Life Sciences, Vol. 48, pp. 2165-2171 Printed in the U.S.A.

Pergamon Press

MU RECEPTOR BINDING OF SOME COMMONLY USED OPIOIDS AND THEIR METABOLITES Zhao Rong Chen 1, Rodney J. Irvine 1, Andrew A. Somogyi 1'2 and Felix Bochner ~,2 Department of Clinical and Expenmental Pharmacology 1, University of Adelaide and Department of Chnical Pharmacology2, Royal Adelmde Hospital, North Terrace, Adelaide, Australia 5000. (Received in final form March 26, 1991)

Summarv The binding affinity to the t~ receptor of some opioids chemically related to morphine and some of their metabol~tes was examined in rat brain homogenates with 3H-DAMGO. The chemical group at position 6 of the molecule had httle effect on binding (e.g. morphine-6-glucuronide K, =0.6 nM; morphine = 1.2 nM), Decreasing the length of the alkyl group at pos=tlon 3 decreased "the K~ values (morphine < codeine < ethylmorphine < pholcodine). Analgesics wfth high climcal potency containing a methoxyl group at position 3 (e.g hydrocodone, K, = 198nM) had relatwely weak receptor binding, whdst their Odemethylated metabohtes (e g hydromorphone, Kj = 0.6 nM) had much stronger binding. Many op=oids may exert their pharmacological actions predominantly through metabolites. Several types of oplo=d receptors have been shown to be present within and outs=de the central nervous system These receptors have been designated I~, ~ and (1,2) The I1 receptor binds agomsts such as morphine, and mediates the antinociceptive actions of such drugs. Binding affimties of several op~oids to th=s receptor have been stud=ed and a hierarchy of affimties (as measured by EDso) has been asmgned by Pert and Snyder (3): morphine 7 nM; (-)-methadone 30 nM; meperidine 1000 nM; (-)-codeine 20000 riM; (-)-oxycodone 30000 nM These workers used 3Hnaloxone as the receptor ligand. Recently more specific i~ receptor ligands have been synthesised, and 3H-DAMGO ([Tyr*-D-Ala-Gly-N-MethyI-Phe-Gly-ol]enkephalin) has been shown to be highly specific for the p. receptor (4). The binding affimty of morphine and its 3- and 6-glucuromde metabolites has been investigated using 3H-DAMGO. Results have shown that the affinity of morphine6-glucuronide (ICso =7.3 nM) to the I~ receptor is similar to that of morphine (ICso =4.3 riM), whilst that of morphine-3-glucuronide is low (ICso >500 riM) (5). Similar results were obtained w=th 3H-naloxone (6,7) and again recently with 3H-DAMGO (8). The very potent in vitro binding of morphme-6-glucuronlde to the I~ receptor is reflected in vivo by its potent analgesic activity (5,8-10). All the clinically used opimds undergo metabolism in humans. There is little reformat=on about the I~ receptor binding of these oploids and their known metabolites. 0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc

2166

Mu Receptor Binding of Opioids

Vol. 48, No. 22, 1991

The aim of this study was to determine the affinity of some commonly used opioids and their metabohtes to the I~ receptor using the specific hgand 3H-DAMGO. By analogy with morphine and =ts htghly actwe 6-glucuromde, m vitro binding information could be used to predict that the actions of other op=oids may be mediated, at least in part, through metaboStes. Methods This study was approved by the Ammal Ethics Commtttee of the Unwersity of Adelaide, 3H-DAMGO was obtained from Du Pont NEN Research Products (Boston, MA, U.S.A.). The specific activity was 35.0 C0/mmol and rad=ochemical punty was more than 99%. Unlabelled DAMGO was obtained from Sigma Chemical Company (St. Louis, MO, U S.A.). The following compounds were tested: morphine-3-glucuronide, oxycodone, hydrocodone, hydromorphone, noroxymorphone, thebaine (Sigma Chemical Company); morphine-6-glucuromde, morphine-3-sulfate, normorphine (National Institute of Drug Abuse, Rockville, Maryland, USA); ethylmorphine (E. Merck, Darmstadt, F.R.G.), (+)methadone (kindly prov0ded by Dr. S. Pond, Department of Medicine, University of Queensland, Brisbane, Australia); morphine hydrochloride, codeine phosphate, pholcodme (F.H Faulding and Co. Limited, Adelaide, Australia); norcodeine (Eh Lilly and Co. Indianapolis, Ind, U.S.A); diamorphine (The Royal Adelaide Hospttal Pharmacy Department); dextromethorphan, dextrorphan (Roche Products Pty Ltd, Sydney, Australia) and codeine-6-glucuronide (for details of synthests and purity see reference 11). Binding was performed by modifications to the method of Pert and Snyder (3). Male Wistar rats (weighing 250-350 g) were decapitated and the brain, after removal of the cerebellum, was homogentsed in 10 ml of 0.1 M Tns buffer (pH 7 4, 0°C). The homogenate was finally diluted to 110 volumes of tissue wtth the cold Tris buffer. 1 8 ml of this freshly prepared homogenate was incubated w=th 1 x 10-9 M SH-DAMGO for 5 mmutes at 20 °C and then 0.1 ml of the test compound solut0on (10-9 - 10-3 M m water) was added and incubated for 15 minutes at 20 °C. Samples were then cooled to about 4 °C in an ice bath and then filtered under reduced pressure through Whatman glass-microfibres (2.5 cm GF/B; Whatman Internatmnal Ltd, Maidstone, England). The filters were washed twice with 8 ml of cold 0.1 M Tris buffer. After add=bon of 10 ml liquid scintillation cocktail (Ready Value, Beckman Instruments Inc., Fullerton, CA, U.S.A.), radioactivity was determined by hquid-scintillation counting (LS 3801 Beckman Instruments). Each sample was counted three times The diluted brain homogenate was incubated wtth a range of 10 concentrations (5 x 10-12 M to 5 x 10-7 M) of 3H-DAMGO and binding curves were constructed. The affinity constant (Kd)for 3H-DAMGO binding was determined by nonlinear least squares regression analysis of the rectangular hyperbola relating 3H-DAMGO concentration in the incubate to the amount bound(as determined by scintillation counting). IC50(the concentratmn to specifically inhibit 3H-DAMGO binding by 50%) was calculated from the log concentration-% mhtbition plot. The inhibition constant (K0 was calculated (12) as Ki = 1(350/(1+ [L] / Kd ) where [L] is the concentration of the ligand 3H-DAMGO.

Vol. 48, No. 22, 1991

Mu Receptor Binding of Oplolds

2167

Results The displacement of 3H-DAMGO by unlabelled DAMGO and morphine ~s shown in Figure 1. The Kd value for DAMGO was 0.28 nM. Displacement curves of 3H-DAMGO by morphine and its metabolites morphine-6-glucuronide, morphme-3-glucuromde, normorphine and morphine-3-sulfate are shown in Figure 1. Their corresponding K~ values are shown in the table. Figure 2 shows the displacement of 3H-DAMGO by codeine and ~ts metabolites codeine-6-glucuronide, norcodeme and morphine The 1 O0

60

.~ •~

40

~

M-6-G

• ~

M-3-G morphine normorphme

-•

M-3-S DAMGO

20

0 10-5.--i

..--,

10 i

10 2

Concentration

10 3

10 4

10 5

(nM)

FIG 1. The displacement curves of 3H-DAMGO by DAMGO, morphine and its metabolites. Each point represents the mean of 6 observations and the ordinate represents % displacement of total 3H-DAMGO binding. (M-6G: morphine-6-glucuronide; M-3-G: morphme-3-glucuronide; M-3-S: morphine-3-sulfate) 80-

El

60 f,,

O

. i

-0--" ,I,

40

n e-

codelne C-6-G morphine norcodelne

2O

ra

0 |..

, ..... ,

O-2

. . . . . -~ . . . . . . , O5

Concentration (nM)

FIG. 2. The displacement curves of 3H-DAMGO by codeine and its metabolites. Each point represents the mean of 6 observations and the ordinate represents % displacement of total 3H-DAMGO binding. (C-6G: codeine-6-glucuronide)

Mu Receptor Binding of Opioids

2168

Vol. 48, No. 22, 1991

8O

A

so C

.o

----0---

40

.Q i,f-

2O

morphine oxycodone hydrocodone hydromorphone noroxymorphone

i

0

5

Concentration (nM) FIG. 3.

The displacement curves of 3H-DAMGO by oxycodone and related compounds w~th comparison to morphine. Each point represents the mean of 4-6 observabons and the ordinate represents % displacement of total 3H-DAMGO binding.

80

~ o

60

C .0

J~ J~

8

40

o

=m

20, J_ aL I

'

0-2

"'""1

"

10-1

"'"'"1

!

10 0

'"--'1

• |'"'1

10 1"

"

10 2

"""'I

"

10 3

" """"1

"

10 4

dextromethorphan dextrorphan morphine pholcodine dlamorphlne thebame ethylmorphme (+)methadone

• ""'=1

10 5

Concentration (nM) FIG. 4.

The displacement curves of 3H-DAMGO by various opioids. Each point represents the mean of 4-6 observations and the ordinate represents % displacement of total 3H-DAMGO binding. corresponding K, values are contained in the table. Figure 3 shows the displacement of 3H-DAMGO by oxycodone and related compounds hydrocodone, hydromorphone and noroxymorphone, with morphme included for comparison. The table contains the corresponding K, values and figure 4 shows the displacement of 3H-DAMGO by various other opioids with morphine included for comparison. The K, values of these compounds are shown in the table. All the binding curves had a similar slope. Non-

Vol. 48, No. 22, 1991

Mu Receptor Binding of Oplolds

2169

TABLE Structures and K, values of some opiolds chemically related to morphine

SR2 v H 7

Compound

R1

R2

other changes

K, (nM)

mean Hydromorphone Morphine-6-glucuronide Morphine Normorphine Diamorphine Hydrocodone Noroxymorphone Methadone* Morphine-3-sulfate Morphine-3-glucuronide Oxycodone Codeine-6-glucuronide Codeine Norcodeine Ethylmorphine Dextrorphan Thebaine Dextromethorphan Pholcodine

OH OH OH OH

=O

OCOCH3 OCH3

OCOCH3

OH OSO3H

OH OH =O =O

OCH3 OCH3 OCH3 OCH3 OC2H5 OH

OH OH =O OC6H905 OH OH OH H

OCH3

OCH3

OCH3

H OH

OC6H905

OC2H4N(C2H4)20

3

OC6H905

1 3 1,2,3

2,3 1 5 4 s

0.6 0.6 1.2 4.7 9.6 19.8 23.7 28.8 30.7 37.1 47.4 238.7 248.3 266.9 345.0 494.7 636.2 1018 5610

SD 0.32 0.45 0.32 0.51 3.90 10.8 8.9 10.4 13.6 19.9 9.3 91.5 101.1 107.5 270.9 116.7 349.4 301.8 1647

Ki: inhibition constant. SD: standard deviation. *: chemical structure not related to morphine. 1: H in place of CH3 at posibon 17; 2: OH in place of H at position 14; 3: single bond between C7 and Ce; 4: double bond between C6 and C7 and C8 and C14 and single bond between C7 and C8; 5: no oxygen between C4 and Cs.

2170

Mu Receptor Binding of Opiolds

Vol. 48, No. 22, 1991

specific binding was about 20+_2% (n=9) of the total binding. Each binding experiment was performed 4 to 6 times. Dtscussion Morphine is regarded as the reference compound agamst which other analgesics are assessed, and its binding affinity to the p. opiold receptor has been studied using various ligands including 3H-naloxone (6,7) and 3H-DAMGO (5,8). Morphine-6glucuronide is an active metabolite of morphine and its binding affinity to the p. opiold receptor was greater than that of morphine, supporting the results of previous studies (5-8). In contrast, the K, of the non-analgesic metabolite morphine-3-glucuronide was 50 times greater than that of morphine, in support of previOUs data (5,6,8). Morphine3-sulfate had a simdar binding affinity as morphine-3-glucuronide and by analogy is unhkely to be an analgesic, ff formed, in man. Our results with codeine and =ts metabohtes have shown that only morphine binds to the I~ receptor with high affinity, and that the parent compound and the metabolites norcodeine and codeme-6-glucuronide have about 1/200th the affinity compared with morphine. These data support those obtained by Pert and Snyder (3) and Christensen and Jorgensen (6). The relatively weak binding of codeine, its N-demethylated and glucuronide metabolites support the original hypothesis of Sanfilippo (13) that codeine exerts its analgesic effect via its O-demethylated metabolite morphine. In contrast to morphine, whose 6-glucuronide has high affinity for the I~ receptor, the bindmg of codeme-6-glucuromde, quantitattvely the most important metabolite of codeine (14-16), has a very low affinity for this receptor, and is therefore unlikely to have analges=c activ=ty Oxycodone, chemically related to codeine, was onginally reported to have in vivo analgesic potency closer to morphine than codeine (17). Recently, it was suggested that oxycodone was more potent on a weight basis than morphine for acute postoperative pain (18). Our binding data and those of Pert and Snyder (3) would suggest that oxycodone should not exhibit such high analgesic activity. One explanat=on for its clinical efficacy ts that tt may have an active metabolite(s). A possible candidate could be the O-demethylated metabohte oxymorphone, by analogy with codeine and its O-demethylated metabolite morphme. Hydromorphone 0s a very potent analgesic in wvo, and this 0s almost certainly a reflection of its high in vitro binding. Hydrocodone, also clinically active, bound less avidly to the I~ receptor. Its clinical potency, however, may be a reflection of the in vivo formation of hydromorphone wa O-demethyla,tion. Our results have shown that diamorphine has a lower I~ receptor binding affinity than morphine, confirming the results of Inturrisi et al (19). In addition, our results w~th racemic methadone are similar to those of Pert and Snyder (3) who studied each enantiomer separately. Pholcodine and dextromethorphan are commonly used antitussive drugs. Previous studies have shown that they are devoid of analgesic activity (20) and our findings support these clinical observations. Our results also indicate that their ant=tussive effect is hkely to be mediated by receptors other than the !~ opmid receptor. An examinatton of the chem=cal structures and binding affinities (Table) of the opioids studied allows one to make certain predictions about structure:activity relationships. Firstly, demethylation at the nitrogen at position 17 results in compounds with reduced.binding affinity (normorphine, norcodeine), and thus less analgesm activ=ty than the parent molecule. Secondly, irrespective of size and

Vol. 48, No. 22, 1991

Mu Receptor Binding of Opioids

2171

physlco-chemical properties, the chemical group at position 6 has little effect on the binding affinity. For example, morphine-6-glucuronide which has a large sugar moiety at position 6 is still a highly potent analgesic and has a high binding affinity to the I~ receptor. Thirdly, the group at posttion 3 ts the most important determinant of binding and also analgesic activity. For example, the hydroxyl group at this position confers the highest potency (morphine); codeine with a methoxyl group, ethylmorphine with an ethoxyl group and pholcodme with a morphohnoethoxyl group have decreasmg binding and loss of analgesic activity. Fourthly, poor m vitro binding of a compound does not necessarily imply poor analgesic actwity, as many of the substances undergo metabolism in vivo, especially at pos=tton 3 to form active analgesics. This is exemplified by codeine, whose active metabolite is morphme. An understanding of an opioid's pharmacolog0cal properties must now include elucidation of the activity (in vitro and m vivo) of its metabolites. Acknowledaements Dr. Chen was supported by the Nabonal Health and Medical Research Council of Australia and now is a Florey Fellow of the Royal Adelaide Hospital. Dr. Somogyi was supported by the Royal Adelaide Hospital Research Fund. We thank the National Institute of Drug Abuse (U.S.A.) for providmg us with some of the compounds. References 1. W.R. MARTIN, Pharmacol. Rev. 35 283-323 (1983) 2. S. WATSON and A. ABBOTT, Trends m Pharmacol. Sci..~.gDi2J,.Z1 (1990). 3 C.B. PERT and S.H. SNYDER, Proc. Nat. Acad. Sci. U.S.A. 70 2243-7 (1973). 4. A. GOLDSTEIN, Trends in Pharmacol. Sci. 8 456-9 (1987). 5. G.W. PASTERNAK, R.J. BODNAR, J A. CLARK and C.E. INTURRISI, Life Scl. 41 2845-2849 (1987). 6. C.B. CHRISTENSEN and L.N. JORGENSEN, Pharmacol. Toxlcol. 60 75-76 (1987). 7 F.V. ABBOTT and R M PALMOUR, I.Jfe Sci. 43 1685-1695 (1988). 8. D. PAUL, K.M. STANDIFER, C.E. INTURRISI and G.W. PASTERNAK, J. Pharmacol. Exp Ther. 251 477-483 (1989). 9. R. OSBORNE, S. JOEL, D. TREW and M. SLEVlN, Lancet i 828 (1988). 10. H. YOSHIMURA, S. IDA, K OGURI and H. TSUKAMOTO, Biochem. Pharmacol. 22 1423-1430 (1973). 11. Z.R. CHEN, G. REYNOLDS, F. BOCHNER and A. SOMOGYI, J. Chromatogr.Biomed. Appl.493 313-324 (1989). 12. Y.-C. CHENG and W.H. PRUSOFF, Biochem. Pharmacol. 22 3099-3108 (1973). 13. G. SANFILIPPO, Boll. Soc. Ital. Brol. Sper. 24 723-726 (1948). 14. T.K. ADLER, J.M. FUJIMOTO, E L. WAY and E.M. BAKER, J Pharmacol. Exp. Ther. 114 251-262 (1955). 15. Q.Y. YUE, J.-O.SVENSSON, C. ALM, F. SJOQVIST and J. S,~,WE, Br. J. Clin. Pharmacol. 28 639-645 (1989). 16. Z.R. CHEN, A.A. SOMOGYI, G. REYNOLDS and F. BOCHNER, Br. J. Clin. Pharmacol. (In Press). 17. W.T. BEAVER, S.L. WALLENSTEIN, A ROGERS and R.W. HOUDE, J. Pharmacol Exp. Ther. 207 92-100 (1978). 18. R. P~)YHIA., E. KALSO, P. ONNELA, I. TIGERSTEDT and T. TAMMISTO, Pain S143 (1990). 19 C.INTURRISI, M. SCHULTZ, S SHIN, J. G. UMANS, L. ANGLE and E.J. SIMON, Life Sci. 33 773-776 (1983). 20. N.B. EDDY, H. FRIEBEL, K.-J. HAHN and H. HALBACH, Bull. W. H. O. 40 639-719 (1969).