J. Med. Chem. 1993,36, 179
179
A Remarkable Change of Opioid Receptor Selectivity on the Attachment of a Peptidomimetic K Address Element to the 6 Antagonist, Natrindole: St-[ (N1-Alkylamidino)methyl]naltrindole Derivatives as a Novel Class of K Opioid Receptor Antagonists
A H O A
S. L. O h t a d > A. E. Takemori: and P. S.P 0 r t o g h ~ ' ~ t 'OH
Department of Medicinal Chemistry, College of Pharmacy, and Department of Pharmacology, Medical School, University of Minnesota, Minneapolis, Minnesota 55455 Received November 11, 1992
The selectivities of the prototypical 6 and K opioid receptor antagonista, naltrindole (1, NTI) and norbinaltorphimine (2, norBNI) have been attributed to the
Y
B
\
OH
Figure 1. Comparison of a 3-dimensional representation of norbinaltorphimine 2 (A) with that of S'-[(N1-akylnmidino)methyllnaltrindole (3-5) (B).Note that the basic groups in the right hand part of these molecules are in similar orientations with respect to the antagonist pharmacophore. 2
presence of nonpeptide 'address" mimics which bear a functional relationship to key elementa in the putative 8 and K addresses of enkephalin and dynorphin, respectively.'B2 Accordingly, the design of 1 employed a model that envisaged the Phe4 phenyl group of enkephalin as a Similarly, the address critical part of the S addre~s.~ element conferring selectivity in 2 has been suggested to be a basic function that mimics the guanidinium moiety of Arg7 in dynorphin.' Here we report on a dramatic and unprecedented change of opioid receptor selectivity from 6 to K , simply by the modification of NTI with a basic group which functions as a K address, using the above model. This molecular modification affords a new class of highly selective K opioid receptor antagonists (3-5).
R &,
(CH2)2CH,
k (CH,)&H,
The design rationale for the series involved the attachment of a basic group to the 5' position of NTI in order t
t
Department of Medicinal Chemistry. Department of Pharmacology.
0022-262319311836-0179$04.00/0
to approximatethe distance between one of the antagonist pharmacophoresof 2 and its second basic group which has been suggested' to function as a K address mimic. Alignment of the antagonist pharmacophore and the basic amidinegroup of 3-5 with those of norBNI (2) is illustrated in Figure 1. In this context the indole moiety functions as a rigid spacer to hold the amidine group in a location similar to that of the right-hand basic group of 2. Another consideration in the design was based on structure-activity relationship studies which showed that &antagonist potency decreased on substitution of the indole moiety of NTI at the 5'-positi0n.~ Thus, 5'substitution of NTI with an appropriate basic group should simultaneouslysuppress &antagonistpotency and greatly enhance K-antagonist potency. Compounds 3-5 in the series were synthesized as outlined in Scheme I. Reaction of naltrexone (6) with 4-hydrazinobenz~nitrile~ under Fischer indole conditions afforded the 5'-nitrile 7, which was reduced to the primary amine 8 using h e y Ni. The amidines 3-5 were prepared by reacting 8 with the appropriate imidate e ~ t e r s . ~ * ~ The pharmacological data (Table I) clearly show the dramatic change in the selectivity of NTI (1) upon modification with a 5'- C(aLkylamidino)methyll substituent. It can be noted that little, if any, of the 6 antagonism of NTI is associated with these derivatives 3-5 and that there is a concomitant increase in the K antagonist potency. Interestingly, 5 possesses greater in vitro K antagonist potency and selectivityin smooth muscle thandoesnorBNI (2). The opioid receptor binding data for the most potent compound, 5, is consistent with the K selectivity (Table 11). Studies in mice indicated that 5 is a selectiveK antagonist Q 1993 American Chemical Society
180 JourrMl of Medicinal Chemistry, 1993, Vol. 36,No.1
receptors.l0 This study also suggesta that a relatively simple group may be employed to mimic a key residue in an aminoacid address sequence. The general implications of this study are that selective peptidomimetics may be designed when the clam of peptides is organized into message and address domains and there is sufficient structureactivity information to determinethe key amino acid residue(s1 that functions as an address. Acknowledgment. We thank Michael Powere, Veronika Doty, Joan Naeseth, and Mary Lunzer for in vitro and in vivo tasting of the compounds. Thiswork was supported by the National Institute on Drug Abuse. References
Scheme I NHNH,
0 CN
6
Table I. Antagonist Potenciea in Smooth Muscle Preparations ICs0 ratios* (*SEMI selectivityratio mmpd EK (K)* Mhdb DADLE We dfi 1 (NTI) 1.3 & 0.2 11.2 1.8 469 & 104 (6/b, 41) ( 6 / K , 353) 17 8.3 i 1.8 10.4 i2.9 22 2 (norBNI)d 181 i 7 3 4 6
159 A 43 11.3 i 2.6 185A 58 19.3 i 5.3 439 i100 15.7 i4.0
2.28 i0.57 14 3.00i0.51 10 4.71 i1.06 28
69 62 93
a The ICw of the agonist in the presence of the antagonist (100 nM) divided by the ICw of the agonist alone in the same preparation. The number of replicate eeeays 1 3. b Determined in the guinea pig ileum preparationll using ethylketezocine (EK) or morphine (M). e Determinedin the m o w vm deferene preparation1'Jusing [ ~ - A l a * p Leu61enkephalin (DADLE). Data taken from ref 13.
Table 11. Opioid Receptor Bindinf of S
Ki,nMb -mPd
rce
6. 5.5 43
wd
3.5 47
Ki selectivity ratio d K a/ K
57 90 150 181 a Guineapig brain membranea were employed usinga modification of the method of Werling et al.14 Valuea are geometric means of at leaet three replicate experiments. [sHlU69593.16 [sHl ala^,MePhe4,Gly-ols]enkephaljn16 (DAMGO). * [WDPDPED f Data from ref 17.
S
2(norBNI)f
0.061 0.28
Communications to the Editor
*
in vivo. At a dose of 4 mglkg sc, 6 increased the antinociceptive ED60 dose of the K selective agonist, trans(*)-3,ddichloro-N-methyl-N-[2-(l-pynolidmyl)cyclohexyllbenzenea~etamide~ (U-50,488) by a factor of 3.77(2.964.76) whereas insignificant increases were observed for the 6 agonist, [D-Pen2,D-PenS]enkephalins (DPDPE), and the p-eelective agonist, morphine. In conclusion,the data are consistent with the idea that the indole moiety in this series functions as a rigid scaffold to orient the amidine group to an address subsite on the K opioid receptor. It has been suggested2that this subsite may recognizethe guanidinegroupof Arg7 which is believed to be important for the recognition of dynorphinat K-Opioid
(1) Schwymr, R. ACTH A Short Introductory Review. Ann. N.Y. Acad. Sci. U.S.A. 1977,297,3-26. (2) Portogheee, P. S. Bivalent Ligands and the Meaaage-Addnru Concept in the D e s i i of Selective Opioid Receptor Antagonbtr. Trendtz Pharmacol. Sci. 1989,10,23&236. (3)Portoghese, P. S.;Sultana,M.; Takemori, A. E. Design of Peptidomimetic &Opioid Receptor Antagonista Using the MwageAddresa Concept. J. Med. Chem. 1990,33,1714-1720. (4) Portoghese, P. S.;Nag-, H.; Takemori, A. E. Only One Phsrmacophore Is Fbquired for the K Opioid Antagonist Selectivityof Norbinaltorphimine. J. Med. Chem. 1988,31,1344-1347. (5) Rivett, D. E.; Ronevear, J.; Wilshire, J. F. K. The Preparation and SpectralProperties of Some Monoeubetitut.4 1,3,5-Triphenyl-2pyrmlines. A u t . J. Chem. 1979,32,1601-1612. (6) Sandler,S.R.; Karo, W.Imidatm. Organic Chemistry;Academic Press: New York, 1972;Chapter 8, VoL 3,pp 268-299. (7) Cereda, E.; Ezhaya, A.; Quintero, M.G.; Bellora, E.; Dubini, E.; Micheletti,R.; Schiavone,A.; Brambde,A.; Schiavi,0.B.; Donetti, A. J. Synthesb and Biological Evaluation of New A n t i m u d n i c Compounds with Amidme k i c Centers. A Ueeful B i o b b r i c Replacementof C h i d CationicH d . J. Med. Chem. 1990,53, 2108-2118. (8) VonVoigtlander, P. F.; Lahti, R. A; Ludem, J. H. U-50,488:A Selectiveand StructurallyNovel Non-mu (Kappa)Opioid Agonist. J. Pharmacol. Exp. Ther. 1989,224,7-12. (9) Moebarg, H. I.; Hurst, R.; Hruby, V. L.; Gee,K.; Ya"ur4 H. I.; Galligan,J. J.; Burke,T. F. Bin-PenicillamineEnkephah Poeaea Highly ImprovedSpecificity Toward Delta-OpioidReceptors. Proc. Not. Acad. Sci. U.S.A. 1988,80,6871-5874. (10)Chavltin, C.; Goldsbin, A. Specific Receptorfor the Opioid Peptide D y n o r p h Structure-Activity Relationships. Proc. Not. Acad. Sei. U S A . 1081, 78,65434647. (11)Rang,H. P. Stimulant Actio- of Volatile haenthetics on Smooth Mwle. Er. J. Phurmacol. 1964,22,ssB-366. (12)Hendenon, G.; Hughes, J.; and Koeterlitz,H. W.A New Example of a Morphine-Sensitive Neuroeffector Junctiox Adrenergic Trammianion in the M o w vas Deferem. Br. J. Pharmmol. 1972, 46,764-766. (13) Portogheee, P. S.;Garzon-Aburbeh, A.; Nag-, H.;Lm,C.-E.; Taksmori,A. E. Role of the Spacerin Conferring K-Opioid Receptor Selectivity to Bivalent Ligands Related to Norbinaltorphiminr. J. Med. Chem. 1991,5V,1202-1298. (14) Werling, L. L.; Zarr,G. D.; Brown, 5. R.;COX,B. M. Opioid Binding to Rat and Guinea Pig Neural Membranes in the Preaence of Phpiological Cations at 37O C. J. Pharmacol. Exp. m e r . 1986, 233,722-728. (15) U t i , R A.; Micklwn, M. M.; McColl, J. M.;vonVoigtlander, P. F. [aHlU69,593A Highly Selective Ligand for the Opioid Kappa Receptor. Eur. J. Pharmacol. l98S,109,281-284. (16) Handa, B. K.; Lane, A. C.; Lord, J. A. H.; Morgan,B. A.; Rance, M. J. Analoguesof B-LPI&* Poewsing SelectiveAgo& Activity at Mu-Opiate Receptors. Eur. J. Pharmacol. 1981,70, 631-640. (17) Takemori, A. E.; Ho, B. Y.;Nameth, J. 9.; Portoghew, P. S. Norbinaltorphimine,a Highly SelectiveKappa-Opioid Antagonist in Analgeaia and Receptor Binding Amp. J. Pharmacol. Exp. Ther. 1988,246,255-258.
