- Email: [email protected]
Genetic influences in opioid analgesic sensitivity in mice
Brain Research, 566 (1991) 295-298 (~) 1991 Elsevier Science Publishers B.V. All rights reserved, 0006.8993/91/$03.50
BRES 24941
295
Short Communications
Genetic influences in opioid analgesic sensitivity in mice C h a i m G . P i c k 1, Jie C h a n g 1, D e n n i s P a u l I a n d G a v r i l W. P a s t e r n a k 1'2 aThe Cotr.ias Laboratory of Neuro.Oncology, Memorial SIoan-Kettering Cancer Center, New York, NY 10021 (U.S.A.) and 2Departments of Neurology and Pharmacology, Cornell University Medical College, New York, NY 10021 (U.S.A.)
(Accepted 27 August 1991) Key words: Morphine; ~ Receptor; ~ Receptor; Analgesia
Studies of various strains of mice revealed marked differences in their analgesic sensitivity towards morphine (~), U50,488H (~:1) and naloxone benzoylhydrazone(NalBzoH; g3). Sensitivityto F and g analgesiavaried independentlyof the other. Analgesicsensitivityto morphine remained relativelyconsistent among 3 different nociceptive assays for each strain. However, the sensitivityof an individualstrain to NalBzoH remained highly dependent upon the assay used. CD-1 mice were sensitive to NalBzoH in all 3 assays, but in BALB/cmice HalBzoH produced analgesia only in the hot plate and cold water tail-flick assays, In Swiss-Webstermice, NalBzoH was active in the radiant heat and cold water tail-flicksbut inactive in the hot plate. Althoughthe levelsof ~, gl and ~3 binding in whole brain homogenatesdid vary somewhat, they did not correlate with analgesic sensitivity. These results suggests that the genetic controls over ~ and ~ analgesia operate independently and further illustrate the many difficulties in evaluatingpotential analgesics. The ability of morphine to relieve pain is dependent upon both the assay and the species. Even within a single species, the potency of morphine can vary enormously t,7,9-tt'u'lT'ls'2°. A number of opioid receptor subtypes have been identified which can influence the perception of pain. Recently studies with naloxone benzoylhydrazone (NalBzoH) identified another subtype, ~3, which could elicit analgesia through a unique supraspinal opioid mechanism3's't3't6. Although NalBzoH analgesia was reversed by the traditional opioid antagonists naloxone and WIN44,441, more selective opioid antagonists, including ~.funaltrexamine (,u), naltrindole (6) and norbinaltorphimine (gl) did not reverse NalBzoH analgesia, Furthermore, NalBzoH analgesia did not demonstrate cross-tolerance with F, 6 or ~t drugs. In the present study we wished to examine potential genetic influences in the sensitivity of strains of mice to NaiBzoH and whether sensitivity to 1¢3ligands covaried with sensitivity to/~ ligands. Male CD-1, CFW(SW)BR (Swiss-Webster), and BALB/cAnNCrlBR (BALB/c) mice (25-35 g) were purchased from Charles River Breeding Laboratories (Wilmington, MA). C57/J6-bgJ/bg J (C57/bgJ), C57BL/J6bgJ/+ (C57/+) and CXBK/Byj (CXBK) mice (20-30 g) were obtained from Jackson Laboratory (Bar Harbor, MA) and HS/Ibg (HS) (25-30 g) from the Institute of Behavioral Genetics (Boulder, CO). All mice were housed in groups of 5 and were maintained on a 12 h light:12 h
dark cycle with Purina rodent chow and water available ad libitum. Morphine sulfate and U50,488-H {trans-3,4dchloro-N-methyl-N-[2-(1-pyrrolindinyl)'cyclohexyl]'ben" zeneacetamide} were obtained from the Research Technology Branch of NIDA. NaiBzoH and [3H]NalBzoH were synthesized as described previouslys't6. Morphine doses were expressed as the salt and NalBzoH and US0, 488H as the free base. NalBzoH was dissolved in 30% ethanol, which did not produce analgesia in control mice. All drugs were administered subcutaneously. All other drugs were dissolved in saline. Radioligands and Formula 963 Scintillation Fluor were purchased from New England Nuclear (Boston, MA). Analgesia was determined quantally as a doubling or greater of individual baseline latencies using the radiant heat tail-flick technique 4'5'7't3, cold water tail-flick t~ or the hot plate assay 13'2t, Baseline latencies in all assays were defined as the mean of two determinations. Maximal latencies were 10 s in the radiant tail-flick assay and 60 s in the cold water tail-flick and hot plate assays. For the cold water tail-flick assay, the tails were submerged approximately half-way into a 0 °C water bath and the latency to remove, or flick, the tail measured. The hot plate assay was performed at 50 °C using a commercially available apparatus (IITC, Inc, Woodland hills, CA). Trials were terminated by either jumping or paw licking. Dose-response curves were analyzed using a modification of the BLISS-20 computer program. This program
Correspondence: G.W. Pasternak, Department of Neurology, MemorialSloan-KetteringCancer Center, 1275York Avenue, New York, NY
10021, U.S.A. Fax: (1) (212) 717-3296.
296 maximizes the log-likelihood function to fit a parallel set of Gaussian normal sigmoid curves to the dose-response data t9. Single-dose antagonist studies were analyzed using Student's t-tests (P < 0.05). Binding assays were performed in mouse homogenares using the procedures previously reported 2'3. Brains were homogenized in 50 vols. of Tris (50 mM; pH 7.4 at 25 °C) buffer containing PMSF (0.1 mM), Na2EDTA (1 raM) and NaCI (100 mM), incubated at 25 °C for 15 min to remove endogenous ligands t2, and centrifuged at 45, 000 g at 4 °C for 30 rain. The pellets were resuspended in 100 vols. of potassium phosphate buffer (50 raM; pH 7.4; I0 mg wet weight tissue/ml) and 1 or 2 ml aliquots used in all binding assays. All binding assays were preformed at 25 °C. g3 Binding was determined using [3H]NalBzoH for 60 min in the presence of K2EDTA (5 mM); [3H]DAMGO (,u) binding for 150 rain in the presence of MgSO4 (1 mM); and [3H]U69,593 (gt) binding for 45 min. When using [3H]U69,593 all filters were treated with polyethylenimine (1%) to decrease filter binding. Non-specific binding was determined with levallorphan (1 pM). All points were determined in triplicate and only specific binding is reported. All values are pre-
sented as the means -+ S.E.M. of the stated number of independent determinations. Administration of single doses of prototypical/~ (morphine), ut (U50,488H) and K3 (NaiBzoH) drugs produced an interesting analgesic profile among the various strains (Table I). With the exception of the CXBK strain, which is known to be very insensitive to morphine Lm'2°, all the strains were sensitive to morphine (5 mg/kg, s.c.), although their responses did vary considerably. We also observed markedly variable responses among the strains for U50,488H and NalBzoH. The CD-1 strain was the most sensitive across the various drugs. Although the BALB/c strain may be more sensitive to morphine than the CD-1 strain, the B A L B / c mice were relatively insensitive to both ~ agents. The Swiss-Webster strain on the other hand, showed similar decreases in analgesic sensitivity for all 3 classes of drugs. We next examined the tail-flick responses of the CD-1, Swiss-Webster and BALB/c strains in greater detail with dose-response curves (Table I). The Swiss-Webster strain was significantly less sensitive to morphine than the other two strains, but was more sensitive to NalBzoH and U50,488H than the BALB/c strain. It was in-
TABLE I Analgesic sensMviry of various strains of mice in the single dose studies, groups of mice (n ;~ 10) were tested with a single dose of morphine (5 mg/kg, s.c.), NalBzoH (50 mg/kg, s.c.)
or US0,488H (5 mg/kg, s.c.) in the radiant heat tail.flick assay. In the dose-response studies, EDso values were determined frome doseresponse curves composed of at least 3 different drug concentrations with at least 10 mice/dose. R~sults are the EDs0 values with 95% confidence limits. Th~ highest dose of NalBzoH which could be tested was 75 mg/kg, s.c., due to solubility problems, Morphine
NalB~oH
USO,488H
76% 40%
54% 29%
70% 30%
BALB/c
90%
14%
10%
C57/bgJ C57/+ CXBK HS
62% 40% 0% 62%
0% 0% 0% 0%
0% 0% 0% 20%
46 (27, 65) 93 (62, 377) 15%a
2.5 (1.2, 3.8) 8.5 (4.6, 17.7) 15.7 (8.4, 39)
Single dose studies Tall.flick CD-I Swiss-.Webster
Dose-response studies Tail.flick CD-I
Swiss-Webster BALB/c
3.2 (2.3,4) 7,4 (5,6, 10,3) 2,4 (1,5, 3.4)
Hot#ate CD-I Swiss-Webster BALBIc
5,0 (3,5, 7,7) 10 (% 16,6) 2,7 (1,6, 3,8)
Cold water tail-flick CD-I Swiss-Webster BALB/c
7.0 (4.3, 14.1) 11.6 (6,9, 33) 5.0 (2.9,8.7)
* Analgesic response at the highest dose tested, 75 mg/kg.
64 (21, >100) 20%t 51 (27, >100) 72 (51, 152) 123 (77, 479) 42 (25, 61)
297 teresting to note that in the BALB/c strain, NalBzoH at the highest dose tested, 75 mg/kg, elicited analgesia in only 15% of the animals. Higher doses of NaiBzoH could not be evaluated due to solubility problems. Morphine retained the same rank-order in analgesic potencies among the 3 strains in the other two antinociceptive assays. Although it was less active in the cold water tail-flick assay, only the CD-1 strain achieved statistical significance. However, when we examined NalBzoH in these other assays, the results were quite different. NalBzoH was virtually inactive in BALB/c mice when examined in the radiant heat tail-flick assay. In contrast, the BALB/c strain was the most sensitive to NalBzoH analgesia in the hotplate and cold water tailflick assays. It is unlikely that these varying sensitivities reflect differences in the bioavailability of the drugs among the strains. First, these differences in sensitivity remained even when the drugs were administered directly into the central nervous system. CXBK mice remain far less sensitive to morphine administered intracerebroventricularly than CD-1 mice and U50,488H was still significamly less a,:tive in the Swiss-Webster strain than in the CD-1 following intrathecal injections (unpublished data). Second, tracer studies with morphine (5 mg/kg, s.c.) and NalBzoH (50 mg/kg, s,c.) indicated very similar concentrations of radioactivity for each ligand in the brains of groups of CD-1 mice and Swiss-Webster mice (n = 5) 30 rain after injection. The radioactivity represents both the administered drug and its metabolites. However, if the tracer refleooted only the administered drug, the levels of radioactivity for morphine would correspond to 1,45 ± 0,07/~g/8 brain and 1.28 ± 0.04 #g/g brain for CD-1 and SwissWebster mice, respectively, and 7.9 ± 0.54/~g/g brain and 7.2 ~tg/g brain, respectively for NalBzoH. Thus, the differing sensitivies of the strains cannot be explained by differences in the bioavailability of the drugs. To determine whether levels of receptors in the brain correlated with analgesic sensitivity, we performed binding studies with [3H]DAMGO (~), [3H]NalBzoH (Ka) and [3H]U69,593 (~1). First, we examined the binding of [3H]DAMGO and [3H]NalBzoH in CD-1 mice in saturation studies. The KD values for [3H]DAMGO and [3H]NalBzoH were quite similar (0.99 ± 0.2 and 0.7 ± 0.05 nM, respectively), but the density of [aH]NalBzoH sites (14 - 1.5 fmol/mg tissue) was approximately twice that for [3H]DAMGO (6.7 --- 1.8 fmoUmg tissue; P < 0.02). The KD value for either [3H]DAMGO or [3H]NalBzoH were comparable among the different strains. Since the ligands had similar affinities for the receptors among the strains, we examined the density of the binding sites using a single radioligand concentration (Table
TABLE II Levels of [3H]opioid binding in different strains of mice Binding was performed with the indicated radioligand (1 nM) as described in the text. Results are expressed as cpm and are the means -- S.E.M. of at least 5 independent replications, each of which was performed in triplicate. Differences in binding among strains for each radioligand was compared using analysis of variance followed by the Tukey-HSD test and significanceestablished at the P < 0.05 level. Within the [3H]DAMGO group, the level of binding in the BALBIc and the Swiss-Webster were significantly different and within the [all]U69,593 group, the SwissWebster mice had binding levels significantly higher than those in the C57/bgJ. Comparisons of binding levels between radioligands cannot be made. [JH]DAMGO [:JHlU69,593 [3HlNalBzoH CD-1 BALB/c Swiss-Webster C57/+ C57/bgJ
5621 - 234 7291 +- 517 4849 +--428 6144 - 77 5989 - 365
763 +-. 53 638 -.+ 23 813 +- 60 682 --- 19 607 - 44
7087 + 482 6111 -.+ 221 6352 - 101 7289 +- 421 6801 --. 172
II). The density of p sites did vary somewhat, with the binding in the Swiss-Webster significantly lower than that in the BALB/c strain. Slight differences also were noted with [3H]U69,593 binding, but only the SwissWebster and C57/bg J significantly differed from each other. Among the strains tested, the levels of [aH]NalBzoH binding did not differ significantly. The current study demonstrates the importance of the strain of mouse and the nociceptive assay when investigating r actions. Of the strains examined, the CD-1 appears to be the most consistently sensitive and should be considered when screening agents for antinociceptive activity in the radiant heat tail-flick assay. Although BALB/c mice are slightly more sensitive to morphine, they are not very sensitive to r drugs in this test. It is interesting that the sensitivity of the strains to/~ and u analgesics varies independently. Compared to CD-1 mice, Swiss-Webster mice show a similar decreased sensitivity towards all 3 drugs while the BALB/c strain is more sensitive to morphine and far less sensitive to both drugs. Among the strains examined, the sensitivity to the two different ~c drugs varied together. However, these systems have been clearly differentiated s'13. U50, 488H, but not NalBzoH, analgesia is readily reversed by nor-BNI and the two compounds show no cross tolerance. Furthermore, U50,488H (ul) analgesia is mediated spinally whereas NalBzoH (r3) analgesia is mediated supraspinally. These results support the existence of multiple opioid analgesic systems under independent genetic control. The choice of nociceptive assay also is important. The sensitivity of the strains to morphine was quite similar among the various assays tested, but not to NalBzoH. Within the CD-1 strain there was little difference in Nal-
298 BzoH sensitivity among the assays. This was not the case with the BALB/c strain. At the highest NalBzoH dose tested (75 mg/~g, s.c.), only 15% of mice were analgesic compared to an EDso in the CD-1 of only 44 mg/kg. In the hot plate assay the sensitivity of the two strains was equivalent and in the cold water tail-flick test the BALB/c strain was the most sensitive. Opposite results were observed with the Swiss-Webster strain, which was far less sensitive in the hot plate assay than the other two tests. The reversal in sensitivity to NalBzoH between the BALB/c and Swiss-Webster strains in the hot plate and cold water tail-flick assays was quite dramatic and illustrates further the variability among the strains. The lack of correlation between binding levels of the various receptors and analgesic sensitivities may have a number of reasons. Perhaps the most important is the low percentage of receptors within the brain involved with antinociception. Autoradiographic studies have documented very high levels of mu receptors in regions unrelated to pain perception and modulation 6.t~. Similarly, the distribution of ~c3 receptors in rodent brain suggests that only a small number are involved with pain modu-
1 Baran, A., Shuster, L., Elefterhiou, B.E. and Bailey, D.W., Opiate receptors in mice: genetic differences, Life Sci., 17 (1975) 633-640. 2 Clark, J.A., Houghten, R. and Pasternak, G.W., Opiate binding in calf th01amic membranes: a selective mut binding assay, Mol. Pharmocol., 34 (1988) 308=317. 3 Clark, J.A., Liu, L., Price, M., Harsh, B., Edelson, M. and Pasternak, G,W., Kappa reeeptol multiplicity: evidence for two U50,488H.sensittve Pet subtypes and a novel ~¢~subtype, J. Phar. macol. Exp. ?'her., 251 (1989)461-468. 4 D'Amour, F,E. and Smith, D.L., A method for determining loss of pain sensation, J, Pharmacol. Exp. Ther., 72 (1941) 7479. 5 Oistrak, M,A., Paul, D., Hahn, E.F. and Pasternak, G.W., Pharmacological actions of a novel mixed opiate agonist/antagonist: naloxone benzolhydrazone, J. Pharmacol. Exp. ?'her., 251 (1989) 469-476. 6 Goodman, R.R. and Pasternak, G.W., Visualization of muz opiate receptors in rat brain using a computerized autoradio. graphic subtraction in technique, Prec. Natl. Acad. Sci. U.$.A., 82 (1985) 6667-6671. 7 Ling, G.S.F., Simantov, R., Clark, J.A. and Pasternak, G.W., Naloxonazine actions in vivo, Eur. J. Pharmacol., 129 (1986) 33-38. 8 Luke, M.C., Hahn, E.F., Price, M. and Pasternak, G.W., Irreversible opiate agonists and antagonists. V. Novel hydrazone derivatives of naltrexone, Life Sci., 43 (1988) 1249-1256. 9 Mathiasen, JR., Raffa, R.B. ~nd Vaught, J.L., C57BL/6J-bgJ (beige) mice: differential sensitivity in the taiifliek test to centrally administered mu and delta opioid receptor agonists, Life Sci., 40 (1987) 1989-1994. 10 Maurer, R., Multiplicity of opiate receptors in different species, Neurosci, Lett., 30 (1982) 303-307, 11 Moskowitz, A.S. and Goodman, R.R., Autoradiographic analysis of rout, mu:, and delta opioid binding in the central nerves system of C57BL/6BY and CXBK (opioid-receptorodeficient)
lation (Paul and Pasternak, unpublished observations). Thus, important changes in the density of receptors involved with antinociception may not be seen due to the large number of sites involved with other actions. In conclusion, our results suggest the presence of multiple pain modulatory systems un~!er different genetic controls. Furthermore, not all pain stimuli are modulated in the same manner among the strains, as documented by the reversal in sensitivity of the Swiss-Webster and BALB/c strains to NalBzoH depending upon the analgesic assay employed. These observations may have important implications in our clinical understanding of pain and the wide range of sensitivities of patients to analgesic drugs.
We would like to thank Drs. Jerome B. Posner and Richard Hawkes for their assistance with these studies. This work was supported, in part, from grants from the National Institute on Drug Abuse to G.W.E (DA02615 and DA07241) and a core grant from the NCI to MSKCC (08748). G.W.P. is supported by a Research Career Development Award from NIDA (DA00138) and D.P. was a recipient of a fellowship from the Pharmaceutical Manufacturer's Association Foundation.
mice, Brain Research, 360 (1985) 108-116, 12 Pasternak, G.W., Wilson, H.A. and Snyder, S.H., Differential effects of protein-modifying reagents on receptor binding of opiate agonists and antagonists, Moi. Pharmacol.. 11 (1975) 340351,
13 Paul, D., Levison, J,A, Howard, D,H., Pick, C.G., Hahn, E,F. and Pasternak, G,W,, Naloxone benzoylhydrazone (NalBzoH) analgesia, J. Ptlarmacol, £xp. That., 255 (1990) 769-774, 14 Peats, LM. and Pomerantz, B., CXBK mice deficient in opiate receptors show poor electropuncture analgesia, Nature, 273 (t978) 675-676. 15 Pizzlketti, RJ,, Pressman, N.S., Geller, E.B., Cowan, A. and Adler, M.W., The rat cold water tail flick: a novel analgesic test that distinguishes opioid agonists from mixed agonist-antagonists, Fur. J. Pharmacol., 119 (1985) 23-29. 16 Price, M., Gistrak, M.A., Itzhak, Y., Hahn, E.F. and Pasternak, G.W., Receptor binding of 3H-naloxone bcnzoylhydrazone: a reversible ~c and slowly dissociable me opiate, Mol, Pharmacol., 35 (1989) 67-74. 17 Reith, M.E.A., Sershen, H., Vadasz, Lajtha, A,, Strain differences in opiate receptors in mouse brain, Eur. J. Pharmacol., 74 (1981) 377-380. 18 Shuster, L., Webster, G.W., Yu, G. and Elefterhiou, B.E., Genetic analysis of the response to morphine in mice: analgesia and running, Psychopharmacologia, 42 (1975) 249-254. 19 Umans, J.G. and tnturrisi, C.E., Pharmamdynamies of subcutaneously administered diacetylmorphine, 6-ace'tylmorphine and morphine in mice, J. Pharmacol. Exp. Ther., 218 (1981) 409415. 20 Vaught, J.L., Mathiasen, J.R. and Raffa, R.B., Examination of the involvement of supraspinal and spinal mu and delta opioid receptors in analgesia using the mu receptor deficient CXBK mouse, J. Pharmacol. Exp. Ther., 245 (1988) 13-16. 21 Woolfe, G. and MacDonald, A.D., The evaluation of the analgesic action of pethidine hydrochloride (Demerol), J. Phar. macol. Exp. That., 80 (1944) 300-307.
BRES 24941
295
Short Communications
Genetic influences in opioid analgesic sensitivity in mice C h a i m G . P i c k 1, Jie C h a n g 1, D e n n i s P a u l I a n d G a v r i l W. P a s t e r n a k 1'2 aThe Cotr.ias Laboratory of Neuro.Oncology, Memorial SIoan-Kettering Cancer Center, New York, NY 10021 (U.S.A.) and 2Departments of Neurology and Pharmacology, Cornell University Medical College, New York, NY 10021 (U.S.A.)
(Accepted 27 August 1991) Key words: Morphine; ~ Receptor; ~ Receptor; Analgesia
Studies of various strains of mice revealed marked differences in their analgesic sensitivity towards morphine (~), U50,488H (~:1) and naloxone benzoylhydrazone(NalBzoH; g3). Sensitivityto F and g analgesiavaried independentlyof the other. Analgesicsensitivityto morphine remained relativelyconsistent among 3 different nociceptive assays for each strain. However, the sensitivityof an individualstrain to NalBzoH remained highly dependent upon the assay used. CD-1 mice were sensitive to NalBzoH in all 3 assays, but in BALB/cmice HalBzoH produced analgesia only in the hot plate and cold water tail-flick assays, In Swiss-Webstermice, NalBzoH was active in the radiant heat and cold water tail-flicksbut inactive in the hot plate. Althoughthe levelsof ~, gl and ~3 binding in whole brain homogenatesdid vary somewhat, they did not correlate with analgesic sensitivity. These results suggests that the genetic controls over ~ and ~ analgesia operate independently and further illustrate the many difficulties in evaluatingpotential analgesics. The ability of morphine to relieve pain is dependent upon both the assay and the species. Even within a single species, the potency of morphine can vary enormously t,7,9-tt'u'lT'ls'2°. A number of opioid receptor subtypes have been identified which can influence the perception of pain. Recently studies with naloxone benzoylhydrazone (NalBzoH) identified another subtype, ~3, which could elicit analgesia through a unique supraspinal opioid mechanism3's't3't6. Although NalBzoH analgesia was reversed by the traditional opioid antagonists naloxone and WIN44,441, more selective opioid antagonists, including ~.funaltrexamine (,u), naltrindole (6) and norbinaltorphimine (gl) did not reverse NalBzoH analgesia, Furthermore, NalBzoH analgesia did not demonstrate cross-tolerance with F, 6 or ~t drugs. In the present study we wished to examine potential genetic influences in the sensitivity of strains of mice to NaiBzoH and whether sensitivity to 1¢3ligands covaried with sensitivity to/~ ligands. Male CD-1, CFW(SW)BR (Swiss-Webster), and BALB/cAnNCrlBR (BALB/c) mice (25-35 g) were purchased from Charles River Breeding Laboratories (Wilmington, MA). C57/J6-bgJ/bg J (C57/bgJ), C57BL/J6bgJ/+ (C57/+) and CXBK/Byj (CXBK) mice (20-30 g) were obtained from Jackson Laboratory (Bar Harbor, MA) and HS/Ibg (HS) (25-30 g) from the Institute of Behavioral Genetics (Boulder, CO). All mice were housed in groups of 5 and were maintained on a 12 h light:12 h
dark cycle with Purina rodent chow and water available ad libitum. Morphine sulfate and U50,488-H {trans-3,4dchloro-N-methyl-N-[2-(1-pyrrolindinyl)'cyclohexyl]'ben" zeneacetamide} were obtained from the Research Technology Branch of NIDA. NaiBzoH and [3H]NalBzoH were synthesized as described previouslys't6. Morphine doses were expressed as the salt and NalBzoH and US0, 488H as the free base. NalBzoH was dissolved in 30% ethanol, which did not produce analgesia in control mice. All drugs were administered subcutaneously. All other drugs were dissolved in saline. Radioligands and Formula 963 Scintillation Fluor were purchased from New England Nuclear (Boston, MA). Analgesia was determined quantally as a doubling or greater of individual baseline latencies using the radiant heat tail-flick technique 4'5'7't3, cold water tail-flick t~ or the hot plate assay 13'2t, Baseline latencies in all assays were defined as the mean of two determinations. Maximal latencies were 10 s in the radiant tail-flick assay and 60 s in the cold water tail-flick and hot plate assays. For the cold water tail-flick assay, the tails were submerged approximately half-way into a 0 °C water bath and the latency to remove, or flick, the tail measured. The hot plate assay was performed at 50 °C using a commercially available apparatus (IITC, Inc, Woodland hills, CA). Trials were terminated by either jumping or paw licking. Dose-response curves were analyzed using a modification of the BLISS-20 computer program. This program
Correspondence: G.W. Pasternak, Department of Neurology, MemorialSloan-KetteringCancer Center, 1275York Avenue, New York, NY
10021, U.S.A. Fax: (1) (212) 717-3296.
296 maximizes the log-likelihood function to fit a parallel set of Gaussian normal sigmoid curves to the dose-response data t9. Single-dose antagonist studies were analyzed using Student's t-tests (P < 0.05). Binding assays were performed in mouse homogenares using the procedures previously reported 2'3. Brains were homogenized in 50 vols. of Tris (50 mM; pH 7.4 at 25 °C) buffer containing PMSF (0.1 mM), Na2EDTA (1 raM) and NaCI (100 mM), incubated at 25 °C for 15 min to remove endogenous ligands t2, and centrifuged at 45, 000 g at 4 °C for 30 rain. The pellets were resuspended in 100 vols. of potassium phosphate buffer (50 raM; pH 7.4; I0 mg wet weight tissue/ml) and 1 or 2 ml aliquots used in all binding assays. All binding assays were preformed at 25 °C. g3 Binding was determined using [3H]NalBzoH for 60 min in the presence of K2EDTA (5 mM); [3H]DAMGO (,u) binding for 150 rain in the presence of MgSO4 (1 mM); and [3H]U69,593 (gt) binding for 45 min. When using [3H]U69,593 all filters were treated with polyethylenimine (1%) to decrease filter binding. Non-specific binding was determined with levallorphan (1 pM). All points were determined in triplicate and only specific binding is reported. All values are pre-
sented as the means -+ S.E.M. of the stated number of independent determinations. Administration of single doses of prototypical/~ (morphine), ut (U50,488H) and K3 (NaiBzoH) drugs produced an interesting analgesic profile among the various strains (Table I). With the exception of the CXBK strain, which is known to be very insensitive to morphine Lm'2°, all the strains were sensitive to morphine (5 mg/kg, s.c.), although their responses did vary considerably. We also observed markedly variable responses among the strains for U50,488H and NalBzoH. The CD-1 strain was the most sensitive across the various drugs. Although the BALB/c strain may be more sensitive to morphine than the CD-1 strain, the B A L B / c mice were relatively insensitive to both ~ agents. The Swiss-Webster strain on the other hand, showed similar decreases in analgesic sensitivity for all 3 classes of drugs. We next examined the tail-flick responses of the CD-1, Swiss-Webster and BALB/c strains in greater detail with dose-response curves (Table I). The Swiss-Webster strain was significantly less sensitive to morphine than the other two strains, but was more sensitive to NalBzoH and U50,488H than the BALB/c strain. It was in-
TABLE I Analgesic sensMviry of various strains of mice in the single dose studies, groups of mice (n ;~ 10) were tested with a single dose of morphine (5 mg/kg, s.c.), NalBzoH (50 mg/kg, s.c.)
or US0,488H (5 mg/kg, s.c.) in the radiant heat tail.flick assay. In the dose-response studies, EDso values were determined frome doseresponse curves composed of at least 3 different drug concentrations with at least 10 mice/dose. R~sults are the EDs0 values with 95% confidence limits. Th~ highest dose of NalBzoH which could be tested was 75 mg/kg, s.c., due to solubility problems, Morphine
NalB~oH
USO,488H
76% 40%
54% 29%
70% 30%
BALB/c
90%
14%
10%
C57/bgJ C57/+ CXBK HS
62% 40% 0% 62%
0% 0% 0% 0%
0% 0% 0% 20%
46 (27, 65) 93 (62, 377) 15%a
2.5 (1.2, 3.8) 8.5 (4.6, 17.7) 15.7 (8.4, 39)
Single dose studies Tall.flick CD-I Swiss-.Webster
Dose-response studies Tail.flick CD-I
Swiss-Webster BALB/c
3.2 (2.3,4) 7,4 (5,6, 10,3) 2,4 (1,5, 3.4)
Hot#ate CD-I Swiss-Webster BALBIc
5,0 (3,5, 7,7) 10 (% 16,6) 2,7 (1,6, 3,8)
Cold water tail-flick CD-I Swiss-Webster BALB/c
7.0 (4.3, 14.1) 11.6 (6,9, 33) 5.0 (2.9,8.7)
* Analgesic response at the highest dose tested, 75 mg/kg.
64 (21, >100) 20%t 51 (27, >100) 72 (51, 152) 123 (77, 479) 42 (25, 61)
297 teresting to note that in the BALB/c strain, NalBzoH at the highest dose tested, 75 mg/kg, elicited analgesia in only 15% of the animals. Higher doses of NaiBzoH could not be evaluated due to solubility problems. Morphine retained the same rank-order in analgesic potencies among the 3 strains in the other two antinociceptive assays. Although it was less active in the cold water tail-flick assay, only the CD-1 strain achieved statistical significance. However, when we examined NalBzoH in these other assays, the results were quite different. NalBzoH was virtually inactive in BALB/c mice when examined in the radiant heat tail-flick assay. In contrast, the BALB/c strain was the most sensitive to NalBzoH analgesia in the hotplate and cold water tailflick assays. It is unlikely that these varying sensitivities reflect differences in the bioavailability of the drugs among the strains. First, these differences in sensitivity remained even when the drugs were administered directly into the central nervous system. CXBK mice remain far less sensitive to morphine administered intracerebroventricularly than CD-1 mice and U50,488H was still significamly less a,:tive in the Swiss-Webster strain than in the CD-1 following intrathecal injections (unpublished data). Second, tracer studies with morphine (5 mg/kg, s.c.) and NalBzoH (50 mg/kg, s,c.) indicated very similar concentrations of radioactivity for each ligand in the brains of groups of CD-1 mice and Swiss-Webster mice (n = 5) 30 rain after injection. The radioactivity represents both the administered drug and its metabolites. However, if the tracer refleooted only the administered drug, the levels of radioactivity for morphine would correspond to 1,45 ± 0,07/~g/8 brain and 1.28 ± 0.04 #g/g brain for CD-1 and SwissWebster mice, respectively, and 7.9 ± 0.54/~g/g brain and 7.2 ~tg/g brain, respectively for NalBzoH. Thus, the differing sensitivies of the strains cannot be explained by differences in the bioavailability of the drugs. To determine whether levels of receptors in the brain correlated with analgesic sensitivity, we performed binding studies with [3H]DAMGO (~), [3H]NalBzoH (Ka) and [3H]U69,593 (~1). First, we examined the binding of [3H]DAMGO and [3H]NalBzoH in CD-1 mice in saturation studies. The KD values for [3H]DAMGO and [3H]NalBzoH were quite similar (0.99 ± 0.2 and 0.7 ± 0.05 nM, respectively), but the density of [aH]NalBzoH sites (14 - 1.5 fmol/mg tissue) was approximately twice that for [3H]DAMGO (6.7 --- 1.8 fmoUmg tissue; P < 0.02). The KD value for either [3H]DAMGO or [3H]NalBzoH were comparable among the different strains. Since the ligands had similar affinities for the receptors among the strains, we examined the density of the binding sites using a single radioligand concentration (Table
TABLE II Levels of [3H]opioid binding in different strains of mice Binding was performed with the indicated radioligand (1 nM) as described in the text. Results are expressed as cpm and are the means -- S.E.M. of at least 5 independent replications, each of which was performed in triplicate. Differences in binding among strains for each radioligand was compared using analysis of variance followed by the Tukey-HSD test and significanceestablished at the P < 0.05 level. Within the [3H]DAMGO group, the level of binding in the BALBIc and the Swiss-Webster were significantly different and within the [all]U69,593 group, the SwissWebster mice had binding levels significantly higher than those in the C57/bgJ. Comparisons of binding levels between radioligands cannot be made. [JH]DAMGO [:JHlU69,593 [3HlNalBzoH CD-1 BALB/c Swiss-Webster C57/+ C57/bgJ
5621 - 234 7291 +- 517 4849 +--428 6144 - 77 5989 - 365
763 +-. 53 638 -.+ 23 813 +- 60 682 --- 19 607 - 44
7087 + 482 6111 -.+ 221 6352 - 101 7289 +- 421 6801 --. 172
II). The density of p sites did vary somewhat, with the binding in the Swiss-Webster significantly lower than that in the BALB/c strain. Slight differences also were noted with [3H]U69,593 binding, but only the SwissWebster and C57/bg J significantly differed from each other. Among the strains tested, the levels of [aH]NalBzoH binding did not differ significantly. The current study demonstrates the importance of the strain of mouse and the nociceptive assay when investigating r actions. Of the strains examined, the CD-1 appears to be the most consistently sensitive and should be considered when screening agents for antinociceptive activity in the radiant heat tail-flick assay. Although BALB/c mice are slightly more sensitive to morphine, they are not very sensitive to r drugs in this test. It is interesting that the sensitivity of the strains to/~ and u analgesics varies independently. Compared to CD-1 mice, Swiss-Webster mice show a similar decreased sensitivity towards all 3 drugs while the BALB/c strain is more sensitive to morphine and far less sensitive to both drugs. Among the strains examined, the sensitivity to the two different ~c drugs varied together. However, these systems have been clearly differentiated s'13. U50, 488H, but not NalBzoH, analgesia is readily reversed by nor-BNI and the two compounds show no cross tolerance. Furthermore, U50,488H (ul) analgesia is mediated spinally whereas NalBzoH (r3) analgesia is mediated supraspinally. These results support the existence of multiple opioid analgesic systems under independent genetic control. The choice of nociceptive assay also is important. The sensitivity of the strains to morphine was quite similar among the various assays tested, but not to NalBzoH. Within the CD-1 strain there was little difference in Nal-
298 BzoH sensitivity among the assays. This was not the case with the BALB/c strain. At the highest NalBzoH dose tested (75 mg/~g, s.c.), only 15% of mice were analgesic compared to an EDso in the CD-1 of only 44 mg/kg. In the hot plate assay the sensitivity of the two strains was equivalent and in the cold water tail-flick test the BALB/c strain was the most sensitive. Opposite results were observed with the Swiss-Webster strain, which was far less sensitive in the hot plate assay than the other two tests. The reversal in sensitivity to NalBzoH between the BALB/c and Swiss-Webster strains in the hot plate and cold water tail-flick assays was quite dramatic and illustrates further the variability among the strains. The lack of correlation between binding levels of the various receptors and analgesic sensitivities may have a number of reasons. Perhaps the most important is the low percentage of receptors within the brain involved with antinociception. Autoradiographic studies have documented very high levels of mu receptors in regions unrelated to pain perception and modulation 6.t~. Similarly, the distribution of ~c3 receptors in rodent brain suggests that only a small number are involved with pain modu-
1 Baran, A., Shuster, L., Elefterhiou, B.E. and Bailey, D.W., Opiate receptors in mice: genetic differences, Life Sci., 17 (1975) 633-640. 2 Clark, J.A., Houghten, R. and Pasternak, G.W., Opiate binding in calf th01amic membranes: a selective mut binding assay, Mol. Pharmocol., 34 (1988) 308=317. 3 Clark, J.A., Liu, L., Price, M., Harsh, B., Edelson, M. and Pasternak, G,W., Kappa reeeptol multiplicity: evidence for two U50,488H.sensittve Pet subtypes and a novel ~¢~subtype, J. Phar. macol. Exp. ?'her., 251 (1989)461-468. 4 D'Amour, F,E. and Smith, D.L., A method for determining loss of pain sensation, J, Pharmacol. Exp. Ther., 72 (1941) 7479. 5 Oistrak, M,A., Paul, D., Hahn, E.F. and Pasternak, G.W., Pharmacological actions of a novel mixed opiate agonist/antagonist: naloxone benzolhydrazone, J. Pharmacol. Exp. ?'her., 251 (1989) 469-476. 6 Goodman, R.R. and Pasternak, G.W., Visualization of muz opiate receptors in rat brain using a computerized autoradio. graphic subtraction in technique, Prec. Natl. Acad. Sci. U.$.A., 82 (1985) 6667-6671. 7 Ling, G.S.F., Simantov, R., Clark, J.A. and Pasternak, G.W., Naloxonazine actions in vivo, Eur. J. Pharmacol., 129 (1986) 33-38. 8 Luke, M.C., Hahn, E.F., Price, M. and Pasternak, G.W., Irreversible opiate agonists and antagonists. V. Novel hydrazone derivatives of naltrexone, Life Sci., 43 (1988) 1249-1256. 9 Mathiasen, JR., Raffa, R.B. ~nd Vaught, J.L., C57BL/6J-bgJ (beige) mice: differential sensitivity in the taiifliek test to centrally administered mu and delta opioid receptor agonists, Life Sci., 40 (1987) 1989-1994. 10 Maurer, R., Multiplicity of opiate receptors in different species, Neurosci, Lett., 30 (1982) 303-307, 11 Moskowitz, A.S. and Goodman, R.R., Autoradiographic analysis of rout, mu:, and delta opioid binding in the central nerves system of C57BL/6BY and CXBK (opioid-receptorodeficient)
lation (Paul and Pasternak, unpublished observations). Thus, important changes in the density of receptors involved with antinociception may not be seen due to the large number of sites involved with other actions. In conclusion, our results suggest the presence of multiple pain modulatory systems un~!er different genetic controls. Furthermore, not all pain stimuli are modulated in the same manner among the strains, as documented by the reversal in sensitivity of the Swiss-Webster and BALB/c strains to NalBzoH depending upon the analgesic assay employed. These observations may have important implications in our clinical understanding of pain and the wide range of sensitivities of patients to analgesic drugs.
We would like to thank Drs. Jerome B. Posner and Richard Hawkes for their assistance with these studies. This work was supported, in part, from grants from the National Institute on Drug Abuse to G.W.E (DA02615 and DA07241) and a core grant from the NCI to MSKCC (08748). G.W.P. is supported by a Research Career Development Award from NIDA (DA00138) and D.P. was a recipient of a fellowship from the Pharmaceutical Manufacturer's Association Foundation.
mice, Brain Research, 360 (1985) 108-116, 12 Pasternak, G.W., Wilson, H.A. and Snyder, S.H., Differential effects of protein-modifying reagents on receptor binding of opiate agonists and antagonists, Moi. Pharmacol.. 11 (1975) 340351,
13 Paul, D., Levison, J,A, Howard, D,H., Pick, C.G., Hahn, E,F. and Pasternak, G,W,, Naloxone benzoylhydrazone (NalBzoH) analgesia, J. Ptlarmacol, £xp. That., 255 (1990) 769-774, 14 Peats, LM. and Pomerantz, B., CXBK mice deficient in opiate receptors show poor electropuncture analgesia, Nature, 273 (t978) 675-676. 15 Pizzlketti, RJ,, Pressman, N.S., Geller, E.B., Cowan, A. and Adler, M.W., The rat cold water tail flick: a novel analgesic test that distinguishes opioid agonists from mixed agonist-antagonists, Fur. J. Pharmacol., 119 (1985) 23-29. 16 Price, M., Gistrak, M.A., Itzhak, Y., Hahn, E.F. and Pasternak, G.W., Receptor binding of 3H-naloxone bcnzoylhydrazone: a reversible ~c and slowly dissociable me opiate, Mol, Pharmacol., 35 (1989) 67-74. 17 Reith, M.E.A., Sershen, H., Vadasz, Lajtha, A,, Strain differences in opiate receptors in mouse brain, Eur. J. Pharmacol., 74 (1981) 377-380. 18 Shuster, L., Webster, G.W., Yu, G. and Elefterhiou, B.E., Genetic analysis of the response to morphine in mice: analgesia and running, Psychopharmacologia, 42 (1975) 249-254. 19 Umans, J.G. and tnturrisi, C.E., Pharmamdynamies of subcutaneously administered diacetylmorphine, 6-ace'tylmorphine and morphine in mice, J. Pharmacol. Exp. Ther., 218 (1981) 409415. 20 Vaught, J.L., Mathiasen, J.R. and Raffa, R.B., Examination of the involvement of supraspinal and spinal mu and delta opioid receptors in analgesia using the mu receptor deficient CXBK mouse, J. Pharmacol. Exp. Ther., 245 (1988) 13-16. 21 Woolfe, G. and MacDonald, A.D., The evaluation of the analgesic action of pethidine hydrochloride (Demerol), J. Phar. macol. Exp. That., 80 (1944) 300-307.