James J. Chambers et al- Synthesis and Pharmacological Characterization of a Series of Geometrically Constrained 5-HT2A/2C Receptor Ligands

8/3/2019 James J. Chambers et al- Synthesis and Pharmacological Characterization of a Series of Geometrically Constrained …

http://slidepdf.com/reader/full/james-j-chambers-et-al-synthesis-and-pharmacological-characterization-of 1/10

S y n t h e s i s a n d P h a r m a c o l o g i c a l C h a r a c t e r i z a t i o n o f a S e r i e s o f G e o m e t r i c a l l y

Cons t rai ned 5- HT2A/2C R e c e p t o r L i g a n d s

Ja mes J . Chambers, Ja son C. Parrish, Niels H . J ensen, Deborah M. Kurrasch-Orbaugh,Danuta Marona-Lewicka, and David E. Nichols*

Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences,

Purdue University, West Lafayette, In diana 47907

Received February 11, 2003

In studies of the S AR of pheneth ylamine-type serotonin 5-HT2A receptor agonists, substitutedconformationally constrained tetrahydronaphthofurans were designed to investigate the optimalconforma tion of the 2-aminoethyl m oiety. These compounds were t ested u sing in vitro assaysfor a ffinity a t 5-HT 1A, 5-HT2A, and 5-HT2C receptors. Th e ben zofur an -conta ining a na logues, 6aa n d 6b , had significantly higher affinity for the 5-HT receptors tested than did the benzodi-hydrofura n-containing compoun ds, 4a , 4b , 5a , a n d 5b . The most potent compound (8-bromo-6-methoxy-4,5-dihydro-3 H -naphtho[1,8-bc]furan-5-yl)aminomethane, 6b , h a d K i values fordisplacement of [125 I]-DOI from 5-HT2A and 5-HT2C cloned rat receptors of 2.6 and 1.1 nM,respectively. Despite their high affinity, the compounds of this naphthofuran series lackedhigh intrinsic activity at the 5-HT2A receptor as mea sur ed using th e phosphoinositide hydrolysisassa y. The most potent compound in vitro, 6b , was t ested in the two-lever dr ug discrimina tionassa y in rat s tra ined to discrimina te LSD from saline, and failed to substitut e, a result t ypical

for compounds with low intrin sic activity. Thu s, alth ough conforma tional const raint has led tohigh-affinity 5-HT 2A ligands with partial agonist activity, all of the spatial and steric propertiesof th e ligand n ecessary for full receptor a ctivation h ave n ot yet been identified.

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

Conformational restriction of bioactive molecules isa valuable tool for investigating the topographical an dchemical features of small-molecule binding sites. Re-str ain t of flexible molecules in to distinct conformat ionsoften allows one t o test t he fun ctiona l effects of a su bsetof the total conformational space. Recently, for example,we have discovered by use of a conformational-restric-tion design m ethod th e optimal orientation for t he 2,5-

dimethoxy groups of typical ha llucinogenic arylalkyl-amines (1, DOM, X ) CH 3; DOB, X ) Br; DOTFM, X )CH 3).1-3 The optimal conformations parallel those wherethe methyl groups are tethered to the phenyl r ing inthe form of dihydrofuran rings as in compound 2.

Further investigation of arylalkylamines in our labo-ratory by the use of conformational r estriction hasrecently focused on the 2-aminoalkyl side chain.4 Con-formationally restricted analogues of arylalkylaminesin which the 2-aminoalkyl side chain has been con-strained by incorporation into a tethered r ing systemhave most recently focused on a series of 4-aminotet-

rahydronaphthofurans.4 The compounds of this serieswere considered to be “hybrid” structures of th e hal-lucinogenic arylalkylamine and ergoline families be-cause i t had been hypothesized since the late 1950sthat the A ring of LSD corresponds to the phenyl ringof the arylalkylamines and that the 5-oxygen of aryl-alkylamines serves as a bioisostere of the indole N-1nitrogen of LSD. Thus, tetrahydronaphthofuran series3 was hypothesized to mimic the A, B, and C rings of LSD.

The m ajority of the compounds in series 3, however,exhibited rather low affinity for the 5-HT

2Areceptor, an d

none of those compounds produced LSD-like effects ina drug discriminat ion anima l model believed to r eflectthe subjective behavioral effects of hallucinogenic drugs.One compound of series 3, the brominated syn analogue,did, however, possess high affinity for the 5-HT2A

receptor but lacked an LSD-like behavioral effect inanimals.4 These results suggested that the hypothesizedstructural correlation between the arylalkylamine andergoline fam ilies was in corr ect. The high bin ding a ffinitybut lack of hallucinogen-like behavioral activity of thesyn brominat ed a nalogue of series 3 indicated that this

* To whom corresp ondence should be addressed. Address: Depar t-ment of Medicinal Chemistry and Molecular Pharmacology, RHPHPhar macy Building, Room 506C, Pu rdue University, West Lafayette,IN 47907-1333. Phone: (765) 494-1461. Fax: (765) 494-1414. E-mail:drda ve@pha rma cy.pur due.edu.

3526 J . M e d . Ch e m . 2003, 46 , 3526-3535

10.1021/jm030064v CCC: $25.00 © 2003 American Chem ical SocietyPublished on Web 07/04/2003



dr ug m i ght be an ant agoni s t and l ack t he abi l i t y t oactivate the receptor.

We, t herefore, aban doned the notion of structuralsimilarity between ergolines and pheneth ylamines andconsidered a de novo approach. Specifically, new tet-rah ydronaph thofura ns, compounds 4-6, were designedon the basis of a molecular scaffold similar to that of series 3 to alter the spa tial orienta tion of the a minoethylmoiety in t he ligand. The compounds in th e new series

were designed also to exploit knowledge we had gainedabout the active orientation of the 2- and 5-alkoxysubstituent s of arylalkylamines. Most importan tly, wedecided to examine conformations of the aminoethylmoiety that were distinctly out of the aromatic r ing

plane (i.e., nonplana r).The alkoxy substituent of compounds 4-6 that cor-

responds to the 5-methoxy group of typical arylalky-lamines is incorporat ed into an appended dihydrofuranring. The peri interaction between the 6-methoxy groupand the buttressing by the adjacent methylene forcest he am i nom et hyl t o a dopt a confor m at i on t hat w ehypothesized might be nea r optimal. Thu s, compounds4-6 localize th e basic am ine m oiety t o a region of spacet hat i s out of t he pl ane def i ned by t he phenyl r i ng.Conformational restriction of the alkylamine was de-signed t o test th e hypothesis that the 2-aminoalkyl sidechain of ar ylalkylamines adopts an anti-periplanarconformat ion upon binding to the 5-HT2A receptor

agonist binding site. The hypothesis that the aminemust lie in an anti-periplanar conformation and out of the pla ne of th e phenyl ring is based, in par t, on solutionNMR studies of amphetamine.5

Molecular modeling of compounds 4b , 5b , and 6b a ndthe compound th at exhibited the h ighest 5-HT2A recep-tor affinity of series 3, the syn brominated analogue,revealed that compounds 4b , 5b , a n d 6b adopt anenergetically favored conformation that resembles closelyan anti-periplanar conformation (Figure 1). The threecompounds 4b , 5b , and 6b all position th e basic amineout of t he plane of t he phenyl r i ng (s em iem pir i calgenerated models using Spart an 6 an d AM-1 potent ials).Although syn br omina ted 3 can mimic an a nti-peripla-

nar conformation as well, the basic amine is locked intoa posit ion closer to the plane of the phenyl r ing tha nthe compounds of the presently described series.

In addition to investigation of th e optimal 2-ami-noalkyl conformat ion, compound s 4-6 were designed toprobe the effect of aromatization of the dihydrofuranring. It wa s predicted, on the basis of previous work,2,3

t hat ar om at i zat i on of t he dihydr ofur an r i ng w oul dincrease the affinity and efficacy of these compoundsat the 5-HT2A receptor.

C h e m i s t r y

Retrosynthetic analysis of the target naphthofurans4-6 revealed that all s ix racemic compounds of the

proposed series could be derived from a single interme-diate, (()-4a , also one of the target compounds. Thenonaromatized compounds of this naphthofuran seriescontain two stereocenters and thus two pairs of enan-t i om er s, t he s yn and ant i r acem at es. A di ver gentsynthetic scheme was designed to generate separatelyeach enantiomeric pair. Various combinations of bro-mina tion, N-protection/deprotection, ar omatization, a ndreduction of (()-4a were envisioned to generate the

remainder of the series from this divergent intermedi-ate.

Synthesis of compounds (()-4, (()-5, and (()-6 (Scheme1) commenced with the regioselective bromination of 4-meth oxyphenol (7) to afford 8 util izing a methoddescribed by Curran et al .7 Substituted phenol 8 wa ssubsequently O-alkylated using Finkelstein conditionswith methyl 4-bromocrotonate (9) to yield 10 . Ut ilizat ionof exactly 1 equiv of pota ssium car bonate a nd a r eactiontemperature of 0 °C resolved difficulties associated withthe synthesis of this γ-aryloxycrotonate.8 The methylester 10 was then subjected to radical reaction condi-tions to effect ring closure and afford benzofuran (()-11 .9 Limitations imposed by t he high-dilution require-

ments of typical radical reactions were overcome byemploying a reaction design t hat maintained a lowconcentration of tributyltin hydride, thus resulting infewer side reactions. Benzofuran (()-11 was then subjected to base hydrolysis to afford substituted acetic acid(()-12 .4 Next, (()-12 was converted to the acyl chloridewith oxalyl chloride and subsequently to diazoketone(()-13 by reaction with diazometh an e.4 Diazoketone (()-13 w as t hen s ubject ed t o Wol ff r ear r angem ent bytreatment with si lver benzoate and tr iethylamine inmethanol to afford the homologated methyl ester (()-14 .10 E s t er (()-14 was subsequently hydrolyzed withbase to afford propionic acid (()-15.

An alternative approach (Scheme 2) was also em-

F i g u r e 1 . Illustration of out-of-plane, anti-periplanar con-form at ion for compound s 4b , 5b , and 6b compared to in-plane,

anti-periplanar naphthofuran syn-Br -3 as viewed from the 6,7-edge of the molecule. The am ino group is at the uppermostt o p o f each s t ru ct u re, w i t h t h e b ro mi n e at o m at t h e v erybottom. Hydrogen atoms have been omitted for clarity.

5-HT 2A/ 2C R eceptor L igan d s J ou rn al of M ed icin al Ch em istry, 2003, V ol. 46, N o. 16 3527



ployed to generate a larger quantity of propionic acid(()-15 . This synthetic route involved reduction of methylester (()-11 to alcohol (()-16 using sodium aluminumhydr ide, followed by protection of th e a lcohol to a ffordtosylat e (()-17 .11 The tosylate group was then displacedby reaction with potassium cyanide in ethanol to afford(()-18 , followed by base hydrolysis of t he nitr i le togenerate homologated acid (()-15 .

Propionic acid (()-15 was then converted to the acylchloride a nd subjected to Fr iedel-Crafts conditions t oeffect ring closure and afford tricyclic ketone (()-19

(Schem e 3).12,13 The k etone moiety of (()-19 was func-tionalized utilizing reaction with trimethylsilyl cyanide

an d zinc iodide in dichlorometha ne.14,15

The adduct thatwas formed by this reaction could not be purified andwas reduced directly with sodium a luminum hydride toafford amino alcohol (()-20 . T his al cohol w as t hendehydrat ed u sing acidic conditions to yield substitut edaminomethylstyrene (()-21 . Subsequently, catalyticredu ction of (()-21 to afford exclusively syn p rodu ct (()-4a was a ccomplished. It was ant icipated t hat reductionof styrene (()-21 would result in a mixture of syn andanti products (()-4a and (()-5a tha t would be amena bleto separation by chromatography. Interestingly, how-ever, 1H N M R exam inat i on of t he pr oduct of t hi sreaction indicated the absence of the anti epimer. X-rayan alysis of hydrochloride sa lt crysta ls of compoun d (()-

4a confirmed the syn geometry of the product. I t ishypothesized that the selectivity observed in the cata-

lytic reduction of (()-21 ar ose from steric-induced facialselectivity. The occurrence of facial selectivity in com-pounds t hat have structures similar to styrene (()-21

has been observed an d reported in th e l i terature.16,17

Synthesis of the remaining target compounds of thecurrent naphthofuran series began with aromatic bro-mina tion of (()-4a to afford the syn br ominat ed product(()-4b (Schem e 4). N ext , a s am ple of (()-4b wa sN-protected to afford (()-22 . Selective oxidation of thedihydrofura n ring of (()-22 was accomplished utilizingrea ction with exactly 1 equiv of DDQ in d ioxan e to a fford(()-23 . Base hydrolysis was then employed to removethe N -protecting group and liberate t he product (()-6b .

Finally, a portion of (()-4a was N-protected as thetr ifluoroacetamide to afford (()-24 (Scheme 5). Trifluo-roacetamide (()-24 was su bsequent ly oxidized ut ilizingreaction with 1 equiv of DDQ to afford (()-25 . T heN-protecting group of (()-25 was th en removed by basehydr olysis to a fford (()-6a . A furt her review of litera tur eperta ining to facially selective hydrogenat ion suggestedthat the two target an ti products, (()-5a and (()-5b , of the current series could be obtained by ut ilizing condi-tions th at favor cata lyst coordinat ion with the pr imaryamine of the substrate molecule.18-20 It was an ticipatedthat metal-mediated reduction in a nonpolar solventw oul d enable t he bas ic am i ne m oi et y of (()-6a t ointeract electronically with the metal catalyst, and thus,

hydrogen delivery from the metal would occur on thes am e f ace as t he am i no gr oup, r es ul t i ng i n t he ant iproduct (()-5a .20 Indeed, cata lytic hydrogenation of pr im ar y a m i ne (()-6a in hexane proceeded as antici-pat ed and 1H NMR studies of the crude product indi-cated that the reduction had produced an approximately97:3 r atio of ant i/syn isomers. Th e 1H NMR spectra of (()-4a and (()-5a are dissimilar in the spectral regionthat corresponds to aliphatic protons. The benzylicprotons (of C-2a and C-5) of (()-4a a p pe a r i n t h espectrum as two distinct groups of peaks at approxi-mately 2.9 and 3.2 ppm, whereas the correspondingprotons of (()-5a a r e f ou n d a s a m u lt ip le t a t a p -proximat ely 3.4 ppm. Th ere a re a lso significan t splitting

S c h e m e 1

S c h e m e 2

3528 J ou rn al of M ed icin al Ch em istry, 2003, V ol. 46, N o. 16 Ch am bers et al.



pattern and frequency differences for the methyleneprotons (of C-3 and C-4) of the two products. A sampleof (()-5a was then brominated using bromine in meth a-nol to afford anti product (()-5b . The anti geometriesof compounds (()-5a a n d (()-5b were ultimately con-firm ed by X-ray an alysis of a crysta l of the t osylat e saltof (()-5b .

D i s c u s s i o n

Compounds (()-4 - (()-6 were assa yed for their a bilityt o b i n d t o t h e 5 - H T 2A r ecept or and t o a ct i vat e t he

phospholipase C pathway by quantification of IP 3 a c-

cumula tion. The receptor a ffinity r esults (Table 1) show

that the three brominated compounds, (()-4b , (()-5b ,

and (()-6b , possess na nomolar a ffinity for t he r eceptor.

The a nalogues lacking the bromine, (()-4a , (()-5a , and

(()-6a , h ad significan tly lower affinity for the 5-HT 2A

receptor. Direct comparison of brominated and non-

brominated coun terpa rts (e.g., (()-4b vs (()-4a , (()-5b

vs (()-5a , and (()-6b vs (()-6a ) shows that the presence

of a bromine a tom on the phenyl ring results in a 33- to

S c h e m e 3

S c h e m e 4

S c h e m e 5




81-fold increase in affinity for members of this series.A general requirement for potent hallucinogenic aryl-alkylamines is that the substituent at t he para position,the 8-position of the currently discussed series, be abulky, hydrophobic group.21-23 If one assumes th at t hesenew analogues are binding to the receptor in the samegener al or i ent at i on as t he s im pler phenet hylam i ne

congeners su ch as 1 or 2, then the disparity in affinitybetween brominated and non-brominated compounds inthis series is not surprising. Indeed, one could use thefact th at the br omine dra mat ically enh ances affinity toargue th at t hese new rigid analogues are binding in th esame region of the receptor as the prototype flexiblephenethylamines.

The benzofura n-containing compounds (()-6a a n d(()-6b exhibited significan tly h igher affinity for the5-HT2A receptor than their corresponding nonaroma-tized counterparts. Again, this result was not surprisingin light of our previous findings that extension of thecor e ar om at i c s yst em t o t he dihydr ofur an r i ngs of compounds conta ining t he tetr ah ydrobenzo[1,2-b;4,5-b′]-

difuran nucleus enhances affinity and efficacy at thisreceptor.2,3 Although i t is n ot clear how this effectoccurs, oxidation of the dihydrofuran ring increases thear om at i c s ur face ar ea of t he dr ug, per haps ai dinglipophilic partitioning into, and enhancement of hydro-phobic interactions within, th e a gonist binding site.

Finally, a comparison of 5-HT2A receptor affinityshows that ther e is little difference between th e pair of s yn and t he pair of ant i com pounds . T hese r es ult sindicate no a ppreciable difference between the bromi-nat ed syn product (()-4b compar ed to an ti product (()-5b and only about a 2-fold difference between the non-brominated syn (()-4a and ant i (()-5a ; it a ppears thatboth syn and anti compounds may be accommodated

about equ ally well within th e binding site. This similar-i ty may be explained by the observation that the synand anti compounds differ only slightly in their respec-tive thr ee-dimensiona l molecular stru cture. Comparisonof syn (()-4b and ant i (()-5b illustrates the apparentlyminor difference between them (Figure 1). Furth er,w h en t h e se s t r uct u r es a r e d ock e d in t o a r e ce n t lydeveloped homology model of the 5-HT 2A receptor,24 t hebulk of the red uced carbocyclic ring projects toward th eextracellular surface of the receptor into a region thatwould a ppear to have considera ble bulk toleran ce. Thus,th e difference with respect to binding to th e agonist siteof th e 5-HT2A receptor could also be construed as beingminor and may explain their similar pharmacology. The

aromatic compound (()-6b is likewise very similar inmolecular shape t o the respective syn and anti (()-4b

a n d (()-5b . This latter observation suggests that theincrease in binding affinity that results from aromati-zation of the dihydrofuran r ing may be largely of anelectronic or hydrophobic nat ure.

Compounds (()-4b , (()-5b , (()-6a , a n d (()-6b weretested for their ability to a ctivate t he ph ospholipase Cpathway by quantification of IP 3 accumu lation. Th esedata (Table 2) indicate that the compounds tested areonly par tial agonists in t his functiona l assay. The m ostpotent of the new compounds, (()-6b , while possessing

a submicromolar EC50, a l s o w as only a w eak par t ialagonist. The brominated compounds of the novel presentseries, (()-4b , (()-5b , a n d (()-6b , however, displayedat least 2-fold increased functional potency comparedto the previously synthesized syn brominated (()-3. Itis apparent tha t th e addition of the bromine substituentalso increases functiona l potency, an other at tribut e tha tparallels results seen in more flexible compounds sucha s 1 a n d 2. In some cases, this effect is very robust,w her e t he E C50 values of (()-6a a n d (()-6b differ byabout 100-fold. Once again, this finding could be inter-preted to mean that these new analogues are bindingin th e sam e region of the receptor as t he m ore flexibleprototype compounds.

Finally, we selected t he compound tha t wa s found tobe the m ost potent during in vitro an alysis ((()-6b ) forbehavioral tests in the two-lever drug discriminationassay in ra ts tr ained to discriminat e either saline fromLSD or saline from the hallucinogenic amphetamineDOI (Table 3). Compound (()-6b i n t hi s par adi gmdisplayed, however, only par tial substi tution at thehighest dose tested. Despite i ts high affinity for the5H T2A receptor, (()-6b did not engender LSD-likeresponding. Although in DOI-trained rats 6b elicited75% drug lever selection at the highest dose tested (6 µmol/kg), our criterion for full substit ut ion is >80% druglever selection. The lack of LSD-like in vivo activity isnot related to pha rma cokinetic factors because rat s did

T a b l e 1 . Results of Radioligand Competition Binding Studiesat Cloned Receptors

K i (SEM), nM

compd5-HT2A

(()-[125I]-DOI5-HT2C

(()-[125I]-DOI5-HT1A

[3H]-8-OH-DPAT

(()-sy n -Br-3 16(1.5) 4.0(0.2) 960(48)(()-4a 700(130) 59(0.6) 3000(460)(()-4b 8.7(0.2) 2.8(0.2) 520(67)(()-5a 310(34) 34(4.7) 1000(336)(()-5b 9.3(0.2) 4.9(0.6) 550(81)

(()-6a 200(9.2) 11(1.2) 860(170)(()-6b 2.6(0.2) 1.1(0.2) 105(6)DOB 2.2(0.3)a 2.8(0.7)a n db

psilocin 25(4.7) ndb n db

LSD 3.5(0.62) 5.5(0.31) 1.1(0.01)

a Data from Chambers et al. 3 b n d ) not determined.

T a b l e 2 . Results of IP 3 Accumulation Studies at Cloned Rat5-HT2A Receptors

compdE C50 (SEM)

at 5-HT2A, n M% max

5-HT stimulat ion (SEM)

(()-sy n -Br-3 2090(125) 63(3.0)(()-4a n da n da

(()-4b 1100(283) 49(8.0)(()-5a n da n da

(()-5b 340(65) 37(8.0)(()-6a >10000 10 @10 µM

((

)-6b 120(14) 33(2.0)DOB 72(3.6) 79(6.0)psilocin 2300(290) 46(2.4)m esca lin e 2600(360) 92(6.3)LSD 9.8(3.7) 22(2.6)

a n d ) not determined.

T a b l e 3 . Results of Drug Discrimination Studies in LSD- orDOI-Trained Rats

LSD-t r a in ed DOI-t r a in ed

compdE D50,

µm ol/k g 95% C IE D50,

µm ol/kg 95% CI

(()-6b a P Sb (25-67%) P Sb (43-75%)DOB 1.06 (0.82-1.36) 0.091 (0.0049-0.017)

a 6b was tested at doses from 1 to 6 µmol/kg (0.36-2.2 mg/kg).b P S ) partial substitution.




emit behavioral responses (25-43% drug lever selection)even at the lowest dose tested (1 µmol/kg). We havepreviously reported on a series of mescaline analoguesi n w hich L SD-l ike act i vi t y appear ed t o depend onintrinsic activity in th e PLC signaling path way.25 Withan intrinsic activity of only 33% that of serotonin, (()-6b may simply fail to provide sufficient activation of thereceptor. The partial agonist character of 4b , 5b , a n d6b was also evident when 1-10 µM concentr ations were

incubated with the 5-HT2A cells in the presence of 10 µM 5-HT and the IP 3 response was inhibited from 25%to 35% (data not shown). The lack of in vivo LSD-likeactivity thus indicates that the novel compounds re-ported here would not possess hallucinogenic activityin man.

We are presently investigating several new areas asa resu lt of th ese findings. First, alth ough rigid ana loguedesign can be very imprecise, on the basis of the effectof the bromine atom substitution in the present series,we hypothesize that the compounds are binding in therelevant region of the receptor. The lack of LSD-likeactivity leads to at least th ree hypotheses: (1) tha t th etethering bulk cannot be tolerated by the receptor, (2)

that the “side chain” is not f ixed into an appropriatedihedra l angle for helical movement in t he receptor th atwould lead to activation, an d (3) th at the side chain m ayrequire more conformational freedom for receptor acti-vation to occur. The argument against the last hypoth-esis, however, is that LSD, a very rigid molecule, isactive. We are currently pursuing the second explana-tion, with the construction of additional constrainedanalogues.

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 . All rea gents were commer cially available andw ere u s ed w it h ou t fu rt h er p u rifi cat i on u n l es s ot h erw is eindicated. Anhydrous THF was obtained by distillation frombenzophenone-sodium under nitrogen immediately before use.

Melting points were determined u sing a Thomas-Hooverappara tus and a re uncorrected. 1H NMR spectra were recordedusing either a 500 MHz Varian DRX-5000s or a 300 MHzBruker ARX-300 NMR spectrometer. Chemical shifts arereported in δ values (ppm) relative to a n internal reference(0.03%, v/v) of tetramethylsilane (TMS) in CDCl 3, except wh erenoted. Chemical ionization mass spectra (CIMS), using isobu-tane as t he carrier gas, were obtained with a Finnigan 4000s p ect ro met er. E l emen t al an al ys es w ere p erformed b y t h ePurdue University Microanalysis Laboratory and are within(0.4% of the calculated values un less otherwise noted. Thin-layer chromatography was performed using J . T. Baker flexsilica gel IB2-F, plastic-backed sheets with fluorescent indica-tor, visualizing with UV light at 254 nm an d with th e producteluting with 4:1 hexanes/ethyl acetate u nless otherwise noted.Column chromatography was carried out using silica gel 60,

230-400 mesh (J. T. Baker). All reactions were carried outunder an inert atmosphere of argon unless otherwise indicated.2 -B r o m o - 4-m e t h o x y p h e n o l ( 8 ).7 Bromine (285 g, 1.78

mol) was added dropwise to a stirred solution of 4-methox-yphenol (7) (200 g, 1.61 mol) in MeOH (840 mL) at 0 °C. Themixture was allowed to warm to room temperatu re an d stirredovernight. The solvent was removed by rotary evaporation andthe r esidue was pu rified by vacuum distillation (130 °C at 25mmHg) to afford a clear liquid (262 g, 80%); 1H NMR (300MHz, CDCl3) δ 3.72 (s, 3 H, OC H 3), 5.18 (bs, 1 H, O H ), 6.78(dd, 1 H, Ar H , J ) 8.9, 3.0 Hz), 6.92 (d, 1 H , Ar H , J ) 8.9 Hz),7.00 (d, 1 H, Ar H , J ) 3.0 Hz); MS (CI) m / z 203 (M + H ).

M e t h y l 4 -( 2 -B r o m o - 4-m e t h o x y p h e n o x y ) b u t -2 -e n o a t e(10). To a mechanically stirred solution of 8 (193 g, 950 mmol)in anhydrous acetone (1.0 L) was added potassium carbonate(131 g, 950 mmol) and potassium iodide (159 g, 958 mmol).

This mixture was stirred and cooled to 0 °C followed bydropwise a ddition over 2 h of methyl 4-bromocrotonate (170g, 951 m mol) in anhydrous acetone (1.5 L). The reactionmixture was allowed to warm to room t emperature and wasstirred for 18 h. Et 2O (500 mL) was added, and the mixturewas then vacuum-filtered. The filter cake was washed wellwith acetone (500 mL), and the filtrate was evaporated. Theresulting r esidue was dissolved in Et 2O (700 mL) and wash edwith cold, aqueous 2 N NaOH (2 × 200 mL). The organic layerwas washed with brine (200 mL), dried (MgSO 4), filtered, andevaporated to leave a light-yellow oil t hat solidified uponcooling. This solid was recrystallized from Et 2O to afford la rge,tan crystals (229 g, 76%): mp 48-49 °C; 1H N MR (300 MHz,CDCl3) δ 3.73 (s, 3 H, ArOC H 3), 3.74 (s, 3 H, CO 2C H 3), 4.66(q, 2 H, ArOC H 2, J ) 2.1 Hz), 6.27 (dt, 1 H , ArOCH2CH C H , J ) 1 5. 6, 2 .1 H z), 6 .7 7 (m, 2 H , Ar H ), 7 .0 2 ( dt , 1 H ,ArOCH 2C H CH , J ) 15.9, 3.6 Hz), 7.09 (m, 1 H, Ar H ); MS (CI)m / z 301 (M + H), Anal. (C12H 13Br O4) C, H.

(()-Methyl 2-(2,3-Dihydro-5-methoxybenzofuran-3-yl)-a c e t a t e ( (()-11). To a stirred solution of compound 10 (32.8g, 109 mmol) in anhydrous benzene (1140 mL) at reflux wasadded portionwise AIBN (1.6 g, 10 mmol). To this mixtu re wa sadded Bu 3SnH (32.3 mL, 120 mmol) dissolved in anhydrousbenzene (420 mL) dr opwise over a period of 1.5 h. The rea ctionmixture was maintained at reflux for an additional 1 h andw as t h en cool ed t o room t emp erat u re. T h e s ol ven t w asremoved by rotary evaporation, t he resulting residue wasdissolved in Et 2O (1.0 L), and a n a queous solution of potas siumfluoride (60% w/v, 150 mL) was added. This mixture wasstirred vigorously overnight. The precipitate t hat form ed wasremoved by vacuum filtrat ion, an d th e filter cake was washedwell with Et 2O (500 mL). The filtrate was evaporated andsubjected to flash chromatography (4:1 hexanes/EtOAc). Frac-tions conta ining the product were pooled and the solvent wasremoved by rotary evaporation to leave a clear oil (24 g, 98%):1H NMR (500 MHz, CDCl 3) δ 2.54 (dd, 1 H, C H 2CO 2CH 3, J )

16.8, 9.0 Hz), 2.74 (dd, 1 H, C H 2CO 2CH 3, J ) 16.8, 5.7 Hz),3.68 (s, 3 H, CO 2C H 3), 3.71 (s, 3 H, ArOC H 3), 3.80 (m, 1 H,ArC H , J ) 2.7 Hz), 4.18 (dd, 1 H, ArOC H 2, J ) 9.0 Hz), 4.68(t, 1 H, ArOC H 2, J ) 9.0 Hz), 6.68 (m, 3 H , Ar H ); MS (CI) m / z223 (M + H). Anal. (C12H 14O4) C, H.

(()-(2,3-Dihydro-5-methoxybenzofuran-3-yl)acetic Acid

((

)-12). A solution of H 2O (100 mL), EtOH (50 mL), and KOH(20.0 g, 357 mmol) was added to a solution of methyl ester(()-11 (11.9 g, 53.6 mmol) in EtOH (100 mL). This mixturew a s h e a t e d a t r e flu x for 1 h a n d t h e n coole d t o r o omtemperatur e. The solvents were removed by rotary evapora-tion, and the resulting residue was dissolved in H 2O (50 mL)and acidified with cold, concentra ted HCl. This solution wasthen extracted with CH 2Cl2 (4 × 40 mL), and th e organic layerswere pooled and washed with brine, dried (MgSO 4), filtered,an d evap orat ed t o l eave a w h it e s ol id . T h e p rod u ct w asrecrysta llized from E tOAc to yield white, gra nula r crysta ls (9.8g, 88%): mp 110-112 °C; 1H NMR (300 MHz, DMSO-d 6) δ2.54 (dd, 1 H, C H 2CO 2H , J ) 14.0, 9.5 Hz), 2.82 (dd, 1 H, C H 2-CO 2H , J ) 14.3, 5.6 Hz), 3.72 (s, 3 H, ArOC H 3), 3.75 (m, 1 H,ArC H , J ) 6.5 Hz), 4.17 (dd, 1 H, ArOC H 2, J ) 8.8, 7.1 Hz),4.69 (t, 1 H, ArOC H 2, J ) 8.9 Hz), 6.67 (s, 2 H, Ar H ), 6.91 (s,

1 H, Ar H ), 12.40 (bs, 1 H, CO 2 H ); MS (CI) m / z 209 (M+

H ).Anal. (C11H 12O4) C, H.

(()-1-Diazo-3-(2,3-dihy dro-5-met hoxy ben zofuran -3-yl)-propan-2-one ((()-13).4 To a solution of acid (()-12 (5.0 g,24.0 mmol) in anh ydrous benzene (40 mL) was a dded DMF (2drops) followed by th e dr opwise add ition of oxalyl chloride (2.72mL, 31.2 mmol). After the mixture was stirred for 3 h, thesolvent and excess oxalyl chloride were removed by rotaryevaporation. Meanwhile, an Et 2O solution of CH2N 2 cooled inan ice/salt bath was generated using an Aldrich Diazald Kitand a Diazald (16.47 g, 76.9 mmol) solution in Et 2O (105 mL)dripped into a 60 °C solution of KOH (5.25 g, 94 mmol) inCarbitol (32 mL) and H 2O (11 mL). The solid acyl chloride of (()-12 was dissolved in anhydrous Et 2O (55 mL) and addeddropwise to t he CH 2N 2 solution. After the mixture wa s stirr edat 0 °C for 1.5 h, a precipitate formed and the excess CH 2N 2




was removed by attaching the apparatus to a water aspirator.After most of the solvent h ad been removed, the remainingvolatiles were removed by rotary evaporation to leave thediazoketone as a yellow solid (5.4 g, 97%). An analytical sampleof the diazoketone was recrystallized from Et 2O : m p 70-72°C (lit.4 mp 73-74 °C); 1H NMR (300 MHz, CDCl3) δ 2.74 (m,2 H, C H 2COCHN2), 3.74 (s, 3 H , ArOC H 3), 3.90 (m, 1 H, ArC H ),4.19 (dd, 1 H, ArOC H 2, J ) 9.1, 6.0 Hz), 4.70 (t, 1 H, ArOC H 2, J ) 9.0 Hz), 5.25 (s, 1 H, CH 2CO C H N 2), 6.73 (m, 3 H, Ar H );MS (CI) m / z 233 (M + H ).

(()-Methyl 3-(2,3-Dihydro-5-met hox ybe nzofu ran-3-yl)-p r o p i o n a t e ( (()-14). To a stirr ed solution of diazoketone (()-13 (5.4 g, 23.3 mmol) in an hydrous MeOH (100 mL) was addedvery slowly a solution of silver benzoate (0.4 g, 1.7 m mol) inanhydrous Et 3N (4.0 mL). This reaction mixture was stirredov ern i gh t an d w as t h en d il u t ed w it h E t 2O (1 0 0 mL ) an dvacuum-filtered to r emove th e solids. The filter cake waswashed well with Et 2O, and th e filtrat e was concentrat ed. Theresidue was dissolved in CH 2Cl 2 (200 mL), washed with a que-o u s 0 . 5 M N aO H (2 × 4 0 mL ) an d b ri n e (2 0 mL ), d ri ed(MgSO4), fi l tered, and evaporated to leave a tan oil (5.2 g,95%): 1H NMR (500 MHz, CDCl3) δ 1.93 (m, 1 H, ArCHC H 2-CH 2), 2.08 (m, 1 H, ArCHC H 2CH 2), 2.38 (m, 2 H, ArCHCH 2-C H 2), 3.42 (m, 1 H , ArC H ), 3.67 (s, 3 H, CO 2C H 3), 3.77 (s, 3 H,ArOC H 3), 4.19 (dd, 1 H, ArCHC H 2O, J ) 8.8, 5.8 Hz), 4.59 (t,1 H, ArCHC H 2O, J ) 8.8 Hz), 6.68 (m, 2 H , Ar H ), 6.75 (s, 1 H,Ar H ); MS (CI) m / z 237 (M + H). Anal. (C13H 16O4) C, H.

(()-3-(2,3-Dihydro-5-methoxybenzofuran-3-yl)propi-o n i c A c i d ( (()-15). Methyl ester (()-14 (5.2 g, 22.0 mmol)was diss olved in E tOH (80 mL), a solution of KOH (5.0 g, 89.0mmol) in H 2O (25 mL) and Et OH (13 mL) was added, an d themixture was stirred overnight. The EtOH was removed byrotary evaporat ion, an d th e resulting mixture was diluted withH 2O (50 mL) and acidified with cold, concentrated HCl. Theresulting suspension was extracted with CH 2Cl2 (4 × 40 mL),and the organic layers were pooled, washed with brine (50 mL),dried (MgSO4), fi lt ered, an d evap orat ed t o l eav e a w h it e,amorphous solid. This solid was recrystallized from Et 2O t oafford white crysta ls (4.6 g, 94%): mp 100-101 °C; 1H NMR(500 MHz, CDCl3) δ 1.77 (m, 1 H, ArCHC H 2CH 2), 2.00 (m, 1H, ArCHC H 2CH 2), 2.34 (m, 2 H, ArCHCH 2C H 2), 3.48 (m, 1 H,ArC H ), 3.73 (s, 3 H, ArOC H 3), 4.22 (dd, 1 H, ArCHC H 2O, J )

8.7, 6.0 Hz), 4.59 (t, 1 H, ArCHC H 2O, J ) 9.5 Hz), 6.72 (m, 2

H, Ar H ), 6.91 (s, 1 H, Ar H ), 12.21 (bs, 1 H, CO 2 H ); MS (CI)m / z 223 (M + H). Anal. (C12H 14O4) C, H.

A l t e r n a t i v e R o u t e t o P r o p i o n i c A c i d (()-15. (()-2-(5-Methoxy-2,3-dihydrobenzofuran-3-yl)ethanol ((()-16). Toa st irred su spension of lithium aluminu m h ydride (32.0 g, 840mmol) in a nhydrous Et 2O (500 mL) was added dropwise ananhydrous Et 2O (450 mL) solution of methyl est er (()-11 (45.5g , 2 0 5 mmo l). T h e react i on mi xt u re w as s t irred at roomtemperatur e overnight and quenched by the careful dropwise

addition of a solution of H 2O (25 mL) in TH F (250 m L) followedby 5 N KOH (20 mL). The solids were removed by vacuumfiltrat ion through a pad of Celite, and the filter cake waswashed well with war m THF (500 mL). The solvents wer e thenremoved by r otary evaporation to leave a clear oil that washomogeneous by TLC analysis (39.0 g, 89%). An analytical

sample wa s pur ified by column chromatography (1:1 hexan es/ EtOAc as eluent): 1H NMR (500 MHz, CDCl3) δ 1.61 (bs, 1 H,CH 2O H ), 1.84 (m, 1 H, ArCHC H 2CH 2), 2.04 (m, 1 H, ArCHC H 2-CH 2), 3.56 (m, 1 H, ArC H ), 3.75 (t, 2 H, ArCHCH 2C H 2, J )

4.2 Hz), 3.76 (s, 3 H, ArOC H 3), 4.25 (dd, 1 H , ArCHC H 2O, J )

8.9, 6.5 Hz), 4.65 (t, 1 H, ArCHC H 2O, J ) 8.9 Hz), 6.68 (m, 2H, Ar H ), 6.77 (m, 1 H, Ar H ); MS (CI) m / z 195 (M + H), 177(M + H - H 2O). Anal. (C11H 14O3) C, H.

C o n t i n u a t i o n o f A l t e r n a t i v e R o u t e . (()-2-(5-Methoxy-2,3-dihydrobenzofuran-3-yl)ethyl p -Toluenesu lfonate ((()-17). A solution of Et 3N (170 mL, 1.22 m ol) in CH 2Cl 2 (400 mL)was added dropwise to a 0 °C solution of alcohol (()-16 (78.6g, 405 mmol) and p-toluenesu lfonyl chloride (163 g, 855 mm ol)in CH 2Cl2 (1300 mL). After being stirr ed for 1 h, t he r eactionmixture was allowed to warm to room temperatu re an d stirredan additional 2 h. The reaction was then quenched by the

addition of satur ated Na 2CO 3 solution (500 mL). The layerswere separated, and the aqueous phase was extracted withCH 2Cl 2 (3 × 350 mL). The organic phases were pooled andw as h ed w i t h s at u rat ed N a 2CO 3 (200 mL), dried (MgSO4),fi lt ered, an d evap orat ed t o l eave a w h it e s ol id t h at w asrecrystallized from EtOAc t o a fford white crystals (117 g,83%): mp 88-89 °C; 1H N MR (500 MHz, CDCl3) δ 1.96 (m, 1H, ArCHC H 2CH 2), 2.08 (m, 1 H, ArCHC H 2CH 2), 2.46 (s, 3 H,ArC H 3), 3.50 (m, 1 H, ArC H ), 3.75 (s, 3 H, ArOC H 3), 4.12 (m,3 H, ArCHCH 2C H 2, ArCHC H 2O), 4.52 (t, 1 H , ArCHC H 2O, J

) 9.0 Hz), 6.67 (m, 3 H, Ar H ), 7.36 (d, 2 H, Ar H , J ) 9.0 Hz),

7.81 (d, 2 H, ArH, J ) 6.0 Hz); MS (CI) m / z 349 (M + H). Anal.(C18H 20O5S) C, H.

C o n t i n u a t i o n o f A l t e r n a t i v e R o u t e . (()-3-(5-Methoxy-2,3-dihydrobe nzofuran-3-yl)propionitrile ((()-18). Potas-sium cyanide (16.0 g, 246 mmol) was added to a solution of tosylate (()-17 (58.8 g, 169 mm ol) in absolute EtOH (1.0 L).The mixture was heat ed at reflux for 9 h and t hen cooled toroom t emp erat u re. T h e s ol id s w ere remov ed b y v acu u mfiltration, and t he filter cake was washed well with EtOH. Thefiltrate volume was reduced under vacuum (ca. 400 mL) andwas then diluted with Et 2O (500 mL). This solution was thenwashed with H 2O (500 mL) and brine (200 mL), dried (MgSO4),filtered, and evaporated to leave a tan oil that was homoge-neous by TLC (33 g, 97%). An analytical sample was purifiedby column chromatography (4:1 hexanes/EtOAc). 1H NMR (500MHz, CDCl3) δ 1.98 (m, 1 H, ArCHC H 2CH 2), 2.07 (m, 1 H,ArCHC H 2CH 2), 2.41 (m, 2 H, ArCHCH 2C H 2), 3.55 (m, 1 H,ArC H ), 3.77 (s, 3 H, ArOC H 3), 4.25 (m, 1 H, ArCH C H 2O), 4.63(dt, 1 H, ArCHC H 2O, J ) 8.9, 2.6 Hz), 6.72 (m, 3 H, Ar H ); MS(CI) m / z 204 (M + H). Anal. (C12H 13NO 2) C, H, N.

C o n t i n u a t i o n o f A l t e r n a t i v e R o u t e . (()-3-(5-Methoxy-2,3-dihydrobenzofuran-3-yl)propionic Acid ((()-15). Ni -trile (()-18 (36.5 g, 180 mmol) was dissolved in EtOH (1.0 L),and then 2 N NaOH (500 mL) was added. The reaction mixturew as h eat ed a t refl u x o vern ig h t an d t h en cool ed t o r oomtemperatur e. The EtOH was removed by r otary evaporation,and the resulting suspension was cooled an d acidified bydropwise addition of 2 N HCl. The mixture was extracted withCH 2Cl 2 (4 × 400 mL). The organic layers were pooled andwashed with brine (300 mL), dried (Na 2SO 4), filtered, andevaporated to leave a white solid. Recrystallization from E tOAcgave off-white n eedles (33 g, 83%). NMR spectr al da ta an d th emelting point were identical to those reported above.

(()-6-Methoxy-2,2a,3,4-tetrahydronaphtho[1,8- bc ]furan-5-one ((()-19). Propionic acid (()-15 (33.0 g, 150 mmol) wasdissolved in anhydrous benzene (850 mL), and then DMF (4mL) was added. Oxalyl chloride (26 mL, 298 mmol) was addeddropwise to the mixture, an d the r eaction mixtur e was stirr edat room temperatur e for 4 h . The solvent and excess oxalylchloride were rem oved by rotary evaporat ion, an d th e residuewas dissolved in CH 2Cl 2 (460 mL). The reaction flask wassubmerged in an ice bath, an d then anhydrous SnCl4 (22 mL,188 mmol) was added dropwise. The ice bath was removed,and t he reaction mixture was stirred for 1 h and t hen pouredo n t o i ce (ca. 3 0 0 g ). T h e l ay ers w ere s ep arat ed , an d t h eaqueous phase was extracted with CH 2Cl 2 (3 × 250 mL). Theorganic layers were then pooled and washed with brine (200

mL), dried (MgSO4), fi ltered, and evaporated to leave a tansolid th at was r ecrystallized from Et OAc (29.7 g, 97%): mp105-106 °C; 1H NMR (500 MHz, CDCl3) δ 1.89 (dq, 1 H,ArCHC H 2CH 2, J ) 7.0, 2.0 Hz), 2.31 (m, 1 H, ArCHC H 2CH 2),2.57 (m, 1 H, ArCHCH2C H 2), 2.76 (m, 1 H, ArCHCH 2C H 2),3.68 (m, 1 H, ArC H ), 3.88 (s, 3 H, ArOC H 3), 4.10 (dd, 1 H,ArCHC H 2O, J ) 6.4, 4.5 Hz), 4.84 (t, 1 H, ArCHC H 2O, J )

8.5 Hz), 6.69 (d, 1 H , Ar H , J ) 3.6 Hz), 6.93 (d, 1 H , Ar H , J )

3.6 Hz); MS (CI) m / z 205 (M + H). Anal. (C12H 12O3) C, H.

(()-5-Aminomethyl-6-methoxy-2a,3,4,5-tetrahydro-2 H -naphtho[1,8- bc] f u r a n -5 -o l H y d r o c h l o r i d e ( (()-20). Zinciodide (0.4 g, 1.3 mmol) was added to a solution of ketone (()-19 (10.0 g, 49 m mol) in CH 2Cl 2 (500 mL). TMSCN (7.2 mL, 54mmol) was then added, and the reaction m ixture was heatedat refl u x for 1 4 h . Comp ari son of t h e IR s p ect ru m of t h ereaction mixture to the IR spectrum of the star ting material






12.9, 6.4 Hz), 3.45 (m, 1 H, ArC H CH 2N), 3.83 (s, 3 H, ArOC H 3),7.07 (s, 1 H, Ar H ), 7.48 (s, 1 H, Ar H ); MS (CI) m / z 296 (M +

H). Anal. (C13H 15BrClNO2) C, H, N.

(()- syn- N -Trifluoroacetyl-(6-methoxy-2a,3,4,5-tetrahy-dro-2 H -naphtho[1,8- bc] f u r a n -5 -y l ) a m i n o m e t h a n e ( (()-24). To a stirred suspension of the hydrochloride salt of (()-4a (3.1 g, 12.1 mmol) and 4- N , N -(dimethylamino)pyridine (0.12g, 0.98 mmol) in CH 2Cl2 (150 mL) was a dded Et 3N (6.0 mL,43.0 mmol), an d t he m ixture was cooled to 0 °C. Trifluoroaceticanh ydride (4.13 mL, 29.0 mmol) was th en added to the r eactionmixture dropwise. The reaction mixture was then allowed towarm to room temperature and stirred for 8 h. The mixturewas diluted with CH 2Cl 2 (200 mL) and washed with 2 N HCl(100 mL), satur ated NaHCO3 (100 mL), and brine (100 mL).T h e o rgan i c p h as e w as t h en d ried (MgSO4), fi l tered, andevapora ted t o leave a white solid th at was r ecrystallized fromEt 2O (3.5 g, 91%): mp 181-183 °C; 1H N MR (500 MHz, CDCl3)δ 1.51 (q, 1 H, ArCHC(3) H 2CH 2, J ) 13.0 Hz), 1.88 (dt, 1 H,ArCHC(3) H 2CH 2, J ) 9.6, 5.7 Hz), 2.01 (m, 2 H, ArCHCH 2C-(4) H 2), 3.23 (p, 1 H, ArC H CH 2N, J ) 7.0 Hz), 3.28 (m, 1 H,ArC H CH 2O ), 3 .4 0 (m , 1 H , Ar C H C H 2N ), 3 .5 7 ( p , 1 H ,ArCHC H 2N, J ) 6.3 Hz), 3.82 (s, 3 H, ArOC H 3), 3.96 (dd, 1 H,ArCHC H 2O, J ) 12.9, 8.2 Hz), 4.72 (t, 1 H, ArCHC H 2O, J )

8.1 Hz), 6.62 (s, 2 H , Ar H ), 7.64 (bs, 1 H, N H COCF 3); MS (CI)m / z 316 (M + H). Anal. (C15H 16F 3NO 3) C, H, N.

(()- N -Trifluoroacetyl-(6-methoxy-4,5-dihydro-3 H -naph-

tho[1,8- b c] f u r a n -5 -y l ) a m i n o m e t h a n e ( ((

)-25). A solutionof DDQ (2.16 g, 9.5 mmol) in dioxane (330 mL) was addedslowly to a solution of (()-24 (3.0 g, 9.5 mm ol) in dioxane (125mL). The reaction m ixture was stirred a t room temperatur efor 8 h and was then diluted with CH 2Cl2 (100 mL) and filter edthrough a short pad of silica gel. The silica gel was washedwell with CH 2Cl 2, and the filtrate and wash were reduced todryness by rotary evaporation. The dark solid t hat resultedwas subjected to column chromatography (1:1 hexanes/EtOAcas eluent), r esulting in a off-white, solid product that wasrecrystallized from Et 2O to a fford whit e, fluffy crysta ls (2.6 g,88%): mp 141-142 °C; 1H NMR (500 MHz, CDCl3) δ 1.95 (m,1 H, ArCCH 2C H 2), 2.13 (dt, 1 H, ArCCH 2C H 2, J ) 13.9, 2.2Hz), 2.73 (t, 1 H, ArCC H 2, J ) 13.6 Hz), 2.85 (d, 1 H , ArCC H 2, J ) 16.2 Hz), 3.37 (dt, 1 H, ArCHC H 2N, J ) 12.3, 3.5 Hz),3.52 (m, 1 H, ArC H CH 2N), 3.68 (m, 1 H, ArCHC H 2N), 3.92 (s,3 H, ArOC H

3), 6.88 (d, 1 H, Ar H , J ) 8.8 Hz), 7.28 (d, 1 H,

Ar H , J ) 8.8 Hz), 7.35 (s, 1 H, Ar H ), 7.83 (bs, 1 H, N H COCF 3);MS (CI) m / z 314 (M + H). Anal. (C15H 14F 3NO 3) C, H, N.

(()-(6-Methoxy-4,5-dihydro-3 H -naphtho[1,8- bc]furan-5 - y l ) a m i n o m e t h a n e H e m i o x a l a t e ( (()-6a). A solution of (()-25 (1.7 g, 5.4 mmol) in MeOH (250 mL) was cooled to 0°C, an d then 5 N KOH solution (30 m L) was a dded slowly.The reaction mixture was allowed to warm to room temper-atu re and st irred overnight, and th en the MeOH was removedby rotary evaporation. The residue was diluted with H 2O (25mL) and extracted with Et 2O (4 × 100 mL), dried (Na 2SO 4),filtered, an d evaporated to a fford a clear oil. This oil wasdissolved in Et 2O (100 mL), filtered through a plug of glasswool, and neutralized by the slow addition of oxalic acid (54mL, 0.1 M in MeOH). The solvents were removed, and theresulting white residue wa s recrysta llized from MeOH to afford

the hemioxalat e salt (0.9 g, 59%): mp 243 °C;1

H NMR (500MHz, CD3OD ) δ 1.82 (m, 1 H, ArCCH 2C H 2), 2.17 (dt, 1 H,ArCCH 2C H 2, J ) 11.9, 2.8 Hz), 2.69 (t, 1 H , ArCC H 2, J ) 13.6Hz), 2.78 (dt, 1 H, ArCC H 2, J ) 12.0, 3.1 Hz), 3.03 (m, 2 H,ArCHC H 2N, J ) 13.0, 6.4 Hz), 3.46 (m, 1 H, ArC H CH 2), 3.81(s, 3 H, ArOC H 3), 6.88 (d, 1 H, Ar H , J ) 8.7 Hz), 7.19 (d, 1 H,Ar H , J ) 8.7 Hz), 7.35 (s, 1 H, Ar H ); MS (CI) m / z 218 (M +

H). Anal. (C15H 17NO 6) C, H, N.

(()- ant i -(6-Methoxy-2a,3,4,5-tetrahydro-2 H - n a p h t h o -[1,8- b c] f u r a n -5 -y l ) a m i n o m e t h a n e H y d r o c h l o r i d e ( (()-5a). T h e p ri mary ami n e of (()-6a (0.17 g, 0.83 mmol) wasd is s ol ved i n h exan e (2 00 mL ) an d ad d ed t o a Parr fl ask containing 10% Pd/C (0.10 g) and hexane (5 mL). The flask was then pressurized and shaken at 65 psi of H 2 for 24 h. Thesolution was filtered through Celite, the filter cake was washedwell with EtOH, and the solvents were removed by rotary

evaporation to leave a clear oil . This oil was dissolved inanhydrous Et 2O (100 mL) an d filtered th rough a small plugof glass wool. The solution was then acidified with ethanolic 1N H Cl an d t h e p reci pi t at e t h at formed w as col lect ed b yvacuum filtration an d then recrystallized from i-Pr OH to affordneedle-like crystals (0.14 g, 81%): mp 185-186 °C; 1H NMR(500 MHz, D2O) δ 1.22 (q, 1 H, ArCHC(3) H 2CH 2, J ) 11.2 Hz),1.52 (q, 1 H, ArCHC(3) H 2CH 2, J ) 9.7 Hz), 2.18 (dd, 1 H,ArCHCH2C(4) H 2, J ) 12.5, 3.3 Hz), 2.30 (d, 1 H, ArCHCH 2C-(4) H 2, J ) 12.2 Hz), 3.30 (dd, 2 H, C H 2N, J ) 12.4, 7.6 Hz),3.34 (m, 1 H, ArC H CH 2O), 3.38 (m, 1 H, ArC H CH 2N), 3.78 (s,3 H, ArOC H 3), 3.97 (t, 1 H, ArCHC H 2O, J ) 10.4 Hz), 4.84 (t,1 H, ArCHC H 2O, J ) 8.4 Hz), 6.71 (d, 1 H, Ar H , J ) 8.3 Hz),6.77 (d, 1 H, Ar H , J ) 8.5 Hz); MS (CI) m / z 220 (M + H), 203(M + H - NH 3). Anal. (C13H 18ClNO2) C, H, N.

(()- an t i-(8-Bromo-6-methoxy-2a,3,4,5-tetrahydro-2 H -n a p h t h o [ 1 , 8 - b c] fu r a n -5 -y l ) a m in o m e t h a n e H y d r o c h l o -r i d e ( (()-5b). Bromi n e (0 .1 6 g , 1 . 0 mmo l, 0 .2 N MeO Hsolution) was added dropwise to a solution of the hydrochloridesalt of (()-5a (0.26 g, 1.0 mmol) in MeOH (25 mL) at 0 °C.The reaction mixture was stirred and allowed to warm to roomtemperature overnight. Et 2O (7 5 mL ) w as ad d ed , an d t h eprecipitate that formed was collected by vacuum filtration.This solid was stirred in concentrated NH 4OH solution (100mL) and extracted with E t 2O (4 × 100 mL). The organic phaseswere pooled and dried (Na 2SO 4), filtered, and evaporated toleave a clear oil. This oil was dissolved in anhydrous Et 2O,filtered through a small plug of glass wool, and acidified with1 N ethanolic HCl. The precipitate was collected by vacuumfiltrat ion a nd t hen recrystallized from i-PrOH to yield needle-like crystals (0.26 g, 78%): mp 231-232 °C; 1H NMR (500MHz, CD3OD ) δ 1.25 (1 H, q, ArCHC(3) H 2CH 2, J ) 12.1 Hz),1.53 (1 H, m, ArCHC(3) H 2CH 2), 2.18 (1 H, dd, ArCHCH 2C(4)- H 2, J ) 12.3, 3.8 Hz), 2.25 (1 H, m, ArCHCH 2C(4) H 2), 3.16 (1H, m, C H 2N), 3.37 (1 H, m, C H 2N), 3.43 (1 H, m, ArC H CH 2N),3.76 (3 H, s, ArOC H 3), 3.96 (1 H, t , ArCHC H 2O, J ) 8.8 Hz),4.75 (1 H, t , ArCHC H 2O, J ) 7.9 Hz), 6.80 (1 H, s, Ar H ); MS(CI) m / z 298 (M + H). Anal. (C13H 17BrClNO2) C, H, N.

P h a r m a c o lo g y Me t h o d s : C e ll C u lt u r e . Cells stablytran sfected to express the rat 5-HT2A, rat 5-HT2C, or h u m an5-HT1A receptor were maintained in Dulbecco’s minimumessential medium, containing 10% dialyzed fetal bovine serum

(Gibco BRL) and supplement ed with L-glutamine, Pen/Strep,and Geneticin.26 The cells were cultured at 37 °C in a H 2Osatu rat ed atm osphere of 95% air a nd 5% CO2. For radioligandbinding assays, cells were split into 100 mm culture disheswhen they reached 90% confluency. Upon reaching 100%confluency in the culture dishes, the cells were washed withster ile filtered phospha te-buffered solution and left to incubatein serum-free Opti-MEM for 5 h. After incubation, the cellswere ha rvested by centrifugation (15000g, 20 min) and placedi mmed iat ely i n a freezer at -8 0 ° C u n t i l t h e a s s a y w a sperformed. For IP 3 accumulation experiments, the cells wereseeded into 24-well plates and assays were performed when70% confluency was achieved.

R a d i o r e c e p t o r C o m p e t i t i o n A s s a y s . For saturation as-says, 0.125-5.0 nM [125 I]-DOI was used for the 5-HT2A a n d5-HT2C receptors and 0.25-10 nM [3H]-8-OH-DPAT was used

for the 5-HT1A receptor. The total volume of the assay was 250 µL. Nonspecific binding was defined in the presence of 10 µMcinanserin (rat 5-HT2A receptor expressing cells), 10 µMmianserin (rat 5-HT2C receptor expressing cells), or 10 µM5-HT (human 5-HT1A receptor expressing cells). Competitionbinding experiment s were carr ied out in a total volume of 500 µL w it h 0 .20 n M [125 I]-D OI or 2 .0 n M [3H]-8-OH-DPAT.Previously harvested cells were resuspended and added to eachwell containing a ssay buffer (50 mM Tr is, 0.1 mM EDTA, 10mM MgCl2; pH 7.4), radioligand, and test compound (or in thecase of the sat ura tion assays, cinan serin, mianserin, or 5-HT).Incubation was carr ied out a t 25 °C for 60 min and t erminat edby rapid filtra tion using a prechilled Packard 96-well ha rvesterwith GF/B Uni-filters that had been preincubated for 30 minin 0.3% polyethylenimine. The filters wer e rinsed using chilledw as h b u ffer (1 0 mM T ri s , 1 5 4 mM N aCl ) an d l eft t o d ry




overnight. The following day, Microscint-O was added andradioactivity was determined using a TopCount (Packard)scintillation counter. GraphPad Prism (GraphPa d Software,San D iego, CA ) w as u s ed t o an al y ze t h e s at u rat i on an dcompetition binding curves.

I n o s i t o l Tr i p h o s p h a t e A c c u m u l a t i o n S t u d i e s i n C e l lsE x p r e s s i n g t h e 5 -H T2A R e c e p t o r . Accumu lation of inositolphosphates was determined using a modified version of apreviously published protocol.27 Briefly, cells expressing th erat 5 -H T2A receptor were labeled for 18-2 0 h i n C R MLmedium containing 1.0 µCi/mL [3H]-myo-inositol. After pre-

treatm ent of the cells with 10 µM pargyline/10 m M LiCl for15 min, the cells were exposed to a test dr ug for 30 m in at 37°C under an atmosphere of 95% O 2 and 5% CO2. T h e as s aywas terminated by aspirating the m edium and a dding 10 mMformic acid. Following incubation for 16 h at 4 °C, the [ 3H ]-inositol phosphates were separated from the cellular debrison Dowex-1 ion-exchange columns and eluted with 1.0 Mammonium forma te and 0.10 M formic acid. The vials werecounted for trit ium using a TriCarb scintillation counter(Packard Instrument Corp.).

D r u g D i s c r i m i n a t i o n S t u d i e s . Male Sprague-Dawleyrats (Harlan Laboratories, Indianapolis, IN) weighing 200-220 g at the beginning of the drug discrimination study weredivided into six groups (n ) 8-15 per group) and trained todiscriminate LSD tartrate or DOI hydrochloride from saline.None of the rats had previously received drugs or behavioral

training. Water was freely available in the individual homecages, and a rationed amount of supplemental feed (PurinaLab Blox) was made available after experimental sessions tomaint ain approximately 80% of free-feeding weight.

Six stan dard experimental chambers (model E10-10RF,Coulbourn Instrum ents, Lehigh Valley, P A) consisted of modular t est cages enclosed within sound-att enua ted cubicleswith fans for ventilation and background white noise. A whitehouse light was centered n ear t he top of the front pa nel of thecage, which was also equipped with two response levers sep-arated by a food hopper, all positioned 2.5 cm above the floor.

A fixed ra tio (FR) 50 schedule of food r einforcement (Noyes45 mg of dustless pellets) in a two-lever paradigm was used.The complete dru g discrimination procedure has been de-scribed in detail elsewhere.28

A c k n o w l e d g m e n t . We thank Stewart Fr escas andAmjad Qandil for their helpful contributions and dis-cussions perta ining to this work. We also than k for t heirkind donations Dr. David Julius (UCSF) for the 5-HT 2A

and 5-HT2C cell lines as well as Dr. Christopher Harber(Pharmacia) for the 5-HT1A cell line. This work wassupported by Gran t DA02189 from NIDA and by a gra ntfrom the Heffter Research Institute.

S u p p o r t i n g I n f o r m a t i o n A v a i la b l e : X-ray crystallo-graphic information for (()-4a a n d (()-5b . This material isavailable free of charge via t he In tern et a t htt p://pubs.acs.org.

R e f e r e n c e s

(1) Monte, A. P.; Marona -Lewicka, D.; Cozzi, N. V.; Nichols, D. E.Synthesis a nd pharma cological examination of benzofuran,

indan, and tetralin analogues of 3,4-(methylenedioxy)amphet-amine. J . M e d . Ch em. 1993, 36 , 3700-3706.(2) Par ker, M. A.; Marona-Lewicka, D.; Lucaites, V. L.; Nelson, D.

L.; Nichols, D. E. A novel (benzodifuran yl)aminoalkane withextremely potent activity at the 5-HT 2A receptor. J. Med. Chem.1998, 41 , 5148-5149.

(3) Cham ber s, J . J . ; Kur r asch-Or baugh, D. M.; Par ker , M. A.;Nichols, D. E . Ena ntiospecific synth esis a nd pharm acologicalevaluation of a series of super-potent, conformationally restricted5-HT(2A/2C) receptor agonists. J. Med. Chem. 2001, 44, 1003-1010.

(4) Monte, A. P.; Marona-Lewicka, D.; Lewis, M. M.; Mailman, R.B.; Wainscott, D. B.; Nelson, D. L.; Nichols, D. E. Substitutednapht hofuran s as hallucinogenic phenethylamine-ergoline hy-brid molecules with unexpected muscarinic antagonist activity.

J. Med. Chem. 1998, 41 , 2134-2145.(5) Makriyann is, A.; Knittel, J. Conforma tional-ana lysis of amphet-

am i ne i n sol ut ion based on un am biguous assignm ent of t hediastereotopic benzylic protons in the H-1-NMR spectra. Tetra-hedron Lett. 1981, 22 , 4631-4634.

(6) Spartan 5.0; Wavefunction, Inc., 18401 Von Karman Ave, Ste370, Irvine, California, 92612.

(7) Curran, D. P.; Xu, J. o-Bromo- p-methoxyphenyl ethers. Protect-ing/radical translocating (PRT) groups that generate radicalsfrom C-H bonds β to oxygen atoms. J . Am . Ch em. S o c. 1996 ,118 , 3142-3147.

(8) Sunitha, K.; Balasubramanian, K. K. A reinvestigation of theClaisen rearra ngement of methyl gamma-aryloxycrotonates. Aconvenient synthesis of 3-ethylidenebenzofuran-2(3H)-ones. Tet-rahedron 1987, 43 , 3269-3278.

(9) Sha nka ran , K.; Sloan, C. P.; Snieckus, V. Synth etic connectionsto t he aromatic directed meta lation reaction. Radical-induced

cyclization to su bstituted benzofurans, benzopyrans, and furo-pyridines. Tetrahedron Lett. 1985, 26 , 6001-6004.(10) Abraham, D. J .; Kennedy, P. E.; Mehanna , A. S.; Patwa, D. C.;

Wil li am s, F. L. Desi gn, synt hesi s, and t est ing of pot ent i alantisickling agents. 4. Str ucture-activity relationships of ben-zyloxy and phenoxy acids. J . M ed . Ch e m. 1984, 27 , 967-978.

(11) Barnett, C. J .; Wilson, T. M.; Wendel, S. R.; Winningham, M.J.; Deeter, J. B. Asymmetric synthesis of Lometrexol ((6 R )-5,-10-dideaza-5,6,7,8-tetrahydrofolic acid). J. Org. Chem. 1994, 59 ,7038-7045.

(12) Horaguchi, S.; Mamoru, H.; Suzuki, T. Fu ran derivatives. IV.On the effects of substituents in t he synthesis of 4,5-dihydro-3 H -naphtho[1,8-bc]furans. Bull. Chem. Soc. Jpn. 1982 , 55 , 865-869.

( 13) Fl augh, M. E.; Mul l en, D. L.; Ful l er , R. W.; Mason, N. R.6-Substituted 1,3,4,5-tetr ahydr obenz[cd ]indol-4-amines: potentserotonin agonists. J . M ed . Ch e m. 1988, 31 , 1746-1753.

(14) Grouta s, W. C.; Felker, D. Synthet ic applications of cyanotrim-ethylsilane, iodotrimethylsilane, azidotrimethylsilane, and me-

thylthiotrimethylsilane. Synthesis 1980, 11 , 861-868.(15) DeBernardis, J. F.; Kyncl, J. J .; Basha, F. Z.; Arendsen, D. L.;

Mar t i n, Y. C.; Wi nn, M.; Ker km an, D. J . Conf or m at i onal l ydefined adrenergic agents. 2. Catechol imidazoline derivatives:biological effects at R1 a n d R2 adrenergic receptors. J . M e d .Chem. 1986, 29 , 463-467.

(16) Ha efliger, W. V.; Kloppner , E. Stereospezifische synt hese einerneuen morphin-teilstrukt ur (Stereospecific synthesis of a newmorphine stru cture). Helv. Chim. A cta 1982, 65 , 1837-1852.

( 17) Leanna, M. R.; Mar t i nel l i , M. J . ; Var i e, D. L.; Kr ess, T. J .Diaster oselectivity in ergoline synthesis: A face selective ep-oxidation. Tetrahedron Lett. 1989, 30 , 3935-3938.

(18) Thompson, H. W. Stereochemical control of reductions. Thedirective effect of carbomethoxy vs h ydroxymethyl groups incatalytic hydrogenation. J. Org. Chem. 1971, 36 , 2577-2581.

(19) Thompson, H. W.; McPher son, E.; Lences, B. L. Stereochemicalcontrol of reductions. 5. Effects of electron density and solventon group haptophilicity. J. Org. Chem. 1976, 41 , 2903-2906.

(20) Rana de, V. S.; Consiglio, G.; Prins, R. Fu nctional-group-directeddiastereoselective hydrogenation of aromatic compounds. 2(1). J. Org. Chem. 2000, 65 , 1132-1138.

( 21) Shul gi n, A. T.; Sar gent , T.; Nar anj o, C. St r uct ur esactivityrelationships of one-ring psychotomimetics. Na t u r e 1969, 22 1,537-541.

(22) Nichols, D. E.; Frescas, S.; Marona -Lewicka, D.; Hua ng, X.; Roth ,B. L.; Gud elsky, G. A.; Na sh, J . F . 1-(2,5-Dimeth oxy-4-(trifluo-romethyl)phenyl)-2-aminopropane: a potent serotonin 5-HT2A/2C

agonist. J . M e d . Ch em. 1994, 37 , 4346-4351.(23) Monte, A. P.; Marona-Lewicka, D.; Kantha samy, A.; Sanders-

Bush, E.; Nichols, D. E. Stereoselective LSD-like activity in aseries of D-lysergic acid am ides of ( R )- and (S )-2-aminoalkanes.

J. Med. Chem. 1995, 38 , 958-966.(24) Cham bers, J . J.; Nichols, D. E. A homology-based m odel of the

hum an 5-HT2A receptor derived from an in silico activatedG-protein coupled receptor. J. Comput.-Aided Mol. Des. 2002,16 , 511-520.

(25) Monte, A. P.; Waldma n, S. R.; Marona -Lewicka, D.; Wainscott,

D. B.; Nelson, D. L.; Sanders-Bush, E.; Nichols, D. E. Dihy-drobenzofura n a nalogues of hallucinogens. 4. Mescaline deriva-tives. J . M ed . Ch e m. 1997, 40 , 2997-3008.

(26) Ju lius, D.; Huang, K. N.; Livelli, T. J.; Axel, R.; J essell, T. M.The 5HT2 receptor defines a family of structurally distinct butfunctionally conserved serotonin receptors. Proc. Natl. Acad. S ci.U.S.A. 1990, 87 , 928-932.

(27) Berg, K. A.; Clarke, W. P.; Sailstad, C.; Saltzma n, A.; Maaya ni,S. Signal tr ansdu ction differences between 5-hydroxytryptaminetype 2A and type 2C receptor systems. Mol. Pharmacol. 1994,46 , 477-484.

(28) Marona-Lewicka, D.; Kurrasch-Orbaugh, D. M.; Selken, J . R.;Cumbay, M. G.; Lisnicchia, J. G.; Nichols, D. E. Reevaluationof lisuride pharma cology: 5-hydroxytryptamine1A r eceptor-mediated behavioral effects overlap its other properties. Psy-chopharmacology 2002, 164 , 93-107.

JM030064V


Documents

James J. Chambers et al- Synthesis and Pharmacological Characterization of a Series of Geometrically Constrained 5-HT2A/2C Receptor Ligands