress ulceration, although, diffusion to other amygdalar structures cannot be ruled out. But the present data also indicated that animals with cannula tips adjacent to the central nucleus showed no differential effects, relative to the controls. Furthermore, a number of other studies have also suggested that the CEA is the critical amygdalar structure during stress ulcer development (see review by Henke, 1988b). Previous studies on the peripheral administration of benzodiazepines on stress ulcers in rats reported that high doses greatly reduced ulcer formation (File & Pearce, 1981; Henke, 1987; Kunchandy & Kulkarni, 1987), yet lower doses (2.5-5 mg/kg) seemingly aggravated the gastric pathology (Ushijima et al., 1986; File & Pearce, 1981). Interestingly, low doses of both CDP and GABA in the present study (2.5/~g/am) produced slight, but nonsignificant, increases in ulcer severity. The fact that a lower dose of GABA (10/xg) was sufficient to significantly reduce stress ulceration, compared to the same amount of CDP, could reflect the slightly greater molar concentration of a given weight of GABA; possibly, equimolar concentrations of the two drugs produce similar degrees of protection against stress ulcers. The intrinsic effects of Ro15-1788 on gastric stress pathology and the blockade of the GABA effect by this compound are somewhat surprising. Ro15-1788 binds with high affinity to benzodiazepine receptors and antagonizes the actions of both benzodiazepines (agonists) and inverse agonists, including the anxiety-producing /3-carbolines (Bonetti et al., 1982; Ninan et al., 1982). But it has been thought to be largely devoid of intrinsic activity in a wide range of behaviors (Bonetti et al., 1982; Ninan et al., 1982; Quintero et al., 1985). In the social interaction test, however, which may be more sensitive than conflict paradigms in detecting anxiety-related drug effects (File, 1980), Ro15-1788 exhibited an anxiogenic
AMYGDALA AND STRESS
267
effect, although less pronounced than/3-carbolines (File, Lister, & Nutt, 1982). The present data on stress ulcerogenesis suggest that Ro15-1788 acts in an opposite fashion to that of CDP. In fact, the pattern of effects seen following Ro15-1788, i.e., ulcer attenuation with a low dose and potentiation with a higher dose, represents virtually the reverse pattern seen with CDP and GABA. The stress ulcer protection afforded by benzodiazepines has previously been suggested to involve central GABAergic mechanisms, as the GABA antagonist bicuculline reversed this benzodiazepine effect (Kunchandy & Kulkarni, 1987). In the present study, attempts to apply small amounts of bicuculline to the CEA produced spontaneous convulsions and were discontinued. GABA antagonists also block benzodiazepine-induced increases in punished responding (Quintero et al., 1985). The reversal of GABAergic effects by benzodiazepine antagonists, however, is less commonly reported. One study did report that Ro151788 reversed the changes in the cortical EEG patterns of rabbits induced by both diazepam and GABA, but not those induced by pentobarbitone (Singh, Gulati, Srimal, & Dhawan, 1986). Ro15-1788 has also been shown to modulate the effects of picrotoxin and pentylenetetrazole on social and exploratory behavior in rats, in a manner seemingly unrelated to the presence of intrinsic effects of Ro15-1788 (File & Pellow, 1985). The latter two compounds exert their effects on the proposed supramolecular complex at the chloride ion channel. The present results, demonstrating a blockade of the GABA effect by Ro15-1788, strengthen the concept of functional interactions within this receptor complex. The sites of the anticonflict actions of benzodiazepines (and the GABA agonist muscimol) have been reported to be the basolateral and posterolateral amygdala (Scheel-Kriiger & Petersen, 1982), not the CEA as initially reported (Shibata et al., 1982). The effective doses reported for anticonflict activity were also substantially lower than those required for stress ulcer attenuation in the present study, consistent with reports that peripheral administration of benzodiazepines produces anxiolytic effects in doses considerably lower than those needed to reduce stress ulcer formation (File, 1980; File & Pearce, 1981), suggesting separate mechanisms for these two effects. One study, however, which does not fit well within the present framework, reported that the benzodiazepine receptor inverse agonist, FG 7142 (given intraperitoneally), inhibited stress-induced ulcer formation in a dose-dependent manner (Trullas, Ginter, & Skolnick, 1987). This same compound unexpectedly failed to reverse the increase in punished responding produced by CDP in the Geller-Seifter conflict test (Quintero et al., 1985). A review of behavioral and biochemical observations, however, has led to the suggestion that FG 7142 and Ro15-1788 may not act at
268
SULLIVAN ET AL.
identical sites on the benzodiazepine receptor complex in producing anxietyrelated effects, whereas Ro15-1788 and CDP apparently do act at the same site (Pellow & File, 1984). A similar explanation may also account for the variable effects of these compounds on stress ulceration. Furthermore, site-specific effects of benzodiazepine-related compounds probably must also be taken into account, given the widespread distribution of benzodiazepine receptors in the CNS (Young & Kuhar, 1980), and the variety of CNS structures which modulate stress-induced ulcer formation (for review see Henke, 1982; Ray, Henke, & Sullivan, 1988). In summary, CDP and GABA produced a comparable attenuation of stress ulcer formation when applied to the CEA, which was reversed by Ro15-1788. These data suggest the involvement of the GABA/benzodiazepine receptor complex in this area in the modulation of stress-induced ulceration. REFERENCES Ben-Aft, Y. (1981). Transmitters and modulators in the amygdaloid complex: A review. In Y. Ben-Aft (Ed.), The amygdaloid complex. INSERM Symposium No. 20 (pp. 163174). Amsterdam: Elsevier/North-Holland Biomedical Press. Bhargava, K. P., Gupta, G. P., & Gupta, M. B. (1985). Central GABA-ergic mechanism in stress-induced gastric ulceration. British Journal of Pharmacology, 84, 619-623. Bonetti, E. P., Pied, L., Cumin, R., Schaffner, R., Pied, M., Gamzu, E. R., Miiller, R. K. M., & Haefely, W. (1982). Benzodiazepine antagonist Ro15-1788: Neurological and behavioral effects. Psychopharmacology, 78, 8-18. Cananzi, A. R., Costa, E., & Guidotti, A. (1980). Potentiation by intraventricularmuscimol of the anticonflict effect of benzodiazepines. Brain Research, 196, 447-453. Costa, E., Corda, M. G., Epstein, B., Forchetti, C., & Guidotti, A. (1983). GABAbenzodiazepine interactions. In E. Costa (Ed.), The benzodiazepines: From molecular biology to clinical practice. New York: Raven Press. File, S. E. 0980). The use of social interaction as a method for detecting anxiolytic activity of chlordiazepoxide-like drugs. Journal of Neuroscience Methods, 2, 219-238. File, S. E., Lister, R. G., & Nutt, D. J. (1982). The anxiogenic action of benzodiazepine antagonists. Neuropharmacology, 21, 1033-1037. File, S. E., & Pearce, J. B. (1981). Benzodiazepines reduce gastric ulcers in rats induced by stress. British Journal of Pharmacology, 74, 593-599. File, S. E., & Pellow, S. (1985). Does the benzodiazepine antagonist Ro15-1788 reverse the actions of picrotoxin and pentylenetetrazole on social and exploratory behavior? Archives International Pharmacodynamics, 277, 272-279. Gloor, P., Olivier, A., & Quesney, L. F. (1981). The role of the amygdala in the expression of psychic phenomena in temporal lobe seizures. In Y. Ben-Aft (Ed.), The amygdaloid complex. INSERM Symposium No. 20 (pp. 489-498). Amsterdam: Elsevier/NorthHolland Biomedical Press. Henke, P. G. (1980). The amygdala and restraint ulcers in rats. Journal of Comparative and Physiological Psychology, 94, 313-323. Henke, P. G. (1982). The telencephalic limbic system and experimental gastric pathology: A review. Neuroscience and Biobehavioral Reviews, 6, 381-390. Henke, P. G. (1985). The amygdala and forced immobilization of rats. Behavioural Brain Research, 16, 19-24.
AMYGDALA AND STRESS
269
Henke, P. G. (1987). Chlordiazepoxide and stress tolerance in rats. Pharmacology, Biochemistry and Behavior, 26, 561-563. Henke, P. G. (1988a). Electrophysiological activity in the central nucleus of the amygdala, emotionality and stress ulcers in rats. Behavioral Neuroscience, 102, 77-83. Henke, P. G. (1988b). Recent studies of the central nucleus of the amygdala and stress ulcers. Neuroscience and Biobehavioral Reviews, 12, 143-150. Kunchandy, J., & Kulkarni, S. K. (1987). Involvement of central type benzodiazepine and GABAA receptor in the protective effect of benzodiazepines in stress-induced gastric ulcers in rats. Archives International Pharmacodynamics, 285, 129-136. Lahti, R. A., & Barsuhn, C. (1974). The effect of minor tranquilizers on stress-induced increases in rat plasma corticosteroids. Psychopharmacology (Berlin), 35, 215-220. Nakamura, K., Hayashi, T., & Nakamura, K. (1984). Effects of bromazepam on cerebral neuronal activity in male Wistar rats with immobilized stress. Folia Pharmacologia Japonica, 83, 401-412. Ninan, P. T., Insel, T. M., Cohen, R. M., Cook, J. M., Skolnick, P., & Paul, S. M. (1982). Benzodiazepine receptor-mediated experimental "anxiety" in primates. Science, 218, 1322-1334. PeUow, S., & File, S. E. (1984). Multiple sites of action for anxiogenic drugs: Behavioral, electrophysiological and biochemical correlations. Psychopharmacology, 83, 304-315. Quintero, S., Henney, S., Lawson, P., Mellanby, J., & Gray, J. A. (1985). The effects of compounds related to 3,-aminobutyrate and benzodiazepine receptors on behavioral responses to anxiogenic stimuli in the rat: Punished barpressing. Psychopharmacology, 85, 244-251. Ray, A., Henke, P. G., & Sullivan, R. M. (1988). Central dopamine systems and gastric stress pathology in rats. Physiology and Behavior, 42, 359-364. Scheel-Kriiger, J., & Petersen, E. N. (1982). Anticonflict effect of the benzodiazepines mediated by a GABAergic mechanism in the amygdala. European Journal of Pharmacology, 82, 115-116. Shephard, R. A. (1986). Neurotransmitters, anxiety and benzodiazepines: A behavioral review. Neuroscience and Biobehavioral Reviews, 10, 449-46t. Shibata, K., Kataoka, Y., Gomita, Y., & Ueki, S. (1982). Localization of the site of the anticonflict action of benzodiazepines in the amygdaloid nucleus of rats. Brain Research, 234, 442-446. Singh, H. K., Gulati, A., Srimal, R. C., & Dhawan, B. N. (1986). Effect of Ro15-1788 on diazepam, GABA and pentobarbitone induced EEG changes in rabbits. Indian Journal of Medical Research, 83(June), 633-641. Trullas, R., Ginter, H., & Skolnick, P. (1987). A benzodiazepine receptor inverse agonist inhibits stress-induced ulcer formation. Pharmacology, Biochemistry and Behavior, 27, 35-39. Ushijima, I., Mizuki, Y., Hara, T., Kudo, R., Watanabe, K., & Yamada, M. (1986). The role of adenosinergic, GABAergic and benzodiazepine systems in hyperemotionality and ulcer formation in stressed rats. Psychopharmacology, 89, 472-476. Young, W. S., & Kuhar, M. J. (1980). Radiohistochemical localization of benzodiazepine receptors in rat brain. Journal of Pharmacology and Experimental Therapeutics, 212, 337-346.
51, 262-269 (1989)
The GABA/Benzodiazepine Receptor Complex in the Central Amygdalar Nucleus and Stress Ulcers in Rats R. M. SULLIVAN, P. G. HENKE, A. RAY, M. A. HEBERT, AND J. M. TRIMPER1 St. Francis Xavier University, Antigonish, Nova Scotia, Canada The effects of bilateral microinjections of chlordiazepoxide and GABA into the central amygdalar nucleus on gastric ulcer formation induced by cold-restraint were examined in chronically implanted Wistar rats. Higher doses of chlordiazepoxide (20 and 30/zg/amygdala) significantly reduced stress ulcer development, whereas a lower dose (2.5/xg) produced a nonsignificant increase in ulcer severity. A similar dose/response pattern was observed following GABA administration. The benzodiazepine receptor antagonist Ro15-1788, applied to the amygdala, abolished the protective effects of both chlordiazepoxide and GABA. In addition, when Ro15-1788 (10/zg) was injected into the amygdala by itself, it aggravated the gastric stress pathology. However, a lower dose (5/xg) had an attenuating effect, opposite to the pattern of effects produced by chlordiazepoxide and GABA. The role of the amygdalar GABA-benzodiazepine receptor complex in stressful conditions is discussed. © 1989AcademicPress, Inc.
Benzodiazepines are among the most widely prescribed drugs for the treatment of anxiety states (Shephard, 1986), and they demonstrate anticonflict activity in a variety of animal models of anxiety (Pellow & File, 1984; Quintero, Henney, Lawson, Mellanby, & Gray, 1985). It has been suggested that benzodiazepine receptors, widespread in the CNS (Young & Kuhar, 1980), may not only mediate the anxiolytic actions of benzodiazepines, but may be crucial in the pathogenesis of anxiety and its associated pathophysiological states in humans (Ninan et al., 1982). Benzodiazepines are reported to decrease stress-induced elevations of plasma corticosteroids (Lahti & Barsuhn, 1974), and they are known to reduce gastric ulcers induced by stress in rats (File & Pearce, 1981; Henke, 1987; Kunchandy & Kulkarni, 1987). Supported by the Natural Sciences and Engineering Research Council of Canada. The authors thank Hoffman-LaRoche Co., Ltd., for the gift of chlordiazepoxide and Ro151788, and also thank Ms. Jane Aker for the secretarial assistance in the preparation of the manuscript. Correspondence and reprint requests should be addressed to Peter G. Henke, Department of Psychology, St. Francis Xavier University, Antigonish, Nova Scotia, Canada, B2G 1C0. 262 0163-1047/89 $3.00 Copyright © 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
AMYGDALA AND STRESS
263
Benzodiazepine receptors are thought to interact with GABA (y-aminobutyric acid) receptors, as part of a supramolecular complex in neuronal cell membranes. Agonists at the benzodiazepine recognition site are believed to enhance GABAergic activity by facilitating the coupling of the GABA recognition site to the associated chloride ion channel, whereas inverse agonists (/~-carboline derivatives) are assumed to hinder this coupling (Costa, Corda, Epstein, Forcfietti, & Guidotti, 1983). GABA agonists, not surprisingly, share a number of behavioral effects with benzodiazepines. In addition to their anticonvulsant and sedative properties, centrally administered GABA agonists have been shown to produce anticonflict activity (Cananzi, Costa, & Guidotti, 1980; ScheelKr~iger & Petersen, 1982). GABA has also been reported to reduce stress ulcers in rats (Bhargava, Gupta, & Gupta, 1985; Ushijima et al., 1986). Furthermore, central GABAergic mechanisms have been suggested to be involved in both the ulcer attenuation and anticonflict activity produced by benzodiazepines (Kunchandy & Kulkarni, 1987; Scheel-Krfiger & Petersen, 1982). Several lines of evidence have implicated the central nucleus of the amygdala (CEA) as an important site in the response of the organism to stressful situations. For example, lesions in this area were found to attenuate stress-induced gastric pathology in rats (Henke, 1980), whereas low-level electrical stimulation of the amygdala produced gastric erosions only when localized in the CEA (Henke, 1985). Changes in multiple-unit activity of the amygdala have been observed in response to restraint stress and the behavior of some of these units was modified by ip injections of chlordiazepoxide (Henke, 1985). Additional recording studies in the CEA revealed that two multiple-unit activity profiles, recorded during restraint stress, were associated with stress ulceration, one with increased and one with decreased stomach pathology. The same activity profiles were differentially distributed in genetically selected Roman high- and low-avoidance rats, as well as Wistar rats differentiated on the basis of emotionality characteristics in an open-field test (Henke, 1988a). Interestingly, stimulation of the amygdala in humans has been reported to result in fear experiences (Gloor, Olivier, & Quesney, 1981). A recent study demonstrated that the increase in glucose utilization rate, seen in a number of brain areas following restraint and waterimmersion stress in rats, was predominant in the CEA, and this elevation was significantly reduced by bromazepam, at a non-muscle relaxant dose (Nakamura, Hayashi, & Nakamura, 1984). Furthermore, microinjections of benzodiazepines into the CEA of rats significantly increased punished responding in a conflict paradigm, suggestive of anxiolytic activity (Shibata, Kataoka, Gomita, & Ueki, 1982), but other amygdalar nuclei are also reportedly involved in this effect (Scheel-Krfiger & Petersen, 1982). The CEA has been shown to contain the highest concentration of
264
SULLIVAN ET AL.
GABA in the amygdalar complex (Ben-Ari, 1981), as well as a high density of benzodiazepine receptors (Young & Kuhar, 1980). We, therefore, investigated the effects of intraamygdalar applications of chlordiazepoxide, Ro15-1788, and GABA on stress ulcer formation in rats. METHOD Male Wistar rats, 90-120 days old (Charles River, Quebec) were individually housed, with free access to food and water. They were maintained at 22 ___ 2°C, on a 12-h light/dark cycle (lights on at 8 AM). For surgery, rats were anesthetized with sodium pentobarbital (50 mg/kg, ip) and secured in a stereotaxic apparatus. Twenty-three-gauge stainless-steel guide cannulae (Plastic Products) were lowered bilaterally to the dorsal edge of the central nucleus of the amygdala, using the following coordinates: 2.0 P (to bregma), 4.1 lateral (to midline), and 7.5 V (to dura), skull horizontal. Stainless-steel screws and dental acrylic secured the position of the cannulae. Following a 1-week recovery period, the rats were deprived of food (but not water) for 24 h prior to testing. The animals then received intraamygdalar injections of drug or vehicle in a volume of 2 ~1, slowly over a period of 1 min. The injection cannula, which did not protrude beyond the tip of the guide cannula, was attached to a 5/xl Hamilton syringe with microbore tubing and was left in position an additional 2 min to allow for diffusion of the injection solution. Following injections, the rats were immediately placed in Plexiglas restrainers (Fisher Scientific). After 3 h of cold-immobilization (4°C), the animals were sacrificed with an overdose of sodium pentobarbital (ip). The stomachs were dissected out and cut along the greater curvature, washed with cold water and examined microscopically for gastric mucosal lesions. The number of lesions, as well as their severity (cumulative length to the nearest 0.1 mm) were recorded. After removal of the stomachs, the rats were perfused intracardially with 0.9% saline and 10% formalin. The brains were removed, embedded in paraffin, sectioned at 15 /~m, and stained with thionin to determine the location of the cannula tips. The drugs used were chlordiazepoxide HCI, Ro15-1788 (HoffmanLaRoche), and GABA (from Sigma). Chlordiazepoxide (CDP) and GABA were dissolved in distilled water, whereas Ro15-1788 was suspended in distilled water with a drop of Tween 20. All injections were intraamygdalar and bilateral. Control animals received injections of either distilled water or distilled water with Tween 20. In the interaction groups, Ro15-1788 was injected 20 min prior to either CDP or GABA. All groups reported in the study consisted of six rats, except for the CDP (1.0/zg/am) group (n = 5). The results were analyzed using the Mann-Whitney U test (two-tailed) for multiple comparisons between groups. A p value of at least .05 was used as the level of significance for the statistical tests.
265
A M Y G D A L A A N D STRESS
TABLE 1 Effects of Intraamygdalar CDP, GABA, and Ro15-1788 on Stress-Induced Gastric Pathology in Rats (Means ± SEM)
Treatment (p-g/am)
n
Mean ulcer frequency
Controls
6
6.0 - 0.45
2.0 --- 0.31
CDP CDP CDP CDP CDP
6 6 6 6 5
1.2 1.3 4.5 7.2 6.0
+-+ -
0.31 a 0.21" 0.67 1.08 0.89
0.2 0.3 1.4 4.2 1.7
+± ± -+ _
0.06 a 0.04 a 0.31 1.21 0.22
6 6 6 6
1.8 2.3 4.3 5.5
-.+ 0.48 a -+ 0.33" -+ 0.49 -+ 0.72
0.4 0.7 3.0 1.7
+--. --±
0.10 ° 0.06 b 0.99 0.30
Ro15-1788 (10) Ro15-1788 (5) Ro15-1788 (2.5)
6 6 6
8.7 ± 1.48 3.8 - 0.48 b 5.8 - 0.70
5.7 -+ 1.80 ~ 1.1 - 0.26 2.0 -+- 0.56
Ro(10) + CDP (20) Ro(10) + G A B A (10)
6 6
6.8 -+ 0.60 5.2 _+ 0.79
2.1 -+ 0.20 2.3 --- 0.64
(30) (20) (10) (2.5) (1)
GABA GABA GABA GABA
(20) (10) (2.5) (1)
Mean severity (ram)
Note. All comparisons were by the M a n n - W h i t n e y U test (two-tailed). a p < .002. b p < .01 (compared to controls).
RESULTS
The control group consisted of three rats receiving distilled water injections, and three additional animals which were also injected with Tween 20. Both types of solutions produced similar pathology measures (mean ulcer frequency was six in each case, and mean severity was 1.8 and 2.2 ram, respectively). Consequently, the data were combined for statistical tests. Table 1 shows that intraamygdalar CDP (both 30 and 20/zg/am) resulted in significant reductions in both ulcer number and severity, compared to controls (19 < .002, Mann-Whitney U test). However, neither the apparent attenuation in gastric stress pathology produced by 10 /xg of CDP nor the apparent facilitation of ulcer development following 2.5/zg of CDP was statistically significant. Similarly, higher doses of GABA (20 and 10/zg/am) significantly inhibited ulcer formation, whereas the lower dose of GABA (2.5/~g/am) produced a slight but nonsignificant increase in ulcer severity. A significant increase in ulcer severity was observed following the largest dose of Ro15-1788 (p < .01 relative to controls). Conversely, a lower dose of Ro15-1788 (5/xg) was found to reduce the number of ulcers
266
SULLIVAN ET AL.
(p < .01). In the interaction groups, Ro15-1788 completely abolished the protective effect of both CDP and GABA, resulting in gastric pathology data almost identical to controls. Histological examination revealed that the majority of the cannula tips were located in the dorsal region of the CEA, or immediately dorsal to the nucleus. When the cannula tip was located outside the central nucleus, or destroyed a major portion of it, the data were not included in the statistical analysis. In general, unilateral injections resulted in gastric pathology similar to that found in controls. Similarly, animals found with accumulated blood near the cannula tips were also eliminated from the data analysis. Virtually in all these latter cases, abnormally severe ulcer development was seen, regardless of the drug treatment. DISCUSSION
The present results implicate the CEA as an important site in the protective effects of CDP and GABA on stress ulceration, although, diffusion to other amygdalar structures cannot be ruled out. But the present data also indicated that animals with cannula tips adjacent to the central nucleus showed no differential effects, relative to the controls. Furthermore, a number of other studies have also suggested that the CEA is the critical amygdalar structure during stress ulcer development (see review by Henke, 1988b). Previous studies on the peripheral administration of benzodiazepines on stress ulcers in rats reported that high doses greatly reduced ulcer formation (File & Pearce, 1981; Henke, 1987; Kunchandy & Kulkarni, 1987), yet lower doses (2.5-5 mg/kg) seemingly aggravated the gastric pathology (Ushijima et al., 1986; File & Pearce, 1981). Interestingly, low doses of both CDP and GABA in the present study (2.5/~g/am) produced slight, but nonsignificant, increases in ulcer severity. The fact that a lower dose of GABA (10/xg) was sufficient to significantly reduce stress ulceration, compared to the same amount of CDP, could reflect the slightly greater molar concentration of a given weight of GABA; possibly, equimolar concentrations of the two drugs produce similar degrees of protection against stress ulcers. The intrinsic effects of Ro15-1788 on gastric stress pathology and the blockade of the GABA effect by this compound are somewhat surprising. Ro15-1788 binds with high affinity to benzodiazepine receptors and antagonizes the actions of both benzodiazepines (agonists) and inverse agonists, including the anxiety-producing /3-carbolines (Bonetti et al., 1982; Ninan et al., 1982). But it has been thought to be largely devoid of intrinsic activity in a wide range of behaviors (Bonetti et al., 1982; Ninan et al., 1982; Quintero et al., 1985). In the social interaction test, however, which may be more sensitive than conflict paradigms in detecting anxiety-related drug effects (File, 1980), Ro15-1788 exhibited an anxiogenic
AMYGDALA AND STRESS
267
effect, although less pronounced than/3-carbolines (File, Lister, & Nutt, 1982). The present data on stress ulcerogenesis suggest that Ro15-1788 acts in an opposite fashion to that of CDP. In fact, the pattern of effects seen following Ro15-1788, i.e., ulcer attenuation with a low dose and potentiation with a higher dose, represents virtually the reverse pattern seen with CDP and GABA. The stress ulcer protection afforded by benzodiazepines has previously been suggested to involve central GABAergic mechanisms, as the GABA antagonist bicuculline reversed this benzodiazepine effect (Kunchandy & Kulkarni, 1987). In the present study, attempts to apply small amounts of bicuculline to the CEA produced spontaneous convulsions and were discontinued. GABA antagonists also block benzodiazepine-induced increases in punished responding (Quintero et al., 1985). The reversal of GABAergic effects by benzodiazepine antagonists, however, is less commonly reported. One study did report that Ro151788 reversed the changes in the cortical EEG patterns of rabbits induced by both diazepam and GABA, but not those induced by pentobarbitone (Singh, Gulati, Srimal, & Dhawan, 1986). Ro15-1788 has also been shown to modulate the effects of picrotoxin and pentylenetetrazole on social and exploratory behavior in rats, in a manner seemingly unrelated to the presence of intrinsic effects of Ro15-1788 (File & Pellow, 1985). The latter two compounds exert their effects on the proposed supramolecular complex at the chloride ion channel. The present results, demonstrating a blockade of the GABA effect by Ro15-1788, strengthen the concept of functional interactions within this receptor complex. The sites of the anticonflict actions of benzodiazepines (and the GABA agonist muscimol) have been reported to be the basolateral and posterolateral amygdala (Scheel-Kriiger & Petersen, 1982), not the CEA as initially reported (Shibata et al., 1982). The effective doses reported for anticonflict activity were also substantially lower than those required for stress ulcer attenuation in the present study, consistent with reports that peripheral administration of benzodiazepines produces anxiolytic effects in doses considerably lower than those needed to reduce stress ulcer formation (File, 1980; File & Pearce, 1981), suggesting separate mechanisms for these two effects. One study, however, which does not fit well within the present framework, reported that the benzodiazepine receptor inverse agonist, FG 7142 (given intraperitoneally), inhibited stress-induced ulcer formation in a dose-dependent manner (Trullas, Ginter, & Skolnick, 1987). This same compound unexpectedly failed to reverse the increase in punished responding produced by CDP in the Geller-Seifter conflict test (Quintero et al., 1985). A review of behavioral and biochemical observations, however, has led to the suggestion that FG 7142 and Ro15-1788 may not act at
268
SULLIVAN ET AL.
identical sites on the benzodiazepine receptor complex in producing anxietyrelated effects, whereas Ro15-1788 and CDP apparently do act at the same site (Pellow & File, 1984). A similar explanation may also account for the variable effects of these compounds on stress ulceration. Furthermore, site-specific effects of benzodiazepine-related compounds probably must also be taken into account, given the widespread distribution of benzodiazepine receptors in the CNS (Young & Kuhar, 1980), and the variety of CNS structures which modulate stress-induced ulcer formation (for review see Henke, 1982; Ray, Henke, & Sullivan, 1988). In summary, CDP and GABA produced a comparable attenuation of stress ulcer formation when applied to the CEA, which was reversed by Ro15-1788. These data suggest the involvement of the GABA/benzodiazepine receptor complex in this area in the modulation of stress-induced ulceration. REFERENCES Ben-Aft, Y. (1981). Transmitters and modulators in the amygdaloid complex: A review. In Y. Ben-Aft (Ed.), The amygdaloid complex. INSERM Symposium No. 20 (pp. 163174). Amsterdam: Elsevier/North-Holland Biomedical Press. Bhargava, K. P., Gupta, G. P., & Gupta, M. B. (1985). Central GABA-ergic mechanism in stress-induced gastric ulceration. British Journal of Pharmacology, 84, 619-623. Bonetti, E. P., Pied, L., Cumin, R., Schaffner, R., Pied, M., Gamzu, E. R., Miiller, R. K. M., & Haefely, W. (1982). Benzodiazepine antagonist Ro15-1788: Neurological and behavioral effects. Psychopharmacology, 78, 8-18. Cananzi, A. R., Costa, E., & Guidotti, A. (1980). Potentiation by intraventricularmuscimol of the anticonflict effect of benzodiazepines. Brain Research, 196, 447-453. Costa, E., Corda, M. G., Epstein, B., Forchetti, C., & Guidotti, A. (1983). GABAbenzodiazepine interactions. In E. Costa (Ed.), The benzodiazepines: From molecular biology to clinical practice. New York: Raven Press. File, S. E. 0980). The use of social interaction as a method for detecting anxiolytic activity of chlordiazepoxide-like drugs. Journal of Neuroscience Methods, 2, 219-238. File, S. E., Lister, R. G., & Nutt, D. J. (1982). The anxiogenic action of benzodiazepine antagonists. Neuropharmacology, 21, 1033-1037. File, S. E., & Pearce, J. B. (1981). Benzodiazepines reduce gastric ulcers in rats induced by stress. British Journal of Pharmacology, 74, 593-599. File, S. E., & Pellow, S. (1985). Does the benzodiazepine antagonist Ro15-1788 reverse the actions of picrotoxin and pentylenetetrazole on social and exploratory behavior? Archives International Pharmacodynamics, 277, 272-279. Gloor, P., Olivier, A., & Quesney, L. F. (1981). The role of the amygdala in the expression of psychic phenomena in temporal lobe seizures. In Y. Ben-Aft (Ed.), The amygdaloid complex. INSERM Symposium No. 20 (pp. 489-498). Amsterdam: Elsevier/NorthHolland Biomedical Press. Henke, P. G. (1980). The amygdala and restraint ulcers in rats. Journal of Comparative and Physiological Psychology, 94, 313-323. Henke, P. G. (1982). The telencephalic limbic system and experimental gastric pathology: A review. Neuroscience and Biobehavioral Reviews, 6, 381-390. Henke, P. G. (1985). The amygdala and forced immobilization of rats. Behavioural Brain Research, 16, 19-24.
AMYGDALA AND STRESS
269
Henke, P. G. (1987). Chlordiazepoxide and stress tolerance in rats. Pharmacology, Biochemistry and Behavior, 26, 561-563. Henke, P. G. (1988a). Electrophysiological activity in the central nucleus of the amygdala, emotionality and stress ulcers in rats. Behavioral Neuroscience, 102, 77-83. Henke, P. G. (1988b). Recent studies of the central nucleus of the amygdala and stress ulcers. Neuroscience and Biobehavioral Reviews, 12, 143-150. Kunchandy, J., & Kulkarni, S. K. (1987). Involvement of central type benzodiazepine and GABAA receptor in the protective effect of benzodiazepines in stress-induced gastric ulcers in rats. Archives International Pharmacodynamics, 285, 129-136. Lahti, R. A., & Barsuhn, C. (1974). The effect of minor tranquilizers on stress-induced increases in rat plasma corticosteroids. Psychopharmacology (Berlin), 35, 215-220. Nakamura, K., Hayashi, T., & Nakamura, K. (1984). Effects of bromazepam on cerebral neuronal activity in male Wistar rats with immobilized stress. Folia Pharmacologia Japonica, 83, 401-412. Ninan, P. T., Insel, T. M., Cohen, R. M., Cook, J. M., Skolnick, P., & Paul, S. M. (1982). Benzodiazepine receptor-mediated experimental "anxiety" in primates. Science, 218, 1322-1334. PeUow, S., & File, S. E. (1984). Multiple sites of action for anxiogenic drugs: Behavioral, electrophysiological and biochemical correlations. Psychopharmacology, 83, 304-315. Quintero, S., Henney, S., Lawson, P., Mellanby, J., & Gray, J. A. (1985). The effects of compounds related to 3,-aminobutyrate and benzodiazepine receptors on behavioral responses to anxiogenic stimuli in the rat: Punished barpressing. Psychopharmacology, 85, 244-251. Ray, A., Henke, P. G., & Sullivan, R. M. (1988). Central dopamine systems and gastric stress pathology in rats. Physiology and Behavior, 42, 359-364. Scheel-Kriiger, J., & Petersen, E. N. (1982). Anticonflict effect of the benzodiazepines mediated by a GABAergic mechanism in the amygdala. European Journal of Pharmacology, 82, 115-116. Shephard, R. A. (1986). Neurotransmitters, anxiety and benzodiazepines: A behavioral review. Neuroscience and Biobehavioral Reviews, 10, 449-46t. Shibata, K., Kataoka, Y., Gomita, Y., & Ueki, S. (1982). Localization of the site of the anticonflict action of benzodiazepines in the amygdaloid nucleus of rats. Brain Research, 234, 442-446. Singh, H. K., Gulati, A., Srimal, R. C., & Dhawan, B. N. (1986). Effect of Ro15-1788 on diazepam, GABA and pentobarbitone induced EEG changes in rabbits. Indian Journal of Medical Research, 83(June), 633-641. Trullas, R., Ginter, H., & Skolnick, P. (1987). A benzodiazepine receptor inverse agonist inhibits stress-induced ulcer formation. Pharmacology, Biochemistry and Behavior, 27, 35-39. Ushijima, I., Mizuki, Y., Hara, T., Kudo, R., Watanabe, K., & Yamada, M. (1986). The role of adenosinergic, GABAergic and benzodiazepine systems in hyperemotionality and ulcer formation in stressed rats. Psychopharmacology, 89, 472-476. Young, W. S., & Kuhar, M. J. (1980). Radiohistochemical localization of benzodiazepine receptors in rat brain. Journal of Pharmacology and Experimental Therapeutics, 212, 337-346.