4-Phenyl tetrahydroisoquinolines as dual norepinephrine and dopamine reuptake inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.09.050

Novel 4-phenyl tetrahydroisoquinolines that inhibit both dopamine and norepinephrine transporters were designed and prepared. In this Letter, we describe the synthesis, in vitro activity and associated structure-activity relationships of this series. We also report the ex vivo NET occupancy of a representative compound, 41.

Full text of DOI:10.1016/j.bmcl.2012.09.050

Bioorganic & Medicinal Chemistry Letters 22 (2012) 7219–7222
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
4-Phenyl tetrahydroisoquinolines as dual norepinephrine and dopamine
reuptake inhibitors
Anthony D. Pechulis a, James P. Beck c, , Matt A. Curry c, , Mark A. Wolf a, Arthur E. Harms a, Ning Xi a,
Chet Opalka a, Mark P. Sweet a, Zhicai Yang a, A. Samuel Vellekoop a, Andrew M. Klos a, Peter J. Crocker a,
Carla Hassler a, Mia Laws c, , Douglas B. Kitchen a, Mark A. Smith c, , Richard E. Olson b,c, , Shuang Liu a,
Bruce F. Molino a,
⇑
a AMRI, 26 Corporate Circle, PO Box 15098, Albany, NY 12212-5098, USA
b Bristol Myers Squibb R&D, 5 Research Parkway, Wallingford, CT 06492, USA
c DuPont Pharmaceuticals, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 June 2012
Revised 13 September 2012
Accepted 17 September 2012
Available online 25 September 2012
Novel 4-phenyl tetrahydroisoquinolines that inhibit both dopamine and norepinephrine transporters
were designed and prepared. In this Letter, we describe the synthesis, in vitro activity and associated
structure–activity relationships of this series. We also report the ex vivo NET occupancy of a representa-
tive compound, 41.
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Norepinephrine reuptake inhibitor
Dopamine reuptake inhibitor
Dual dopamine and norepinephrine
reuptake inhibitors
DAT
NET
ADHD
Depression
Monoamine reuptake inhibitors block the uptake of synaptic
serotonin, norepinephrine and dopamine by the presynaptic sero-
tonin, norepinephrine and dopamine transporters (SERT, NET and
DAT, respectively). For several decades, selective serotonin reup-
take inhibitors (SSRIs) and dual serotonin-norepinephrine reup-
take inhibitors (SNRIs) have been used to treat a number of CNS
disorders, such as depression, anxiety, obsessive compulsive disor-
der and pain.1,2
Nomifensine (1) is a monoamine reuptake inhibitor, with a un-
ique dual inhibition activity at the norepinephrine and dopamine
transporters and much less activity at the serotonin transporter.
It was marketed for use as an antidepressant in the 1970s and
was also investigated for the treatment of attention deficit hyper-
activity disorder (ADHD).3,4 Nomifensine was withdrawn from the
market in the 1980s mainly due to its association with immune
hemolytic anemia.5,6 Recent studies have explored the potential
role of metabolism in these adverse events.7 In particular, the pres-
ence of an aniline group has been implicated as a source of reactive
metabolites.7a,b We wanted to explore novel dual NET/DAT reup-
take inhibitors as such agents could provide benefit in the treat-
ment of depression, and might also have an improved profile for
the treatment of ADHD as compared to current therapies. Although
stimulants remain the treatment of choice for patients with ADHD
due to exceptional response rate, their use is associated with abuse
potential.8 On the other hand, atomoxetine, a NET inhibitor and the
only approved non-stimulant for the treatment of ADHD, was gen-
erally considered less efficacious compared to the stimulants.8
A
NET/DAT dual inhibitor might overcome these limitations of exist-
ing ADHD therapies.
N
CH3
⇑
Corresponding author.
NH2
E-mail address: bruce.molino@amriglobal.com (B.F. Molino).
These aurhors were employees of DuPont Pharmaceuticals when this work was
conducted.
1
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.09.050

7220
A. D. Pechulis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7219–7222
We initially explored substitution on the THIQ core (R1 through
R4 in Table 1). To simplify the comparison of binding data, these
initial analogues did not have substitution on the 4-phenyl ring.
As seen in Table 1, 1 has dual activity at NET and DAT and is essen-
tially inactive at SERT. Removal of the anilino group decreased po-
tency at NET and DAT by 2- to 3-fold (Compound 12 vs 1).
Substitution at R1 with methyl, methoxy and hydroxy groups
(compounds 13–15) had no benefit. We were pleased to find out
that methyl or methoxy substitution at R2 (compounds 16 and
18) increased potency at NET with Ki below 100 nM along with
2.5- to5-fold selectivity for NET over DAT and weaker SERT activity
as desired. The R1–R2-dimethyl analogue 20 did not have a favor-
able potency at either NET or DAT. Potency and NET/DAT selectiv-
ity suffered when R3 was methyl or methoxy (21–22). Substitution
at R4 attenuated activity at both NET and DAT (Compound 23).
From this data, we concluded that substitution at the 7-position
(R2) of the THIQ is preferred for the dual DAT/NET activities.
We next elaborated the pendant aryl ring (Table 2). We focused
on analogues with substitutions on the 6- and 7-positions of the
THIQ core (R1 and R2, respectively), since these compounds seemed
to have substantial activity for NET and DAT and demonstrated the
desired selectivity over SERT. These compounds were typically sep-
arated into enantiomers by chiral HPLC prior to testing. It was
determined that the more active enantiomer was (+) in all cases,
as seen in Table 2 [comparison of compounds 24-(+) and 24-(ꢀ)
and 25-(+) and 25-(ꢀ)]. It was later determined using X-ray crys-
In pursuit of novel NET/DAT inhibitors, our medicinal chemistry ef-
forts explored novel tetrahydroisoquinolines (THIQs) lacking the
aniline moiety of 1 with high NET potency (Ki <15 nM), 2- to 10-fold
selectivity over DAT and >25-fold selectivity for NET over SERT.
Here we describe the preparation of a series of novel 4-phenyl THIQ
derivatives, along with the in vitro and in vivo biological evaluation
of representative compounds.
Our general synthetic method is shown in Schemes 1 and 2.9,12
Reductive amination of aldehydes 2 afforded methylamines 3.
Haloacetophenones 5 were prepared by bromination of methyl ke-
tones 4 or by addition of an aryl Grignard reagent to Weinreb
amides 6. Alkylation of amine 3 with 5 afforded ketones 7. Reduc-
tion of 7 provided the hydroxyl intermediates 8.
Scheme 2 shows the acid catalyzed cyclization of 8 which pro-
vides the THIQ core. When R2 = R3 = H and R1 – H (8a), 9 was ob-
tained as the sole product. Similarly, for the synthesis of
compounds 21 and 22 (8a, R1 = R2 = H, R3 = CH3 and OCH3, respec-
tively) the cyclization gave a single regioisomer due to the symme-
try of the cyclization precursor. When R1 = H and R2 – H (8b), there
are two open positions for the cyclization to occur, thus a mixture
of regioisomers 10 and 11 was obtained with 10 as the predomi-
nant regioisomer in most cases. Compounds 10 and 11 were read-
ily separated by column chromatography. Single enantiomers of
compounds 9–11 typically were obtained by chiral HPLC. The final
products 9–11 were generally converted to the corresponding salt
(maleate, fumarate or hydrochloride) for biological testing.
R3
R2
R3
R2
a, b
NHCH3
CHO
R
R1
R1
3
2
R
BrMg
O
O
O
c
R
X'
O
X'
H3C
N
CH3
X' = Cl only
d
CH3
6
4
5
O
OH
R3
R2
R
R3
R2
R3
R2
R
e
f
NHCH3
N
N
CH3
CH3
R1
R1
R1
3
7
8
Scheme 1. Reagents and conditions: (a) CH3NH2, MeOH; (b) NaBH4, H2O, MeOH, 0 °C to rt, 2.5 h; (c) HBr, HOAc, rt, 2 h or Bu4NBr3, THF, rt, 18 h; (d) THF, ꢀ10 °C to rt, 20 min;
(e) 5, Et3 N, CH2Cl2, rt, 18 h; (f) NaBH4, EtOH, 0 °C to rt, 1.5 h.
R
R2 = H, R1 = H or R1 = R2 = H, R3 = H
OH
R3
R2
R
a
R3
R2
N
N
CH3
CH3
R1
R1
8a
9
R
N
R
R1 = H, R2 = H
OH
R2
R
b
N
R2
CH3
+
R1
N
R2
CH3
CH3
8b
R1
(33–52%)
H
(19–39%)
10
11
Scheme 2. Reagents and conditions (a) H2SO4, CH2Cl2, 0 °C to rt, 20 min, 58–73%; (b) H2SO4, CH2Cl2, 0 °C to rt, 20 min.

A. D. Pechulis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7219–7222
7221
Table 1
Table 3
In vitro inhibition of NET, DAT and SERTa
NET Ki and occupancy data (10 mg/kg)
Compd.
NET Ki (nM)
Occupancy (%)
Human
Rat
20
R4
R3
R2
1
23
67
70
49
Duloxetine
Compound 41
5.97
7.1
3.9
6.3
N
CH3
R1
Compd.
Substitution
Ki (nm)
Two of these compounds (27 and 28) provided the first compounds
with single-digit nM NET activity. Introduction of a methyl group
at R1 (compounds 30 and 31) resulted in loss of selectivity over
DAT.
R1
R2
R3
R4
NET
47
100
136
137
238
35
DAT
SERT
1
NHH2
CH3
OCH3
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
84
276
157
129
371
112
233
82
210
227
107
173
1840
2057
1430
4224
2050
1510
1220
767
1450
352
1750
2850
1570
12
13
14
15
16
17
18
19
20
21
22
23
H
H
H
H
Compounds 32–43 are all 7-substituted THIQ analogues. The
7-F THIQ 32 had relatively low activity at NET and DAT. Com-
pounds 33–36 (all 7-methoxy or phenolic analogues) displayed
good potency for NET and DAT and generally good selectivity over
SERT. Compounds substituted with methyl at the 7-position were
generally very potent at NET (37–43). Selectivity over SERT was
also generally good, with the exception of the p-methyl phenyl
analogue 37, which was not sufficiently selective with respect to
DAT or SERT. A number of compounds such as 26, 40, and 41 in
Table 2 met our initial binding criteria for the three transporters.
We selected compound 41 as a representative example for further
biological testing.
H
H
H
H
CH3
Et
H
H
54
31.8
OCH3
OEt3
CH3
H
H
H
H
H
H
CH3
OCH3
H
115
135
116
1020
806
CH3
H
H
H
CH3
N.D.
a
Data are for racemates. N.D. not determined.
In order to assess the in vivo activity and brain penetrance of
compound 41, we selected a rat ex vivo binding assay and mea-
sured the occupancy levels at NET in frontal cortex. Animals were
dosed orally with vehicle, duloxetine (as a NET positive control), 1
or compound 41, then sacrificed after one hour.10,11 The results dis-
played in Table 3 show the NET occupancy levels after oral admin-
istration of duloxetine, 1 and compound 41(10 mg/kg) after 1 hour.
Compound 41 demonstrated significant NET occupancy under
these conditions, comparable to that of duloxetine and 1.
Compound 41 was found to be selective when screened for off-
target activity. The compound displayed excellent selectivity
against a panel of approximately 50 receptors and enzymes (Nova-
screen), including MAO-A and MAO-B. Compound 41 was also neg-
ative in an Ames assay. However, when tested in a CYP450
inhibition assay, 41 exhibited moderate inhibition of CYP2D6 with
Table 2
In vitro inhibition NET, DAT and SERTa
R6
R7
R5
N
R2
CH3
R1
Compd.
Substitution
R5
Ki (nm)
R1
R2
R6
R7
NET
47
21.5
6450
28
630
12.3
7.2
4.4
DAT
SERT
1
NH2
NH2
NH2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
H
H
H
Cl
Cl
F
Cl
CH3
H
H
H
H
H
H
H
H
84
43
13400
38
2057
935
1550
25
1776
266
81.7
24-(+)
24-(ꢀ)
25-(+)
25-(ꢀ)
26
an IC50 value of 0.48 lM. A number of other analogues in this series
also exhibited various amounts of CYP2D6 inhibition (data not
shown).
356
98.8
44.7
30.1
In summary, a series of substituted 4-phenyl tetrahydroisoquin-
olines was prepared in an effort to find new treatments for CNS
disorders. Removal of the aniline group of 1 and optimization pro-
vided potent NET/DAT inhibitors. A representative compound in
the series, compound 41, demonstrated potent NET and DAT dual
inhibition with desirable selectivity over SERT. These values met
our selectivity criteria. Compound 41 was shown to penetrate
the CNS and occupy the site of action (NET). However, the possibil-
ity of drug–drug interactions predicted by CYP2D6 inhibition hin-
dered progression of this series. Further optimization of this series
was pursued and the results will be reported in due course.
27
28
H
H
F
F
102
29
30
31
H
CH3
CH3
H
H
H
CH3
F
H
F
F
CH3
H
H
H
13.5
37
19
61
58
10.6
118
ND
822
32
33
34
35
36
37
38
H
H
H
H
H
H
H
F
H
F
CH3
F
F
H
F
H
CH3
F
H
H
H
H
H
H
H
29
9
687
77
25
21.5
64
27
1470
530
479
111
287
121
594
OCH3
OCH3
OCH3
OH
CH3
CH3
7.3
16.5
13
15
14
H
F
CH3
F
106
39
40
41
42
43
H
H
H
H
H
CH3
CH3
CH3
CH3
CH3
Cl
F
H
H
F
H
H
F
Cl
H
H
F
H
H
H
5
5.8
7.1
47
35
36
78
36.2
40.5
291
338
2149
231
116
2310
Acknowledgment
The ex vivo occupancy study was carried out by RenaSci., Not-
tingham, UK. The authors thank Stephanie Koch and Charles L. Ing-
alls for their synthetic assistance.
a
Except where noted, compound is the (+)-enantiomer N.D.: not determined.
References and notes
tallography that the (+)-enantiomer of a representative compound
41 has the (S) configuration (data not shown). Halogenated or
methylated phenyl analogues (such as compounds 26–29) afforded
high potency at NET, and varied selectivities versus DAT and SERT.
1. Walter, M. W. Drug Dev. Res. 2005, 65, 97.
2. Liu, S.; Molino, B. F. Annu. Rep. Med. Chem. 2007, 42, 13.
3. Corman, S. L.; Fedutes, B. A.; Culley, C. M. Am. J. Health Syst. Pharm. 2004, 61,
2391.

7222
A. D. Pechulis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7219–7222
4. Hunt, P.; Hannengiesser, M. H.; Raynaud, J. P. J. Pharm. Pharmacol. 1974, 26, 370.
5. (a) Habibi, B.; Cartron, J. P.; Bretagne, M.; Rouger, P.; Salmon, C. Vox Sang. 1981,
40, 79.
chloride (3 ꢁ 100 ml). The combined organic extracts were washed with brine
and dried over anhydrous sodium sulfate, followed by filtration and
concentration in vacuo. Purification by column chromatography on silica gel
eluting with hexanes/ethyl acetate (3/1) provided the amino alcohol (4.3 g) as
a yellow oil; 1H NMR (300 MHz, CDCl3) d 7.08 7.30 (m, 7H), 4.73 (t, J = 6.0 Hz,
1H), 3.60 (ABq, JAB = 14.0 Hz, 2H), 2.55 (d, J = 8.0 Hz, 2H), 2.36 (s, 3H), 2.31 (s,
3H); CI MS m/z = 274 [C17H20NFO+H]+.
6. Harrison, J. H., Jr.; Jollow, D. J. J. Pharm. Exp. Ther. 1986, 238, 1045.
7. (a) Yu, J.; Brown, D. G.; Burdette, D. Drug Metab. Dispos. 2010, 38, 1767; (b) Yu,
J.; Mathisen, D. E.; Burdette, D.; Brown, D. G.; Becker, C.; Aharony, D. Drug
Metab. Dispos. 2010, 38, 46; (c) Obach, R. S.; Dalvie, D. K. Drug Metab. Dispos.
2006, 34, 1310; (d) Stephan, A. F.; Walker, D. P.; Bauman, J.; Price, D. A.; Baillie,
T. A.; Kalgutkar, A. S.; Aleo, M. D. Chem. Res. Toxicol. 2011, 24, 1345.
8. (a) Wigal, S. B. CNS Drugs 2009, 23, 21; (b) Findling, R. L. Clin. Ther. 2008, 30,
942.
9. Beck, J. P.; Curry, M. A.; Smith, M. A. U.S.7,163,949. The binding assay is
described in this patent.
10. Bymaster, F. P.; Dreshfield-Ahmad, L. J.; Threlkeld, P. G. Neuropsychopharmacology
2001, 25, 871. and references therein.
Step B: The product from Step A (1.0 g, 4.0 mmol) was stirred in methylene
chloride (100 ml) and treated dropwise with concentrated sulfuric acid (98%,
7.0 ml) over 3 min. After stirring for 1 h, the reaction mixture was diluted with
ice chips and made basic with 25% aqueous ammonium hydroxide. The
reactions mixture was extracted with methylene chloride (3 ꢁ 100 ml) and the
organic extracts combined, dried over anhydrous sodium sulfate, filtered, and
concentrated in vacuo. Purification by column chromatography, eluting with
hexanes/ethyl acetate (3/1), afforded the desired tetrahydroisoquinoline as a
11. Molino, B.F.; Liu, S.; Guzzo, P. R.; Beck, J.P. U.S.7,541,375 B2.
12. The synthesis of these THIQ analogues is exemplified by the preparation of 43
as described below:
yellow oil: 1H NMR (300 MHz, CDCl3)
d 6.89 7.00 (m, 5H), 6.75 (d,
J = 8.0 Hz,1H), 4.21 (t, J = 7.0 Hz, 1H), 3.64 (ABq, JAB = 15.0 Hz, 2H), 3.02 (m,
1H), 2.56 (m, 1H), 2.41 (s, 3H), 2.29 (s, 3H); CI MS m/z = 256 [C17H18NF+H]+.
Step C: The product from Step B was subjected to chiral HPLC separation
employing a Chiral Technologies.
Step A: m-Tolualdehyde (1.66 g, 14.0 mmol) was treated with methylamine
(40% aqueous, 1.39 ml, 18.0 mmol) in methanol (20 ml) at room temperature.
The reaction was stirred for 20 min and treated with sodium borohydride
(0.26 g, 7.0 mmol) portionwise. The reaction was stirred for 1 h and treated
with 30-fluoro-2-bromoacetophenone (3.0 g, 14.0 mmol) followed by stirring
for 45 min at room temperature. The reaction was finally treated with sodium
borohydride (0.52 g, 14.0 mmol) portionwise, and stirring continued overnight.
The reaction was diluted with water (100 ml) and extracted with methylene
Chiracel™ AD column (5 ꢁ 50 cm) eluting with hexanes/isopropanol (9/1) to
afford the (R), ½a D25
ꢂ
ꢀ16.3 (c = 0.498, MEOH) and (S), ½a D25
ꢂ
+16.3 (c = 0.476, MeOH)
enantiomers in order of elution. The (S)-(+) enantiomer was treated with maleic
acid (1.0 equiv) and the resultant maleate salt filtered and dried to constant
weight. (S)-(+)-2,7-dimethyl-4-(3-fluorophenyl)-1,2,3,4-tetrahydroisoquinoline,
maleate salt: mp 172–173.5 °C.