Agonist and antagonist activities of ligands derived from naltrexone and oxymorphone

Life Sciences, Vol. 50, pp. 1491-1495 Printed in the USA

Pergamon Press

AGONIST AND ANTAGONIST ACTIVITIES OF LIGANDS DERIVED FROM NALTREXONE AND OXYMORPHONE 1 A. E. Takemori*, M. Sultana**, H. Nagase** and P. S. Portoghese** *Department of Pharmacology, Medical School and **Department of Medicinal Chemistry, College of Pharmacy, University of Minnesota, Minneapolis, MN 55455 (Received

in final form March 3, 1992)

Summary The pharmacological profile of naltrindole (NTI) and three of its analogues, N-methyl-NTI (N-Me-NTI), oxymorphindole (OMI) and naltriben (NTB) were studied in antinociceptive assays. The compounds were found to have agonist activities that appear to be mediated mainly by r opioid receptors because norbinaltorphimine (nor-BNI), the selective r opioid receptor antagonist inhibited their effects significantly. All of the compounds, behaved as antagonists at doses that were lower than those that produced agonist effects and they possessed a profile that was very selective for inhibiting the antinociceptive activities of 6 opioid receptor agonists. Differential antagonism by NTB of the activities of DSLET and DPDPE was demonstrated. It is essential to have highly selective opioid receptor antagonists as pharmacologic tools to sort out the pharmacologic effects that are mediated by different types of opioid receptors, e.g. Ix, K and 8. Recently, we reported on a series of selective, non-peptide ~ opioid receptor antagonists (1- 4). These compounds are illustrated in fig. 1 and shows the relationship of NTI, N-Me-NTI and NTB to naltrexone and that of OMI to oxymorphone. N-CH3

H

OXYMORPHONE

NALTREXONE

N-CH3

H

OXYMORPHINDOLE (OMI)

X NALTRINDOLE (NTI)

NH

N-METHy L-NTI

N.Me

NALTRIBEN (NTB)

O

Figure 1. Chemical structures of NTI analogues. Studies in this report were carried out in accordance with the Declaration of Helsinld and/or with the guide for the care and use of laboratory animals as adopted and promulgated by the National Institutes of Health. 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd All rights reserved.

1492

Naltrindole Analogues

Vol. 50, No. 20, 1992

Among these compounds, NTI is the most potent and selective in smooth muscle preparations. In the mouse deferens preparation, NTI (Ke = 0.13 nM) is more potent than N-Me-NTI (Ke = 1.0 nM) which is equipotent with NTB. OMI is a partial agonist that had little or no antagonist activity at concentrations that displayed no agonism. In receptor binding assays, NTB exhibits the highest affinity (Ki=0.013 nM) for 8 opioid receptors and is about 1,500 and 12,000 times more selective for 8 than for Ix and ~: opioid receptors, respectively. However, the other compounds also are potent and selective for 8 opioid receptors with Ki values in the sub-nM range. We have now examined these compounds for their agonist and antagonist activities using antinociceptive assays in mice. Materials and Methods Animals. Random bred male Swiss-Webster mice weighing 20 to 25 g (BioLab, White Bear Lake, MN) were used in assessing the antinociceptive activity of the compounds. All animals were supplied food and water ad libitum and were housed in a temperature controlled (23"C) animal room for at least a day before using them for experiments. Each animal was used only once. Antinociceptive Assay. The abdominal stretching (writhing) assay described previously was used (5). Statistics. The parallel line assay of Finney (6) was used to estimate the EDs0 values and 95% confidence limits with the aid of a computer. Drugs. [D-Ala2, MePhe4, Gly-ol5]enkephalin (DAMGO), [D-Pen 2, D-PenS]enkephalin (DPDPE), [D-Ala2,D-Leu5]enkephalin (DADLE) and [D-Ser2,Leu5]enkephalin-Thr (DSLET) were procured from Sigma Chemical Co. (St Louis, MO). Morphine sulfate was bought from Mallinckrodt,Inc.(Paris, KY). Trans(:t:)-3,4-dichloro-N-methyI-N-[2-(1-pyrrolidinyl)cyclohexyl]benzeneacetamide (U-50,488H) and ethylketazocine (EK) were gifts from Dr. Philip VonVoigtlander (The Upjohn Co., Kalamazoo, MI) and Dr. Alfred Farah (Sterling-Winthrop Research Institute, Rensselaer, NY), respectively. 15-Funaltrexamine ([3-FNA) and norbinaltorphimine (norBNI) were synthesized as described previously (7, 8). Results The compounds were initially examined with the use of the abdominal stretching (writhing) assay to see whether or not they possessed antinociceptive activity. The antinociceptive property of these compounds could not be detected by the tail flick assay at the doses used in the present agonist or antagonist studies. All compounds possessed antinociceptive activity with moderate potencies when administered i.c.v. (Table 1). When compared to the activities of other 8 opioid receptor agonists such as DSLET or DADLE, N-Me-NTI was about half as potent, OMI was about 1/4th as potent, NTI was was about 1/Sth as potent and NTB was only about 1/12th as potent. NTB, which was designed to distribute to the brain more readily than NTI displayed a relatively potent antinociceptive activity when administered s.c. Although the antagonist studies described below suggested that the other compounds distributed to the central nervous system, the estimation of their s.c. antinociceptive activity was not pursued due to limited amounts of the compounds. Selective opioid receptor antagonists (9, 10), 15-FNA (IX), norBNI (r), NTI (81 & 82)) and NTB (82) were employed to determine the opioid receptor type at which these compounds were interacting to oroduce their antinociceptive effect. The doses of antagonists used were those that had been previously shown to antagonize substantially the antinociceptive activity of the prototypic opioid receptor agonists, morphine and (DAMGO) (Ix), U-50,488H (r), (DPDPE) (81) and DSLET (82). The agonists chosen for further characterization were the representative naltrexone derivative, NTI and the oxymorphone derivative, OMI. Inspection of the data in Table 2 reveals that only norBNI inhibited significantly the antinociceptive activity of both OMI and NTI which suggested that these compounds were mediatingtheir antinociceptive effects through K opioid receptors.

Vol. 50, No. 20, 1992

NaltrindoleAnalogues

1493

TABLE 1 Antinociceptive Activities of NTI Derivatives COMPOUND

EDs0 (95% Confidence Limits) pmol/mouse, i.c.v, a /~mol/kg, s.c.

NTI

39 (5 - 21)

N-Me-NTI

10 (9 - 11)

OMI

19 (11 - 33)

NTB

62 (40 - 120)

DSLET

4 (3 - 7)

DADLE

6 (5 - 7)

3.4 (3.0 - 3.8)

a The abdominal stretching assay was used. The peak time for antinociceptive activity of all compounds was 10 min.

TABLE 2 Effect of Selective Opioid Receptor Antagonists on the Antinociceptive activity of NTI and OMI EDNI RATIO (EDs0 + Antagonist/Control EDs0) Antagonist Pretreatment

OMI

I~-FNA, 20/~mol/kg, 24 hr

1.8 (0.5 - 5.7)

norBNI, 12/~mol/kg, s.c., 2hr

4.8 (2.7 - 16.4)*

NTB, 1.2/~mol/kg, s.c., 30 min NTI, 44/~mol/kg, s.c., 30 min

NTI 3.8 (1.0 - 9.5) 15.2 (4.7 - 44.3)* 0.9 (0.4 -2.2)

2.6 (0.7- 8.8)

2.7 (1.0 - 6.5)

* Values are significantly greater than 1.0. Next, the compounds were examined for their antagonist activity at doses lower than those that exhibited antinociceptive activity. The antagonist doses used were determined to be the highest possible dose at which antagonism could be studied without interference from the agonist component of the compounds. The compounds were administered s.c. and their ability to antagonize the antinociceptive activity of morphine, U-50,488H, DPDPE and DSLET was studied (Table 3). The analogues proved to be highly selective for inhibiting the antinociceptive activity of the 6 agonist, DSLET, without altering the potencies of morphine or U-50,488H. Thus, presumably these compounds did gain access to the central nervous system. When one examines the s.c. data and calculates the doses to produce equivalent antagonism, i.e. doses required to double the EDs0 of DSLET, NTB was the most potent antagonist and was about 26 and 39 times more potent than NTI and N-Me-NTI, respectively. Thus it appeared that NTB distributed to the central nervous system more readily than NTI or N-ME-NTI. NTB, not only exhibited selectivity for antagonizing the action of DSLE'T but it did not alter the action of DPDPE. This selectivity for

1494

Naltrindole Analogues

Vol. 50, No. 20, 1992

one 8 opioid receptor agonist but not for the other corroborates previous reports using the tail flick assay in mice (11, 12). When administered i.c.v., these compounds displayed the same 8 selective antagonist profile that was observed after s.c. administration (Table 3). Unexpectedly, OMI, the oxymorphnone derivative, displayed the most potent, selective antagonist activity at 8 opioid receptors by the i.c.v. route of administration. Antagonism by OMI when administered by the s.c. route could not be tested due to limited amounts of this compound. TABLE 3 The Antagonist Activities of NTI Analogues EDs0 RATIO (95% Confidence Limits) a Antagonist Pretreatment

Morphine

U-50,488H

DSLET

DPDPE

NTI, 20 min, 10 pmol i.c.v.

1.0 (0.5 - 1.9)

0.9 (0.6 - 1.5)

5.3 (2.6 - 20.0)*

NTI, 30 min, 44 Fmol/kg s.c.

1.2 (0.6 - 2.8)

1.2 (0.6 - 2.9)

5.3 (2.7 - 11.1)*

N-Me-NTI, 10 min, 6 pmol i.c.v.

1.1 (0.6 - 2.3)

0.9 (0.7 - 1.4)

4.2 (2.3 - 8.3)*

N-Me-NTI, 30 min, 40 Fmol/kg s.c.

1.2 (0.3 - 4.7)

1.3 (0.7 - 2.6)

3.6 (1.9 - 7.1)*

NTB, 10 min, 25 pmol i.c.v.

0.9 (0.4 - 1.7)

0.6 (0.4 - 1.7)

3.1 (1.8 - 5.6)*

0.6 (0.3 - 1.0)

NTB, 30 min, 1.2 (0.4 - 4.5) 1.2 Fmol/kg s.c.

1.2 (0.8 - 1.9)

4.3 (2.0 - 7.7)*

1.0 (0.7 - 1.5)

OMI, 10 min, 1.25 pmol i.c.v.

0.7 (0.5 - 1.1)

7.7 (4.2 - 14.3)*

0.8 (0.4 - 2.5)

2.0 (1.1 - 3.6)*

a EDs0 ratio is defined in Table 2. The peak times for the agonists were 30 rain for morphine, 20 min for DPDPE, 15 min for U-50,488H and 10 rain for DSLEF. * Values are significantly greater than 1.0. Discussion All the analogues of naltrexone and oxymorphine studied in this report p o s s e s s e d antinociceptive activity only when the abdominal stretching assay, and not the tail flick assay, was used. The antinociceptive activity of these compounds was mediated probably through 1
Vol. 50, No. 20, 1992

Naltrindole Analogues

1495

possesses a methyl group at the 17-N position, whereas antagonists generally have an allyl, cyclopropylmethyl or a related moiety at this position. As these compounds exhibit high 8 selectivity ratios in both smooth muscle and receptor binding assays, it is surprising that they exhibit fairly potent r-mediated agonism in vivo. This apparent discrepancy may be explained by either an indirect effect through the release of dynorphins perhaps in the spinal cord or a direct effect on a subtype of K opioid receptors in the brain that is not discernible in smooth muscle or receptor binding assays. Acknowledgments This investigation was supported by U. S. Public Service Grants from the National Institute on Drug Abuse. We thank the technical assistance of Mary Lunzer. References 1. P. S. PORTOGHESE, M. SULTANA, H. NAGASE and A. E. TAKEMORI, J. Med. Chem. 3_!.1281-282 (1988). 2. P. S. PORTOGHESE, M. SULTANA and A. E. TAKEMORI, Eur. J. Pharmacol. 146185186 (1988). 3. P. S. PORTOGHESE, M. SULTANA and A. E. TAKEMORI, J. Med. Chem. 3._331714-172( (1990). 4. P. S. P O R T O G H E S E , H. N A G A S E , K. E. M A L O N E Y H U S S , C.-E. LIN and A. E. T A K E M O R I , J. Med. Chem. 3..441715-1720 (1991). 5. G. H A Y A S H I and A. E. T A K E M O R I , Eur. J. Pharmacol. I_6663-66 (1971). 6. D.J. FINNEY, StatisticalMethods in BiologicalAssay, 2nd ed.,Hafner PublishingCo., New York, 1964. 7. P. S. P O R T O G H E S E , D. L. L A R S O N , L. M. SAYRE, D. S. FRIES and A. E. T A K E M O R I , J. Mcd. Chem. 23 (1980). 8. P. S. P O R T O G H E S E , A. W. L I P K O W S K I and A. E. T A K E M O R I , J. Mcd. Chem. 30 238239 (1987). 9. A. E. TAKEMORI and P. S. PORTOGHESE, Ann. Rev. Pharmacol. Toxicol. 25 193-223 (1985). 10. A. E. TAKEMORI and P. S. PORTOGHESE, Ann. Rev. Pharmacol. Toxicol. 32 239-269 (1992). 11. M. SOFUOGLU, P. S. PORTOGHESE and A. E. TAKEMORI, J. Pharmacol. Exp. Ther. 257 676-680 (1991). 12. Q. JIANG, A. E. TAKEMORI, M. SULTANA, P. S. PORTOGHESE, W. D. BOWEN, H. I MOSBERG and F. PORRECA, J. Pharmacol. Exp. Ther. 257 1069-1075 (1991).