The selective 5-HT(1B) receptor inverse agonist 1-methyl-5-[[2'-methyl- 4'(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-yl]carbonyl]-2,3,6,7- tetrahydrospiro[furo[2,3-f]indole-3,4'-piperidine] (SB-224289) potently blocks terminal 5-HT autoreceptor function bo

1218

J . Med. Chem. 1998, 41, 1218-1235

Th e Selective 5-HT_1BRecep tor In ver se Agon ist 1′-Meth yl-5-[[2′-m eth yl-4′-

(5-m eth yl-1,2,4-oxa d ia zol-3-yl)bip h en yl-4-yl]ca r bon yl]-2,3,6,7-tetr a h yd r o-

sp ir o[fu r o[2,3-f]in d ole-3,4′-p ip er id in e] (SB-224289) P oten tly Block s Ter m in a l

5-HT Au tor ecep tor F u n ction Both in Vitr o a n d in Vivo

Laramie M. Gaster,*^,‡Frank E. Blaney,^‡Susannah Davies,^‡D. Malcolm Duckworth,^‡Peter Ham,^‡

Sarah J enkins,^‡Andrew J . J ennings,^‡Graham F. J oiner,^‡Frank D. King,^‡Keith R. Mulholland,^‡

Paul A. Wyman,^‡J im J . Hagan,^†J on Hatcher,^†Brian J . J ones,^†Derek N. Middlemiss,^†Gary W. Price,^†

Graham Riley,^†Claire Roberts,^†Carol Routledge,^†J ulie Selkirk,^†and Paula D. Slade^†

SmithKline Beecham Pharmaceuticals, Discovery Research, New Frontiers Science Park (North), Third Avenue,

Harlow, Essex CM19 5AW, England

Received J uly 14, 1997

5-HT₁receptors are members of the G-protein-coupled receptor superfamily and are negatively

linked to adenylyl cyclase activity. The human 5-HT_1Band 5-HT_1Dreceptors (previously known

as 5-HT_1Dâand 5-HT_1DR, respectively), although encoded by two distinct genes, are structurally

very similar. Pharmacologically, these two receptors have been differentiated using nonselective

chemical tools such as ketanserin and ritanserin, but the absence of truly selective agents has

meant that the precise function of the 5-HT_1Band 5-HT_1Dreceptors has not been defined. In

this paper we describe how, using computational chemistry models as a guide, the nonselective

5-HT_1B/5-HT_1Dreceptor antagonist 4 was structurally modified to produce the selective 5-HT_1B

receptor inverse agonist 5, 1′-methyl-5-[[2′-methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-

4-yl]carbonyl]-2,3,6,7-tetrahydrospiro[furo[2,3-f]indole-3,4′-piperidine] (SB-224289). This com-

pound is a potent antagonist of terminal 5-HT autoreceptor function both in vitro and in vivo.

Ch a r t 1

In tr od u ction

5-HT (5-hydroxytryptamine, serotonin) receptors are

classified into seven distinct groups¹(5-HT_1-7). Al-

though 5-HT₃receptors belong to the family of ligand-

gated ion channels, the remainder are all members of

the G-protein-coupled receptor (GPCR) superfamily.

5-HT₁and 5-HT₂receptors have been further divided

into subtypes based on considerations of structural

features, functional pharmacology, and secondary mes-

senger coupling systems. The GPCRs of the 5-HT₁

group are negatively coupled to adenylyl cyclase and

have been characterized as 5-HT_1A, 5-HT_1B, 5-HT_1D

,

2-5

5-HT_1E, and 5-HT_1F

.

The human 5-HT_1Band 5-HT_1D

receptors (previously known as 5-HT_1Dâand 5-HT_1DR

,

respectively) are especially similar in sequence despite

being encoded by two distinct genes.^6,7Although human

5-HT_1Band 5-HT_1Dreceptors have been pharmacologi-

cally differentiated using nonselective 5-HT_1B/Dreceptor

antagonists such as ketanserin (1) and ritanserin^8-10

(2) (Chart 1), the precise function of these receptors

remains undefined, and progress toward this has been

hampered by the lack of selective ligands. The 5-HT_1B

receptor is of particular interest since it has been

demonstrated that this receptor is present in the brain

11

in higher densities than 5-HT_1D

.

It has also been

shown that the majority of terminal 5-HT autoreceptors

in rat, guinea pig, and human are of the 5-HT_1Breceptor

subtype and that these receptors elicit negative feedback

control on the release of 5-HT from serotonergic nerve

* To whom correspondence should be addressed.

^‡Department of Medicinal Chemistry.

terminals in the central nervous system (CNS).^12-17

Since there is evidence that enhancement of 5-HT

^†Department of Neuroscience Research.

Published on Web 03/20/1998

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1219

F igu r e 1. Attempted docking of 4 into model of human 5-HT_1Breceptor with piperazine binding to Asp 129 on TM3.

neurotransmission underlies the therapeutic response

to several types of antidepressant agent,^18,19it has been

postulated that a selective brain-penetrant 5-HT_1B

receptor antagonist might act as a fast-acting antide-

pressant.²⁰The 5-HT_1B/Dreceptor antagonist methio-

thepin (3) and the 5-HT_1B/Dreceptor antagonist 4 (GR

127935) (Chart 1) have been reported to increase

extracellular 5-HT when perfused into the frontal cortex

of the guinea pig, although when administered systemi-

cally they both decrease extracellular 5-HT in this

region.^21-23Although attempts have been made to

rationalize these findings, neither compound is an ideal

tool for studying terminal 5-HT autoreceptor function.

Methiothepin has affinity for a number of receptors

including 5-HT, dopamine, histamine, and adrenocep-

tors. Compound 4 does not discriminate between

5-HT_1Band 5-HT_1Dreceptor subtypes and also has

affinity for 5-HT_1Aand 5-HT_2Areceptors.²⁴It has also

been demonstrated that 4 displays partial agonist

activity in recombinant cell lines expressing functional

human cloned receptors.²⁵Using 4 as a lead for

chemistry, we have now identified 5, 1′-methyl-5-[[2′-

methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-yl]-

carbonyl]-2,3,6,7-tetrahydro[spiro[furo[2,3-f]indole-3,4′-

piperidine] (SB-224289) (Chart 1), as the first selective

5-HT_1Breceptor (terminal 5-HT autoreceptor) antago-

nist.

tate (Asp 129)²⁸on transmembrane helix 3 (TM3).^29,30

However attempts to dock 4 into the receptors in this

way indicated that this mode of binding was energeti-

cally unfavorable and resulted in part of the molecule

extending between helices 6 and 7 into the lipid bilayer

(Figure 1). As the protonated nitrogen of these ligands

is an essential feature for binding, alternative acidic

residues within the transmembrane bundle were sought.

All members of the rhodopsin GPCR family (which

includes the various 5-HT sequences) have a totally

conserved aspartate (Asp 95) on transmembrane helix

2 (TM2), and this has been implicated in GPCR receptor

activation.³¹On the other hand, sequence analysis of

the 5-HT_1B, 5-HT_1D, and 5-HT_1Ereceptors reveals the

presence of a third aspartate residue (Asp 352) on

transmembrane helix 7 (TM7). This residue is not

present in other 5-HT₁or 5-HT₂receptor subtypes.

Alternative modes of binding, involving these two as-

partates, were therefore evaluated. The principal dif-

ference between the 5-HT_1B/Dand 5-HT_1Ereceptors is

the presence of a “Lys-Glu” pair at the extracellular side

of TM6 in the 5-HT_1Ereceptor instead of the less polar

“Ile-Ser” (5-HT_1B) or “Val-Ser” (5-HT_1D). In our models

of these receptors, this lysine forms a salt bridge with

the aspartate on TM7, and this has been used to explain

the observed approximate 1000-fold lower affinity of the

agonist 5-carboxamidotryptamine (5-CT) for the 5-HT_1E

over the 5-HT_1Breceptor.³²Furthermore, it has been

reported that mutation of this “Lys-Glu” pair to the

corresponding 5-HT_1Bsequence results in a switch in

the pharmacology of these receptors.³³Compound 4 is

known to bind only weakly to 5-HT_1Ereceptors (pK_i5.5

vs [³H]-5-HT), and this would therefore suggest that it

is the aspartate on TM7 which is primarily involved in

the binding of this compound. When 4 was docked into

Computational models of the 5-HT_1B, 5-HT_1D, 5-HT_1E

,

and 5-HT_2Areceptors were constructed using the GPCR

Builder program²⁶and a protocol previously described

for our model of the 5-HT_2Creceptor.²⁷Studies were

then conducted to investigate possible modes of binding

of 4 to these receptors. It is generally accepted that

5-HT receptor agonists and antagonists bind their

protonated amino sites to the highly conserved aspar-

1220 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

F igu r e 2. Docking of 4 into model of human 5-HT_1Breceptor with piperazine binding to Asp 95 on TM2.

F igu r e 3. Docking of 4 into model of human 5-HT_1Breceptor with piperazine binding to Asp 352 on TM7.

the 5-HT_1Breceptor model with the protonated nitrogen

bound either to the aspartate on TM7 or to the aspartate

on TM2, it was found that both binding modes were

predicted to be energetically favorable. In each case the

biphenyl moiety was stabilized within a hydrophobic

pocket, and in addition a number of favorable hydrogen-

bonding interactions were possible (Figures 2 and 3).

Since the aspartate on TM2 has been implicated in

agonist function, it might be concluded therefore that

ligands which interact directly with this residue would

function as agonists or partial agonists. Hutchins³⁴has

suggested that dopamine agonists act by a bridging

interaction between the aspartates on TM2 and TM3

rather than the more generally accepted bridging mode

between the aspartate on TM3 and the serines on

TM5.²⁸On the other hand, compounds which bind to

the aspartate on TM7 only would be expected to act as

antagonists. Following on from this hypothesis, com-

pounds which can bind to either site and sit in an

equilibrium between the two binding modes would be

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1221

Sch em e 1^a

a

Reagents: (a) ClCH₂CH₂NMe₂‚HCl, Na₂CO₃, EtOH, ∆; (b) KNO₃, cH₂SO₄; (c) (^tBoc)₂O, Et₃N, CH₂Cl₂; (d) H₂, 10% Pd/C; (e) NaOH,

H₂O, THF; (f) TFA, CH₂Cl₂.

expected to be partial agonists with the degree of partial

agonism related to the relative binding energies of the

two modes. Since our aim was to identify a 5-HT_1B

receptor-selective compound with no intrinsic activity,

our working hypothesis was to identify structural

features and optimize these to favor binding with the

aspartate on TM7 over that on TM2. It was also

rationalized that a compound which binds preferentially

to TM7 might have reduced affinity for receptors other

than 5-HT_1Band 5-HT_1Dwhich do not have the third

aspartate residue (Asp 352) on TM7 (notably 5-HT₂).

However, although the computational models served as

a guide for achieving these targets, the close similarity

between the 5-HT_1Band 5-HT_1Dreceptor models pre-

cluded their use for discriminating between these two

receptor subtypes. In the model of 4 bound to the

aspartate on TM2, the o-methoxy group sat out of plane

from the phenyl ring and was predicted to be involved

in a hydrogen bond with the conserved histidine 139

on TM3. No corresponding role could be found for the

methoxy group in the alternative binding mode to the

TM7 aspartate. Furthermore in the TM2 binding mode,

the carbonyl of the amide could H-bond to the hydroxyl

of serine 136 on TM3 provided that the amide was

rotated out of plane with the piperazine-containing

phenyl ring. In the alternative TM7 binding mode, the

in-plane amide carbonyl was stabilized by a strong

H-bond with the NH of the highly conserved tryptophan

327 on TM6. In both modes of binding, the 1,2,4-

oxadiazole was capable of acting as a hydrogen bond

acceptor (with either Ser 334, TM2, or His 139, TM7),

so no preference for either binding mode was evident.

Areas pinpointed for modification therefore were the

arylpiperazine and the amide linker group.

Ch em istr y

The methods of preparation of test compounds 5-15

are shown in Schemes 1-8 and have been previously

reported in the patent literature.^35-40Alkylation of

2-methoxyaniline was effected with N,N-dimethylami-

noethyl chloride hydrochloride and sodium carbonate in

ethanol at reflux to give the ethylenediamine 6a which

was nitrated using potassium nitrate in concentrated

sulfuric acid to give 6b. Reaction with di-tert-butyl

dicarbonate gave the carbamate 6c which was reduced

to the aniline 6d by catalytic hydrogenation. 2′-Methyl-

1222 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

Sch em e 2^a

Ta ble 1. 5-HT_1B, 5-HT_1D, and 5-HT_2AReceptor Binding

Affinities (pK_iValues) of Compounds 4-15^a,b

compd

5-HT_1B

5-HT_1D

5-HT_2A

4

5

6

7

8

9.0 ( 0.07 (10)

8.2 ( 0.05 (12)

8.6 ( 0.05 (4)

9.0 ( 0.04 (23)

8.5 ( 0.12 (4)

8.8 ( 0.04 (5)

8.6 ( 0.07 (9)

9.3 ( 0.12 (6)

8.2 ( 0.4 (3)

8.6 ( 0.05 (5)

6.3 ( 0.09 (12)

7.4 ( 0.19 (4)

7.5 ( 0.03 (23)

7.0 ( 0.11 (4)

7.6 ( 0.08 (5)

7.7 ( 0.07 (9)

8.3 ( 0.04 (4)

7.9 ( 0.02 (3)

6.7 ( 0.06 (3)

7.4 ( 0.13 (4)

5.7 ( 0.05 (5)

7.8 ( 0.05 (6)

5.8 ( 0.06 (3)

8.3, 8.6

7.3, 7.4

8.2, 8.5

7.3, 7.3

8.1, 8.4

6.5, 6.6

6.7, 6.9

9

10

11

12

13

14

15

6.9 ( 0.11 (3)

7.9 ( 0.02 (3)

7.5 ( 0.05 (9)

6.4, 6.5

6.9, 6.9

6.0, 6.2

a

b

All human cloned receptors. Values are represented as mean

( SEM; n is in parentheses. For n ) 2 determinations, individual

values are shown.

9d which was reduced to aniline 9e by catalytic hydro-

genation. Coupling as before gave the amide 9 (Scheme

3).

The benzoxazine 10 was prepared according to Scheme

4. 6-Nitro-2H-1,4-benzoxazin-3(4H)-one⁴³was con-

verted to the (dimethylamino)ethyl derivative 10a by

treatment with sodium hydride and 2-(dimethylamino)-

ethyl chloride in toluene. Reduction of this amide with

diborane gave the nitrobenzoxazine 10b which was

reduced by catalytic hydrogenation to the aniline 10c.

Coupling with 2′-methyl-4′-(5-methyl-1,2,4-oxadiazol-3-

yl)biphenyl-4-carboxylic acid chloride using triethyl-

amine in dichloromethane gave 10. 2,2-Dimethoxyac-

etaldehyde was converted to the dimethyl acetal 11a

by reductive amination with 3-[2-(dimethylamino)et-

hoxy]-4-methoxyaniline, 7b. Cyclization to the indole

11b was carried out by heating under reflux with

trifluoroacetic anhydride in trifluoroacetic acid. Reduc-

tion of 11b to the 2,3-dihydroindole 11c was effected

with sodium cyanoborohydride in acetic acid. Coupling

with 2′-methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphe-

nyl-4-carboxylic acid chloride under Schotten-Baumann

conditions gave 11 (Scheme 5).

a

Reagents: (a) ClCH₂CH₂NMe₂‚HCl, K₂CO₃, Me₂CO, H₂O; (b)

H₂, 10% Pd/C, EtOH; (c) CH₂dCHCONMe₂, Pd(OAc)₂, Et₃N, DMF;

(d) LiAlH₄, THF, ∆; (e) NaOH, H₂O, THF; (f) Et₃N, CH₂Cl₂.

The aryl piperazines 12-14 were prepared in a

similar manner from the previously reported anilines³⁵

(Scheme 6). The ether 15a was prepared from 2-io-

dophenol using Mitsunobu conditions. Cyclization to

the spiropiperidine 15b was carried out using tributyltin

hydride in benzene in the presence of AIBN. Nitration

with copper nitrate in acetic anhydride gave 15c which

was reduced by catalytic hydrogenation to the aniline

15d . Coupling with 2′-methyl-4′-(5-methyl-1,2,4-oxa-

diazol-3-yl)biphenyl-4-carboxylic acid gave 15 (Scheme

7). Compound 5 was prepared in a similar way from

1-acetyl-6-bromo-2,3-dihydro-1H-indol-5-ol⁴⁴(Scheme 8).

4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-carboxylic

acid⁴¹was converted to the acid chloride and coupled

to the aniline 6d using Schotten-Baumann conditions

to give the amide 6e, which was converted to 6 by

treatment with TFA in dichloromethane (Scheme 1).

Compound 7 was prepared using similar methodology

from 2-methoxy-5-nitrophenol. N,N-Dimethyl-3-(5-

amino-2-methoxyphenyl)acrylamide (8a ) was prepared

from 4-amino-2-bromoanisole via a Heck reaction and

then reduced by catalytic hydrogenation followed by

treatment with LiAlH₄to give N,N-dimethyl-3-(5-amino-

2-methoxyphenyl)propylamine (8b). Coupling with 2′-

methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-car-

boxylic acid was effected as before to give the amide 8

(Scheme 2).

Resu lts a n d Discu ssion

Structure-activity relationships were determined

from affinity values obtained using radioligand binding

in human cloned receptors (summarized in Table 1). In

vitro functional data were obtained using [³⁵S]GTPγS

radioligand binding in CHO cells expressing human

cloned 5-HT_1Breceptors as previously described.^45,46

(Table 2). Removal of conformational restraint of the

piperazine in 4 to give the (dimethylamino)ethylamine

6 resulted in an improvement in 5-HT_1Bvs 5-HT_1D

receptor selectivity to 15-fold, although 5-HT_2Aaffinity

2,3-Dihydrobenzofuran-7-carboxylic acid⁴²was con-

verted to the corresponding acetyl derivative 9a by

reaction with methyllithium. Nitration of 9a with

potassium nitrate in concentrated sulfuric acid gave 9b

which was converted to the phenol 9c via Baeyer-

Villiger rearrangement and base hydrolysis. Alkylation

of 9c with N,N-dimethylaminoethyl chloride hydrochlo-

ride and potassium carbonate in DME gave the ether

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1223

Sch em e 3^a

a

Reagents: (a) MeLi, Et₂O; (b) KNO₃, cH₂SO₄; (c) mCPBA, CHCl₃, TsOH; (d) NaOH, H₂O, EtOH; (e) ClCH₂CH₂NMe₂‚HCl, K₂CO₃,

DME, ∆; (f) H₂, 10% Pd/C, EtOH; (g) NaOH, H₂O, THF.

Ta ble 2. Effect of Compounds 4-10 and 15 on [³⁵S]GTPγS

In these three compounds the methoxy group could be

oriented so as to interact with histidine 139. However

Binding in CHO Cells Expressing 5-HT_1BReceptors^a

% stimulation

(mean ( SEM)

intrinsic activity

(mean ( SEM)

it was also feasible that with compounds 6 and 7, the

heteroatom of the (dimethylamino)ethyl-X chain could

also be involved in an H-bonding interaction with

histidine 139. In compound 6 this would consist of a

bond from the NH of the ligand to the unprotonated

nitrogen of the histidine; in 7 the histidine NH could

donate an H-bond to the oxygen of the chain. Com-

pound 8 had no such possible second interaction. To

complicate matters further, the increased flexibility of

all these compounds made it possible for the protonated

amino group to now interact with the aspartate on TM3

(Asp 129). It was clearly evident however that the

orientation of the methoxy oxygen electron lone pairs

was important. Interaction with histidine 139 would

be favored if they were oriented toward the aminoethyl

chain and disfavored if they were oriented in the

opposite direction.

compd

5-HT

4

5

6

7

8

9

10

15

219 ( 31

143 ( 25

-62 ( 4

108 ( 19

139 ( 25

-44 ( 4

207 ( 33

91 ( 20

1.0

0.64 ( 0.01

<0

0.49 ( 0.02

0.63 ( 0.02

<0

0.94 ( 0.02

0.40 ( 0.02

<0

-64 ( 4

a

All values represent the mean of at least three independent

determinations. Data are expressed as the percent stimulation

above basal when compounds are added at 1 µM. Basal stimulation

is defined as the amount of [³⁵S]GTPγS specifically bound in the

absence of compounds. The intrinsic activity for each compound

is calculated as a percentage of the 5-HT response in the same

assay. Thus the intrinsic activity of 5-HT is always 1, for partial

agonists between 0 and 1, and for inverse agonists below 0.

To investigate this further, the methoxy oxygen was

incorporated into two different ring systems: the dihy-

drobenzofuran 9 and the benzoxazine 10. Both com-

pounds were potent 5-HT_1Breceptor ligands, but whereas

9 displayed high intrinsic activity in the human cloned

receptors, 10 was a weak partial agonist, supporting the

idea that the direction of the oxygen electron density

was influencing receptor activation by favoring primary

binding to the aspartate on TM2 (Table 2). As already

mentioned, the computational models suggested that

restricting the conformational freedom of the amide

linker group within a ring to evoke an in-plane carbonyl

was increased 5-fold. Replacement of the anilinic

nitrogen of 6 with oxygen to give the ether 7 improved

the selectivity for 5-HT_1Bover 5-HT_1Dto 30-fold, yet

interestingly 5-HT_2Aaffinity was reduced 15-fold over

that of 6. The corresponding methylene derivative 8

was slightly less potent than 7 at 5-HT_1B/Dreceptors and

showed increased 5-HT_2Aaffinity, although it retained

30-fold selectivity for the 5-HT_1Bvs 5-HT_1Dreceptor.

However, an equally interesting finding was that com-

pound 8 showed no intrisic activity in the [³⁵S]GTPγS

functional model, whereas the two related compounds

6 and 7 were partial agonists in this system (Table 2).

1224 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

Sch em e 4^a

Sch em e 6^a

a

Reagents: (a) NaOH, H₂O, THF.

affinity while retaining 10-fold selectivity for the 5-HT_1B

over the 5-HT_1Dreceptor. The models also suggested

that the piperazine in compound 4 preferred to lie out

of plane with the plane of the adjacent aromatic ring.

It was therefore feasible that the methoxy group could

also be involved in stabilizing this conformation. The

role of this substituent was briefly investigated in a

model system based on compound 12. Removal of the

methoxy substituent of 12 to give 13 brought about a

20-fold reduction in 5-HT_1Baffinity while 5-HT_1Daffinity

was reduced 15-fold. The corresponding methyl ana-

logue 14 was only marginally less potent than 12. This

is in keeping with the hypothesis that the primary role

of the ortho substituent in the TM7 binding mode is to

orientate the protonated amino group, as no hydrogen-

bonding role can be found for the methoxy in this pocket.

Given that 5-HT_1Baffinity could be retained when the

anilinic nitrogen was replaced by methylene (compound

8) and also considering that the orientation of the

oxygen lone pairs in the benzoxazine reduced intrinsic

activity, it was postulated that the spirocyclic piperidine

15, in which the basic nitrogen-containing ring lies

virtually orthogonal to the adjacent aromatic, could be

of interest. Although this compound was less potent

than the conformationally less restricted analogue 8 or

the parent piperazine 4, it was surprisingly much more

selective for the 5-HT_1Bvs 5-HT_1Dreceptor and, as

expected, showed no intrinsic activity in the [³⁵S]GTPγS

model (Table 2). In addition, 5-HT_2Aaffinity was

reduced by 50- and 150-fold over 4 and 8, respectively.

Applying the knowledge that cyclization to the indo-

line 11 had given improved 5-HT_1Bpotency and reduced

5-HT_2Aaffinity, we targeted compound 5 for synthesis.

Computational studies with compound 5 indicated that

binding into TM2 would not be favored. In contrast 5

could bind very favorably into TM7. In addition to

stabilization evoked by the hydrogen-bonding interac-

tions with His 319 and Trp 327, an interaction between

the oxygen of the dihydrobenzofuran and Ser 128 was

also favored (Figure 4). Compound 5, SB-224289,

retained high (80-fold) selectivity for the 5-HT_1Bover

5-HT_1Dreceptor and showed >200-fold selectivity for

5-HT_1Bover 5-HT_2Areceptors. In the [³⁵S]GTPγS

functional model in human cloned 5-HT_1Breceptors, 5

also displayed inverse agonist activity. (Table 2, Figure

5). In radioligand binding studies, 5 was found to be

highly selective over a range of other receptors (Table

3) and was therefore selected for further evaluation both

in vitro and in vivo.

a

Reagents: (a) ClCH₂CH₂NMe₂‚HCl, NaH, MePh, ∆; (b) BH₃,

THF, ∆; (c) H₂, 10% Pd/C, EtOH; (d) Et₃N, CH₂Cl₂.

Sch em e 5^a

a

Reagents: (a) (MeO)₂CHCHO, EtOH, H₂, 10% Pd/C; (b) TFA,

TFAA; (c) NaBH₃CN, AcOH; (d) NaOH, H₂O, THF.

“right-hand side” aromatic might favor the interaction

with the TM7 aspartate resulting in improved selectiv-

ity over 5-HT₂receptors. This was due to an interaction

with the NH of tryptophan 327 as opposed to the H-bond

with serine 136 which arose from the out-of-plane

orientation favored for binding to aspartate 95. Cy-

clization of the amide of compound 7 to give the indoline

11 resulted in a 6-fold loss of 5-HT_2Aaffinity over 7. This

cyclization also resulted in a slight increase in 5-HT_1B

Secon d a r y Eva lu a tion in Vitr o. In both human

5-HT_1Band 5-HT_1Dreceptor-expressing cell lines, 5-HT

produced a concentration-dependent stimulation of basal

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1225

Sch em e 7^a

a

Reagents: (a) DEAD, Ph₃P, THF; (b) Bu₃SnH, AIBN, C₆H₆, ∆; (c) Cu(NO₂)₂, Ac₂O; (d) H₂, 10% Pd/C, EtOH; (e) Et₃N, CH₂Cl₂.

F igu r e 4. Docking of 5 into model of human 5-HT_1Breceptor with spiropiperidine binding to Asp 352 on TM7.

[³⁵S]GTPγS binding with a maximal stimulation of 290%

and 137% above basal levels, respectively. In contrast,

5 evoked a concentration-dependent inhibition of basal

[³⁵S]GTPγS binding in both cell lines (Figures 5 and 6).

Data for compound 4 are included for comparison. In

antagonist studies at human 5-HT_1Band 5-HT_1Drecep-

1226 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

Ta ble 3. Receptor Binding Profile of 5^a

Sch em e 8^a

receptor

affinity (pK_i)

<5.5

receptor

affinity (pK_i)

5-HT_1A

5-HT_1B

5-HT_1D

5-HT_1E

5-HT_1F

5-HT_2A

5-HT_2C

5-HT₄

R_1A

R_2A

R_2B

â₁

<6.0

8.2 ( 0.05 (12)

6.3 ( 0.09 (12)

<5.0

<6.0

â₂

D₂

D₃

H₁

H2

M₁

M₂

6.2 ( 0.06 (9)

<6.0

R_1B

a

All human cloned receptors except 5-HT₄(piglet hippocampus,

[

¹²⁵I]SB 207710), R_1A(rat submaxillary gland, [³H]prazosin), R_1B

,

(rat liver, [³H]prazosin), R_2B(NG 108-15 cells, [³H]RX821002), H₁

(guinea pig cerebellum, [³H]pyrilamine), and H₂(guinea pig

striatum, [¹²⁵I]APT).

a

Reagents: (a) DEAD, Ph₃P, THF; (b) Bu₃SnH, AIBN, C₆H₆,

∆; (c) HCl, EtOH, ∆; (d) Et₃N, CH₂Cl₂.

F igu r e 6. Inverse agonist effect of 5 in [³⁵S]GTPγS radioli-

gand binding in human cloned 5-HT_1Dreceptors expressed in

CHO cells (mean ( SEM, n ) 3).

F igu r e 5. Inverse agonist effect of 5 in [³⁵S]GTPγS radioli-

gand binding in human cloned 5-HT_1Breceptors expressed in

CHO cells (mean ( SEM, n ) 7).

tors, compound 5 produced a rightward shift of the

concentration curve to 5-HT with apparent pA₂values

of 8.4 ( 0.2 and 6.9 ( 0.1, respectively, demonstrating

functional selectivity for the 5-HT_1Bover the 5-HT_1D

receptor (Figures 7 and 8). To evaluate terminal 5-HT

autoreceptor function in vitro, 5 was investigated for

its ability to influence 5-HT release in electrically

stimulated guinea pig brain cortex slices. At concentra-

tions of 100 nM and 1 µM, 5 evoked an increase in [³H]-

5-HT release, indicative of terminal 5-HT receptor

antagonist activity (Figure 9).

F igu r e 7. Effect of 5 on dose response to 5-HT in [³⁵S]GTPγS

radioligand binding in human cloned 5-HT_1Breceptors ex-

pressed in CHO cells (mean ( SEM, n ) 16).

Secon d a r y Eva lu a tion in Vivo. Compound 5 was

evaluated in the previously reported SK&F 99101H-

induced hypothermia model of central 5-HT_1B/Dreceptor

function in the guinea pig.⁴⁷Compound 5 (2.2-22.0 mg/

kg po) had no effect on body temperature alone (Table

4) but reversed SK&F 99101H-induced hypothermia

giving an ED₅₀of 3.62 mg/kg po (Table 5). When dosed

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1227

Ta ble 6. Effects of 5 (5.0 mg/kg po) on SK&F 99101H (30

mg/kg ip)-Induced Hypothermia (2-24 h Postdose)^a

mean maximum

temp drop (°C) (SEM)

treatment

% of SK&F 99101H

control

(h postdose) SK&F 99101H

5

2

4

8

-1.72 (0.09) -0.38 (0.19)**

-1.73 (0.17) -0.27 (0.13)**

-1.22 (0.27) -0.53 (0.35)*

-2.23 (0.13) -1.10 (0.37)**

-2.60 (1.37) -1.31 (0.57)**

22

15

43

48

50

8^b

24^b

a

b

n ) 6/group. Data from second, independent experiment. *p

< 0.05 and **p < 0.01 vs SK&F 99101H controls.

F igu r e 8. Effect of 5 on dose response to 5-HT in [³⁵S]GTPγS

radioligand binding in human cloned 5-HT_1Dreceptors ex-

pressed in CHO cells (mean ( SEM, n ) 6).

F igu r e 10. Effects of 5 on sumatriptan-induced inhibition of

5-HT release in the frontal cortex of the freely moving guinea

pig.

F igu r e 9. Effects of 5 on electrically evoked 5-HT release in

guinea pig frontal cortex slices (mean ( SEM, n ) 4).

cortical 5-HT release using microdialysis in the freely

moving guinea pig as an in vivo model of terminal

5-HT_1Bautoreceptor function. The 5-HT_1B/1Dreceptor

agonist sumatriptan (1 µM), administered locally via the

dialysis probe, significantly reduced 5-HT levels to 53%

of controls, indicative of presynaptic terminal 5-HT_1B

receptor stimulation. Compound 5 (4 mg/kg po) signifi-

cantly reversed this inhibition of 5-HT demonstrating

its ability to act as a terminal 5-HT autoreceptor

antagonist in vivo (Figure 10). Full experimental

details are described elsewhere.⁴⁹Recently we have also

shown that compound 5 (4 mg/kg po) is able to signifi-

cantly increase 5-HT levels in a different brain region,

the guinea pig dentate gyrus, and this work is described

Ta ble 4. Lack of Effects of 5 (po) Alone on Body Temperature

of the Guinea Pig^a

treatment

body temp drop

(dose, mg/kg)

(°C, (SEM)

vehicle

-0.07 ( 0.08

-2.00 ( 0.44**

-0.23 ( 0.11

-0.05 ( 0.11

-0.03 ( 0.07

-0.03 ( 0.11

SK&F 99101H^b

5 (2.2)

5 (4.6)

5 (10)

5 (22)

a

b

n ) 6/group. SK&F 99101H used as positive control. **p <

0.01 vs vehicle/vehicle controls.

Ta ble 5. Effects of 5 (po) on Hypothermia Induced by the

50

in a separate publication.

5-HT_1B/DReceptor Agonist SK&F 99101H (30 mg/kg ip)^a

treatment

(dose, mg/kg)

body temp drop % vehicle/SK&F 99101H

Con clu sion

(°C (SEM)

controls

vehicle/vehicle

-0.12 ( 0.07

-1.92 ( 0.23

-1.77 ( 0.27

-1.63 ( 0.18

-1.00 ( 0.38*

Using molecular modeling as a basis for stuctural

modification of the nonselective 5-HT_1B/Dreceptor par-

tial agonist 4, we have identified the selective 5-HT_1B

receptor ligand 5. This compound behaves as an inverse

agonist in CHO cells expressing human cloned 5-HT_1B

receptors and shows excellent oral activity with a long

duration of action in a pharmacodynamic model of

central 5-HT_1Breceptor-mediated function. In addition,

5 has been shown to act as a potent terminal 5-HT

autoreceptor antagonist both in vitro and in vivo. As

such, compound 5 provides an excellent new chemical

tool for the investigation of 5-HT_1Bautoreceptor function

in the CNS.

vehicle/SK&F 99101H

5 (0.3)/SK&F 99101H

5 (1.0)/SK&F 99101H

5 (3.0)/SK&F 99101H

100

93

85

52

35

5 (10.0)/SK&F 99101H -0.67 ( 0.19**

ED₅₀(95% confidence limits) 3.62 (1.62-19.88)

a

n ) 6/group. *p < 0.05 vs vehicle/SK&F 99101H controls. **p

< 0.01 vs vehicle/SK&F 99101H controls.

at 5 mg/kg po, the antagonist effects of 5 were still

evident at 24 h postdose, indicating that this compound

has a long pharmacodynamic duration of action (Table

6). Experimental details have been published.⁴⁸Com-

pound 5 was therefore evaluated for its effects on

1228 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

housed in groups of five or six, in a temperature-controlled

environment (20 ( 1 °C), and maintained on a 12-h light/dark

cycle for at least 5 days prior to use with free access to food

and water. On test days animals were either housed in pairs

in opaque observation cages (24 cm × 45 cm × 20 cm) at least

2 h prior to the start of the experiment for intraperitoneal (ip)

dosing or food-deprived overnight in pairs in clear grid-

bottomed starving cages for perioral (po) dosing. Rectal

temperature was monitored using an electric thermometer

(Comark, model 9001) coupled to a rectal probe (Comark,

model BS4937). Animals were manually restrained and

temperatures measured by inserting the probe approximately

4-5 cm into the rectum for a period of 15 s or until a stable

reading was obtained. Compound 5 was administered perio-

rally (po) at 45 min prior to injection of SK&F 99101H. SK&F

99101H or vehicle was administered ip at time 0. Tempera-

ture measurements were taken at 60 min (-60), 45 min (-45),

and immediately prior to administration of SK&F 991091H.

Further measurements were taken 45, 60, and 90 min after

injection of SK&F 99101H. Temperature at each time point

was expressed as a function of the temperature at time 0. To

simplify the data analysis the peak temperature drop was then

calculated for each animal. These peak drops provide a simple

measure of the maximal drug effect and were averaged for each

treatment group. A time course for each drug was also plotted.

Changes in temperature were assessed using a one-way

analysis of variance (ANOVA) with post hoc analysis carried

out using Dunnett’s t-tests. ED₅₀values and 95% confidence

limits were calculated using a linear fit procedure in RS1.⁵⁸

Exp er im en ta l Section

Biologica l Assa ys. 1. Recep tor Bin d in g. Affinities of

compounds were determined using competition binding assays

to determine pK_ivalues at the various receptors. Human

cloned 5-HT_1A, 5-HT_2A, and 5-HT_2Creceptors were expressed

in HEK 293 cells. Human 5-HT_1B, 5-HT_1D, 5-HT_1E, 5-HT_1F

,

and dopamine D₂and D₃receptors were all expressed in CHO

cells. Affinities were measured as the displacement of [³H]-

5-HT from 5-HT_1B, 5-HT_1D, 5-HT_1E, and 5-HT_1Freceptors, [³H]-

8-OH-DPAT from 5-HT_1Areceptors, [³H]ketanserin from 5-HT_2A

receptors, [³H]mesulergine from 5-HT_2Creceptors, and [³H]-

spiperone from dopamine D₂and D₃receptors. The incubation

buffer used for 5-HT₁receptors was Tris-Mg²⁺(50 mM Trizma-

HCl, pH 7.7 at 25 °C; 6 mM ^L-ascorbic acid; 10 mM MgCl₂;

0.5 µM pargyline), and 50 mM Tris-HCl (pH 7.7 at 25 °C) was

used for 5-HT_2Aand 5-HT_2Creceptors. For dopamine D₂and

D₃receptors the incubation buffer was 50 mM Tris (pH 7.7 at

25 °C), 120 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂.

Affinity for the 5-HT₄receptor was evaluated in piglet hip-

pocampus, by displacement of [¹²⁵I]SB-207710 according to the

method of Brown et al.⁵¹Affinities for R-adrenergic, â-adren-

ergic, histamine H₁and H₂, and muscarinic M₁and M₂

receptors were determined using literature procedures.^52-56

Briefly, the appropriate radioligand and a range of concentra-

tions of test compound were incubated with the receptor

preparation for between 30 and 45 min at 37 °C. At the end

of the incubation period, samples were rapidly filtered and

washed with ice-cold incubation buffer. Filters were dried, and

specifically bound radiolabel was measured on a scintillation

counter. Results are expressed as mean values ((SEM, n) or

as individual values where n ) 2.

5. Micr od ia lysis in th e F r eely Movin g Gu in ea P ig.

Male Dunkin-Hartley guinea pigs weighing between 350 and

450 g were used in all experiments. Animals were maintained

on a 12-h light/dark cycle at 22 °C and given free access to

food and water. Guinea pigs were anesthetized with 2%

methoxyfluorane delivered with O₂(1 L/min) and N₂O (3

L/min) in an induction chamber. On attaining surgical

anesthesia the guinea pigs were transferred to a stereotaxic

frame which had been adapted to accommodate an anesthetic

mask and scavenging unit.⁵⁹Anesthesia was maintained on

1-2% methoxyfluorane with O₂(1 L/min), NO₂(2 L/min).

Dialysis probes were implanted into the frontal cortex (coor-

dinates from the intra-aural zero: AP+15.0 mm, ML+2.0 mm,

and 3.0 mm vertical from the dura, using the atlas of

Rapisarda and Bacchelli).⁶⁰Probes were secured with two

skull screws and dental acrylic cement, and the wound was

sealed. Animals were allowed 24 h for recovery, after which

time the probes were perfused with artificial CSF at a rate of

2 µL/min. After 2-h perfusion, samples were collected every

20 min into 10 µL of 0.4 M perchloric acid. Three samples

were taken to measure basal extracellular levels of 5-HT before

drug treatment. Following drug treatments, 5-HT levels were

measured for at least nine samples. Guinea pigs were dosed

orally with vehicle or 5 after the third dialysis sample (i.e., 60

min from the start of the experiment), and samples were

collected for 360 min in total; 45-µL dialysis samples were

injected onto an HPLC system using center loop filling, and

5-HT was separated from other substances using reverse-

phase ion pair chromatography. Separation was achieved at

a flow rate of 1 mL/min with a 4.6-mm × 7.5-mm ODS2, 3-µm

column (HPLC Technology Ltd.) and a mobile phase consisting

of 0.15 M NaH₂PO₄, 0.8 mM sodium octylsulfate, 0.4 mM

EDTA (pH 3.8), and 14% MeOH. The mobile phase was

filtered through a 0.22-µm GS filter and degassed with helium.

Data from experiments are reported as area under chromato-

gram peaks. The first three samples were averaged to yield

a basal level of extracellular 5-HT. All samples were expressed

as a percentage of basal levels. Percentages of basal levels

for the individual time points posttreatment were accumulated

and averaged, producing a value for the mean percent basal,

which was an estimate of the mean area under the curve

(AUC). Statistical comparisons of mean AUC following drug

administration were calculated on an SAS-Research Scientist

Application (v.1.4, release 6.08 (1992); SAS Institute Inc., Cary,

NC 27513). Analyses were performed using a one-way ANO-

2. [³⁵S]GTP γS Ra d ioliga n d Bin d in g. [³⁵S]GTPγS bind-

ing studies in CHO cells expressing human cloned 5-HT_1Bor

5-HT_1Dreceptors were performed as previously described.^45,46

In brief, membranes from 1 × 10⁶cells were preincubated at

30 °C for 30 min, in HEPES buffer (20 mM HEPES, 3 mM

MgCl₂, 100 mM NaCl, 0.2 mM ascorbate), containing 10 µM

GDP, with or without test compounds. The reaction was

started by the addition of 10 µL of [³⁵S]GTPγS (100 pM)

followed by a further 30-min incubation at 30 °C. Nonspecific

binding was determined by addition of unlabeled GTPγS (10

µM), prior to the addition of cells. The reaction was stopped

by rapid filtration using Whatman GF/B grade filters followed

by washing with ice-cold HEPES buffer. Radioactivity was

determined by liquid scintillation spectrometry. Basal stimu-

lation is the amount of [³⁵S]GTPγS specifically bound in the

absence of compounds and was taken as 0% in each assay.

Compounds were added at 1 µM, and the percent stimulation

above basal was determined. The intrinsic activity for each

compound was calculated as a percentage of the 5-HT response

in each assay. Thus the intrinsic activity of 5-HT is always

1, for partial agonists between 0 and 1, and for inverse agonists

below 0. Percent stimulation and intrinsic activity values are

expressed as the mean ((SEM) of three or more separate

determinations.

3. Effects on Electr ica lly Stim u la ted 5-HT Relea se in

Gu in ea P ig F r on ta l Cor tex. Methodology has previously

been described by Roberts et al.⁵⁷Cerebral cortices were

removed from male guinea pigs, cross-chopped into slices, and

then incubated with [³H]-5-HT. Slices were placed in a

Brandel Superfusion System and superfused with oxygenated

Kreb’s solution. Transmitter release was evoked twice (S1 and

S2), through electrical stimulation using a Brandel stimulator

(2-ms biphasic square wave pulses, 20 mA in amplitude at a

frequency of 3 Hz, to optimize the biphase concentration of

endogenous 5-HT). Compounds were superfused prior to and

during the S2 stimulation period. Released radioactivity in

each sample was measured, and after basal release was

subtracted, the S2/S1 ratio was calculated. Results represent

the mean ((SEM) of four determinations.

4. SK&F 99101H-In d u ced Hyp ot h er m ia Mod el of

Cen tr a l 5-HT_1B/DRecep tor F u n ction in th e Gu in ea P ig.⁴⁷

Male Dunkin-Hartley guinea pigs (Charles River UK Ltd.)

weighing between 270 and 400 g were used. Animals were

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1229

VA followed by a post hoc Tukey-Kramer t-test. Significance

was taken at the 5% level.

(2H, t, CH₂NMe₂), 2.24 (6H, s, N(CH₃)₂), 1.45-1.35 (9H, br,

C(CH₃)₃).

Ch em istr y. ¹H NMR spectra were recorded on a Bruker

AC-200 or AC-250 instrument using tetramethylsilane as

internal standard. High-resolution mass spectra were ob-

tained using a J EOL DX303 double focusing mass spectrom-

eter with 5000 resolution using manual peak matching.

Chromatography was performed on Merck Art. 7734 silica gel.

Where required, tetrahydrofuran (THF) was freshly distilled

from sodium/benzophenone prior to use. HPLC analysis of test

compounds was carried out on a Gilson 712 HPLC system,

using a model 231 sample injector and 306 pump with 806

manomeric module detector. HPLC method Asa Hypersil

BDS C18 3-µm (100- × 3-mm i.d.) column was used with

elution using the following conditions: eluant A 0.1% TFA/

H₂O v/v, eluant B 0.1% TFA/CH₃CN v/v; flow rate 0.7 mL/

min. The eluation gradient was 0% B held for 0.5 min, then

linearly increased to 75% B over 24.5 min, then held for 5 min.

A detection wavelength of 218 nm was used. HPLC method

Bsa Merck RP-select B (125 × 4.6 mm) column was used, with

elution using A 0.1% v/v TFA in H₂O, eluant B 0.1% v/v TFA

in MeCN, with a flow rate of 1.0 mL/min. The elution gradient

increase was from 5% B to 80% B over 40 min, then held at

80% B for 40 min. A detection wavelength of 218 nm was used.

HPLC method Csa Symmetry C8 5-µm (150- × 3.9-mm i.d.)

column was used with elution using eluant A 0.1% TFA v/v in

H₂O, eluant B 0.1% TFA v/v in MeCN, with a flow rate of 1.0

mL/min. The elution gradient was held at 0% B for 0.5 min,

then linearly increased to 90% over 24.5 min, then held for 5

min. The UV detection wavelength was 218 nm. Elemental

analysis was carried out using a Control Equipment Corp.

(CEC) 440 elemental analyzer. Where reported only by

symbols, values were within 0.4% of theoretical limits.

N,N-Dim eth yl-N′-(ter t-bu tyloxyca r bon yl)-N′-(5-a m in o-

2-m eth oxyp h en yl)eth ylen ed ia m in e (6d ). A solution of

N,N-dimethyl-N′-(tert-butyloxycarbonyl)-N′-(2-methoxy-5-ni-

trophenyl)ethylenediamine, 6c (0.50 g, 0.0015moL), in EtOH

(25 mL) was hydrogenated over 10% Pd/C at room temperature

and pressure until H₂uptake was complete. The catalyst was

removed by filtration through Kieselguhr and the filtrate

concentrated in vacuo to give 6d (0.37 g, 80%) as a brown oil.

¹H NMR (CDCl₃): δ 6.65 (1H, d, H6), 6.50 (2H, d, H3, H4),

3.68 (3H, s, OCH₃), 3.60-3.02 (4H, m, NH₂, NCH₂), 2.49, (2H,

t, CH₂NMe₂), 2.16 (6H, s, NMe₂), 1.45-1.28 (9H, br, C(CH₃)₃).

N-[3-[N-[2-(Dim eth yla m in o)eth yl]-N-(ter t-bu tyloxyca r -

bon yl)a m in o]-4-m eth oxyp h en yl]-2′-m eth yl-4′-(5-m eth yl-

1,2,4-oxadiazol-3-yl)biph en yl-4-car boxam ide (6e). A stirred

solution of 2′-methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphe-

nyl-4-carboxylic acid (0.34 g, 0.0012 mol) in SOCl₂(15 mL) was

heated under reflux for 1 h. After cooling to room temperature,

the excess SOCl₂was removed in vacuo to give the acid

chloride. A solution of N,N-dimethyl-N′-(tert-butyloxycarbo-

nyl)-N′-(5-amino-2-methoxyphenyl)ethylenediamine, 6d (0.37

g, 0.0012 mol), in THF (15 mL) was treated with a solution of

NaOH (0.4 g, 0.1 mol) in H₂O (15 mL), and a solution of the

acid chloride in THF (15 mL) was added. The reaction mixture

was stirred overnight at room temperature. After removal of

the solvent in vacuo, the residue was dissolved in CH₂Cl₂,

washed with H₂O, dried (MgSO₄), and concentrated in vacuo

to give 6e (0.40 g, 57%) as a brown oil. ¹H NMR (CDCl₃): δ

8.10-7.61 (6H, m, biphenyl, H2, H3, H5, H6, H3′, H5′), 7.53-

7.20 (4H, m, biphenyl H6′, H4, H6, CONH), 6.97-6.62 (1H,

m, H3), 3.82 (3H, s, OCH₃), 3.51 (2NH, br, NCH₂), 2.70 (3H, s,

CH₃), 2.62-2.41 (2H, m, CH₂NMe₂), 2.41-2.11 (9H, m, NMe₂,

CH₃), 1.58-1.35 (9H, br, C(CH₃)₃).

N ,N -D i m e t h y l-N ′-(2-m e t h o x y p h e n y l)e t h y le n e d i -

a m in e (6a ). A mixture of 2-methoxyaniline (6.20 g, 0.05 mol),

N,N-dimethylaminoethyl chloride hydrochloride (8.6 g, 0.06

mol), and Na₂CO₃(7.42 g, 0.07 mol) in EtOH (75 mL) was

heated under reflux for 7 h. After cooling, the mixture was

filtered and the filtrate concentrated in vacuo. The residue

was dissolved in H₂O and the pH adjusted to 12 by the addition

of 10% aqueous NaOH. The mixture was extracted into Et₂O,

dried (Na₂SO₄), concentrated in vacuo, and purified by flash

chromatography, eluting with EtOAc to give 6a as a pale-

orange oil (3.28 g, 48%). ¹H NMR (CDCl₃): δ 6.87 (1H, dt,

H4), 6.75 (1H, dd, H6), 6.68 (1H, dd, H5), 6.60 (1H, dd, H3),

4.62 (1H, brs, NH), 3.82 (3H, s, OCH₃), 3.15 (2H, m, NHCH₂),

2.58 (2H, t, CH₂NMe₂), 2.25 (6H, s, N(CH₃)₂).

N-[3-[[2-(Dim eth yla m in o)eth yl]a m in o]-4-m eth oxyp h e-

n yl]-2′-m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-yl)bip h en yl-

4-ca r b oxa m id e (6). N-[3-[N-[2-(Dimethylamino)ethyl]-N-

(tert-butyloxycarbonyl)amino]-4-methoxyphenyl]-2′-methyl-4′-

(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-carboxamide, 6e (0.40

g, 0.0007 mol), was dissolved in CH₂Cl₂(20 mL) and TFA (5

mL). The solution was stirred at room temperature for 2.5 h,

and the solvent was removed in vacuo. The residue was

dissolved in CH₂Cl₂, washed with saturated aqueous NaHCO₃

solution, dried (Na₂SO₄), and concentrated in vacuo. The

residue was purified by flash column chromatography on silica

gel, eluting with CH₂Cl₂with increasing concentrations of

MeOH. The product 6 was isolated by conversion to its oxalate

salt (0.25 g, 75%) as an off-white solid. ¹H NMR (CDCl₃): δ

10.08 (1H, s, COOH), 8.05 (2H, d, biphenyl, H2, H6), 8.00 (1H,

s, NH), 7.92 (1H, d, biphenyl, H′5), 7.57 (2H, d, biphenyl, H3,

H5), 7.44 (1H, d, biphenyl, H3′), 7.11 (2H, m, biphenyl, H6,

H6), 6.82 (2H, d, H3, H4), 5.51-3.90 (1H, brs, NH), 3.78 (3H,

s, OCH₃), 3.41 (2H, t, NHCH₂), 3.27 (2H, t, CH₂NMe₂), 2.79

(6H, s, N(CH₃)₂), 2.70 (3H, s, CH₃), 2.36 (3H, s, CH₃). HRMS

(FAB): calcd for 486.250 (M⁺), found 486.253. Purity was

determined as 96.8% by HPLC (method A), retention time

17.35 min.

N,N-Dim eth yl-N′-(2-m eth oxy-5-n itr op h en yl)eth ylen e-

d ia m in e (6b). N,N-Dimethyl-N′-(2-methoxyphenyl)ethylene-

diamine, 6a (1.5 g, 0.008 mol), was dissolved in 5 N H₂SO₄

(0.86 mL) and the resulting solution concentrated to dryness

in vacuo. Concentrated H₂SO₄(6.5 mL) was added, and the

mixture was stirred until homogeneous and then cooled to 0

°C. Potassium nitrate (1.0 g, 0.01 mol) was added portionwise,

maintaining the temperature below 10 °C; then the mixture

was stirred for 4 h at room temperature. The reaction mixture

was poured onto ice (150 mL) and made slightly alkaline by

the addition of Na₂CO₃. The product was extracted into EtOAc

and the extract dried (Na₂SO₄) and concentrated in vacuo to

give 6b (0.90 g, 49%) as an orange oil. ¹H NMR (CDCl₃): δ

7.62 (1H, dd, H4), 7.36 (1H, d, H6), 6.75 (1H, d, H3), 4.95 (1H,

brs, NH), 3.94 (3H, s, OCH₃), 3.23 (2H, m, NHCH₂), 2.63 (2H,

t, CH₂NMe₂), 2.29 (6H, s, N(CH₃)₂).

3-[2-(Dim eth yla m in o)eth oxy]-4-n itr oa n isole (7a ).

A

stirred solution of of 2-methoxy-5-nitrophenol (5.0 g, 0.03 mol)

and K₂CO₃(8.3 g, 0.06 mol) in acetone (200 mL) and H₂O (60

mL) was treated with N,N-dimethylaminoethyl chloride hy-

drochloride (8.64 g, 0.60 mol). The reaction mixture was

heated under reflux for 10 h and then concentrated in vacuo

to approximately 80-mL volume. The residue was acidified

with 2 M HCl (150 mL) and washed with EtOAc. The acid

solution was basified with K₂CO₃and extracted with EtOAc

(2 × 100 mL). These extracts were combined, dried (Na₂SO₄),

and concentrated in vacuo to give 7a (4.87 g, 70%) as a yellow

solid. ¹H NMR (CDCl₃): δ 7.92 (1H, dd, H3), 7.77 (1H, d, H5),

6.91 (1H, d, H2), 4.18 (2H, t, OCH₂), 3.96 (3H, s, OCH₃), 2.82

(2H, t, CH₂NMe₂), 2.37 (6H, s, NMe₂).

N,N-Dim eth yl-N′-(ter t-bu tyloxyca r bon yl)-N′-(2-m eth -

oxy-5-n itr op h en yl)eth ylen ed ia m in e (6c). N,N-Dimethyl-

N′-(2-methoxy-5-nitrophenyl)ethylenediamine, 6b (0.90 g, 0.0038

mol), in CH₂Cl₂(50 mL) was treated with Et₃N (0.42 g, 0.004

mol) and di-tert-butyl dicarbonate (1.0 g, 0005 mol), and the

reaction mixture stirred at room temperature for 24 h. The

reaction mixture was washed with H₂O, dried (Na₂SO₄), and

concentrated in vacuo to give 6c (0.50 g, 39%) as an orange

oil. ¹H NMR (CDCl₃): δ 8.28-8.10 (2H, m, H4, H5), 6.98 (1H,

d, H3), 3.96 (3H, s, OCH₃), 3.72-3.28 (2H, br, NCH₂), 2.45

3-[2-(Dim eth yla m in o)eth oxy]-4-m eth oxya n ilin e (7b). A

solution of 3-[2-(dimethylamino)ethoxy]-4-nitroanisole, 7a (4.8

g, 0.02 mol), in EtOH (200 mL) was hydrogenated over 10%

1230 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

Pd/C at room temperature and pressure until hydrogen uptake

was complete. The catalyst was removed by filtration through

Kieselguhr and the filtrate concentrated in vacuo to afford 7b

(4.0 g, 95%) as a pink solid. ¹H NMR (CDCl₃): δ 6.71 (1H, d,

H2), 6.33 (1H, d, H6), 6.24 (1H, d, H5), 4.07 (2H, t, OCH₂),

3.78 (3H, s, OCH₃), 3.46 (2H, brs, NH₂), 2.76 (2H, t, CH₂NMe₂),

2.33 (6H, s, NMe₂).

was added. The reaction mixture was stirred for 1 h and then

washed with H₂O followed by NaHCO₃solution. The organic

layer was dried (Na₂SO₄) and concentrated in vacuo to give a

brown oil, which was purified by column chromatography on

silica gel using 7.5% MeOH in CH₂Cl₂as eluant to give 8 (0.052

g, 16%), isolated as the oxalate salt. ¹H NMR (DMSO-d₆): δ

10.20 (1H, s, COOH), 8.08 (2H, d, biphenyl H2, H6), 7.98 (1H,

brs, CONH), 7.91 (2H, d, biphenyl H3′, H5′), 7.65-7.58 (4H,

m, biphenyl H6′, H3, H5, H2), 7.45 (1H, d, H6), 7.00 (1H, d,

H5), 3.80 (3H, s, OCH₃), 3.08 (2H, t, CH₂), 2.75 (6H, s,

N(CH₃)₂), 2.65 (3H, s, CH₃), 2.58 (2H, t, CH₂NMe₂), 2.32 (3H,

s, CH₃), 1.90 (2H, m, CH₂CH₂CH₂). HRMS (FAB): calcd for

485.255 (M⁺), found 485.254. Purity was determined as >99%

by HPLC (method A), retention time 18.04 min.

7-Acetyl-2,3-d ih yd r oben zofu r a n (9a ). To a cooled sus-

pension of 2,3-dihydrobenzofuran-7-carboxylic acid (2 g, 0.012

mol), in dry Et₂O (100 mL) under argon, was added methyl-

lithium (16.2 mL of a 1.5 M solution in Et₂O, 0.02 mol). The

mixture was stirred at room temperature for 18 h and then

added to 5 M HCl (50 mL). The product was extracted into

EtOAc and the organic extract washed with aqueous K₂CO₃

and then H₂O and dried (Na₂SO₄). Evaporation of solvent gave

9a (1.5 g, 76%) as a pale-yellow solid. ¹H NMR (CDCl₃): δ

7.69 (1H, d, H6), 7.35 (1H, d, H4), 6.89 (1H, t, H5), 4.70 (2H,

t, OCH₂), 3.25 (2H, t, CH₂), 2.61 (3H, s, COCH₃).

7-Acetyl-5-n itr o-2,3-d ih yd r oben zofu r a n (9b). Potas-

sium nitrate (0.50 g, 0.005 mol) was added portionwise over

0.5 h to a solution of 7-acetyl-2,3-dihydrobenzofuran, 9a (0.50

g, 0.003 mol), in concentrated H₂SO₄(5 mL) at 5-10 °C. The

resulting solution was stirred for a further 2 h, then added to

ice, basified with 40% aqueous NaOH, and extracted with

EtOAc. The organic extract was dried (Na₂SO₄) and concen-

trated in vacuo to give 9b (0.52 g, 81%) as a cream powder.

¹H NMR (CDCl₃): δ 8.65 (1H, s, 1H), 8.20 (1H, s, H4), 4.91

(2H, t, OCH₂), 3.36 (2H, t, CH₂), 2.65 (3H, s, COCH₃).

7-Hyd r oxy-5-n itr o-2,3-d ih yd r oben zofu r a n (9c). A solu-

tion of 7-acetyl-5-nitro-2,3-dihydrobenzofuran, 9b (0.50 g,

0.0024 mol), in CHCl₃(3 mL) was added to m-chloroperoxy-

benzoic acid (0.83 g, 55%, 0.008 mol) in CHCl₃(8 mL)

containing a few crystals of p-toluenesulfonic acid. A further

quantity of m-chloroperoxybenzoic acid (0.83 g, 0.008 mol) was

added after 24 h and a further 0.40 g (0.004 mol) after another

24 h. After stirring at room temperature for 3 days under

argon, the reaction mixture was washed with aqueous NaH-

CO₃solution, dried (Na₂SO₄), and concentrated in vacuo. The

residue was dissolved in EtOH (50 mL) containing 40%

aqueous NaOH (5 mL) and stirred for 30 min at room

temperature. The mixture was acidified by the addition of 5

M HCl and evaporated to dryness. The residue was parti-

tioned between CHCl₃and H₂O and the organic layer sepa-

rated, dried (Na₂SO₄), and concentrated in vacuo. Flash

chromatography on silica eluting with 60-80 °C petroleum

ether with increasing concentrations of EtOAc gave 9c (0.29

g, 67%) as a pale-yellow oil. ¹H NMR (CDCl₃): δ 7.80-7.67

(2H, m, H4, H6), 5.59 (1H, brs, OH), 4.80 (2H, t, OCH₂), 3.22

(2H, t, CH₂).

N-[3-[2-(Dim eth yla m in o)eth oxy]-4-m eth oxyp h en yl]-2′-

m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-yl)bip h en yl-4-ca r -

boxa m id e (7). A stirred suspension of 2′-methyl-4′-(5-methyl-

1,2,4-oxadiazol-3-yl)biphenyl-4-carboxylic acid (0.13 g, 0.0005

mol) in SOCl₂(5 mL) was heated under reflux for 1 h. After

cooling to room temperature, the solvent was removed in

vacuo. A solution of 3-[2-(dimethylamino)ethoxy]-4-methoxya-

niline, 7b (0.10 g, 0.0005 mol), in THF (5 mL) was treated with

a solution of NaOH (0.04 g) in H₂O (0.6 mL), and a solution of

the acid chloride in THF (15 mL) was added. The reaction

mixture was stirred overnight at room temperature. After

removal of the solvent in vacuo, the residue was dissolved in

CH₂Cl₂, washed with H₂O, dried (MgSO₄), and concentrated

in vacuo. Flash chromatography on silica gel eluting with CH₂-

Cl₂with increasing concentrations of MeOH gave 7 (0.10 g,

43%) as a white solid. ¹H NMR (CDCl₃): δ 8.07 (5H, m,

biphenyl H2, H6, H′3, H′5, NH), 7.55-7.40 (3H, m, biphenyl

H3, H5, H′6), 7.35 (1H, d, H2), 7.08 (1H, dd, H6), 6.87 (1H, d,

H5), 4.18 (2H, t, OCH₂), 3.87 (3H, s, OCH₃), 2.82 (2H, t, CH₂-

NMe₂), 2.70 (3H, s, CH₃), 2.38 (9H, s, CH3, N(CH₃)₂). HRMS

(FAB): calcd for 486.227 (M⁺), found 486.227. Purity was

determined as >99% by HPLC (method A), retention time

23.21 min.

N ,N -Dim e t h yl-3-(5-a m in o-2-m e t h oxyp h e n yl)a cr yl-

a m id e (8a ). 4-Amino-2-bromoanisole (1.70 g, 0.008 mol), tri-

o-tolylphosphine (0.21 g, 0.007 mol), N,N-dimethylacrylamide

(0.95 mL, 0.009 mol), Et₃N (2.9 mL, 0.002 mol), Pd(OAc)₂(0.04

g, 0.002 mol), and dry DMF (4 mL) were heated together with

stirring, under argon at 110 °C. After 4 h, the reaction mixture

was allowed to cool and was partitioned between EtOAc and

H₂O. The aqueous layer was extracted with EtOAc, and the

combined organic layers were dried (Na₂SO₄) and concentrated

in vacuo. The residual oil was purified by column chroma-

tography on silica using EtOAc increasing polarity to 10%

MeOH/EtOAc to give 8a (0.36 g, 20%) as a yellow solid. ¹H

NMR (CDCl₃): δ 7.83 (1H, d, H6), 6.95 (1H, d, H4), 6.87 (1H,

d, H3), 6.75 (1H, d, dCH), 6.68 (1H, dd, dCH), 3.80 (3H, s,

OCH₃), 3.48 (2H, s, NH₂), 3.18 (3H, s, NCH₃), 3.05 (3H, s,

NCH₃).

N,N-Dim et h yl-3-(5-a m in o-2-m et h oxyp h en yl)p r op yl-

a m in e (8b). A solution of N,N-dimethyl-3-(5-amino-2-meth-

oxyphenyl)acrylamide, 8a (0.22 g, 0.001 mol), in EtOH (30 mL)

was hydrogenated over 10% Pd/C until H₂uptake was com-

plete. The catalyst was removed by filtration through Kie-

selguhr and the filtrate concentrated in vacuo. The residue

was dissolved in dry THF (15 mL) under an argon atmosphere

and treated with LiAlH₄(0.05 g, 0.0014 mol). The reaction

mixture was heated under reflux for 4 h. After cooling, H₂O

(0.05 mL) was added, followed by 10% aqueous NaOH (0.08

mL), followed by H₂O (0.13 mL). After stirring for a further

30 min, the mixture was filtered. Evaporation of the filtrate

under reduced pressure gave 8b (0.15 g, 80%) as a brown oil.

¹H NMR (CDCl₃): δ 6.69 (1H, d, H6), 6.52 (2H, m, H3, H4),

3.75 (3H, s, OCH₃), 3.40 (2H, s, NH₂), 2.53 (2H, t, CH₂), 2.30

(2H, t, CH₂NMe₂), 2.20 (6H, s, N(CH₃)₂), 1.75 (2H, m, CH₂CH₂-

CH₂).

N-[3-[3-(Dim eth yla m in o)p r op yl]-4-m eth oxyp h en yl]-2′-

m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-yl)bip h en yl-4-ca r -

boxa m id e (8). 2′-Methyl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)-

biphenyl-4-carboxylic acid (0.21 g, 0.007 mol) was suspended

in CH₂Cl₂(5 mL) and treated with oxalyl chloride (0.09 mL,

0.001 mol), followed by 1 drop of dry DMF with stirring. After

4 h, the reaction mixture was concentrated in vacuo to give

the crude acid chloride as a yellow solid. A solution of N,N-

dimethyl-3-(5-amino-2-methoxyphenyl)propylamine, 8b (0.014

g 0.007 mol), in CH₂Cl₂(10 mL) was treated with Et₃N (0.1

mL, 0.007 mol); then the crude acid chloride in CH₂Cl₂(4 mL)

7-[2-(Dim eth yla m in o)eth oxy]-5-n itr o-2,3-d ih yd r oben -

zofu r a n (9d ). A solution of 7-hydroxy-5-nitro-2,3-dihydroben-

zofuran, 9c (280 mg, 0.0015 mol), in DME (20 mL) was treated

with N,N-dimethylaminoethyl chloride hydrochloride (0.49 g,

0.0033 mol) and K₂CO₃(3 g, 0.02 mol), and the mixture was

heated to reflux under argon for 18 h. The solvent was

removed in vacuo and the residue partitioned between H₂O

and EtOAc. The organic phase was removed and extracted

with 5 M HCl. The acid layer was separated, washed with

EtOAc, and then basified by the addition of K₂CO₃. The

product was extracted into EtOAc, the resulting solution

washed with H₂O and dried (Na₂SO₄), and the solvent removed

in vacuo to give 9d (0.28 g, 73%) as a yellow solid. ¹H NMR

(CDCl₃): δ 7.80 (1H, s, H6), 7.71 (1H, s, H4), 4.80 (2H, t,

OCH₂), 4.2 (2H, t, OCH₂CH₂NMe₂), 3.31 (2H, t, CH₂), 2.8 (2H,

t, CH₂NMe₂), 2.36 (6H, s, N(CH₃)₂).

5-Am in o-7-[2-(d im eth yla m in o)eth oxy]-2,3-d ih yd r oben -

zofu r a n (9e). A solution of 7-[2-(dimethylamino)ethoxy]-5-

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1231

nitro-2,3-dihydrobenzofuran, 9d (0.27 g, 0.001 mol), in EtOH

was hydrogenated over 10% Pd/C at room temperature and

atmospheric pressure until hydrogen uptake ceased. The

catalyst was removed by filtration and the filtrate concentrated

in vacuo to give 9e (0.28 g, 94%) as a yellow oil. ¹H NMR

(CDCl₃): δ 6.25 (1H, s, H6), 6.17 (1H, s, H4), 4.51 (2H, t,

OCH₂), 4.19 (2H, t, OCH₂CH₂NMe₂), 3.95 (2H, brs, NH₂), 3.12

(2H, t, CH₂), 2.89 (2H, t, CH₂NMe₂), 2.44 (6H, s, N(CH₃)₂).

tion and the filtrate concentrated in vacuo to give 10c (0.84 g,

100%) as a brown oil. ¹H NMR (CDCl₃): δ 6.58 (1H, d, H5),

6.08 (1H, d, H7), 5.98 (1H, dd, H8), 4.13 (2H, t, NCH₂CH₂-

NMe₂), 3.35 (6H, m, OCH₂CH₂N, NH₂), 2.50 (2H, t, CH₂NMe₂),

2.30 (6H, s, N(CH₃)₂).

N-[4-[2-Dim eth yla m in o)eth yl]-3,4-d ih yd r o-2H-1,4-ben -

zoxa zin -6-yl)-2′-m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-yl)-

bip h en yl-4-ca r boxa m id e (10). 2′-Methyl-4′-(5-methyl-1,2,4-

oxadiazol-3-yl)biphenyl-4-carboxylic acid (0.19 g, 0.007 mol)

was suspended in CH₂Cl₂(15 mL) and treated with oxalyl

chloride (0.065 mL, 0.074 mol), followed by 1 drop of dry DMF

with stirring. After 4 h, the reaction mixture was concentrated

in vacuo to give the crude acid chloride as a yellow solid. The

solid was redissolved in CH₂Cl₂(10 mL) and added to a

solution of 6-amino-3,4-dihydro-4-[2-(dimethylamino)ethyl]-

2H-1,4-benzoxazine, 10c (0.14 g, 0.007 mol), in CH₂Cl₂(10 mL)

containing Et₃N (0.19 mL, 0.0014 mol) under argon. After 19

h at room temperature, the reaction mixture was treated with

H₂O (20 mL) and extracted with CH₂Cl₂, and the combined

organic layers were dried (Na₂SO₄) and evaporated under

reduced pressure. The residue was purified using flash

chromatography on silica and CH₂Cl₂to give 10 (0.90 g, 30%)

as a pale-yellow solid. ¹H NMR (CDCl₃): δ 8.02 (1H, s,

CONH), 7.98 (3H, m, biphenyl H2, H6, H5′), 7.78 (1H, brs,

biphenyl H3′), 7.48 (2H, d, biphenyl H3, H5), 7.35 (1H, d,

biphenyl H6′), 7.22 (1H, s, H5), 6.77 (2H, brs, H7, H8), 4.23

(2H, t, OCH₂), 3.45 (4H, m, CH₂NCH₂), 2.70 (3H, s, CH₃), 2.59

(2H, t, CH₂NMe₂), 2.35 (3H, s, CH₃), 2.31 (6H, s, N(CH₃)₂).

HRMS m/e 497.255 (M⁺), found 497.256. Purity was deter-

mined as 95.5% by HPLC (method B), retention time 24.15

min.

N-[3-[2-(Dim et h yla m in o)et h oxy]-4-m et h oxyp h en yl]-

a m in oa ceta ld eh yd e Dim eth yl Aceta l (11a ). A solution of

3-[2-(dimethylamino)ethoxy]-4-methoxyaniline, 7b (5.7 g, 0.027

mol), in EtOH (120 mL) was treated with a solution of 2,2-

dimethoxyacetaldehyde in methyl tert-butyl ether (9.5 g, 40%

solution, 0.036 mol), and the mixture was allowed to stand at

room temperature for 16 h. The solution was then hydroge-

nated over 10% Pd/C at room temperature and atmospheric

pressure for 7 h. The catalyst was removed by filtration

through Kieselguhr and the filtrate concentrated in vacuo. The

residue was dissolved in EtOAc and the solution washed with

H₂O, dried (Na₂SO₄), and concentrated in vacuo. The product

was purified by column chromatography on silica gel eluting

with 5% MeOH, 95% CHCl₃to give 11a (5.4 g, 67%) as a red

oil. ¹H NMR (CDCl₃): δ 6.75 (1H, d, H6), 6.30 (1H, d, H2),

6.19 (1H, dd, H5), 4.56 (1H, t, NH), 4.08 (2H, t, OCH₂), 3.78

(3H, s, OCH₃), 3.61 (1H, brt, CH(OMe)₂), 3.42 (6H, s, O(CH₃)₂),

3.20 (2H, t, NHCH₂CH), 2.78 (2H, t, CH₂NMe₂), 2.34 (6H, s,

N(CH₃)₂).

6-[2-(Dim e t h yla m in o)e t h oxy]-5-m e t h oxy-1H -in d ole

(11b). A stirred solution of N-[3-[2-(dimethylamino)ethoxy]-

4-methoxyphenyl]aminoacetaldehyde dimethyl acetal, 11a (5.3

g, 0.018 mol), in TFA (22 mL) at -5 °C under argon was

treated dropwise over 40 min with TFAA (22 mL). After a

further 30 min, TFA (33 mL) was added and the reaction

mixture was heated at reflux for 7 h. The solvent was removed

in vacuo and the residue basified with 10% aqueous Na₂CO₃

solution and extracted with EtOAc. The extract was dried

(Na₂SO₄) and concentrated in vacuo and the residue dissolved

in MeOH (100 mL), treated with K₂CO₃(10 g), and stirred at

room temperature for 3 h. The mixture was concentrated in

vacuo and the residue treated with H₂O (60 mL) and extracted

with EtOAc. The organic solution was dried (Na₂SO₄) and

concentrated in vacuo to give 11b (3.5 g, 83%) as a brown solid.

¹H NMR (DMSO-d₆) δ 7.14 (1H, brt, NH), 7.05 (1H, s, H7),

6.96 (1H, s, H4), 6.27 (1H, brs, H2), 4.03 (2H, t, OCH₂), 3.75

(3H, s, OCH₃), 3.45 (1H, brs, H3), 2.70 (2H, t, CH₂NMe₂), 2.27

(6H, s, N(CH₃)₂).

N -[7-[2-(Dim e t h yla m in o)e t h oxy]-2,3-d ih yd r ob e n zo-

fu r a n -5-yl]-2′-m et h yl-4′-(5-m et h yl-1,2,4-oxa d ia zol-3-yl)-

bip h en yl-4-ca r boxa m id e (9). A stirred solution of 2′-meth-

yl-4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-carboxylic acid

(0.17 g, 0.0006 mol) in SOCl₂(5 mL) was heated under reflux

for 1 h. After cooling to room temperature, the solvent was

removed in vacuo. The resulting acid chloride was redissolved

in THF (15 mL) and the solution treated with 5-amino-7-[2-

(dimethylamino)ethoxy]-2,3-dihydrobenzofuran, 9e (0.13 g,

0.0006 mol), in THF (10 mL) and a solution of NaOH (0.045

g, 0.0011 mol) in H₂O (0.5 mL). The mixture was stirred at

room temperature overnight and the solvent removed in vacuo.

The residue was partitioned between H₂O and CH₂Cl₂and the

organic phase separated, dried (Na₂SO₄), and concentrated in

vacuo to give the crude product which was chromatographed

on silica eluting with CH₂Cl₂with increasing quantities of

methanol to give 9 (0.12 g, 45%) as an off-white solid. ¹H NMR

(CDCl₃): δ 8.07 (1H, s, CONH), 8.02-7.90 (4H, m, biphenyl,

H2, H6, H3′, H5′), 7.45 (2H, d, biphenyl H3, H5), 7.33 (1H, d,

biphenyl H6′), 7.25 (1H, s, H4), 7.12 (1H, s, H6), 4.60 (2H, t,

OCH₂), 4.23 (2H, t, OCH₂CH₂NMe₂), 3.23 (2H, t, CH₂), 2.92

(2H, t, CH₂NMe₂), 2.70 (3H, s, CH₃), 2.48 (6H, s, N(CH₃)₂), 2.32

(3H, s, CH₃). HRMS (FAB): calcd for 499.234 (M⁺), found

499.236. Purity was determined as 93.1% by HPLC (method

A), retention time 17.45 min.

4-[2-(Dim eth yla m in o)eth yl]-6-n itr o-2H-1,4-ben zoxa zin -

3(4H)-on e (10a ). To a suspension of 6-nitro-2H-1,4-benzox-

azin-3(4H)-one (1.0 g, 0.0057 mol) in dry THF (20 mL) at 0 °C

under argon was added NaH (0.2 g, 80% dispersion in mineral

oil, 0.0057 mol). A solution of N,N-dimethylaminoethyl chlo-

ride (2.3 g, 0.028 mol) in dry toluene (15 mL) was added, and

the reaction mixture was heated under reflux for 19 h. After

cooling, H₂O was added dropwise until effervescence ceased;

then the mixture was separated and the aqueous further

extracted with EtOAc. The organic layers were combined,

dried (Na₂SO₄), and evaporated under reduced pressure to give

10a (1.10 g, 79%) as a pale-brown solid. ¹H NMR (CDCl₃): δ

8.03 (1H, d, H5), 7.94 (1H, dd, H7), 7.04 (1H, d, H8), 4.73 (2H,

s, OCH₂CO), 4.11 (2H, t, NCH₂), 2.60 (2H, t, CH₂NMe₂), 2.35

(6H, s, N(CH₃)₂).

3,4-Dih ydr o-4-[2-(d im eth yla m in o)eth yl]-6-n itr o-2H-1,4-

ben zoxa zin e (10b). Boron trifluoride diethyl etherate (2 mL,

0.0016 mol) was added dropwise to a suspension of sodium

borohydride (0.46 g, 0.0012 mol) in dry THF (30 mL) at 0 °C

under argon. After 1 h, a solution of 4-[2-(dimethylamino)-

ethyl]-6-nitro-2H-1,4-benzoxazin-3(4H)-one, 10a (1.10 g, 0.004

mol), in dry THF (20 mL) was added, and the reaction mixture

was heated under reflux for 2 h. After cooling in ice, aqueous

NaHCO₃was added dropwise until effervescence ceased; then

the solvent was removed under reduced pressure and the

residue dissolved in a mixture of EtOH (10 mL) and 5 M HCl

(10 mL) and heated under reflux for 1 h. After cooling, the

solvent was removed under reduced pressure. The residue was

treated with saturated K₂CO₃solution to pH 8 and then

extracted with EtOAc. The combined organic extracts were

dried (Na₂SO₄) and evaporated under reduced pressure to give

10b (0.94 g, 92%) as a pale-yellow oil. ¹H NMR (CDCl₃): δ

7.52 (2H, m, H5, H7), 6.78 (1H, d, H8), 4.30 (2H, t, NCH₂CH₂-

NMe₂), 3.42 (4H, m, OCH₂CH₂N), 2.56 (2H, t, CH₂NMe₂), 2.31

(6H, s, N(CH₃)₂).

6-Am in o-3,4-d ih yd r o-4-[2-(d im et h yla m in o)et h yl]-2H -

1,4-ben zoxa zin e (10c). A solution of 3,4-dihydro-4-[2-(dim-

ethylamino)ethyl]-6-nitro-2H-1,4-benzoxazine, 10b (0.94 g,

0.004 mol), in EtOH (100 mL) was hydrogenated over 10%

Pd/C at room temperature and atmospheric pressure until

hydrogen uptake ceased. The catalyst was removed by filtra-

2,3-Dih yd r o-6-[2-(d im eth yla m in o)eth oxy]-5-m eth oxy-

1H-in d ole (11c). A stirred solution of 6-[2-(dimethylamino)-

ethoxy]-5-methoxy-1H-indole, 11b (0.50 g, 0.002 mol), in AcOH

(10 mL) at room temperature was treated portionwise over

15 min with sodium cyanoborohydride (0.25 g, 0.004 mol).

1232 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

After 2 h the solution was diluted with H₂O (50 mL), basified

by addition of Na₂CO₃, and then extracted with EtOAc. The

organic extract was dried (Na₂SO₄) and concentrated in vacuo

to give 11c (0.37 g, 73%) as a brown oil. ¹H NMR (DMSO-d₆):

δ 6.75 (1H, s, H7), 6.37 (1H, s, H4), 4.05 (2H, t, OCH₂), 3.90

(1H, brs, NH), 3.77 (3H, s, OCH₃), 3.52 (2H, t, NHCH₂), 2.96

(2H, t, CH₂), 2.73 (2H, t, CH₂NMe₂), 2.32 (6H, s, N(CH₃)₂).

402.118 (M⁺), found 402.119. Purity was determined as 95.2%

by HPLC (method C), retention time 13.99 min.

4-[(2-Iod op h en oxy)m et h yl]-1-m et h yl-1,2,3,6-t et r a h y-

d r op yr id in e (15a ). A stirred soluton of 2-iodophenol (0.5 g,

0.002 mol), triphenylphosphine (0.53 g, 0.002 mol), and

1-methyl-1,2,3,6-tetrahydropyridine-4-methanol (0.32 g, 0.0025

mol) in THF (20 mL) at 0 °C under argon was treated with a

solution of diethyl azodicarboxylate (0.4 mL, 0.0025 mol) in

THF (3 mL). The reaction mixture was stirred at room

temperature for 4 h and then concentrated in vacuo and the

residue basified with 10% aqueous Na₂CO₃solution and

extracted with EtOAc. The EtOAc layer was extracted with

1 M HCl and the acid layer basified by addition of K₂CO₃and

extracted with EtOAc. The organic extract was dried (Na₂-

SO₄) and concentrated in vacuo. The residue was purified by

column chromatography on silica gel using 15% EtOAc, 85%

Et₂O to give 15a (0.50 g, 74%) as a yellow oil. ¹H NMR: δ

7.76 (1H, dd, H3), 7.27 (1H, m, H5), 6.81 (1H, dd, H4), 6.68

(1H, td, H6), 5.84 (1H, m, dCH), 4.47 (2H, s, OCH₂), 2.99 (2H,

m, CHNCH), 2.58 (2H, t, dCHCH₂CH₂), 2.40-2.25 (5H, m,

CHNMeCH, NCH₃).

2,3-Dih yd r o-1′-m e t h ylsp ir o[b e n zofu r a n -3,4′-p ip e r i-

d in e] (15b). A stirred solution of 4-[(2-iodophenoxy)methyl]-

1-methyl-1,2,3,6-tetrahydropyridine, 15a (0.50 g, 0.0015 mol),

and AIBN (0.005 g) in benzene (75 mL) was heated to reflux

under argon and then treated dropwise over 1 h with a solution

of tributyltin hydride (1 mL, 0.0035 mol) in benzene (10 mL).

The reaction mixture was heated under reflux for a further 3

h, then more tributyltin hydride (0.5 mL, 0.0017 mol) and

AIBN (0.005 g) were added, and heating was continued for 4

h. The reaction mixture was allowed to cool and then

concentrated in vacuo and the residue treated with 2 M HCl,

washed with EtOAc, basified with K₂CO₃, and extracted with

EtOAc. This organic extract was dried (Na₂SO₄) and concen-

trated in vacuo. The residue was purified by column chroma-

tography on silica gel using 2% MeOH, 98% CHCl₃to give 15b

(0.21 g, 69%) as a pale-yellow oil. ¹H NMR (CDCl₃): δ 7.15

(2H, m, H4, H6), 6.88 (1H, t, H5), 6.79 (1H, d, H7), 4.35 (2H,

s, OCH₂), 2.87 (2H, m, CHNMeCH), 2.33 (3H, s, NCH₃), 2.02

(4H, m, CHNMeCH, CHCH), 1.78 (2H, m, CHCH).

2,3-Dih yd r o-6-[2-(d im eth yla m in o)eth oxy]-5-m eth oxy-

1-[4-[2-m et h yl-4-(5-m et h yl-1,2,4-oxa d ia zol-3-yl)p h en yl]-

ben zoyl]-1H-in d ole (11). A stirred suspension of 2′-methyl-

4′-(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-carboxylic acid (0.17

g, 0.0055 mol) in thionyl chloride (5 mL) was heated under

reflux for 2 h. The solution was concentrated in vacuo to leave

the acid chloride as a yellow solid. This was redissolved in

THF (5 mL) and added to a stirred solution of 2,3-dihydro-6-

[2-(dimethylamino)ethoxy]-5-methoxy-1H-indole, 11c (0.12 g,

0.0005 mol), in a mixture of H₂O (5 mL) and THF (5 mL)

containing NaOH (0.045 g, 0.0011 mol). The mixture was kept

at room temperature for 18 h, then concentrated to ap-

proximately 50% volume, diluted with 10% Na₂CO₃solution

(20 mL), and extracted with EtOAc. The organic extract was

dried (Na₂SO₄) and concentrated in vacuo and the product

purified by column chromatography on silica gel eluting with

5% MeOH, 95% CHCl₃. Crystallization from Et₂O afforded

11 (0.055 g, 21%) as a white solid. ¹H NMR (DMSO-d₆at 80

°C): δ 7.96 (1H, s, H3′), 7.90 (1H, d, biphenyl H5′), 7.65 (2H,

d, biphenyl H2, H6), 7.50 (2H, d, biphenyl H3, H5), 7.43 (1H,

1H, biphenyl H6′), 7.35 (1H, br, H7), 6.94 (1H, s, H4), 4.09

(2H, t, OCH₂), 3.93 (2H, brt, NCH₂), 3.76 (3H, s, OCH₃), 3.10-

3.00 (2H, m, CH₂), 2.68 (3H, s, CH₃), 2.61 (2H, t, CH₂NMe₂),

2.36 (3H, s, CH₃), 2.22 (6H, s, N(CH₃)₂). HRMS m/e 512.239

(M⁺), found 512.238. Purity was determined as 95.3% by

HPLC (method B), retention time 22.85 min.

N-[3-(4-Meth yl-1-p ip er a zin yl)p h en yl]-4-br om o-3-m eth -

ylben za m id e (13). A stirred suspension of 4-bromo-3-meth-

ylbenzoic acid (3.5 g, 0.016 mol) in thionyl chloride (100 mL)

was heated under reflux for 2 h and then concentrated to

dryness in vacuo. The resultant acid chloride was redissolved

in THF (25 mL) and the solution added to a mixture of 3-(4-

methyl-1-piperazinyl)aniline (2.0 g, 0.01 mol) in THF contain-

ing NaOH (0.56 g, 0.0014 mol) and H₂O (25 mL). The reaction

mixture was stirred at room temperature for 4 h, then poured

into H₂O, and extracted with CH₂Cl₂. The organic extract was

dried (Na₂SO₄) and concentrated in vacuo. The residue was

purified by column chromatography on silica gel using 1%

MeOH, 99% CH₂Cl₂as eluant to give 13 (1.1 g 28%) as an off-

white solid. ¹H NMR (CDCl₃): δ 7.80 (1H, brs, CONH), 7.72

(1H, d, H6 adjacent to CO), 7.63 (1H, d, H2 adjacent to CO),

7.40-7.52 (2H, m, H5 adjacent to Br, H5), 7.22 (1H, d, H6),

6.95 (1H, d, H2), 6.71 (1H, dd, H4), 3.25 (4H, t, CH₂NCH₂),

2.58 (4H, t, CH₂NMeCH₂), 2.44 (3H, s, CH₃), 2.32 (3H, s,

NCH₃). HRMS (FAB): calcd for 388.102 (M⁺), found 388.103.

Purity was determined as 96.0% by HPLC (method A),

retention time 15.63 min.

2,3-Dih yd r o-1′-m eth yl-5-n itr osp ir o[ben zofu r a n -3,4′-p i-

p er id in e] (15c). A stirred solution of 2,3-dihydro-1′-methyl-

spiro[benzofuran-3,4′-piperidine], 15b (0.20 g, 0.0013 mol), in

acetic anhydride (5 mL) at 0 °C under argon was treated

portionwise over 15 min with copper(II) nitrate hemipentahy-

drate (0.35 g, 0.0015 mol). The reaction mixture was kept at

0 °C for a total of 1.5 h and then allowed to warm to room

temperature over 0.5 h. The mixture was poured into water/

ice (100 mL), basified by the careful addition of K₂CO₃, and

then extracted with EtOAc. The organic extract was dried

(Na₂SO₄) and concentrated in vacuo to afford 15c (0.25 g, 79%)

as a pale-brown oil. ¹H NMR: δ 8.11 (1H, dd, H4), 8.03 (1H,

d, H6), 6.83 (1H, d, H7), 4.53 (2H, s, OCH₂), 2.91 (2H, m,

CHNMeCH), 2.34 (3H, s, NCH₃), 2.02 (4H, m, CHNMeCH,

CHCH), 1.81 (2H, m, CHCH).

N-[4-Meth yl-3-(4-m eth yl-1-piper azin yl)ph en yl]-4-br om o-

3-m eth ylben za m id e (14). A stirred suspension of 4-bromo-

3-methylbenzoic acid (0.56 g, 0.0026 mol) in thionyl chloride

(25 mL) was heated under reflux for 2 h and then concentrated

to dryness in vacuo. The resultant acid chloride was redis-

solved in THF (3 mL) and the solution added to a mixture of

4-methyl-3-(4-methyl-1-piperazinyl)aniline (0.36 g, 0.0018 mol)

in THF containing NaOH (0.1 g, 0.0025 mol) and H₂O (3 mL).

The reaction mixture was stirred at room temperature for 4

h, then poured into H₂O, and extracted with CH₂Cl₂. The

organic extract was dried (Na₂SO₄) and concentrated in vacuo

to give a beige foam which was purified by column chroma-

tography on silica gel using 1% MeOH, 99% CH₂Cl₂as eluant

to give 14 (0.44 g, 63%) as a pale-cream solid. ¹H NMR

(CDCl₃): δ 7.78 (1H, brs, CONH),7.72 (1H, d, H6 adjacent to

CO), 7.66 (1H, d, H2, adjacent to CO), 7.52 (1H, dd, H5

adjacent to Br), 7.25 (2H, m, H5, H6), 7.14 (1H, d, H2), 3.00

(4H, t, CH₂NCH₂), 2.62 (4H, 4H, CH₂NMeCH₂), 2.50 (3H, s,

CH₃), 2.38 (3H, s, CH₃), 2.29 (3H, s, NCH₃). HRMS calcd for

5-Am in o-2,3-d ih yd r o-1′-m et h ylsp ir o[b en zofu r a n -3,4′-

p ip er id in e] (15d ). A solution of 2,3-dihydro-1′-methyl-5-

nitrospiro[benzofuran-3,4′-piperidine], 15c (0.25 g, 0.001 mol),

in EtOH (30 mL) was hydrogenated over 10% Pd/C at room

temperature and atmospheric pressure until uptake of hydro-

gen ceased. The catalyst was removed by filtration through

Kieselguhr and the filtrate concentrated in vacuo to give 15d

(0.13 g, 60%) as a brown oil. ¹H NMR (DMSO-d₆): δ 6.42 (2H,

m, H4, H6), 6.33 (1H, dd, H7), 4.22 (2H, s, OCH₂), 2.84 (2H,

d, CHCHNMe), 2.29 (3H, s, NCH₃), 2.12 (2H, t, CHNMeCH),

1.82 (2H, m, CHCCH), 1.61 (2H, d, CHCCH).

N-(2,3-Dih yd r o-1′-m eth ylsp ir o[ben zofu r a n -3,4′-p ip er i-

d in e]-5-yl)-2′-m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-yl)bi-

p h en ylca r boxa m id e (15). A suspension of 2′-methyl-4′-(5-

methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-carboxylic acid (0.20 g,

0.0007 mol) in SOCl₂(3 mL) was heated under reflux for 1 h.

After cooling to room temperature, the solvent was removed

in vacuo. This was dissolved in CH₂Cl₂(2 mL) and added to

a stirred solution of 5-amino-2,3-dihydro-1′-methylspiro[ben-

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1233

zofuran-3,4′-piperidine], 15d (0.13 g, 0.0006 mol), and Et₃N

(0.17 mL, 0.0013 mol) in CH₂Cl₂(5 mL). The reaction mixture

was stirred at room temperature for 20 h, then treated with

10% aqueous Na₂CO₃solution, and extracted with CH₂Cl₂. The

organic extracts were dried (Na₂SO₄) and concentrated in

vacuo and the residue chromatographed on silica gel eluting

with 15% MeOH, 85% CH₂Cl₂, to give 15 isolated as the

oxalate salt (0.28 g, 80%). ¹H NMR (DMSO-d₆): δ 10.31 (1H,

s, COOH), 8.15-7.90 (4H, m, biphenyl H2, H6, H3′, H5′), 7.70-

7.55 (4H, m, biphenyl H3, H5, H6′, H5), 7.45 (1H, d, H4), 6.83

(1H, d, H7), 4.50 (2H, s, OCH₂), 3.41 (2H, d, CHNMeCH), 3.02

(2H, t, CHNMeCH), 2.75 (3H, s, CH₃), 2.70 (3H, s, CH₃), 2.37

(3H, s, NCH₃), 2.20-1.85 (4H, m, CH₂CCH₂). HRMS (FAB):

calcd for 495.239 (M⁺), found 495.237. Purity was determined

as >99% by HPLC (method C), retention time 14.45 min.

1-Acetyl-6-br om o-2,3-d ih yd r o-5-[(1-m eth yl-1,2,3,6-tet-

r a h yd r op yr id in -4-yl)m eth oxy]-1H-in d ole (5a ). A stirred

suspension of powdered 1-acetyl-6-bromo-2,3-dihydro-1H-in-

dol-5-ol (19.0 g, 0.074 mol) in dry THF (1700 mL) at room

temperature under argon was treated with triphenylphosphine

(19.6 g, 0.074 mol) and 1-methyl-1,2,3,6-tetrahydropyridine-

4-methanol (9.5 g, 0.075 mol) followed by the dropwise addition

over 15 min of a solution of diethyl azodicarboxylate (11.8 mL,

0.075 mol) in THF (40 mL). A mild exotherm occurred, and

the insoluble material dissolved. The solution was warmed

at 32 °C for 1 h and then concentrated in vacuo to ap-

proximately 500-mL volume. The solid which had formed was

collected by filtration and dried affording 16.1 g of solid. The

filtrate was concentrated in vacuo, the residue treated with

EtOAc (700 mL) and 1 M HCl (500 mL) and shaken well, and

the acid layer separated. This was washed with EtOAc,

basified with 40% NaOH, and extracted with EtOAc, followed

by CHCl₃. The combined organic extracts were dried (Na₂-

SO₄) and concentrated in vacuo to leave a solid, which was

crystallized from EtOAc giving a total yield of 19.9 g (75%) of

5a , isolated as a yellow solid. ¹H NMR (CDCl₃): δ 8.42 (1H,

s, H7), 6.72 (1H, s, H4), 5.80 (1H, brs, dCH), 4.41 (2H, s,

OCH₂), 4.04 (2H, t, CH₂), 3.12 (2H, t, NCH₂), 2.97 (2H, brs,

dCHCH₂NMe), 2.58 (2H, t, dCHCH₂CH₂NMe), 2.38 (3H, s,

COCH₃), 2.28 (2H, brs, CHNMeCH), 2.18 (3H, s, NCH₃).

5-Acetyl-1′-m eth yl-2,3,6,7-tetr a h yd r osp ir o[fu r o[2,3-f]-

in dole-3,4′-piper idin e] (5b). A stirred suspension of 1-acetyl-

6-bromo-2,3-dihydro-5-[(1-methyl-1,2,3,6-tetrahydropyridin-4-

yl)methoxy]-1H-indole, 5a (20.8 g, 0.057 mol), in benzene (1500

mL) was treated with AIBN (400 mg) and heated to 75 °C

under argon and then treated dropwise over 0.5 h with a

solution of tributyltin hydride (23 mL, 0.085 mol) in benzene

(200 mL). The mixture was heated under reflux for 6 h and

then concentrated in vacuo. The residue was treated with 2

M HCl (900 mL) and EtOAc (600 mL) and then shaken well

and the acid layer separated, washed with EtOAc, and then

basified with 40% aqueous NaOH solution, keeping the tem-

perature below 20 °C. A precipitate was formed which was

filtered off, washed with H₂O, and dried to afford 5b (14.8 g,

91%) as a white solid. ¹H NMR (CDCl₃): δ 8.11 (1H, s, H4),

6.60 (1H, s, H8), 4.36 (2H, s, OCH₂), 4.03 (2H, t, CH₂), 3.10

(2H, t, NCH₂), 2.92-2.78 (2H, m, CHNMeCH), 2.30 (3H, s,

CH₃), 2.18 (3H, s, NCH₃), 2.15-1.90 (4H, m, CHNMeCH,

CHCCH), 1.80-1.63 (2H, m, CHCCH).

1′-Meth yl-2,3,6,7-tetr a h yd r osp ir o[fu r o[2,3-f]in d ole-3,4′-

p ip er id in e] (5c). A stirred solution of 5-acetyl-1′-methyl-

2,3,6,7-tetrahydrospiro[furo[2,3-f]indole-3,4′-piperidine], 5b (14.5

g, 0.051 mol), in a mixture of 5 M HCl (250 mL) and EtOH

(100 mL) was heated under reflux under argon for 2 h and

then stirred for 16 h at room temperature. The EtOH was

removed in vacuo and the remaining solution cooled in an ice

bath and basified to pH 12 by addition of 40% aqueous NaOH.

The resultant white precipitate was collected by filtration,

washed with H₂O, and dried to afford 6.0 g of 5c. The filtrate

was extracted with EtOAc, followed by CHCl₃. The combined

extract was dried (Na₂SO₄) and concentrated in vacuo leaving

a beige solid (6.2 g) and affording a total of 12.2 g (99%) of

compound 5c as a beige solid. ¹H NMR (CDCl₃): δ 6.61 (1H,

s, H4), 6.46 (1H, s, H8), 4.31 (2H, s, OCH₂), 3.53 (2H, t, CH₂),

2.96 (2H, t, NCH₂), 2.90 (3H, m, 3H, CHNMeCH, NH), 2.33

(3H, s, NCH₃), 1.99 (4H, m, CHNMeCH, CHCCH), 1.74 (2H,

m, CHCCH).

1′-Meth yl-5-[[2′-m eth yl-4′-(5-m eth yl-1,2,4-oxa d ia zol-3-

yl)bip h en yl-4-yl]ca r bon yl]-2,3,6,7-tetr a h yd r osp ir o[fu r o-

[2,3-f]in d ole-3,4′-p ip er id in e] (5). A stirred suspension of 2′-

m et h yl-4′-(5-m et h yl-1,2,4-oxa dia zol-3-yl)biph en yl-4-ca r -

boxylic acid (0.036 g, 0.0012 mol) in thionyl chloride (3 mL)

was heated under reflux for 1 h. After cooling to room

temperature, the solvent was removed in vacuo to leave the

acid chloride as a yellow solid. This was dissolved in CH₂Cl₂

(2 mL) and added to a stirred solution of 1′-methyl-2,3,6,7-

tetrahydrospiro[furo[2,3-f]indole-3,4′-piperidine], 5c (0.030 g,

0.0012 mol), and Et₃N (0.06 mL, 0.0043 mol) in CH₂Cl₂and

kept at room temperature for 3 days. The mixture was washed

with aqueous K₂CO₃solution and brine, then dried (Na₂SO₄),

and concentrated in vacuo. The residue was chromatographed

on silica gel eluting with CH₂Cl₂with increasing concentra-

tions of MeOH to afford 5 (0.019 g, 30%) which was converted

to the hydrochloride salt, isolated as a white solid, mp >280

°C. ¹H NMR (DMSO-d₆): δ 7.95-7.80 (3H, m, biphenyl H2,

H6, H3′), 7.62 (2H, m, biphenyl H5′, H6′), 7.47 (2H, d, biphenyl

H3, H5), 7.39 (1H, d, H4), 6.72 (1H, s, H8), 4.44 (2H, s, OCH₂),

4.01 (2H, t, CH₂), 3.42-3.22 (4H, m, NCH₂, CHNMeCH), 2.92

(2H, brt, CHNMeCH), 2.66 (3H, brs, NCH₃), 2.62 (3H, s, CH₃),

2.31 (3H, s, CH₃), 2.02-1.84 (4H, m, CH₂CCH₂). HRMS

(FAB): calcd for 521.255 (M⁺), found 521.255. Purity was

determined as >99% by HPLC (method A), retention time

18.86 min. Anal. (C₃₂H₃₂N₄O₃‚HCl‚ H₂O) C, H, N.

Com p u ta tion a l Meth od s. Molecular mechanics and dy-

namics calculations on the protein structures with and without

bound ligands were performed using the CHARMm program.

Geometries for the ligands were generated by optimization

with the COSMIC molecular mechanics force field or, where

parameters were unavailable, with the semiempirical molec-

ular orbital program VAMP, using the AM1 Hamiltonian.

Natural atomic orbital charges were used for the docking

calculations, and these were generated from single-point

Hartree-Fock calculations on the optimized structures, using

the Spartan suite of programs (Wave Function Inc.) with a

3-21G* basis set. The results for the CHARMm calculations

were visualized using the program QUANTA (Molecular

Simulations, Inc.). All calculations were performed on either

a Silicon Graphics 4D-380 Powerserver or a DEC alpha 2100

and the results visualized on a Silicon Graphics Indigo-2

Extreme workstation.

Refer en ces

(1) Hoyer, D.; Clarke, D. E.; Fozard, J . R.; Hartig, P. R.; Martin, G.

R.; Mylechrane, E. J .; Saxena, P. R.; Humphrey, P. P. A.

International Union of Pharmacology classification of receptors

for 5-hydroxytryptamine (serotonin). Pharmacol. Rev. 1994, 46,

157-203.

(2) Leonhardt, S.; Herrick-Davis, K.; Titeler, M. Detection of a novel

serotonin receptor subtype (5-HT_1E) in human brain: interaction

with a GTP binding protein. J . Neurochem. 1989, 53, 465-471.

(3) Schoeffter, P.; Hoyer, D. 5-Hydroxytryptamine 5-HT_1Band

5-HT_1Dreceptors mediating inhibition of adenylate cyclase

activity: pharmacological comparison with special reference to

the effects of yohimbine, rauwolscine and some â-adrenoreceptor

antagonists. Naunyn Schmiedeberg’s Arch. Pharmacol. 1989,

340, 285-292.

(4) Hamon, M.; Emerit, M. B.; El Mestikawy, S.; Gallisot, M. C.;

Gozlan, H. Regional differences in the transduction mechanisms

of 5-HT receptors in the mammalian brain. In Cardiovascular

pharmacology of 5-HT: prospective therapeutic applications;

Sanea, P. R., Walis, D. I., Wouters, W., Bevan, P., Eds.; Kluwer

Academic Publishers: Dordrecht, The Netherlands, 1991; pp

41-59.

(5) Adham, N.; Borden, L. A.; Schechter, L. E.; Gustafson, E. L.;

Cochran, T. L.; Vaysse, P. J . J .; Weinshank, R. L.; Branchek, T.

A. Cell specific coupling of the cloned human 5-HT_1Freceptor to

multiple signal transduction pathways. Naunyn Schmiedeberg’s

Arch. Pharmacol. 1993, 348, 566-575.

(6) Hartig, P. R.; Hoyer, D.; Humphrey, P. P. A.; Martin, G. R.

Alignment of receptor nomenclature with the human genome:

classification of 5-HT_1Band 5-HT_1Dreceptor subtypes. Trends

Pharmacol. Sci. 1996, 17, 103-105.

1234 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8

Gaster et al.

(7) Weinshank, R. L.; Zgombick, J . M.; Macchi, M. J .; Branchek, T.

A.; Hartig, P. R. Human serotonin 1D receptor is encoded by a

subfamily of two distinct genes: 5-HT_1DRand 5-HT_1Dâ. Proc. Natl.

Acad. Sci. U.S.A. 1992, 89, 3630-3634.

(8) Kaumann, A. J .; Frenken, M.; Posival, H.; Brown, A. M. Variable

participation of 5-HT₁-like receptors and 5-HT₂receptors in

serotonin-induced contraction of human isolated coronary arter-

ies. Circulation 1994, 90, 1141-1153.

(9) Peroutka, S. J . Pharmacological differentiation of human 5-HT_1B

and 5-HT_1Dreceptors. Biol. Signals 1994, 3, 217-222.

(10) Pauwels, P. J .; Reihsaus, E.; Palmier, C.; J ournot, L.; Lyons, J .;

Colpaert, F. C. Ketanserin differentiates between cloned human

5-HT_1DRreceptors and cloned human 5-HT_1Dâ, sheep and bovine

caudate nucleus 5-HT_1Dreceptor sites. Cell Pharmacol. 1995,

2, 183-191.

(11) Beer, M. S.; Middlemiss, D. N. Serotonin-5-O-carboxymethyl-

glycyl[¹²⁵I]tyrosin-amide labels the 5-HT_1Dâreceptor subtype in

human cortex. Eur. J . Pharmacol. 1993, 242, 195-198.

(12) Engel, G.; Go¨thert, M.; Hoyer, D.; Schlocker, E.; Hillenbrand,

K. Identity of inhibitory pre-synaptic 5-HT autoreceptors in the

rat brain cortex with 5-HT_1Bbinding sites. Naunyn Schmiede-

berg’s Arch. Pharmacol. 1986, 332, 1-7.

(31) Strader, C. D.; Sigal, I. S.; Candelore, M. R.; Rands, E.; Hill, W.

S.; Dixon, R. A. F. Conserved aspartate residues 79 and 113 of

the â-adrenergic receptor have different roles in receptor func-

tion. J . Biol. Chem. 1988, 263, 10267-10271.

(32) Leonhardt, S.; Herrick-Davis, K.; Titeler, M. Detection of a novel

serotonin receptor subtype (5-HT_1E) in human brain: interaction

with a GTP-binding protein. J . Neurochem. 1989, 53, 465-471.

(33) Parker, E. M.; Izzarelli, D. G.; Lewis-Higgins, L.; Palmer, D.;

Shapiro, R. A. Two amino acid differences in the sixth trans-

membrane domain are partially responsible for the pharmaco-

logical differences between the 5-HT_1Dâand 5-HT_1E5-hydrox-

ytryptamine receptors. J . Neurochem. 1996, 67, 2096-2103.

(34) Hutchins, C. Three-dimensional models of the D1 and D2

dopamine receptors. Endocrine J . 1994, 2, 7-23.

(35) Oxford, A. W.; Mitchell, W. L.; Bradshaw, J .; Clitherow, J . W.

Benzanilide derivatives as 5-HT_1Dantagonists. Eur. Pat. Appl.

0 533 267 A1, March 1993; Chem. Abstr. 1993, 119, 49414c.

(36) Gaster, L. M.; King, F. D.; Wyman, P. A. Heterocyclic bipheny-

lamides usefuls as 5-HT_1Dantagonists. Int. Pat. Appl. WO 95/

15954, J une 1995; Chem. Abstr. 1995, 123, 313977n.

(37) Gaster, L. M. Dihydrobenzofuranyl-biphenyl carboxamides hav-

ing 5-HT_1Dantagonist activity. Int. Pat. Appl. WO 95/17401,

J une 1995; Chem. Abstr. 1995, 123, 340139x.

(38) Ham, P.; Gaster, L. M.; King, F. D.; Duckworth, D. M. Amide

derivatives having 5-HT_1Dantagonist activity. Int. Pat. Appl.

WO95/32967, December 1995; Chem. Abstr. 1995, 124, 232466p.

(39) Wyman, P. A.; Gaster, L. M.; J ennings, A. J . Tricyclic spiro

compounds, process for their preparation and their use as 5-HT_1D

receptor antagonists. Int. Pat. Appl. WO 96/11934, April 1996;

Chem. Abstr. 1996, 125, 114514a.

(13) Fink, K.; Zentner, J .; Gothert, M. Subclassification of presynaptic

5-HT autoreceptors in the human cerebral cortex as 5-HT_1Dâ

receptors. Naunyn Schmiedeberg’s Arch. Pharmacol. 1995, 352,

451-454.

(14) Middlemiss, D. N.; Bremer, M. E.; Smith, S. M. A pharmacologi-

cal analysis of the 5-HT receptors mediating inhibition of 5-HT

release in the guinea pig frontal cortex. Eur. J . Pharmacol. 1988,

157, 101-107.

(15) Davidson, C.; Stamford, J . A. Evidence that 5-HT release in rat

dorsal raphe´ nucleus is controlled by 5-HT_1A, 5-HT_1Band 5-HT_1D

autoreceptors. Br. J . Pharmacol. 1995, 114, 1107-1109.

(16) Buhlen, M.; Fink, K.; Boing, C.; Go¨thert, M. Evidence for

presynaptic location of inhibitory 5-HT_1Dâ-like receptors in the

guinea pig cortex. Naunyn Schmiedeberg’s Arch. Pharmacol.

1996, 353, 281-289.

(17) Roberts, C.; Watson, J .; Burton, M.; Price, G. W.; J ones, B. J .

Functional characterisation of the 5-HT terminal autoreceptor

in the guinea pig brain cortex. Br. J . Pharmacol. 1996, 117, 384-

388.

(18) Pinder, R. M.; Wieringa, J . H. Third generation antidepressants.

Med. Res. Rev. 1993, 13, 259-325.

(19) Norman, T. R.; Leonard, B. E. Fast-acting antidepressants. Can

the need be met? CNS Drugs 1994, 2, 120-131.

(20) Blier, P.; De Montigny, C. Current advances and trends in the

treatment of depression. Trends Pharmacol. Sci. 1994, 15, 220-

226.

(21) Roberts, C.; Thorn, L.; Price, G. W.; Middlemiss, D. N.; J ones,

B. J . Effect of the selective 5-HT_1Dreceptor antagonist GR

127935 on in vivo 5-HT release, synthesis and turnover in the

guinea pig frontal cortex. Br. J . Pharmacol. 1994, 112, 489P.

(22) Skingle, M.; Sleight, A. J .; Fenuik, W. Microdialysis study

investigating the effects of GR 127935, a potent 5-HT_1Dreceptor

antagonist, on cortical levels of 5-HT in the guinea pig. Br. J .

Pharmacol. 1994, 112, C57.

(23) Skingle, M.; Sleight, A. J .; Fenuik, W. Effects of the 5-HT_1D

receptor antagonist GR 127935 on extracellular levels of 5-HT

in the guinea pig frontal cortex as measured by microdialysis.

Neuropharmacology 1995, 34, 377-382.

(24) Skingle, M.; Beattie, D. T.; Scopes, D. I. C.; Starkey, S. J .;

Connor, H. E.; Feniuk, W.; Tyers, M. B. GR 127935: a potent

and selective 5-HT_1Dreceptor antagonist. Behav. Brain Res.

1996, 73, 157-161.

(25) Watson, J .; Burton, M.; Price, G. W.; J ones, B. J .; Thomas, D.;

Faruq, A.; Middlemiss, D. N. GR 127935 acts as a partial agonist

at recombinant human 5-HT_1DRand 5-HT_1Dâreceptors. Br. J .

Pharmacol. 1995, 114, 362P.

(40) Gaster, L. M.; King, F. D.; Wyman, P. A. Tetracyclic spiro

compounds, process for their preparation and their use as 5-HT_1D

receptor antagonists. Int. Pat. Appl. WO 96/19477, J une 1996;

Chem. Abstr. 1996, 125, 142702d.

(41) Oxford, A. W.; Mitchell, W. L.; Bradshaw, J .; Clitherow, J . W.;

Carter, M. Azolylpiperazine benzamide derivatives as 5-HT_1D

antagonists. Eur. Pat. Appl. 0 533 268 A1, March 1993; Chem.

Abstr. 1993, 119, 117270e.

(42) Murphy, R. A.; Kung, H. F.; Kung, M.-P.; Billings, J . Synthesis

and characterisation of iodobenzamide analogues: Potential D-2

dopamine receptor imaging agents. J . Med. Chem. 1990, 33,

171-178.

(43) Chapleo, C. B.; Butler, R. C. M.; England, D. C.; Myers, P. L.;

Roach, A. G.; Smith, C. F.; Stillings, M. R.; Tulloch, I. F.

Heteroaromatic analogues of the R2-adrenoceptor partial agonist

clonidine. J . Med. Chem. 1989, 32, 1627-1630.

(44) Svensson, K. G.; Selander, H.; Karlsson, M.; Nilsson, J . L. G.

The directing effect of annelated rings in aromatic systems -

XI. Bromination of bicyclic phenols with heterocyclic annelated

rings. Tetrahedron 1973, 29, 1115-1118.

(45) Thomas, D. R.; Faruq, S. A.; Balcarek, J . M.; Brown, A. M.

Pharmacological characterisation of [³⁵S]-GTPγS binding to

chinese hamster ovary cell membranes stably expressing cloned

human 5-HT_1Dreceptor subtypes. J . Rec. Sig. Trans. Res. 1995,

15, 199.

(46) Watson, J . M.; Burton, M. J .; Price, G. W.; J ones, B. J .;

Middlemiss, D. N. GR 127935 acts as a partial agonist at

recombinant human 5-HT_1DRand 5-HT_1Dâreceptors. Eur. J .

Pharmacol. 1996, 314, 365-372.

(47) Hatcher, J . P.; Slade, R., C.; Hagan, J . J . 5-HT_1Dreceptors

mediate SK&F 99101H induced hypothermia in the guinea pig.

J . Psychopharmacol. 1995, 9, 234-241.

(48) Hagan, J . J .; Slade, P. D.; Gaster, L. M.; J effrey, P. M.; Hatcher,

J . P.; Middlemiss, D. N. Stimulation of 5-HT_1Breceptors causes

hypothermia in the guinea pig. Eur. J . Pharmacol. 1997, 331,

169-174.

(49) Roberts, C.; Price, G. W.; Gaster, L. M.; J ones, B. J .; Middlemiss,

D. N.; Routledge, C. Importance of 5-HT_1Breceptor selectivity

for 5-HT terminal autoreceptor activity: an in vivo microdialysis

study in the freely moving guinea pig. Neuropharmacology 1997,

36, 549-557.

(50) Roberts, C.; Belenguer, A.; Middlemiss, D. N.; Routledge, C.

Comparison of 5-HT autoreceptor control in guinea pig dorsal

and median raphe innervated brain regions. Br. J . Pharmacol.

1997, 331, 207P.

(51) Brown, A. M.; Young, T. J .; Patch, T. L.; Cheung, C. W.;

Kaumann, A. J .; Gaster, L. M.; King, F. D. [125I]-SB-207710, A

potent selective radioligand for 5-HT₄receptors. Br. J . Phar-

macol. 1993, 110, 10P.

(52) Michel, A. D.; Loury, D. N.; Whiting, R. L. Identification of a

single R1-adrenoceptor corrresponding to the R_1Asubtype in rat

submaximillary gland. Br. J . Pharmacol. 1989, 98, 883-889.

(53) Bylund, D. B.; Ray-Prenger, C.; Murphy, T. J . Alpha-2A and

alpha-2B adrenergic receptor subtypes: antagonist binding in

tissues and cell lines containing only one subtype. J . Pharmacol.

Exp. Ther. 1988, 245, 600-607.

(26) Blaney, F. E.; Tennant, M. Computational tools and results in

the construction of

G protein-coupled receptor models. In

Membrane protein models; Findlay, J ., Ed.; Bios. Scientific

Publishing: Oxford, U.K., 1996.

(27) Forbes, I. T.; Dabbs, S.; Duckworth, D. M.; Ham, P.; J ones, G.

E.; King, F. D.; Saunders: D. V.; Blaney, F. E.; Naylor, C. B.;

Baxter, G. S.; Blackburn, T. P.; Kennett, G. A.; Wood, M. D.

Synthesis, biological activity and molecular modeling studies of

selective 5-HT_2C/Breceptor antagonists. J . Med. Chem. 1996, 39,

4966-4977.

(28) The numbers refer to the corresponding residue in the human

5-HT_1Bsequence.

(29) Trumpp-Kallmeyer, S.; Hoklack, J .; Bruinvels, A.; Hibert, M.

Three-dimensional models of neurotransmitter G-binding protein-

coupled receptors. Mol. Pharmacol. 1991, 40, 8-15.

(30) Kristiansen, K.; Dahl, S. G. Molecular modeling of serotonin

ketanserin, ritanserin and their 5-HT_2Creceptor interactions.

Eur. J . Pharmacol. 1996, 306, 195-210.

SB-224289 Blocks Terminal 5-HT Autoreceptor Function

J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 8 1235

(54) Dini, S.; Caselli, G. F.; Ferrari, M. P.; Giani, R.; Clavenna, G.

Heterogeneity of [³H]-mepyramine binding sites in guinea pig

cerebellum and lung. Agents Actions 1991, 33, 181-184.

(55) Ruat, M.; Traiffort, E.; Bouthenet, M. L.; Schwartz, J . C.;

Hirschfeld, J .; Buschauer, A.; Schunack, W. Reversible and

irreversible labeling and autobiographic localization of the

cerebral histamine H₂receptor using [¹²⁵I]iodinated probes. Proc.

Natl. Acad. Sci. U.S.A. 1990, 87, 1658-1662.

(56) Dorje, F.; Wess, J .; Labrecht, G.; Tacke, R.; Mutschler, E.; Brann,

M. R. Antagonist binding profiles of five cloned human musca-

rinic receptor subtypes. J . Pharmacol. Exp. Ther. 1991, 256,

727-733.

(57) Roberts, C.; Watson, J . M.; Burton, M. J .; Price, G. W.; J ones,

B. J . Functional characterisation of the 5-HT terminal autore-

ceptor in guinea pig brain cortex. Br. J . Pharmacol. 1996, 117,

384-388.

(58) Draper, N. R.; Smith, H. Applied regression analysis; J ohn Wiley

and Sons: New York, 1966.

(59) Klapwyk, J . C.; Pallett, M. J .; Roberts, C. An anaesthetic mask

for use on guinea pigs (Cavia porcellus) in a stereotaxic frame.

Lab. Anim. Sci. 1995, 29, 407-410.

(60) Rapisadra, C.; Bacchelli, B. The brain of the guinea pig in

stereotaxic coordinates. Arch. Sci. Biol. 1977, 61, 1-37.

J M970457S

Article Doi

The selective 5-HT(1B) receptor inverse agonist 1-methyl-5-[[2'-methyl- 4'(5-methyl-1,2,4-oxadiazol-3-yl)biphenyl-4-yl]carbonyl]-2,3,6,7- tetrahydrospiro[furo[2,3-f]indole-3,4'-piperidine] (SB-224289) potently blocks terminal 5-HT autoreceptor function bo

DOI: 10.1021/jm970457s

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm970457s

Full text of DOI:10.1021/jm970457s

Products guided by the article

R&D Labs maybe for 179064-39-2

Relevant to this article

Hot Product