nichols-naphtofurans

7/31/2019 nichols-naphtofurans

1/12

S u b s t it u t e d N a p h t h o f u r a n s a s H a ll u c i n o g e n i c P h e n e t h y l a m in e-E r g o l i n e

H y b r i d M o l e c u l e s w i t h U n e x p e c t e d M u s c a r i n i c A n t a g o n i s t A c t i v i t y

Aar on P. Monte, Danu ta Marona-Lewicka, Mechelle M. Lewis, Richar d B. Mailman, David B. Wainscott,

David L. Nelson, an d David E . Nichols*

Dep ar tm en t of M ed ici n al Ch em is tr y an d M olecu la r Ph ar m acol ogy , S ch ool of Ph ar m acy , Pu rd u e U n iv ers it y,

West Lafayette, Indiana 47907, Departments of Pharmacology and Psychiatry, Neuroscience Center, University of North

Carolina School of Medicine, Chapel Hill, North Carolina 27599, and Lilly Research Laboratories, Eli Lilly and Company,In d ia n ap oli s, In d ia n a 46 28 5

R eceiv ed Feb ru ar y 2, 19 98

A series of substitu ted r acemic napht hofur an s were synth esized a s hybrid molecules of thetwo major pr ototypical ha llucinogenic drug classes, th e phenet hylamines a nd t he t ryptam ines/ergolines. Although it was hypothesized that these new agents might possess high affinity forthe serotonin 5-HT2A/2C receptor subtypes, unexpected affinity for muscarinic receptors wasobserved. The compounds initially synthesized for this st udy were (()-anti- and sy n -4-amino-6-methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]furan (4a ,b), respectively, an d t heir 8-bromoderivatives 4c ,d , respectively. The brominated primary am ines 4c ,d were assayed initiallyfor activity in the two-lever drug discrimination (DD) paradigm in rats trained to discriminatesaline from LSD ta rt rat e (0.08 mg/kg). Also, 4c ,d were evaluat ed for their ability t o compete

against agonist and antagonist radioligands at cloned human 5-HT 2A, 5-HT2B, and 5-HT2Creceptors. After the syn diastereomers were found to have th e highest activity in th eseprelimina ry assa ys, the N-alkylated a nalogues sy n -N ,N -dimeth yl-4-amino-6-methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]furan (4e ) and sy n -N ,N -dipropyl-4-amino-6-methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]furan (4f) were prepared and assayed for their affinities at[3H]ketanserin-labeled 5-HT2A and [3H]-8-OH-DPAT-labeled 5-HT1A sites. All of th e moleculestested ha d r elatively low a ffinity for serotonin receptors, yet a preliminary screen indicatedtha t compound 4d ha d affinity for mu scarinic receptors. Thus, 4b ,d ,e were evaluated for theiraffinity at muscarinic M1-M5 receptors a nd also assessed for t heir functiona l chara cteristicsat the M1 and M2 isoforms. Compound 4d ha d a ffinities of 12-33 nM at all of the mu scarinicsites, with 4b ,e ha ving much lower affinity. All thr ee compounds fully an ta gonized the effectsof carbachol at the M1 receptor, while only 4d completely antagonized carbachol at the M 2receptor. The fact that the napht hofura ns lack LSD-like a ctivity suggests t hat they do notbind to the serotonin receptor in a way such that the t r icycl ic naphthofuran nucleus is

bioisosteric with, a nd directly superimposable upon, th e A, B, and C rings of LSD. This alsoimplies, ther efore, that the hallucinogenic pheneth ylam ines cannot be directly su perimposedon LSD in a common binding orientation for these two chemical classes, contrary to previoushypotheses.

I n t r o d u c t i o n

One of the most intriguing issues in exploring thest ructure-activity relationships (SAR) of p sychedelic(hallucinogenic) agents has been to determine howdifferen t chemical class es of ha llucinogens with similarphar macologic and behaviora l effects may act t hr ougha common receptor. Much of our previous work on theSAR of hallucinogens has been driven by our efforts todiscern the common pharmacophoric elements withinthe phenethylamine (e.g., 1) and tryptamine/ergoline(e.g., 2) classes a s they r elate to binding interactionswithin the serotonin 5-HT2 family of receptors, theputative primary si te of action of these drugs.1-3 Todat e, however, no entirely satisfactory model correlat es

the activity of th ese ha llucinogens with their uniquestructural features.

The m ost widely accepted ph ar ma cophoric hypoth esisis one in which th e ar yl ring of a phen ethylamine, suchas 1, is super imposed over th e A ring of LSD 2. Thi sidea was first developed in t he 1950s by Mar ini-Bettolo,Bovet, and co-workers as i t related to the oxytocicproperties of the ergolines.4 In this model, the proto-nated primary amine of1 an d N(6) of2 would bind to acommon anionic residue within the receptor bindingsite, now thought to be a conserved aspartate presentin transmembrane helix 3 of all characterized monoam-ine receptors. Indeed, Wang et al .5 have shown thatmutation of Asp155 to an Asn in the 5-HT2A receptor gavea mutant with greatly decreased affinity for 5-HT2agonists and anta gonists. This model also suggests tha tth e 5-meth oxy oxygen of1 and the indolic NH of2 mighthydrogen-bond to a common serine residue in thehuma n 5-HT2A receptor, possibly Ser 242 in TM III,6 whileth e 2-meth oxy oxygen at om of1 and the carbonyl oxygen

* Address correspondence to: Dr. David E. Nichols at PurdueUniversity. Phone: (765) 494-1461. Fax: (765) 494-1414. E -mail:drda ve@pha rm acy.purdu e.edu.

UNC School of Medicine. Eli Lilly and Co. Present address: Departmen t of Chemistry, University of Wis-

consin-La Crosse, La Crosse, WI 54601.

2134 J . M e d . Ch e m . 1998, 41 , 2134-2145

S0022-2623(98)00076-4 CCC: $15.00 1998 American Chem ical SocietyPublished on Web 05/01/1998


2/12

of2 could int era ct with a common H-bond donor with inthe 5-HT2 binding si te. This hypothesis is supportedby t he f i nd i ng t ha t i t i s t he R enant iomer of thephenyl ispropylamines that i s the more potent hal -lucinogen, and the asymmetric center in these agentsis homochiral with C(5) of the ergolines in t his su per-imposition.7,8 Lat er s t ud ies t ha t com pared ch i ra ltryptamines with phenethylamines gave data consistentwith th is notion.9

Recently, we have shown that constraining the meth-oxy groups of the 2,5-dimethoxy-substituted hallucino-genic pheneth ylamines into dihydrofur an moieties (asin 3) provided compoun ds of high p otency and selectivity

for 5 -HT2A/2C receptors th at exh ibited behavioral effectssimilar to 2 in the rat drug discriminat ion assay.10

Thus, 3 clearly established the binding conformation ofthe a lkoxy substituents at 5-HT2 sites in this class ofcompounds. As an extension of that work, and as ameans to test the pharmacophoric hypotheses discussedabove, i t was reasoned that an extremely rigidifiedphenethylamine-like species could be constructed bytethering the side chain of3 to the adjacent dihydro-furan r ing. This stra tegy would produce a molecule tha tis, in essence, a hybrid of molecules 1 and 2, in whichthe indolic A and B rings of 2 are mimicked by thebenzodihydrofuran portion of3 and the C ring of2 isequivalent t o the teth ered side chain of the n ew hybrid.

Thus, a series of compounds (4) was proposed asmethoxylated phenethylamine analogues that poten-tially would be bioisosteric with rings A, B, and C ofthe ergolines. Following an evaluation of ph ysicalmodels, molecular modeling studies were initially car-ried out to examine conformations of the syn isomers(e.g., 4b ,d ) and t o determine wheth er they were super-imposable on the A, B, and C rings of 2. E n er gyminimization us ing semiempirical meth ods (AM1 Ham il-tonian) gave low-energy conforma tional minima for 2and 4d . A least-squares fit of the framework atoms of4d to th e A/B/C rings a nd N (6) of2 gave a su perpositionwhere th e distance between an y two paired at oms wasgenerally less than 0.1 , with the exceptions of the

basic nitrogens and atoms C(2a) in both molecules,where th e corresponding dista nces between th em were0.44 an d 0.48 , respectively. Given th at a sm all degreeof conforma tiona l flexibility exist s in both molecules, webelieve t hat the fit between m olecules 4 and the A, B,and C rings of2 can be consider ed to be good. Her e wereport on the synthesis and pharmacological evaluationof both the anti (4a ,c ) and syn (4b ,d-f) racemic dia-stereomeric naphthofurans as phenethylamine-ergolinehybrid molecules.

Each of the compounds 4a-f synth esized in th is studywas a ssayed for LSD-like beha viora l activity in th e dru gdiscrimination model, a paradigm t hat has been cor-related with h allucinogenic activity in ma n.11 Radioli-

gand binding studies init ially were carried out usingboth r at bra in and cloned huma n 5-HT2 receptors. Aftera receptor screen demonstr ated th at 4d ha d affinity formuscar inic receptors, the syn diastereomers were evalu-a t ed a t cloned hum an M1-M5 muscarinic receptorsubtypes.

C h e m i s t r y

The series of target racemic tetrahydronaphthofurans4a-f was synthesized successfully as i l lustrated inSchemes 1 an d 2. The early stages of the synth esis ar epresented in Scheme 1. The acyl chloride of 2,5-dimethoxybenzoic acid (5) was converted to the diaz-omethyl ketone and efficiently cyclized to the benzo-

furanone 6 using the method of Jung and Abrecht.12This method provided an excellent means of producinglarge quan tities of6. Following the procedur e of Chanet al.,13 6 underwent Witt ig r eaction with the ylide,meth yl (triphenylphosphorylidene)acetat e. These au-thors claimed tha t for t he an alogous Wittig addu ct usingthe ethyl ester ylide, the mixtur e of dihydrobenzofuran(exo double bond) and benzofuran products could beisomerized to the more stable benzofuran isomer bystoring the mixtur e in chloroform overnight. While asimilar mixture of products (7a ,b) was obtained usingthe methyl ester ylide, isomerization to th e reportedmore stable benzofuran 7b did not always proceed tocompletion. Nonetheless, the benzofuran alone, or themixtur e of isomers, could be r educed cata lytically in th enext st ep to a fford th e 2,3-dihydrobenzofur an ester 8 .Basic hydrolysis then gave the benzofuranylacetic acid9 in excellent yield.

Homologation of the side chain of 9 by one carbon,and formation of the third ring, was accomplished byinitially forming diazometh yl ketone 10 . Cyclization to11 was accomplished by reflux of10 in dichloromethanewith a catalytic amount of rhodium(II) acetate dimer,followed by tr eat men t with tr ifluoroacetic acid, accord-ing to th e met hod of McKervey et a l.14 Because therewas only one possible aromatic insertion point in 10, i twas a nticipated tha t this cyclization would proceed in

good yield. Neverth eless, pur e napht hofur an one 11 wa sobtained only in modest amounts (30-35% yields).Therefore, a n alternate cyclization was attempted inwhich trifluoroacetic acid (TFA) was used to protonatethe d iazometh yl ketone a nd t o cata lyze intra molecularat tack of the arene on the diazonium moiety wi thconcomitant displacement of N2.15 Cushman et al .16,17

had previously used t his cyclization meth od in a morecomplex system with some success, attaining yields ofup to 37%. When 10 was treated with TFA, however,the only product isolated (55%) arose from direct nu-cleophilic displacement of th e diazo group by tr ifluoro-acetate anion. Similar discoura ging r esults were ob-tained when m etha nesulfonic acid was used in place of

S u b st it u ted N a ph t h of u ra n s a s H yb ri d M ol ecu l es J ou r na l of M ed i ci n al C hem i st ry , 1 99 8, V ol . 4 1, N o. 1 2 2135


3/12

TFA. It was concluded th at the rhodium-catalyzedcyclization approach gave the best results for thissystem, and this method was used to produce ketone11 . Reductive amination of11 with benzylamine in the

presence of sodium cyanoborohydride readily gave thediastereomeric mixture of ant i- and sy n -N-benzyltet-rahydronaphthofurans 12 a ,b , respectively.

Fina l elaboration of the 4-aminonaph thofuran nucleuswas carried out a s i llustrated in Scheme 2. The dia-ster eomeric 1:2.4 mixtur e ofant i-12 a :sy n-12 b N-benzylproducts was separ ated by column chr omatography. Theabsorptions and coupling constan ts of al l h ydrogenswere a ssigned following 2D-NOESY NMR experiment s.The racemic diastereomers were differentiated by ob-serving the pr esence or a bsence of a nuclear Overha usereffect (NOE) between hydr ogens H (2a) an d H(4). The2D-NOESY NMR spectrum of the syn compound 12 bclearly showed a small cross-peak between these pro-

tons, but t his coupling was completely absent in th e 2D-NOESY spectrum of 12 a , indicating th at the protonswere on opposite faces of the plan ar rin g system in th isdiastereomer. These init ial NMR experiments wereconfirmed later by compar ing th e 2D-NMR spectra ofthe subsequent diastereomeric pairs (4a with 4b ; 4cwith 4d ). The coupling constan ts for the methylenehydrogens at C(3) in 4a ,b were also diagnostic. In 4bH(3)a is pseudoaxial, observed as a quartet at 1.19,with dihedral angles to both H(2a) and H(4) of ap-proximat ely 175, giving lar ge and ess ent ially ident icalvicina l coupling const an ts. The gemina l coupling con-

stant to H(3)eq is large and also virtually identical tothe two vicinal coupling constan ts. Thus, H(3)a iscoupled to three protons with nearly identical couplingconstants. In 4a no clear ant i relat ionships existbetween vicinal h ydrogens, with the dihedral anglesbetween H(3)a-H(2a) and H (3)a-H(4) of 139 an d 29,respectively, and between H(3)eq-H(2a) and H(3)eq-H(4) of 22 an d 145, respectively, with the resonan ceat 1.45 observed as a doublet of quar tets.

Once the stereochemistry of the pure diastereomers12 a ,b was establ ished, each was carried through aseries of an alogous rea ctions as shown in Scheme 2 toafford the desired target compounds 4a-d . Bot h t he

catalytic hydrogenolysis to give primary amines4a

,b

and the subsequent aromatic brominations affording4c ,d , proceeded sm ooth ly for each dias ter eomer, givingth e desired tar get molecules in good yield. Fina lly, th esy n-tetrahydronaphthofuran 4b was converted to i ts

N,N-dimethyl (4e ) a n d N ,N-dipropyl (4f) congenersusing previously described meth ods with the appr opri-ate aldehydes and sodium cyanoborohydride.18 Afterchromatographic purification, oxalate salts of both 4e ,fwere prepared, and these were crystallized from etha-nol. The intent of preparing these compounds was toascertain whether N-alkylation might ei ther abolish5-HT2A receptor affinity or lead to ligands with higheraffinity for the 5-HT1A receptor.

S c h e m e 1a

a Reagents: (a) 1. ClCOCOCl, benzene, DMF, 2. CH2N2, (CH 3CH 2)2O; (b) CH 3COOH; (c) Ph 3PCHCOOCH 3, xylene, ; (d) 10% Pd-C/H2, CH 3CH 2OH; (e) KOH, CH 3CH 2OH , ; (f) 1. ClCOCOCl, benzen e, DMF, 2. CH 2N2, (CH3CH 2)2O; (g) 1. Rh2(OAc)4, CH 2Cl2, , 2. CF3COOH;

(h) benzylamine, benzene, 2. Na CNBH 3, CH 3CH 2OH .

S c h e m e 2a

a Reagents : (a) 10% Pd-C/H 2, CH 3OH; (b) Br 2, CHCl3; (c)HCHO, NaCNBH 3, CH 3OH; (d) CH 3CH 2CHO, NaCNBH 3, CH 3OH .

2136 J ou rn al of M ed icin al Ch em istry, 1998, V ol. 41, N o. 12 M on te et al.


4/12

P h a r m a c o l o g y

Compounds 4c ,d initially were screened for in vivoactivity using th e two-lever dr ug discrimina tion assa yin rat s tra ined to discriminate saline from LSD tart ra te(0.08 mg/kg). Subsequent ly, the n ew compounds wereassessed for affinity at the serotonin 5-HT 2 family ofreceptors. The sy n -naphthofuran 4d , shown to havehighest affinity in the serotonin receptor assays, thenwas su bmitted t o NovaScreen to assess its a ffinity forother monoamine r eceptor subtypes. Following identi-ficat ion of possible high a ffinity for mus carin ic recept orsthrough this screen, compounds 4c ,d were tested fortheir affinity at t he cloned M1-M5 mus carinic receptors.These agents also were evaluated functionally by a s-sessing the abil ity of the compounds to st imulate

phosphoinositide hydrolysis in mus carinic M1 and M2receptor system s.

R e s u l t s a n d D i s c u s s i o n

D r u g D i s c r i m i n a t i o n a n d S e r o t o n i n R e c e p t o r

A ffinity. Table 1 shows th e dru g discriminat ion da tafor compounds 4c ,d , and includes compar ison da ta tha twe had obtained previously for compounds 1 and 3.Clearly, the rigid naphthofurans have lost the LSD-likebehavioral effects that are characteristic of compounds1 and 3. This result was surpr ising because we hadpreviously shown t ha t LSD-like beha viora l poten cy wasincreased in the rigid analogue 3 compared with 1.Based on th e h ypothesis tha t the hallucinogenic am-

phetamines superimpose over the ergoline structure,with correspondence between the phenyl ring of theamph etam ines an d ring A of the ergolines, it h ad beenanticipated that 4d might have enhanced LSD-likeeffects, relat ive to 3. Although design of rigid analoguescan be problema tic, an d data for inactive stru ctur es canbe difficult to inter pret , based on our exper ience to date,it seemed the most likely explanation for the lack ofactivity was simply that the hallucinogenic amphet-amines do not bind to th e 5-HT2A receptor in a confor-mat ion th at r esembles the ergolines, as described above,and as embodied within the structure of4d .

The two diastereomers 4c ,d then were assessed fortheir a ffinity at [3H]ketan serin-labeled 5-HT2A receptors

in rat br ain homogenate (Table 2). Their affinities werevery low, especially when compared with compoundssuch as 1 and 3. Nonetheless, these data did establishthat the syn diastereomer 4d had greater complemen-tar ity to the 5-HT2A receptor than the anti diastereomer.

Table 3 shows radioligand competition data for com-pounds 4c ,d , and compar ison da ta for 3, in the clonedhuma n 5-HT2 receptor family. From the affinity data,it can be clearly seen th at the syn compound 4d againgives a markedly bet ter fi t to the receptor. Whencompar ed with 3, however, 4d has approximately 20-30-fold less affinity a t th e an ta gonist -labeled 5-HT 2A and5-HT2C receptors and a 164-fold lower affinity for the5-HT2B receptor. These data ar e consistent with t heobserved loss of behavioral activity in the drug discrimi-nat ion a ssay a nd r einforce the conclusion t ha t 4d doesnot repr esent a n active conform at ion for ha llucinogenic

amphetamines such as 1.Although naphthofuran 4d has low affinity for theant agonist-labeled 5-HT2A receptor, it does have sig-nificant affinity for the agonist-labeled high-affinitystates of both the 5-HT2A and 5-HT2C human recep-tors.1,19 The absence of an LSD-like interoceptive cueleads u s t o speculate th at 4d may be a partial agonistor an tagonist, since all full agonists with high a ffinityfor [125 I]DOI-labeled 5-HT2A receptors that we haveexamined so far have possessed LSD-like behavioralactivity in th e drug discriminat ion par adigm. Since wehave previously elucidated the active conformations ofthe methoxy groups for 1 ,10 these resul ts lead us toconclude that the active conformation of the hallucino-

T a b l e 1 . Data from Drug Discrimination Tests in Rats Trained To Discriminate Saline from LSD Tartrate (0.08 mg/kg ip)

E D50

com pd mg/kg mol/kg N %Da %SDLb mol/kg m g/kg

LSD (2) 0.01 0.023 13 0 31 0.037 0.0160.02 0.046 14 7 62 (0.023-0.057) (0.01-0.025)0.04 0.093 15 0 810.08 0.186 16 0 100

1 c 8-12 1.12(0.086-1.46)

3 c 7-15 0.061

(0.031-0.12)4c 0.365 1 7 0 0

0.78 2 9 0 111.46 4 10 0 20 NS NS2.92 8 11 0 275.84 16 9 11 37.5

4d 0.365 1 10 0 00.78 2 11 0 01.46 4 10 0 10 NS NS2.92 8 12 0 335.84 16 13 15 44

a Percentage of animals failing to emit 50 presses on either lever within the 15-min test period. b Percentage of animals emitting 50presses on the LSD appropriate lever during the test period. c Data from Monte et al.10

T a b l e 2 . Results of the Radioligand Competition Stu dies of4c-f at 5-HT2A and 5-HT1A Receptors in Rat Brain

Homogenate (Ki values(

SEM)

drug[3H]ketanserin-labeled

5-HT2A sites (nM)[3H]-8-OH-DPAT-labeled

5-HT1A sites (M)

1 a 22 ( 3 6.1 ( 0.53 a 18 ( 1 4.3 ( 0.44c 3880 ( 170 NT4d 1050 ( 170 NT4e 967 ( 124 38 ( 64f 644 ( 80 144 ( 10

a Data from Monte et al.10



5/12

genic amphetamines may n ot be one where the eth y-lamine side chain l ies in an approximately coplanararra ngement with t he aromatic system. It also seemspossible tha t side-chain flexibility m ay be required toaccommodate the dynamics of receptor activation forpheneth ylamine type agonists, but n ot ergolines. Thisconclusion would a lso imply, however, t ha t t he classicalview of the relationship between th e ergolines an d th ebinding conformation of the phenethylamine hallucino-gens is incorrect.

Following these investigations, the nonbrominatedsy n-N,N-dimeth yl (4e ) and sy n -N,N-dipropyl (4f) naph-thofura ns were prepar ed and as sayed for affinity at the

5-HT1A and 5-HT2A receptors in rat brain homogenate(Table 2). The results show that n either of these ta rgetnaphth ofuran s have significant affinity for [3H]ket-ans erin-labeled 5-HT2A sites or [3H]-8-OH-DPAT-labeled5-HT1A sites. It is curious to note, however, that theaffinity of the syn diastereomers for 5-HT2A sitesincreases slightly with N-alkylation and with exten sionof th e N-alkyl groups from methyl to propyl groups, anobservat ion that i s contrary to the known SAR ofphenethylamine type 5-HT2A agonists.

E ffects at Muscarinic R eceptors. One un expectedfindin g resulted from a r adioreceptor screen of th e mostactive sy n -bromonaphthofuran 4d . This compound hadsignificant affinity for muscarinic receptors, a surprising

findin g in light of the fact th at 4d did not resemble anyknown mu scarinic ligands.20-23 Although new m usca-rinic ligands have been developed th at can cross theblood-brain ba rr ier, many of the pr ototypical muscar -inics possess a quat erna ry amm onium functionality tha tprevents thei r penet rat ion into the central nervoussystem (CNS).24 Muscarin ic agonists with CNS a ctivityare current ly highly sought, since it is now known t hatpatients suffering from Alzheimers disease exhibitselective degeneration of the muscarinic M2 subtypereceptors in the poster ior pa rieta l cortex.24 Because t hemolecular and three-dimensional structures of musca-rinic receptors ar e believed to resemble closely those ofthe other GPCRs,23 i t is possible that compounds

designed as ligands for a different receptor type mayactually have affinity for any one of the muscarinicreceptor su btypes. With this in mind, the sy n -naph-thofurans 4b ,d ,e were su bjected to compreh ensive ra-dioligand competition and functional studies at all fiveof the muscar inic receptor subt ypes, M1-M5.

The receptor binding results obtained for 4b ,d ,e atthe five cloned huma n m uscarinic subtypes are sh ownin Table 4. In agreement with th e preliminar y NovaS-creen data, compound 4d has significant affinity forseveral of the mu scarinic receptor subtypes. In addi-tion, an importa nt an d striking structura l featur e is thehydrophobic aromatic substi tuent para to the alky-lamine side chain (compare 4b ,d ) that increases r ecep-

tor affinity. This subst i tuent i s also an importantrecognition element at the 5-HT2A receptor. Thus, theremay be a region of hydrophobic interaction within th emuscarinic binding si te analogous to the one that isproposed to exist in th e serotonin 5-HT2A receptor. Theresults also show tha t N-substitution great ly attenuat esligand affinity for all subtypes, a structural feature thatis divergent from the SAR of most other muscarinicreceptor ligands.

Figures 1 a nd 2 i l lustra te tha t 4b ,d are a ntagonists

a t t he M1 a n d M2 receptors, respectively. While th eaffinity of 4d for the various muscarinic r eceptorsubtypes is lower th an tha t of atr opine, it is similar toseveral currently available, hydrophobic M2 agonistsand antagonists.20,21 Furthermore, whi le i t has ap-proximat ely one-tenth the affinity of atr opine, it is aneffective muscarinic anta gonist, at least a t th e M1 andM2 subtypes. As a potential muscarinic ligand, 4drepresen ts a compound of novel molecular str uctur e tha tis of sufficient hydr ophobicity t o cross th e blood-brainbarr ier and enter th e CNS. Specifically, 4d may serveas a valuable lead compound in the future developmentof mus carinic ligands a nd gener ally may lead to a betterunder sta nding of the SAR of muscarinic receptors.

T a b l e 3 . Results of the Ra dioligand Competition Stu dies of4c ,d at Cloned Huma n 5-HT2A, 5-Ht 2B, and 5-HT2C Receptors (Ki valuesin nM ( SEM)

[3H]a ntagonist ra dioligands agonist ra dioliga nds

compdketanserin

5-HT2Arauwolscine

5-HT2Bmesulergine

5-HT2C[125 I]DOI

5-HT2A[3H]-5-HT

5-HT2B[125 I]DOI

5-HT2C

3 a 10.7 ( 1.0 1.15 ( 0.04 2.28 ( 0.38 0.48 ( 0.03 1.60 ( 0.25 0.30 ( 0.024c NT NT NT 8500 ( 150 11200 ( 1800 4190 ( 1204d 240 ( 16 255 ( 6 89.6 ( 21.1 13.0 ( 1.6 263 ( 32 5.96 ( 0.62

a Data from Monte et al.10

T a b l e 4 . Results of Radioligand Competition Stu dies atCloned Muscarinic Receptors (K0.5 values in nM ( SEM)

drug M1 M2 M3 M4 M5

4b 149 ( 9 33400 ( 5 129 ( 6 240 ( 25 236 ( 224d 20 ( 8 41 ( 18 12 ( 3 30 ( 8 33 ( 54e 846 ( 26 953 ( 74 654 ( 56 913 ( 42 1100 ( 50atropine 0.6 ( 0.1 1.8 ( 0.2 0.6 ( 0.1 0.5 ( 0.1 0 .7 ( 0.1

F i g u r e 1 . Functional activity of4 derivatives at the clonedM1 muscarinic receptor. B82 cells tranfected with the M 1receptor were incubated with carbachol (Carb; 1 mM) orcarbachol plus drug. Atropine (1 M) was a ble to inhibit t hecarbachol-stimulated response. Compounds 4b ,d ,e tested at10 M ha d l i t t le a c t iv i ty on the i r own but we r e a ble toa t te nua te the ca r ba chol r e sponse to a s im i la r degr ee a satr opine. Values represent t he mean ( SEM for du plicates runat thr ee different t imes.



6/12

Fr om a molecular s tan dpoint, 4d was designed to bea hybrid of the ha llucinogenic pheneth ylamines, exem-plified by DOB (1) or the rigid congener 3, a n d t h eergolines, exemplified by LSD (2). Because 4d is lesspotent than 1 or 3 and appears to lack the LSD-likeactivity th at is chara cteristic of the latter two molecules,one hypothesis to be drawn from these data is that 4ddoes not bind to th e serotonin r eceptor in a way thatthe tr icyclic na phth ofura n nucleus is bioisosteric with,an d directly superimposable upon, the A, B, an d C rings

of LSD. This also implies, therefore, that the hal -lucinogenic pheneth ylamines cannot be directly super-imposed on LSD in a common binding orienta tion forthese two chemical classes, contra ry to previous hy-potheses . The loss of activity for 4d relative to 1 and 3also suggests that the flexible aminoalkyl side chain inthe ph eneth ylamines does not adopt a conform ation tha tis relatively coplana r with t he a romatic ring system, asis the case with the ergolines. While the present datado not absolutely prove such an assertion, i t is theconclusion most parsimonious with our results. If true,one could infer that the agonist s i te of the 5-HT 2Areceptor accommodates either the trypta mine/ergolineor the oxygenated ph enethylamine t emplates in differ-

ent bindin g orient at ions, in which at least some distinctamino a cid r esidues a re involved in the recognition ofeach molecular species.

E x p e r im e n t a l S e c t i o n

C h e m i s t r y . M elt ing poin ts wer e dete r m ine d us ing aThomas-Hoover apparatus and are uncorrected. 1H NM Rspectra were recorded using either a 500-MHz Varian VXR-500S or 300-MHz Bruk er ARX-300 NMR spectr ometer. Chemi-cal sh ifts are reported in values (ppm) relative to tet rame-thylsi la ne (TMS ) a s a n in te r na l r e fe r e nce (0 .03% v/v).Abbreviations used in NMR analyses are as follows: s )singlet, d ) doublet, t ) triplet, dd ) doublet of doublets, dt )doublet of triplets, td ) triplet of doublets, q ) quartet , p )pentet, m ) multiplet, b ) broad, Ar ) aromatic. Chemical

ionization m ass spectra (CIMS) using meth ane a s th e carriergas were obtained with a Finnigan 4000 spectrometer . IRmeasurements were taken with a Perkin-Elmer 1600 seriesFTIR spectrophotometer. Element al analyses were performedby the P urdu e University Microanalysis Laborat ory and werewithin (0.4% of th e calculat ed values u nless other wise noted.Thin-layer chromatography (TLC) was typically performedusing Baker-flex silica gel IB2-F, plastic-backed plates withfluorescent indicator (2.5 7.5 cm; J . T. Baker), eluting withCH 2Cl2, and visualiz ing with UV light a t 254 nm and/or I2vapor unless oth erwise noted. Plat es used for radial centr ifu-gal chromatograph y (Chromat otron; Ha rrison Research, PaloAlto, CA) were pr epar ed from silica gel 60 PF2-54 cont ainin ggypsum. Reactions were carried out un der an a tmosphere ofdry nitr ogen.

5-Methoxybenzofuran-3(2H)-one (6).25 Oxalyl chloride(36 mL, 0.41 mol) was added dr opwise over 1 h to a stirredsusp ension of 57.5 g (0.32 m ol) of 2,5-dimeth oxybenzoic acid(5) in 250 mL of benzene an d drops of DMF. As the rea ctionprogressed, th e acid grad ua lly dissolved to give a clear-yellowsolution. After 2.5 h, volatiles were removed on th e rotaryevaporat or, and th e acyl chloride was dist illed in a Ku gelrohrapparat us (bp 105-110 C at 0.05 m mH g) to give 61.3 g (97%)of a pa le-yellow oil. The pur e acyl chloride was u sed portion-wise in the following diazomethane reaction.

Using a large diazomethan e appar atu s (Aldrich, Milwauk ee,

WI), diazomethane was generated by adding a solution of 38g (0.18 mol) ofN-methyl-N-nitroso-p-toluen esulfona mide (Di-azald) in 250 mL of anhydrous ether to a stirred solution of10 g of KOH, 17 mL of H 2O, 50 mL of 2-(2-ethoxyethoxy)-ethan ol (carbitol), and 15 mL of ether held over a 40 C oilbath.26,27 The distilled ether-diazometha ne was cooled in anice-salt bath and stirred while 11 g (0.055 mol) of the 2,5-dimeth oxybenzoyl chloride (see above) in 30 mL of ether wascautiously added over a 10-min period. After the vigorousevolution of HCl gas had subsided, the ice-sa l t ba th wa sremoved, and the mixture was allowed to warm to roomtemperat ure while stirring in the hood. Remaining volatileswere removed by attaching the reaction flask directly to awater as pirator in th e hood and immer sing the flask in a war mwater bath . The intermediate diazoketone was obtained as abright-yellow solid caked to the walls of the flask.

Glacial acetic acid (70 mL) th en was a dded cautiously tothe intermediate diazoketone, causing the solid to dissolve withthe vigorous evolution of nitrogen. The mixture was stirredfor 1.5 h, and t he solvent wa s removed on t he rota ry evapora-tor . Furth er drying under h igh vacuum produced a yellowsolid that was recrystallized from ethyl acetate-petroleumeth er t o give 7.3 g (82%) of6 as fluffy, pale-yellow flakes: mp92 C (lit.25 mp 8 8 C); 1H NMR (CDCl3) 3.80 (s, 3, ArOCH3),4.64 (s, 2, ArOCH2CO), 7.06 (d, 1, ArH, J ) 2.5 H z), 7.07 (d, 2,ArH, J ) 9.0 Hz), 7.25 (dd, 1, ArH, J ) 9.0, 2.8 Hz).

Methyl 2-(5-Methoxybenzofuran-3-yl)acetate (7a).25 Fol-lowing the procedure of Chan et al.,13 a m ixtur e of 20.0 g (0.122mol) of ketone 6 and 41.54 g (0.124 mol) of methyl (triph-enylphosphoran ylidene)acetat e (Aldrich, Milwau kee, WI) wasstir red vigorously in 400 m L of toluene a t r eflux for 50 h. Thesolvent was then removed on the rotary evaporator, and the

r e sidue wa s d issolve d in a la r ge qua nt ity of e the r . Thetriphen ylphosphine oxide that did not dissolve was filtered anddiscarded, and a dditional oxide was removed by allowing it t ocrystallize out of the ether solution in the freezer withcollection by vacuum filtration. Final purification was ac-complished by repeat ed flash column chromatograph y, elutin gwith 10% ethyl acetat e-petroleum ether until the product wasfree of triphenylphosphine oxide and appea red as a single spoton TLC. The f irst band to e lute f rom t he columns was amixture of predominan tly 7a and its isomer 7b (assignmentsbased on 1H NMR spectra ) as a yellow solid th at weighed 15.1g (68%). The mixture was stirr ed in a solution of CHCl3 a troom t emperatu re for 12 h to isomerize the product mixtureto the more stable isomer 7b th at wa s a yellow oil: bp 103 Cat 0.01 mmHg; 1H NMR (CDCl3) 3.70 (s, 2, CH2COOCH 3),3.75 (s, 3, CH 2COOCH3), 3.83 (s, 3, ArOCH3), 6.90 (dd, 1 , ArH,

F i g u re 2 . Functional assessment of 4 derivatives a t thecloned M2 mu scarinic receptor. B82 cells tran sfected with th eM2 receptor were stimu lated with forskolin (FSK; 10 M), andthen cyclase activity was inhibited by coincubation withcarbachol (FSK + Carb (1 mM)). Addition of atr opine (1 M)was able to r everse t he carbachol-induced inhibition of FSKstimu lation. The new compound s ha d no effect by themselves,bu t 4d is able to reverse totally the carbachol-induced inhibi-tion, while 4b ,e are able to reverse this effect only partially.Values represent the mean ( SEM for duplicates run a t t wodifferent times. All 4 analogues were tested at 10 M.



7/12

J ) 8.9, 2.5 Hz), 7.00 (d, 1, ArH, J ) 2.1 Hz), 7.36 (d, 1H,ArH, J ) 8.9 Hz), 7.60 (s, 1H, ArCCH); CIMS m/z 221 (M +1). The second ban d to elute from th e column contain ed 3.5 gof unreacted 6 tha t could be recycled for su bsequent r uns.

Methyl 2-(2,3-Dihydro-5-methoxy-3-benzofuranyl)ac-etate (8).28 A solution of 4.7 g (0.021 mol) of the Witt ig addu ct7b was sh aken in 55 mL of absolute ethan ol containing 750mg of 10% Pd-C in a Par r hydrogenation apparat us at 50 psiof H 2. After 34 h, the th eoretical amount of H 2 had been takenup, and the reaction mixture was f i l tered through Celite .Solvent removal un der r educed pressu re gave 4.4 g (92%) of8as a clear oil that was sat isfactorily pure by TLC and 1H NMRanalyses: 1H NMR (CDCl3) 2.60 (dd, 1, CH2COOCH 3, J )16.6, 9.4 Hz), 2.80 (dd, 1, CH2COOCH 3, J ) 16.5, 5.3 H z), 3.70(s, 3, COOCH3), 3.75 (s, 3, ArOCH3), 3.82 (m, 1, ArCH), 4.22(dd, 1, ArOCH2, J ) 9.2, 6.4 Hz), 4.73 (t, 1, ArOCH2, J ) 8.9Hz), 6.68 (dd, 1, ArH, J ) 8.6, 2.4 Hz), 6.71 (d, 1, ArH, J ) 8.6Hz), 6.74 (d, 1, ArH, J ) 2.4 Hz).

(2,3-Dihydro-5-meth oxy-3-benzo furanyl)acetic Acid (9).A solution of 4.2 g (0.019 mol) of the reduced methyl ester 8in 20 mL of ethanol was added to a solution of 2.5 g of KOHin 5 mL of H 2O and 20 mL of etha nol. The mixture was stirr edove r a s te a m ba th f or 30 m in , a nd the a lc ohol the n wa sremoved un der reduced pressure. The residue was dilutedwith 30 m L of wa te r a nd cooled on a n ice ba th . C old,concentrated HCl was added, and the acidic mixture wasextracted with 3 20 mL of CH

2Cl

2. The organic extracts

were combined, washed once with brine, dried over MgSO4,and filtered thr ough Celite. Solvent r emoval on the rotaryevaporator gave 3.82 g (97%) of a white solid that was purifiedby recrystallization from ethyl acetate-eth er t o afford 3.38 g(86%) of9 as hea vy whit e cryst als: mp 114 C (lit.28 mp 111-112 C); 1H N MR (DMSO-d6) 2.50 (dd, 1, CH2COOCH3, J )16.7, 9.4 Hz), 2.78 (dd, 1, CH2COOCH 3, J ) 16.7, 5.3 H z), 3.66(s, 3, ArOCH3), 3.70 (m, 1, ArCH), 4.12 (dd, 1, ArOCH2, J )8.8, 7.2 Hz), 4.64 (t, 1, ArOCH2, J ) 8.9 Hz), 6.65 (s, 2, ArH),6.85 (s, 1, ArH); CIMS m/z 209 (M + 1), 191, 149.

1-Diazo-3-(2,3-dihydro-5-methoxy-3-benzofuranyl)pro-pan-2-one (10). The a cyl chloride was prepared by slowlyadding 1.37 mL (0.016 mol) of oxalyl chloride to a stirredmixture of 2.5 g (0.012 mol) of the acid 9 in 20 mL of benzeneand 1 dr op of DMF. After st irring at r oom t emperatur e for 1

h, the acid had completely dissolved and the solution wasorange in color. The volatiles were removed u nder r educedpressu re t o give, quan tita tively, 2.7 g of a clear-yellow oil thatcrystallized to a yellow solid on further drying under highvacuum.

The acyl chloride was dissolved in 30 mL of anhydrous etherand was cautiously added to a stirred, ice-salt bath-cooledsolution of ethereal diazomethane that had been prepared anddistilled from 8.3 g (0.038 mol) of Diazald (Aldrich, Milwauk ee,WI), 15 mL of carbitol, 2.5 g of solid KOH, 4 mL of H 2O, and100 mL of anh ydrous ether . Once addit ion was complete, theice-salt bath was removed, and the mixture was allowed towarm t o room temperatur e over the next hour. The volatileswer e the n r e m oved unde r r e duce d pr essur e to g ive thediazoketone 10 as a pale-yellow solid th at weighed 2.71 g (98%)and could be used without further purification. An analytical

sample was prepared by r ecrystallization from ether-

petro-leum eth er to give fine yellow needles: mp 73-74 C; 1H NMR(CDCl3) 2.74 (m, 2, CH2COCHN 2), 3.75 (s, 3, ArOCH3), 3.90(m, 1, ArCH), 4.20 (dd, 1, ArOCH2, J ) 9.1, 6.0 Hz), 4.70 (t, 1,ArOCH2, J ) 9.0 Hz), 5.25 (s, 1, CH 2COCHN 2), 6.73 (m, 3,ArH) ; CIMS m/z 233 (M + 1), 205, 161, 149. Ana l .(C12H 12O3N 2) C, H, N.

6-Methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]fu-ran-4-one (11). Two grams (8.61 mmol) of the diazoketone10 was dissolved in 150 mL of CH 2Cl2 and added dropwiseover 2 h to a suspension of 5 mg (0.132 mol %) of Rh 2(OAc)4catalyst in 150 mL of CH 2Cl2 at r eflux. After addition wascomplete, th e mixtur e was held at r eflux for an addit ional 1.5h, and then 1 drop of tr ifluoroacetic acid was added. Themixture was stirred a nd a llowed to cool to room temperatu reover 1 h. The cooled organic solution was wash ed with 150

mL of H 2O, 2 150 mL of 5% NaHCO 3, and 150 mL of brine,dried over MgSO4, and filtered through a pad of Celite andsilica gel. Removal of solvent un der redu ced pressur e yielded1.25 g (72%) of a yellow oil tha t appeared to be nearlyhomogeneous on TLC. The crude product was pur ified bycentr ifugal r adial chromatography (Chromatotron; H arrisonResearch, Palo Alto, CA) on a 4-mm silica gel plate, elutingwith 15% ethyl acetat e-hexane, to afford 600 mg (34%) of11a s a yel low oi l. An a na lyt ica l sa m ple wa s pr e pa r ed bycrystallizing a small portion from ether-hexan e to give a waxy,off-white solid: mp 83 C; 1H NMR (CDCl3) 2.36 (t, 1,ArCHCH2CO), 2.90 (dd, 1, ArCHCH2CO), 3.30 (d, 1, ArCH2-CO), 3.62 (d, 1, ArCH2CO), 3.78 (s, 3, ArOCH3), 3.82 (m, 1,ArCH, methine), 4.15 (t, 1, ArOCH2), 4.88 (t, 1, ArOCH2), 6.65(dd, 2, ArH); IR (neat) (CdO) 1711 cm-1; CIMS m/z 205 (M +1). Anal. (C12H 12O3) C, H.

a n t i -N-Benzyl-4-amino-6-methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]furan (12a) and syn-N-Benzyl-4-amino-6 - m e t h o x y - 2 a , 3 , 4 , 5 - t e t r a h y d r o - 2H- n a p h t h o [ 1 , 8 -b c ] f u -ran (12b). In a typical procedure for the preparation of adiastereomeric mixture of the N-benzylamines, th e na phth o-furanone 11 (1.1 g, 5.40 mmol) was stirred for 7 h under anitr ogen at mospher e in 20 mL of benzene cont ainin g 0.58 mL(5.40 mmol) of benzylam ine a nd 0 .5 g of 4- molecula r sieves .Reaction completion wa s indicated by th e disappear ance of th ecarbonyl IR band at 1711 cm-1. The dark mixture was filtered

through Celite, and the solvent was removed on the rotaryevaporator t o afford a red oil. Drying under high vacuumovernight gave the enamine as a red solid. The enamine wasthen t aken u p into 15 mL of absolute ethanol and stirred un derN 2, while anhydrous 1 N H Cl in ether was added via syringeto bring the reaction mixture to about pH 4 on moist pH pa per.Sodium cyanoborohydride, 0.28 g (4.04 mmol), was added, andthe mixture was stirred at room temperature for 5 h, addingdr y HC l-e t he r a s n e ed ed t o m a i n t ai n a bou t p H 5 (a p -proximat ely 6 mL of 1 N HCl-ether was u sed.) The volatileswe r e r e m ove d unde r r e duc e d pr e ssur e a nd the da r k r e dresidue was taken u p and sh aken gently between 2.5 N KOHand ether. The layers were separated, and the aqueous phasewas extracted with 3 20 mL of eth er. The organic fractionswere combined, washed with 3 20 mL of H 2O and 50 mL ofbrine, dr ied (MgSO4), and filtered through Celite. Solvent

removal under r educed pressure gave a dark -brown oil. Theoil was t aken up into dry ether , an d the hydrochloride saltwas formed by the addition of 1 N HCl in anhydrous ethanol.Removal of the solvent on t he rotar y evaporat or gave a dar ksolid that was t ritur ated with acetone-ether and collected ona Buchn er funn el, a procedure t ha t r emoved much of th e darkcolor. The ta n-gray solid collected was a m ixtur e of diast er-eomeric N-benzylamine hydrochlorides 12 a ,b that weighed0.90 g (50%).

The crude products from this a nd other run s were pur ifiedon the Chromat otron. In a typical separ at ion, 1.5 g of a crudemixture of12 a ,b hydrochlorides was treated with excess 5 NKOH, and the free bases were extracted into 4 25 mL ofCH 2Cl2. The organic extracts were combined, washed withbrine, dried over MgSO4, and filtered th rough Celite. Afterconcentration on the rotary evaporator , the residue was

dissolved in eth yl acetate, applied to a 4-mm silica plate, an deluted with 15% ethyl acetate-hexane under an ammoniaatmosphere. The first band t o elute from the Chromat otronplate was an ti isomer 12 a , the m inor product wh ose identitywas confirmed subsequently by NOESY NMR. The free basewas a yellow oil weighing 0.35 g (26%). It wa s ta ken u p intodry ether and converted to i ts hydrochloride salt by theaddition of 1 N HCl in anhydrous ethanol. After solventremoval in vacuo, the solid was crystallized from ethanol-eth er to give fine whit e crystals of12 aHCl: mp 235-236 C;1H NMR (free base in CDCl3) 1.40 (td, 1, ArCHCH2CHN, J) 12.5, 2.7 Hz), 1.50 (bs, 1, NH), 2.25 (dt, 1, ArCH CH2CHN,J ) 12.8, 4.2 Hz), 2.75 (bs, 2, ArCH2CHN), 3.40 (bt, 1,ArCH 2CHN, J ) un identifiable), 3.65 (m, 1, ArCH), 3.78 (s, 3,ArOCH3), 3.9 (s, 2, benzylic met hylen e), 4.0 (dd, 1, OCH2, J )12.2, 8.0 Hz), 4.78 (t, 1, OCH2, J ) 8.1 Hz), 6.52 (d, 1, ArH,



8/12

J ) 8.5 Hz), 6.58 (d, 1, ArH, J ) 8.5 Hz), 7.30 (m, 5, ArH,benzylic); CIMS m/z 296 (M + 1). Anal. (C19H 21NO2HCl) C,H, N.

The second band to elute from the Chromatotron was themajor product and was the syn isomer 12 b , as confirmed bysubsequent NOESY NMR experiments. The free base was ayellow oil weighing 0.84 g (63%). It was ta ken u p int o dryether and converted t o its hydrochloride salt by the additionof 1 N H Cl in anh ydrous ethan ol. After solvent removal invacuo, the solid was crystallized from ethanol-ether to givefine white crystals of12 bHCl: mp 269-270 C; 1H N MR (freebase in CDCl3) 1.33 (q, 1, ArCH CH2CHN, J ) 11.1 Hz), 1.80(bs, 1, NH), 2.29-2.39 (m, 2, a m ixtur e of ArCHCH2CHN andAr CH2CHN), 3.06-3.26 (m, a mixture of ArCH2C HN a ndArCH 2CHN), 3.38 (m, 1, ArCH), 3.78 (s, 3, ArOCH3), 3.9-4.1( a m ixtur e of s , 2 , a m ixtur e of OCH2 and the benzylicmeth ylene), 4.73 (t, 1, OCH2, J ) 8.2 H z), 6.53 (d, 1, ArH, J )8.4 H z), 6.57 (d, 1, ArH, J ) 8.4 H z), 7.32 (m, 5, ArH, benzylic);CIMS m/z 296 (M + 1). Anal. (C19H 21NO 2HCl) C, H, N.

a n t i-4-Amino-6-methoxy-2a,3,4,5-tetrahydro-2H-naph-tho[1,8-bc] f u r a n H y d r o c h l o r i d e ( 4 a ). To a su spension of100 mg of 10% Pd-C in 100 mL of absolute methanol wasadded 160 mg (0.48 mmol) of 12 a HCl in a Parr flask. Themixture was shaken under 60 psi of H 2 for 12 h and thenfiltered thr ough Celite. Removal of solvent under reducedpressu re gave a clean -white solid tha t was r ecrystallized frommethanol-ethyl acetate-ether to yield 97 m g (84%) of4a -HCl as fine white crystals: mp 229 C; 1H N MR (free bas e inCDCl3) 1.35 (bs, 2, NH2), 1.45 (qd, 1, ArCHCH2CHN, J )12.1, 2.9 Hz), 1.98 (dt, 1, ArCHCH2CHN, J ) 12.5, 4.5 H z),2.53 (d, 1, ArCH2CHN, J ) 18.2 H z), 2.71 (dd, 1, ArCH2CHN,J ) 18.2, 4.5 Hz), 3.59 (m, 1, ArCH), 3.62 (m, 1, ArCH 2CHN),3.69 (s, 3, ArOCH3), 3.9 (dd, 1, OCH2, J ) 12.3, 8.0 Hz), 4.70(t, 1, OCH2, J ) 8.0 H z), 6.45 (d, 1, ArH, J ) 8.5 Hz), 6.50 (d,1, ArH, J ) 8.5 Hz); CIMS m/z 206 (M + 1), 189. Anal.(C12H 15NO 2HCl) C, H, N.

syn -4-Amino-6-methoxy-2a,3,4,5-tetrahydro-2H- n a p h -tho[1,8-bc]furan Hydrochloride (4b). In a method identicalto that for the preparation of the anti isomer 4a HCl above,550 mg (1.66 mmol) of12 bHCl was added to a suspension of150 mg of 10% Pd-C in 100 mL of absolute methan ol andshaken under 60 psi of H2 for 12 h. Filtration through Celite

and removal of solvent under reduced pressure gave a clean-white solid that was recrystallized from methanol-ethylacetate-ether to yield 381 mg (95%) of4b HCl as fine whitecrystals: mp >310 C; 1H NMR (free base in CDCl3) 1.19 (q,1, ArCHCH2CHN, J ) 11.6 Hz), 1.42 (bs, 2, NH2), 2.09 (dt, 1,ArCHCH2CHN, J ) 11.6, 3.9 Hz), 2.18 (dd, 1, ArCH2CHN, J) 17.3, 11.9 Hz), 3.01 (dd, 1, ArCH2CHN, J ) 17.4, 6.3 Hz),3.21 (m, 1, ArCH 2CHN), 3.38 (m, 1, ArCH), 3.69 (s, 3, ArOCH3),3.9 (dd, 1, OCH2, J ) 12.3, 8.3 Hz), 4.70 (t, 1, OCH2, J ) 8.2Hz), 6.45 (d, 1, ArH, J ) 8.5 H z), 6.49 (d, 1, ArH, J ) 8.5 Hz);CIMS m/z 206 (M + 1), 189. Anal. (C12H 15NO2HCl) C, H, N.

syn -4-Amino -8-bromo-6-metho xy-2a,3,4,5-tetrah ydro-2H-naphtho[1,8-bc]furan (4d). To an ice-bath-cooled solu-tion of 85 mg (0.41 mmol) of the naphthofuran 4b in 5 mL ofCHCl3 was slowly added via syringe th rough a sept um 2.1 mLof a 0.20 N s olut ion of Br 2 in CHCl3. The mixture was stirred

for 30 min, during which time the solution turned orange anda white precipitate formed. Eth er was added, and the solidwas collected by suction filtration, rinsing well with ether onth e filter t o afford 119 m g (79%) of4d HBr as a pristine-whitesolid. The solid was recrystallized from met ha nol to give finewhite crystals: mp 277-278 C; 1H NMR (HBr salt in CD3-OD) 1.45 (q, 1, ArCHCH2CHN, J ) 11.7 Hz), 2.35 (m, 1,ArCHCH2CHN), 2.40 (ddd, 1, ArCH2CHN, J ) 17.4, 10.3, 1.9Hz), 3.1 (dd, 1, ArCH2CHN, J ) 17.4, 6.6 Hz), 3.55 (m, 2, amixture of ArCH and ArCH 2CHN), 3.70 (s, 3, ArOCH3), 3.95(dd, 1, OCH2, J ) 12.2, 8.4 Hz), 4.82 (t, 1, OCH2, J ) 8.4 Hz),6.68 (s, 1, ArH); CIMS m/z 284, 286 (M + 1), 206. Anal.(C12H 14BrNO2HBr) C, H, N.

a n t i -4-Amino-8-bromo-6-methoxy-2a,3,4,5-tetrahydro-2H-naphtho[1,8-bc]furan (4c). In a man ner identical to tha tfor the preparat ion of the syn isomer 4d above, 30 mg (0.16

mmol) of a n ti-naphthofuran 4a and 1.5 mL of 0.11 N Br 2-CHCl3 solution in 5 m L of CHCl3 gave 40 mg (69%) of4c HBras a white solid that was purified by recrystallization frommethanol: mp 269-270 C; 1H NMR (HBr salt in CD3OD) 1.65 (ddd, 1, ArCHCH2CHN, J ) 15.9, 12.4, 3.6 Hz), 2.30 (dt,1, ArCHCH2CHN, J ) 14, 4.4 Hz), 2.65 (d, 1, ArCH2CHN, J) 19.0 Hz), 2.92 (ddd, 1, ArCH2CHN, J ) 19, 6.5, 1.8 Hz),3.45-3.65 (m, 1, ArCH), 3.68 (s, 3, ArOCH3), 3.92 (m, 1,ArCH 2CHN), 4.0 (dd, 1, OCH2, J ) 11.8, 8.4 Hz), 4.82 (t, 1,OCH2, J ) 8.4 H z), 6.70 (s, 1, ArH); CIMS m/z 284, 286 (M +1), 206. Anal. (C12H 14BrNO2HBr) C, H, N.

syn -N,N-Dimethyl-4-amino-6-methoxy-2a,3,4,5-tetrahy-dro-2H-naphtho[1,8-b c] f u r a n ( 4 e ) . Sodium cyanoborohy-dride, 78 mg (1.24 mmol), was a dded to a stirred mixture of70 mg (0.31 mm ol) of4b HCl in 5 mL of meth an ol an d 0.5 mLof 37% aqueous formaldehyde u nder N 2. The clear mixturewas stirred at room temperature for 26 h, and the methanolwas removed under reduced pressure. The aqueous residuewas diluted with 2.5 N NaOH a nd extracted with 4 15 mLof CH 2Cl2. The extracts were combined, washed with 2 15mL of 5% NaHCO3 solution a nd 25 mL of brine, dried overMgSO4, and filtered th rough Celite. Removal of the solventon the rotary evaporator gave a yellow oil that was pu rifiedby applying it to a 1-mm Chromatotron plate and eluting withCH 2Cl2 under a N 2-NH 3 at mospher e. This gave 36 mg (50%)of th e free base 4e as a pale-yellow oil. The oxalat e salt wasformed by dissolving the base in ether and adding 1 equiv ofoxalic acid that had been dissolved in a small am ount of eth er.The pr ecipita ted sa lt was collected on a Buchn er funn el andwash ed well with et her to yield 48 mg (49%) of4e C2H 2O4 a sa white solid: mp 186-187 C; 1H N MR (free bas e in CDCl3) 1.35 (q, 1, ArCHCH2CHN, J ) 11.4 Hz), 2.24 (ddd, 1,ArCHCH2CHN, J ) 11.3, 4.2, 3 Hz), 2.40 (s, 6, N(CH3)2), 2.52(dd, 1, ArCH2CHN, J) 15.8, 10.6 H z), 2.9-3.1 (m, 2, a mixtu reof ArCH2CHN and ArCH 2CHN), 3.40 (m, 1, ArCH), 3.77 (s, 3,ArOCH3), 4.0 (dd, 1, OCH2, J ) 12.1, 8.2 Hz), 4.82 (t, 1, OCH2,J ) 8.2 H z), 6.52 (d, 1, ArH, J ) 8.6 H z), 6.56 (d, 1, ArH, J )8.6 Hz); CIMS m/z 234 (M + 1), 205, 162. Anal. (C14H 19-NO2C2H 2O4) C, H, N.

sy n -N,N -Dipropyl-4-amino-6-methoxy-2a,3,4,5-tetrahy-dro-2H-naphtho[1,8-b c]furan (4f). Sodium cyanoborohy-dride, 130 mg (2.07 mmol), was added to a stirred mixture of

100 mg of4b

HCl in 3 mL of methanol and 0.3 mL (4.14 mmol)of propionaldehyde u nder N 2. The white slurry was stir redat r oom tempera tur e for 72 h, and the met han ol was removedunder redu ced pressure. The residue was diluted with 2.5 NNaOH an d extracted with 4 15 mL of CH 2Cl2. The extractswere combined, washed with 2 15 m L of 5% Na HC O3solution and 25 mL of brine, dried over MgSO4, and filteredthr ough Celite. Removal of solvent on t he rotar y evaporat orga ve a n o i l tha t wa s pur i f ie d by a pply ing i t to a 1- m mChromatotron plate and eluting with CH 2Cl2 under a N 2-NH 3atmosphere. Concentrat ion of the eluted fractions gave aquant itative yield of the free base as a yellow oil. Severalattempts were made to crystallize the product as its hydro-chloride or metha nesulfonat e salt, but these formed gumlikesolids tha t could not be induced t o crystallize. The productfinally was precipitated as its oxalate salt and recrystallized

from eth yl acetat e-

ether-

hexane to give 4f

C2H 2O4 as a whitegranular solid: m p 132-133 C; 1H NMR (oxalat e in CD3OD) 1.01 (t, 6, N(CH 2CH 2CH3)2, J ) 7.4 Hz), 1.7 (q, 1, ArCH CH2-CHN, J ) 11.4 Hz), 1.73 (m, 4, N(CH 2CH2CH 3)2) 2.45 (dt, 1,ArCHCH2CHN, J ) 10.9, 3.8 Hz), 2.78 (dd, 1, ArCH2CHN, J) 17.3, 11.8 H z), 3.1-3.26 (m, 5, a mixture of ArCH2CHN andN(CH2CH 2CH 3)2), 3.5 (m, 1, ArCH 2CHN), 3.77 (s, 3, ArOCH3),4.01 (m, 1, ArCH and OCH2), 4.78 (t, 1, OCH2, J ) 8.2 Hz),6.54 (d, 1, ArH, J ) 8.5 Hz), 6.63 (d, 1, ArH, J ) 8.5 Hz); CIMSm/z 290 (M + 1). Anal. (C18H 27NO2C2H 2O4) C, H, N.

Pharmacological Methods. Drug Discrimination Stud-i e s . The procedures for t he dr ug discriminat ion assays wereessentially as described in previous reports. 10,29-31 Twentymale Sprague-Dawley rats (Harlan Laboratories, Indianapo-lis, IN), weighing 200-220 g at the beginning of the study,were trained to discriminate LSD tart rat e from saline. None

S u b st it u ted N a p ht h of u ra n s a s H yb ri d M ol ecu l es J ou r na l of M ed i ci n al C hem i st ry , 1 99 8, V ol . 4 1, N o. 1 2 2141


9/12

of th e rats had previously received drugs or behavioraltraining. Water was freely available in th e individual homecages, and a rationed amount of supplemental feed (PurinaLab Blox) was m ade a vailable after experimental sessions tomainta in a pproximately 80% of free-feeding weight. Lightswere on from 0700 to 1900. The laboratory and a nimal facilitytemperature was 22-24 C, and th e relative humidity was 40-50%. Experiments were performed between 0830 and 1700each day, Monday-Friday.

Six standard operant chambers (model E10-10RF, Coul-

bourn Instruments, Lehigh Valley, PA) were enclosed withinsound-att enuat ed cubicles with fans both for ventilation an das a source of background white noise . A white l ight wascent ered near th e top of the front pan el of the cage, which wasalso equipped with two response levers, separated by a foodhopper (combinat ion dipper pellet trough, model E14-06,module size 1/2), all positioned 2.5 cm above th e floor. Contr olof reinforcement and acquisition of data was accomplishedth rough a Med Associates in terface to an ISA microcomput er,using custom software wr itten in this laboratory.

A fixed r at io (FR) 50 schedu le of food r einforcemen t (Bioserv45 mg of dustless pellets) in a two-lever pa radigm was used.Initially, rats were shaped to lever-press on a n F R1 scheduleso that one food pellet was dispensed for each press. To avoidpositional preference, half of the rats were trained on drug-L(left), saline-R (right) and the other half on drug-R, saline-L.

Training sessions lasted 15 min and were conducted at thesame t ime each between anima ls with 10% etha nol solutionto avoid olfactory cues. Only one appropriat e lever was presen tdur ing the first 10 sessions of initia l learning (after beginningto a dm iniste r sa line or t r a in ing dr ug, ip 30 m in befor esessions). Afterward, both levers were present during allfollowing phases of training, but reinforcements were deliveredonly after responses on the appropriate lever. Presses on theincorrect lever had no programmed consequences.

As responding rat es stabilized (dur ing the n ext 15 sessions),the schedule of reinforcement was gradually increased to aFR50. Once at th e FR50, tra ining continued un til an accura cyof at least 85% (number of correct presses 100/number oftotal presses) was attained for 8 of 10 consecutive sessions(approximately 40-60 sessions). Once criterion perform an cewas att ained, test sessions were interspersed between trainin gsessions, eith er one or two times per week. At least one drugand one saline session separated each test session. Rats wererequired to maintain the 85% correct responding criterion ontra ining days in order to be test ed. In addition, test dat a werediscarded when the accuracy criterion of 85% was not achievedon the t wo training sessions following a t est session. Testsessions were run under conditions of extinction, with ratsremoved from the operant chamber when 50 presses wereemitted on one lever. If 50 presses on one lever were notcompleted within 5 min, the session was ended and scored asa disruption. Treatment s were randomized at the beginningof the study.

The training drug was (+)-lysergic acid diethylamide tar-tr at e (LSD; 0.08 mg/kg, 186 nm ol/kg; NIDA). All dru gs weredissolved in 0.9% saline and were injected intraperitoneally

in a volume of 1 mL/kg, 30 min before th e sessions. Data fromthe dru g discriminat ion stu dy were scored in a quant al fashion,with the lever on which the rat first emitted 50 presses in atest s ession scored as t he selected lever. The percenta ge ofrats selecting the drug lever (%SDL) for each dose of testcompound was determined. The degree of substitution wasdetermined by th e maximum %SDL for a ll doses of the testdrug. No substitut ion is defined as 59% SDL or less, andpart ial substitut ion is 60-79% SDL. If the drug was onetha t completely substituted for t he tr aining drug (at least onedose resulted in a %SDL ) 80% or higher), the method ofLitchfield and Wilcoxon 32 wa s use d to dete r m ine th e ED 50(log-probit analysis as the dose producing 50% drug-leverresponding) and 95% confidence interval (95% CI). Thismethod a lso allowed for tests of parallelism between dose-response curves of the dr ug and the t raining dru g. If 50% or

more of the animals tested were disrupted at a dose wherethe n ondisrupted rat s gave 80% SDL, no ED50 was calculated.

S e r o t o n i n R a d i o r e c e p t o r A s sa y s U s i n g R a t B r a i nH o m o g e n a t e . Male Sprague-Dawley rat s (Harlan Labora-tories, Indian apolis, IN) weighing 175-199 g were used. Theanimals were kept in groups of five rats per cage, using theconditions described above, except for free access to food andwater .

[3H]Ketan serin and [3H]-8-OH-DPAT were pu rchas ed fromNew E nglan d Nu clear (Boston, MA) at specific activities of 61

and 135.5-

216 Ci/mmol, respectively. Cinan serin was a giftfrom the E. R. Squibb & Sons, Inc. (New Brunswick, NJ),and 5-HT was purchased from Sigma Chemical Co. (St. Louis,MO).

The procedures of J ohnson et a l.33 were employed. Briefly,the fronta l cortex or hippocampal brain regions from 20-40rats were pooled and homogenized (Brinkman Polytron, setting6 for 2 20 s) in 4 or 8 volumes of 0.32 M sucrose for front alcort ex or hippocampus, respectively. The homogenates werecentr ifuged at 36000g for 10 min, and the resulting pelletswere resuspended in the same volume of sucrose. Separat ealiquots of tissu e suspens ion were then frozen at -70 C un tilassay. For each separate experiment, a t issue aliquot wasthawed slowly and diluted 1:25 with 50 mM Tris-HCl (pH 7.4).The homogenate was then incubated at 37 C for 10 min andcentrifuged twice at 36500g for 20 min with an intermittent

wash. The resulting pellet was resuspended in 50 mM Tris-HCl with 0.5 mM Na 2EDTA, 0.1% Na ascorbate, and 10 mMpargyline HCl (pH 7.4). In experiments with [3H]ketanserinor [3H]-8-OH-DPAT, either 5.7 mM CaCl2 or 10 mM MgCl2was included, respectively. A second preincubation for 10 minat 37 C was conducted, and the tissues were t hen cooled inan ice bath.

All experiments were performed in tr iplicate using theappr opriate bu ffer to which 200-40 0 g of protein was added,giving a final volume of 1 mL. The tu bes were allowed toequilibrate for 15 min at 37 C before fi lter ing th roughWhatman GF/C filters using a cell harvester (Brandel, Gaith-ersburg, MD) followed by two 5 mL washes using ice-cold Trisbuffer. Specific binding was defined usin g 10 M cinanserinin the [3H]ketanserin binding study and 10 M 5-HT in t he[3H]-8-OH-DPAT binding stu dy. Filters were a ir-dried, placed

into scintillation vials with 10 mL of Ecolite scintillationcocktail, and allowed to sit overnight before counting fortr i t ium.

R a d i o l i g a n d C o m p e t i t i o n E x p e r i m e n t s U s i n g C l o n e dHuman 5-HT2 R e c e p t o r s . All chemicals were obtained fromth e sources previously described.34 [3H]-5-HT was purchasedfrom DuPont-NEN (Wilmington, DE) or Amersham Corp.(Arlington Heights, IL) a t 22.8-26.7 or 81-91 Ci/mmol,respectively. [3H]Mesulergine (76-83 Ci/mmol) was pur-chased from Amersham Corp. (Arlington Heights, IL). [125 I]-DOI (2200 Ci/mmol), [3H]rauwolscine (70-90 Ci/mmol), and[3H]ketanserin (60-78.7 Ci/mmol) were purchased from Du-Pont-NEN (Wilmington, DE).

Membranes were prepared essentially as previously de-scribed usin g AV12 cell lines (Syrian h am ster fibroblast , ATCCno. CRL 9595) stably transformed with the human 5-HT 2A,

5-HT2B , or 5-HT2C receptors.34

In brief, cells expressing th ereceptor of interest were grown in suspension and har vestedby centrifugation. The cell pellets were th en resuspen ded ina m inimal volume of a h ypotonic buffer, 50 m M Tris-HCl, pH7.4, and frozen at -70 C until needed. On the day of theassay, the membrane suspension was thawed and diluted to35 mL per 0.5 10 9 cells with 50 mM Tris-HCl, pH 7.4. Thecombina tion of hypotonic buffer a nd vort exing was sufficientto lyse the cells for t he membra ne prepar ation. After vortex-ing, the prepara tion was centrifuged at 39000g for 10 min at4 C, and the resulting membrane pellet was resuspended andincubated at 37 C for 10 min and then centrifuged at 39000gfor 10 min at 4 C. This pellet was resuspended an d centr i-fuged one more t ime, and the f inal membrane pelle t wasresupended (using a Tissumizer, setting 65 for 15 s) in Tris-HCl, pH 7.4, for cells expressing the human 5-HT 2B receptor,



10/12

in Tr is-HCl, pH 7.4, contain ing MgCl2 and EDTA for [125 I]DOIbinding t o 5-HT 2A or 5-HT2C receptors, or in Tris-HCl, pH 7.6,for [3H]ketanserin and [3H]mesulergine binding t o 5-HT2A and5-HT2C receptors, r espectively.

5-HT2B [3H ] -5 -H T B i n d i n g S t u d i e s . Human 5-HT2B r e-ceptor binding assays using [3H]-5-HT were performed aspreviously described.34 The a ssa y wa s a utom a te d us ing aBiomek 1000 (Beckman Instruments, Fullerton, CA). [3H]-5-HT in Tris-HCl containing CaCl2, pargyline, and L-ascorbicacid, adjusted t o pH 7.4, was added t o drug dilutions, spanning6 log units, in water. Then 200 L of membrane resuspension(approximat ely 100-150 g of protein) was added wit h m ixingfol lowed by incuba t ion for 15 m in a t 37 C . The to ta lincubation volume was 800 L, and all incubations wereperformed in t riplicate. The final concentrat ion of CaCl2,par gyline, Tris, and L-ascorbic acid was 3 m M, 10 M, 50 mM,and 0.1%, respectively. The assa y was term inat ed by vacuumfi lt r a t ion thr ough Wha tm a n GF /B fi lte r s tha t ha d bee npresoaked with 0.5% poly(ethylenimine) (w/v) and precooledwith 4 mL of ice-cold wa sh buffer (50 m M Tris-HCl, pH 7.4),using a Brandel cell harvester (model MB-48R, Brandel,Gaither sburg, MD). The filters then were wash ed rapidly fourtimes with 1 mL of ice-cold wash buffer. The am ount of [3H]-5-HT t rapped on the filters was determined by liquid scintil-lation spectrometry (Ready Protein, LS 6000IC, BeckmanInstru ments, Fullerton, CA). The final [3H]-5-HT concentr a-tion for competition stu dies was approximately 2 nM (ran ge) 1.7-2.5 nM). The actua l free radioligand concentra tion wasdetermined by sampling t he supernatant of identical t ubeswhere bound ligand was removed by centrifugation. Nonspe-cific binding wa s defined with 10 M 5-HT or 10 M 1-naph -thylpiperazine (1-NP). The amount of protein was determinedby the method of Bradford, with bovine serum albumin as thestandard.35

5-HT2A/2C [125 I ]D O I B i n d i n g S t u d i e s . Human 5-HT2A or5-HT2C binding studies were performed essentially as describedfor [3H]-5-HT binding to the 5-HT2B receptor with the followingexceptions. The assa y buffer conta ined, in final concentr ation,10 M pargyline, 9.75 mM MgCl2, 0.5 mM disodium EDTA,0.1% sodium ascorbat e, and 50 mM Tris-HCl, pH 7.4. Incuba-tions were performed at 37 C for 30 min with approximately40 and 30 g of protein for the 5-HT2A and 5-HT2C receptors,respectively, followed by filtration and washing as describedabove. The amount of [125 I]DOI trapped on the filters wasdetermined using a -counter. Nonspecific binding was de-termined with 10 M mianserin for 5-HT2C a n d 1 M ket-anser in for 5-HT2A receptors. The final concentration of [125 I]DOI was approximately 0.07-0.15 nM.

[3H ] K e t a n s e r i n B i n d i n g t o t h e H u m a n 5 - H T 2A R e c e p -tor. The assay conditions were essentially as previouslydescribed.36 Assays consisted of 0.8-mL total volume cont ain-ing 50 mM Tris-HCl, 100 nM prazosin (to block potentialbinding of [3H]ketanserin to R1-adrener gic receptors), 0.4-0.5nM [3H]ketan serin, and va rying concentra tions of the compet-ing compound of inter est (final pH 7.6). Mianserin , 3 M, wasused t o define t he level of nonspecific binding. Tubes wereincubated at 37 C for 15 min and then rapidly filtered andwashed as descr ibed above. The amount of [3H]ketanserintrapped on th e filters was determ ined by liquid scintillationspectrometry.

[3H ] R a u w o l s c i n e B i n d i n g t o t h e H u m a n 5 - H T2B Re -c e p t o r . This assay is based on a previously descr ibedprocedure.37 Conditions specific to th is as say wer e as follows(all concentrations given as final concentrations): 2 nM [3H]-rauwolscine, 500 nM efaroxan (to mask rauwolscine bindingto R2-adrenergic receptors), and 50 mM Tris-HCl, pH 7.4.Tubes were incubated at 37 C for 20 min and then rapidlyfiltered a s described above. Nonspecific binding was definedin the presence of 10 M 1-naph thylpiperazine.

[3H]Mesulergine Binding to the Human 5-HT2C R e c e p -tor. This assay was adapted from that described by Pazos etal.38 Membranes were prepared as described above. Finalconcentr at ions for th e 0.8-mL a ssays were 0.74-0.82 nM [3H]-mesu lergine, varying concentr at ions of competin g compoun d,

and 50 nM Tris-HCl, final pH 7.6. Nonspecific binding wasdetermined using 3 M mianserin. Assay tubes were incu-bated for 30 min at 37 C, af ter which the samples werefiltered, and washed, and radioactivity was determined as forthe [3H]ketanserin binding assay described above.

Mu s c a r i n ic R e c e p t o r S t u d i e s : 1 . C e ll C u l tu r e . B82(murine fibroblasts) tr ansfected with the muscarinic M1-M5receptors were prepa red as previously described by Lai et al.39

(for the M1 receptor), Lai et al.40 (for the M2 receptor), andKa shiha r a e t a l .41 (for the M3-M5 receptors) and were agenerous gift from Dr. Henry Yamamura (University of Arizona). Cells were maint ained in DMEM/F12 media supp le-ment ed with 5% feta l calf serum, 5% newborn calf serum , andpen/strep. Cells were grown in media supplemented with 500ng/mL G418 every fourt h pa ssage to ma intain receptor expres-sion. Cells were grown in eith er Costar 75-cm 2 flasks (bindingstudies) or 24-well plates (functional studies) and kept in ahumidified incubator maintained at 37 C with 95% O2 a nd5% CO2 levels.

2 . M e m b ra n e P r e p a r a t i o n . Confluent flasks of cells wereincubated with 7 mL of 5 mM H epes, 5 mM EDTA (pH 7.4)buffer, homogenized with a Whea ton Teflon-glass h omogenizer(7 strokes), and centrifuged at 1000g for 10 min at 4 C. Theresulting supernatant was centrifuged at 27000 g for 30 minat 4 C, and the pellet was resuspended in 10 mM Hepes, 1mM MgCl2, 10 mM Na/K-PO4 buffer, pH 7.4, and stored in

1-mL aliquots at-

80 C until use in the binding assays.Aliquots contained a pproximately 1 mg/mL pr otein, a s m ea-sur ed using th e BCA protein a ssay rea gent (Pierce, Rockford,IL).

3 . M u s c a r i n i c R a d i o l i g a n d B i n d i n g S t u d i e s . [3H]QNBbinding was performed essentially as described in Lai et al.39

with slight modifications. Membranes were diluted with 50mM Hepes, 4 mM MgCl2 (pH 7.4), and 100 L (approximately25 g of protein) was incubated with 0.25 nM [3H]QNB andincreasing concentrat ions of competing dru gs. Nonspecificbinding was determined u sing 1 M atropine. Incubationswere ru n in d uplicate at 37 C for 60 min in a final volume of1 mL. Tubes then were filtered rapidly through Skat ron glassfiber filter mats (cat. 11734), and rinsed with 5 mL of ice-cold buffer using a Skatron micro cell harvester (SkatronInstru ments Inc., Sterling, VA). Filters were allowed to dry

and th en punched into scintillation vials (Skat ron Instru mentsInc., Ster ling, VA). OptiPh ase HiSa fe II scintillation cocktail(1 mL) was a dded to each vial. After t he vials shook for 30min, radioactivity in each sample was determined on an LKBWallac 1219 Rackbeta liquid scintillation counter (Wallac Inc.,Gaithersburg, MD).

I n o s i t o l L i p i d H y d r o l y s i s S t u d i e s . These experimentswere performed as described by Mei et al.42 Briefly, M1-B82cells were plated onto 24-well plates and grown to 50 %confluency. Cells th en were incubat ed overnight with 0.2 Mmyo-2-[3H]inositol (15 Ci/mmol; New En gland N uclear) in 0.5mL of media. Excess m yo-2-[3H]inositol wa s removed, a ndcells were rinsed with 1 mL of plain media. Following thisrinse, cells were incubated with 10 mM LiCl for 10 min, andth e reaction was initiat ed by the add ition of carbachol or othertest drugs. The reaction was terminated 60 min la ter by

aspirating th e media and adding 0.31 mL of metha nol. Thecells were scraped a nd placed in 1.7-mL centr ifuge tu bes. Thewells were rinsed with an additional 0.31 mL of methanol,which was then added to th e centr ifuge tubes. Chloroform(0.62 mL) and distilled wat er (0.31 m L) were added to eachtube; the tubes were shaken for 15 s and then centrifuged for5 min at 1000g. The top phase (0.7 mL) was th en added to 2mL of disti l led water , and this mixture was passed overminicolumns consisting of 2 mL of a 10% slurry of anion-exchan ge resin in forma te form (AG 1-X8, 100-200 mesh ; Bio-Rad Labs). The columns were wash ed five times with 5 mLof distilled water. The [3H]IP 1 was th en eluted using 2 mL of0.2 M amm onium format e/0.1 M formic acid. Nine millilitersof AquaMix (ICN Radiochemicals, San Diego, CA) then wasadded to the eluates that were then quantif ied by l iquidscintillation spectroscopy.



11/12

cAMP Assay. B82 cells t ran sfected with the M2 receptorwere plat ed onto 24-well plates an d grown t o ca. 90% conflu-ency. Duplicate wells were r un for each dr ug concentra tion.cAMP levels were increa sed by th e addit ion of 100 M forskolinin media supplemented by 500 M isobutyl methylxanthine(IBMX), after which the ability of drugs to inhibit cAMPsynthesis e levation was assessed. Carbachol was run as acont rol in each assa y. Cells were incubated with dr ugs for 10min at 37 C, after which the reaction was terminat ed by theadd ition of 0.5 mL of ice-cold 10 mM Tris/2 mM E DTA. Cellswere then scraped and placed into 1.7-mL centrifuge tubes.The wells were rinsed with a n a dditional 0.5 mL of Tris/EDTAbuffer th at wa s added t o the centr ifuge tubes before th ey werespun at 14000g for 5 min t o pellet large debris. The level ofcAMP was determined as described below.

R a d i o im m u n o a s s a y o f c AMP . The concentration ofcAMP in each sample was determined with an RIA of acety-lated cAMP, modified from th at previously described.43 Iodi-nation of cAMP was performed using a method describedpreviously.44 Assay buffer was 50 mM sodium acetate bufferwith 0.1% sodium a zide (pH 4.75). Sta nda rd curves of cAMPwere prepar ed in buffer at concentr ations of 2-500 fmol/assaytube. To improve assa y sensitivity, all samples and stan dard swere acetylated with 10 L of a 2:1 solution of trieth ylamine-acetic anh ydride. Samples were assayed in duplicate. Eachassay tube contained 10 L of sample, 100 L of buffer, 100

L of primary antibody (sheep anti-cAMP, 1:100000 dilutionwith 1% BSA in bu ffer), and 100 L of [125I]cAMP (50 000 dpm/100 L of buffer) giving a total a ssay volume of ca. 300 L.Tube s wer e vor te xe d a nd s tor e d a t 4 C over night (a p-proximately 18 h ). Antibody-bound radioactivity was t henseparated by the addition of 10 L of BioMag r abbit an ti-goatIgG (Advanced Magnetics, Cambridge, MA), followed by vor-texing and further incubat ion at 4 C for 1 h. To these sam pleswas added 1 m L of 12% poly(ethylene glycol)/50 mM s odiumacetate buffer (pH 6.75), and all tubes were centrifuged at1700g for 10 min. Supernatants were aspirated, and radio-activity in th e resulting pellet was determined using an LKBWallac -counter (Wallac Inc., Gaithersburg, MD).

D a t a A n a l y s i s . The resulting data from each competitionassay were an alyzed by nonlinear regression u sing the modelfor sigmoid cur ves in th e curve-fitt ing program P rism (Graph -

pad Inc., San Fra ncisco, CA). This program generated IC50values and a H ill coefficient for each curve. The appa rentaffinities (K0.5) were calculated by the Cheng-Pru soff equa-tion.45

For M1 functional data, basal levels of IP 1 were subtracteda nd the da ta wer e the n e xpr essed a s pe rce nt ca r ba cholstimulation at 100 M, to minimize interassay variability.Values were then averaged across assays. For M2 functionaldata, baseline values of cAMP were subtracted from the totalamount of cAMP produced for each drug condition. Tominimize interassay variation, data for each drug wereexpressed relative to the percentage of the stimulation pro-duced by 100 M forskolin in each assa y. These values thenwere averaged to obtain the cAMP stimulation for each drug.

A c k n o w l e d g m e n t . This work was supported byU.S. Public Health Service Grant DA02189 from theN at ional Ins t it u t e on D rug Abuse , C ent er G rant sMH33127 and HD03310, an d Train ing Grant DA07244.A. Monte was supported by a National Insti tutes ofHealth-Na tional Inst itut e of Gener al Medical SciencesPredoctoral Training Grant. We would also l ike toexpress our grati tude to John Kozlowski and AmjadQandil for assistance with the NMR studies. TheNovaScreen assay was provided through support by acont ra ct from th e NIMH. The mu scarinic cell lines weregraciously provided by P rofessor Hen ry Yamam ura ofthe Department of Pharmacology, School of Medicine,University of Arizona.

R e f e r e n c e s

(1) Branchek, T.; Adham, N.; Macchi, M.; Kao, H.-T.; Hartig, P. R.[3H]DOB (4-Bromo-2,5-Dimet hoxyphenylisopropylamine) an d[3H]Ketanserin Label Two Affinity States of the Cloned Human5-Hydroxytryptamine2 Receptor. Mol. Pharmacol. 1990, 38 ,604-609.

(2) Pierce, P. A.; Peroutka, S. J . Hallucinogenic Drug I ntera ctionswith Neurotr ansm itter Receptor Binding Sites in Hum an Cortex.Psychopharmacology 1989, 97, 118-122.

(3) Titeler , M.; Leonha rdt , S.; Weisberg, E. L.; Hoffman, B. J. 4-[125I]-Iodo-(2,5-dimet hoxy)phen ylisopropylamin e an d [3H]KetanserinLabeling of 5-Hydroxytryptam ine2 (5-HT2) Receptors in Ma m-

malian Cells Transfected with a Rat 5-HT2 cDNA: Evidence forMultiple Stat es and Not Multiple 5-HT2 Receptor Su btypes. Mol.Pharmacol. 1990, 38 , 594-598.

(4) Marini-Bettolo, G. B.; Chiavarelli, S. ; Bovet, D. SyntheticSympat holytic Substa nces of the Ergotam ine Series. I. Nitr ogen-Substituted Derivatives of the d l-1,2,3,4-Tetrahydro-2-Naph-thylamine-Containing Amino and Amido Functional Groups.Gazz. Chim. Ital. 1951, 80 , 281-298.

(5) Wang, C.-D.; Gallaher, T. K.; Shih, J. C. Site-Directed Muta gen-esis of the Ser otonin 5-Hydroxytrypam ine2 Receptor: Identifica-tion of Amino Acids Necessar y for Ligan d Binding a nd ReceptorActivation. Mol. Pharmacol. 1993, 43 , 931-940.

(6) Kao, H.-T.; Adha m, N.; Olsen, M. A.; Weinsha nk, R. L.; Branchek,T. A.; Hartig, P. R. Site-Directed Mutagenesis of a Single ResidueChanges the Binding Properties of the Serotonin 5-HT 2 receptorfrom a h uman to a rat pharmacology. FEBS Le t t . 1992, 307,324-328.

(7) Nichols, D. E.; Barfknecht, C. F.; Rusterholz, D. B.; Benington,F.; Morin, R. D. Asymmetric Synthesis of Psychotomimetic

Phenylisopropylamines. J . M ed . Ch e m. 1973, 16, 480-483.(8) Shulgin, A. T.; Dyer, D. C. P sychotomimetic Phenylisopropy-lamin es. 5. 4-Alkyl-2,5-Dimethoxyphenylisopropylamin es. J .

Med . Ch em . 1975, 18 , 1201-1204.(9) Glennon, R. A.; Young, R.; Ja cyno, J. M. Indolalkylamine and

Phenalkylamine Hallucinogens. Biochem . Pharm acol. 1983 , 32 ,1267-1273.

(10) Monte, A. P.; Marona-Lewicka, D.; Parker , M. A.; Mayleben, M.;Mailman, R. B.; Wainscott, B.; Nelson, D. L.; Nichols, D. E.Dihydrobenzofuran Analogues of Hallucinogens. 3. Models of4-Substituted (2,5-Dimethoxyphenyl)alkylamine Derivatives withRigidified Methoxy Groups. J. Med. Chem. 1996 , 39, 2953-2961.

(11) Appel, J. B.; Cunningh am, K. A. The Use of Drug Discrimin ationProcedures to Ch aracterize Hallucinogenic Drug Actions. Psy-chopharmacol. Bull. 1986, 22 , 959-967.

(12) Jung, M. E.; Abrecht, S. Improved Synthesis of 3-Substituted7-Methoxybenzofurans, Useful Intermediates for Preparation ofMorphine Analogues. J. Org. Chem. 1988, 53 , 423-425.

(13) Chan, J . H.-T.; Elix, J. A.; Ferguson, B. A. Annelated Fur ans.

XVII The Wit tig Reaction of Benzofuran s. Aust. J. Chem. 1975,28 , 1097-1111.(14) McKervey, M. A.; Tuladhar, S. M.; Twohig, M. F. Efficient

Synth esis of Bicyclo[5.3.0]decat rienones and of 2-Tetr alones viaRhodium(II) Acetate-Catalysed Cyclization ofR-DiazoketonesDerived from 3-Arylpropionic Acids. J . C h em . S o c. , C h em .Co mmu n . 1984, 129-130.

(15) Beames, D. J.; Mander, L. N. Studies on In tra molecular Alky-lation. IV. The Pr eparat ion of Spirodienones from PhenolicDiazoketones. Au s t . J . Ch e m. 1974, 27, 1257-1268.

(16) Cushman, M.; Choong, T.-C.; Valko, J. T.; Koleck, M. P. TotalSynthesis of (()-Chelidonine. J . Or g. Ch e m. 1980, 45 , 5067-5073.

(17) Cushma n, M.; Abbaspour, A.; Gupta, Y. P. Total Synthesis of(()-14-Epicorynoline, (()-Corynoline, an d (()-6-Oxocorynoline.

J . Am . Ch em . S oc. 1983, 105 , 2873-2879.(18) Borch, R. F.; Hassid, A. I. A New Meth od for the Met hylat ion of

Amines. J. Org. Chem. 1972, 37, 1673-1674.(19) Johnson, M. P.; Audia, J. E .; Nissen, J. S.; Nelson, D. L. N(1)-

Substituted Ergolines and Tryptamines Show Species Differ-ences for the Agonist-labeled 5-HT2 Receptor. Eur. J. Pharmacol.1993, 23 9, 111-118.

(20) Doods, H. N.; Quirion, R.; Mihm, G.; Engel, W.; Rudolf, K.;Entzeroth, M.; Schiavi, G. B.; Ladinsky, H.; Bechtel, W. D.;Ensinger, H. A.; Mendla, K. D.; Eberlein, W. TherapeuticPotent ial of CNS-Active M2 Antagonists: Novel Structures andPharmacology. Life S ci. 1993, 52 , 497-503.

(21) Freedman, S. B.; Dawson, G. R.; Iversen, L. L.; Baker, R.;Hargreaves, R. J . The Design of Novel Muscarinic PartialAgonists th at Have F unctional Selectivity in Phar macologicalPreparations in Vitro and Reduced Side-Effect Profile in Vivo.

L ife S ci. 1993, 52 , 489-495.(22) Lambrecht, G.; Moser, U.; Grimm, U .; Pfaff, O.; Herma nni, U.;

Hildebrandt, C.; Waelbroeck, M.; Christophe, J .; Mutschler, E.New Fu nctionally Selective Muscarinic Agonists. Life Sci. 1993,52 , 481-488.

(23) Wess, J. Molecular Basis of Muscarinic Acetylcholine ReceptorFunction. TiPS 1993, 14 , 308-312.



12/12

(24) Gitler, M. S.; Cohen, V. I.; De La Cruz, R.; Boulay, S.; Jin, B.;Zeeberg, B. R.; Reba, R. C. A Novel Muscarin ic Receptor Ligandwhich Penet rat es the Blood Brain Bar rier an d Displays In VivoSelectivity for the M2 Subtype. Life S ci. 1993, 53 , 1743-1751.

(25) Tomaszewski, Z.; Johnson, M. P.; Huang, X.; Nichols, D. E.Benzofuran Isosteres of Ha llucinogenic Tryptam ines. J . M e d .Chem. 1992, 35 , 2061-2064.

(26) De Boer, T. J.; Backer, H. J. Diazomethane. Organic Syntheses;Wiley: New York, 1963; Collect. Vol. IV, pp 250-253.

(27) Black, T. H. The Pr eparat ion a nd Reactions of Diazomethane.Al d rich im . Act a 1983, 16, 3-10.

(28) Boegesoe, K. P.; P erregaard, J. K. 3-Aminomethyl Derivativesof Indane, Indoline, and Dihydrobenzofuran and Thiophene.

Eur opean Pat ent Application No. 0 281 261 A2; July 9, 1988.(29) Nichols, D. E.; Snyder, S. E .; Oberlender, R.; J ohnson, M. P .;

Hu ang, X. 2,3-Dihydrobenzofur an Analogues of Hallu cinogenicPhenethylamines. J . M e d . Ch em. 1991, 34 , 276-281.

(30) Oberlender , R.; Nichols, D. E. (+)-N-Methyl-1-(1,3-benzodioxol-5-yl)-2-butan amine a s a Discriminitive Stimu lus in Studies of3,4-Methylenedioxyamphetam ine-Like Behavioral Activity. J .Pharmacol. Exp. Ther. 1990, 255 , 1098-1106.

(31) Oberlender, R.; Nichols, D. E. Drug Discrimin at ion Studies withMDMA and Ampheta mine. Psychopharmacology 1988, 95 , 71-76.

(32) Litchfield, J. T.; Wilcoxon, F. A. A Simplified Method of Evaluat-ing Dose-Effect Experiment s. J. Pharmacol. Exp. T her. 1949,96, 99-113.

(33) Johnson, M. P.; Mat his, C. A.; Shulgin, A. T.; Hoffman, A. J .;Nichols, D. E. [125I]-2-(2,5-dimethoxy-4-iodophenyl)Aminoethane([125I]-2C-I) as a Label for the 5-HT2 Receptor in Rat FrontalCortex. Pharmacol. Biochem. Behav. 1990, 35 , 211-217.

(34) Wainscott, D. B.; Cohen, M. L.; Schenk, K. W.; Audia, J . E.;

Nissen, J. S.; Baez, M.; Kursar , J . D.; Lucaites, V. L.; Nelson,D. L. Pharmacological Characteristics of the Newly Cloned Rat5-Hydroxytryptamine2F Receptor. Mol. Pharmacol. 1993, 43 ,419-426.

(35) Bradford, M. M. A. A Rapid and Sensitive Method for theQuant itation of Microgram Quant ities of Protein Ut ilizing thePrin ciple of Protein-Dye Binding. Anal. B iochem. 1976 , 72, 248-254.

(36) Nelson, D. L.; Lucaites , V. L.; Audia, J . E.; Nissen, J . S. ;Wainscott, D. B. Species Differences in th e Ph ar macology of the5-Hydroxytryptamine2 Receptor: Stru cturally Specific Differ-entiation by Er golines and Tryptamines. J. Pharmacol. Exp.Ther. 1993, 26 5, 1272-1279.

(37) Wainscott, D. B.; Sasso, D. A.; Kursar , J. D.; Baez, M.; Lucaites,V. L.; Nelson, D. L. [3H]-Rauwolscine: an Antagonist Radioli-gand for the Cloned Human 5-Hydroxytryptamine2B (5-HT2B)Receptor. Naunyn-Schmiedebergs Arch. Pharmacol. 1998, 357,17-24.

(38) Pazos, A.; Hoyer, D.; Palacios, J . M. Mesulergine, a SelectiveSerotonin-2 Ligand in the Rat Cortex: Evidence for SpeciesDifferences in Brain Serotonin-2 Receptors. Eur. J . Pharmacol.1985, 106, 531-538.

(39) Lai, J.; Mei, L.; Roeske, W. R.; Chung, F.-Z.; Yamamura, H. I.;Venter, J . C. The Cloned Murine M1 Muscarinic Receptor is

Associated with the Hydrolysis of P hosphatidylinositols inTransfected B82 Cells. Life S ci. 1988, 42 , 2489-2502.(40) Lai, J .; Bloom, J. W.; Yama mur a, H. I.; Roeske, W. R. Amplifica-

tion of the rat M2 muscarinic receptor gene by th e polymerasechain reaction: functional expression of t he M2 m uscarinicreceptor. Life S ci. 1990, 47, 1001-1013.

(41) Kashihara, K.; Varga, E. V.; Waite, S. L.; Roeske, W. R.;Yamamur a, H. I. Cloning of the r at M3, M4 and M5 muscarinicacetylcholine receptor genes by the polymerase chain reaction(PCR) and th e phar macological chara cterization of the expressedgenes. Life S ci. 1992 , 51 , 955-971.

(42) Mei, L.; Lai, J.; Yamamur a, H. I.; Roeske, W. R. The RelationshipBetween Agonist States of the M1 Muscarinic Receptor an d th eHydrolysis of Inositol Lipids in Transfected Murine FibroblastCells (B82) Expressing Different Receptor Densities. J . Ph a r -macol. E xp. Th er. 1989, 251 , 90-97.

(43) Harper, J. F.; Brooker, G. Femtomole Sensitive Radioimmu-

noassay for Cyclic AMP and Cyclic GMP after 2O-Acetylationby Acetic Anhydride in Aqueous Solution. J. Cyclic Nucleotide

R es. 1975, 1, 207-218.(44) Patel, A.; Linden, J. Purification of125I-Labeled S uccinyl Cyclic

Nucleotide Tyrosine Meth yl Esters by High Per form ance LiquidChromatography. Anal. Biochem. 1988, 16 8, 417-420.

(45) Cheng, Y.; Prusoff, W. H. Relationship Between the InhibitionConstan t (Ki) and the Concentra tion of an Inhibitor that Causes50% Inhibition (I 50) of an Enzymatic Reaction. Biochem. Phar-macol. 1973, 22 , 3099-3108.

JM980076U


Documents

nichols-naphtofurans