8/8/2019 J.Med.Chem.1981.24.1429-1432

1/4

J.Med. C hem. 1981,24,1429-1432 1429under way between our laboratory and th at of Kr au t andMatthews in La Jolla employing the triazines used todevelop the QSAR in this report for dihydrofolate re-ductase.

members of sets of congeneric inhibitors bound to theenzym e. These studie s, which are now possible, will firmu p our QSA R-derived conclusions an d enable us to m oreconfidently refine QSAR techniques. Such a p roject is now

Structure-Activity Relationships of the Cycloalkyl Ring of PhencyclidineRoy L. McQuinn, Edward J. Cone,* Harlan E. Shan non, and Tsung-Ping SuNat ional Zmt i tute on Drug Abuse, Divis ion of Research, Addict ion Research Center , Lexington, Ke ntuck y 40583.Received August 20, 1981

In order to investigate the structural requirements for a cycloalkyl moiety in the potent hallucinogen 141-phenylcyc1ohexyl)piperidine (PCP, l) ,a series of structural analogueswas synthesized in which th e sizeof the cycloalkylring was varied from three carbons to eight carbons. Biological activities of thes e compounds were assessed in a nin vitro assay (phencyclidine binding assay) and an in vivo assay (discriminative stimulus assay). Aa he cycloalkylring size decreased from that of cyclohexane (PCP), PCP -like activity declined in b oth aasays, but as th e cycloalkylring size became larger than cyclohexane, a sharp decline in PCP-like activity was observed in the in vivo assay,while activity in th e in vitro assay was only slightly less tha n th at of PCP. l-(l-Phenylcycloocty1)piperidine 8)had potent competitive binding properties in th e in vitro binding w a y bu t produced no observable PCP-like effectsin the in vivo assay. Th e importance of th e cycloalkyl ring in the struc ture of PC P was demonstrated by testingbenzylpiperidine (2), which had alm ost no measurable activity in either assay.Phencyclidine (PCP, 1) is an easily synthesized and th us

readily available drug that has become a substance ofmajor abuse, particularly among teenagers. Agitated,hostile, and aggressive behavior, depressive states, a toxicpsychosis, and a nu mbe r of deaths, including suicides andmurders, have been associated with PCP abuse.2 PC P(Sernyl) was introduced more th an 15 years ago as ananesthetic medication in man, b ut t he high incidence ofundesirable side-effects during recovery from anesthesiahas precluded its further use in man.3 In spite of PCPsmany negative aspects, the eupho rigenic properties of P C Phave made the drug popular among recreational users.PC P appears to represent a unique class of drugs whosespectrum of action is readily distinguishable from that ofother classes of psychoactive drugs, includ ing narcotics,LSD-like hallucinogens, amphetamine-like stimulants,barbiturate-like depressants, and cholinergic agonist^.^Although th e pharmacology of P CP h as been extensivelystudied , little is known about th e mechanism of action ofPC P a t the molecular level. In order to gain some un-derstanding of the mechanism of action of PCP, we haveundertaken studies to explore the structural requirementsfor PCP-like activity. Th e first of these studies is reportedin this paper.An exam ination of th e P CP molecule reveals a sem irigidstruc ture in which th e interatomic distance between thebenzene and piperidine rings is governed by the inte rnalbond angles of the cyclohexane ring. Th e flexibility of thecyclohexane ring allows these bond angles to conformclosely to t he the oretical 109.5 bond angles found fornormal sp3 hybrid bonding. Th is relationship was con-f m e d by Argos e t alS6n a study of the molecular structureof PCP.HC1 by X-ray crystallography. They found theangle 8 see Table I) to be 109.0. Also, th e distance be-tween th e bo nding carbon (C7) of th e cyclohexyl group and

~~~ ~ ~(1) Riker Laboratories, Inc., St.Paul, MN 55101.(2) R. C. Peterson and R. C . Stillman,NZDA Res. Monoqr.,no. 21,(3) E.Domino, Znt. Reu. Neurobiol., 6, 303 (1964).(4) D. . Jasinski, H. E. Shannon, E. J. Cone, D. B.Vaupel, M.E. Risner, R. L. McQuinn, T.-P.Su, and W. B. Pickworth, inPC P (Phencyclidine): Historical and Current Perspectives,E. F. Domino, Ed., NN P Books, Ann Arbor, MI, pp 331-400.(5) P. Argos, R. E. Barr, nd A. H. Weber, Acta Crystallogr., Sec t.B, 26, 53 (1970).

1-17 (1978).

th e nitrogen atom (N13) of th e piperidine group was foundto be 1.550 A, while the distance between C7 and thebonding carbon of th e phenyl group (C l) was 1.543 A.Since the value for 8 nd the distances between C 7 4 1 andC7-Nl3 have been experimentally measured, the distance(a)between the nitrogen atom of the piperidine ring andth e C1 ato m of the phenyl ring can be calculated by uti-lizing a simp le trigometric relationship (a2= b2+ c2- 2bccos 8, where b and c are the lengths of th e adjacent sidesof the angle 8). Th e distance a was calculated to be 2.518A.PC P has been shown to exhibit antich olinergic activity!In th e case of acetylcholine, the distance t ha t separatesthe areas of minimum potential between the negativelycharged ester oxygen and the positively charged tri-methylammonium group has been recognized as a criticalparam eter in determining cholinergic activity.a Since th ePCP-HC1molecule also can be visualized as having an areaof localized positive charge (th e nitrogen atom of the pi-peridine ring) a t a fixed distance (2.518 A) from a n areaof negative charge (th e delocalized electron cloud of t hebenzene ring): it has been suggested th at th is distancemight be a critical parameter in determining PCP-likeactivity.8~~n order to test whether th e distance betweenthe benzene and piperidine ring is critical for PCP-likeactivity, a series of com pounds was prepared in which theangle 8, and consequently the distance a, was systemati-cally altered w hile leaving th e com position of the benzeneand piperidine rings unaltered. Assuming that th e dis-tances between C7-Nl3 and C7-C1 of PCP are inde-pendent of bond angles and remain constant, th e distancea should be directly dependent upon the value of 8. Fo rsatura ted cycloalkyl ring sy stems of sixcarbons or less, th ebond angles are depend ent upon t he ring size. By alteringthe size of th e cycloalkyl ring in PC P, th e value for 8 and,consequently, th e distance between th e benzene an d pi-peridine ring can be altered. For rings larger than six(6) S. Maayani, H. Weinstein, N. Ben-Zvi,S.Cohen, and M. Sok-(7) J.B. Robinson,B.Belleau, and B. Cox,J.Med. Chem., 12,848( 8 ) S . Maayani, H. Weinstein,S. Cohen, and M. Sokolovsky,h o c .(9) H. Weinstein, S.Maayani, S. Srebrenik, S. Cohen, and M.

olovsky, Biochem. Pharmacol., 23,1263 (1974).(1969).Natl . Acad. Sci. U.S.A., 7 0 , 3103 (1973).Sokolovsky, Mol. Pharmacol., 9,820 (1973).

This article not subject to U.S. Copyright. Published 1981 by th e American Chemical Society

8/8/2019 J.Med.Chem.1981.24.1429-1432

2/4

1430 Journal of Medicinal Chem is try , 1981, Vol. 24 , No. 12 McQuinn et al .Table I. Physical Properties of the Hydrochlorides of Some Cycloalkyl-Substituted PCP Analogues\,C*\

\ J(CH&yield,no. n % 8 , b deg a , mp, "C recrystn solvent formula

2 0 43 184-1 86 CH,CN ether c 2H 8 ClN3 2 21 116 2.623 195-197 CHCl, ether CI,H,,ClN4 3 28 11 1 2.549 243-245 CH,CN CISH22C1N5 4 22 10 9. 5 2.526 239- 240 CH,CN C,,H2,ClN1 5 63 109 2.518 243-244 CH, CN CI,H,ClN7 6 53 10 9 2.518 220 -22 2 CH,CN Cl,H*8ClN8 7 19 10 9 2.518 182-1 85 2-propanol C,,H,OC1NYields are for the last step. The angle e for compounds where n = 6 or 7 was assumed to be equal to that of phen-cyclidine. For compounds where n = 2, 3, or 4, the angle 8 was obtained from ref 10.analysis for C, H, and N. All compounds gave satisfactory

carbons, the value for B will remain essentially the same(109.0), since the ir flexibility allows them to achieve thepreferred bonding angles for norm al sp3 bonding.A series of P CP analogues was synthesized in which th esize of t he cycloalkyl ring was varied from three to eightcarbons. Th e theoretical values for the distance a for thesecompounds are presented in Table I. Th e values used forthe angle B are those reported for unsubstituted saturatedcycloalkyl ring systems.1 In addition t o th e cycloalkylanalogues, compound 2 was prepared to test the n ecessityof having a cyclic alkyl moiety in the structure of P CP-likecompounds.Chemistry. Compounds 1 and 5 were synthesized ac-cording to the procedures outlined by Kalir etCompounds 4 and 7 were prepared a s described by Kalire t al.12Th e method used to synthesize 3 involved th e produc-tion of 1-phenylcyclopropanecarbonyl azide an d its sub-sequent rearrangement a nd reduction to l-phenylcyclo-propylamine, followed by condensation with 1,5-di-bromopen tane to form t he pipe ridine ring.13 The syn-thesis of 8 involved a two-step reaction to produce 1-phenylcyclooctyl azide (10) and reduction of th e azide to1-phenylcyclooctylamine (1 ), ollowed by co ndensationwith 1,5-dibromopentane (see Scheme I). Overall yieldswere generally low, but purity was high. Compound 2 wa smade in good yield by refluxing benzylamine (6) in thepresence of 1,5-dibromopentan e and potassium carbo natein dimethylformamide. The physical properties of thecycloalkyl homologues of P C P a nd 2 are listed in TableI.

Biological Assay. The biological potency of each ofth e PC P analogues, relative to PCP, was measured by bothan in vitro and an in vivo w a y for PCP-like activity. Th ein vivo assay measures the potency of an analogue, relativeto P CP , in producing correct responding in a two-choice,disc reb tria l avoidance task by rata trained to discriminatebetween P CP and saline. Selective responding on th e P CPlever is taken t o indicate tha t the analogue is PCP-like or(10) C. A. Coulson and W. Moffitt, Philos. Mag., 40 , 1 (1949).(11) A. Kalir, H. Edery, Z. Pelah, D. Balderman, and G. Porath,J.Med. Chem., 12, 473 (1969).(12) A. Kalir, S. Sadeh, H. Karoly, E. Shirin, D. Balderman, H.Edery, and G. Porath, Isr. J . Chem., 13, 125 (1975).(13) C. Kaiser, B. M. Lester, C. L. Zirkle, A. Burger, C. S. Davis,T. J. Delia, and L. Zirngibl, J.M e d . Chem., 5, 1243 (1962).

Scheme I

th at ita effects generalize to those of PCP . This assay isbased upon the previous findings that PCP producesdiscriminative stimuli in the rat and that the stimuliproduced by several PC P structural analogues are gener-alized to those produced by PCP.1"16 Th e in vitro assaymeasures the displacement by the PCP analogues of[%]PCP specifically bound to rat brain homogenates usingthe p rocedure described by Zukin a nd Zukin."Results and Discussion

Th e cycloalkyl homologues of P CP (Table I) were syn-thesized by systematically altering the size of th e cycloalkylring in increments of a single methylene unit whilemaintaining a constant molecular composition for thephenyl and piperidine rings. Th e biological activ ity of eachcompound wa s measured in both an in vitro and in vivoassay. Relative potencies were determined for each com-pound by comparing its activity in a particular assay tothat of 1 (PCP ), which produced maximal activity in bothassays. As the cycloalkyl ring size decreased from t ha t ofcyclohexane (compounds 3 4 , a parallel reduction ofrelative potency in bo th assays was observed (see Table11),results consistent with those reported by K a l i P forcompounds 4 and 5. Although th e reduction in ring sizeand t he con comitant increase of a correlate with the ob-(14) T. U. C. Jarbe, J. 0. Johansson, and B. G. Henriksson,Psy-(15) D. A. Overton,Arch. Int. Pharmacodyn. Ther.,215,180 (1975).(16) A. D. Poling, F. J. White, and J. B. Appel, Neuropharmacol-(17) S.R. Zukin and R. S.Zukin, Proc. Natl. Acad. Sci. U.S.A .,7 6 ,(18) A. Kalir, Psychopharm. Bull., 16, 54 (1980).

chopharmacologia, 43, 33 (1975).ogy, 18, 459 (1979).5372 (1979).

8/8/2019 J.Med.Chem.1981.24.1429-1432

3/4

Phencyclidine Structure-Activity RelationshipsTable 11. Relative Potencies of PCP and Analoguesin the [ 'HIPCP Binding Assay and th e RatDiscriminative Stim ulus T est (DS)

r e i i i t e n c y['HIPCPno. binding assay rat DS

2 0.02 IACsd3 0.07 IA4 0.08 0.19 (k0.07)5 0.12 0.37 (k0 .13 )1 1.0 1 .07 0.23 ;O:k0.05)8 0.72

In th e binding assay, relative potencies were obtainedfrom the ICso values comp ared with PCP. Th e IC,, ofPCP was 270 nM. The SD for each sample was less than4%. The relative po tenc y f or each analogue is expressedas the milligrams of PCP equivalent to 1.0 mg of the indi-cated analogue. Did no t generalize to PCP. ProducedCNS depression a t 56 mg/kg. e No activity seen a t doses4 56 mg/kg. f Tonic seizures were observed a t 56 mg/kg.served decrease in relative potency, it is doubtful th at thisis th e only factor operatin g in th e diminution of PCP-likeactivity. Th e change in a is calculated to be only 0.008 Ain going from 1 o 5, whereas th e relative potency of 5 wasonly approximately one-eighth that of 1 in the in vitroassay an d about one-third of 1 n the in vivo assay. Com-pound 2 also was tested bu t showed almost no measurableactivity in either assay.As th e cycloalkyl ring of PC P analogues becomes largertha n th e cyclohexane ring of PC P, th e distance betweenthe benzene and piperidine rings is expected to remainapproximately the same. Thus, an d 8 should differ fromP CP only in the bulkiness of the cycloalkyl ring. Suchsimilarity in stru ctu re and m olecular composition wouldbe expected to yield compounds with similar biologicalactivities. Th e finding th at the relative potencies of 7 an d8 did not markedly differ from 1 as measured by t he invitro binding assay was not unexpected, bu t when 7 and8 were assayed for in vivo PCP-like activity, they werefound to have almost no measurable activity. Changes inth e preferred conformations of the P CP derivatives wereconsidered as a possible explanation of th e lack of activityof compounds 7 and 8. PCP has been shown to adopt aconformation with th e phenyl group in th e axial positionand t he piperidine ring in an equatorial position.lg OtherPC P derivatives which could adopt a similar conformationwere reported to b e more active tha n th e correspondingcis-tra ns isomer. How ever, th e cycloalkyl rings of com-pounds 7 and 8 are sufficiently flexible to exist in similarconformations to th at of PCP , and it seems unlikely tha tthis accounts for their lack of activity, It would appear,then, t ha t the molecular volume or shape of th e cycloalkylring in PCP -like compound s is equally as important as thedistance between the phenyl and piperidine rings in de-terminin g in vivo activity.It seems apparen t that even very minor perturbationsof the cy clohexane ring are no t well tolerated in retainingPCP-like activity. It has been suggested tha t th e cyclo-hexane ring of PC P co ntributes to th e biological activityof this drug by m aintaining th e proper structural rigidityof the other two rings and by contributing to hydrophobicbinding a t the receptor site.20 Th e latter hypothesis is

Journa l of Medicinal Chemistry, 1981, Vol. 24, No. 12 1431consistent with our findings that 7 an d 8 displace [3H]PCPin vitro, bu t is inconsistent with our findings tha t 8 hasvirtually no m easurable PCP-like activity in vivo. Th e facttha t 7 an d 8 bind well at th e P CP specific binding site invitro and yet produce very little m easurable activity in vivosuggested the possibility th at these two com pounds mayhave partial agonist an d/o r antagonist properties. How-ever, coadministration of 7 or 8 with l did not produceantagonistic effects in the ra t discriminative stimulus assay.Another possible explanation for th e observed disparitybetween t he binding assay an d the discriminative stimulusassay is th at the re may be more th an one binding site forPC P in the brain homogenate preparation. Thus, 7 and8 may have high affinity for one of these binding sites bu tlittle affinity for the site tha t is most correlated with theeasily observable effects of PCP-like d r u g s . Some supportfor such a possibility is offered by our observation that 7and 8 produce a marked pupillary dilatation in th e rat,while PC P and PCP-like drugs are characterized by theirability to produce pupillary co nstriction in t his species.It should be pointed out t ha t some doubt has been castover the specificity of the in vitro binding assay2' as ori-ginally reported by Zukin and Zukin," since PC P has atendency t o adsorb t o th e glass-fiber filters used in theseassays. However, our technique of pretr eat me nt of th efilters with tert-amyl alcohol reduced binding to th e filterby approximately 90%. Th e residual binding to th e filterwas nonspecific an d was subtracted as pa rt of total non-specific binding from the tota l binding to yield specificbinding. Vincent et al.n also have reported techniques forelimination of nonspecific binding to the glass fiber filters.Experimental Section

Melting points were determined on a F isher-Johns (ho t stage)appara tus and are unco rrected. All new com pounds were char-acterized by m elting point, IR , MS, and elem ental analysis (C,N, and N). Elemental analyses were performed by GalbraithLabo ratories, Inc., Knoxville, TN , and were within k0 .3% of thetheoretical values. IR sped ra were obtained on a Beckman IR-18AusingCHC& olutions or KBr pellets. Mass spectra were obtainedon a Finnigan M odel 3300 quadrupole GC-mass spectrometeropera ting n the m ethane chemical-ionization mode and equippedwith a Finnigan Model 6000 Interactive Data System.1-PhenylcyclooctylAzide (10). A solution of gP (5.0 g, 0.025mol) in 25 mL of CHC13 was add ed slowly with stirri ng to asolution of sodium azide (3.18 g, 0.049 M) a nd trich loroacetic acid(8.33 g, 0.049 mol) in 24 mL of CHCI3 a t -5 OC. Th e tem per atu rewas not allowed to exceed 5 OC during the addition. After th eaddition was complete, the mixture was stirred for 3 h at 0 "C.Th e solution was neutralized with NHlOH and ex tracted withCHCIB. Th e CHC13 extr act was thoroug hly washed with H20an dthen dried over N a2S 04. The CHC13 was evaporated und er re-duced pressure to yield 4.8 g of a clear viscous liquid: yield 84%;MS, m/e (relative intensity) 229 (M', 3), 201 (lo),187 (100). Thismaterial was used without fu rther purification.1-Phenylcyclooctylamine 11). The crude 10 (4.8 g, 0.02 mol)was dissolved in 40 mL of dry ether and added dropwise to asuspension of LiAIHl (1.3 g, 0.035 mol) in 100 mL of ether at 0"C. After th e addition was com plete, the mixture was refluxedfor 2 h. Th e mixture was cooled to -10 OC , and ethyl acetateether(l:l, 6 mL) was slowly adde d to the mixture. After the evolutionof gas ceased, the mixture was poured into an ice-cold solutionof 20% HC1. Th e aqueou s portion was collected, neutralized w ithN H 4 0 H ,and extracted with diethyl ether. Th e ether solutionwas washed with a 5% solution of K&03 and dried over an-hydrous NazS04. Evap oration of the ether gave 2.4 g of a hea vy,

(19) P.Geneste, J. M. Kamenka, M. S. N. Ung, P. Herrman, R.Goudal, and G. Trouiller, Eu r. J.Med. Chem., 14,30 1 (1979).(20) S.Maayani and H. Wein stein, in "M embrane M echanisms ofDrugs of Abuse", Alan R . Liss, Inc., New York, 1979, pp91-106.

(21) S.Maayani and H. Weinstein, Life Sci., 26, 2011 (1980).(22) J. P.Vincent, J. Vignon, B. Kartalovski, and M. Lazdunski,(23) A . C. Cope and R.B. Kinnel, J . Am . Chem. SOC., 8, 752Eur. J . Pharmocol., 68, 73 (1980).(1966).

8/8/2019 J.Med.Chem.1981.24.1429-1432

4/4

1432 J . Med. C h e m . 1981, 24 , 1432-1437clear, iscous liquid yield 59%; MS, m l e (relative intensity) 203(M+,2), 187 (72). T his ma terial was used without further pu-rification.1-(1-Phenylcycloocty1)piperidine 8) was made by refluxing11 with 1,5-dibromopentane and K 2 C 4 n DMF. yield 19% (0.699); m p 183-185 'C; IR (KBr) 3420,2940,3500-3670,1450,1205,1030, 950, 870, 760, 700 cm-'; MS , m l e (relative intensi ty) 272

[;H]PCP Binding Assay . Th e binding assay was performedas described by Zu kii and Z ~ k i n . '~ liquota of freshly preparedrat brain homogenate (1.0mL)were incubated at 4 C for 45 minwith 7.0 nM [3H ]PCP,and t he binding of [3H ]PCP to the mem-brane prep aration was mea sured by filtration assay. Th e glass-fiber filter was pretreated with water saturated with tert-amylalcohol to reduce binding o the filter. This treatment eliminatedapproximately 90%of [sH]PCP binding to th e glass-fiber filter.Th e nonspecific binding in th e presence of 0.1 mM PCP wassubtracted from the total binding to yield specific binding. Aconcentration-displacement curve for each analogue was visuallyfitted from a log-probit plot, and t he IC Mof each analogue wascompared to tha t of PCP , which was used as a standa rd in eachassay.

(68), 271 (M', 100). Anal. (C && lN), C, H, N.

Disc rimina tive S tim ul us Assay. Male, Fischer-derived CDFrats were trained to discriminate between PC P and saline in atwo-choice, discrete-trial avoidance task similar to that describedin detail elsewhere.' PCP (3.0 mg/k g) or saline was administere dintraperiton eally 30 min prior to a trainin g session. Each ex-perimental seasion consisted of 20 trials, nd the ra ts were traineduntil they reliably completed a t least 90% of the 20 trials on theappropria te choice lever. When acquisition of th e discrim inationwas com pleted, drug testing sessions were interposed amongtraining sessions. During training sessions, only a response onthe appropriate choice lever term inat ed a trial; uring test sessions,a respon se on either choice lever term inate d a trial. A dose ofa test drug was considered to produce stim uli which generalizedto those produced by PCP if the group of rats com pleted a meanof a t leas t 90% of the 20 trials on the PCP-appropriate choicelever. Relativ e potenc ies were determ ined by using sta nda rdbioassay statistics. PC P produced dose-rela ted discrimin ativestimuli when tested over a greater than tenfold dose range.

Acknowledgment. The authors thank Mr. W. G.Marqu ardt and Ms. W. M. Roberts for their aid in thepreparation of this manuscript.

a-Adrenergic Agents. 1. Direct-Acting a1Agonists Related to Methoxamine'R. M. DeMarinis,* W. M. Bryan, D. H. Shah, J. P. Hieble, and R. G. PendletonResearch and Developmen t Divis ion, Smi th Kl ine & French Laboratories, Philadelphia, Pennsylvania 19101.Received A pril 17, 1981

A aeries of pheny lethylamines re lated to methoxamine has been prepare d and evaluated for direct al-recepto r agonistactivity. It has been observed that for open-chain compounds suc h as methoxamine, n which the amine-containingportion is free to adopt numerous conformations, an hydroxyl group is necessary for direct al-adrenergic activity.When the hydroxyl is removed, however, the direc t component.of activity is greatly reduced unless the amine isincorporated into a more sterically defined structure. From our studies we have concluded that in order for aphenylethylamineto be activeas a direct al-receptor agonist it should have a @ nitrogen n a fully extended conformationrelative to a substituted phenyl ring. For optimum potency, the nitrogen should be exocyclic to a saturatedsix-mem bered ring. It may be fur ther inc orporated exocyclic or endocyclic into an additional ring so long as h eamine oc cupies a well-defined region of space relative to the a romatic portion of a molecule. Th e EDSOalues ofsome of th e more pote nt compounds as al-receptor agonists are on the order of 1 X lo-' M.The phenylethylamines are a well-studied class ofpharmacological agents which over the years have been t hesubject of numerou s investigations. In th e last decadethere have been many studies into the a-adrenergic effectsof a variety of phenylethylamines with rigid, as well asflexible, structuresH (and references therein). At the sam etime, muc h evidence has been accumulating which indi-cates tha t ct receptors ca n be subdivided into a t least twopharmacologically distinct classes which can be function-ally differentiated by their ability to interact with a seriesof agonists and ant ag on i~ ts. ~J ~n addition to the classical(1) This paper has been presented in part. See "Abstracts ofPapers", 181st National Meeting of the American ChemicalSociety, New York, Aug, 1981,American Chemical Society,Washington, DC, 1981,Abstr MEDI 12.(2) P. Burn, P. A. Crooks, B. L. Ratcliff, and J. M. H. Rees, J.Pharm. Pharmacol., 32,87 (1980).(3) G. L. Gruenwald, G. L. Ruth ,J. A. Kroboth, T. R. Kamdar, B.V. Patil, and K. N. Salman, J. Pharm. Sci., 65 , 920 (1980).(4)H. . Chang, . P. Long, D. E. Nichols, and C. F.Barfknecht,J . Pharmacol. E x p . Ther., 188,114 (1974).(5) D. B.Rusterholz, J. P. Long, . R. Flynn, nd J. R. Glyn,Arch.Znt.Pharmacodyn. Ther., 232,246 1978).(6)F. M. Sharabi,J. P. Long, D. B. Rusterho lz, and C. F.Barf-knecht, Res. Comm un. Chem. Pathol. Pharmacol., 19,l 1978).(7) J.G. Cannon, J. A. Perez, J. P. Pease, J. P. Long, J. R. Flynn,D. B. Rusterholz, and S. E. Dryer, J. Med. Chem., 23, 745(1980).(8)D. B.Rusterholz,S.E. Dryer, J. P. Long, C. F.Barfknecht, andJ. Mott, Eur. J. Pharmacol., 65 , 201 (1980).

posbynaptic a adrenoceptor (a1)hat mediates the re-sponses of effe dor organs to norepinephrine, there are alsoa receptors (az)n noradrenergic nerve terminals whichmodu late the release of norepinephrine in th e peripheryand the central nervous This subclassificationof a adrenoceptors opens new possibilities for drug dis-covery through t he d evelopm ent of agonists or antagonistshaving a high degree of selectivity for each receptor sub-type.Norepinephrine (l), he endogenous transm itter of thePH

IOH1 2 3

sympathetic nervous system, acta with almost equal po-tency on both a1 nd az receptor^.^ Other agents, such asmethoxamine (2) or phenylephrine (31, are much morespecific for th e a1 han for the ct2 receptor an d representattractive points of departure for the development of se-(9) S.Berthelsen and w. A. Pettinger, Life Sci . , 21, 595 (1977).(10) C. Melchiorre, Farmaco, Ed . Sci., 35, 535 (1980).(11)S. Z.Langer, Trends Neurosci., 3,110 (1980).(12) S.2.Langer, Br . J. Pharmacol., 60 , 481 (1977).(13) K. tarke, Rev. P hysiol., Biochem. Pharmacol., 7 7 , 1 (1977).

0022-2623/81/1824-1432$01,25/0 1981 American Chemical Society