- Email: [email protected]
Effects of stress and β-funal trexamine pretreatment on morphine analgesia and opioid binding in rats
Life Sciences, Vol. 41, pp. 2835-2844 Printed in the U.S.A.
Pergamon Journal~
EFFECTS OF STRESS ANO B-FUNALTREXAMINE PRETREATMENT ON MORPHINE ANALGESIA AND OPIOID BINDING IN RATS J.U. Adams ~, J.S. Andrewsl, 2, J.M. Hiller s , E.J. Simon3, ~ and S.G. Holtzman l ~Department of Pharmacology, Emory University School of Medicine Atlanta, Georgia 30322 and 3Departments of Psychiatry and ~Pharmacology, New York University Medical Center, 550 First Avenue, New York, New York lOO16 (Received in final form November 3, 1987) Summary This study was essentially an in vivo protection experiment designed to test further the hypothesis that stress induces release of endogenous opioids which then act at opioid receptors. Rats that were either subjected to restraint stress for 1 hr or unstressed were injected ICY with either saline or 2.5 pg of Bfunaltrexamine (B-FNA), an irreversible opioid antagonist that alkylates the mu-opioid receptor. Twenty-four hours later, subjects were tested unstressed for morphine analgesia (tail-flick assay) oK w@re sacrificed and opioid binding in brain was determined. [3HJD-Ala2NMePhe~-GlyS(ol)enkephalin (DAGQ) served as a specific ligand for mu-- opioid receptors, and L3HJ-bremazocine as a general ligand for all opioid receptors. Rats injected with saline while stressed were significantly less sensitive to the analgesic action of morphine 24 hr later than were their unstressed counterparts. B-FNA pretreatment attenuated morphine analgesia in an insurmountable manner. Animals pretreated with B-FNA while stressed were significantly more sensitive to the analgesic effect of morphine than were animals that received 8-FNA while unstressed, consistent with the hypothesis that stress induces release of endogenous opioids that would protect opioid receptors from alk~la~ion by B-FNA. 6-FNA caused small and similar decreases in [3HJ-DAGO binding in brain of both stressed and unstressed animals. Stressed rats injected with saline tended to have increased levels of [3H]DAGO and [~H]bremazocine binding compared to the other groups. This outcome may be relevant to the tolerance to morphine analgesia caused by stress. Endogenous opioids have been implicated in many of the biochemical, physiological, and behavioral responses to stress. For example, opioid involvement in the phenomenon of stress-induced analgesia has been widely investigated (for reviews see 1-3). Not only can stressful events induce analgesia, they can also potentiate the analgesic affect of exogenously administered opioid drugs (4-6). For example, both the magnitude and dura2present address: Berlin 65.
Department of Neuropsychopharmacology, Schering AG, D-IO00
0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
2836
Stress, 6-FNA and Morphine
Vol. 41, No. 26, 1987
tion of the analgesic effect of morphine and methadone administered subcutaneously were enhanced significantly in rats that were either restrained or placed in a cold environment at the time of testing as compared to unrestrained and presumably minimally stressed animals (6). Similar results were obtained when morphine or enkephalin analogs were administered intracerebroventricularly (ICY) (7,8), indicating that the interaction between stress and exogenous opioids is mediated centrally. The mechanisms that underlie stress-induced potentiation of opioid analgesia are unknown, but also may involve endogenous opioids. Stress has been associated with increased brain levels of opioid peptides (9) and with a decreased number of opioid binding sites in brain. The latter finding is due presumably to the presence of competing endogenous ligands (lO,11) or to receptor downregulation from prolonged exposure of receptors to elevated levels of endogenous ligands (12). Beta-Funaltrexamine (8-FNA), the fumarate methyl ester derivative of the opioid antagonist naltrexone, is a non-equilibrium opioid receptor antagonist (13). It is believed to alkylate relatively selectively the mu-opioid receptor (14). Receptors can be protected from alkylation in vitro if they are exposed to reversible receptor ligands, agonists or antagonists, before being exposed to the inactivating agent (e.g., 15-17). For example, incubation of the guinea pig ileum with 6-FNA produced a rightward shift in the morphine concentration-effect curve even after the tissue was washed numerous times. However, if naltrexone was added to the tissue bath before 8-FNA, after the tissue was washed there was little antagonism of the effects of morphine, presumably because naltrexone prevented 8-FNA from occupying and alkylating the receptors (18)° The present study was essentially an in vivo protection experiment designed to provide further evidence for the stress-induced release of endogenous opioids and their subsequent interaction at opioid receptors. We postulated that stress increases the occupation of opioid receptors by their endogenous ligands. Therefore, 8-FNA would alkylate fewer receptors in rats subjected to stress (i.e. restraint) than in rats that were not stressed at the time of 8-FNA administration. The experimental design was similar to the one used to demonstrate stress-induced potentiation of opioid analgesia (5-8), with one important difference. Stress-induced potentiation of opioid analgesia occurs in rats that are tested for analgesia while restrained. In the present study, 8-FNA or saline was administered ICV to separate groups of rats that were either restrained or unrestrained. Twenty-four hours later, the analgesic effect of morphine was determined in all animals while they were unrestrained. The magnitude of the analgesic response to morphine should be an inverse function of the number of opioid receptors inactivated by 8-FNA. Other groups of animals were also treated with either 8-FNA or saline while stressed or unstressed. Twenty-four hours later, instead of being tested with morphine, these rats were sacrificed and the number of opioid binding sites was determined in several regions of the brain. Methods
Animals. Subjects were Sprague-Dawley derived male rats (Sasco Inc., Omaha, NE)--weighing between 250 and 275 g at the beginning of the study. Rats were housed in pairs under a 12-hour light/dark cycle, with food and water available ad libitum. Surgery. Rats were anesthetized with 50 mg/kg sodium pentobarbital and, if additional anesthetic was needed, lO0 mg/kg chloral hydrate. A 22 g stainless steel guide cannula (Plastic Products, Inc., Roanoke, VA) was cut
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2837
to a length of 5.5 mm and beveled and was surgically implanted into the lateral ventricle. Using a stereotaxic apparatus with the upper incisor bar set at 3.3 mm below horizontal zero, the cannula was placed at 0.8 mm posterior and 1.4 mm lateral relative to bregma, and 3.5 mm ventral to the skull, and was permanantly secured with skull screws (no. 0.-80 x 3/32 in) and dental acrylic (Lang Dental Mfg. Co., Chicago, IL). A 28 g dummy cannula (Plastic Products), cut to protude 0.5 mm beyond the tip of the guide cannula, was kept in the implanted cannula except during ICV injections. Procedure. Rats were divided into four groups according to pretreatmerit: l) saline/no stress, 2) saline/stress, 3) B-FNA/no stress, 4) 8-FNA/stress. Animals were habituated to handling and test procedures and to restraint stress (i.e., groups 2 and 4) for 5 days before drug testing began. Tests were conducted on two consecutive days, once each week. On the pretreatment day, animals in groups 2 and 4 were subjected to restraint stress by immobilizing them in Plexiglas cylinders (5 and 6 cm in diameter) for I hr. All animals received an ICV injection, either 2.5 pl of sterile bacteriostatic isotonic saline (groups i and 2) or 2.5 pg B-FNA (groups 3 and 4) in a volume of 2.5 pl. Injections were performed on freely moving rats with a 25 pl Hamilton syringe (Hamilton Co., Reno, NV) connected to a 28 g injection cannula (Plastic Products) by PE 20 tubing. Injections were given over a 1-min period and the injection cannula was kept in place for an additional 30-60 sec. Stressed animals were removed from restraint tubes 30 min after the start of restraint stress to receive their injections and then were replaced for the remaining 30 min. Analgesia. Twenty-four hours later, animals were tested for morphine analgesia using the tail-flick method of D'Amour and Smith (19), with modifications (20). Radiant heat from a 20 V high amperage bulb was focused on the lower half of the rat's tail with an ellipsoidal reflector (Optical Industries, Costa Mesa, CA). Movement of the tail allowed the light to activate a photocell, which stopped both the stimulus and the reaction timer. Stimulus intensity was controlled by a variable transformer set to give baseline latencies of 2.0-2.5 sec. A cut-off time of 6.0 sec was used to minimize damage to the tail. After a habituation trial (no stimulus) and a baseline trial, rats were injected SC with one of four doses of morphine, and were tested for latency to tail-flick 5, 15, 30, 45, 60, and 120 min later. After the last trial, rats were administered a dose of naltrexone one-tenth that of the morphine test dose in order to ensure termination of the effects of morphine. Each animal was tested at weekly intervals. Half of the animals in each group received morphine doses in ascending order, beginning with 3.0 mg/kg, and the other half received morphine doses in descending order, ending with 3.0 mg/kg. Binding. A different set of rats was divided into four groups, habituated to handling and stress, and then subjected to the same four pretreatments as described above. On the following day, the animals were decapitated and their brain was removed and rapidly cooled on a thermoelectric cold plate. The caudate, thalamus, and central gray areas were dissected out. Brain areas from three animals were pooled, weighed and homogenized in lO volumes of 50 Tris-HC1 buffer, pH 7.4, containing 1 mM dipotassium EDTA, using a Brinkmann Polytron (Brinkmann Instruments, Inc., Westbury, NY) at a setting of 6 for 20 sec. The homogenate was centrifuged at 20,000 x g for 15 min and the pellet was resuspended in six volumes of 0.32 M sucrose using the Polytron set at 4, and stored at -60°C until used. Prior to equilibrium binding experiments, membrane preparations were thawed and diluted lO-fold in the above Tris buffer so as to give a final
2838
Stress,
B-FNA and Morphine
Vol. 41, No. 26, 1987
protein concentration of approximately l.O mg/ml, as determined by the method of Lowry et al. (21). Binding assays on triplicate 1.O ml samples were carried out essentially as previously described (22) at 25°C for i hr in the presence and absence of a lO00-fold excess of unlabeled naloxone. In these preliminary experiments to screen for possible changes in binding levels, a single concentration of each labeled ligand was used. The concentration of the labeled ligands was within a factor of 2 of their respective Kds. To assess mu site ~in@ing, [3HI D-Ala2-MeRhe"-GlyS(ol)-enkephalin (3H-DAGO, i nM) was u---sad. L3HJ-Bremazocine (1 nM), possessing almost equal affinity for mu, delta and kappa binding sites, was used to assess total opioid binding levels. Separation of bound from free ligand was accomplished by filtration through GF/B filters followed by two washes with 4.0 ml of Tris buffer. Radioactivity remaining on dried filters was determined by scintillation spectroscopy in a 8eckman LS-515]-T counter. Data Analysis. Analgesia data are expressed as percent maximum possible effect (~MPE] (23): postdrug latency - baseline latency %MPE .......................................... cut-off time (6.0 sec) - baseline latency Dose-effect curves were generated by calculating the area under the time-ZMRE curve using the trapezoidal rule. Differences between the groups' timeeffect curves and differences between dose-effect curves were analyzed using two-way ANOVA and post hoc Newman-Keuls test. Binding data were analyzed by one-way ANOVA. Drugs. The drugs used in the analgesia experiments were morphine sulfate (Penick Corp., Newark, NJ), naltrexone hydrochloride, and B-FNA hydrochloride (both from National Institute on Drug Abuse, Rockville, MD). Morphine and naltrexone were dissolved in 0.gZ saline; B-FNA was dissolved in sterile bacteriostatic saline for icy administration. Subcutaneous injections were delivered in a volume of 1.0 ml/kg. All doses are expressed ~n terms of the free base. Ligands used in the binding ex_periments were 3 [~HJDAGO (49 Ci/mMol) (Amersham Corp., Des Plaines, IL), [3HJ-bremazocine (32 Ci/mMol) (New England Nuclear Corp., Boston, MA), and naloxone hydrochloride (National Institute on Drug Abuse). TABLE I Pre-Drug Response Latencies in Tail-Flick Test
Group
Pretceatment
Seconds (mean ± SEM; n = i0)
i
SAL/NO STRESS
2.24 + .07
2
SAL/STRESS
2.14 _+ .i0
5
B-FNA/NO STRESS
2.11 + .07
4
B-FNA/STRESS
2.16 _+ .lO
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2839
Results Pretreatment conditions had no effect on baseline latencies in the tailflick test performed 24 hr later (Table i; p > .05). However, pretreatment conditions did modify significantly the responses of the animals to morphine. Morphine produced increases in latencies to tail flick in all four groups (Figure i). There were dose-related increases in both peak effect and duration of effect. In the control rats (group l: saline/no stress), 5.0 mg/kg morphine produced a peak effect of 95 ± 5 ~4PE at 45 min post-injection; at 2 hr the effect was reduced to 67 ± 9 ~ P E (Figure 1A). With 5.6 mg/kg morphine, the peak effect was 100 %MPE at 50 min post-injection (Figure IB); that is, rats did not move their tail from the heat stimulus before the 6.0 sec cutoff time. At 2 hr the ~ P E was still 96 ± 4. At the highest dose of morphine tested in group l, lO mg/kg, rats reached i00 ~MPE in 50 min and remained at that level of analgesia throughout the 2-hr session (Figure iC).
A.
100 ~
C,
i
10 mg/kg Morphine
1O0
O. 80
80
~ (@
60
60
@
40
40
e~ ,<
20
20
5
0 I
I O SAL/NOSTRESS I1 B-FNA/NOSTRESS I • SAL/STRESS • 8-FNA/STRESS
I
B.
5.6 mg/kg Morphine
100 P,
<[
D.
180[ O0
17,5 mg/kg
Morphine
80
l 2or
60
60
40
40
20 0 5 15 30
5
0
120
5 15 30 45 60
M i n u t e s After M o r p h i n e
FIG. i Time-effect curves for morphine analgesia in rats 24 hr after one of the following pretreatments, ICV saline while unstressed; ICV saline while stressed; 2.5 Mg ICV 8-FNA while unstressed; 2.5 pg ICV 8-FNA while stressed. Abscissae: Time in min after SC injection of morphine. Ordinates: Change in tail-flick latency expressed as % maximum possible effect. Each point represents mean for responses in lO rats except panel D where n = 9. P values indicate a significant difference between Ewo time-effect curves.
120
2840
Stress, ~-FNA and Morphine
Vol. 41, No. 26, 1987
Rats restrained and injected with saline (group 2: saline/stress) 24 hr before testing were less sensitive to analgesic actions of morphine than were their unrestrained counterparts. There was a significant difference in time-effect curves for 3.0 mg/kg morphine between groups 1 and 2 (Figure 1A; p < .05). The previously stressed animals reached a peak effect of 73 ~ ll 9#4PE at 30 min, and at 2 hr were down to 29 ± 7 ~#4PE. With 5.6 mg/kg morphine, the mean ~4PE in the previously stressed animals was consistently lower and there was a more rapid offset of analgesic effect between the 1 and 2 hr time points than in the unstressed animals, although the overall curves were not statistically different from one another (Figure 1B). Differences between groups i and 2 were completely absent at lO mg/kg of morphine where the time-effect curves are practically superimposable (Figure iC). Prior restraint stress resulted in an apparent i/4 log-unit shift to the right of the morphine dose-effect curve relative to the curve for animals that had not been stressed (Figure 2).
12000
.,m Y
c ,m
E 10000
IisALINO ESSI SAL/STRESS B-FNA/NO STRESS B-FNA/STRESS
0.
8000
"~
@
p < .05
6000
@@ " 4000
O m
2000
p < .05
I
I
|
I
3.0
5.6
10.0
17.5
Morphine
Dose (mg/kg)
FIG. 2 Oose-effect curves for morphine analgesia in rats treated 24 hr earlier with either ICV saline or ICY 6-FNA (2.5 pg) while either unstressed or stressed. Abscissa, SC morphine dose in mg/kg. Ordinate, Area under the time-effect curves in Figure l, with units of ~PE-min. Symbols and sample sizes as in Figure t. Vertical lines indicate ± SEM and are absent where the SEM is less than the radius of tile point. ~ values indicate a significant difference between two dose-response curves.
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
284]
8-FNA (2.5 ug icv), administered 24 hr prior to testing (group 3: B-FNA/ no stress), effectively antagonized the analgesic response to morphine. With 3.0 m g A g morphine, the blockade was nearly complete as the peak effect was only 18 ± 5 ~4PE (Figure 1A). Ten mg/kg morphine, which produced i00 9#4PE from 30 min to 2 hr in saline-pretreated control animals, produced a peak effect of only 45 ± 9 ~ P E at 60 min in 8-FNA pretreated rats; the analgesic effect decreased to 17 ± 6 ~MPE at 2 hr (Figure IC). Thus, 8-FNA decreased both the magnitude and duration of the effect of morphine on tailflick latencies. The highest dose of morphine tested, 17.5 mg/kg, produced a peak effect of just 57 ± 9 ~NPE (Figure ID); thus, the antagonism was not surmounted. This can be seen as a shift of the morphine dose-response curve to the right and down (Figure 2, compare groups 1 and 3). The maximum analgesic effect attained in B-FNA-pret~eated rats (4596 ~NPE-min with 17.5 mg/kg morphine) was only 41% of that seen in the corresponding control group (11,142 ~NPE-min with lO mg/kg morphine). Animals pretreated with 2.5 ug of 8-FNA icv while stressed (group 4, Beta-FNA/stress) were significantly more sensitive to morphine 24 hr later than were animals that received 8-FNA while unstressed, particularly at the intermediate doses of 5.6 and lO mg/kg of morphine (Figure LB,C; p < .05). The analgesic effect of morphine was characterized by a more rapid onset, a greater peak, and a longer duration in rats subjected previously to stress. For example, lO mg/kg morphine produced a maximum of 65 ± ll ~e4PE at 45 min post-injection in animals that had been stressed while receiving B-FNA compared to 45 ± 9 ~e4PE at 60 min in animals unstressed while receiving 8-FNA. Further, at the 2 hr time point, previously stressed rats exhibited 42 ± ll ~MPE compared to 17 ± 6 ~NPE in unstressed animals. With 17.5 mg/kg morphine, however, the curves were practically superimposable. There was an apparent 1/4-i/2 log-unit shift to the left in the dose-effect curve, but no significant difference in maximum (Figure 2, compare groups 3 and 4). Statistically significant differences in [3H]DAGO binding among the four pretreatment groups were found in the caudate and thalamus, and in [3H]bremazocine binding in the thalamus (Table 2). There was a consistent trend towards increases in binding for both radioligands in the stressed rats pretreated with saline, which was statistically significant for [3H]-bremazocine binding in thalamus. Pretreatment with 8-FNA reduced the binding of [3H]DAGO in the caudate of stressed rats and in the thalamus of stressed and s~ressed rats, but had no effect on binding in the central gray (Table 2). HJ-Bremazocine binding was not altered in brain regions from animals pretreated with 8-FNA.
~
Discussion As little as 2.5 ug of 8-FNA (5.1 nM) administered into the lateral cerebral ventricle of rats significantly and insurmountably antagonized the analgesic effect of morphine 24 hr later. This outcome is consistent with the characterization of 8-FNA as an irreversible antagonist at the mu-opioid receptor (14). Long-lasting morphine-antagonist activity of beta-F~ in vivo also has been shown in tests of analgesia in mice (24), and discrimlna~and locomotor activity in rats (25,26). The insurmountability of the antagonist has also been demonstrated. For example, in monkeys physically dependent on morphine, doses of morphine 20 times higher than those that completely suppressed withdrawal induced by naltrexone, a competitive antagonist, failed to suppress withdrawal induced by 8-FNA (27). Rats that received B-FNA while stressed had a significantly greater analgesic response to morphine 24 hr later than did those that were
2842
Stress, B-FNA and Morphine
Vol. 41, No. 26, 1987
unstressed when 8-FNA was administered. This is consistent with the hypothesis that restraint stress enhances the neuronal release of opioid receptor ligands in the brain, which would protect opioid receptors from alkylation by 8-FNA. B-FNA was administered under experimental conditions,
TABLE ll
Effect of Prior Stress and Pretreatment with 8-FNA on Opioid Binding Sites in Rat Brain
Binding Sites (fmol/mg proteinl mean ± SEMI n = 3) Group
Pretreatment
Caudate
Thalamus
Central Gray
[aH]-DAGO 1
SAL/NO STRESS
68 ± 6
56 + 4
47 _+ 1
2
SAL/STRESS
77 ± 2
58 ± 4
59 ± 4
3
8-FNA/NO STRESS
62 ± 4
45 _+ I*
47 + 3
4
B-FNA/STRESS
53 ± 3 * *
44 + 2*
47 + 1
Significance a,
P < .05
P < .05
N.S.
[3HI-BREMAZOC INE 1
SAL/NO STRESS
347 ± 29
249 _+ 5
222 _+ 4
2
SAL/STRESS
375 ± 14
321 _+ 27 +
258 + 22
3
8-FNA/NO STRESS
329 ± lO
227 + 6
230 ± 25
4
B-FNA/STRESS
320 ± 27
244 ± 22
246 4 14
N.S.
P < .05
Significance,
NoS.
*Significantly
lower than mean of corresponding saline group, p < 0.05
**Significantly
lower than mean of corresponding saline group, p < O.O1
+Significantly higher than other means in column, p < 0.05 aResults of one-way ANOVA on difference by treatment group for column
i.e., restraint, that have been shown to potentiate the analgesic effect of opioid drugs when tests were performed while subjects were restrained (6-8). Therefore, these results support the concept that endogenous opioid peptides are involved in the phenomenon of stress-induced potentiation of opioid analgesia.
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2843
However, consistent differences in mu binding between brains of previously stressed and unstressed rats could not be demonstrated directly in binding assays. In fact, irrespective of stress, t~ea~ment with B-FNA resulted in only small decreases in the binding of L~HJDAGO. To our knowledge, there are no reports on how 8-FNA affects the number of opioid binding sites in vivo. 8-FNA may be alkylating a large percentage of functional mureceptors that mediate morphine analgesia, but a small percentage of total sites that selectively bind mu agonists. Thus, it is possible that the approximate tO-20~ decreases in [ ~ D A G O binding seen in rats pretreated with 8-FNA is sufficient to account for the greater than 50% decrease in the maximum analgesic effect of morphine. On the other hand, chronic treatment with opioid antagonists enhances the analgesic potency of morphine and concommitantly increases the number of opioid binding sites in brain (28-51). This suggests a second possibility. I[ synthesis of new receptors begins within 24 hr of 8-FNA administration, L3HJDAGO would bind to these newly synthesized sites, which might be unlinked to second messenger systems, and thus incapable of mediating opioid analgesia. Stress may affect central opioid systems in many ways. In the present study, rats subjected to restraint stress and injected only with saline ICV were less sensitive to the analgesic effect of low doses of morphine 24 hr later than were the unstressed controls. There are numerous reports of chronic daily stress resulting in tolerance to morphine-induced analgesia (e.g., 32-35). Apparently such tolerance can also occur with periodic stress. Perhaps this phenomenon has relevance to the increased ligand binding observed in brain regions of stressed saline-treated animals (i.e., group 2, Table 2). Because binding experiments were done at a single ligand concentration, it is not yet possible to determine if the changes observed reflect changes in the affinity or the number of binding sites. The effect of restraint stress alone on the subsequent responsiveness of rats to morphine-induced analgesia was opposite in direction to the effect of stress in combination with 6-FNA. Thus, the tolerance to morphine analgesia caused by stress probably served to limit the magnitude of the difference in sensitivity to morphine between the stressed and unstressed animals treated with 6-FNA.
Acknowledgments This research was supported in part by Grants DAO0017 (E.J.S.) and DAO0541 (S.G.H.) and by Research Scientist Award K05 DAO0008 (S.G.H.), all from National Institute on Drug Abuse. References
1. R.J. BODNAR, D.D. KELLY, M. BRUTUS and M. GLUSMAN, Neurosci. Biobehav. Rev. 4 87-i00 (1980). 2. L.R. WATKINS and D.J. MAYER, Science 216, i185-1192 (1982). 3. G.W. TERMAN, Y. SHAVIT, 3.W. LEWIS, J.T. CANNON and J.C. LIEBESKIND, Science 226 1270-1277 (1984). 4. H. SCHLEN and G.A. BENTLEY, Pain 9 73-84 (1980). 5. G.L. BELENKY and J.W. HOLADAY, LTfe Sci. 29 553-563 (1981). 6. B.D. APPELBAUM and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 231555-565
(1984). 7. 8.
B.D. APPELBAUM and S.G. HOLTZMAN, Brain Res. 358 303-308 (1985). B.D. APPELBAUM and S.G. HOLTZMAN, Brain Res. 377 330-336 (1986).
2844
9. lO. 11. 12. 13. 14. 15. 16.
Stress, B-FNA and Morphine
Vol. 41, No. 26, 1987
J. MADDEN, H. AKIL, R.L. PATRICK and J.D. BARCHAS, Nature 265 358-360 (1977). W.T. CHANCE, A.C. WHITE, G.M. KRYNOCK and J.A. ROSECRANS, Eur. J. Pharmacol. 44 283-284 (1977). W.T. CHANCE, A.C. WHITE, G.M. KRYNOCK and J.A. ROSECRANS, Brain Res. 141 371-374 (1978). M.R. HNATOWICH, F.S. LABELLA, K. KIERNAN and G.B. GLAVIN, Brain Res. 380 I07-113 (1986). P.R. PORTOGHESE, D.L. LARSON, L.M. SAYRE, D.S. FRIED and A.E. TAKEMORI, J. Med. Chem. 23 233-234 (1980). A.E. TAKEMORI and P.S. PORTOGHESE, Annu. Rev. Pharmacol. 25 193-223 (1985). L.E. ROBSON and H. KOSTERLITZ, Proc. R. Soc. Lond. B. Biol. Sci. 205 425-432 (1979). J.R. SMITH and E.J. SIMON, Proc. Natl. Acad. Sci. USA 77 281-284
(198o). 17. 18. 19. 20. 21. 22. 23.
24. 25.
26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
I.F. JAMES and A. GOLDSTEIN, Mol. Pharmacol. 25 337-342 (1984). P.S. PORTOGHESE and A.E. TAKEMORI, J. Med. Chem. 26 1341-1343 (1983). F.E. D'AMOUR and D.L. SMITH, J. Pharmacol. Exp. T~-er. 72 74-79 (1941). V.F. GELLERT and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 205 536-546 (1978). O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R.J. RANDALL, O. Biol. Chem. 193 265-275 (1951). E.J. SIMON, J.M. HILLER, 3. GROTH and I. EDELMAN, J. Pharmacol. Exp. Ther. 192 531-537 (1975). W.L. OEWEY and L.S. HARRIS, In, Methods in Narcotic Research, ed. by S. Ehrenpreis and A. Neidle, pp. I01-109, Marcel Dekker, Inc., New York (1975). S.J. WARD, P.S. PORTOGHESEand A.E. TAKEMORI, O. Pharmacol. Exp. Ther. 220 494-498 (1982). K.W. LOCKE and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 238 990-996 ( 1986a). K.W. LOCKE and S.G. HOLTZMAN, O. Pharmacol. Exp. Ther. 238 997-1003 (1986). D.E. GMEREK and O.H. WOODS, J. Pharmacol. Exp. Ther. 235 296-301 (1985). R.A. LAHTI and R.J. COLLINS, Eur. O. Pharmacol. 51 185-186 (1978). R.S. ZUKIN, J.R. SUGARMAN, M.L. FITZ-SYANGE, E.L. GARDNER, S.R. ZUKIN and A.R. GINTZLER, Brain Res. 245 285-292 (1982). A. TEMPEL, E.L. GARDNER and R.S. ZUKIN, J. Pharmacol. Exp. Ther. 232 439-444 (1984). B.C. YOBURN, R.R. GOODMAN, A.H. COHEN, G.W. PASTERNAK and C.E. INTURRISI, Life Sci. 36 2325-2332 (1985). D.J. MAYER and R.L. HAYNES, Science 188 941-943 (1975). G.B. CHESHER and B. CHAN, Life Sci. 21 1569-1574 (1977). A. SPIAGGIA, R.J. BODNAR, D.D. KELLY and M. GLUSMAN, Pharmacol. Biochem. Behav. i0 761-765 (1979). A.E. SNOW and W.L-- DEWEY, J. Pharmacol. Exp. Ther. 227 42-50 (1983).
Pergamon Journal~
EFFECTS OF STRESS ANO B-FUNALTREXAMINE PRETREATMENT ON MORPHINE ANALGESIA AND OPIOID BINDING IN RATS J.U. Adams ~, J.S. Andrewsl, 2, J.M. Hiller s , E.J. Simon3, ~ and S.G. Holtzman l ~Department of Pharmacology, Emory University School of Medicine Atlanta, Georgia 30322 and 3Departments of Psychiatry and ~Pharmacology, New York University Medical Center, 550 First Avenue, New York, New York lOO16 (Received in final form November 3, 1987) Summary This study was essentially an in vivo protection experiment designed to test further the hypothesis that stress induces release of endogenous opioids which then act at opioid receptors. Rats that were either subjected to restraint stress for 1 hr or unstressed were injected ICY with either saline or 2.5 pg of Bfunaltrexamine (B-FNA), an irreversible opioid antagonist that alkylates the mu-opioid receptor. Twenty-four hours later, subjects were tested unstressed for morphine analgesia (tail-flick assay) oK w@re sacrificed and opioid binding in brain was determined. [3HJD-Ala2NMePhe~-GlyS(ol)enkephalin (DAGQ) served as a specific ligand for mu-- opioid receptors, and L3HJ-bremazocine as a general ligand for all opioid receptors. Rats injected with saline while stressed were significantly less sensitive to the analgesic action of morphine 24 hr later than were their unstressed counterparts. B-FNA pretreatment attenuated morphine analgesia in an insurmountable manner. Animals pretreated with B-FNA while stressed were significantly more sensitive to the analgesic effect of morphine than were animals that received 8-FNA while unstressed, consistent with the hypothesis that stress induces release of endogenous opioids that would protect opioid receptors from alk~la~ion by B-FNA. 6-FNA caused small and similar decreases in [3HJ-DAGO binding in brain of both stressed and unstressed animals. Stressed rats injected with saline tended to have increased levels of [3H]DAGO and [~H]bremazocine binding compared to the other groups. This outcome may be relevant to the tolerance to morphine analgesia caused by stress. Endogenous opioids have been implicated in many of the biochemical, physiological, and behavioral responses to stress. For example, opioid involvement in the phenomenon of stress-induced analgesia has been widely investigated (for reviews see 1-3). Not only can stressful events induce analgesia, they can also potentiate the analgesic affect of exogenously administered opioid drugs (4-6). For example, both the magnitude and dura2present address: Berlin 65.
Department of Neuropsychopharmacology, Schering AG, D-IO00
0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
2836
Stress, 6-FNA and Morphine
Vol. 41, No. 26, 1987
tion of the analgesic effect of morphine and methadone administered subcutaneously were enhanced significantly in rats that were either restrained or placed in a cold environment at the time of testing as compared to unrestrained and presumably minimally stressed animals (6). Similar results were obtained when morphine or enkephalin analogs were administered intracerebroventricularly (ICY) (7,8), indicating that the interaction between stress and exogenous opioids is mediated centrally. The mechanisms that underlie stress-induced potentiation of opioid analgesia are unknown, but also may involve endogenous opioids. Stress has been associated with increased brain levels of opioid peptides (9) and with a decreased number of opioid binding sites in brain. The latter finding is due presumably to the presence of competing endogenous ligands (lO,11) or to receptor downregulation from prolonged exposure of receptors to elevated levels of endogenous ligands (12). Beta-Funaltrexamine (8-FNA), the fumarate methyl ester derivative of the opioid antagonist naltrexone, is a non-equilibrium opioid receptor antagonist (13). It is believed to alkylate relatively selectively the mu-opioid receptor (14). Receptors can be protected from alkylation in vitro if they are exposed to reversible receptor ligands, agonists or antagonists, before being exposed to the inactivating agent (e.g., 15-17). For example, incubation of the guinea pig ileum with 6-FNA produced a rightward shift in the morphine concentration-effect curve even after the tissue was washed numerous times. However, if naltrexone was added to the tissue bath before 8-FNA, after the tissue was washed there was little antagonism of the effects of morphine, presumably because naltrexone prevented 8-FNA from occupying and alkylating the receptors (18)° The present study was essentially an in vivo protection experiment designed to provide further evidence for the stress-induced release of endogenous opioids and their subsequent interaction at opioid receptors. We postulated that stress increases the occupation of opioid receptors by their endogenous ligands. Therefore, 8-FNA would alkylate fewer receptors in rats subjected to stress (i.e. restraint) than in rats that were not stressed at the time of 8-FNA administration. The experimental design was similar to the one used to demonstrate stress-induced potentiation of opioid analgesia (5-8), with one important difference. Stress-induced potentiation of opioid analgesia occurs in rats that are tested for analgesia while restrained. In the present study, 8-FNA or saline was administered ICV to separate groups of rats that were either restrained or unrestrained. Twenty-four hours later, the analgesic effect of morphine was determined in all animals while they were unrestrained. The magnitude of the analgesic response to morphine should be an inverse function of the number of opioid receptors inactivated by 8-FNA. Other groups of animals were also treated with either 8-FNA or saline while stressed or unstressed. Twenty-four hours later, instead of being tested with morphine, these rats were sacrificed and the number of opioid binding sites was determined in several regions of the brain. Methods
Animals. Subjects were Sprague-Dawley derived male rats (Sasco Inc., Omaha, NE)--weighing between 250 and 275 g at the beginning of the study. Rats were housed in pairs under a 12-hour light/dark cycle, with food and water available ad libitum. Surgery. Rats were anesthetized with 50 mg/kg sodium pentobarbital and, if additional anesthetic was needed, lO0 mg/kg chloral hydrate. A 22 g stainless steel guide cannula (Plastic Products, Inc., Roanoke, VA) was cut
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2837
to a length of 5.5 mm and beveled and was surgically implanted into the lateral ventricle. Using a stereotaxic apparatus with the upper incisor bar set at 3.3 mm below horizontal zero, the cannula was placed at 0.8 mm posterior and 1.4 mm lateral relative to bregma, and 3.5 mm ventral to the skull, and was permanantly secured with skull screws (no. 0.-80 x 3/32 in) and dental acrylic (Lang Dental Mfg. Co., Chicago, IL). A 28 g dummy cannula (Plastic Products), cut to protude 0.5 mm beyond the tip of the guide cannula, was kept in the implanted cannula except during ICV injections. Procedure. Rats were divided into four groups according to pretreatmerit: l) saline/no stress, 2) saline/stress, 3) B-FNA/no stress, 4) 8-FNA/stress. Animals were habituated to handling and test procedures and to restraint stress (i.e., groups 2 and 4) for 5 days before drug testing began. Tests were conducted on two consecutive days, once each week. On the pretreatment day, animals in groups 2 and 4 were subjected to restraint stress by immobilizing them in Plexiglas cylinders (5 and 6 cm in diameter) for I hr. All animals received an ICV injection, either 2.5 pl of sterile bacteriostatic isotonic saline (groups i and 2) or 2.5 pg B-FNA (groups 3 and 4) in a volume of 2.5 pl. Injections were performed on freely moving rats with a 25 pl Hamilton syringe (Hamilton Co., Reno, NV) connected to a 28 g injection cannula (Plastic Products) by PE 20 tubing. Injections were given over a 1-min period and the injection cannula was kept in place for an additional 30-60 sec. Stressed animals were removed from restraint tubes 30 min after the start of restraint stress to receive their injections and then were replaced for the remaining 30 min. Analgesia. Twenty-four hours later, animals were tested for morphine analgesia using the tail-flick method of D'Amour and Smith (19), with modifications (20). Radiant heat from a 20 V high amperage bulb was focused on the lower half of the rat's tail with an ellipsoidal reflector (Optical Industries, Costa Mesa, CA). Movement of the tail allowed the light to activate a photocell, which stopped both the stimulus and the reaction timer. Stimulus intensity was controlled by a variable transformer set to give baseline latencies of 2.0-2.5 sec. A cut-off time of 6.0 sec was used to minimize damage to the tail. After a habituation trial (no stimulus) and a baseline trial, rats were injected SC with one of four doses of morphine, and were tested for latency to tail-flick 5, 15, 30, 45, 60, and 120 min later. After the last trial, rats were administered a dose of naltrexone one-tenth that of the morphine test dose in order to ensure termination of the effects of morphine. Each animal was tested at weekly intervals. Half of the animals in each group received morphine doses in ascending order, beginning with 3.0 mg/kg, and the other half received morphine doses in descending order, ending with 3.0 mg/kg. Binding. A different set of rats was divided into four groups, habituated to handling and stress, and then subjected to the same four pretreatments as described above. On the following day, the animals were decapitated and their brain was removed and rapidly cooled on a thermoelectric cold plate. The caudate, thalamus, and central gray areas were dissected out. Brain areas from three animals were pooled, weighed and homogenized in lO volumes of 50 Tris-HC1 buffer, pH 7.4, containing 1 mM dipotassium EDTA, using a Brinkmann Polytron (Brinkmann Instruments, Inc., Westbury, NY) at a setting of 6 for 20 sec. The homogenate was centrifuged at 20,000 x g for 15 min and the pellet was resuspended in six volumes of 0.32 M sucrose using the Polytron set at 4, and stored at -60°C until used. Prior to equilibrium binding experiments, membrane preparations were thawed and diluted lO-fold in the above Tris buffer so as to give a final
2838
Stress,
B-FNA and Morphine
Vol. 41, No. 26, 1987
protein concentration of approximately l.O mg/ml, as determined by the method of Lowry et al. (21). Binding assays on triplicate 1.O ml samples were carried out essentially as previously described (22) at 25°C for i hr in the presence and absence of a lO00-fold excess of unlabeled naloxone. In these preliminary experiments to screen for possible changes in binding levels, a single concentration of each labeled ligand was used. The concentration of the labeled ligands was within a factor of 2 of their respective Kds. To assess mu site ~in@ing, [3HI D-Ala2-MeRhe"-GlyS(ol)-enkephalin (3H-DAGO, i nM) was u---sad. L3HJ-Bremazocine (1 nM), possessing almost equal affinity for mu, delta and kappa binding sites, was used to assess total opioid binding levels. Separation of bound from free ligand was accomplished by filtration through GF/B filters followed by two washes with 4.0 ml of Tris buffer. Radioactivity remaining on dried filters was determined by scintillation spectroscopy in a 8eckman LS-515]-T counter. Data Analysis. Analgesia data are expressed as percent maximum possible effect (~MPE] (23): postdrug latency - baseline latency %MPE .......................................... cut-off time (6.0 sec) - baseline latency Dose-effect curves were generated by calculating the area under the time-ZMRE curve using the trapezoidal rule. Differences between the groups' timeeffect curves and differences between dose-effect curves were analyzed using two-way ANOVA and post hoc Newman-Keuls test. Binding data were analyzed by one-way ANOVA. Drugs. The drugs used in the analgesia experiments were morphine sulfate (Penick Corp., Newark, NJ), naltrexone hydrochloride, and B-FNA hydrochloride (both from National Institute on Drug Abuse, Rockville, MD). Morphine and naltrexone were dissolved in 0.gZ saline; B-FNA was dissolved in sterile bacteriostatic saline for icy administration. Subcutaneous injections were delivered in a volume of 1.0 ml/kg. All doses are expressed ~n terms of the free base. Ligands used in the binding ex_periments were 3 [~HJDAGO (49 Ci/mMol) (Amersham Corp., Des Plaines, IL), [3HJ-bremazocine (32 Ci/mMol) (New England Nuclear Corp., Boston, MA), and naloxone hydrochloride (National Institute on Drug Abuse). TABLE I Pre-Drug Response Latencies in Tail-Flick Test
Group
Pretceatment
Seconds (mean ± SEM; n = i0)
i
SAL/NO STRESS
2.24 + .07
2
SAL/STRESS
2.14 _+ .i0
5
B-FNA/NO STRESS
2.11 + .07
4
B-FNA/STRESS
2.16 _+ .lO
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2839
Results Pretreatment conditions had no effect on baseline latencies in the tailflick test performed 24 hr later (Table i; p > .05). However, pretreatment conditions did modify significantly the responses of the animals to morphine. Morphine produced increases in latencies to tail flick in all four groups (Figure i). There were dose-related increases in both peak effect and duration of effect. In the control rats (group l: saline/no stress), 5.0 mg/kg morphine produced a peak effect of 95 ± 5 ~4PE at 45 min post-injection; at 2 hr the effect was reduced to 67 ± 9 ~ P E (Figure 1A). With 5.6 mg/kg morphine, the peak effect was 100 %MPE at 50 min post-injection (Figure IB); that is, rats did not move their tail from the heat stimulus before the 6.0 sec cutoff time. At 2 hr the ~ P E was still 96 ± 4. At the highest dose of morphine tested in group l, lO mg/kg, rats reached i00 ~MPE in 50 min and remained at that level of analgesia throughout the 2-hr session (Figure iC).
A.
100 ~
C,
i
10 mg/kg Morphine
1O0
O. 80
80
~ (@
60
60
@
40
40
e~ ,<
20
20
5
0 I
I O SAL/NOSTRESS I1 B-FNA/NOSTRESS I • SAL/STRESS • 8-FNA/STRESS
I
B.
5.6 mg/kg Morphine
100 P,
<[
D.
180[ O0
17,5 mg/kg
Morphine
80
l 2or
60
60
40
40
20 0 5 15 30
5
0
120
5 15 30 45 60
M i n u t e s After M o r p h i n e
FIG. i Time-effect curves for morphine analgesia in rats 24 hr after one of the following pretreatments, ICV saline while unstressed; ICV saline while stressed; 2.5 Mg ICV 8-FNA while unstressed; 2.5 pg ICV 8-FNA while stressed. Abscissae: Time in min after SC injection of morphine. Ordinates: Change in tail-flick latency expressed as % maximum possible effect. Each point represents mean for responses in lO rats except panel D where n = 9. P values indicate a significant difference between Ewo time-effect curves.
120
2840
Stress, ~-FNA and Morphine
Vol. 41, No. 26, 1987
Rats restrained and injected with saline (group 2: saline/stress) 24 hr before testing were less sensitive to analgesic actions of morphine than were their unrestrained counterparts. There was a significant difference in time-effect curves for 3.0 mg/kg morphine between groups 1 and 2 (Figure 1A; p < .05). The previously stressed animals reached a peak effect of 73 ~ ll 9#4PE at 30 min, and at 2 hr were down to 29 ± 7 ~#4PE. With 5.6 mg/kg morphine, the mean ~4PE in the previously stressed animals was consistently lower and there was a more rapid offset of analgesic effect between the 1 and 2 hr time points than in the unstressed animals, although the overall curves were not statistically different from one another (Figure 1B). Differences between groups i and 2 were completely absent at lO mg/kg of morphine where the time-effect curves are practically superimposable (Figure iC). Prior restraint stress resulted in an apparent i/4 log-unit shift to the right of the morphine dose-effect curve relative to the curve for animals that had not been stressed (Figure 2).
12000
.,m Y
c ,m
E 10000
IisALINO ESSI SAL/STRESS B-FNA/NO STRESS B-FNA/STRESS
0.
8000
"~
@
p < .05
6000
@@ " 4000
O m
2000
p < .05
I
I
|
I
3.0
5.6
10.0
17.5
Morphine
Dose (mg/kg)
FIG. 2 Oose-effect curves for morphine analgesia in rats treated 24 hr earlier with either ICV saline or ICY 6-FNA (2.5 pg) while either unstressed or stressed. Abscissa, SC morphine dose in mg/kg. Ordinate, Area under the time-effect curves in Figure l, with units of ~PE-min. Symbols and sample sizes as in Figure t. Vertical lines indicate ± SEM and are absent where the SEM is less than the radius of tile point. ~ values indicate a significant difference between two dose-response curves.
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
284]
8-FNA (2.5 ug icv), administered 24 hr prior to testing (group 3: B-FNA/ no stress), effectively antagonized the analgesic response to morphine. With 3.0 m g A g morphine, the blockade was nearly complete as the peak effect was only 18 ± 5 ~4PE (Figure 1A). Ten mg/kg morphine, which produced i00 9#4PE from 30 min to 2 hr in saline-pretreated control animals, produced a peak effect of only 45 ± 9 ~ P E at 60 min in 8-FNA pretreated rats; the analgesic effect decreased to 17 ± 6 ~MPE at 2 hr (Figure IC). Thus, 8-FNA decreased both the magnitude and duration of the effect of morphine on tailflick latencies. The highest dose of morphine tested, 17.5 mg/kg, produced a peak effect of just 57 ± 9 ~NPE (Figure ID); thus, the antagonism was not surmounted. This can be seen as a shift of the morphine dose-response curve to the right and down (Figure 2, compare groups 1 and 3). The maximum analgesic effect attained in B-FNA-pret~eated rats (4596 ~NPE-min with 17.5 mg/kg morphine) was only 41% of that seen in the corresponding control group (11,142 ~NPE-min with lO mg/kg morphine). Animals pretreated with 2.5 ug of 8-FNA icv while stressed (group 4, Beta-FNA/stress) were significantly more sensitive to morphine 24 hr later than were animals that received 8-FNA while unstressed, particularly at the intermediate doses of 5.6 and lO mg/kg of morphine (Figure LB,C; p < .05). The analgesic effect of morphine was characterized by a more rapid onset, a greater peak, and a longer duration in rats subjected previously to stress. For example, lO mg/kg morphine produced a maximum of 65 ± ll ~e4PE at 45 min post-injection in animals that had been stressed while receiving B-FNA compared to 45 ± 9 ~e4PE at 60 min in animals unstressed while receiving 8-FNA. Further, at the 2 hr time point, previously stressed rats exhibited 42 ± ll ~MPE compared to 17 ± 6 ~NPE in unstressed animals. With 17.5 mg/kg morphine, however, the curves were practically superimposable. There was an apparent 1/4-i/2 log-unit shift to the left in the dose-effect curve, but no significant difference in maximum (Figure 2, compare groups 3 and 4). Statistically significant differences in [3H]DAGO binding among the four pretreatment groups were found in the caudate and thalamus, and in [3H]bremazocine binding in the thalamus (Table 2). There was a consistent trend towards increases in binding for both radioligands in the stressed rats pretreated with saline, which was statistically significant for [3H]-bremazocine binding in thalamus. Pretreatment with 8-FNA reduced the binding of [3H]DAGO in the caudate of stressed rats and in the thalamus of stressed and s~ressed rats, but had no effect on binding in the central gray (Table 2). HJ-Bremazocine binding was not altered in brain regions from animals pretreated with 8-FNA.
~
Discussion As little as 2.5 ug of 8-FNA (5.1 nM) administered into the lateral cerebral ventricle of rats significantly and insurmountably antagonized the analgesic effect of morphine 24 hr later. This outcome is consistent with the characterization of 8-FNA as an irreversible antagonist at the mu-opioid receptor (14). Long-lasting morphine-antagonist activity of beta-F~ in vivo also has been shown in tests of analgesia in mice (24), and discrimlna~and locomotor activity in rats (25,26). The insurmountability of the antagonist has also been demonstrated. For example, in monkeys physically dependent on morphine, doses of morphine 20 times higher than those that completely suppressed withdrawal induced by naltrexone, a competitive antagonist, failed to suppress withdrawal induced by 8-FNA (27). Rats that received B-FNA while stressed had a significantly greater analgesic response to morphine 24 hr later than did those that were
2842
Stress, B-FNA and Morphine
Vol. 41, No. 26, 1987
unstressed when 8-FNA was administered. This is consistent with the hypothesis that restraint stress enhances the neuronal release of opioid receptor ligands in the brain, which would protect opioid receptors from alkylation by 8-FNA. B-FNA was administered under experimental conditions,
TABLE ll
Effect of Prior Stress and Pretreatment with 8-FNA on Opioid Binding Sites in Rat Brain
Binding Sites (fmol/mg proteinl mean ± SEMI n = 3) Group
Pretreatment
Caudate
Thalamus
Central Gray
[aH]-DAGO 1
SAL/NO STRESS
68 ± 6
56 + 4
47 _+ 1
2
SAL/STRESS
77 ± 2
58 ± 4
59 ± 4
3
8-FNA/NO STRESS
62 ± 4
45 _+ I*
47 + 3
4
B-FNA/STRESS
53 ± 3 * *
44 + 2*
47 + 1
Significance a,
P < .05
P < .05
N.S.
[3HI-BREMAZOC INE 1
SAL/NO STRESS
347 ± 29
249 _+ 5
222 _+ 4
2
SAL/STRESS
375 ± 14
321 _+ 27 +
258 + 22
3
8-FNA/NO STRESS
329 ± lO
227 + 6
230 ± 25
4
B-FNA/STRESS
320 ± 27
244 ± 22
246 4 14
N.S.
P < .05
Significance,
NoS.
*Significantly
lower than mean of corresponding saline group, p < 0.05
**Significantly
lower than mean of corresponding saline group, p < O.O1
+Significantly higher than other means in column, p < 0.05 aResults of one-way ANOVA on difference by treatment group for column
i.e., restraint, that have been shown to potentiate the analgesic effect of opioid drugs when tests were performed while subjects were restrained (6-8). Therefore, these results support the concept that endogenous opioid peptides are involved in the phenomenon of stress-induced potentiation of opioid analgesia.
Vol. 41, No. 26, 1987
Stress, B-FNA and Morphine
2843
However, consistent differences in mu binding between brains of previously stressed and unstressed rats could not be demonstrated directly in binding assays. In fact, irrespective of stress, t~ea~ment with B-FNA resulted in only small decreases in the binding of L~HJDAGO. To our knowledge, there are no reports on how 8-FNA affects the number of opioid binding sites in vivo. 8-FNA may be alkylating a large percentage of functional mureceptors that mediate morphine analgesia, but a small percentage of total sites that selectively bind mu agonists. Thus, it is possible that the approximate tO-20~ decreases in [ ~ D A G O binding seen in rats pretreated with 8-FNA is sufficient to account for the greater than 50% decrease in the maximum analgesic effect of morphine. On the other hand, chronic treatment with opioid antagonists enhances the analgesic potency of morphine and concommitantly increases the number of opioid binding sites in brain (28-51). This suggests a second possibility. I[ synthesis of new receptors begins within 24 hr of 8-FNA administration, L3HJDAGO would bind to these newly synthesized sites, which might be unlinked to second messenger systems, and thus incapable of mediating opioid analgesia. Stress may affect central opioid systems in many ways. In the present study, rats subjected to restraint stress and injected only with saline ICV were less sensitive to the analgesic effect of low doses of morphine 24 hr later than were the unstressed controls. There are numerous reports of chronic daily stress resulting in tolerance to morphine-induced analgesia (e.g., 32-35). Apparently such tolerance can also occur with periodic stress. Perhaps this phenomenon has relevance to the increased ligand binding observed in brain regions of stressed saline-treated animals (i.e., group 2, Table 2). Because binding experiments were done at a single ligand concentration, it is not yet possible to determine if the changes observed reflect changes in the affinity or the number of binding sites. The effect of restraint stress alone on the subsequent responsiveness of rats to morphine-induced analgesia was opposite in direction to the effect of stress in combination with 6-FNA. Thus, the tolerance to morphine analgesia caused by stress probably served to limit the magnitude of the difference in sensitivity to morphine between the stressed and unstressed animals treated with 6-FNA.
Acknowledgments This research was supported in part by Grants DAO0017 (E.J.S.) and DAO0541 (S.G.H.) and by Research Scientist Award K05 DAO0008 (S.G.H.), all from National Institute on Drug Abuse. References
1. R.J. BODNAR, D.D. KELLY, M. BRUTUS and M. GLUSMAN, Neurosci. Biobehav. Rev. 4 87-i00 (1980). 2. L.R. WATKINS and D.J. MAYER, Science 216, i185-1192 (1982). 3. G.W. TERMAN, Y. SHAVIT, 3.W. LEWIS, J.T. CANNON and J.C. LIEBESKIND, Science 226 1270-1277 (1984). 4. H. SCHLEN and G.A. BENTLEY, Pain 9 73-84 (1980). 5. G.L. BELENKY and J.W. HOLADAY, LTfe Sci. 29 553-563 (1981). 6. B.D. APPELBAUM and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 231555-565
(1984). 7. 8.
B.D. APPELBAUM and S.G. HOLTZMAN, Brain Res. 358 303-308 (1985). B.D. APPELBAUM and S.G. HOLTZMAN, Brain Res. 377 330-336 (1986).
2844
9. lO. 11. 12. 13. 14. 15. 16.
Stress, B-FNA and Morphine
Vol. 41, No. 26, 1987
J. MADDEN, H. AKIL, R.L. PATRICK and J.D. BARCHAS, Nature 265 358-360 (1977). W.T. CHANCE, A.C. WHITE, G.M. KRYNOCK and J.A. ROSECRANS, Eur. J. Pharmacol. 44 283-284 (1977). W.T. CHANCE, A.C. WHITE, G.M. KRYNOCK and J.A. ROSECRANS, Brain Res. 141 371-374 (1978). M.R. HNATOWICH, F.S. LABELLA, K. KIERNAN and G.B. GLAVIN, Brain Res. 380 I07-113 (1986). P.R. PORTOGHESE, D.L. LARSON, L.M. SAYRE, D.S. FRIED and A.E. TAKEMORI, J. Med. Chem. 23 233-234 (1980). A.E. TAKEMORI and P.S. PORTOGHESE, Annu. Rev. Pharmacol. 25 193-223 (1985). L.E. ROBSON and H. KOSTERLITZ, Proc. R. Soc. Lond. B. Biol. Sci. 205 425-432 (1979). J.R. SMITH and E.J. SIMON, Proc. Natl. Acad. Sci. USA 77 281-284
(198o). 17. 18. 19. 20. 21. 22. 23.
24. 25.
26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
I.F. JAMES and A. GOLDSTEIN, Mol. Pharmacol. 25 337-342 (1984). P.S. PORTOGHESE and A.E. TAKEMORI, J. Med. Chem. 26 1341-1343 (1983). F.E. D'AMOUR and D.L. SMITH, J. Pharmacol. Exp. T~-er. 72 74-79 (1941). V.F. GELLERT and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 205 536-546 (1978). O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R.J. RANDALL, O. Biol. Chem. 193 265-275 (1951). E.J. SIMON, J.M. HILLER, 3. GROTH and I. EDELMAN, J. Pharmacol. Exp. Ther. 192 531-537 (1975). W.L. OEWEY and L.S. HARRIS, In, Methods in Narcotic Research, ed. by S. Ehrenpreis and A. Neidle, pp. I01-109, Marcel Dekker, Inc., New York (1975). S.J. WARD, P.S. PORTOGHESEand A.E. TAKEMORI, O. Pharmacol. Exp. Ther. 220 494-498 (1982). K.W. LOCKE and S.G. HOLTZMAN, J. Pharmacol. Exp. Ther. 238 990-996 ( 1986a). K.W. LOCKE and S.G. HOLTZMAN, O. Pharmacol. Exp. Ther. 238 997-1003 (1986). D.E. GMEREK and O.H. WOODS, J. Pharmacol. Exp. Ther. 235 296-301 (1985). R.A. LAHTI and R.J. COLLINS, Eur. O. Pharmacol. 51 185-186 (1978). R.S. ZUKIN, J.R. SUGARMAN, M.L. FITZ-SYANGE, E.L. GARDNER, S.R. ZUKIN and A.R. GINTZLER, Brain Res. 245 285-292 (1982). A. TEMPEL, E.L. GARDNER and R.S. ZUKIN, J. Pharmacol. Exp. Ther. 232 439-444 (1984). B.C. YOBURN, R.R. GOODMAN, A.H. COHEN, G.W. PASTERNAK and C.E. INTURRISI, Life Sci. 36 2325-2332 (1985). D.J. MAYER and R.L. HAYNES, Science 188 941-943 (1975). G.B. CHESHER and B. CHAN, Life Sci. 21 1569-1574 (1977). A. SPIAGGIA, R.J. BODNAR, D.D. KELLY and M. GLUSMAN, Pharmacol. Biochem. Behav. i0 761-765 (1979). A.E. SNOW and W.L-- DEWEY, J. Pharmacol. Exp. Ther. 227 42-50 (1983).