Discovery of a new series of monoamine reuptake inhibitors, the 1-amino-3-(1H-indol-1-yl)-3-phenylpropan-2-ols

Article abstract of DOI:10.1016/j.bmcl.2009.07.053

A novel series of monoamine reuptake inhibitors, the 1-amino-3-(1H-indol-1-yl)-3-phenylpropan-2-ols, have been discovered by combining virtual and focused screening efforts with design techniques. Synthesis of the two diastereomeric isomers of the molecule followed by chiral resolution of each enantiomer revealed the (2R,3S)-isomer to be a potent norepinephrine reuptake inhibitor (IC₅₀ = 28 nM) with excellent selectivity over the dopamine transporter and 13-fold selectivity over the serotonin transporter.

Full text of DOI:10.1016/j.bmcl.2009.07.053

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5029–5032
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a new series of monoamine reuptake inhibitors,
the 1-amino-3-(1H-indol-1-yl)-3-phenylpropan-2-ols
a
a
a,
Callain Y. Kim a, , Paige E. Mahaney , Oliver McConnell , Yingru Zhang , Eric Manas a,à
,
*
Douglas M. Ho b,§, Darlene C. Deecher c, Eugene J. Trybulski a
a Chemical Sciences, Wyeth Research, Arcola Road, Collegeville, PA 19426, USA
b Department of Chemistry, Princeton University, Princeton, NJ 08543, USA
c Women’s Health, Wyeth Research, Arcola Road, Collegeville, PA 19426, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of monoamine reuptake inhibitors, the 1-amino-3-(1H-indol-1-yl)-3-phenylpropan-2-ols,
have been discovered by combining virtual and focused screening efforts with design techniques. Synthe-
sis of the two diastereomeric isomers of the molecule followed by chiral resolution of each enantiomer
revealed the (2R,3S)-isomer to be a potent norepinephrine reuptake inhibitor (IC50 = 28 nM) with excel-
lent selectivity over the dopamine transporter and 13-fold selectivity over the serotonin transporter.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 22 May 2009
Revised 6 July 2009
Accepted 8 July 2009
Available online 23 July 2009
Keywords:
Vasomotor symptoms
Monoamine reuptake inhibitor
Norepiephrine reuptake inhibitor
NRI
NE
5-HT
Vasomotor symptoms (VMS), referred to as hot flushes and
night sweats, are the result of fluctuating sex steroid hormone lev-
els and are the principal menopausal symptom1 for which women
seek medical treatment. To date, estrogens are the most effective
therapeutics for the alleviation of these symptoms; however, these
therapies are not acceptable or indicated for all women. Accord-
ingly, there is a clear unmet medical need to develop effective
non-hormonal therapies to expand the choice of available thera-
peutic options. Norepinephrine (NE) has been shown to stimulate
areas of the hypothalamus that are important in temperature reg-
ulation.2 Estrogens modulate the activity of both the norepineph-
rine (NE) and the serotonin (5-HT) systems thus maintaining
optimal neurotransmitter levels in the brain.3 Depletion of estro-
gens may produce unstable concentrations of these specific neuro-
transmitters resulting in fluctuations in body temperature.
Previously, we disclosed that restoration of NE levels in ovariecto-
mized rats by the administration of a non-selective norepinephrine
reuptake inhibitor (NRI), desipramine (1) could alleviate VMS and
restore normal thermoregulation.4 Thus, the goal of this program
was to identify and develop novel and selective NRIs for the treat-
ment of temperature dysregulation associated with menopause
and for evaluation toward additional conditions that have been re-
ported to be ameliorated by NRIs including major depressive disor-
der (MDD),5 attention deficit hyperactivity disorder (ADHD),6 and
certain pain disorders including fibromyalgia7,8 and low back
pain.7
Our search for a novel and potent NRI scaffold was initiated by
screening a subset of the Wyeth compound inventory that was se-
lected based on similarity to biogenic amines (2000 compounds).
In parallel, the entire Wyeth compound collection was virtually
screened using a Unity flexible 3-D search of a published 3-point
pharmacophore containing a basic amine and two aromatic rings.9
The distances were adjusted10 to ensure the inclusion of the known
NRIs desipramine (1), nisoxetine (2), and reboxetine (3) (Fig. 1). Fil-
ters aimed at identifying brain penetrable, lead-like compounds
(MW <400, cLog P <5, rotatable bonds <9, TPSA <80) were applied
* Corresponding author. Tel.: +1 484 865 4429; fax: +1 484 865 9399.
E-mail address: kimcy@wyeth.com (C.Y. Kim).
N
O
NH
O
NHCH3
OEt
O
OCH3
NHCH3
desipramine (1)
Present address: Bristol-Myers Squibb, Princeton, NJ 08543, USA.
Present address: GlaxoSmithKline, King of Prussia, PA 19406, USA.
à
nisoxetine (2)
( )-reboxetine (3)
§
Present address: Department of Chemistry and Chemical Biology, Harvard
University, Cambridge, MA 02138, USA.
Figure 1. Structures of known NRIs.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.053

5030
C. Y. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5029–5032
to the list of virtual hit compounds. A selected set of compounds
was screened for norepinephrine reuptake inhibition in MDCK-
Net6 cells stably transfected with the human norepinephrine
transporter (hNET) using previously described procedures.11 Both
screens identified b-hydroxyamines 4 and 5 (Fig. 2) that signifi-
O
R1
a
R1
OH
+
OH
N
H
N
OH
( )-8
( )-9
cantly inhibited NE uptake at a concentration of 1 lM. We were
most interested in indoles 5 due to their potential novelty, and
when the 3-phenyl group, found in known monoamine reuptake
inhibitors, was incorporated, a potent, albeit large, lead molecule
( )-6 that exhibited some selectivity over the human serotonin
transporter (hSERT) was identified. Efforts to reduce the molecular
weight to obtain a more lead-like molecule led to compound ( )-7a
which maintained potent NE uptake inhibition but lost selectivity
over hSERT. Compound ( )-7a became an early lead molecule in
our program. Initial SAR studies of the phenylpropanolamine scaf-
fold have recently been disclosed.12
R1
R1
b
c
R2
2
N
OTs
N
3
N
R3
OH
OH
( )-7
( )-10
Scheme 1. Synthesis of ( )-erythro-1-amino-3-(1H-indol-yl)-3-phenylpropan-2-
ols. Reagents and conditions: (a) KOH, DMSO, 70 °C or NaH, t-BuOH, Ti(i-PrO)4,
CH2Cl2, rt; (b) TsCl, pyridine, rt; (c) NHR2R3, K2CO3, CH3CN, reflux (when R3 = H;
excess NH2Me, or NH2Et, MeOH, rt).
Synthesis of the erythro diastereomeric pair of 1-amino-3-(1H-
indol-1-yl)-3-phenylpropan-2-ols13 [( )-7, Scheme 1] began with
trans-1-phenylglycidol ( )-8 which was prepared from trans-cin-
namyl alcohol using peracetic acid.14 Regioselective epoxide open-
ing with indole was accomplished using either pulverized
potassium hydroxide in DMSO or sodium t-butoxide in the pres-
ence of titanium isopropoxide. Selective conversion of the primary
alcohol in compound ( )-9 proceeded smoothly with p-toluenesul-
fonyl chloride in pyridine to form tosylate ( )-10 which was dis-
placed with an amine to form target compounds ( )-7 each as an
erythro diastereomeric pair. Diol ( )-9 was also converted to the
threo diastereomer ( )-13 using the procedure13 outlined in
Scheme 2. Selective protection of the primary alcohol in diol ( )-
9 was accomplished with p-nitrobenzoyl chloride in pyridine at
À10 °C followed by conversion of the secondary alcohol into the
mesylate using methansulfonyl chloride and triethylamine to form
compound ( )-11. Deprotection of the primary alcohol using aque-
ous sodium hydroxide allowed an in situ intramolecular displace-
ment of the mesylate with inversion to form epoxide ( )-12.
Subsequent regioselective ring opening of the epoxide with
methylamine in methanol afforded target compound ( )-13 as
the threo diastereomeric pair.
R1
R1
a,b
OPNB
OMs
OH
N
N
OH
( )-9
( )-11
R1
R2
R1
2
c
d
N
N
3
N
R3
O
OH
( )-12
( )-13
Scheme 2. Synthesis of ( )-threo-1-amino-3-(1H-indol-yl)-3-phenylpropan-2-ols.
Reagents and conditions: (a) p-nitrobenzoyl chloride, pyridine, À10 °C (b) MsCl,
TEA, CH2Cl2, 0–5 °C (c) NaOH, dioxane, rt (d) excess NH2Me, MeOH, rt.
H2
H
HO
(S)
Compounds 7a and 13a (R1 and R3 = H, R2 = CH3) were sepa-
rated into single enantiomers (7a.a, 7a.b, 13a.a and 13a.b) using
a chiral supercritical fluid chromatography (SFC) technique.15 Sub-
sequently, the absolute stereochemistry of each enantiomer was
examined using vibrational circular dichroism (VCD), electronic
circular dichroism (ECD), and NMR methods.16 The absolute
stereochemistry of eutomer 7a.a, was confirmed by single crystal
X-ray analysis17 to possess the (2R,3S) configuration18 (Fig. 3).
All target compounds were tested initially as racemates for
their ability to block NE reuptake. Compounds that exhibited at
N
CH3
(R)
H3
N
7a.a
Figure 3. ORTEP view of compound 7a.a from single crystal X-ray analysis.
least 40% inhibition at a concentration of 1
l
M were tested for
selectivity over hSERT and the human dopamine transporter
(hDAT). These assays were performed as previously described.11
An examination of the effect of amine moiety size in the erythro-
diastereomers (7a–l, Table 1) revealed that only primary amines
and secondary amines with small alkyl groups were tolerated for
hNET activity. Consequently, while ethylamine 7j provided modest
hNET potency (IC50 = 1200 nM), a drop-off in activity was observed
with the iso-propylamine analog 7h which exhibited only 13% inhi-
Z
Z
X
N
R
Aryl
Aryl
N
N
R
Aryl
OH
OH
4
5
2
2
Z
N
NHCH3
N
N
R
3
3
bition of hNET at 1 lM. The secondary methylamine analog 7a was
OH
OH
the most potent NRI of the group (IC50 = 137 nM) while the pri-
mary amine 7l was tolerated but was less effective (IC50 = 560 nM).
When either a 3-methyl (7n–p) or 5-F substituent was incorpo-
rated (7q–s) on the indole ring, the size of the amine moiety
provided an identical trend toward hNET potency that was
observed in the unsubstituted analogs, that is, NHCH3 preferred
over NH2 which was preferred over NHCH2CH3. Incorporation of
( )-6
( )-7a
MW = 280
MW = 451
hNET IC50 = 137 nM
hSERT IC50= 126 nM
hNET IC50 = 197 nM
hSERT IC50> 1000 nM
Figure 2. Conversion of NRI scaffolds identified from screening into a Discovery
lead molecule.

C. Y. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5029–5032
5031
Table 1
Characterization of 1-amino-3-(1H-indol-yl)-3-phenylpropan-2-ol analogs at the human norepinephrine, serotonin and dopamine transporters
R1
R2
2
N
N
3
R3
OH
R1
R2
R3
hNET uptake IC50
nM (StDev)
,
hSERT uptake IC50
nM (StDev)
,
Ratio of hSERT uptake
IC50 /hNET uptake IC50
hDAT binding %
a
b
Compd
Stereochem
inhibition @1 l
Md
c
7a
( )-Erythro
2R,3S
2S,3R
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
CH3
CH3
H
H
H
137 (56)e
28 (11)e
104 (41)e
5%f
126 (31)e
358 (108)e
144 (50)e
NT
NT
NT
1
13
1
—
—
—
—
—
—
—
7%
19%
18%
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
7%
NT
0%
7a.a
7a.b
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
7p
7q
7r
7s
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Erythro
( )-Threo
1S,2S
–(CH2)2N(CH3)(CH2)2–
–(CH2)2NH(CH2)2–
– (CH2)2O(CH2)2–
4-PyridylCH2
CyclohexylCH2
Benzyl
(CH3)2CH–
CH3CH2
CH3CH2
CH3
H
CH3
CH3CH2
CH3
H
CH3CH2
CH3
H
CH3
CH3
CH3
CH3
CH3
34%f
18%f
H
H
H
H
CH3
H
CH3
H
H
H
H
H
H
H
H
H
H
H
24%f
NT
41%f
2833 (283)e
NT
26%f
13%f
NT
NT
38%f
1200g
25%f
298 (74)e
NT
0.2
—
0.1
—
2
560g
50g
NT
2-CH3
3-CH3
3-CH3
3-CH3
5-F
5-F
5-F
5-Cl
5-OCH3
H
45%f
1100g
42 (6)e
321g
2200g
251 (20)e
299(100)e
NT
6
1
0%
6%
NT
NT
NT
NT
2%
4%
0%
11%f
—
1
—
0.2
0.1
1
118 (32)e
32%f
95g
NT
7t
49g
10g
7u
13a
13a.a
13a.b
230g
31g
79 (32)e
415 (134)e
150 (42)e
74 (14)e
918g
H
H
H
H
2
1
1R,2R
161 (4)e
31%
a
Inhibition of NE uptake in MDCK-Net6 cells, stably transfected with hNET. Desipramine (IC50 = 3.4 + 1.6 nM) was used as a standard.
Inhibition of serotonin uptake in JAR cells, stably transfected with human SERT. Fluoxetine (IC50 = 9.4 + 3.1 nM) was used as a standard.
Unitless value as a ratio in which higher numbers represent relatively greater NET selectivity. A value of 1 represents no selectivity.
b
c
d
e
f
Inhibition of [3H]WIN35,428 binding to membranes from CHO cells expressing recombinant hDAT. Mazindol (22.1 + 6.5 nM) was used as a standard.
Data is the average of between 2 and 6 independent experiments, each run in triplicate.
Percent inhibition measured at a concentration of 1000 nM.
Data is the average of three triplicate runs; standard error (SE) < 30% of the mean. NT = not tested.
g
the 3-methyl substituent on indole (7n–p) also provided a modest
improvement in selectivity for hNET over hSERT when compared
to the unsubstituted analogs.
In summary, we have identified a series of 1-amino-3-(1H-in-
dol-1-yl)-3-phenylpropan-2-ols as NRIs. Diastereoselective syn-
thesis and separation of enantiomers led to the finding that the
(2R,3S)-isomer, 7a.a, was a potent NRI with modest selectivity over
the serotonin transporter. Compound 7a.a was also selective over
the dopamine transporter as well as a select panel of 5-HT and
adrenergic receptors. In addition, 7a.a possessed suitable pharma-
ceutical properties. Together, these data supported 7a.a as a viable
and promising Discovery lead molecule. Additional SAR toward the
development of this series will be the subject of future
communications.
The effect of stereochemistry on hNET potency and selectivity
was examined and a surprising result was observed. Diastereo-
meric pairs 7a and 13a exhibited similar potencies for both hNET
and hSERT reuptake inhibition. The same was true for the threo-
enantiomers 13a.a and 13a.b. Separation of the erythro-enantio-
mers 7a.a and 7a.b, however, revealed that although (2S,3R)-enan-
tiomer 7a.b exhibited equivalent potencies for hNET and hSERT,
the (2R,3S)-enantiomer 7a.a provided modest selectivity (13-fold)
for hNET over hSERT. All compounds tested exhibited only weak
binding to hDAT.
Acknowledgements
Compound 7a.a was further assessed in a CNS panel for selec-
tivity and in pharmaceutical property profiling. In a competition
radioligand-binding assay in whole MDCK cells stably transfected
with hNET using [3H]-nisoxetine as the radioligand,11 7a.a exhib-
ited an IC50 value of 23 nM, similar to its functional IC50 of
28 nM. It also had no affinity for the 5-HT1a and 5-HT2a receptors
The authors are grateful to Rebecca Dooley, a member of our
chemical technologies group, for NMR structural elucidation of
compound 7a.a to assist VCD analysis.
References and notes
or the
a1-, and a2-isoforms of the adrenergic receptor. The com-
1. (a) Waldinger, M. D.; Berendsen, H. H. G.; Schweitzer, D. H. Maturitas 2000, 36,
165; (b) Heath, H. III; Plouffe, L, Jr. WO9944601A1, 1999; (c) Freedman, R.;
Dinsay, R. Fertility Sterility 2000, 74, 20; (d) Plouffe, L., Jr. Del. Med. J. 1997, 69,
481.
2. (a) Kumar, V. N. Proc. Natl. Acad., Part B: Biol. Sci. 2003, 69, 507; (b) Riedel, W.;
Dorward, P. K.; Korner, P. I. J. Auton. Nerv. 1981, 3, 525; (c) Lahti, H.; Pyornila, A.;
Hissa, R. Experientia 1980, 36, 1188.
pound also exhibited good aqueous solubility at pH 7.4 (>50 lg/
mL), good PAMPA permeability (4.6 Â 10À6 cm/s) and <20% inhibi-
tion of CYP3A4, CYP2D6 and CYP2C9 at 3 lM. When incubated in
liver microsomes at a concentration of 1 lM, the t of 7a.a across
species was 9.8 min in rat, 15.9 min in dog, 9.7 min in monkey and
>60 min in human.
½
3. Panek, D.; Dixon, W. J. Pharm. Exp. Therapeutics 1986, 236, 646.

5032
C. Y. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5029–5032
4. Deecher, D.; Merchenthaler, I.; Leventhal, L.; Sipe, K.; O’Connor, L.
WO2004035035A1, 2004.
5. (a) Dubini, A.; Bosc, M.; Polin, V. J. Psychopharmacol. 1997, 11, S17; (b) Ban, T.
A.; Gaszner, P.; Aguglia, E.; Batista, R., et al Hum. Psychopharmacol. 1998, 13,
S29.
6. Bymaster, F. P.; Katner, J. S.; Nelson, D. J., et al Neuropsychopharmacology 2002,
27, 699.
7. Krell, H. V.; Leuchter, A. F.; Cook, I. A.; Abrams, M. Psychosomatics 2005, 46, 379.
8. Bergan, T. Can. J. Psychiatry 2004, 49, 499.
Lett. 2008, 18, 6067; (c) McComas, C. C.; Vu, A. T.; Mahaney, P. E., et al Bioorg.
Med. Chem. Lett. 2008, 18, 4929; (d) Mahaney, P. E.; Vu, A. T.; McComas, C. C.,
et al Bioorg. Med. Chem. 2006, 14, 8455.
13. For detail synthesis of the erythro and threo diastereomers of 1-amino-3-(1H-
indol-1-yl)-3-phenylpropan-2-ols see: Kim, C. Y., Mahaney, P. E.; Trybulski, E. J.
et al. U.S. Patent 7517899, 2009.
14. Henegar, K. E.; Mancini, S. E.; Maisto, K. D. WO 2000039072A1, 2000.
15. McConnell, O.; Bach, A., II; Balibar, C., et al Chirality 2007, 19, 658.
16. McConnell, O.; He, Y.; Nogle, L.; Sarkahian, A. Chirality 2007, 19, 716.
17. Stout, G.; Jensen, L. X-ray Structure Determination: A Practical Guide, 2nd ed.;
Wiley: New York, 1989.
9. Enyedy, I. J.; Zaman, W. A.; Sakamuri, S.; Kozikowski, A. P.; Johnson, K. M.;
Wang, S. Bioorg. Med. Chem. Lett. 2001, 11, 1113.
10. The distances used were: center of aromatic ring A (RA) to the basic amine
(BA) = 6.20 0.50 Å; center of aromatic ring B (RB) to BA = 5.60 0.70 Å; center
of RA to center of RB = 6.05 0.80 Å).
18. Crystallographic data for compound 7a.a reported in this Letter have been
deposited with the Cambridge Crystallographic Data Centre as supplementary
publication no. CCDC 737858.
11. Mahaney, P. E.; Vu, A. T.; McComas, C. C., et al J. Med Chem. 2008, 51, 4038.
12. (a) Vu, A. T.; Cohn, S. T.; Terefenko, E. A., et al Bioorg. Med. Chem. Lett. 2009, 19,
2464; (b) Zhang, P.; Terefenko, E. A.; McComas, C. C., et al Bioorg. Med. Chem.