Advances toward new antidepressants with dual serotonin transporter and 5-HT1A receptor affinity within a class of 3-aminochroman derivatives. Part 2

6980

J. Med. Chem. 2008, 51, 6980–7004

Advances toward New Antidepressants with Dual Serotonin Transporter and 5-HT_1AReceptor

Afﬁnity within a Class of 3-Aminochroman Derivatives. Part 2

Nicole T. Hatzenbuhler,*^,‡Reinhardt Baudy,^‡Deborah A. Evrard,^‡Amedeo Failli,^‡Boyd L. Harrison,^‡Steven Lenicek,^‡

Richard E. Mewshaw,^†Annmarie Saab,^‡Uresh Shah,^‡Jean Sze,^‡Minsheng Zhang,^‡Dahui Zhou,^‡Michael Chlenov,^‡

Michael Kagan,^‡Jeannette Golembieski,^§Geoffrey Hornby,^§Margaret Lai,^§Deborah L. Smith,^§Kelly M. Sullivan,^§

Lee E. Schechter,^§and Terrance H. Andree^§

Chemical and Screening Sciences and DiscoVery Neuroscience, Wyeth Research, CN 8000, Princeton, New Jersey 08543, and Chemical and

Screening Sciences, Wyeth Research, 500 Arcola Road, CollegeVille, PennsylVania 19426

ReceiVed June 12, 2008

Novel compounds combining a 5-HT_1Amoiety (3-aminochroman scaffold) and a 5-HT transporter (indole

analogues) linked through a common basic nitrogen via an alkyl chain attached at the 1- or 3-position of the

indole were evaluated for dual afﬁnity at both the 5-HT reuptake site and the 5-HT_1Areceptor. Compounds

of most interest were found to have a 5-carbamoyl-8-ﬂuoro-3-amino-3,4-dihydro-2H-1-benzopyran linked

to a 3-alkylindole (straight chain), more speciﬁcally substituted with a 5-ﬂuoro ((R)-(-)-35c), 5-cyano ((-)-

52a), or 5,7-diﬂuoro ((-)-52g). Several factors contributed to 5-HT_1Aafﬁnity, serotonin rat transporter afﬁnity,

and functional antagonism in vitro. Although most of our analogues showed good to excellent afﬁnities at

both targets, speciﬁc features such as cyclobutyl substitution on the basic nitrogen and stereochemistry at

the 3-position of the chroman moiety seemed necessary for antagonism at the 5-HT_1Areceptor. Branched

linkers seemed to impart antagonism even as racemates; however, the potency of these analogues in the

functional assay was not desirable enough to further pursue these compounds.

Introduction

inhibition and a more pronounced increase in serotonergic

activity compared to acute treatment.⁶Antagonism of the 5-HT_1A

autoreceptor blocks the reduced rate of neuronal ﬁring, thus

presumably allowing the effects of the SSRI to be seen more

rapidly.⁷More recently, drugs targeting multiple sites associated

withdepressionhavebeendiscovered.^8,9Forexample,venlafaxine^10,11

and duloxetine¹²have been developed as serotonin and nore-

pinephrine reuptake inhibitors (SNRIs). Several studies have

shown that antidepressant effects of an SSRI can be accelerated

by the coadministration of a 5-HT_1Aantagonist.^13-15This effect

has not been seen by all investigators.^13,16It should be noted

that while pindolol is the only clinically available 5-HT_1A

antagonist, it is also a ꢀ-adrenoreceptor antagonist that has

5-HT_1Apartial agonist properties in rats.¹⁷Animal models have

shown that the 5-HT_1Aantagonist 1 (WAY-100635) can

potentiate the antidepressant effects of several SSRIs when given

in combination.^14,18There is currently an unﬁlled need for a

single agent with a dual mechanism of antidepressant action

(5-HT uptake inhibition plus 5-HT_1Aantagonism). Therefore,

the concept of combining 5-HT_1Aantagonism and selective

serotonin reuptake inhibition (SSRI) in one molecule to reduce

the latency period to efﬁcacy has been proposed for the

discovery of a new generation of antidepressant drugs. Several

reports have appeared in the literature by researchers trying to

create such a molecular entity.^19,20In our own laboratories, there

has been an ongoing effort to discover compounds with dual

5-HT_1Areceptor and serotonin transporter afﬁnity through

utilization of a common basic nitrogen linking the 5-HT_1Aand

SSRI moieties.²¹This generic strategy was previously termed

“the overlapping type approach”.²²

Depression can have devastating effects and can lead to

suicide. Affected individuals lack the energy and motivation to

perform everyday activities, which results in a poor quality of

life. Although depression is one of the most frequently diagnosed

psychiatric disorder, current therapies have several drawbacks.^1,2

The discovery and development of new antidepressants remain

an active area of research. Several undesired side effects have

been associated with the traditional tricyclic antidepressants

(TCAs^a),^3,4but the discovery and development of selective

serotonin reuptake inhibitors (SSRIs) have resulted in improved

drugs for the treatment of depression and related illnesses.⁴

SSRIs such as ﬂuoxetine, sertraline, paroxetine, and citalopram

are the most prescribed drugs since the 1980s. Although SSRIs

are devoid of anticholinergic and cardiovascular side effects,

they are generally effective in less than two-thirds of patients.

Additionally, like the TCAs, their delayed onset of action (2-6

weeks) in therapeutic efﬁcacy⁵is less than desirable in the

treatment of depression. This delay of action may be attributed

to the nonselective stimulation of serotonergic sites such as

somatodendritic 5-HT_1Aautoreceptors, which reduce cell-ﬁring

activity, thus limiting the amount of synaptic 5-HT in desired

brain regions. Eventually, following chronic antidepressant

treatment, desensitization of the somatodendritic 5-HT_1Aau-

toreceptors occurs, resulting in less autoregulatory feedback

* To whom correspondence should be addressed. Phone: 732-274-4047.

Fax: 732-274-4505. E-mail: hatzenn@wyeth.com.

‡

Chemical and Screening Sciences, Wyeth Research, PA.

Discovery Neuroscience, Wyeth Research, NJ.

Chemical and Screening Sciences, Wyeth Research, PA.

†

§

Several years ago, 5-methoxy-3-(di-n-propylamino)chroman

(5-OMe-DPAC, 2) was discovered as a selective 5-HT_1Aligand

vs other 5-HT and D₂sites in rat brain membranes.²³Subse-

quently, different 3-amino-3,4-dihydro-2H-1-benzopyran deriva-

tives were explored and developed as 5-HT_1Aagonists.²⁴The

discovery of robalzotan (3, NAD-299) as a potent 5-HT_1A

^aAbbreviations: TCAs, tricyclic antidepressants; SSRIs, selective sero-

tonin reuptake inhibitors; SNRIs, selective norepinephrine reuptake inhibi-

tors; 8-OH-DPAT, 8-hydroxydipropylaminotetralin; EDC, 1-(3-dimethyl-

aminopropyl)-3-ethylcarbodiimide hydrochloride; cPr, cyclopropyl; cBu,

cyclobutyl; cHexyl, cyclohexyl; MecPr, methylcyclopropyl; RA, reductive

amination; VCD, vibrational circular dichroism; SFC, supercritical ﬂuid

chromatography.

10.1021/jm8007097 CCC: $40.75

2008 American Chemical Society

Published on Web 10/04/2008

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6981

Figure 1. Structures of 3 and 3-aminochroman analogues A and B.

Figure 2. Design strategy and different 5-HT_1Ascaffolds.

Figure 3. Design strategy and different indole scaffolds.

receptor antagonist for the treatment of depression and anxiety

by other researchers²⁵(Figure 1) led us to the ﬁnding of novel

compounds that possess dual afﬁnity at both the 5-HT_1Areceptor

and the 5-HT reuptake site. We recently reported²⁶on the

discovery of derivatives that contain a 5-carbamoyl-8-ﬂuoro-

3-amino-3,4-dihydro-2H-1-benzopyran (or 3-amino-3,4-dihydro-

2H-1-benzopyran) and a straight chain aminoalkyl indole moiety

linked through a common basic nitrogen. Interestingly, com-

pound A (Figure 1) was shown to possess neurochemical activity

in vivo by producing acute and rapid increases in 5-HT in the

rat frontal cortex as measured by microdialysis.²⁶We also found

that similar analogues of structure B (Figure 1) showed dual

afﬁnities at both 5-HT_1Areceptor and serotonin transporter but

they lacked 5-HT_1Aantagonism in preliminary structure-activity

relationship (SAR) studies.

These results prompted us to investigate further the SAR

around the 5-HT_1Amoiety and the stereochemistry at the C-3

of the chroman ring of the dual acting molecule using the

3-aminoalkyl-5-ﬂuoro indole as the SSRI entity.

In this report, we look at several chroman scaffolds (sum-

marized in Figure 2) and discuss the synthesis and structure-

activity relationship (SAR) of these derivatives by looking at

Figure 4. Design strategy and substitution on linker.

the effects of substitution on the basic nitrogen (more speciﬁcally

R₁) Pr or cBu) and the nature of the substituents on the

aromatic ring of the chroman, on the binding afﬁnities at the

5-HT_1Areceptor and 5-HT reuptake site. We also elaborate on

what features of these novel derivatives might be responsible

for their agonist or antagonist properties at the 5-HT_1Areceptor.

In addition, we address the effect of varying the 5-HT reuptake

component of the dual-acting molecule on the binding afﬁnities

at 5-HT_1Aand transporter receptors using the 5-carboxamide

8-ﬂuoro 3-aminochroman as the 5-HT_1Amoiety. Several

6982 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

Scheme 1^a

i

^aReagents: (i) nitroethanol, Bu₂NH·HCl, isoamylacetate; (ii) NaBH₄, SiO₂, CHCl₃, PrOH; (iii) NH₂NH₂·H₂O, Raney Ni, EtOH; (iv) chiral HPLC; (v)

BnBr, K₂CO₃, DMF; (vi) NBS, CH₃CN; (vii) N-ﬂuorobenzenesulfonimide, n-BuLi/hexane, THF; (viii) ammonium formate, Pd/C, MeOH-THF; (ix) chiral

resolution with ^L-(+)-tartaric acid followed by base treatment.

Scheme 2^a

^aReagents: (i) trimethylorthoformate, H₂SO₄, MeOH; (ii) H₂SO₄, MeOH; (iii) propargyl bromide, K₂CO₃, acetone; (iv) N,N-diethylaniline; (v) NaOH,

EtOH-H₂O; (vi) CDI, THF followed by NH₃(g); (vii) 18-crown-6, KNO₂, I₂, THF-pyridine, sonication; or ethylene glycol, THF, NaNO₂, I₂; (viii) NaBH₄,

SiO₂, CHCl₃-ⁱPrOH; or NaBH₄, MeOH; (ix) NH₂NH₂·H₂O, Raney Ni, EtOH-THF; (x) chiral resolution with L-(+)-tartaric acid.

substituted indole scaffolds were investigated as summarized

in Figure 3. Finally, we look at substitution on the linker

tethering the 5-HT_1Aand SSRI moieties as illustrated in Figure

4.

with published data.²⁴Intermediate 21 was prepared from

racemic 19, and the enantiomers were separated by chiral

resolution with L-(+)-tartaric acid. The R/S stereochemistry was

assigned by matching their optical rotations with literature

data.²⁹

Chemistry

For compounds in series 6, 7, and 8, the required moieties

29a (3-amino-8-ﬂuorochromane-5-carboxamide), 29b (3-amino-

8-chlorochromane-5-carboxamide), and 29c (3-aminochroman-

5-carboxamide) were prepared as illustrated in Scheme 2. Their

syntheses were carried out as previously described^26,30,31using

commercially available 4-ﬂuoro-3-hydroxybenzoic acid 22,

2-chloro-5-triﬂuoromethylphenol 23, and methyl 3-hydroxy-

benzoate 24c (R₁) H) as starting materials, respectively. The

triﬂuoromethyl derivative 23 was converted to the methyl ester

Intermediates 19 and 21, needed for the synthesis of ﬁnal

targets in series 4 and 5, respectively, were synthesized as

racemates from the benzaldehyde 17 as previously described

(Scheme 1).^26,27The stereochemistry of the enantiomers of 19

had been determined earlier by other researchers through

resolution with L-(+)-tartaric acid.²⁸We separated these enan-

tiomers by chiral HPLC, and the R/S stereochemistry was

determined by comparing the optical rotation of each enantiomer

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6983

Scheme 3^a

Compound (R)-41 was prepared from (R)-33c through reductive

amination with propionaldehyde followed by triple Boc protec-

tion with tert-butyl dicarbonate. Treatment of (R)-41 with

methylamine in tetrahydrofuran yielded a mixture of two

compounds, the desired methylated carboxamide (R)-(-)-42

(30%) and a di-Boc derivative (R)-(-)-43 (62%). Compound

(R)-(-)-43 was reacted with iodomethane in the presence of

sodium hydride in THF to generate the di-Boc methylated

derivative (R)-44 in good yield. Finally upon treatment of (R)-

(-)-42 and (R)-44 with HCl in ethanol, a quantitative yield of

the desired N-methylated carboxamide derivative (R)-(-)-34f

was obtained. An alternative route B was also looked at where

(R)-33c was converted to the methyl ester under acidic condi-

tions followed by reductive amination with cyclobutanone to

generate (R)-38. The methyl ester was then converted to the

acid, which was coupled with methylamine in THF in the

presence of EDC and HOBt to generate the desired target (R)-

(-)-35f.

The synthesis of the 5-HT reuptake scaffolds of general structure

12 is shown in Scheme 7. They were all prepared using a Fischer

indole synthesis, followed by conversion of the generated alcohol

intermediates 45 or 47 to either the bromo derivatives 32a,c,d using

carbon tetrabromide and triphenylphosphine or the aldehydes

32b,e-i under modiﬁed Swern conditions. Only the 5-cyanoindole

alcohol 47 was prepared from the 5-bromo compound 46 by

treatment with copper(I) cyanide in DMF. Most alkylindoles were

converted to the aldehyde except for 5-methoxyindole, which turned

out to be unstable under the oxidation conditions, and the

5-cyanoindole. In these instances, the bromides 32c and 32d were

then prepared and alkylation reaction conditions were used for

synthesis of the ﬁnal targets.

^aReagents: (i) CH₃NH₂, EDC, HOBt, THF; (ii) 18-crown-6, KNO₂, I₂,

THF-pyridine, sonication; (iii) NaBH₄, SiO₂, CHCl₃-ⁱPrOH; (iv)

NH₂NH₂·H₂O, Raney Ni, EtOH-THF.

by heating with sulfuric acid at 100 °C for 1 h followed by

treatment with methanol to generate 24b (R₁) Cl). The chloro

intermediate 29b and the des-ﬂuoro analogue 29c were prepared

using a slightly modiﬁed protocol published earlier for the ﬂuoro

compound 29a.²⁶The enantiomers of 29a were separated by

chiral resolution with L-(+)-tartaric acid, and the stereochemistry

was conﬁrmed by NMR studies and VCD experiments. The

stereochemistry was further elucidated by X-ray crystallography

of one of the ﬁnal targets. The enantiomers of 29b and 29c

were not separated at this step, and the racemates were carried

through to the secondary amines 33d and 33e.

Several routes can be used for the synthesis of the substituted

carboxamide derivatives 9. For example, the N-methyl analogue

33f was prepared using intermediate 31 where the chroman

moiety is already N-methylated as shown in Scheme 3. The

acid 26a was reacted with methylamine in the presence of EDC

and HOBt to generate a quantitative yield of 30, which was

then converted to the amine 31 using chemistry described in

Scheme 2.

The synthesis of scaffold 13 was carried out in one step by

alkylation of the appropriately substituted indole with 1,4-

dibromobutane to generate the desired intermediate 48 as

illustrated in Scheme 8.

Scaffold 14 was prepared as summarized in Scheme 9

according to a literature procedure.³³Commercially available

(4-ﬂuorophenyl)hydrazine hydrochloride and 4-oxocyclohex-

anecarboxylic acid ethyl ester were combined in the presence

of ethanol under reﬂux to yield 49. Intermediate 49 was reduced

with lithium aluminum hydride followed by Dess-Martin

oxidation to generate the aldehyde 50.

Final targets 52a-g were synthesized using route 1 as

illustrated in Scheme 10. After the 5-carboxamide 8-ﬂuoro-3-

aminochroman template 29a was coupled with the alkylindole

moieties (32a-i) through alkylation or reductive amination to

generate the secondary amines 51a-g, the ﬁnal products were

obtained by reductive amination with cyclobutanone or an

aldehyde. Only the enantiomers of 51g and the ﬁnal target 52a

were separated by chiral HPLC. However, their stereochemistry

was not determined and these compounds are identiﬁed as (+)

and (-) based on their optical rotation. Novel compounds

54a-d, 55a-d, and 56a-d linked at the indole nitrogen were

prepared as racemates under alkylation conditions, generating

the secondary amines 53a-d as shown in Scheme 10 (route

2). Introduction of alkyl groups on the basic nitrogen was carried

out under reductive amination conditions. Finally, targets 57-59

were synthesized by coupling (R)-(+)-29a with intermediate

50 under reductive amination conditions followed by introduc-

tion of different R₁groups on the basic nitrogen as illustrated

in Scheme 10 (route 3) to generate one pair of enantiomers.

Target 60 was synthesized using racemic 29a to provide two

diastereomeric pair of enantiomers of undetermined stereo-

chemistry.

Final targets 34a-c,e,f and 35a,c-f were synthesized using

the route illustrated in Scheme 4. Following coupling of the

chroman amine template with the alkylindole moiety 32a or

32b (prepared using chemistry already published²⁶) through

alkylation or reductive amination, the desired targets were

obtained by reductive amination of the secondary amines 33a-f

with the desired aldehyde or cyclobutanone in the presence of

sodium cyanoborohydride and acetic acid in methanol. The

enantiomers of intermediates 33b-e were separated by chiral

HPLC, but only the stereochemistry of 33c was further

conﬁrmed as R/S. The absolute conﬁguration of all other

intermediates was not determined, and it is identiﬁed as (+)

and (-). The secondary amine intermediates 33a and 33f were

not chirally separated at this step.

Most of the ﬁnal targets with mono- or bis-substitution of

the carboxamide functionality (series 9, compounds 35f and

40a-h) were prepared using a more convergent synthetic

pathway amenable to combinatorial chemistry and summarized

in Scheme 5. Intermediate 36 was prepared from derivative 25a

using the chemistry elaborated earlier in Scheme 2. Coupling

of 36 with indole aldehyde 32a under reductive amination

conditions generated desired product 37, which was then reacted

with cyclobutanone to yield compound 38 (series 11). The ester

38 was converted to the acid derivative 39 (series 10), which

was then subjected to standard coupling conditions with desired

amine (EDC and HOBt) to generate the substituted carboxamide

targets 35f and 40a-h. Since this synthetic route generated both

methyl ester 38 and carboxylic acid 39, these were also tested

in the in vitro biological assays. A different route³²was

investigated for the synthesis of substituted carboxamide deriva-

tives (more speciﬁcally, the (R)-N-Me enantiomer) from the

primary carboxamide as summarized in Scheme 6 (route A).

6984 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

Scheme 4^a

^aReagents: (i) 32b, TEA, DMSO; (ii) 32a, NaBH₃CN, HOAc, MeOH; or NaBH(OAc)₃, HOAc, 1,2-dichloroethane; (iii) aldehyde or cyclobutanone,

NaBH₃CN, HOAc, MeOH.

Scheme 5^a

^aReagents: (i) 18-crown-6, KNO₂, I₂, THF-pyridine, sonication; (ii) NaBH₄, SiO₂, CHCl₃-ⁱPrOH; (iii) NH₂NH₂·H₂O, Raney Ni, EtOH-THF; (iv)

NaBH₃CN, HOAc, MeOH; (v) cyclobutanone, NaBH₃CN, HOAc, MeOH; (vi) NaOH, EtOH-H₂O; (vii) R₄R₅NH, EDC, HOBt, THF.

The R-methylalkylindole scaffold 15 (compound 61) was

prepared in one step as summarized in Scheme 11 using

5-ﬂuoroindole as starting material. The ꢀ-methylindole deriva-

tive 16 (compound 65) was synthesized using the racemic route

illustrated in Scheme 12. 5-Fluorogramine 62 prepared from

5-ﬂuoroindole using a well-known procedure³⁴was reacted with

diethyl methyl malonate in the presence of tributylphosphine

to generate intermediate 63, which upon treatment with base

followed by decarboxylation yielded derivative 64.³⁵Finally,

reduction with lithium aluminum hydride and oxidation of the

resulting alcohol to the aldehyde generated starting material 65.

tion of 68a,b. In the synthesis of target 70, (R)-(+)-29a was

used. The stereochemistry at the second chiral center was not

determined.

Discussion

1. Effect of Chroman Moiety for N-Propyl Derivatives.

We recently published some SAR on the effect of substitution

on the basic nitrogen in the methoxy and carboxamide series.²⁶

As observed earlier,²⁶a secondary amine is far from optimal in

terms of afﬁnity for the 5-HT_1Areceptor. Propyl, methylcyclo-

propyl, and cyclobutyl were favored substituents for desired dual

activity at both 5-HT_1Areceptor and 5-HT reuptake site. Further

studies were carried out investigating the 5-HT_1Amoiety in the

N-propyl series, and the in vitro biological data are summarized

in Table 1. Using racemic compounds, we found that the

5-carboxamide-8-ﬂuoro derivative (34c), the 5-N-methylcar-

boxamide-8-ﬂuoro analogue (34f), and the 5-carboxamide-8-

des-ﬂuoro analogue (34e) have the best afﬁnities for the 5-HT_1A

receptor. Furthermore, derivatives 34c and 34f had excellent

afﬁnities at the 5-HT reuptake site while the des-ﬂuoro analogue

Targets 67, 68, and 70 were synthesized as shown in Scheme

13 under reductive amination conditions. The secondary amines

66 and 69 were then converted to tertiary amines by reaction

with the appropriately substituted alkylaldehyde or ketone. The

diastereomers of the secondary amine 66 were separated by

HPLC, followed by SFC separation of the enantiomers. On the

basis of in vitro biological data for compounds 67a-d (see

Table 7), only two enantiomers of 66 were chosen for prepara-

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6985

Scheme 6^a

^aReagents: (i) propionaldehyde, NaBH₃CN, AcOH, MeOH; (ii) (Boc)₂O, DMAP, CH₂Cl₂; (iii) MeNH₂, THF; (iv) HCl/EtOH; (v) CH₃I, NaH, THF; (vi)

H₂SO₄, MeOH; (vii) cyclobutanone, NaBH₃CN, AcOH, MeOH; (viii) LiOH/H₂O, THF-MeOH; (ix) MeNH₂/THF, EDC, HOBt.

Scheme 7^a

^aReagents: (i) acid, dioxane, reﬂux; (ii) CBr₄, PPh₃, CH₂Cl₂; (iii) TFA, pyridine, DMSO-benzene, DCC; (iv) Cu(I)CN, DMF.

Scheme 8^a

afﬁnities at the 5-HT_1Areceptor and the 5-HT reuptake site

than their 5-methoxy counterparts.

The 5-carboxamide-8-ﬂuoro (34c) and 5-methoxy-8-ﬂuoro

(34b) chromans were directly compared with respect to intrinsic

functional activity at the 5-HT_1Areceptor in a GTPγS assay.

^aReagents: (i) 1,4-dibromobutane, NaH, DMF.

The racemates of 34c and 34b were separated into their

enantiomers (R)-(-)-34c, (S)-(+)-34c and (R)-(-)-34b, (S)-(+)-

34b, respectively. Interestingly, we found that (R)-(-)-34c was

a 5-HT_1Aantagonist in the assay while (S)-(+)-34c was a partial

agonist with excellent afﬁnities at both 5-HT_1Areceptor and

5-HT reuptake site. In contrast, the enantiomers of 34b ((R)-

(-)-34b and (S)-(+)-34b) were both partial agonists at the

34e had a slightly weaker afﬁnity. The 5-methoxy (34a) and

5-methoxy-8-ﬂuoro (34b) analogues had moderate afﬁnity

at both the 5-HT_1Areceptor and SSRI site. In general, the

5-carboxamide derivatives (8-F or 8-des-F) have better

6986 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

Scheme 9^a

^aReagents: (i) EtOH, reﬂux; (ii) LiAlH₄, THF; (iii) Dess-Martin periodinane, CH₃CN.

Scheme 10^a

^aReagents: (i) 32a,c,d, TEA, DMSO; (ii) 32b,e-i, NaBH₃CN, HOAc, MeOH; or NaBH(OAc)₃, HOAc, 1,2-dichloroethane; (iii) aldehyde or cyclobutanone,

NaBH₃CN, HOAc, MeOH; (iv) DMSO, Hu¨nig’s base, 85 °C; (v) 29a or (R)-(+)-29a, NaBH₃CN, HOAc, MeOH.

5-HT_1Areceptor with moderate SSRI and 5-HT_1Aafﬁnities.

Introduction of a methyl on the carboxamide functionality

generated a racemate (34f) with partial agonist activity in the

GTPγS assay (48% E_max), while the R-enantiomer ((R)-(-)-

34f) was an antagonist at the 5-HT_1Areceptor (0% E_max).

Although it is known that the stereochemistry at the 3-position

of the chroman ring plays a role in the agonist/antagonist

properties of these dual acting molecules, the above results

suggest that the carboxamide (or substituted carboxamide) may

also have some inﬂuence on this intrinsic functional activity at

the 5-HT_1Areceptor (cf. (R)-(-)-34c, (S)-(+)-34c, and (R)-(-)-

34f vs (R)-(-)-34b, (S)-(+)-34b).

2. Effect of Chroman Moiety for N-Cyclobutyl Derivatives.

Similar to the N-propyl series, the 5-carboxamide-8-ﬂuoro (35c)

and 8-chloro (35d) derivatives, the 5-N-methylcarboxamide-8-

ﬂuoro analogue (35f), and the 5-carboxamide-8-des-ﬂuoro

compound (35e) all had excellent to good afﬁnities at both the

5-HT_1Areceptor and the 5-HT reuptake site as summarized in

Table 2. Additionally, the 5-methoxy derivative 35a had

relatively good afﬁnity at both the 5-HT_1Areceptor and the

5-HT reuptake site, although slightly weaker than the

carboxamide derivatives. Interestingly, as racemates, most

of the cyclobutyl substituted compounds were either antago-

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6987

nists or partial agonists at 5-HT_1Ain the GTPγS assay both

in 5-carboxamide and in 5-methoxy series. Furthermore, when

the enantiomers of compounds 35a, 35c, 35d, 35e, and 35f

were separated, (R)-(-)-35a, (R)-(-)-35c, (-)-35d, and (R)-

(-)-35f were 5-HT_1Aantagonists except for the 5-carboxa-

mide-8-des-ﬂuorochroman analogue (-)-35e, where partial

agonism was observed. Even more interesting was the 5-HT_1A

antagonism of the 5-methoxy-8-des-ﬂuoro derivative (R)-(-)-

35a, although with a drop-off in potency at the transporter

(K_iof 41 nM vs 13 nM for racemate). In comparison with

the data observed for the N-propyl derivatives (see Table 1),

these results suggest that the cyclobutyl seems to play a fairly

important role in the antagonistic properties of these mol-

ecules along with the stereochemistry at the 3-position of

the chroman. In general, the other enantiomers (data not

shown) were found to have decreased afﬁnity along with

partial agonism at the 5-HT_1Areceptor. We also found that

the 5-carboxamide-8-des-ﬂuoro derivative 35e (partial agonist

as a racemate), once separated into its enantiomers, resulted

in two partial agonists, (-)-35e and (+)-35e. Moreover, one

of the enantiomers of the 5-carboxamide-8-chloro analogue

((-)-35d) was relatively potent at both receptors and an

antagonist. This suggests that substitution at the 8-position

might also be important for agonism/antagonism of these

molecules at the 5-HT_1Areceptor.

4. Effect of Substitution on Indole Moiety. All the SAR

studies were so far focused on the use of a 3-alkyl-5-ﬂuoroindole

as the SSRI moiety. Cyclobutyl substitution on the basic nitrogen

was also favored, especially for antagonism at the 5-HT_1A

receptor. We then decided to investigate further the SAR around

the indole functionality as summarized in Table 4. The replace-

ment of 5-ﬂuoro (35c) by 5-cyano ((-)-52a), 5-methoxy (52b),

and 5-chloro (52c) did not affect the binding afﬁnity at the

5-HT_1Areceptor. Although the afﬁnity at the 5-HT reuptake

site was very good for the 5-cyano derivative (-)-52a, a 25-

fold drop in afﬁnity was observed for the 5-chloro analogue

52c while the 5-methoxy compound 52b was rather weak. Both

derivatives 52b and 52c showed weaker functional activity at

both 5-HT_1Aand 5-HT reuptake sites. Furthermore, as race-

mates, the 5-methoxy analogue (52b) was found to be a partial

agonist at 5-HT_1Ain GTPγS assay and the 5-chloro compound

(52c) was found to be an antagonist. Interestingly, both

enantiomers of the 5-cyano analogue ((-)-52a and (+)-52a)

were antagonists at the 5-HT_1Areceptor, the best one being (-)-

52a although its IC₅₀(195 nM) is weaker than its 5-ﬂuoro

counterpart (R)-(-)-35c (48 nM). The 6-ﬂuoroindole analogue

52d did show an 80-fold decrease in afﬁnity at the 5-HT

reuptake site (K_iof 237 nM vs 3 nM for 35c) without any

signiﬁcant effect on the binding afﬁnity at the 5-HT_1Areceptor.

Methoxy (52e) and chloro (52f) substitution at the 7-position

of the indole generated compounds with relatively good binding

afﬁnity at the 5-HT_1Areceptor. However, the 7-methoxy

derivative 52e lost signiﬁcant afﬁnity at the 5-HT reuptake site

with the 7-chloro (52f) showing a 20-fold decrease in afﬁnity

(K_iof 66 nM vs 3 nM for 35c). We decided to investigate a

5,7-diﬂuoroindole as an SSRI moiety in order to block a

potential site of metabolism of (R)-(-)-(35c). As shown in Table

4, the 5,7-diﬂuoro analogue (-)-52g showed comparable

afﬁnities to the 5-ﬂuoroindole derivative (R)-(-)-35c at both

5-HT_1Aand 5-HT reuptake sites. However, no improvement in

microsomal metabolic stability was observed. Regardless of the

substitution on the indole moiety, most of these compounds were

antagonists at the 5-HT_1Areceptor as racemates, suggesting that

agonism or antagonism is mainly affected by the changes on

the 5-HT_1Amoiety or substitution at the basic nitrogen, as

discussed earlier.

5. Chroman-Indole Nitrogen Linker. One potential site

of metabolism for all compounds discussed is the indole

nitrogen, thus leading us to investigate derivatives where the

chroman is tethered to this nitrogen through an alkyl chain.

Several such analogues were prepared with ﬂuoro substitution

at the 4-, 5-, 6-, and 7-position of the indole as summarized

in Table 5. Furthermore, substitution on the basic nitrogen

was also looked at for each analogue. In terms of afﬁnity at

the 5-HT_1Areceptor, propyl (54a-c), methylcyclopropyl

(55a-c), and cyclobutyl (56a-c) were all well tolerated. In

all cases, the 7-ﬂuoroanalogues (54d, 55d, and 56d) showed

slightly weaker afﬁnity at the 5-HT_1Areceptor. Although all

compounds (54a-d, 55a-d, 56a-d) had good to excellent

afﬁnity at the 5-HT_1Areceptor (K_i) 3.25-39.95 nM), the

position of the ﬂuoro on the indole moiety affected the

binding afﬁnity at the 5-HT reuptake site. The 7-ﬂuoro

analogues had the best binding afﬁnities at the 5-HT reuptake

site except for analogue 55c, which with methylcyclopropyl

on the basic nitrogen and 6-ﬂuoro substitution on the indole

had the best proﬁle. However, derivative 55c lacks intrinsic

functional activity at the transporter site. When a propyl group

is present on the basic nitrogen, partial agonism was observed

at the 5-HT_1Areceptor for 54c and 54d racemates. For the

The substitution of a methyl ester (38) or carboxylic acid

(39) for a primary carboxamide resulted in a signiﬁcant decrease

in afﬁnity at both 5-HT_1Areceptor and 5-HT reuptake site for

these molecules (see Table 2). To further understand this SAR,

we decided to investigate the effect of substitution on the

carboxamide functionality.

3. Substitution on the Carboxamide Functionality. The

structure-activity relationship studies carried out so far have

shown that the most preferred compound is a 5-carboxamide-

8-ﬂuoro substituted chroman, with a cyclobutyl substitution

on the basic nitrogen and a 3-alkyl-5-ﬂuoroindole as the SSRI

moiety (compound 35c). The in vitro biological results

summarized in Table 3 show the effect of substitution on

the carboxamide functionality. The substituent on the second-

ary amide has minimal effect on the SSRI afﬁnity. However,

the various substituents investigated did have an effect on

the binding afﬁnity at the 5-HT_1Areceptor. Methyl (35f), ethyl

(40a), and cyclopropyl (40d) groups were well tolerated with

5-HT_1Aafﬁnities similar to the primary carboxamide. How-

ever, their potency in the GTPγS was weaker than that of

the unsubstituted carboxamide. An isopropyl (40c) substituent

resulted in a 4-fold drop in afﬁnity at the 5-HT_1Areceptor

compared to 35c. Finally, all other groups investigated

(propyl, cyclobutyl, cyclohexyl, and methylcyclopropyl,

40b,e-g) generated compounds with lower binding afﬁnity

at the 5-HT_1Areceptor. Moreover, disubstitution (e.g.,

dimethyl, 40h) also resulted in lower binding afﬁnity at the

5-HT_1Areceptor with moderate afﬁnity at the 5-HT reuptake

site. Having a secondary or tertiary carboxamide did not

affect the intrinsic functional activity of these racemic

compounds resulting in antagonism at the 5-HT_1Areceptor.

However, their potency was decreased as shown by their IC₅₀

values in the GTPγS assay. Therefore, a primary carboxa-

mide, secondary carboxamide (with relatively small substit-

uents), or methoxy functionalities are necessary for good

5-HT_1Aafﬁnity.

In keeping with the SAR of this chroman series, all these

compounds demonstrated acceptable binding selectivity over the

R₁receptor.

6988 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

Scheme 11^a

compound (R)-(-)-35c. Interestingly, only one diastereomer

(67a and 68a) shows excellent binding afﬁnity at both receptors,

with the methylcyclopropyl derivative 68a having the best

proﬁle. This compound possesses the R-stereochemistry at the

C-3 of the chroman amine, since it was synthesized from (R)-

(+)-29a, but the stereochemistry at the R-methyl position has

not been determined. The introduction of a ꢀ-methyl group on

the chain generated targets 70a and 70b, with 70a having

excellent binding afﬁnity at the 5-HT_1Areceptor and reasonable

afﬁnity at the 5-HT reuptake site. However, it is a little weaker

in intrinsic functional activity at the 5-HT_1Areceptor. The

substitution on the linker did not impart greater microsomal

metabolic stability to these derivatives. Antagonism at the

5-HT_1Areceptor was observed for most of the compounds

regardless of the substitution on the basic nitrogen. Unlike the

tetrahydrocarbazole derivatives, which are analogues of the

ꢀ-methyl derivatives, these compounds are showing better

afﬁnities at the 5-HT_1Areceptor and SSRI site (cf. 67a vs 58b

and 68a vs 59b).

^aReagents: (i) BF₃·Et₂O/EtOH, CH₃NO₂, -20 °C.

Scheme 12^a

All compounds generated in the course of investigating the

various SSRI moieties demonstrated acceptable binding selec-

tivity over the R₁receptor.

^aReagents: (i) Eschenmoser’s salt, HOAc, CH₃CN; (ii) diethylmethyl-

malonate, P(Bu)₃, CH₃CN, reﬂux; (iii) NaOH/EtOH; (iv) bromobenzene,

reﬂux; (v) LiAlH₄, THF; (vi) TFA, pyridine, DMSO-toluene, DCC.

Conclusion

The combination of a 5-HT_1Ascaffold and a transporter

moiety (indole analogue) sharing a common basic nitrogen

resulted in compounds with desirable afﬁnities at the sero-

tonin reuptake site and 5-HT_1Areceptor. A detailed

structure-activity relationship investigation of this series has

shown that several factors inﬂuence the binding afﬁnity and

functional agonism or antagonism of these compounds. First,

the 5-HT_1Aafﬁnity seems to be dependent on (1) the

substitution on the basic nitrogen with cyclobutyl, propyl,

and methylcyclopropyl as favored substituents, (2) the

substitution on the carboxamide functionality with unsub-

stituted or nonbulky substitution (methyl or ethyl) preferred,

(3) a carboxamide at the 5-position of the chroman moiety

favored over a methoxy, carboxylic acid, or methyl ester,

and (4) the SSRI moiety with straight chain alkyl linked to

the 3-position or nitrogen of the indole as preferred over

substitution on the alkyl linker or a more constrained

alkylindole such as a tetrahydrocarbazole. Interestingly, in

the tetrahydrocarbazole series, only one of the diastereomers

was favored in terms of afﬁnity at the 5-HT_1Areceptor.

Second, the SSRI afﬁnity was mostly inﬂuenced by (1) the

substitution on the basic nitrogen with propyl, cyclobutyl,

and methylcyclopropyl as favored substituents, (2) the

presence of a methoxy, primary carboxamide, or secondary

carboxamides (such as methyl or ethyl) at the 5-position of

the aromatic portion of the chroman, and (3) the substitution

on the indole with ﬂuoro or cyano at the 5-position or 5,7-

diﬂuoro as favored for the straight chain alkyl derivatives.

The ﬂuoro at the 6- or 7-position of the indole was preferred

in N-linked analogues. Any substitution on the alkyl chain

did lower the afﬁnity for the 5-HT reuptake site. Third, the

agonist or antagonist properties of these dual-acting com-

pounds seem to be mostly inﬂuenced by the stereochemistry

at the 3-position of the chroman moiety with one enantiomer

(R for compounds where stereochemistry is known) usually

preferred. Another critical observation seems to be the

presence of a cyclobutyl on the basic nitrogen for the straight

chain alkylindoles or N-linked derivatives. Even as racemates,

many of these compounds are antagonists at the 5-HT_1A

receptor. However, antagonism was not an issue when

methylcyclopropyl (55a and 55c) and cyclobutyl (56b-d)

derivatives, antagonism was shown by most of the com-

pounds. However, all these targets were characterized by

weaker IC₅₀values in the GTPγS assay, notwithstanding their

good binding afﬁnity at both receptors.

6. Tetrahydrocarbazole as Transporter Moiety. Earlier

work in our laboratories had shown that a tetrahydrocarbazole

imparts binding afﬁnity at the 5-HT reuptake site when linked

to different ligands (unpublished results). We can view this

tricyclic moiety as a constrained propylindole. Although it

introduces another chiral center, it may also impart more

microsomal metabolic stability to the targets. We decided to

look at its effect when linked through the basic nitrogen of the

5-carboxamide-8-ﬂuoro-3-aminochroman scaffold on dual af-

ﬁnity at both receptors. The in vitro biological data are

summarized in Table 6. The binding afﬁnity at the 5-HT_1A

receptor was very much dependent on the stereochemistry at

both chiral centers. One diastereomer is preferred over all others

for each substituent on the basic nitrogen: propyl (58b),

methylcyclopropyl (59b), and cyclobutyl (60b). All the dia-

stereomers were separated by HPLC, and we know that the

R-stereochemistry at the C-3 position of the chroman is present

in compounds 57a,b, 58a,b, and 59a,b, since they were

synthesized from (R)-(+)-29a. Regardless of the substitution

on the basic nitrogen, all these compounds are antagonists at

the 5-HT_1Areceptor. In the case of cyclobutyl substitution, the

compound with the best binding afﬁnity at 5-HT_1A(60b) did

show some partial agonism in the GTPγS assay. Unlike the

5-ﬂuoroindole analogues, the best afﬁnity at the 5-HT reuptake

site was obtained with the unsubstituted derivatives (R₁) H).

The intrinsic functional activity at the transporter was weaker

for all the compounds, and a slight increase in metabolic stability

in rat microsomes was observed. In summary, this novel series

of compounds did show good dual activity at both receptors

but did not have the potency needed for further evaluation.

7. Substitution on Linker. R-Substitution to the basic

nitrogen was investigated in order to potentially slow down

dealkylation as one of the metabolic pathway for novel

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6989

Scheme 13^a

^aReagents: (i) NaBH(OAc)₃, HOAc, 1,2-dichloroethane or NaBH₃CN, HOAc, MeOH; (ii) aldehyde or cyclobutanone, NaBH₃CN, HOAc, MeOH.

Table 1. N-Propyl Analogues with Different Chroman Moieties^a

5-HT transporter

5-HT_1Areceptor

afﬁnity^b

K_i(SEM), nM

function^c

IC₅₀(SEM), nM

afﬁnity^d

K_i(SEM), nM

GTPγS^e

E_max, % (IC₅₀, nM)

R₁receptor^g

K_i, nM

compd

34a^h

34b

X

Y

OMe

CONH₂

CONHMe

CONH₂

H

F

H

9.0 (3.0)

12.5 (2.5)

17.5 (1.5)

13.5 (5.5)

7.1 (1.7)

8.0 (0.9)

4.7 (0.7)

2.40 (0.5)

11.3 (4.9)

23.4 (2.7)

3.9 (0.35)

386 (153)

442 (9.5)

379

452 (61)

128 (44)

181 (22)

59.9 (14.3)

24.3 (11.3)

47 (1.6)

15.8 (1.5)

19.4 (1.06)

10.9 (4.4)

12.6 (0.51)

3.5 (0.9)

1.5 (0.1)

3.2 (1.0)

0.81 (0.05)

2.64 (0.27)

0.93 (0.31)

100

79.5

88

94

78

0 (405)

60

48 (617)

0 (752)

74

820

1059

50%

345

(R)-(-)-34b

(S)-(+)-34b

34c^h

nd

(R)-(-)-34c^h

(S)-(+)-34c^h

34f

1150

46%

476

39%

31%

(R)-(-)-34f

34e

ﬂuoxetine

1

8-OH-DPAT

146 (87)

39.4 (3.1)

0.96 (0.21)

3.0 (0.1)

0 (4.04)

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function were performed in triplicate, determined from eight concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}Data for these

compounds were published in a previous paper.²⁶

relative to DMSO-d₆at 2.49 ppm or MeOH-d₄at 3.31 ppm as an

internal standard. Mass spectra were recorded on a Micromass LCT

spectrometer. CHN combustion analyses were determined on a

Perkin-Elmer 2400 analyzer or were performed by Robertson

Microlit (Madison, NJ). All analyzed compounds are within (0.4%

of the theoretical value unless otherwise indicated. Optical rotations

were measured using a Jasco P-1020 polarimeter. Solvents and

reagents were used as purchased. All ﬁnal targets were converted

to the HCl salt by dissolution in ethyl acetate and addition of 1 M

HCl/Et₂O, followed by ﬁltration unless otherwise indicated. Chiral

HPLC screens were carried out on two Agilent 1100 series HPLC

equipped with quaternary pumps (model G13411A), autosamplers

(model G1367A), diode-array detectors (model G1315B) from

Agilent Technologies (CA), and LC Spiderling Deluxe automated

column selection systems (Chiralizer Services, Newton, PA).

Preparative supercritical ﬂuid chromatography (SFC) separations

were carried out on a Berger Multigram III SFC equipped with

two SD-1 Varian pumps, a Knauer K-2501 spectrophotometer, 6-ton

substitution was introduced on the alkyl chain. This holds

true especially for the tetrahydrocarbazole compounds re-

gardless of the substitution on the basic nitrogen as long as

R-stereochemistry is present at the 3-amino position of the

chroman. These substituted derivatives did lack potency in

the functional assay. In summary, a combination of several

factors seems to contribute to the agonist or antagonist

properties of these derivatives. From this SAR study,

compound (R)-(-)-35c was found to have the best proﬁle,

and a more complete in vivo proﬁling will be the subject of

future publications.

Experimental Section

Chemistry. Melting points were determined on a MEL-TEMP

apparatus and are uncorrected. ¹H NMR spectra were recorded on

a Varian Unity Plus 400 spectrometer or a Unity INOVA Varian

500 MHz spectrometer. Chemical shifts δ are reported in ppm

6990 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Table 2. N-Cyclobutyl Analogues with Different Chroman Moieties^a

Hatzenbuhler et al.

5-HT transporter

5-HT_1Areceptor

afﬁnity^d

afﬁnity^b

function^c

GTPγS^eE_max

,

R₁receptor^g

K_i, nM

compd

X

Y

K_i(SEM), nM

IC₅₀(SEM), nM

K_i(SEM), nM

% (IC₅₀, nM)

35a^h

OMe

H

F

13 (4.0)

41.4

400 (169)

732

15.6 (0.2)

2.14 (1.1)

9.3 (2.1)

1.2 (0.35)

6.21 (1.06)

10.9 (0)

1.75 (0.43)

0.66 (0.09)

29.6

5.35 (0.17)

1.32 (0.09)

73.8 (3.8)

4%

8.5 (>2000)

0 (488)

0 (202)

0 (48)

0 (>1000)

0 (125)

28 (184.5)

30 (66)

71

15%

16%

21%

567

22%

500

10%

6%

27%

0%

6000

8%

(R)-(-)-35a^h

35c^h

CONH₂

CONHMe

COOMe

COOH

3.0 (0.5)

1.5 (0.04)

3.74 (0.6)

1.98 (1.1)

9.55 (0.4)

5.9 (2.4)

7.2

8.3 (1.9)

10.2 (5.7)

161

25.2 (8.3)

18.7 (1.8)

56.5 (7.1)

61.4 (19)

114 (61)

109

28.3

15 (9.0)

147

(R)-(-)-35c^h

35d

(-)-35d

35e

(-)-35e

(+)-35e

35f

(R)-(-)-35f

38

F

Cl

H

F

0 (812)

0 (262)

nd

1860

1026

39.4 (3.1)

39

ﬂuoxetine

1

498.5

3.9 (0.35)

0 (>2000)

0.96 (0.21)

3.0 (0.1)

0 (4.04)

8-OH-DPAT

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function were performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}Data for these

compounds were published in a previous paper.²⁶

Table 3. Substituted Carboxamide Chroman Analogues^a

5-HT transporter

5-HT_1Areceptor

afﬁnity^d

afﬁnity^b

function^c

GTPγS^eE_max

,

R₁receptor^g

K_i, nM

compd

35c^h

35f

40a

40b

40c

40d

40e

40f

40g

40h

ﬂuoxetine

1

8-OH-DPAT

R₄

R₅

K_i(SEM), nM

IC₅₀(SEM), nM

K_i(SEM), nM

% (IC₅₀, nM)

H

Me

3.0 (0.5)

8.3 (1.9)

9.47 (2.9)

7.96 (3.1)

11.28 (3.8)

8.05 (1.8)

7.26 (2.6)

19.8

25.2 (8.3)

15 (9.0)

27.9

32.7

44.7

69.6

70.6

144

43.6

9.3 (2.1)

0 (202)

0 (812)

0 (>1000)

0 (800)

0 (>2000)

0 (395)

0 (91.5)

0 (>1000)

0 (>2000)

16%

27%

12%

21%

4%

12%

21%

31%

55.5

Me

5.35 (0.17)

17.8 (2.09)

196 (1.06)

46.1 (7.5)

19.2 (1.66)

213.5 (26.5)

44%

Et

Pr

ⁱPr

cPr

cBu

cHexyl

MecPr

Me

6.91 (1.2)

28.8

3.9 (0.35)

160.5 (11.6)

372 (8.49)

218

39.4 (3.1)

0.96 (0.21)

3.0 (0.1)

0 (4.04)

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function were performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}Data for this

compound were published in a previous paper.²⁶

bulk CO₂tank with built-in chiller and heater, and G700 compressor

(Mettler-Toledo, Newark, DE). Columns and conditions used for

HPLC were as speciﬁed for each compound. Analytical thin-layer

chromatography (TLC) was carried out on precoated plates (silica

AdVances toward New Antidepressants

Table 4. Substituted Indole Analogues^a

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6991

5-HT transporter

5-HT_1Areceptor

afﬁnity^b

K_i(SEM), nM

function^c

IC₅₀(SEM), nM

afﬁnity^d

K_i(SEM), nM

GTPγS^e

E_max, % (IC₅₀, nM)

R₁receptor^g

K_i, nM

compd

R₇

R₈

35c^h

F

CN

OMe

Cl

H

F

H

3.0 (0.5)

1.5 (0.04)

5.3 (0.3)

2.44

9.05

>1000

75.5

237.5

>1000

66

3.34 (0.19)

12.15

3.9 (0.35)

25.2 (8.3)

18.7 (1.8)

17.4 (10)

24.1 (1.55)

10.1 (0.43)

834

612 (18.5)

899

>1000

9.3 (2.1)

1.2 (0.35)

253 (31)

0 (202)

0 (48)

0 (>1000)

0 (195)

0 (>1000)

84 (419)

0 (735)

0 (512)

0 (557)

16%

21%

nd

19%

27%

659

20%

30%

15%

29%

788

(R)-(-)-35c^h

(S)-(+)-35c^h

(-)-52a

(+)-52a

52b

1.98 (0.78)

258 (21.9)

1.19 (0.15)

3.98 (0.8)

5.02 (2.43)

3.22 (0.39)

26.1 (1.66)

1.73 (0.49)

68.9 (0.71)

52c

52d

52e

52f

(-)-52g

(+)-52g

ﬂuoxetine

1

6-F

7-OMe

7-Cl

7-F

0 (>1000)

0 (206)

0 (>1000)

71.6 (2.8)

225

39.4 (3.1)

7-F

5%

0.96 (0.21)

3.0 (0.1)

0 (4.04)

8-OH-DPAT

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function were performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}Data for these

compounds were published in a previous paper.²⁶

Table 5. Linkage of Chroman Moiety to the Indole Nitrogen^a

5-HT transporter

5-HT_1Areceptor

afﬁnity^b

K_i(SEM), nM

function^c

IC₅₀(SEM), nM

afﬁnity^d

K_i(SEM), nM

GTPγS^e

E_max, % (IC₅₀, nM)

R₁receptor^g

K_i, nM

compd

54a

54b

54c

54d

55a

55b

55c

55d

56a

56b

56c

56d

ﬂuoxetine

1

8-OH-DPAT

R₁

Pr

MecPr

cBu

R₉

4-F

5-F

6-F

7-F

4-F

5-F

6-F

7-F

4-F

5-F

6-F

7-F

210.5

114

70

32.9

160

>1000

776 (182)

435 (246)

>1000

138

14 (1.63)

6.54 (0.48)

10.6 (0.18)

16.6 (0.85)

9.64 (2.8)

5.44 (0.89)

3.25 (0.29)

19.25 (0.74)

17.35 (2.51)

10.78 (0.87)

12.83 (2.25)

39.95 (0.32)

nd

40

39

23%

655

15%

39%

18%

387

898

27%

5%

0 (>1000)

nd

77

9.96 (5.99)

38.7

288.5

241.5

123.5

39.3

0 (371)

31 (204)

nd

0 (>1000)

0 (428)

0 (>1000)

103

cBu

>1000

39.4 (3.1)

670

7%

14%

3.9 (0.35)

0.96 (0.21)

3.0 (0.1)

0 (4.04)

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function were performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.⁴¹

gel, 60 F-254), and spots were visualized with UV light and stained

in iodine chamber.

Intermediates 18-21 were prepared according to literature

procedures.^27,29Intermediates 24a-27a were prepared according

6992 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Table 6. Tetrahydrocarbazole as Transporter Moiety^a

Hatzenbuhler et al.

5-HT transporter

5-HT_1Areceptor

afﬁnity^b

K_i(SEM), nM

function^c

IC₅₀(SEM), nM

afﬁnity^d

K_i(SEM), nM

GTPγS^e

E_max, % (IC₅₀, nM)

R₁receptor^g

K_i, nM

compd

57a

57b

58a

58b

59a

59b

60a

60b

60c

60d

ﬂuoxetine

1

8-OH-DPAT

R₁

optical rotation^h

H

-52.8

59.8

-90.8

12.4

-89

10

-26.2

27.4

20.9

47.6

>1000

180.5

347

116

855

192

291

38.05(0.25)

31.4 (11.8)

60.3 (7.35)

7.34 (1.35)

117.5(8.84)

9.48 (1.15)

49%@1 µM

0.85 (0.11)

185 (14.1)

73.3 (10.5)

0 (429)

0 (174)

0 (>1000)

0 (181)

0 (>1000)

0 (611)

0 (>1000)

25 (>1000)

55 (978)

0 (>1000)

23%

250

5%

15%

3%

22%

9%

5%

>1000

39.4 (3.1)

Pr

MecPr

cBu

50.2

-50.2

>1000

322

3.9 (0.35)

1%

0%

0.96 (0.21)

3.0 (0.1)

0 (4.04)

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function was performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}Optical rotations of

compounds 57-59 were recorded in DMSO and 60 in MeOH.

Table 7. Substitution on the Linker^a

5-HT transporter

5-HT_1Areceptor

optical

afﬁnity^b

K_i(SEM), nM

function^c

IC₅₀(SEM), nM

afﬁnity^d

K_i(SEM), nM

GTPγS^eE_max, %

R₁receptor^g

K_i, nM

compd

67a

67b

67c

67d

68a

68b

70a

70b

R₁

Pr

MecPr

rotation^h

(IC₅₀, nM)

-22.1

23.4

-67.4

61.4

-17

15.4

-22.2

-59.6

22.0 (7.07)

50

121

152

10.2

46.7

48.8

144.5

3.9 (0.35)

330 (75)

>1000

193

162

498

>1000

39.4 (3.1)

3.19 (0.71)

19.39 (5.48)

48.77 (0.48)

44%

1.88 (0.57)

31.65 (1.59)

4.16 (0.35)

127.5 (6.01)

0 (323)

73.5

0 (>1000)

7 (>1000)

0 (333)

32

48%

5%

14%

nd

15%

7%

24%

16%

0 (151)

nd

ﬂuoxetine

1

8-OH-DPAT

0.96 (0.21)

3.0 (0.1)

0 (4.04)

100 (26)^f

^aK_iand IC₅₀values are the mean of at least two experiments ( SEM (performed in singlet, determined from 11 concentrations for 5-HT_1Abinding and

function; 5-HT transporter binding and function was performed in triplicate, determined from 8 concentrations). Values without SEM are for a single

determination only. nd: not determined. Percentages represent inhibition of binding at 1 µM. ^bBinding afﬁnity for the 5-HT transporter was determined by

displacement of [³H]-paroxetine from rat cortical membranes.³⁶K_ivalues were calculated from IC₅₀values using the method of Cheng and Prusoff.³⁷

^cInhibition of [³H]-5-HT uptake by human 5-HT transporter in Jar cells.^{38 d}Binding afﬁnity at human 5-HT_1Areceptor in CHO cells labeled with

[³H]-8-OH-DPAT.^{39 e}Stimulation of GTPγS binding in CHO cells expressing 5-HT_1Areceptors.⁴⁰E_maxrefers to maximal agonist effect relative to DPAT.

^fReported as an EC₅₀of 26 nM for 8-OH-DPAT. ^gBinding afﬁnity at rat cortical R₁adrenergic receptor labeled with [³H]-prazosin.^{41 h}All optical rotations

were recorded in DMSO.

to procedures elaborated in U.S. Patent No. 6,197,978, and

intermediates 28a,29a were prepared as we recently described.²⁶

New intermediates 24c, 25b,c, 26b,c, and 27b,c were synthesized

using the same procedure as for 24a-27a. New intermediate 28b

was prepared as described for 28a,²⁶while 28c was synthesized

using the procedure published in U.S. Patent No. 6,197,978.

Intermediates 32a,b were prepared according to procedures elabo-

rated in U.S. Patent No. 6,121,307.

Methyl 4-Chloro-3-hydroxybenzoate (24b). 2-Chloro-5-triﬂuo-

romethylphenol 23 (5 g, 25 mmol) was added to concentrated

sulfuric acid (37 g, 375 mmol) under exclusion of moisture and

heated under stirring to 100 °C. The reaction mixture was stirred

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6993

at this temperature for 1 h, then cooled to ambient temperature

and poured slowly into anhydrous MeOH (300 mL). The obtained

solution was reﬂuxed for 3 h, followed by stirring at ambient

temperature overnight. The solvent was removed in vacuo to a ﬁnal

volume of ∼150 mL and then poured carefully into 10% aqueous

Na₂CO₃(300 mL). The product was extracted with Et₂O (3×), and

the combined organic layers were washed with brine (100 mL),

dried over anhydrous MgSO₄, ﬁltered, and concentrated. The 4.2 g

(90.5%) of 24b was isolated as an off-white solid: mp 100-101

°C; ¹H NMR (400 MHz, DMSO-d₆) δ 3.83 (s, 3H), 7.35-7.41

(m, 1H), 7.48 (d, J ) 8.1 Hz, 1H), 7.55 (d, J ) 2.0 Hz, 1H,) 10.65

(br s, 1H); MS (APCI) m/z 187 ([M + H]⁺).

3-Amino-8-ﬂuorochromane-5-carboxamide (29a). To 28a (3.73

g, 15.7 mmol) in CHCl₃(340 mL)-ⁱPrOH (125 mL), under nitrogen

at room temperature, was added silica gel (11 g). To the slurry

was slowly added over a 15 min period NaBH₄(1.48 g, 39.2 mmol).

After 30 min, the reaction mixture was quenched with acetic acid

(28 mL) and stirred for another 30 min. The reaction mixture was

then ﬁltered and the silica gel washed thoroughly with CH₂Cl₂.

The ﬁltrate was concentrated and the residue taken up in EtOAc/

H₂O. The organic layer was separated, treated with brine, dried

over anhydrous MgSO₄, ﬁltered, and concentrated. The 3.65 g

(97%) of 8-ﬂuoro-3-nitrochromane-5-carboxamide was isolated as

an off-white solid: mp 184-185.5 °C; ¹H NMR (500 MHz, DMSO-

d₆) δ 3.41-3.53 (m, 1H), 3.52-3.67 (m, 1H), 4.33-4.52 (m, 1H),

4.74-4.92 (m, 1H), 5.42 (s, 1H), 6.98-7.24 (m, 2H), 7.43 (s, 1H),

7.84 (s, 1H); MS (ESI) m/z 239 ([M - H]^-). Anal. (C₁₀H₉FN₂O₄)

C, H, N.

8-Fluoro-N-methyl-2H-chromene-5-carboxamide (30). To 26a

(3.5 g, 0.018 mol) in anhydrous THF (100 mL), under nitrogen at

room temperature, was added 1-(3-dimethylaminopropyl)-3-ethyl-

carbodiimide hydrochloride (EDC, 6.9 g, 0.036 mol), 1-hydroxy-

benzotriazole hydrate (HOBt, 4.86 g, 0.036 mol), and methylamine

(2 M/THF, 36 mL, 0.072 mol). The mixture was stirred at room

temperature overnight. It was concentrated and the residue taken

up in CH₂Cl₂/H₂O. The aqueous layer was extracted one more time

with CH₂Cl₂, and the organic extracts were treated with brine, dried

over anhydrous MgSO₄, ﬁltered, and concentrated. Chromatography

(3:1 EtOAc-hexanes) afforded 3.69 g (99%) of 30 as a white solid:

1

mp 150-153 °C; H NMR (400 MHz, DMSO-d₆) δ 2.73 (d, J )

4.6 Hz, 3H), 4.74-4.90 (m, 2H), 5.94-6.08 (m, 1H), 6.70-6.83

(m, 1H), 6.90-7.02 (m, 1H), 7.04-7.18 (m, 1H), 8.19-8.33 (m,

1H); MS (ES) m/z 206.1 ([M - H]^-).

3-Amino-8-ﬂuoro-N-methylchromane-5-carboxamide (31). Ni-

tration of 30 was carried out using the procedure described in U.S.

Patent No. 6,197,978. Reduction with NaBH₄was carried out as

described above for 29a, generating 2.54 g (98%) of 8-ﬂuoro-N-

methyl-3-nitrochromane-5-carboxamide: ¹H NMR (400 MHz, DM-

SO-d₆) δ 2.65-2.80 (d, 3H), 3.40-3.50 (m, 2H), 4.38-4.48 (m,

1H), 4.75-4.90 (m, 1H), 5.35-5.45 (m, 1H), 6.97-7.08 (m, 1H),

7.08-7.20 (m, 1H), 8.20-8.40 (m, 1H). Reduction of the nitro to

the amine using Raney Ni and hydrazine hydrate was carried out

as described above for 29a, affording 1.35 g (60%) of 31 as an

off-white solid which was converted to the monohydrochloride salt:

1

mp 131 °C/dec; H NMR (400 MHz, DMSO-d₆) δ 2.73 (d, J )

4.6 Hz, 3H), 2.93-3.06 (m, 1H), 3.20-3.42 (m, 1H), 3.55-3.66

(m, 1H), 3.71-3.84 (m, 1H), 4.15-4.35 (m, 2H), 7.01-7.08 (m,

1H), 7.13-7.21 (m, 1H), 8.15-8.38 (m, 3H); MS (ES) m/z 225.1

([M + H]⁺). Anal. (C₁₁H₁₃FN₂O₂·1HCl·0.5H₂O) C, H, N.

Methyl 3-Amino-8-ﬂuorochromane-5-carboxylate (36). Nitration

of 25a was carried out using the procedure already published.²⁶

Reduction with NaBH₄was carried out as described above for 29a,

generating 1.72 g (84%) of methyl 8-ﬂuoro-3-nitrochromane-5-

carboxylate: ¹H NMR (400 MHz, DMSO-d₆) δ 3.45-3.60 (m, 1H),

3.75-3.90 (m, 4H), 4.38-4.50 (m, 1H), 4.78-4.90 (m, 1H),

5.40-5.50 (m, 1H), 7.18-7.30 (m, 1H), 7.50-7.60 (m, 1H).

Reduction of the nitro to the amine using Raney Ni and hydrazine

hydrate was carried out as described above for 29a, affording 1.46 g

(55%) of 36 as a thick orange liquid: ¹H NMR (400 MHz, DMSO-

d₆) δ 1.60-2.00 (m, 2H), 2.62-2.78 (m, 1H), 3.05-3.28 (m, 2H),

3.62-3.74 (m, 1H), 3.78 (s, 3H), 4.08-4.20 (m, 1H), 7.05-7.20

(m, 1H), 7.32-7.45 (m, 1H).

1-(4-Bromobutyl)-4-ﬂuoro-1H-indole (48a). To a solution of the

appropriate indole (1 g) in 20 mL of DMF was added sodium

hydride (60% in mineral oil, 8.14 mmol). The solution was stirred

for 1-2 h and then treated with 1,4-dibromobutane (2.66 mL, 22

mmol). The mixture was stirred for 45 min to 2 h, quenched with

20 mL of H₂O, and extracted with EtOAc. The combined extracts

were dried over anhydrous MgSO₄and concentrated. Puriﬁcation

carried out by ﬂash chromatography using a Biotage Quad 12/25

(Dyax Corp) with KP Sil 32-63 mM, 60 Å cartridge, and a gradient

of 100% Hex to 2% EtOAc/Hex generated a 72% yield of 48a as

a clear oil. ¹H NMR (400 MHz, DMSO-d₆) δ 1.67-1.79 (m, 2H),

1.86 (q, 2H), 3.52 (t, J ) 6.6 Hz, 2H), 4.23 (t, J ) 6.8 Hz, 2H),

6.48 (d, J ) 2.9 Hz, 1H), 6.78 (t, J ) 9.3 Hz, 1H), 7.10 (dd, J )

16.1, 5.4 Hz, 1H), 7.34 (d, J ) 8.3 Hz, 1H), 7.42 (d, 1H); MS

(ESI) m/z 269.1, 271.1([M + H]⁺).

To 8-ﬂuoro-3-nitrochromane-5-carboxamide (2 g, 8.3 mmol) in

absolute EtOH (90 mL) was added THF (15 mL). The mixture

was heated to 60 °C to help solubilize the starting material and

cooled down to 45 °C. Wet Raney Ni was added followed by, over

a 30 min period, a solution of hydrazine hydrate (4.7 mL) in

absolute EtOH (12 mL). The reaction mixture was kept at 45 °C

for 1 h. While still warm, the mixture was ﬁltered over Celite and

the Raney Ni washed thoroughly with hot EtOH. The ﬁltrate was

concentrated under vacuum. Chromatography (4:1 CH₂Cl₂-MeOH

(1% NH₄OH)) afforded 1.06 g (61%) of 29a as an off-white solid:

mp 229 °C/dec; ¹H NMR (500 MHz, DMSO-d₆) δ 2.96-3.15 (m,

1H), 3.23-3.43 (m, 1H), 3.67-3.89 (m, 1H), 4.15-4.37 (m, 2H),

7.03-7.23 (m, 2H), 7.43 (s, 1H), 7.80 (s, 1H), 8.40 (s, 3H); MS

(ESI) m/z 211 ([M + H]⁺). Anal. (C₁₀H₁₁FN₂O₂·1HCl) C, H, N.

The enantiomers of 29a were separated by chiral resolution with

L-(+)-tartaric acid to generate (R)-(+)-29a as a tartrate salt: [R]_D²⁵

+71.0° (c 1.0, H₂O).

3-Amino-8-chlorochromane-5-carboxamide (29b). This inter-

mediate was prepared as described above for 29a, generating 1.28 g

(85%) of 8-chloro-3-nitrochromane-5-carboxamide as a light-beige

solid: mp 193-4 °C; MS (ESI) m/z 255/257 ([M - H]^-). Reduction

in the presence of Raney Ni and hydrazine hydrate followed by

chromatography (1:4 MeOH-CHCl₃(1% NH₄OH)) afforded 0.78 g

(83%) of 29b as an off-white solid: mp 231-6 °C; MS (ES) m/z

225 ([M - H]^-). It was then converted to the HCl salt by dissolving

the free base in methanol, adding ethereal HCl, and evaporating

the solution in vacuo. The residue was crystallized from methanol/

ether, affording the mono-HCl salt of 29b as off-white microcrys-

1

tals: mp >250 °C; H NMR (400 MHz, DMSO-d₆) δ 2.97-3.12

(m, 1H), 3.21-3.41 (m, 1H), 3.74-3.86 (m, 1H), 4.20-4.35 (m,

2H), 7.02-7.15 (m, 1H), 7.33-7.44 (m, 1H), 7.46-7.54 (m, 1H),

7.79-7.91 (m, 1H), 8.21-8.40 (m, 3H); MS (ES) m/z 225.0 ([M

- H]^-).

1-(4-Bromobutyl)-5-ﬂuoro-1H-indole (48b). 48b was obtained

according to the procedure for 48a, and a 69% yield of 48b was

isolated as a clear oil. ¹H NMR (400 MHz, DMSO-d₆) δ 1.62-1.76

(m, 2H), 1.76-1.90 (m, 2H), 3.42-3.52 (m, 2H), 4.10-4.22 (m,

2H), 6.34-6.40 (m, 1H), 6.88-6.98 (m, 1H), 7.20-7.30 (m, 1H),

7.38-7.42 (m, 1H), 7.42-7.50 (m, 2H); MS (ESI) m/z 269.1, 271.1

([M + H]⁺).

1-(4-Bromobutyl)-6-ﬂuoro-1H-indole (48c). 48c was obtained

according to the procedure for 48a, and a 68% yield of 48c was

isolated as a clear oil. ¹H NMR (500 MHz, DMSO-d₆) δ 1.69-1.78

(m, 2H), 1.79-1.88 (m, 2H), 3.47-3.58 (m, 2H), 4.09-4.23 (m,

3-Aminochromane-5-carboxamide (29c). This intermediate was

prepared as described above for 29a, generating 2 g (95%) of

3-nitrochromane-5-carboxamide as off-white microcrystals; mp

200-202 °C. MS (ES) m/z 223.0. Reduction in the presence of

Raney Ni and hydrazine hydrate afforded 1.75 g (∼100%) of 29c

as waxy faintly green solid. ¹H NMR (400 MHz, DMSO-d₆) δ

2.10-2.40 (m, 2H), 2.55-3.20 (m, 2H), 3.40-3.60 (m, 1H),

3.90-4.20 (m, 2H), 6.75 (d, 1H), 6.85 (d, 1H), 7.02 (t, 1H), 7.30

(s, 1H), 7.60 (s, 1H); MS (ES) m/z 193.1 ([M + H]⁺).

6994 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

2H), 6.40-6.48 (m, 1H), 6.78-6.92 (m, 1H), 7.31-7.41 (m, 2H),

7.45-7.56 (m, 1H); MS (ESI) m/z 269.0, 271.1 ([M + H]⁺).

1-(4-Bromobutyl)-7-ﬂuoro-1H-indole (48d). 48d was obtained

according to the procedure for 48a, and a 44% yield of 48d was

isolated as a clear oil by preparative HPLC (Primesphere Silica, 5

cm × 25 cm column, ﬂow rate 95 mL/min, sample dissolved in

hexane, mobile phase of 5% ethyl acetate in hexane). ¹H NMR

(400 MHz, DMSO-d₆) δ 1.64-1.79 (m, 2H), 1.79-1.93 (m, 2H),

3.51 (t, J ) 6.7 Hz, 2H), 4.30 (t, J ) 6.7 Hz, 2H), 6.48 (s, 1H),

6.83-6.99 (m, 2H), 7.28-7.43 (m, 2H); MS (ESI) m/z 270.0, 272.0

([M + H]⁺).

6.82-6.93 (m, 1H), 7.08-7.14 (m, 1H), 7.15-7.21 (m, 1H),

7.24-7.36 (m, 1H), 10.97-11.09 (m, 1H).

3-(5-Fluoro-1H-indol-3-yl)-2-methylpropanoic Acid (64). A solu-

tion of 63 (15.7 g, 48.9 mmol) in EtOH (160 mL) was treated with

2.5 N aqueous NaOH (80 mL, 200 mmol) and reﬂuxed for 1.5 h.

The cooled solution was concentrated to remove EtOH. The residue

was acidiﬁed with concentrated HCl and extracted with EtOAc

(3×). The combined organic extracts were treated with brine, dried

over anhydrous MgSO₄, and concentrated. The residue was

triturated with CH₂Cl₂/hexane and dried under vacuum to afford

12.1 g (93%) of [(5-ﬂuoro-1H-indol-3-yl)methyl](methyl)malonic

acid as a pinkish-white solid: mp 135-137 °C/dec.

Ethyl 6-Fluoro-2,3,4,9-tetrahydro-1H-carbazole-3-carboxylate

(49). 4-Cyclohexanonecarboxylic acid ethyl ester (25 g, 0.14 mol)

and 4-ﬂuorophenyhydrazine hydrochloride (22.5 g, 0.13 mol) were

dissolved in EtOH (450 mL) and heated under reﬂux for 16 h. After

the mixture was cooled, the white solid was ﬁltered off and the

ﬁltrate concentrated. The residue was taken up in EtOAc, washed

with H₂O, dried over anhydrous MgSO₄, and concentrated to

generate 35.5 g (93%) of 49 which was recrystallized from heptane:

A suspension of [(5-ﬂuoro-1H-indol-3-yl)methyl](methyl)malonic

acid (12.3 g, 46.4 mmol) in bromobenzene (50 mL) was reﬂuxed

for 1.5 h and then concentrated. The residue was triturated with

CH₂Cl₂/hexane. Air drying afforded 8.86 g (86%) of 64 as a tan

solid: mp 112-113 °C; ¹H NMR (400 MHz, DMSO-d₆) δ

0.99-1.11 (m, 3H), 2.58-2.76 (m, 2H), 2.87-3.01 (m, 1H),

6.81-6.92 (m, 1H), 7.14-7.19 (m, 1H), 7.20-7.34 (m, 2H), 10.89

(br s, 1H), 11.88-12.20 (m, 1H); MS (ES) m/z 220.1 ([M - H]^-).

3-(5-Fluoro-1H-indol-3-yl)-2-methylpropanal (65). A solution of

64 (9.24 g, 41.8 mmol) in anhydrous THF (30 mL) was chilled to

0 °C and treated with 1 N LiAlH₄/THF (50 mL, 50 mmol) at room

temperature for 3 h. Additional 1 N LiAlH₄/THF (34 mL, 34 mmol)

was added and the mixture stirred for an additional 2 h. The reaction

was quenched with ice-water, diluted with saturated aqueous

potassium sodium tartrate, and extracted with EtOAc (3×). The

combined organic extracts were treated with brine, dried over

anhydrous MgSO₄, and concentrated to afford 7.74 g (89%) of (5-

ﬂuoro-1H-indol-3-yl)-2-methylpropan-1-ol as a viscous orange oil.

A solution of pyridine (8.3 mL, 100 mmol) in anhydrous toluene

(100 mL) was chilled to 0 °C and treated sequentially with TFA

(4.0 mL, 51 mmol), anhydrous DMSO (100 mL), (5-ﬂuoro-1H-

indol-3-yl)-2-methylpropan-1-ol (7.06 g, 34.1 mmol), and dicyclo-

hexylcarbodiimide (21.1 g, 102 mmol). It was stirred at room

temperature for 5 h. The reaction was quenched with ice-water

and stirred for 1 h. A white precipitate was ﬁltered out and the

resulting solution extracted with EtOAc (3×). The combined

organic extracts were washed with brine, dried over anhydrous

MgSO₄, and concentrated. Chromatography ((85:15) hexane-EtOAc)

1

mp 115-117 °C; H NMR (400 MHz, DMSO-d₆) δ 1.21 (t, J )

7.1 Hz, 3H), 1.78-1.93 (m, 1H), 2.11-2.24 (m, 1H), 2.64-2.93

(m, 5H), 4.02-4.18 (m, 2H), 6.75-6.85 (m, 1H), 7.06-7.13 (m,

1H), 7.16-7.23 (m, 1H), 10.78 (s, 1H); MS (ES) m/z 262 ([M +

H]⁺).

6-Fluoro-2,3,4,9-tetrahydro-1H-carbazole-3-carbaldehyde (50).

Lithium aluminum hydride (800 mg) was added portionwise to a

solution of 49 (5.77 g, 22.1 mmol) in dry THF (100 mL). The

mixture was stirred at ambient temperature under nitrogen for 16 h,

followed by quenching with the addition of an aqueous Rochelle

salt solution. The mixture was diluted with Et₂O, and the aqueous

phase was extracted once more with Et₂O. The organic extracts

were pooled, dried over anhydrous MgSO₄, and concentrated.

Chromatography ((2:1) Hex-EtOAc) afforded 3.90 g (80%) of (6-

ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-3-yl)methanol: mp 107-109

°C; MS (ES) m/z 218.1 ([M - H]^-).

Dess-Martin periodinane (7.37 g 17.4 mmol) was added

portionwise to a stirred solution/suspension of (6-ﬂuoro-2,3,4,9-

tetrahydro-1H-carbazol-3-yl)methanol (2.64 g, 11.8 mmole) in

CH₂Cl₂(120 mL). The alcohol completely dissolved after the

Dess-Martin reagent was added. The mixture was stirred at ambient

temperature for 30 min and quenched with EtOH. The mixture was

diluted with Et₂O, washed with saturated aqueous NaHCO₃(2×)

followed by 5% sodium thiosulfate pentahydrate, treated with brine,

dried over anhydrous MgSO₄, and concentrated. Chromatography

((4:1) Hex-EtOAc) afforded 1.2 g (47%) of 50: mp 96-98 °C;

¹H NMR (400 MHz, DMSO-d₆) δ 1.78-1.94 (m, 1H), 2.12-2.23

(m, 1H), 2.76 (t, J ) 11.6 Hz, 5H), 6.75-6.85 (m, 1H), 7.07-7.15

(m, 1H), 7.15-7.24 (m, 1H), 9.73 (s, 1H), 10.78 (br s, 1H); MS

(ES) m/z 216.1 ([M - H]^-).

1

afforded 5.45 g (78%) of 65 as a tan solid. H NMR (400 MHz,

DMSO-d₆) δ 0.95 (d, 3H), 2.62-2.80 (m, 2H), 2.96-3.08 (m, 1H),

6.80-6.90 (m, 2H), 7.12-7.32 (m, 3H), 9.63 (s, 1H), 10.90 (s,

1H).

General Procedure A (Alkylation Reaction). N-[3-(5-Fluoro-

1H-indol-3-yl)propyl]-5-methoxychroman-3-amine (33a) and (3R)-

N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-methoxychroman-3-

amine ((R)-33a). A solution of 19 (0.20 g, 1.12 mmol), 32b (0.19

g, 0.75 mmol), and triethylamine (0.21 mL, 1.5 mmol) in anhydrous

DMSO (6 mL) was stirred at 90 °C for 9 h. The mixture was cooled

to room temperature, diluted with EtOAc, and washed with H₂O

(2×). The organic layer was treated with brine, dried over anhydrous

MgSO₄, ﬁltered, and concentrated. Chromatography (6:3:1 hexanes/

EtOAc/MeOH (1% NH₄OH)) afforded 0.16 g (60%) of 33a as a

dark-yellow solid. Conversion to the oxalate salt using oxalic acid

in isopropanol generated a white solid: mp 133-136 °C; ¹H NMR

(400 MHz, DMSO-d₆) δ 1.89-2.06 (m, 2H), 2.65-2.83 (m, 3H),

2.92-3.20 (m, 3H), 3.68-3.87 (m, 5H), 4.15-4.30 (m, 2H),

6.45-6.52 (m, 1H), 6.56-6.65 (m, 1H), 6.85-6.97 (m, 1H),

7.07-7.16 (m, 1H), 7.21-7.27 (m, 1H), 7.27-7.40 (m, 2H), 10.94

4-(5-Fluoro-1H-indol-3-yl)butan-2-one (61). To 5-ﬂuoroindole (5

g, 37mmol) in nitromethane (150 mL) was added methyl vinyl

ketone (3.7 mL, 44 mmol). The mixture was cooled to -20 °C,

and a mixture of BF₃·Et₂O (0.94 mL, 7.4 mmol) and EtOH (4 mL)

was added dropwise. The mixture was stirred at -20 °C for 45

min. It was quenched with 1 N NaOH, diluted with EtOAc,

extracted with H₂O (2×), treated with brine, dried over anhydrous

MgSO₄,ﬁltered,andconcentrated.Chromatography((2:1)Hex-EtOAc))

1

afforded 6.23 g (81%) of 61. H NMR (400 MHz, DMSO-d₆) δ

2.08 (s, 3H), 2.72-2.86 (m, 4H), 6.84-6.92 (m, 1H), 7.14-7.18

(m, 1H), 7.22-7.32 (m, 2H), 10.85 (br s, 1H).

(s, 1H); MS (ESI) m/z 355 ([M

+

H]⁺). Anal.

(C₂₁H₂₃FN₂O₂·C₂H₂O₄) C, H, N. Similarly, (R)-(-)-19 was reacted

with 32b followed by chromatography (6:3:1 hexanes/EtOAc/

MeOH (1% NH₄OH)) to afford 0.2 g (71%) of (R)-33a as an orange

gum.

Diethyl [(5-Fluoro-1H-indol-3-yl)methyl](methyl)malonate (63).

A solution of 5-ﬂuorogramine 62 (15.5 g, 80.6 mmol) in CH₃CN

(450 mL) was treated with diethyl methylmalonate (20.8 mL, 121

mmol) and tributylphosphine at reﬂux for 17 h. The cooled reaction

was concentrated, dissolved in EtOAc, washed with 1 N aqueous

HCl, treated with brine, dried over anhydrous MgSO₄, and

concentrated. Chromatography ((3:1) hexane-EtOAc) afforded

16.5 g (64%) of 63 as an oil which solidiﬁed on standing to a white

solid: mp 76-77 °C; ¹H NMR (400 MHz, DMSO-d₆) δ 1.15 (t, J

) 7.1 Hz, 6H), 1.27 (s, 3H), 3.21 (s, 2H), 3.98-4.17 (m, 4H),

3-{[3-(5-Cyano-1H-indol-3-yl)propyl]amino}-8-ﬂuorochromane-

5-carboxamide (51a). Using general procedure A, 29a was reacted

with 32d as described above for 33a followed by chromatography

((96:4) CH₂Cl₂-MeOH (5% NH₄OH)) to generate 0.677 g (47%)

1

of 51a as an off-white solid: H NMR (400 MHz, DMSO-d₆) δ

1.65-1.80 (m, 2H), 2.52-2.78 (m, 4H), 2.85-3.10 (m, 2H),

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6995

3.20-3.32 (m, 2H), 3.68-3.85 (m, 1H), 4.12-4.25 (m, 1H),

6.80-7.10 (m, 2H), 7.20-7.50 (m, 4H), 7.58-7.70 (m, 1H), 8.04

(s, 1H).

solid: ¹H NMR (400 MHz, DMSO-d₆) δ 1.95-2.08 (m, 2H),

2.71-2.79 (m, 2H), 2.80-2.91 (m, 1H), 2.98-3.15 (m, 3H),

3.73-3.84 (m, 4H), 4.26-4.41 (m, 2H), 6.49-6.57 (m, 1H),

6.87-6.96 (m, 1H), 7.03-7.12 (m, 1H), 7.24-7.28 (m, 1H),

7.29-7.37 (m, 2H), 8.88-9.31 (m, 2H), 10.97 (s, 1H); MS (ESI)

m/z 373 ([M + H]⁺). Anal. (C₂₁H₂₂F₂N₂O₂· 1HCl · 0.25H₂O) C,

H, N.

(R)-33b and (S)-33b were prepared according to general proce-

dure A using (R)-(-)-21 and (S)-(+)-21, respectively, as starting

material. Both products were found to be >99% ee by chiral HPLC

using either Chiralcel OJ column (20 mm × 250 mm) with

100%EtOH or a Chiralpak AS column (4.6 mm × 250 mm) with

50% EtOH/hexane.

8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (33c), (3R)-8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-

yl)propyl]amino}chromane-5-carboxamide ((R)-33c), and (3S)-8-

Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-5-

carboxamide ((S)-33c). Using general procedure A, 29a was reacted

with 32b as described above for 33a followed by chromatography

(5:4:1 EtOAc-hexanes-MeOH (1% NH₄OH)) to generate 0.48 g

(60%) of 33c as a peach solid which was converted to the mono-

HCl salt, yielding an off-white solid: mp 122 °C/dec; ¹H NMR

(500 MHz, DMSO-d₆) δ 1.90-2.08 (m, 2H), 2.68-2.82 (m, 2H),

2.98-3.25 (m, 3H), 3.27-3.45 (m, 2H), 3.71-3.88 (m, 1H),

4.27-4.46 (m, 2H), 6.82-6.98 (m, 1H), 7.08-7.21 (m, 2H),

7.22-7.27 (m, 1H), 7.27-7.37 (m, 2H), 7.43 (s, 1H), 7.81 (s, 1H),

9.00 (s, 1H), 10.95 (s, 1H); MS (ESI) m/z 384([M - H]^-). Anal.

(C₂₁H₂₁F₂N₃O₂·1.20HCl) C, H, N.

8-Fluoro-3-{[3-(5-methoxy-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (51b). Using general procedure A, 29a was reacted

with 32c as described above for 33a followed by chromatography

((95:5) CH₂Cl₂-methanol(5% NH₄OH) to generate 0.75 g (62%)

of 51b which was converted to the mono-HCl salt as a reddish-

white solid: mp 171-176 °C (melts with decomposition); ¹H NMR

(400 MHz, DMSO-d₆) δ 1.95-2.08 (m, 2H), 2.69-2.81 (m, 2H),

2.99-3.45 (m, 4H), 3.72-3.85 (m, 4H), 4.29-4.45 (m, 2H),

6.66-6.77 (m, 1H), 6.97-7.05 (m, 1H), 7.07-7.26 (m, 4H),

7.37-7.48 (m, 1H), 7.75-7.85 (m, 1H), 8.98-9.11 (m, 2H),

10.63-10.72 (m, 1H); MS (ES) m/z 398.1 ([M + H]⁺). Anal.

(C₂₂H₂₄FN₃O₃·HCl·0.30H₂O) C, H, N (N calcd 9.56, found 8.55).

8-Fluoro-3-[4-(4-ﬂuoroindol-1-yl)butylamino]chroman-5-car-

boxylic Acid Amide (53a). Using general procedure A, 29a was

reacted with 48a as described above for 33a. Puriﬁcation was carried

out using a Biotage Quad 12/25 (Dyax Corp) with KP Sil 32-63

mM, 60 Å cartridges, and a gradient from 100% CH₂Cl₂to 4%

methanolic ammonia/CH₂Cl₂), generating 53a (63%) as a pale-

1

yellow foam: H NMR (400 MHz, DMSO-d₆) δ 1.23-1.39 (m,

2H), 1.66-1.82 (m, 3H), 2.51-2.69 (m, 4H), 2.82-2.94 (m, 1H),

2.94-3.06 (m, 1H), 3.67-3.78 (m, 1H), 4.07-4.24 (m, 3H),

6.38-6.47 (m, 1H), 6.67-6.78 (m, 1H), 6.83-6.93 (m, 1H),

6.95-7.10 (m, 2H), 7.24-7.31 (m, 2H), 7.33-7.39 (m, 2H), 7.63

(br s, 1H); MS (ESI), m/z 400.17 [M + H]⁺, MS (ESI) m/z 398.2

[M - H]^-.

It was also prepared using general procedure B where 29a is

reacted with 32a as described above for 33b, generating 2.17 g

(92%) of 33c as a white foamy solid which was then subjected to

chiral separation of the enantiomers by SFC on a Chiralcel AS

column (2 cm × 25 cm) using 40% MeOH in CO₂(100 bar). The

enantiomers were isolated and characterized as mono-HCl salts.²⁶

In some instances, enantiomers were prepared from (R)-(+)-29a

and (S)-(-)-29a, respectively, as speciﬁed in each experimental of

the ﬁnal targets.

8-Fluoro-3-[4-(5-ﬂuoroindol-1-yl)butylamino]chroman-5-car-

boxylic Acid Amide (53b). Using general procedure A, 29a was

reacted with 48b as described above for 53a, generating 53b (58%)

as a pale-yellow foam: ¹H NMR (500 MHz, DMSO-d₆) δ

1.28-1.39 (m, 2H), 1.70-1.81 (m, 3H), 2.55-2.69 (m, 3H),

2.85-2.96 (m, 1H), 2.98-3.08 (m, 1H), 3.71-3.81 (m, 1H),

4.08-4.25 (m, 3H), 6.36-6.42 (m, 1H), 6.88-6.98 (m, 2H),

6.99-7.07 (m, 1H), 7.22-7.35 (m, 2H), 7.39-7.49 (m, 2H), 7.66

(s, 1H); MS (ESI) m/z 400.2 [M + H]⁺, MS (ESI) m/z 398.2 [M

- H]^-.

8-Chloro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (33d). Using general procedure B, 29b was reacted

with 32a as described above for 33b, generating 0.37 g (54%) of

8-Fluoro-3-[4-(6-ﬂuoroindol-1-yl)butylamino]chroman-5-car-

boxylic Acid Amide (53c). Using general procedure A, 29a was

reacted with 48c as described above for 53a, generating 53c (60%)

1

33d as a dense white foam: mp 65-67 °C; H NMR (400 MHz,

1

DMSO-d₆) δ 1.68-1.81 (m, 2H), 1.87 (br s, 1H), 2.67 (d, J ) 7.1

Hz, 4H), 2.91-3.12 (m, 2H), 3.24-3.38 (m, 1H), 3.78-3.90 (m,

1H), 4.23-4.33 (m, 1H), 6.82-6.97 (m, 2H), 7.17 (br s, 1H),

7.21-7.35 (m, 3H), 7.41 (br s, 1H), 7.74 (br s, 1H), 10.83 (s, 1H);

MS (ES) m/z 402.1 ([M + H]⁺).

as a white solid: mp 146-148 °C; H NMR (400 MHz, DMSO-

d₆) δ 1.29-1.41 (m, 2H), 1.67-1.83 (m, 3H), 2.53-2.69 (m, 3H),

2.86-2.96 (m, 1H), 2.97-3.09 (m, 1H), 3.69-3.81 (m, 1H),

4.07-4.25 (m, 3H), 6.38-6.45 (m, 1H), 6.78-6.96 (m, 2H),

6.96-7.09 (m, 1H), 7.25-7.39 (m, 3H), 7.44-7.54 (m, 1H), 7.66

(br s, 1 H); MS (ESI) m/z 400.2 [M + H]⁺, MS (ESI) m/z 398.2

[M - H]^-.

The enantiomers of 33d were separated by chiral HPLC using a

Chiralpak AS column (2 cm × 25 cm) with (3:7) hexanes-EtOH

as mobile phase to generate each enantiomer which were both

isolated and characterized as free bases.

3-{[3-(5-Fluoro-1H-indol-3-yl)propyl]amino}chromane-5-car-

boxamide (33e). Using general procedure B, 29c was reacted with

32a as described above for 33b, generating 1.37 g (42%) of 33e as

a white amorphous powder: mp 153-155 °C; ¹H NMR (400 MHz,

DMSO-d₆) δ 1.70-1.85 (m, 2H), 2.57-2.75 (m, 5H), 2.90-3.13

(m, 2H), 3.32 (br s, 1H), 3.65-3.80 (m, 1H), 4.12-4.25 (m, 1H),

6.78-6.84 (m, 1H), 6.88 (td, J ) 9.2, 2.4 Hz, 2H), 7.05-7.14 (m,

1H), 7.15-7.21 (m, 1H), 7.21-7.39 (m, 3H), 7.68 (s, 1H), 10.83

(s, 1H); MS (ES) m/z 368.2 ([M + H]⁺).

The enantiomers of 33e were separated by chiral HPLC using a

Chiracel AS column (2 cm × 25 cm) with 75% EtOH/Hex as

mobile phase to generate each enantiomer which were both isolated

and characterized as free bases.

8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-N-meth-

ylchromane-5-carboxamide (33f). Using general procedure B, 31

was reacted with 32a as described above for 33b, generating 0.52 g

(87%) of 33f as a white solid which was converted to the HCl salt:

mp 148 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ 1.88-2.09 (m,

2H), 2.66-2.82 (m, 5H), 2.96-3.23 (m, 3H), 3.25-3.43 (m, 1H),

3.71-3.89 (m, 1H), 4.29-4.45 (m, 2H), 6.85-6.96 (m, 1H),

7.01-7.10 (m, 1H), 7.13-7.22 (m, 1H), 7.22-7.27 (m, 1H),

8-Fluoro-3-{[4-(7-ﬂuoro-1H-indol-1-yl)butyl]amino}chromane-

5-carboxamide (53d). Using general procedure A, 29a was reacted

with 48d as described above for 53a, generating 53d as a white

solid: mp 166-168 °C; ¹H NMR (400 MHz, DMSO-d₆) δ

1.26-1.40 (m, 2H), 1.68-1.84 (m, 3H), 2.53-2.70 (m, 3H),

2.84-2.95 (m, 1H), 2.95-3.09 (m, 1H), 3.68-3.79 (m, 1H),

4.13-4.34 (m, 3H), 6.41-6.51 (m, 1H), 6.82-6.97 (m, 3H),

6.98-7.09 (m, 1H), 7.24-7.42 (m, 3H), 7.66 (br s, 1H); MS (ESI)

m/z 400.2 [M + H]⁺, MS (ESI) m/z 398.2 [M - H]^-.

General Procedure B (Reductive Amination). 8-Fluoro-N-[3-

(5-ﬂuoro-1H-indol-3-yl)propyl]-5-methoxychroman-3-amine (33b),

(3R)-8-Fluoro-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-methoxychro-

man-3-amine ((R)-33b), and (3S)-8-Fluoro-N-[3-(5-ﬂuoro-1H-indol-

3-yl)propyl]-5-methoxychroman-3-amine ((S)-33b). To 21 (0.24 g,

1.22 mmol) in anhydrous MeOH (18 mL) was added 32a (0.198

g, 1.04 mmol), acetic acid (0.15 mL, 2.92 mmol), and sodium

cyanoborohydride (0.115 g, 1.83 mmol). The mixture was stirred

at room temperature overnight, quenched with 1 N NaOH/H₂O,

and concentrated. The residue was taken up in EtOAc/H₂O and

extracted with EtOAc (2×). The organic extracts were treated with

brine, dried over anhydrous MgSO₄, ﬁltered, and concentrated.

Chromatography (2:1 EtOAc-hexanes) afforded 0.34 g (87%) of

33b as a gum. Conversion to the mono-HCl salt generated a beige

6996 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

7.28-7.38 (m, 2H), 8.27 (s, 1H), 8.87-9.12 (m, 2H), 10.95 (s,

1H); MS (ES) m/z 400.2 ([M + H]⁺). Anal. (C₂₂H₂₃F₂N₃O₂·

1HCl·0.5H₂O) C, H, N.

m/z 412.1 ([M + H]⁺); [R]_D²⁵-52.8° (c 1.0, DMSO). 57b: mp

1

198-202 °C; H NMR (400 MHz, DMSO-d₆) δ 1.54-1.68 (m,

1H), 2.02-2.44 (m, 3H), 2.63-2.93 (m, 3H), 3.07-3.21 (m, 2H),

3.23-3.50 (m, 2H), 3.80-3.90 (m, 1H), 4.35-4.50 (m, 2H),

6.76-6.87 (m, 1H), 6.94-7.06 (m, 1H), 7.08-7.26 (m, 3H),

7.37-7.48 (m, 1H), 7.76-7.87 (m, 1H), 8.98-9.15 (m, 2H), 10.83

(s, 1H); MS (ES) m/z 412.1 ([M + H]⁺); [R]_D²⁵+59.8° (c 1.0,

DMSO).

3-{[3-(5-Chloro-1H-indol-3-yl)propyl]amino}-8-ﬂuorochromane-

5-carboxamide (51c). Using general procedure B, 29a was reacted

with 32e as described above for 33b, generating 0.60 g (64%) of

51c which was converted to the mono-HCl salt and isolated as a

1

white solid: mp 218-221 °C; H NMR (400 MHz, DMSO-d₆) δ

1.92-2.07 (m, 2H), 2.67-2.81 (m, 2H), 2.97-3.45 (m, 4H),

3.74-3.86 (m, 1H), 4.28-4.46 (m, 2H), 7.00-7.28 (m, 4H),

7.30-7.45 (m, 2H), 7.55-7.63 (m, 1H), 7.75-7.84 (m, 1H),

8.92-9.09 (m, 2H), 11.00-11.10 (m, 1H); MS (ES) m/z 402.1 ([M

+ H]⁺). Anal. (C₂₁H₂₁ClFN₃O₂·HCl) C, H, N.

8-Fluoro-3-{[3-(6-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (51d). Using general procedure B, 29a was reacted

with 32f as described above for 33b, generating 0.24 g (67%) of

51d: ¹H NMR (400 MHz, DMSO-d₆) δ 1.88-2.10 (m, 2H),

2.67-2.85 (m, 2H), 2.97-3.48 (m, 4H), 3.69-3.88 (m, 1H),

4.23-4.47 (m, 2H), 6.74-6.92 (m, 1H), 7.00-7.24 (m, 4H),

7.35-7.59 (m, 2H), 7.80 (br s, 1H), 9.01 (br s, 2H), 10.91 (br s,

1H); MS (ES) m/z 384.2 ([M - H]^-).

General Procedure C (Reductive Amination with Cyclobu-

tanone). 3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

8-ﬂuorochromane-5-carboxamide (35c). To 33c (0.14 g, 0.35

mmol) in anhydrous MeOH (6 mL), under nitrogen at room

temperature, was added cyclobutanone (0.070 mL, 0.876 mmol),

acetic acid (0.050 mL, 0.84 mmol), and sodium cyanoborohydride

(0.044 g, 0.70 mmol). The reaction mixture was stirred at room

temperature overnight. More cyclobutanone (0.026 mL), acetic acid

(0.21 mL), and sodium cyanoborohydride (0.22 g) were added after

24 and 48 h, at which time the reaction went to completion. Workup

was the same as for procedure B. Chromatography ((5:4:1)

EtOAc-hexane-MeOH (1% NH₄OH)) afforded 0.12 g (78%) of

35c as a sticky gum. Conversion to the mono-HCl salt generated

1

an off-white solid: mp 109 °C/dec; H NMR (500 MHz, DMSO-

8-Fluoro-3-{[3-(7-methoxy-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (51e). Using general procedure B, 29a was reacted

with 32g as described above for 33b, generating 0.28 g (68%) of

51e which was converted to the mono-HCl salt and isolated as a

d₆) δ 1.51-1.75 (m, 2H), 1.85-2.57 (m, 7H), 2.59-2.76 (m, 2H),

3.06-3.48 (m, 3H), 3.88-4.14 (m, 2H), 4.31-4.43 (m, 1H),

4.45-4.59 (m, 1H), 6.84-6.95 (m, 1H), 7.10-7.25 (m, 3H),

7.26-7.35 (m, 2H), 7.38-7.49 (m, 1H), 7.72-7.89 (m, 1H),

10.11-10.42 (m, 1H), 10.93 (s, 1H); MS (ES) m/z 438.2 ([M -

H]^-). Anal. (C₂₅H₂₇F₂N₃O₂·1.10 HCl ·0.50H₂O) C, H, N.

(3R)-(-)-3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

8-ﬂuorochromane-5-carboxamide ((R)-(-)-35c). The enantiomers

of 35c were separated by chiral HPLC using a Chiralcel AD column

(2 cm × 25 cm) and 16% IPA in hexane/DEA as mobile phase.

They were isolated and characterized as mono-HCl salts. (R)-(-)-

35c (white solid): mp 129 °C/dec; MS (ES) m/z 440.1 ([M + H]⁺);

[R]²_D⁵-26.76° (c 1.0, DMSO); 99.5% ee by chiral HPLC. Anal.

(C₂₅H₂₇F₂N₃O₂·1HCl ·0.40H₂O) C, H, N. (S)-(+)-35c (white solid):

mp 129 °C/dec; MS (ES) m/z 438.2([M - H]^-); [R]_D²⁵+27.56° (c

1.0, DMSO); 95% ee by chiral HPLC. Anal. (C₂₅H₂₇F₂N₃O₂·1

HCl·0.50H₂O) C, H, N.

1

white solid: mp 196-198 °C; H NMR (400 MHz, DMSO-d₆) δ

1.93-2.08 (m, 2 H), 2.69-2.81 (m, 2 H), 2.97-3.44 (m, 4 H),

3.74-3.84 (m, 1 H), 3.88 (s, 3 H), 4.27-4.45 (m, 2 H), 6.56-6.68

(m, 1 H), 6.84-6.96 (m, 1 H), 7.00-7.24 (m, 4 H), 7.35-7.48

(m, 1 H), 7.74-7.86 (m, 1 H), 8.95-9.11 (m, 2 H), 10.84-10.94

(m,

(C₂₂H₂₄FN₃O₃·HCl·0.30H₂O) C, H, N.

1 H); MS (ES) m/z 398.1 ([M +

H]⁺). Anal.

3-{[3-(7-Chloro-1H-indol-3-yl)propyl]amino}-8-ﬂuorochromane-

5-carboxamide (51f). Using general procedure B, 29a was reacted

with 32h as described above for 33b, generating 0.73 g (79%) of

51f which was converted to the mono-HCl salt and isolated as a

1

white solid: mp 222-224 °C; H NMR (400 MHz, DMSO-d₆) δ

1.92-2.08 (m, 2H), 2.73-2.84 (m, 2H), 2.99-3.42 (m, 4H),

3.73-3.86 (m, 1H), 4.28-4.45 (m, 2H), 6.94-7.05 (m, 1H),

7.07-7.29 (m, 4H), 7.36-7.47 (m, 1H), 7.49-7.58 (m, 1H),

7.73-7.86 (m, 1H), 8.94-9.11 (m, 2H), 11.16-11.26 (m, 1H);

MS (ES) m/z 402.1 ([M + H]⁺). Anal. (C₂₁H₂₁ClFN₃O₂·HCl) C,

H, N.

(R)-(-)-35c was also prepared according to procedures B and

C using (R)-(+)-29a as starting material, and X-ray analysis further

conﬁrmed the stereochemistry of this ﬁnal target.

General Procedure D (Amide Functionalization). 3-{Cyclobu-

tyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuoro-N-methyl-

chromane-5-carboxamide (35f). To 39 (0.1 g, 0.227 mmol) in

anhydrous THF (8 mL) was added 1-(3-dimethylaminopropyl)-3-

ethylcarbodiimide hydrochloride (EDC, 0.087 g, 0.454 mmol),

1-hydroxybenzotriazole hydrate (HOBt, 0.061 g, 0.454 mmol), and

methylamine (2 M/THF, 0.45 mL, 0.908 mmol). The mixture was

stirred at room temperature overnight. It was concentrated and the

residue taken up in CH₂Cl₂/H₂O. The aqueous layer was extracted

one more time with CH₂Cl₂, and the organic extracts were treated

with brine, dried over anhydrous MgSO₄, ﬁltered, and concentrated.

Chromatography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)

afforded 0.1 g (97%) of 35f. Conversion to the mono-HCl salt

generated a white solid: mp 132 °C/dec; ¹H NMR (400 MHz,

DMSO-d₆) δ 1.49-1.78 (m, 2H), 1.91-2.44 (m, 5H), 2.63-2.81

(m, 5H), 3.05-3.23 (m, 2H), 3.32 (s, 3H), 3.85-4.16 (m, 2H),

4.30-4.44 (m, 1H), 4.49-4.64 (m, 1H), 6.85-6.95 (m, 1H),

7.03-7.12 (m, 1H), 7.15-7.27 (m, 2H), 7.28-7.37 (m, 2H),

8.21-8.37 (m, 1H), 10.29-10.60 (m, 1H), 10.93 (s, 1H); MS (ES)

m/z 454.2 ([M + H]⁺). Anal. (C₂₆H₂₉F₂N₃O₂·1.2HCl·0.25H₂O)

C, H, N.

3-[3-(5,7-Diﬂuoro-1H-indol-3-yl)propylamino]-8-ﬂuorochroman-

5-carboxylic Acid Amide (51g). Using general procedure B, 29a

was reacted with 32i as described above for 33b, generating 0.70 g

1

(99%) of racemic 51g as an off-white foam. H NMR (500 MHz,

DMSO-d₆) δ 1.66-1.82 (m, 2H), 2.59-2.76 (m, 6H), 2.93-3.14

(m, 2H), 3.74-3.89 (m, 1H), 4.19-4.28 (m, 1H), 6.83-6.98 (m,

2H), 7.00-7.19 (m, 2H), 7.20-7.39 (m, 2H), 7.63-7.75 (m, 1H),

11.29-11.39 (m, 1H); MS (ES) m/z 404.1 ([M + H]⁺). The

enantiomers were separated by SFC using a Chiralpak AS column

(25 cm × 2 cm) and 40% MeOH (0.1% DEA)/CO₂(100 bar) as

mobile phase. Both enantiomers were isolated as white foam: (+)-

51g, MS (ES) m/z 404.2 ([M + H]⁺), [R]²_D⁵+19.8° (c 1.0, DMSO);

(-)-51g, MS (ES) m/z 402.3 ([M - H]^-), [R]_D²⁵-21.8° (c 1.0,

DMSO).

(3R)-(-)-8-Fluoro-3-{[(6-ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-

3-yl)methyl]amino}chromane-5-carboxamide (57a) and (3R)-(+)-

8-Fluoro-3-{[(6-ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-3-

yl)methyl]amino}chromane-5-carboxamide (57b). Using general

procedure B, 57a and 57b were prepared by reaction of (R)-(+)-

29a with 50. The diastereomers were separated by chiral HPLC

using a Chiracel AD column (2 cm × 25 cm) and 30% Hex/EtOH

as mobile phase, isolated as free base, and converted to the mono-

HCl salt to generate 57a and 57b as white solids. 57a: mp 197-202

°C; ¹H NMR (400 MHz, DMSO-d₆) δ 1.51-1.68 (m, 1H),

2.04-2.43 (m, 3H), 2.63-2.94 (m, 3H), 3.01-3.51 (m, 4H),

3.75-3.90 (m, 1H), 4.35-4.53 (m, 2H), 6.74-6.86 (m, 1H),

6.97-7.07 (m, 1H), 7.08-7.27 (m, 3H), 7.38-7.48 (m, 1H),

7.76-7.88 (m, 1H), 9.01-9.27 (m, 2H), 10.83 (s, 1H); MS (ES)

Final Targets. N-[3-(5-Fluoro-1H-indol-3-yl)propyl]-5-methoxy-

N-propylchroman-3-amine (34a). 34a was obtained according to

procedure B using 33a and propionaldehyde. Chromatography (2%

MeOH/CH₂Cl₂) afforded 0.44 g (100%) of 34a as a yellow oil.

Conversion to the mono-HCl salt generated an off-white solid: mp

1

117-120 °C; H NMR (400 MHz, DMSO-d₆) δ 0.81-0.94 (m,

3H), 1.58-1.79 (m, 2H), 1.97-2.16 (m, 2H), 2.63-2.77 (m, 2H),

2.82-2.95 (m, 1H), 2.95-3.45 (m, 5H), 3.79 (s, 3H), 3.84-3.95

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6997

(m, 1H), 4.16-4.29 (m, 1H), 4.38-4.57 (m, 1H), 6.45-6.52 (m,

1H), 6.56-6.65 (m, 1H), 6.86-6.96 (m, 1H), 7.07-7.17 (m, 1H),

7.20-7.39 (m, 3H), 10.39-10.57 (m, 1H), 10.96 (s, 1H); MS (ESI)

m/z 397 ([M + H]⁺). Anal. (C₂₄H₂₉FN₂O₂·HCl·0.50H₂O) C, H,

N (N calcd 6.34, found 5.89).

chromane-5-carboxamide ((S)-(+)-34c). The enantiomers of 34c

were separated by chiral HPLC on a Chiralpak AS column (2 cm

× 25 cm) using (1:1) hexane-EtOH as mobile phase. They were

isolated and characterized as mono-HCl salts. (R)-(-)-34c (white

solid): mp 126 °C/dec; ¹H NMR (500 MHz, DMSO-d₆) δ

0.77-0.94 (m, 3H), 1.56-1.76 (m, 2H), 1.97-2.19 (m, 2H),

2.66-2.80 (m, 2H), 3.00-3.56 (m, 6H), 3.90-4.04 (m, 1H),

4.35-4.49 (m, 1H), 4.50-4.63 (m, 1H), 6.79-7.00 (m, 1H),

7.11-7.28 (m, 3H), 7.29-7.36 (m, 2H), 7.39-7.53 (m, 1H),

7.77-7.94 (m, 1H), 10.03 (s, 1H), 10.94 (s, 1H); MS (ESI) m/z

428 ([M + H]⁺); [R]²_D⁵-31.49° (c 1.0, DMSO); >99.9% ee by

chiral HPLC. Anal. (C₂₄H₂₇F₂N₃O₂·1.20HCl·0.25H₂O) C, H, N.

N-Cyclobutyl-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-methoxy-

chroman-3-amine (35a). 35a was obtained according to procedure

C using 33a and cyclobutanone. Chromatography (9:1:1 hexanes-

EtOAc-MeOH (1% NH₄OH)) afforded 0.17 g (65%) of 35a.

Conversion to the mono-HCl salt generated an off-white solid: mp

70.5 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ 1.51-1.79 (m, 2H),

1.92-2.24 (m, 4H), 2.26-2.48 (m, 2H), 2.58-2.76 (m, 2H),

2.75-3.24 (m, 4H), 3.67-3.91 (m, 4H), 3.97-4.27 (m, 2H),

4.33-4.47 (m, 1H), 6.43-6.51 (m, 1H), 6.57-6.65 (m, 1H),

6.85-6.96 (m, 1H), 7.08-7.15 (m, 1H), 7.16-7.25 (m, 1H),

7.26-7.36 (m, 2H), 10.44-10.62 (m, 1H), 10.92 (s, 1H); MS (ESI)

m/z 407 ([M - H]^-). Anal.(C₂₅H₂₉FN₂O₂·1.20 HCl ·0.40C₄H₈O₂)

C, H, N.

1

(S)-(+)-34c (white solid): mp 126 °C/dec; H NMR (500 MHz,

DMSO-d₆) δ 0.80-0.96 (m, 3H), 1.58-1.73 (m, 2H), 2.01-2.16

(m, 2H), 2.65-2.77 (m, 2H), 3.03-3.50 (m, 6H), 3.91-4.08 (m,

1H), 4.33-4.49 (m, 1H), 4.51-4.68 (m, 1H), 6.85-6.97 (m, 1H),

7.11-7.27 (m, 3H), 7.28-7.38 (m, 2H), 7.41-7.50 (m, 1H),

7.76-7.96 (m, 1H), 10.12 (s, 1H), 10.95 (s, 1H); MS (ESI) m/z

426 ([M - H]^-); [R]²_D⁵+30.67° (c 1.0, DMSO); >99.9% ee by

chiral HPLC. Anal. (C₂₄H₂₇F₂N₃O₂·1.20HCl·0.20H₂O) C, H, N.

8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl](propyl)amino}-

N-methylchromane-5-carboxamide (34f). 34f was obtained accord-

ing to procedure B using 33f and propionaldehyde. Chromatography

(6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded 0.097 g

(89%) of 34f. Conversion to the mono-HCl salt generated a white

solid: mp 123 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ 0.78-0.96

(m, 3H), 1.57-1.73 (m, 2H), 1.98-2.13 (m, 2H), 2.64-2.82 (m,

5H), 3.01-3.45 (m, 6H), 3.91-4.04 (m, 1H), 4.38-4.48 (m, 1H),

4.51-4.66 (m, 1H), 6.85-6.96 (m, 1H), 7.04-7.13 (m, 1H),

7.15-7.27 (m, 2H), 7.29-7.38 (m, 2H), 8.24-8.36 (m, 1H),

9.87-10.03 (m, 1H), 10.94 (s, 1H); MS (ES) m/z 440.2 ([M -

H]^-). Anal. (C₂₅H₂₉F₂N₃O₂·1HCl·0.25H₂O) C, H, N.

3-{[3-(5-Fluoro-1H-indol-3-yl)propyl](propyl)amino}chromane-

5-carboxamide (34e). 34e was obtained according to procedure B

using 33e and propionaldehyde. Without further puriﬁcation, it was

converted to the mono-HCl salt, generating 34e as a colorless oil:

¹H NMR (400 MHz, DMSO-d₆) δ 0.87 (t, J ) 7.2 Hz, 3H),

1.64-1.75 (m, 2H), 1.99-2.15 (m, 2H), 2.64-2.77 (m, 2H),

3.01-3.51 (m, 6H), 3.86-3.99 (m, 1H), 4.28-4.37 (m, 1H),

4.45-4.56 (m, 1H), 6.85-7.01 (m, 2H), 7.07-7.16 (m, 1H),

7.17-7.28 (m, 2H), 7.29-7.37 (m, 2H), 7.38-7.47 (m, 1H),

7.77-7.88 (m, 1H), 10.16-10.30 (m, 1H), 10.95 (s, 1H); MS (ES)

m/z 410.2 ([M + H]⁺).

(3R)-(-)-N-Cyclobutyl-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-

methoxychroman-3-amine ((R)-(-)-35a). (R)-(-)-35a was obtained

according to procedure C using (R)-33a (synthesized from (R)-

(-)-19) and cyclobutanone. Chromatography (1:1 hexanes-EtOAc)

afforded 0.15 g (69%) of (R)-(-)-35a. Conversion to the mono-

HCl salt generated an off-white solid: mp 115 °C/dec; [R]²_D⁵-33.02°

1

(c 1.0, DMSO); H NMR (500 MHz, DMSO-d₆) δ1.52-1.79 (m,

2H), 1.91-2.45 (m, 6H), 2.58-2.76 (m, 2H), 2.78-3.23 (m, 4H),

3.68-3.90 (m, 4H), 3.98-4.26 (m, 2H), 4.32-4.48 (m, 1H),

6.44-6.51 (m, 1H), 6.57-6.65 (m, 1H), 6.86-6.94 (m, 1H),

7.09-7.16 (m, 1H), 7.17-7.25 (m, 1H), 7.26-7.36 (m, 2H),

10.43-10.58 (m, 1H), 10.94 (s, 1H); MS (ES) m/z 409.2 ([M +

H]⁺). Anal. (C₂₅H₂₉FN₂O₂·HCl·0.50H₂O) C, H, N.

8-Fluoro-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-methoxy-N-pro-

pylchroman-3-amine (34b). 34b was obtained according to proce-

dure B using 33b and propionaldehyde. Chromatography (4:1)

EtOAc-hexanes afforded 0.16 g (96%) of 34b. Conversion to the

mono-HCl salt generated a white solid: ¹H NMR (400 MHz,

DMSO-d₆) δ 0.87 (t, J ) 7.1 Hz, 3H), 1.60-1.77 (m, 2H),

2.02-2.16 (m, 2H), 2.65-2.78 (m, 2H), 2.88-3.00 (m, 1H),

3.01-3.31 (m, 5H), 3.77 (s, 3H), 3.87-4.00 (m, 1H), 4.26-4.39

(m, 1H), 4.52-4.66 (m, 1H), 6.48-6.59 (m, 1H), 6.84-6.96 (m,

1H), 7.02-7.13 (m, 1H), 7.21-7.29 (m, 1H), 7.29-7.39 (m, 2H),

10.59-10.81 (m, 1H), 10.97 (s, 1H); MS (ESI) m/z 415 ([M +

H]⁺). Anal. (C₂₄H₂₈F₂N₂O₂·1HCl·0.75H₂O) C, H, N.

(3R)-(-)-8-Fluoro-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-meth-

oxy-N-propylchroman-3-amine ((R)-(-)-34b). (R)-(-)-34b was

obtained according to procedure B using (R)-33b. Chromatography

(4:1) EtOAc-hexanes afforded 0.026 g (78%) of (R)-(-)-34b.

Conversion to the mono-HCl salt generated a white solid: MS

(APCI) m/z 415 ([M + H]⁺); [R]²_D⁵-30.2° (c 1.0, DMSO). Anal.

(C₂₄H ₂₈F₂N₂O₂·1HCl) C, H, N.

8-Chloro-3-{cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

chromane-5-carboxamide (35d). 35d was obtained according to

procedure C using 33d and cyclobutanone. Chromatography (5%

MeOH in EtOAc) afforded 0.13 g (89%) of 35d as a colorless oil:

¹H NMR (400 MHz, DMSO-d₆) δ 1.45-1.60 (m, 2H), 1.66-1.77

(m, 2H), 1.78-1.97 (m, 3H), 2.55-2.69 (m, 4H), 2.81-3.02 (m,

2H), 3.04-3.16 (m, 1H), 3.26-3.43 (m, 1H), 3.85-3.96 (m, 1H),

3.98-4.08 (m, 1H), 4.24-4.35 (m, 1H), 6.82-6.98 (m, 2H),

7.15-7.34 (m, 4H), 7.37-7.46 (m, 1H), 7.66-7.82 (m, 1H), 10.82

(s, 1H); MS (ES) m/z 456.1 ([M + H]⁺).

(3S)-(+)-8-Fluoro-N-[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-5-meth-

oxy-N-propylchroman-3-amine ((S)-(+)-34b). (S)-(+)-34b was

obtained according to procedure B using (S)-33b. Chromatography

(4:1) EtOAc-hexanes afforded 0.12 g (94%) of (S)-(+)-34b.

Conversion to the mono-HCl salt generated a white solid: [R]_D²⁵

+32.2° (c 1.0, DMSO). Anal. (C₂₄H₂₈F₂N₂O₂·1HCl) C, H, N.

8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl](propyl)amino}-

chromane-5-carboxamide (34c). 34c was obtained according to

procedure B using 33c and propionaldehyde. Chromatography (5:

4:1 EtOAc-hexanes-MeOH (1% NH₄OH)) afforded 0.12 g (90%)

of 34c as a sticky gum. Conversion to the mono-HCl salt generated

a pale-yellow solid: mp 125 °C/dec; ¹H NMR (500 MHz, DMSO-

d₆) δ 0.75-0.95 (m, 3H), 1.56-1.72 (m, 2H), 1.96-2.16 (m, 2H),

2.66-2.78 (m, 2H), 3.00-3.53 (m, 6H), 3.92-4.04 (m, 1H),

4.34-4.49 (m, 1H), 4.50-4.68 (m, 1H), 6.82-6.96 (m, 1H),

7.09-7.38 (m, 5H), 7.39-7.51 (m, 1H), 7.75-7.98 (m, 1H), 10.11

(s, 1H), 10.95 (s, 1H); MS (ESI) m/z 426 ([M - H]^-). Anal.

(C₂₄H₂₇F₂N₃O₂·1HCl·0.20H₂O) C, H, N.

(-)-8-Chloro-3-{cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-

amino}chromane-5-carboxamide ((-)-35d) and (+)-8-Chloro-3-

{cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-5-

carboxamide ((+)-35d). The compounds were prepared according

to procedure C using each enantiomer of 33d separately and

cyclobutanone, generating (-)-35d and (+)-35d. (-)-35d: mp

1

104-107 °C; H NMR (500 MHz, DMSO-d₆) δ 1.53-1.75 (m,

2H), 1.93-2.25 (m, 2H), 2.29-2.50 (m, 3H), 2.62-2.74 (m, 2H),

3.05-3.25 (m, 2H), 3.27-3.50 (m, 3H), 3.86-4.12 (m, 2H),

4.31-4.50 (m, 1H), 4.57-4.73 (m, 1H), 6.86-6.94 (m, 1H),

7.07-7.13 (m, 1H), 7.20-7.27 (m, 1H), 7.28-7.36 (m, 2H),

7.36-7.42 (m, 1H), 7.47-7.54 (m, 1H), 7.82-7.94 (m, 1H), 10.96

(s, 1H); MS (ES) m/z 456.2 ([M + H]⁺); [R]_D²⁵-37.8° (c 1.0,

DMSO). (+)-35d: mp 190-194 °C; MS (ES) m/z 456.2 ([M +

25

H]⁺); [R] +37.2° (c 1.0, DMSO).

D

(3R)-(-)-8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl](pro-

pyl)amino}chromane-5-carboxamide ((R)-(-)-34c) and (3S)-

(+)-8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl](propyl)amino}-

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chromane-

5-carboxamide (35e). 35e was obtained according to procedure C

using 33e and cyclobutanone to generate 35e: mp 102-5 °C; H

1

6998 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

NMR (400 MHz, DMSO-d₆) δ 1.87-2.46 (m, 6H), 2.59-2.76 (m,

2H), 3.01-3.46 (m, 6H), 3.79-3.92 (m, 1H), 3.95-4.11 (m, 1H),

4.21-4.35 (m, 1H), 4.37-4.49 (m, 1H), 6.85-6.97 (m, 2H),

7.08-7.15 (m, 1H), 7.17-7.25 (m, 2H), 7.26-7.36 (m, 2H),

7.40-7.48 (m, 1H), 7.75-7.85 (m, 1H), 9.94-10.27 (m, 1H), 10.93

(s, 1H); MS (ES) m/z 422.3 ([M + H]⁺).

δ 1.54-1.76 (m, 2H), 1.92-2.25 (m, 3H), 2.25-2.46 (m, 2H),

2.59-2.76 (m, 2H), 3.05-3.28 (m, 3H), 3.36-3.60 (m, 2H),

3.78-3.87 (m, 3H), 3.89-4.14 (m, 2H), 4.33-4.45 (m, 1H),

4.49-4.63 (m, 1H), 6.84-6.97 (m, 1H), 7.15-7.39 (m, 4H),

7.49-7.62 (m, 1H), 10.36-10.60 (m, 1H), 10.85-11.00 (m, 1H);

MS (ES) m/z 455.2 ([M + H]⁺). Anal. (C₂₆H₂₈F₂N₂O₃·HCl) C,

H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuo-

rochromane-5-carboxylic Acid (39). To 38 (0.25 g, 0.558 mmol)

in absolute EtOH (3.5 mL) was added 2.5 N NaOH/H₂O (0.3 mL,

0.781 mmol). The mixture was brought to reﬂux and kept under

reﬂux for 50 min. It was concentrated, and the residue was taken

up in CH₂Cl₂/H₂O. The organic layer was separated and the aqueous

layer made acidic with 2 N HCl/H₂O. The aqueous layer was then

extracted with CH₂Cl₂(3×) followed by EtOAc (1×). The organic

extracts were pooled, dried over anhydrous MgSO₄, ﬁltered, and

concentrated to generate 0.216 g (88%) of 39 as an off-white solid

which was converted to a di-HCl salt: mp 140 °C/dec; MS (ES)

m/z 441.3 ([M + H]⁺). Anal. (C ₂₅H₂₆F₂N₂O₃·2HCl·1.5H₂O) C,

H, N.

tert-Butyl 3-{3-[{(3R)-5-[Bis(tert-butoxycarbonyl)carbamoyl]-8-

ﬂuoro-3,4-dihydro-2H-chromen-3-yl}(propyl)amino]propyl}-5-ﬂuoro-

1H-indole-1-carboxylate ((R)-41). To (R)-34c (which was synthe-

sized from (R)-33c) (0.77 g, 1.8 mmol) in anhydrous CH₂Cl₂(10

mL), under nitrogen at room temperature, was added N,N-

dimethylaminopyridine (DMAP, 0.22 g, 1.8 mmol) and di-tert-butyl

dicarbonate (1.45 mL, 6.3 mmol) in CH₂Cl₂(10 mL). The mixture

was stirred at room temperature for 1.5 h. It was concentrated and

puriﬁed by chromatography (7:1 hexanes-EtOAc) to generate

1.17 g (89%) of (R)-41 as a foamy white solid: LC-MS m/z 728.25

([M + H]⁺); ¹H NMR (400 MHz, DMSO-d₆) δ 0.78-0.88 (t, 3H),

1.28 (s, 18H), 1.30-1.40 (m, 2H), 1.40-1.48 (m, 1H), 1.58 (s,

9H), 1.65-1.80 (m, 2H), 2.55-2.68 (m, 4H), 2.75-3.15 (m, 4H),

3.89-4.05 (m, 1H), 4.30-4.40 (m, 1H), 7.00-7.25 (m, 3H),

7.30-7.40 (m, 1H), 7.47 (s, 1H), 7.92-8.05 (m, 1H).

(-)-3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}chro-

mane-5-carboxamide ((-)-35e) and (+)-3-{Cyclobutyl[3-(5-ﬂuoro-

1H-indol-3-yl)propyl]amino}chromane-5-carboxamide ((+)-35e).

The compounds weres obtained according to procedure C using

either (-)-33e or (+)-33e and cyclobutanone to generate (-)-35e

and (+)-35e. (-)-35e: MS (ES) m/z 422.2 ([M + H]⁺); [R]_D²⁵

-68.4° (c 1.0, DMSO). (+)-35e: MS (ES) m/z 422.2 ([M + H]⁺);

25

D

[R] +69.4° (c 1.0, DMSO).

Methyl (3R)-3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-

amino}-8-ﬂuorochromane-5-carboxylate ((R)-38). To (R)-33c (male-

ate salt) (0.85 g, 1.69 mmol) in MeOH (20 mL) was added

concentrated sulfuric acid (0.5 mL). The mixture was stirred under

reﬂux overnight. It was cooled to room temperature and extracted

with CH₂Cl₂from 2.5 N NaOH. The organic extracts were dried

over anhydrous sodium sulfate and concentrated. It was puriﬁed

by chromatography (0-10% MeOH-CH₂Cl₂) to generate 0.45 g

(66%) of methyl (3R)-8-ﬂuoro-3-{[3-(5-ﬂuoro-1H-indol-3-

yl)propyl]amino}chromane-5-carboxylate as a white solid. To this

methyl ester (0.45 g, 1.12 mmol) in MeOH (10 mL) was added

acetic acid (0.3 mL), cyclobutanone, and sodium cyanoborohydride

(0.16 g, 2.54 mmol). The mixture was stirred at room temperature

over the weekend. The mixture was diluted with CH₂Cl₂, extracted

from saturated NaHCO₃/NaOH (pH 10), dried over anhydrous

sodium sulfate, and concentrated. It was puriﬁed by chromatography

(0-15% MeOH-CH₂Cl₂) to afford 0.55 g (100%) of (R)-38 as a

colorless oil.

(3R)-(-)-3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

8-ﬂuoro-N-methylchromane-5-carboxamide ((R)-(-)-35f). To (R)-

38 (0.1 g, 0.22 mmol) in MeOH (1 mL)-THF (3 mL) was added

1 M LiOH/H₂O (1 mmol, 1 mL), and the mixture was stirred at

room temperature for 2 days. It was concentrated to afford crude

(3R)-3-{cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuo-

rochromane-5-carboxylic acid (0.12 g). This crude acid was

dissolved in THF (10 mL) and treated with EDC (0.15 g, 0.78

mmol) and HOBt (0.156 g, 1.15 mmol) followed by addition of 2

M MeNH₂/THF (0.5 mL, 1 mmol). The mixture was stirred at room

temperature for 5 days. The mixture was then diluted with EtOAc,

extracted from brine, dried over anhydrous sodium sulfate, and

concentrated. Preparative HPLC on a Luna 5 µm C₁₈column (150

mm × 21.2 mm) using a gradient (5-50% CH₃CN/H₂O (0.075%

TFA)) afforded (R)-(-)-35f which was then converted to the mono-

tert-Butyl 5-Fluoro-3-(3-{[(3R)-8-ﬂuoro-5-(methylcarbamoyl)-

3,4-dihydro-2H-chromen-3-yl](propyl)amino}propyl)-1H-indole-1-

carboxylate ((R)-(-)-42) and tert-Butyl 3-{3-[{(3R)-5-[(tert-Butox-

ycarbonyl)carbamoyl]-8-ﬂuoro-3,4-dihydro-2H-chromen-3-

yl}(propyl)amino]propyl}-5-ﬂuoro-1H-indole-1-carboxylate ((R)-

(-)-43). To (R)-41 (1.16 g, 1.59 mmol) in anhydrous THF (15 mL),

under nitrogen at room temperature, was added 2 M MeNH₂/THF

(0.95 mL, 1.9 mmol). The mixture was stirred at room temperature

overnight. It was concentrated and puriﬁed by chromatography (3:1

hexanes-EtOAc to EtOAc) to generate 0.62 g (62%) of (R)-(-)-

43 as a white foamy solid: mp 58-63 °C; [R]²_D⁵-46.6° (c 1.0,

1

DMSO); H NMR (400 MHz, DMSO-d₆) δ 0.76-0.89 (m, 3H),

1

HCl salt: H NMR (400 MHz, MeOH-d₄) δ 2.09-2.31 (m, 2H),

1.38 (s, 11H), 1.61 (s, 9H), 1.67-1.79 (m, 2H), 2.56-2.69 (m,

4H), 2.79-2.94 (m, 2H), 3.03-3.14 (m, 1H), 3.27-3.37 (m, 2H),

3.88-4.00 (m, 1H), 4.26-4.37 (m, 1H), 6.90-7.01 (m, 1H),

7.02-7.20 (m, 2H), 7.35-7.43 (m, 1H), 7.46-7.53 (m, 1H),

7.95-8.07 (m, 1H), 10.58-10.69 (m, 1H); MS (ES) m/z 628.3 ([M

+ H]⁺). Also generated was 0.26 g (30%) of (R)-(-)-42 as a white

foamy solid: mp 58-63 °C; [R]²_D⁵-42.8° (c 1.0, DMSO); ¹H NMR

(400 MHz, DMSO-d₆) δ 0.77-0.89 (m, 3H), 1.30-1.45 (m, 2H),

1.61 (s, 9H), 1.68-1.80 (m, 2H), 2.56-2.67 (m, 4H), 2.72 (d, J )

4.4 Hz, 3H), 2.85-2.97 (m, 2H), 3.00-3.14 (m, 1H), 3.25-3.38

(m, 2H), 3.88-4.00 (m, 1H), 4.24-4.36 (m, 1H), 6.85-6.94 (m,

1H), 7.02-7.11 (m, 1H), 7.11-7.20 (m, 1H), 7.37-7.44 (m, 1H),

7.48-7.52 (m, 1H), 7.97-8.07 (m, 1H), 8.10-8.19 (m, 1H); MS

(ES) m/z 542.2 ([M + H]⁺).

2.42-2.84 (m, 7H), 3.15-3.29 (m, 2H), 3.56-3.70 (m, 2H),

3.71-3.89 (m, 5H), 4.33-4.57 (m, 2H), 4.70-4.93 (m, 2H),

7.24-7.36 (m, 1H), 7.44-7.57 (m, 3H), 7.60-7.68 (m, 1H),

7.69-7.76 (m, 1H); MS (APPI) m/z 400.0 ([M + H]⁺); [R]_D²⁵

-21.6° (c 1.0, DMSO).

Methyl 8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

chromane-5-carboxylate (37). Using general procedure B, 36 was

reacted with 32a in the presence of acetic acid and NaBH₃CN.

Chromatography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH))

afforded 0.46 g (91%) of 37 as a white solid which was converted

to the mono-HCl salt: mp 219 °C/dec; ¹H NMR (500 MHz, DMSO-

d₆) δ 1.94-2.06 (m, 2H), 2.70-2.80 (m, 2H), 3.06-3.16 (m, 2H),

3.29-3.37 (m, 1H), 3.44-3.55 (m, 1H), 3.77-3.90 (m, 4H),

4.31-4.39 (m, 1H), 4.42-4.53 (m, 1H), 6.83-6.98 (m, 1H),

7.18-7.38 (m, 4H), 7.50-7.62 (m, 1H), 8.86-9.08 (m, 2H), 10.95

(s, 1H); MS (ES) m/z 401.2 ([M + H]⁺). Anal. (C₂₂H₂₂F₂N₂O₃·

1HCl) C, H, N.

tert-Butyl 3-{3-[{(3R)-5-[(tert-Butoxycarbonyl)(methyl)carbam-

oyl]-8-ﬂuoro-3,4-dihydro-2H-chromen-3-yl}(propyl)amino]propyl}-

5-ﬂuoro-1H-indole-1-carboxylate ((R)-44). To NaH (60% suspen-

sion in oil, 9.6 mg, 0.24 mmol) in anhydrous THF (3 mL), under

nitrogen at room temperature, was added (R)-(-)-43 (0.1 g, 0.16

mmol) in anhydrous THF (3 mL). The mixture was stirred at room

temperature for 30 min. Iodomethane (0.012 mL, 0.19 mmol) was

added and the mixture stirred at room temperature overnight. It

was quenched with H₂O and concentrated and the residue taken

up in EtOAc/H₂O. The aqueous layer was extracted once more with

Methyl 3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-

8-ﬂuorochromane-5-carboxylate (38). Using general procedure C,

37 was reacted with cyclobutanone in the presence of acetic acid

and NaBH₃CN. Chromatography (2:1 hexanes-EtOAc) afforded

0.33 g (81%) of 38 as a white solid which was converted to the

1

mono-HCl salt: mp 110 °C/dec; H NMR (500 MHz, DMSO-d₆)

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6999

EtOAc, and the organic extracts werre pooled, treated with brine,

dried over anhydrous MgSO₄, ﬁltered, and concentrated. Chroma-

tography (4:1 hexanes-EtOAc) afforded 61 mg (60%) of (R)-44

as a white gummy solid: MS (ES) m/z 642 ([M + H]⁺); ¹H NMR

(400 MHz, DMSO-d₆) δ 0.75-0.88 (t, 3H), 1.04 (s, 9H), 1.28-1.45

(m, 2H), 1.69 (s, 9H), 1.65-1.80 (m, 2H), 2.53-2.80 (m, 6H),

2.88-3.10 (m, 2H), 3.14 (s, 3H), 3.28-3.38 (m, 1H), 3.80-3.97

(m, 1H), 4.28-4.40 (m, 1H), 6.70-6.82 (m, 1H), 7.00-7.20 (m,

2H), 7.30-7.42 (m, 1H), 7.47 (s, 1H), 7.92-8.05 (m, 1H).

(3R)-(-)-8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)propyl](propyl)-

amino}-N-methylchromane-5-carboxamide ((R)-(-)-34f). To either

(R)-(-)-42 or (R)-44 was added 1.25 N HCl/EtOH (25 equiv), and

the mixture was stirred at 60 °C for 4 h. It was concentrated to

generate (R)-(-)-34f (quantitative yield) as the HCl salt: mp 142

°C, dec; [R]²_D⁵-25.8° (c 1.0, DMSO); ¹H NMR (400 MHz, DMSO-

d₆) δ 0.87 (t, J ) 7.3 Hz, 3H), 1.58-1.77 (m, 2H), 2.00-2.18 (m,

2H), 2.66-2.80 (m, 6H), 3.01-3.44 (m, 5H), 3.91-4.02 (m, 1H),

4.37-4.49 (m, 1H), 4.55-4.69 (m, 1H), 6.86-6.94 (m, 1H),

7.04-7.11 (m, 1H), 7.14-7.28 (m, 2H), 7.30-7.37 (m, 2H),

8.25-8.38 (m, 1H), 10.41-10.56 (m, 1H), 10.96 (s, 1H); MS (ES)

m/z 442.2 ([M + H]⁺). Anal. (C₂₅H₂₉F₂N₃O₂·1HCl·0.8H₂O) C,

H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-N-ethyl-

8-ﬂuorochromane-5-carboxamide (40a). 40a was obtained accord-

ing to procedure D using 39 and ethylamine. Chromatography (6:

3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded 0.092 g

(86%) of 40a. Conversion to the mono-HCl salt generated a white

powder: mp 130 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ

1.04-1.17 (m, 3H), 1.52-1.78 (m, 2H), 1.89-2.47 (m, 6H),

2.59-2.76 (m, 2H), 3.02-3.43 (m, 6H), 3.83-4.17 (m, 2H),

4.31-4.45 (m, 1H), 4.46-4.62 (m, 1H), 6.85-6.96 (m, 1H),

7.02-7.11 (m, 1H), 7.15-7.27 (m, 2H), 7.28-7.37 (m, 2H),

8.24-8.46 (m, 1H), 10.13-10.51 (m, 1H), 10.92 (s, 1H); MS (ES)

m/z 468.2 ([M + H]⁺). Anal. (C₂₇H₃₁F₂N₃O₂·1.1HCl·0.25H₂O)

C, H, N.

6.99-7.08 (m, 1H), 7.13-7.27 (m, 2H), 7.28-7.37 (m, 2H),

8.34-8.46 (m, 1H), 10.30-10.61 (m, 1H), 10.95 (s, 1H); MS (ES)

m/z 480.2 ([M + H]⁺). Anal. (C₂₈H₃₁F₂N₃O₂·1HCl·0.75H₂O) C,

H, N.

N-Cyclobutyl-3-{cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]-

amino}-8-ﬂuorochromane-5-carboxamide (40e). 40e was obtained

according to procedure D using 39 and cyclobutylamine. Chroma-

tography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded

0.092 g (82%) of 40e. Conversion to the mono-HCl salt generated

1

a white solid: mp 135 °C/dec; H NMR (400 MHz, DMSO-d₆) δ

1.52-1.77 (m, 4H), 1.87-2.45 (m, 10H), 2.61-2.73 (m, 2H),

3.06-3.23 (m, 2H), 3.22-3.40 (m, 2H), 3.86-4.15 (m, 2H),

4.28-4.43 (m, 2H), 4.46-4.61 (m, 1H), 6.84-6.96 (m, 1H),

7.03-7.12 (m, 1H), 7.14-7.26 (m, 2H), 7.27-7.38 (m, 2H),

8.52-8.64 (m, 1H), 10.16-10.48 (m, 1H), 10.95 (s, 1H); MS (ES)

m/z 494.2 ([M + H]⁺). Anal. (C₂₉H₃₃F₂N₃O₂·1HCl·0.5H₂O) C,

H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-N-cyclo-

pentyl-8-ﬂuorochromane-5-carboxamide (40f). 40f was obtained

according to procedure D using 39 and cyclohexylamine. Chro-

matography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) af-

forded 0.09 g (76%) of 40f. Conversion to the mono-HCl salt

generated a white solid: mp 134 °C/dec; ¹H NMR (400 MHz,

DMSO-d₆) δ 1.02-1.39 (m, 5H), 1.50-1.88 (m, 7H), 1.88-2.48

(m, 6H), 2.61-2.76 (m, 2H), 3.04-3.24 (m, 2H), 3.25-3.46 (m,

2H), 3.62-3.78 (m, 1H), 3.84-4.14 (m, 2H), 4.30-4.44 (m, 1H),

4.47-4.63 (m, 1H), 6.84-6.97 (m, 1H), 6.98-7.08 (m, 1H),

7.13-7.27 (m, 2H), 7.28-7.39 (m, 2H), 8.12-8.24 (m, 1H),

10.27-10.62 (m, 1H), 10.95 (s, 1H); MS (ES) m/z 522.2 ([M +

H]⁺). Anal. (C₃₁H₃₇F₂N₃O₂·1.2HCl·0.25H₂O) C, H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-N-(cyclo-

propylmethyl)-8-ﬂuorochromane-5-carboxamide (40g). 40g was

obtained according to procedure D using 39 and methylcyclopro-

pylamine. Chromatography (6:3:1 hexanes-EtOAc-MeOH (1%

NH₄OH)) afforded 0.10 g (92%) of 40g. Conversion to the mono-

1

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuoro-

N-propylchromane-5-carboxamide (40b). 40b was obtained ac-

cording to procedure D using 39 and propylamine. Chromatography

(6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded 0.084 g

(77%) of 40b. Conversion to the mono-HCl salt generated an off-

white solid: mp 67 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ

0.74-0.97 (m, 3H), 1.42-1.74 (m, 4H), 1.85-2.44 (m, 6H),

2.62-2.76 (m, 2H), 3.07-3.23 (m, 4H), 3.23-3.43 (m, 2H),

3.88-4.15 (m, 2H), 4.32-4.45 (m, 1H), 4.45-4.62 (m, 1H),

6.85-6.95 (m, 1H), 7.03-7.10 (m, 1H), 7.14-7.26 (m, 2H),

7.27-7.36 (m, 2H), 8.27-8.41 (m, 1H), 10.18-10.48 (m, 1H),

10.94 (s, 1H); MS (ES) m/z 482.2 ([M + H]⁺). Anal.

(C₂₈H₃₃F₂N₃O₂·1.2HCl·0.25H₂O) C, H, N.

HCl salt generated a white solid: mp 114 °C/dec; H NMR (400

MHz, DMSO-d₆) δ 0.15-0.27 (m, 2H), 0.32-0.49 (m, 2H),

0.93-1.08 (m, 1H), 1.50-1.76 (m, 2H), 1.86-2.46 (m, 7H),

2.60-2.76 (m, 2H), 3.00-3.23 (m, 4H), 3.24-3.46 (m, 1H),

3.86-4.14 (m, 2H), 4.31-4.46 (m, 1H), 4.47-4.63 (m, 1H),

6.83-6.96 (m, 1H), 7.01-7.12 (m, 1H), 7.14-7.27 (m, 2H),

7.26-7.36 (m, 2H), 8.38-8.52 (m, 1H), 10.25-10.60 (m, 1H),

10.94 (s, 1H); MS (ES) m/z 494.2 ([M + H]⁺). Anal.

(C₂₉H₃₃F₂N₃O₂·1HCl·0.5H₂O) C, H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuoro-

N,N-dimethylchromane-5-carboxamide (40h). 40h was obtained

according to procedure D using 39 and dimethylamine. Chroma-

tography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded

0.116 g (86%) of 40h. Conversion to the mono-HCl salt generated

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuoro-

N-isopropylchromane-5-carboxamide (40c). 40c was obtained

according to procedure D using 39 and isopropylamine. Chroma-

tography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) afforded

0.10 g (93%) of 40c. Conversion to the mono-HCl salt generated

1

an off-white solid: mp 126 °C/dec; H NMR (500 MHz, DMSO-

d₆) δ 1.52-1.78 (m, 2H), 1.88-2.24 (m, 3H), 2.23-2.47 (m, 2H),

2.58-2.81 (m, 5H), 2.87-3.24 (m, 7H), 3.26-3.45 (m, 1H),

3.90-4.15 (m, 2H), 4.35-4.62 (m, 2H), 6.81-6.87 (m, 1H),

6.87-6.94 (m, 1H), 7.13-7.25 (m, 2H), 7.26-7.36 (m, 2H),

10.42-10.57 (m, 1H), 10.88-11.01 (m, 1H); MS (ES) m/z 466.2

([M - H]^-). Anal. (C₂₇H₃₁F₂N₃O₂·1HCl·1H₂O) C, H, N.

(-)-3-[[3-(5-Cyano-1H-indol-3-yl)propyl](cyclobutyl)amino]-8-

ﬂuorochromane-5-carboxamide ((-)-52a) and (+)-3-[[3-(5-Cyano-

1H-indol-3-yl)propyl](cyclobutyl) amino]-8-ﬂuorochromane-5-

carboxamide ((+)-52a). The compounds were obtained according

to procedure C using 51a and cyclobutanone. Chromatography ((96:

4) CH₂Cl₂-MeOH (5%NH₄OH)) afforded 0.381 g (84%) of

racemic 52a as a white solid. The enantiomers were separated by

chiral HPLC using a Whelk O column (2 cm × 25 cm) and 52%

EtOH-DEA/Hex-DEA as mobile phase, isolated, and character-

ized as mono-HCl salts. (-)-52a (white solid): mp 166-171 °C/

dec; ¹H NMR (400 MHz, DMSO-d₆) δ 1.49-1.77 (m, 2H),

1.86-2.21 (m, 4H), 2.23-2.38 (m, 1H), 2.65-2.81 (m, 2H),

3.03-3.22 (m, 2H), 3.22-3.49 (m, 3H), 3.84-4.13 (m, 2H),

4.31-4.62 (m, 2H), 7.07-7.25 (m, 2H), 7.28-7.55 (m, 4H),

7.76-7.88 (m, 1H), 8.09-8.17 (m, 1H), 10.38-10.67 (m, 1H),

1

a white solid: mp 127 °C/dec; H NMR (400 MHz, DMSO-d₆) δ

1.05-1.19 (m, 6H), 1.52-1.77 (m, 2H), 1.88-2.47 (m, 6H),

2.60-2.77 (m, 2H), 3.06-3.24 (m, 2H), 3.24-3.44 (m, 2H),

3.87-4.16 (m, 3H), 4.32-4.45 (m, 1H), 4.47-4.63 (m, 1H),

6.85-6.95 (m, 1H), 7.00-7.09 (m, 1H), 7.14-7.27 (m, 2H),

7.28-7.38 (m, 2H), 8.12-8.25 (m, 1H), 10.24-10.59 (m, 1H),

10.94 (s, 1H); MS (ES) m/z 482.2 ([M + H]⁺). Anal.

(C₂₈H₃₃F₂N₃O₂·1HCl·0.5H₂O) C, H, N.

3-{Cyclobutyl[3-(5-ﬂuoro-1H-indol-3-yl)propyl]amino}-N-cyclo-

propyl-8-ﬂuorochromane-5-carboxamide (40d). 40d was obtained

according to procedure D using 39 and cyclopropylamine. Chro-

matography (6:3:1 hexanes-EtOAc-MeOH (1% NH₄OH)) af-

forded 0.092 g (85%) of 40d. Conversion to the mono-HCl salt

generated a white solid: mp 134 °C/dec; ¹H NMR (400 MHz,

DMSO-d₆) δ 0.48-0.59 (m, 2H), 0.63-0.74 (m, 2H), 1.51-1.77

(m, 2H), 1.89-2.47 (m, 6H), 2.61-2.75 (m, 2H), 2.76-2.89 (m,

1H), 3.03-3.23 (m, 2H), 3.24-3.41 (m, 2H), 3.84-4.16 (m, 2H),

4.30-4.44 (m, 1H), 4.48-4.62 (m, 1H), 6.86-6.95 (m, 1H),

7000 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

11.41-11.51 (m, 1H); MS (ES) m/z 445.2 ([M - H]^-); [R]_D²⁵

-27.8° (c 1.0, DMSO); 99.4% ee by chiral HPLC. Anal.

(C₂₆H₂₇FN₄O₂·HCl·1.3H₂O) C, H, N. (+)-52a (white solid): mp

generating a 70% yield of each enantiomer. (-)-52g: mp 59-61

°C, ¹H NMR (400 MHz, DMSO-d₆) δ 1.42-1.59 (m, 2H),

1.62-1.76 (m, 2H), 1.76-1.97 (m, 4H), 2.54-2.70 (m, 4H),

2.82-3.00 (m, 2H), 3.03-3.15 (m, 1H), 3.24-3.42 (m, 1H),

3.78-3.93 (m, 1H), 4.17-4.29 (m, 1H), 6.83-7.16 (m, 4H),

7.21-7.36 (m, 2H), 7.68 (s, 1H), 11.33 (s, 1H); MS (ES) m/z

1

170-177 °C/dec; H NMR (400 MHz, DMSO-d₆) δ 1.53-1.76

(m, 2H), 2.00-2.22 (m, 4H), 2.25-2.43 (m, 2H), 2.68-2.81 (m,

2H), 3.06-3.24 (m, 2H), 3.29-3.45 (m, 2H), 3.85-4.12 (m, 2H),

4.31-4.62 (m, 2H), 7.07-7.23 (m, 2H), 7.29-7.55 (m, 4H),

7.76-7.89 (m, 1H), 8.12 (s, 1H), 10.47-10.76 (m, 1H), 11.40-11.52

(m, 1H); MS (ES) m/z 445.2 ([M - H]^-); [R]_D²⁵+19.0° ( c 1.0,

DMSO); 99.5% ee by chiral HPLC. Anal. (C₂₆H₂₇FN₄O₂·

HCl·0.80H₂O) C, H, N.

3-{Cyclobutyl[3-(5-methoxy-1H-indol-3-yl)propyl]amino}-8-ﬂuo-

rochromane-5-carboxamide (52b). 52b was obtained according to

procedure C using 51b and cyclobutanone, generating after chro-

matography 0.13 g (93%) of 52b which was converted to the mono-

HCl salt: mp 151-158 °C/dec; ¹H NMR (400 MHz, DMSO-d₆) δ

1.48-1.78 (m, 2H), 1. 86-2.44 (m, 6H), 2.59-2.75 (m, 2H),

3.04-3.21 (m, 2H), 3.31 (s, 2H), 3.75 (s, 3H), 3.85-4.14 (m, 2H),

4.28-4.44 (m, 1H), 4.47-4.62 (m, 1H), 6.67-6.74 (m, 1H),

6.97-7.03 (m, 1H), 7.04-7.26 (m, 4H), 7.35-7.51 (m, 1H),

7.71-7.93 (m, 1H), 10.32-10.76 (m, 2H); MS (ES) m/z 452.3 ([M

+ H]⁺).

3-[[3-(5-Chloro-1H-indol-3-yl)propyl](cyclobutyl)amino]-8-ﬂuo-

rochromane-5-carboxamide (52c). 52c was obtained according to

procedure C using 51c and cyclobutanone, generating 0.135 g (96%)

of 52c which was converted to the mono-HCl salt and isolated as

a white solid: mp 152-160 °C/dec; ¹H NMR (400 MHz, DMSO-

d₆) δ 1.50-1.77 (m, 2H), 1.84-2.43 (m, 6H), 2.61-2.79 (m, 2H),

3.05-3.24 (m, 2H), 3.26-3.47 (m, 2H), 3.83-4.15 (m, 2H),

4.28-4.45 (m, 1H), 4.47-4.63 (m, 1H), 7.01-7.08 (m, 1H),

7.09-7.27 (m, 3H), 7.31-7.37 (m, 1H), 7.38-7.47 (m, 1H), 7.59

(s, 1H), 7.74-7.91 (m, 1H), 10.34-10.72 (m, 1H), 10.95-11.12

(m, 1H); MS (ES) m/z 454.1 ([M - H]^-).

1

458.2 ([M + H]⁺). (+)-52g: H NMR (500 MHz, DMSO-d₆) δ

1.42-1.57 (m, 2H), 1.63-1.75 (m, 2H), 1.76-1.96 (m, 4H),

2.55-2.69 (m, 4H), 2.82-3.00 (m, 2H), 3.03-3.14 (m, 1H),

3.27-3.40 (m, 1H), 3.80-3.92 (m, 1H), 4.17-4.30 (m, 1H),

6.83-7.17 (m, 4H), 7.21-7.37 (m, 2H), 7.69 (s, 1H), 11.33 (s,

1H); MS (ES) m/z 458.2 ([M + H]⁺); [R]_D²⁵+51.0° (c 1.0,

DMSO).

8-Fluoro-3-[[4-(4-ﬂuoro-1H-indol-1-yl)butyl](propyl)amino]chro-

mane-5-carboxamide (54a). 54a was obtained according to

procedure B using 53a and propionaldehyde, generating 54a

(69%) as a glassy foam which was converted to the mono-HCl

1

salt to afford a white solid: H NMR (400 MHz, DMSO-d₆) δ

0.76 (t, J ) 7.3 Hz, 3H), 1.24-1.39 (m, 4H), 1.68-1.81 (m,

2H), 2.33-2.43 (m, 2H), 2.45-2.57 (m, 2H), 2.85-3.05 (m,

3H), 3.84-3.94 (m, 1H), 4.10-4.29 (m, 3H), 6.43-6.49 (m,

1H), 6.69-6.80 (m, 1H), 6.90-6.98 (m, 1H), 6.99-7.13 (m,

2H), 7.27-7.43 (m, 3H), 7.68 (br s, 1H); MS (ESI) m/z 442.2

([M + H]⁺); MS (ESI) m/z 440.3 ([M - H]^-).

8-Fluoro-3-[[4-(5-ﬂuoro-1H-indol-1-yl)butyl](propyl)amino]chro-

mane-5-carboxamide (54b). 54b was obtained according to

procedure B using 53b and propionaldehyde, generating 54b

(63%) as a glassy foam which was converted to the mono-HCl

1

salt to afford an off-white solid: H NMR (400 MHz, DMSO-

d₆) δ 0.70-0.81 (m, 3H), 1.24-1.38 (m, 4H), 1.67-1.78 (m,

2H), 2.33-2.44 (m, 2H), 2.44-2.54 (m, 2H), 2.84-3.05 (m,

3H), 3.83-3.94 (m, 1H), 4.09-4.27 (m, 3H), 6.39 (s, 1H),

6.86-6.98 (m, 2H), 7.00-7.10 (m, 1H), 7.21-7.36 (m, 2H),

7.38-7.50 (m, 2H), 7.65-7.73 (m, 1H); MS (ESI) m/z 442.2

([M + H]⁺); MS (ESI) m/z 440.2 ([M - H]^-).

3-{Cyclobutyl[3-(6-ﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuo-

rochromane-5-carboxamide (52d). 52d was obtained according to

procedure C using 51d and cyclobutanone, generating 0.16 g (99%)

of 52d which was converted to the mono-HCl salt and isolated as

8-Fluoro-3-[[4-(6-ﬂuoro-1H-indol-1-yl)butyl](propyl)amino]chro-

mane-5-carboxamide (54c). 54c was obtained according to

procedure B using 53c and propionaldehyde, generating 54c

(73%) as a glassy foam which was converted to the mono-HCl

1

a white solid: mp 126 °C/dec; H NMR (500 MHz, DMSO-d₆) δ

1.49-1.76 (m, 2H), 1.89-2.44 (m, 6H), 2.60-2.76 (m, 2H),

3.03-3.22 (m, 2H), 3.26-3.46 (m, 2H), 3.85-4.13 (m, 2H),

4.30-4.44 (m, 1H), 4.45-4.61 (m, 1H), 6.77-6.88 (m, 1H),

7.02-7.25 (m, 4H), 7.35-7.56 (m, 2H), 7.75-7.88 (m, 1H),

10.32-10.60 (m, 1H), 10.90 (s, 1H); MS (ES) m/z 440.2 ([M +

H]⁺). Anal. (C₂₅H₂₇F₂N₃O₂·1.10 HCl ·H₂O) C, H, N.

1

salt to afford an off-white solid: H NMR (400 MHz, DMSO-

d₆) δ 0.71-0.82 (m, 3H), 1.24-1.37 (m, 4H), 1.65-1.78 (m,

2H), 2.33-2.45 (m, 2H), 2.43-2.54 (m, 2H), 2.82-3.06 (m,

3H), 3.82-3.93 (m, 1H), 4.06-4.16 (m, 2H), 4.17-4.29 (m,

1H), 6.41 (s, 1H), 6.78-6.88 (m, 1H), 6.90-6.97 (m, 1H),

6.99-7.10 (m, 1H), 7.26-7.38 (m, 3H), 7.45-7.55 (m, 1H),

7.67 (br s, 1H); MS (ESI) m/z 442.2 ([M + H]⁺); MS (ESI) m/z

440.2 ([M - H]^-).

8-Fluoro-3-[[4-(7-ﬂuoro-1H-indol-1-yl)butyl](propyl)amino]chro-

mane-5-carboxamide (54d). 54d was obtained according to

procedure B using 53d and propionaldehyde, generating a

quantitative yield of 54d as a colorless glass which was converted

to the mono-HCl salt to afford a white solid: ¹H NMR (400

MHz, DMSO-d₆) δ 0.69-0.80 (m, 3H), 1.21-1.36 (m, 4H),

1.68-1.81 (m, 2H), 2.32-2.44 (m, 2H), 2.43-2.53 (m, 2H),

2.82-3.03 (m, 3H), 3.79-3.92 (m, 1H), 4.17-4.31 (m, 3H),

6.43-6.50 (m, 1H), 6.82-6.96 (m, 3H), 6.98-7.08 (m, 1H),

7.28-7.41 (m, 3H), 7.68 (s, 1H); MS (ESI) m/z 442.2 ([M +

H]⁺); MS (ESI) m/z 440.2 ([M - H]^-).

3-{Cyclobutyl[3-(7-methoxy-1H-indol-3-yl)propyl]amino}-8-ﬂuo-

rochromane-5-carboxamide (52e). 52e was obtained according to

procedure C using 51e and cyclobutanone, generating 0.11 g

(81%) of 52e which was converted to the mono-HCl salt and

isolated as a white solid: mp 156-162 °C/dec; ¹H NMR (400

MHz, DMSO-d₆) δ 1.50-1.76 (m, 2H), 1.89-2.46 (m, 6H),

2.62-2.77 (m, 2H), 3.02-3.22 (m, 2H), 3.25-3.48 (m, 2H),

3.88 (s, 3H), 3.92-4.13 (m, 2H), 4.27-4.43 (m, 1H), 4.46-4.62

(m, 1H), 6.56-6.69 (m, 1H), 6.83-6.96 (m, 1H), 6.97-7.27

(m, 4H), 7.36-7.50 (m, 1H), 7.73-7.90 (m, 1H), 10.41-10.71

(m, 1H), 10.90 (s, 1H); MS (ES) m/z 452.2 ([M + H]⁺).

3-[[3-(7-Chloro-1H-indol-3-yl)propyl](cyclobutyl)amino]-8-ﬂuo-

rochromane-5-carboxamide (52f). 52f was obtained according to

procedure C using 51f and cyclobutanone, generating 0.133 g

(97%) of 52f which was converted to the mono-HCl salt and

isolated as a white solid: mp 154-163 °C/dec; ¹H NMR (400

MHz, DMSO-d₆) δ 1.50-1.77 (m, 2 H), 1.87-2.46 (m, 6 H),

2.61-2.79 (m, 2 H), 3.03-3.22 (m, 2 H), 3.31 (s, 2 H),

3.85-4.14 (m, 2 H), 4.29-4.44 (m, 1 H), 4.46-4.61 (m, 1 H),

6.92-7.04 (m, 1 H), 7.07-7.29 (m, 4 H), 7.37-7.57 (m, 2 H),

7.74-7.88 (m, 1 H), 10.27-10.55 (m, 1 H), 11.20 (s, 1 H); MS

m/z 455.2.

3-{(Cyclopropylmethyl)[4-(4-ﬂuoro-1H-indol-1-yl)butyl]amino}-

8-ﬂuorochromane-5-carboxamide (55a). 55a was obtained ac-

cording to procedure B using 53a and cyclopropane carboxal-

dehyde, generating 55a (73%) as a glassy foam which was

1

converted to the mono-HCl salt to afford an off-white solid: H

NMR (400 MHz, DMSO-d₆) δ -0.05 to 0.08 (m, 2H), 0.31-0.42

(m, 2H), 0.66-0.81 (m, 1H), 1.30-1.43 (m, 2H), 1.69-1.82

(m, 2H), 2.31-2.42 (m, 2H), 2.53-2.64 (m, 2H), 2.85-2.98

(m, 2H), 3.06-3.20 (m, 1H), 3.84-3.94 (m, 1H), 4.12-4.30

(m, 3H), 6.46 (d, J ) 2.9 Hz, 1H), 6.69-6.82 (m, 1H),

6.89-6.98 (m, 1H), 6.99-7.14 (m, 2H), 7.26-7.44 (m, 3H),

7.68 (br s, 1H); MS (ESI) m/z 454.3 ([M + H]⁺); MS (ESI) m/z

452.3 ([M - H]^-).

(-)-3-{Cyclobutyl[3-(5,7-diﬂuoro-1H-indol-3-yl)propyl]amino}-

8-ﬂuorochromane-5-carboxamide ((-)-52g) and (+)-3-{Cyclobu-

tyl[3-(5,7-diﬂuoro-1H-indol-3-yl)propyl]amino}-8-ﬂuorochromane-

5-carboxamide ((+)-52g). The compounds were obtained according

to procedure C using each enantiomer of 51g and cyclobutanone,

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 7001

3-{(Cyclopropylmethyl)[4-(5-ﬂuoro-1H-indol-1-yl)butyl]amino}-

8-ﬂuorochromane-5-carboxamide (55b). 55b was obtained ac-

cording to procedure B using 53b and cyclopropane carboxal-

dehyde, generating 55b (71%) as a glassy foam which was

3-{Cyclobutyl[4-(7-ﬂuoro-1H-indol-1-yl)butyl]amino}-8-ﬂuoro-

chromane-5-carboxamide (56d). 56d was obtained according to

procedure C using 53d and cyclobutanone, generating 56d (93%)

as a colorless glass which was converted to the mono-HCl salt to

afford a white solid: ¹H NMR (400 MHz, DMSO-d₆) δ 1.24-1.37

(m, 2H), 1.39-1.55 (m, 2H), 1.65-1.94 (m, 6H), 2.42-2.53 (m,

3H), 2.74-3.06 (m, 3H), 3.76-3.86 (m, 1H), 4.12-4.20 (m, 1H),

4.21-4.33 (m, 2H), 6.43-6.50 (m, 1H), 6.83-6.98 (m, 3H),

7.00-7.11 (m, 1H), 7.28-7.42 (m, 3H), 7.68 (br s, 1H); MS (ESI)

m/z 454.2 ([M + H]⁺); MS (ESI) m/z 452.2 ([M - H]^-).

(3R)-(-)-8-Fluoro-3-[{[6-ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-

3-yl]methyl}(propyl)amino]chromane-5-carboxamide (58a) and

(3R)-(+)-8-Fluoro-3-[{[6-ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-3-

yl]methyl}(propyl)amino]chromane-5-carboxamide (58b). The com-

pounds were obtained according to procedure B using 57a and

propionaldehyde, generating 0.085 g (96%) of 58a which was

converted to the mono-HCl salt to afford a white solid: mp 174-180

°C; ¹H NMR (400 MHz, DMSO-d₆) δ 0.81-1.00 (m, 3H),

1.49-1.66 (m, 1H), 1.70-1.91 (m, 2H), 2.09-2.39 (m, 3H),

2.67-2.81 (m, 2H), 2.83-3.04 (m, 1H), 3.06-3.57 (m, 6H),

3.96-4.09 (m, 1H), 4.32-4.53 (m, 1H), 4.63-4.77 (m, 1H),

6.74-6.87 (m, 1H), 6.99-7.27 (m, 4H), 7.40-7.51 (m, 1H),

7.80-7.91 (m, 1H), 9.93-10.07 (m, 1H), 10.82 (s, 1H); MS (ES)

m/z 454.2 ([M + H]⁺); [R]_D²⁵-90.8° (c 1.0, DMSO). Similarly,

57b was reacted with propionaldehyde to generate 0.072 g (82%)

of 58b which was converted to the mono-HCl salt to afford a white

solid: mp 173-178 °C; ¹H NMR (400 MHz, DMSO-d₆) δ 0.92 (t,

3H), 1.52-1.69 (m, 1H), 1.69-1.91 (m, 2H), 2.02-2.39 (m, 3H),

2.61-2.83 (m, 2H), 2.84-3.05 (m, 1H), 3.08-3.59 (m, 5H),

3.93-4.12 (m, 1H), 4.36-4.55 (m, 1H), 4.59-4.76 (m, 1H),

6.71-6.86 (m, 1H), 6.97-7.27 (m, 4H), 7.39-7.49 (m, 1H),

7.77-7.91 (m, 1H), 9.77-9.96 (m, 1H), 10.83 (s, 1H); MS (ES)

m/z 454.2 ([M + H]⁺); [R]²_D⁵+12.4° (c 1.0, DMSO).

(3R)-(-)-3-((Cyclopropylmethyl){[6-ﬂuoro-2,3,4,9-tetrahydro-

1H-carbazol-3-yl]methyl}amino)-8-ﬂuorochromane-5-carboxam-

ide (59a) and (3R)-(+)-3-((Cyclopropylmethyl){[6-ﬂuoro-2,3,4,9-

tetrahydro-1H-carbazol-3-yl]methyl}amino)-8-ﬂuorochromane-5-

carboxamide (59b). The compounds were obtained according to

procedure B using 57a and cyclopropanecarboxaldehyde, gen-

erating 0.071 g (79%) of 59a which was converted to the mono-

HCl salt to afford a white solid: mp 175-179 °C; ¹H NMR (400

MHz, DMSO-d₆) δ 0.37-0.55 (m, 2H), 0.60-0.76 (m, 2H),

1.17-1.32 (m, 2H), 1.51-1.66 (m, 1H), 2.05-2.39 (m, 3H),

2.68-2.83 (m, 2H), 2.83-3.20 (m, 2H), 3.21-3.55 (m, 4H),

4.03-4.18 (m, 1H), 4.35-4.52 (m, 1H), 4.63-4.76 (m, 1H),

6.74-6.87 (m, 1H), 6.99-7.26 (m, 4H), 7.39-7.52 (m, 1H),

7.79-7.91 (m, 1H), 9.90-10.05 (m, 1H), 10.82 (s, 1H); MS

(ES) m/z 466.2 ([M + H]⁺); [R]²_D⁵-89.0° (c 1.0, DMSO).

Similarly, 57b was reacted with cyclopropanecarboxaldehyde,

generating 0.083 g (92%) of 59b which was converted to the

mono-HCl salt to afford a white solid: mp 174-178 °C; ¹H NMR

(400 MHz, DMSO-d₆) δ 0.34-0.55 (m, 2H), 0.57-0.74 (m, 2H),

1.14-1.33 (m, 2H), 1.52-1.69 (m, 1H), 2.02-2.41 (m, 3H),

2.68-2.84 (m, 2H), 2.85-3.07 (m, 1H), 3.07-3.62 (m, 4H),

4.01-4.17 (m, 1H), 4.39-4.52 (m, 1H), 4.61-4.76 (m, 1H),

6.73-6.87 (m, 1H), 6.99-7.26 (m, 4H), 7.40-7.52 (m, 1H),

7.78-7.90 (m, 1H), 9.84-9.99 (m, 1H), 10.83 (s, 1H); MS (ES)

m/z 466.2 ([M + H]⁺); [R]²_D⁵+10.0° (c 1.0, DMSO).

3-{Cyclobutyl[(6-ﬂuoro-2,3,4,9-tetrahydro-1H-carbazol-3-yl)-

methyl]amino}-8-ﬂuorochromane-5-carboxamide(60a-d). 60a-d

were obtained according to procedure C using 57 and cyclob-

utanone to give a residue which was puriﬁed by chiral HPLC

[column, Chiralcel AD, 0.46 cm × 25 cm; mobile phase, 1:1

hexane/ethanol], generating the four diastereoisomers below as

the free bases, which were then converted to the mono-HCl salts

to afford 60a-d. 60a: mp 192 °C; MS (ES) m/z 466.2 ([M +

H]⁺); [R]²_D⁵-26.2° (c 1.0, MeOH). 60b: mp 218 °C; MS (ES)

m/z 464.2 ([M - H]^-); [R]_D²⁵+27.4° ( c 1.0, MeOH). 60c: mp

227 °C; MS (ES) m/z 464.2 ([M - H]^-); [R]_D²⁵+50.2° (c 1.0,

MeOH). 60d: mp 210 °C; MS (ES) m/z 464.2 ([M - H]^-); [R]²_D⁵

-50.2° (c 1.0, MeOH). ¹H NMR (400 MHz, DMSO-d₆) (for

1

converted to the mono-HCl salt to afford an off-white solid: H

NMR (400 MHz, DMSO-d₆) δ -0.06-0.07 (m, 2H), 0.32-0.42

(m, 2H), 0.68-0.78 (m, 1H), 1.29-1.41 (m, 2H), 1.67-1.79

(m, 2H), 2.31-2.41 (m, 2H), 2.53-2.63 (m, 2H), 2.84-2.97

(m, 2H), 3.07-3.19 (m, 1H), 3.81-3.95 (m, 1H), 4.09-4.31

(m, 3H), 6.35-6.44 (m, 1H), 6.88-7.10 (m, 3H), 7.22-7.35

(m, 2H), 7.38-7.51 (m, 2H), 7.68 (br s, 1H); MS (ESI) m/z

454.2 ([M + H]⁺); MS (ESI) m/z 452.2 ([M - H]^-).

3-{(Cyclopropylmethyl)[4-(6-ﬂuoro-1H-indol-1-yl)butyl]amino}-

8-ﬂuorochromane-5-carboxamide (55c). 55c was obtained accord-

ing to procedure B using 53c and cyclopropane carboxaldehyde,

generating 55c (81%) as a glassy foam which was converted to

1

the mono-HCl salt to afford an off-white solid: H NMR (400

MHz, DMSO-d₆) δ -0.04-0.08 (m, 2H), 0.33-0.42 (m, 2H),

0.68-0.79 (m, 1H), 1.30-1.42 (m, 2H), 1.67-1.79 (m, 2H),

2.31-2.43 (m, 2H), 2.52-2.62 (m, 2H), 2.86-2.97 (m, 2H),

3.08-3.19 (m, 1H), 3.83-3.95 (m, 1H), 4.08-4.17 (m, 2H),

4.20-4.31 (m, 1H), 6.38-6.45 (m, 1H), 6.77-6.88 (m, 1H),

6.89-6.98 (m, 1H), 6.99-7.11 (m, 1H), 7.28-7.38 (m, 3H),

7.46-7.54 (m, 1H), 7.68 (br s, 1H); MS (ESI) m/z 454.2 ([M +

H]⁺); MS (ESI) m/z 452.2 ([M - H]^-).

3-{(Cyclopropylmethyl)[4-(7-ﬂuoro-1H-indol-1-yl)butyl]amino}-

8-ﬂuorochromane-5-carboxamide (55d). 55d was obtained accord-

ing to procedure B using 53d and cyclopropane carboxaldehyde,

generating 55d (97%) as a colorless glass which was converted to

1

the mono-HCl salt to afford a white solid: H NMR (400 MHz,

DMSO-d₆) δ -0.04 to 0.06 (m, 2H), 0.31-0.40 (m, 2H), 0.66-0.76

(m, 1H), 1.28-1.40 (m, 2H), 1.70-1.82 (m, 2H), 2.31-2.41 (m,

2H), 2.53-2.62 (m, 2H), 2.84-2.97 (m, 2H), 3.04-3.18 (m, 1H),

3.81-3.92 (m, 1H), 4.18-4.32 (m, 3H), 6.42-6.49 (m, 1H),

6.83-7.10 (m, 4H), 7.28-7.42 (m, 3H), 7.68 (br s, 1H); MS (ESI)

m/z 454.2 ([M + H]⁺); MS (ESI) m/z 452.2 ([M - H]^-).

3-{Cyclobutyl[4-(4-ﬂuoro-1H-indol-1-yl)butyl]amino}-8-ﬂuoro-

chromane-5-carboxamide (56a). 56a was obtained according to

procedure C using 53a and cyclobutanone, generating 56a (68%)

as a glassy foam which was converted to the mono-HCl salt to

afford an off-white solid: ¹H NMR (400 MHz, DMSO-d₆) δ

1.26-1.39 (m, 2H), 1.40-1.56 (m, 2H), 1.64-1.95 (m, 6H),

2.39-2.55 (m, 3H), 2.76-3.06 (m, 3H), 3.78-3.90 (m, 1H),

4.11-4.23 (m, 3H), 6.43-6.48 (m, 1H), 6.71-6.80 (m, 1H),

6.89-6.98 (m, 1H), 7.00-7.12 (m, 2H), 7.27-7.43 (m, 3H), 7.68

(br s, 1H); MS (ESI) m/z 454.2 ([M + H]⁺); MS (ESI) m/z 452.3

([M - H]^-).

3-{Cyclobutyl[4-(5-ﬂuoro-1H-indol-1-yl)butyl]amino}-8-ﬂuoro-

chromane-5-carboxamide (56b). 56b was obtained according to

procedure C using 53b and cyclobutanone, generating 56b (71%)

as a glassy foam which was converted to the mono-HCl salt to

afford an off-white solid: ¹H NMR (400 MHz, DMSO-d₆) δ

1.25-1.38 (m, 2H), 1.38-1.55 (m, 2H), 1.61-1.92 (m, 6H),

2.40-2.57 (m, 3H), 2.73-3.08 (m, 3H), 3.75-3.89 (m, 1H),

4.08-4.22 (m, 3H), 6.35-6.42 (m, 1H), 6.88-6.98 (m, 2H),

7.00-7.10 (m, 1H), 7.22-7.36 (m, 2H), 7.39-7.51 (m, 2H), 7.68

(br s, 1H); MS (ESI) m/z 454.2 ([M + H]⁺); MS (ESI) m/z 452.2

([M - H]^-).

3-{Cyclobutyl[4-(6-ﬂuoro-1H-indol-1-yl)butyl]amino}-8-ﬂuoro-

chromane-5-carboxamide (56c). 56c was obtained according to

procedure C using 53c and cyclobutanone, generating 56c (75%)

as a glassy foam which was converted to the mono-HCl salt to

afford an off-white solid: ¹H NMR (400 MHz, DMSO-d₆) δ

1.25-1.38 (m, 2H), 1.39-1.56 (m, 2H), 1.62-1.94 (m, 6H),

2.40-2.59 (m, 3H), 2.74-3.07 (m, 3H), 3.77-3.89 (m, 1H),

4.05-4.21 (m, 3H), 6.38-6.43 (m, 1H), 6.78-6.87 (m, 1H),

6.88-6.98 (m, 1H), 6.99-7.10 (m, 1H), 7.28-7.38 (m, 3H),

7.44-7.54 (m, 1H), 7.68 (br s, 1 H); MS (ESI) m/z 454.2 ([M +

H]⁺); MS (ESI) m/z 452.2 ([M - H]^-).

7002 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

60a-d) δ 1.45-1.83 (m, 3H), 2.04-2.37 (m, 6H), 2.36-2.62

(m, 3H), 2.68-2.82 (m, 2H), 2.85-3.20 (m, 2H), 3.28-3.53

(m, 1H), 3.86-4.21 (m, 2H), 4.34-4.50 (m, 1H), 4.59-4.72

(m, 1H), 6.75-6.86 (m, 1H), 6.99-7.26 (m, 4H), 7.39-7.49

(m, 1H), 7.74-7.90 (m, 1H), 10.19-10.40 (m, 1H), 10.77-10.86

(m, 1H).

1H), 4.32-4.52 (m, 2H), 6.83-6.95 (m, 1H), 7.04-7.48 (m,

6H), 7.81 (br s, 1H), 8.92-9.22 (m, 2H), 10.97 (br s, 1H).

(3R)-3-{(Cyclopropylmethyl)[3-(5-ﬂuoro-1H-indol-3-yl)-2-meth-

ylpropyl]amino}-8-ﬂuorochromane-5-carboxamide (70a,b). 70a,b

were obtained according to procedure B using 69 and cyclo-

propanecarboxaldehyde to generate 0.214 g (94%) of racemic

70 as a white solid. The diastereomers were separated by HPLC

using a Whelk O column (2 cm × 25 cm) and 40% EtOH/Hex

as mobile phase, isolated, and converted to the mono-HCl salt

to afford the following products as white solids. 70a: mp

155-159 °C/dec; MS (ES) m/z 454.2 ([M + H]⁺); [R]²_D⁵-22.2°

(c 1.0, DMSO); 99.9% ee by chiral HPLC. 70b: mp 155-160

°C/dec; MS (ES) m/z 454.2 ([M + H]⁺); [R]_D²⁵-59.6° (c 1.0,

DMSO); 99.9% ee by chiral HPLC. ¹H NMR (400 MHz, DMSO-

d₆) (for 70a,b) δ 0.16-0.69 (m, 4H), 0.92-1.24 (m, 5H),

2.21-2.37 (m, 1H), 2.54-2.80 (m, 1H), 2.88-3.30 (m, 4H),

3.29-3.58 (m, 2H), 3.94-4.10 (m, 1H), 4.29-4.73 (m, 2H),

6.82-6.98 (m, 1H), 7.06-7.51 (m, 6H), 7.84 (br s, 1H),

9.68-10.12 (m, 1H), 10.99 (br s, 1H).

8-Fluoro-3-[[3-(5-ﬂuoro-1H-indol-3-yl)-1-methylpropyl]ami-

no]chromane-5-carboxamide (66). To a slurry of 29a (0.59 g, 2.79

mmol) in anhydrous 1,2-dichloroethane (25 mL), under nitrogen

at room temperature, was added 61 (0.57 g, 2.79 mmol), acetic

acid (0.29 mL, 5.58 mmol), and sodium triacetoxyborohydride

(0.83 g, 3.91 mmol). The mixture was stirred at room temperature

overnight. Workup was the same as described in procedure B.

Chromatography ((14:1) CH₂Cl₂-MeOH (1% NH₄OH)) afforded

0.82 g (74%) of 66: MS (ESI) m/z 400 ([M + H]⁺). The

diastereomers were ﬁrst separated by normal phase HPLC using

a Luna SIL column (5 cm × 25 cm) and 3% MeOH (5% H₂O)

in CH₂Cl₂as mobile phase. The enantiomers were then separated

by SFC using a Chiralpak AS column (2 cm × 25 cm) and 50%

MeOH (0.1% DEA) in CO₂as mobile phase and isolated. ¹H

NMR (400 MHz, DMSO-d₆) δ 0.95-1.10 (d, 3H), 1.45-1.85

(m, 3H), 2.52-2.70 (m, 3H), 2.70-2.88 (m, 1H), 2.98-3.10

(m, 2H), 3.62-3.80 (m, 1H), 4.10-4.28 (m, 1H), 6.78-6.97

(m, 2H), 6.97-7.10 (m, 1H), 7.10-7.40 (m, 4H), 7.67 (s, 1H),

10.75 -10.82 (m, 1H).

Biological Assays. 5-HT_1ABinding Afﬁnity Assay. The afﬁnity

of compounds for the 5-HT_1Areceptor was performed using a

standard [³H]-8-OH-DPAT binding assay using human 5-HT_1A

receptors stably cloned in CHO cells.⁴²

³H-Paroxetine Binding To Assess Afﬁnity of Drugs for the

Serotonin Transporter. A protocol similar to that used by

Cheetham et al.³⁶was used to determine the afﬁnity of

compounds for the serotonin transporter. Brieﬂy, frontal cortical

membranes prepared from male Sprague-Dawley rats were

incubated with ³H-paroxetine (0.1 nM) for 60 min at 25 °C. All

tubes also contained vehicle, test compound (one to eight

concentrations), or a saturating concentration of ﬂuoxetine (10

µM) to deﬁne speciﬁc binding. All reactions are terminated by

the addition of ice cold Tris buffer followed by rapid ﬁltration

using a Tom Tech ﬁltration device to separate bound from free

³H-paroxetine. Bound radioactivity was quantitated using a

Wallac 1205 Beta Plate counter. Nonlinear regression analysis

was used to determine IC₅₀values which were converted to K_i

values using the method of Cheng and Prusoff;³⁷K_i) IC₅₀/

[(radioligand concn)/(1 + KD)].

8-Fluoro-3-[[3-(5-ﬂuoro-1H-indol-3-yl)-1-methylpropyl](pro-

pyl)amino]-3,4-dihydro-2H-chromene-5-carboxamide (67a-d).

Each isomer of compound 66 was converted to the desired

products 67a-d by reaction of 66 with propionaldehyde ac-

cording to procedure B, generating the following desired pro-

ducts as white solids which were converted to their mono-HCl

salts. 67a, 0.114 g (100%): mp 133 °C/dec; MS (ESI) m/z 442

([M

+

H]⁺); [R]_D²⁵-22.1° (c 1.0, DMSO). Anal.

(C₂₅H₂₉F₂N₃O₂· HCl · 0.60H₂O) C, H, N. 67b, 0.11 g (96%): mp

133 °C/dec; MS (ESI) m/z 442 ([M + H]⁺); [R]_D²⁵+23.4° (c

1.0, DMSO). Anal. (C₂₅H₂₉F₂N₃O₂· HCl · 0.60H₂O) C, H, N. 67c,

0.093 g (94%): mp 133 °C/dec; MS (ESI) m/z 442 ([M + H]⁺);

[R]_D²⁵-67.4° (c 1.0, DMSO). Anal. (C₂₅H₂₉F₂N₃O₂·

HCl · 0.60H₂O) C, H, N. 67d, 0.095 g (95%): mp 133 °C/dec;

MS (ESI) m/z 442 ([M + H]⁺); [R]²_D⁵+61.4° (c 1.0, DMSO).

Anal. (C₂₅H₂₉F₂N₃O₂· HCl · 0.60H₂O) C, H, N. ¹H NMR (400

MHz, DMSO-d₆) (for 67a-d) δ 0.75-0.94 (m, 3H), 1.31-1.56

(m, 3H), 1.57-1.79 (m, 1H), 1.81-2.37 (m, 1H), 2.58-2.88

(m, 4H), 3.06-3.63 (m, 5H), 3.89-4.13 (m, 1H), 4.22-4.39

(m, 1H), 4.41-4.73 (m, 1H), 6.84-6.97 (m, 1H), 7.09-7.24

(m, 2H), 7.25-7.47 (m, 4H), 7.78-7.90 (m, 1H), 9.55-9.73

(m, 1H), 10.90-11.00 (m, 1H).

Inhibition of ³H-5-HT Uptake by Cells Possessing the Human

5-HT Transporter. A human carcinoma cell line (Jar cells)

possessing low endogenous levels of the 5-HT transporter were

seeded into 96-well plates and treated with staurosporine at least

18 h prior to assay. [Staurosporine greatly increases the

expression of the 5-HT transporter.] On the day of assay, vehicle,

excess of ﬂuoxetine, or test compound was added to various

wells on the plate. All wells then received ³H-5-HT and were

incubated at 37 °C for 5 min. The wells are then washed with

ice cold 50 mM Tris HCl (pH 7.4) buffer and aspirated to remove

free ³H-5-HT. An amount of 25 µL of 0.25 M NaOH is then

added to each well to lyse the cells and 75 µL scintillation

cocktail (Microscint 20) added prior to quantitation on a Packard

TopCount machine. Tubes with vehicle represent total possible

uptake, and radioactivity counted in tubes with ﬂuoxetine

represent nonspeciﬁc binding/uptake and is subtracted from the

total possible uptake to give total possible speciﬁc uptake. This

nonspeciﬁc binding (usual low in number) is then subtracted

from the counts obtained in wells with various test compounds

(or different concentrations of test drug) to give speciﬁc uptake

in the presence of drug. Speciﬁc uptake is then expressed as a

% of control values and is analyzed using nonlinear regression

analysis (Prizm) to determine IC₅₀values. If the compound is

active at inhibiting 5-HT uptake, its counts will be close to that

obtained with ﬂuoxetine.

(-)-3-{(Cyclopropylmethyl)[3-(5-ﬂuoro-1H-indol-3-yl)-1-meth-

ylpropyl]amino}-8-ﬂuorochromane-5-carboxamide (68a) and (+)-

3-{(Cyclopropylmethyl)[3-(5-ﬂuoro-1H-indol-3-yl)-1-methylpro-

pyl]amino}-8-ﬂuorochromane-5-carboxamide (68b). Two isomers

of compound 66 were converted to the desired products 68a,b

by reaction of 66 with cyclopropanecarboxaldehyde according

to procedure B, generating the following desired products as

white solids which were converted to their mono-HCl salts. 68a,

0.08 g (90%): mp 135 °C/dec; MS (ES) m/z 454.2 ([M + H]⁺);

[R]²_D⁵-17.0° (c 1.0, DMSO). Anal. (C₂₆H₂₉F₂N₃O₂· HCl) C, H,

N. 68b, 0.075 g (90%): mp 135 °C/dec; MS (ES) m/z 452.2 ([M

- H]^-); [R]_D²⁵+15.4° (c 1.0, DMSO). Anal. (C₂₆H₂₉F₂N₃O₂·

HCl · 0.60H₂O) C, H, N. ¹H NMR (500 MHz, DMSO-d₆) (for

68a,b) δ 0.25-0.69 (m, 4H), 0.92-1.28 (m, 1H), 1.36-1.53

(m, 3H), 1.78-1.99 (m, 1H), 2.10-2.40 (m, 1H), 2.55-3.11

(m, 3H), 3.14-3.45 (m, 2H), 3.47-3.81 (m, 2H), 3.98-4.72

(m, 3H), 6.83-6.97 (m, 1H), 7.08-7.49 (m, 6H), 7.80-7.91

(m, 1H), 9.49-9.72 (m, 1H), 10.89-11.00 (m, 1H).

Assessment of Agonism/Antagonism at the 5-HT_1AReceptor

Using [³⁵S]-GTPγS Binding to Cloned Human 5-HT_1ARecep-

tors. The [³⁵S]-GTPγS binding assay was similar to that used

by Larenzo and Birdsall.⁴⁰Brieﬂy, 5-HT_1Acloned receptor

membrane fragments (as used for 5-HT_1Areceptor binding

assays) were stored at -70 °C until needed. These membranes

were then incubated for 30 min at 37 °C with [³⁵S]-GTPγS (1

(3R)-8-Fluoro-3-{[3-(5-ﬂuoro-1H-indol-3-yl)-2-methylpropyl]-

amino}chromane-5-carboxamide (69). 69 was obtained accord-

ing to procedure B by reacting (R)-29 with 65 to generate 1.07 g

(80%) of 69 as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ

0.93 (d, J ) 6.6 Hz, 3H), 2.16-2.30 (m, 1H), 2.82-2.98 (m,

2H), 3.07-3.17 (m, 1H), 3.21-3.48 (m, 3H), 3.71-3.85 (m,

AdVances toward New Antidepressants

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 7003

Raphe pathway: role of somatodendritic autoreceptors. J. Neurochem.

1997, 68, 2593–2603.

nM) in the presence of vehicle, test compound (11 concentra-

tions), or excess 8-OH-DPAT to deﬁne maximum agonist

response. All reactions were terminated by the addition of ice

cold Tris buffer (50 mM Tris, 3 mM MgCl₂; 100 mM NaCl, pH

8.0) followed by rapid ﬁltration using a Brandel ﬁltration device

to separate bound from free [³⁵S]-GTPγS. Agonists produce an

increase in the amount of [³⁵S]-GTPγS bound, whereas antago-

nists produce no increase in binding. Bound radioactivity was

counted and analyzed as above.

(15) Blier, P.; Bergeron, R. J. The use of pindolol to potentiate antidepres-

sant medication. J. Clin. Psychiatry 1998, 59 (Suppl. 5), 16–23.

(16) Berman, R. M.; Anand, A.; Cappiello, A.; Miller, H. L.; Xu, X. S.;

Oren, D. A.; Charney, D. S. The use of pindolol with ﬂuoxetine in

the treatment of major depression: ﬁnal results from a double-blind,

placebo-controlled trial. Biol. Psychiatry 1999, 45, 1170–1177.

(17) Gartside, S. E.; Clifford, E. M.; Cowen, P. J.; Sharp, T. Effects of

(-)-tertatolol, (-)-penbutolol and (()-pindolol in combination with

paroxetine on presynaptic 5-HT function: an in vivo microdialysis and

electrophysiological study. Br. J. Pharmacol. 1999, 127, 145–152.

(18) Beyer, C. E.; Boikess, S.; Luo, B.; Dawson, L. A. Comparison of the

effects of antidepressants on norepinephrine and serotonin concentra-

tions in the rat frontal cortex: an in-vivo microdialysis study.

J. Psychopharmacol 2002, 16, 297–304.

(19) (a) Martinez-Esparza, J.; Oﬁcialdegui, A. M.; Perez-Silanes, S.; Heras,

B.; Orus, L.; Palop, J. A.; Lasheras, B.; Roca, J.; Mourelle, M.; Bosch,

A.; Del Castillo, J. C.; Tordera, R.; Del Rio, J.; Monge, A. New 1-aryl-

3-(4-arylpiperazin-1-yl)propane derivatives, with dual action at 5-HT_1A

serotonin receptors and serotonin transporter, as a new class of

antidepressants. J. Med. Chem. 2001, 44, 418–428, and references

therein. (b) Meagher, K. L.; Mewshaw, R. E.; Evrard, D. A.; Zhou,

P.; Smith, D. L.; Scerni, R.; Spangler, T.; Abulhawa, S.; Shi, X.;

Schechter, L. E.; Andree, T. H. Studies towards the next generation

of antidepressants. Part 1. Indolylcyclohexylamines as potent serotonin

reuptake inhibitors. Bioorg. Med. Chem. Lett. 2001, 11, 1885–1888.

(20) Takeuchi, K.; Kohn, T. J.; Honigschmidt, N. A.; Rocco, V. P.;

Spinazze, P. G.; Hemrick-Luecke, S. K.; Thompson, L. K.; Evans,

D. C.; Rasmussen, K.; Koger, D.; Lodge, D.; Martin, L. J.; Shaw, J.;

Threlkeld, P. G.; Wong, D. T. Advances toward new antidepressants

beyond SSRIs: 1-aryloxy-3-piperidinylpropan-2-ols with dual 5-HT_1A

receptor antagonism/SSRI activities. Part 5. Bioorg. Med. Chem. Lett.

2006, 16, 2347–2351, and references therein.

³H-Prazosin Binding To Assess Afﬁnity of Drugs for the r₁

Receptor. Brieﬂy, cortical membranes prepared from male

Sprague-Dawley rats were incubated with ³H-prazosin (0.2nM)

for 30 min at 25 °C. All tubes also contained vehicle, test

compound (one to eight concentrations), or a saturating con-

centration of phentolamine (10 µM) to deﬁne speciﬁc binding.

All reactions are terminated by the addition of ice cold Tris

buffer followed by rapid ﬁltration using a Tom Tech ﬁltration

3

device to separate bound from free H-prazosin. Bound radio-

activity was quantitated using a Wallac 1205 ꢀ plate counter.

Nonlinear regression analysis was used to determine IC₅₀values,

which were converted to K_ivalues using the method of Cheng

and Prusoff;³⁷K_i) IC₅₀/[(radioligand concn)/(1 + KD)].

Acknowledgment. The authors thank members of the Wyeth

Chemical Technology Department for preparative chiral HPLC

separations, optical rotation, mass spectometry, and NMR data,

and Dr. Douglas M. Ho for X-ray studies. We also thank

Jennifer Lo, Susan Croce, and Robert Vogel for the synthesis

of a few intermediates.

(21) Evrard, D. A.; Zhou, P.; Yi, S.; Zhou, D.; Smith, D. L.; Sullivan,

K. M.; Hornby, G. A.; Schechter, L. E.; Andree, T. H.; Mewshaw,

R. E. Studies toward the next generation of antidepressants. Part 4:

Derivatives of 4-(5-ﬂuoro-1H-indol-3-yl) cyclohexylamine with afﬁnity

for the serotonin transporter and the 5-HT_1Areceptor. Bioorg. Med.

Chem. Lett. 2005, 15, 911–914, and references therein.

Supporting Information Available: Synthesis and analytical

data of some intermediates and elemental analysis data for most

ﬁnal targets. This material is available free of charge via the Internet

at http://pubs.acs.org.

(22) Perez, M.; Pauwels, P. J.; Pallard-Sigogneau, I.; Fourrier, C.; Chopin,

P.; Palmier, C.; Colovray, V.; Halazy, S. Design and synthesis of new

potent, silent 5-HT_1Aantagonists by covalent coupling of aminopro-

panol derivatives with selective serotonin reuptake inhibitors. Bioorg.

Med. Chem. Lett. 1998, 8, 3423–3428.

(23) Podona, T.; Guardiola-Lemaitre, B.; Caignard, D. H.; Adam, G.;

Pfeiffer, B.; Renard, P.; Guillaumet, G. 3,4-Dihydro-3-amino-2H-1-

benzopyran derivatives as 5-HT_1Areceptor ligands and potential

anxiolytic agents. 1. Synthesis and structure-activity relationship

studies. J. Med. Chem. 1994, 37, 1779–1793.

(24) Hammarberg, E.; Nordvall, G.; Leideborg, R.; Nylof, M.; Hanson,

S.; Johansson, L.; Thorberg, S.-O.; Tolf, B.-R.; Jerning, E.; Svantesson,

G. T.; Mohell, N.; Ahlgren, C.; Westlind-Danielsson, A.; Csoregh, I.;

Johansson, R. Synthesis of novel 5-substituted 3-amino-3,4-dihydro-

2H-1-benzopyran derivatives and their interactions with the 5-HT_1A

receptor. J. Med. Chem. 2000, 43, 2837–2850.

(25) Ross, S. B.; Thorberg, S.-O.; Jerning, E.; Mohell, N.; Stenfors, C.;

Wallsten, C.; Milchert, I. G.; Ojteg, G. A novel selective 5-HT_1A

receptor antagonist. CNS Drug ReV. 1999, 5, 213–232.

(26) Hatzenbuhler, N. T.; Evrard, D. A.; Harrison, B. L.; Huryn, D.;

Inghrim, J.; Kraml, C.; Mattes, J. F.; Mewshaw, R. E.; Zhou, D.;

Hornby, G.; Lin, Q.; Smith, D. L.; Sullivan, K. M.; Schechter, L. E.;

Beyer, C. E.; Andree, T. H. Synthesis and biological evaluation of

novel compounds within a class of 3-aminochroman derivatives with

dual 5-HT_1Areceptor and serotonin transporter afﬁnity. J. Med. Chem.

2006, 49, 4785–4789.

(27) Al Neirabeyeh, M.; Reynaud, D.; Podona, T.; Ou, L.; Perdicakis, C.;

Coudert, G.; Guillaumet, G.; Pichat, L.; Gharib, A.; Sarda, N. Methoxy

and hydroxyl derivatives of 3,4-dihydro-3-(di-n-propyl-amino)-2H-1-

benzopyrans: new synthesis and dopaminergic activity. Eur. J. Med.

Chem. 1991, 26, 497–504.

(28) Hammarberg, E. M.; Johansson, L. G.; Ross, S. B.; Thorberg, S. O.

3-(N-Isopropyl-N-n-propylamino)-5-(N-isopropyl)carbamoylchro-

man. WO 93/07135, April 15, 1993.

(29) Evenden, J. L.; Hammarberg, E. M.; Hansson, H. S.; Hellberg, S. E.;

Johansson, L. G.; Lundkvist, J. R. M.; Ross, S. B.; Sohn, D. D.;

Thorberg, S. O. (R)-5-Carbamoyl-8-ﬂuoro-3-N,N-disubstituted-amino-

3,4-dihydro-2H-1-benzopyrans. U.S. 5,616,610, April 1, 1997.

(30) Hanson, S.; Johansson, L.; Sohn, D. D. Process for the manufacture

of 3-N,N-Dicyclobutylamino-8-ﬂuoro-3,4-dihydro-2H-1-benzopyran-

5-carboxamide. U.S. 6,197,978 B1, March 6, 2001.

References

(1) Greenberg, P. E.; Stiglin, L. E.; Finkelstein, S. N.; Berndt, E. R. The

economic burden of depression in 1990. J. Clin. Psychiatry 1993, 54,

405–418.

(2) Trivedi, M. H.; Kleiber, B.; Greer, T. L. Remission and recovery in

depression treatment. Drug DeV. Res. 2005, 65, 335–343.

(3) Blackwell, B. Adverse effects of antidepressant drugs. Part 1: monoamine

oxidase inhibitors and tricyclics. Drugs 1981, 21, 201–219.

(4) Evrard, D. A.; Harrison, B. L. Recent approaches to novel antidepres-

sant therapy. Annu. Rep. Med. Chem. 1999, 34, 1–9.

(5) Katz, M. M.; Tekell, J. L.; Bowden, C. L.; Brannan, S.; Houston,

J. P.; Berman, N.; Frazer, A. Onset and early behavioral effects of

pharmacologically different antidepressants and placebo in depression.

Neuropsychopharmacology 2004, 29, 566–579, and references therein.

(6) Kreiss, D. S.; lucki, I. Effects of acute and repeated administration of

antidepressant drugs on extracellular levels of 5-hydroxytryptamine

measured in vivo. J. Pharmacol. Exp. Ther. 1995, 274, 866–876.

(7) (a) Briner, K.; Dodel, R. C. New approaches to rapid onset antidepres-

sants. Curr. Pharm. Des. 1998, 4, 291–302. (b) Schechter, L. E.; Kelly,

M. G. An overview of 5-HT_1Areceptor antagonists: historical

perspective and therapeutic targets. Serotonin 1997, 2, 299–309.

(8) Artigas, F. Selective serotonin/noradrenaline reuptake site inhibitors

(SNRIs). CNS Drugs 1995, 4, 79–89.

(9) Evrard, D. A. Recent strategies for the development of new antidepres-

sant drugs. Annu. Rep. Med. Chem. 2006, 41, 23–38.

(10) Montgomery, S. A. Venlafaxine: a new dimension in antidepressant

pharmacotherapy. J. Clin. Psychiatry 1993, 54, 119–126.

(11) Shelton, C.; Entsuah, R.; Padmanabhan, S. K.; Vinall, P. E. Venlafaxin

XR demonstrates higher rates of sustained remission compared to

ﬂuoxetine, paroxetine or placebo. Int. Clin. Psychopharmacol. 2005,

20, 233–238.

(12) Bymaster, F. P.; Lee, T. C.; Knadler, M. P.; Detke, M. J.; Iyengar, S.

The dual transporter inhibitor duloxetine: a review of its preclinical

pharmacology, pharmacokinetic proﬁle, and clinical results in depres-

sion. Curr. Pharm. Des. 2005, 11, 1475–1493.

(13) Ballesteros, J.; Callado, L. F. Effectiveness of pindolol plus serotonin

uptake inhibitors in depression, a meta-analysis of early and late

outcomes from randomized controlled trials. J. AffectiVe Disord. 2004,

79, 137–147, and references therein.

(14) Romero, L.; Arigas, F. Preferential potentiation of the effects of

serotonin uptake inhibitors by 5-HT_1Areceptor antagonists in the dorsal

(31) Hatzenbuhler, N. T.; Evrard, D. A.; Mewshaw, R. E.; Zhou, D.; Shah,

U. S.; Inghrim, J. A.; Lenicek, S. E.; Baudy, R. B.; Butera, J. A.;

7004 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21

Hatzenbuhler et al.

Sabb, S. L.; Failli, A. A.; Ramamoorthy, P. S. A Preparation of

3-Aminochroman and 2-Aminotetralin Derivatives, Useful in the

Treatment of Serotonin-Mediated Disorders. WO 2005/012291, Febru-

ary 10, 2005.

(37) Cheng, Y.-C.; Prusoff, W. H. Relationship between the inhibition

constant (K_i) and the concentration of inhibitor which causes 50%

inhibition (IC₅₀) of an enzymatic reaction. Biochem. Pharmacol.

1973, 22, 3099–3108.

(32) Davidsen, S. K.; May, P. D.; Summers, J. B. Di-tert-butyl N-

acylimidodicarbonates as isolable acylating agents: mild conversion

of primary carboxamides to substituted amides. J. Org. Chem. 1991,

56, 5482–5485.

(33) Block, M. H.; Boyer, S.; Brailsford, W.; Brittain, D. R.; Carroll,

D.; Chapman, S.; Clarke, D. S.; Donald, C. S.; Foote, K. M.;

Godfrey, L.; Ladner, A.; Marsham, P. R.; Masters, D. J.; Mee,

C. D.; O’Donovan, M. R.; Pease, J. E.; Pickup, A. G.; Rayner,

J. W.; Roberts, A.; Schoﬁeld, P.; Suleman, A.; Turnbull, A. V.

Discovery and optimization of a series of carbazole ureas as NPY5

antagonists for the treatment of obesity. J. Med. Chem. 2002, 45,

3509–3523.

(34) Ley, S. V.; Priour, A. Total synthesis of the cyclic peptide argyrin B.

Eur. J. Org. Chem. 2002, 3995–4004.

(35) Somei, M.; Karasawa, Y.; Kaneko, C. Selective mono-alkylation

of carbon nucleophiles with gramine. Heterocycles 1981, 16, 941–

949.

(38) Ramamoorthy, J. D.; Ramamoorthy, S.; Papapetropoulos, A.;

Catravas, J. D.; Leibach, F. H.; Ganapathy, V. Cyclic AMP-

independent up-regulation of the human serotonin transporter by

staurosporine in chloriocarcinoma cells. J. Biol. Chem. 1995, 270,

17189–17195.

(39) Hall, M. D.; El Mestikawy, S.; Emerit, M. B.; Pichat, L.; Hamon,

M.; Gozlan, H. [³H]8-Hydroxy-2-(di-n-propylamino) tetralin bind-

ing to pre- and postsynaptic 5-hydroxytryptamine sites in var-

ious regions of the rat brain. J. Neurochem. 1985, 44, 1685–1696.

(40) Larenzo, S.; Birdsall, N. J. Pharmacological characterization of

acetylcholine-stimulated [³⁵S]-GTP gamma S binding mediated by

human muscarinic m1-m4 receptors: antagonist studies. Br. J.

Pharmacol. 1993, 109, 1120–1127.

(41) Morrow, A. L.; Creese, I. Characterization of R₁-adrenergic receptor

subtypes in rat brain: a reevaluation of [³H]WB4104 and [³H]prazosin

binding. Mol. Pharmacol. 1986, 29, 321–330.

(42) Dunlop, J.; Zhang, Y.; Smith, D. E.; Schechter, L. E. Characteriza-

tion of the 5-HT_1Areceptor functional coupling in cells expressing

the human 5-HT_1Areceptor as assessed with the cytosensor

microphysiometer. J. Pharmacol. Toxicol. Methods 1998, 40, 41–55.

(36) Cheetham, S. C.; Viggers, J. A.; Slater, N. A.; Heal, D. J.; Buckett,

W. R. [³H]-Paroxetine binding in rat frontal cortex strongly

correlates with [³H]5-HT uptake: effect of administration of var-

ious antidepressant treatments. Neuropharmacology 1993, 32, 737–

743.

JM8007097

Article Doi

Advances toward new antidepressants with dual serotonin transporter and 5-HT1A receptor affinity within a class of 3-aminochroman derivatives. Part 2

DOI: 10.1021/jm8007097

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm8007097

Full text of DOI:10.1021/jm8007097

Products guided by the article

R&D Labs maybe for 843653-51-0

Relevant to this article

Hot Product