Synthesis and structure-activity relationships of selective norepinephrine reuptake inhibitors (sNRI) with a heterocyclic ring constraint

Article abstract of DOI:10.1016/j.bmcl.2008.07.050

The design, synthesis and SAR of a series of heterocyclic ring-constrained norepinephrine reuptake inhibitors are described. As racemates, the best compounds compare favorably with atomoxetine (IC₅₀'s < 10 nM) in potency at the transporter.

Full text of DOI:10.1016/j.bmcl.2008.07.050

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4495–4498
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and structure–activity relationships of selective norepinephrine
reuptake inhibitors (sNRI) with a heterocyclic ring constraint
a
a
a
a
a
Sarah Hudson a, , Mehrak Kiankarimi , Wendy Eccles , Yalda S. Mostofi , Marc J. Genicot , Wesley Dwight ,
*
Beth A. Fleck b, Kathleen Gogas c, Warren S. Wade a,
*
a Department of Medicinal Chemistry, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
b Department of Pharmacology, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
c Department of Neuroscience, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design, synthesis and SAR of a series of heterocyclic ring-constrained norepinephrine reuptake inhib-
itors are described. As racemates, the best compounds compare favorably with atomoxetine
(IC50’s < 10 nM) in potency at the transporter.
Received 28 May 2008
Revised 11 July 2008
Accepted 14 July 2008
Available online 17 July 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Norepinephrine
Serotonin
Monoamine reuptake inhibition has been an effective therapeu-
tic intervention in a variety of CNS diseases, starting with depres-
sion and recently expanding to include chronic pain, ADHD, and
stress urinary incontinence.1–4 We were interested in multiple
indications in the therapeutic spectrum of selective norepineph-
rine reuptake inhibitors (sNRI), and initiated work directed toward
finding potent and selective compounds with good pharmaceutical
properties and minimal risk of drug–drug interactions. In the pre-
vious paper NBI 80532 (compound 2), a rigid and potent sNRI with
a significant risk of drug–drug interactions was discovered.5 Fur-
ther work toward improving the drug-like properties of active
compounds is reported here.
CF3
O
O
O
N
H
N
N
H
H
1, atomoxetine
2
5
Figure 1. Structures of atomoxetine 1, NBI 80532 2, and the SSRI MDL 27777A
(racemic) 5. MDL 28618A (R,S) is the active enantiomer.17
The previous paper described a new ring constraint for the 3-
aryloxypropylamine scaffold of atomoxetine that maintained po-
tency and selectivity for NET but was significantly more rigid than
atomoxetine 1.5 However, in compound 2 (Fig. 1), potent inhibition
of CYP2D6 was also observed. In an effort to separate the two activ-
ities, other ring connectivities were explored. Our attention was
drawn to the SSRI MDL 28618 (compound 5),6 an additional isomer
of the indane core described previously.5 Compound 5 had similar
activity to fluoxetine, but the series selectivity profile had not been
published. In order to determine the scope and selectivity of this
ring constraint, both carbocyclic and heterocyclic cores were syn-
thesized and studied.
Synthesis of the indane compounds followed the literature pro-
cedure.7,8 Synthesis of other core structures was generalized as
shown in Scheme 1. All bicyclic ketones are commercially available
except for isochromanone 12, which was made in three steps from
2-iodobenzyl alcohol 30. In all cases, Mannich reactions generated
the tertiary amine ketones 15–19 that could then be converted
predominantly to the cis amino alcohols 20–24 with L-selectride
(ꢀ9:1 cis:trans) or to a mixture favoring the trans alcohols 25–29
with sodium borohydride (1:2–5 cis:trans). The two diastereomers
were readily distinguishable by 1H NMR coupling constant of the
benzyloxy proton. When X was a heteroatom, selectivities with
L-SelectrideÒ decreased to 2:1 but could be restored with the more
bulky reducing agent LS-SelectrideÒ. In general, 15–29 were di-
methyl substituted because the Mannich reactions gave high yields
of crystalline products, and demethylation was accomplished in
good yield.
* Corresponding authors. Tel.: +1 858 617 7600; fax: +1 858 617 7601.
E-mail addresses: shudson@neurocrine.com (S. Hudson), warrenswade@gmail.-
com (W.S. Wade).
URL: http://www.neurocrine.com (S. Hudson).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.050

4496
S. Hudson et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4495–4498
OH
R1
N
R2
Y
b
O
O
X
X
R1
20-24
a
N
R2
HCl
c
Y
Y
X
OH
.
R1
10 X-Y = OCH2
11 X-Y = SCH2
12 X-Y = CH2O
13 X-Y = CH2S
14 X-Y = CH2CH2
15-19
N
R2
a R1 = R2 = Me
Y
X
b R1 = R2 = allyl
c R1 = Me, R2 = Bn
25-29
O
d
O
e, f
OH
O
I
I
12
30
31
Scheme 1. Reagents and conditions: (a) R1R2NH, paraformaldehyde, concd HCl, EtOH or IPA, 16 h, 70 °C, 50–75%; (b) L-selectride, THF, À30 °C, 60–80%; (c) NaBH4, MeOH,
À15 °C, 60–90%; (d) allyl bromide, NaH, THF, 16 h, rt, quant; (e) Pd(OAc)2, PPh3, Et3N, CH3CN, 2 h, DR; (f) O3, pyr., DCM, MeOH, 5 h, À78 °C then DMS, rt, 3 h, 78% (3 steps).
Formation of final compounds followed the sequence shown in
Scheme 2 for the chromane example. Starting from the cis alcohol
20a, SNAr substitution with an aryl fluoride9 followed by deprotec-
tion with 1-chloroethyl chloroformate generated the target sec-
ondary methyl amines predominantly with retention of the cis
configuration. Higher temperatures and extended reaction times
led to epimerization as has been observed with 1.10 Conversely, a
Mitsunobu reaction starting with either cis or trans alcohols gener-
ated predominantly the trans product, for example, 44a. When X or
Y is a heteroatom, the product was almost exclusively trans, prob-
ably due to neighboring group participation by the aminomethyl
side chain in the more stabilized intermediate.11 This was consis-
tent with the expectation that the trans isomer, having only pseud-
oequatorial substitution, was more stable thermodynamically.
For the key 2-methyl substitution, generation of the cis product
35a using 2-fluorotoluene was extremely low yielding, so the alter-
native literature method was employed.12 SNAr with the tert-butyl
imine of 2-fluorobenzaldehyde followed by hydrolysis to the alde-
Cl
O
R
O
b
N
N
H
O
O
32a
33 R = Cl
36 R = Me
a
O
b
OH
O
O
c, d
N
e, f, g
N
N
O
O
O
20a
34a
35a
h
O
O
b
N
N
H
O
O
44a
45
Scheme 2. Reagents and conditions: (a) aryl fluoride, for example, 2-chlorofluorobenzene for 32a, NaH and potassium benzoate or KOtBu, DMSO or DMA, 16 h, 70 °C, 30–80%;
(b) ACE-Cl, diisopropylethylamine, 1,2-DCE, 1 h, 45 °C, then MeOH, rt, 32–75%; (c) tert-butyl-[1-(2-fluoro-phenyl)-meth-(Z)-ylidene]-amine, NaH, potassium benzoate, DMSO,
2.5 h, 70 °C; (d) AcOH, THF, H2O, 3 h, rt, 57% (2 steps); (e) NaBH4, MeOH, 30 min, 0 °C, 97%; (f) SOCl2, DCM, 30 min, 0 °C; (g) Zn, AcOH, H2O, 3 h, rt, 86% (2 steps); (h) substituted
phenol, for example, o-cresol for 44a, TMAD, n-Bu3P, THF, 16 h, 0 °C–rt, 25–80%.

S. Hudson et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4495–4498
4497
hyde 34a proceeded in reasonable yield. Conversion to the methyl
in the reduction, alcohol to chloride, reduction sequence, then
demethylation under the standard conditions gave 36. This alter-
native route was used to produce the other cis 2-methyl aryl
substituted products 37, 39, 41, and 43 as well as to control substi-
tution of the asymmetric difluoroaromatic used to produce 61.
More electron rich 2-substituents gave even lower yields of the
nucleophilic substitution product. Fortunately, the ethyl deriva-
tives 52 and 64 could be generated by SNAr with the appropriate
2-fluorovinylbenzene followed by reduction with hydrazine.13 In
this manner, the thiomethyl compound 57 could be generated
using 2-fluorophenyl methyl sulfoxide followed by reduction with
dimethylsulfide/trimethylsilyl chloride.14 The 2-methoxy and 2-
ethoxy substitutents were synthesized only in the trans
diastereomer.
Several other compounds in the tables required additional syn-
thetic steps. The primary amines 37 and 41 were best generated
using diallylamine in the Mannich reaction to produce 15b and
17b. After installation of the aryloxy ring, the allyl groups were re-
moved by palladium mediated deprotection.15 Compound 47 was
generated by oxidation of Boc protected 46 with mCPBA followed
by acidic deprotection.16
Table 1
SAR of indanesa
R2
R2
All compounds were tested for their ability to inhibit norepi-
nephrine, serotonin, and dopamine uptake in HEK cell lines that
had been stably transfected with the human transporters.16 Com-
pounds were initially tested in two independent dose response
experiments, and those with reasonable potency at NET were re-
tested multiple times. Atomoxetine was included on all assay
plates as a standard control, and assay variability was reasonable
for a functional assay with SEM values typically below 0.2 log
units. A table of SEM values at all three transporters is included
O
O
N
R1
N
R1
3-7
8, 9
Compound
R1
R2
NETb
SERTb
DATb
5300
>10,000
>10,000
>10,000
5700
>10,000
4000
3000
as Supplementary material. Potency values are reported as IC50
,
3
4
5
6
7
8
9
H
Me
H
Me
H
H
2-Me
2-Me
4-CF3
4-CF3
Naphthyl
2-Me
Naphthyl
10
5000
880
4500
30
35
400
5
1600
6500
190
230
140
1500
90
180
1400
47
though with the neurotransmitter concentrations in each assay
well below their respective Km values, little difference would be
expected between the measured IC50 and Ki of the compounds. Ac-
tive compounds were further tested in transporter binding assays
by competition with the appropriate radioligand.
Compound 5 has been described in the literature as an analog of
fluoxetine,6 and the active enantiomer is known.17 However, the
available literature does not describe the SAR of the series. As this
ring structure represents an alternate ring connection to our previ-
ous work,5 it presented a possible starting point for a new series of
H
Atomoxetine, 1
Nisoxetine
Fluoxetine
15
2000
2400
6000
a
Data are the average of two or more independent measurements.
IC50 (nM) for monoamine uptake.
b
Table 2
SAR of six-membered ring hetereocycles with 2 substitutiona
R3
O
R3
O
R3
O
R3
O
R2
N
R2
R2
R2
N
N
N
R1
R1
Y
R1
Y
R1
X
X
33-38
39-43
45-47
R3
48-50
Compound
X
Y
R1
R2
NETb
SERTb
DATb
33
35a
36
37
38
39
40
41
42
43
45-Ic
46
47
48
49
50
O
O
O
O
S
Me
Me
Me
H
Me
Me
Me
H
Me
Me
Me
Me
Me
Me
Me
Me
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2-Cl
36
700
26
46
145
18
17
95
140
40
200
430
8000
140
100
180
6000
1100
1400
3800
3600
2500
1800
6000
2700
2600
1200
3100
10,000
3500
1100
2700
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
4200
2-Me
2-Me
2-Me
2-Cl
2-Me
2-Cl
O
O
O
S
2-Me
2-Cl
C
2-Me
2-Me
2-Me
2-Me
2-Cl
O
S
SO2
7500
>10,000
>10,000
4100
O
O
C
2-Me
2-Me
>10,000
a
Data are the average of two or more independent measurements.
IC50 (nM) for monoamine uptake.
Active enantiomer, racemic not tested.
b
c

4498
S. Hudson et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4495–4498
Table 3
CF3 > 2-ethyl = 2-SMe > 2-F. As had been observed in the previous
series, 2-methoxy and 2-ethoxy were significantly less active.5
SAR of the aryloxy ring in the isochromanesa
A
combination of 2-methyl and fluoro substitution to modify the
hydrophobicity of the ring gave compounds with single digit nM
activity. Selectivity was found to be a function of the fluorine posi-
tion, with 4-F the least selective at 10-fold (61), and the other iso-
mers more selective. Increasing the size of the 2-substituent also
improved the selectivity of the trans diastereomer, so that com-
pounds like 53 had encouraging profiles. However, increasing the
size of the 3-substituent from fluorine to chlorine led to a greater
than 10-fold drop in selectivity (compare 59 and 60, Table 3).
Overall, this series produced a number of potent and selective
compounds. Further development required characterization in
safety and in vivo efficacy studies as separate enantiomers. To
determine whether these compounds could have improved profiles
over atomoxetine, multiple racemic compounds were tested for
inhibition of CYP2D6 and CYP3A4, and their oxidative stability in
human liver microsomes was measured. Multiple compounds,
including 40 and 49, exhibited inhibition of CYP2D6 in the micro-
molar range and low oxidative clearance. Because it was possible
that both enantiomers contributed to the observed selectivity pro-
file (see Boot et al. for an example18), further characterization was
reserved for the separate individual enantiomers. Results will be
reported in the next publication in this series.
R2
R2
R1
O
R1
O
NH
NH
O
O
51-52, 57-65
53-56
Compound
R1
R2
NETb
SERTb
DATb
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
F
H
75
15
36
480
240
45
14
6
8
9
13
11
7
3100
4500
3000
>10,000
>10,000
640
6000
5600
1100
100
>10,000
>10,000
>10,000
>10,000
>10,000
4000
>10,000
>10,000
>10,000
1400
>10,000
>10,000
>10,000
>10,000
>10,000
Et
H
Et
H
OMe
OEt
SMe
SMe
CF3
Me
Me
Me
Me
Me
Et
H
H
H
H
H
3-F
3-Cl
4-F
5-F
6-F
6-F
6-F
150
4000
3100
1400
3200
8
4
CF3
a
Data are the average of two or more independent measurements.
IC50 (nM) for monoamine uptake.
b
Acknowledgments
The authors are indebted to Mr. Rajesh Huntley and Mr. Brock
Brown for technical support in the transporter assays and Mr.
Nathan Kozon, Mr. William Ban, Ms. Anna Aparicio, Ms. Hua Wang,
and Mr. Andrew Fisher for technical support in the CYP inhibition
and microsomal stability assays.
sNRI compounds. Synthesis of the minimal set of analogs was
undertaken to determine the correlation with the atomoxetine/flu-
oxetine (Lilly) series. As shown in Table 1, the activity of this series
correlated well with the Lilly compounds. 2-Substitution on the
aromatic ring produced compound 3, a potent and selective NET
inhibitor with properties very similar to atomoxetine. The di-
methyl tertiary amine 4 was much less active at NET. Naphthyl
substitution generated significant activity at both transporters,
see 7 and 9. Comparing cis and trans diastereomers, the cis was
more potent at NET and more selective for NET. As was observed
with NBI 80532 (2), the more potent compounds have one of the
aryloxy or aminomethyl substituents in a pseudoaxial orientation.
Thus this ring connection produced compounds with the desired
primary potency and selectivity. Additionally, this series was more
flexible than 2 but less flexible than the Lilly series.
Characterization of the active enantiomer of 3 revealed that it
was also similar to atomoxetine in both CYP2D6 inhibition and oxi-
dative stability as described in the following paper. Accordingly, a
number of ring variants were tested with the objective of modifying
theshapeandgeometryenoughtoimprovethesepropertieswithout
losing potency or selectivity at NET. As shown in Table 2, the tetralin
analog 43 was 4-fold less potent than the similar indane. Again the
cis diastereomer 43 was more potent than the trans 50. Replacement
of the benzylic carbon to generate the chromane 36 gave about a 2-
fold improvement, and at this stage the isochromane 39 was similar
though generally trended toward higher potency. As with the ind-
anes, the trans diastereomer was less potent and less selective for
NET. The same trends were observed for 2-chloro substitution.
Replacement with sulfur was less successful, generally resulting in
a 10-fold drop in potency, and conversion of thio to sulfone (47) re-
sulted in a 20-fold loss of activity. Overall, the results are consistent
with potency increasing as the ring size decreases.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.07.050.
References and notes
1. Jann, M. W.; Slade, J. H. Pharmacotherapy 2007, 27, 1571.
2. Babu, R. P. K.; Maiti, S. N. Heterocycles 2006, 69, 539.
3. Zhou, J. Drugs Future 2004, 29, 1235.
4. Wernicke, J. F.; Iyengar, S.; Ferrer-Garcia, M. D. Curr. Drug Ther. 2007, 2, 161.
5. Wu, D.; Pontillo, J.; Ching, B.; Hudson, S.; Gao, Y.; Fleck, B. A.; Gogas, K.; Wade,
W. S. Bioorg. Med. Chem. Lett. 2008, 18, 4224.
6. Olivier, B.; Soudijn, W.; van Wijngaarden, I. Prog. Drug Res. 2000, 54, 59.
7. Cregge, R. J.; Wagner, E. R.; Freedman, J.; Margolin, A. L. J. Org. Chem. 1990, 55,
4237.
8. Freedman, J.; Dudley, M. W. Preparation and formulation of
(aryloxy)indanamines and related compounds useful as antidepressants and
as inhibitors of synaptic norepinephrine and serotonin uptake. US 5149714.
9. Berglund, R. A. Org. Process Res. Dev. 1997, 1, 328.
10. Koenig, T. M.; Mitchell, D. Tetrahedron Lett. 1994, 35, 1339.
11. Okuda, M.; Tomioka, K. Tetrahedron Lett. 1994, 35, 4585.
12. Heath, P. C.; Ratz, A. M.; Weigel, L. O. Preparation of tomoxetine. WO
2000058262.
13. Hu, T.; Stearns, B. A.; Campbell, B. T.; Arruda, J. M.; Chen, C.; Aiyar, J.;
Bezverkov, R. E.; Santini, A.; Schaffhauser, H.; Liu, W.; Venkatraman, S.; Munoz,
B. Bioorg. Med. Chem. Lett. 2004, 14, 2031.
14. Samanen, J. M.; Brandeis, E. J. Org. Chem. 1988, 53, 561.
15. Garro-Helion, F.; Merzouk, A.; Guibe, F. J. Org. Chem. 1993, 58, 6109.
16. Kiankarimi, M.; Hudson, S.; Dwight, W. J.; Wade, W. S. Preparation of
benzofuranylmethylamines, chromanylmethylamines, and thiochromanyl-
methylamines as monoamine reuptake inhibitors. US 2006252818.
17. Michals, D.; Smith, H. E. Chirality 1993, 5, 20.
Selectivity was further explored with the most potent heterocy-
cle, the isochromane, as shown in Table 3. Potency was correlated
with the size and hydrophobicity of the aryloxy substitution, 2-
18. Boot, J.; Cases, M.; Clark, B. P.; Findlay, J.; Gallagher, P. T.; Hayhurst, L.; Man, T.;
Montalbetti, C.; Rathmell, R. E.; Rudyk, H.; Walter, M. W.; Whatton, M.; Wood,
V. Bioorg. Med. Chem. Lett. 2005, 15, 699.