Discovery of an 8-methoxytetrahydroisoquinoline derivative as an orally active N-type calcium channel blocker for neuropathic pain without CYP inhibition liability

Article abstract of DOI:10.1016/j.bmc.2015.05.053

In lead optimization efforts starting from the tetrahydroisoquinoline (S)-1, we identified 2-{[(2R)-2-hydroxypropyl]amino}-1-[(1S)-8-methoxy-1-phenyl-3,4-dihydroisoquinolin-2(1H)-yl]ethanone ((1S)-8t) as a novel orally active small-molecule N-type calcium channel blocker without CYP inhibition liability. CYP3A4 inhibition profile was improved by reducing the lipophilicity of compound (S)-1. Moreover, introduction of a methoxy group to the C-8 position of tetrahydroisoquinoline led to identification of (1S)-8t, which eliminated CYP2D6 inhibition liability. Oral administration of (1S)-8t exerted efficacy in a rat spinal nerve ligation (SNL) model of neuropathic pain with an ED₅₀ value of 2.8 mg/kg.

Full text of DOI:10.1016/j.bmc.2015.05.053

Bioorganic & Medicinal Chemistry 23 (2015) 4638–4648
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Discovery of an 8-methoxytetrahydroisoquinoline derivative
as an orally active N-type calcium channel blocker for neuropathic
pain without CYP inhibition liability
a
a
a
a
Takashi Ogiyama a, , Koichi Yonezawa , Makoto Inoue , Naoko Katayama , Toshihiro Watanabe ,
Seiji Yoshimura a, Takayasu Gotoh a, Tetsuo Kiso a, Akiko Koakutsu b, Shuichiro Kakimoto a, Jun-ichi
Shishikura a
⇑
a Drug Discovery Research, Astellas Pharma Inc., 21 Miyukigaoka, Tsukuba, Ibaraki 305-8585, Japan
b Drug Discovery Research, Astellas Pharma Inc., 2-1-6 Kashima, Yodogawa-ku, Osaka 532-8514, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In lead optimization efforts starting from the tetrahydroisoquinoline (S)-1, we identified 2-{[(2R)-2-hy-
droxypropyl]amino}-1-[(1S)-8-methoxy-1-phenyl-3,4-dihydroisoquinolin-2(1H)-yl]ethanone ((1S)-8t)
as a novel orally active small-molecule N-type calcium channel blocker without CYP inhibition liability.
CYP3A4 inhibition profile was improved by reducing the lipophilicity of compound (S)-1. Moreover,
introduction of a methoxy group to the C-8 position of tetrahydroisoquinoline led to identification of
(1S)-8t, which eliminated CYP2D6 inhibition liability. Oral administration of (1S)-8t exerted efficacy in
a rat spinal nerve ligation (SNL) model of neuropathic pain with an ED50 value of 2.8 mg/kg.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 20 April 2015
Revised 28 May 2015
Accepted 29 May 2015
Available online 3 June 2015
Keywords:
N-type calcium channel
CaV2.2
Pain
CYP2D6
CYP3A4
1. Introduction
Unlike the case with ziconotide, no abnormal behavior was
observed in test animals. In addition, reducing the lipophilicity of
N-type (Cav2.2) calcium channels are primarily localized at
presynaptic terminals in afferent Ad- or C-fiber endings of lamina
I and II, and play a prominent role in controlling the transmission
of nociceptive information. Numerous lines of evidence have
proved that N-type calcium channels are profoundly involved in
the pain pathway.1 The marketed peptide drug ziconotide exhibits
clinical efficacy in the treatment of chronic pain via the blockade of
N-type calcium channels. However, the clinical utility of ziconotide
is limited by its requirement for intrathecal administration and
severe CNS adverse effects.2 Therefore, recent studies have focused
on identifying small-molecule N-type calcium channel blockers
which overcome these drawbacks.3,4
We previously identified 1-isopropyltetrahydroisoquinoline
(R)-2 as a potent N-type calcium channel blocker for the treatment
of neuropathic pain in lead optimization efforts from 1-cyclohexyl-
tetrahydroisoquinoline (S)-1 (Table 1).5a Oral administration of (R)-
2 improved mechanical allodynia in a rat spinal nerve ligation
(SNL) model of neuropathic pain with an ED50 value of 2.5 mg/kg.
compound (S)-1 by replacing the 1-cyclohexyl group with a 1-
isopropyl moiety led to the elimination of CYP3A4 inhibition
liability while maintaining N-type calcium channel inhibitory
potency ((S)-1 versus (R)-2).
Human cytochrome P450 (CYP) enzymes are membrane-associ-
ated proteins which contain heme as a prosthetic group. These
enzymes are found in the inner membrane of the mitochondria
and endoplasmic reticulum of cells. Some isoforms of these
enzymes (i.e. CYP1A2, 2D6, 2C9, 2C19, and 3A4) play a crucial role
in the metabolic elimination of drugs. CYP2D6 participates in the
metabolism of about 25% of clinically available drugs, although
hepatic abundance of CYP2D6 is less than 5% of all CYP enzymes.6
Inhibition of CYP2D6 occasionally causes unwanted drug-drug
interactions (DDIs), as shown by combination of the selective sero-
tonin reuptake inhibitors (SSRIs) fluoxetine and paroxetine, which
are potent CYP2D6 inhibitors, and tricyclic antidepressants (TCAs),
which are metabolized by CYP2D6.6 To mitigate risks in drug
development and maximize convenience in clinical use, drug dis-
covery efforts should focus on minimizing issues related to
CYP2D6-mediated DDIs.
⇑
In an attempt to eliminate CYP2D6 inhibition liability while
maintaining potency for N-type calcium channel inhibition,
Corresponding author. Tel.: +81 29 863 6683; fax: +81 29 852 5387.
E-mail address: takashi.ogiyama@astellas.com (T. Ogiyama).
http://dx.doi.org/10.1016/j.bmc.2015.05.053
0968-0896/Ó 2015 Elsevier Ltd. All rights reserved.

T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
4639
Table 1
In vitro pharmacological properties of (S)-1 and (R)-2
b
Compound
N-type FLIPR IC50
(l
M)
CYP2D6 IC50
(
lM)
CYP3A4a
clogD7.4
Preincubation time
0 min
30 min
OH
N
N
1.0
1.9
0.60
2.9
87%
99%
67%
92%
3.48
2.55
H
O
(S)-1
MeO
OH
N
N
(RH)-2
O
a
Residual activities (%) of HLM were evaluated using midazolam as a probe substrate. Details are described in the experimental section.
clogD values at pH 7.4 were calculated with ACD/PhysChem Batch (version 12.01).
b
introduction of various substituents on the phenyl ring of 1-iso-
of the formalin test in mice. Selected compounds were assessed for
inhibitory activity against CYP2D6 using a fluorescence substrate.
The reversible and time-dependent inhibitory (TDI) effects of
selected compounds were evaluated using midazolam as a probe
substrate to monitor changes in CYP3A4 activity.
propyl tetrahydroisoquinoline was examined.5 Through this effort,
we identified 6-methoxy derivative (R)-2. The inhibitory activity of
(R)-2 against CYP2D6 was approximately 5-fold weaker than that
of lead compound (S)-1 ((S)-1 versus (R)-2) however, (R)-2 still
exhibited moderate potency for CYP2D6 inhibition. Further optimiza-
tion of reducing CYP2D6 inhibition liability starting from 1-isopropyl
tetrahydroisoquinoline (R)-2 therefore appeared difficult. In addition,
the previous results raised the possibility that conformational alter-
ations might help to reduce CYP2D6 inhibition liability, particularly
those induced by steric repulsion of two bulky substituents between
the C-1 and C-8 positions of tetrahydroisoquinoline.5b
In keeping with the implications of the last investigation,5 we
conducted lead optimization from (S)-1 while retaining the bulky
1-cyclohexyl moiety. It is well documented that lipophilic com-
pounds tend to have CYP inhibition liability.8b,9 In addition,
reducing the lipophilicity of 1-substituted tetrahydroisoquinoline
led to an improvement in CYP3A4 inhibition liability while
retaining potency for N-type calcium channel inhibition in our
investigation.5b To explore another approach to reducing the
lipophilicity of (S)-1, we investigated the effect of transformation
of a second hydrophobic cyclohexyl moiety which linked to a basic
nitrogen through a glycine linker (Table 2). To reduce CYP3A4
liability, we designed compounds mainly reference to the clogD7.4
value. To ensure efficient screening, we prepared and evaluated the
diastereomixture of compounds except for (1S)-8q and 8t. Drastic
differences in in vitro pharmacological profiles between the
diastereomixture and biologically active isomer were not observed,
as exemplified by 8a and (1S)-8a (Table 2).
We therefore pursued a second approach to lead optimization
starting from (S)-1. Here, we describe the design and synthesis of
a novel series of tetrahydroisoquinolines, and the discovery of
(1S)-8t as an orally active small-molecule N-type calcium channel
blocker with the elimination of CYP inhibition liability.
2. Chemistry
Synthesis of tetrahydroisoquinoline derivatives 5–11 is illus-
trated in Scheme 1. Tetrahydroisoquinoline derivatives (5–7,
8a–8e, and 9–11) were prepared by substitution of the chloroac-
etamide 3a–3e with amines. (1S)-8a was synthesized from (1S)-
3a5a,13 with (2R)-1-amino-2-propanol. Tetrahydroisoquinoline
derivatives (8f–8p, (1S)-8q, 8r, 8s, (1S)-8t, and (1R)-8t) were
synthesized from the tetrahydroisoquinolines (4f–4p, (1S)-4q, 4r,
4s, (1S)-4t, and (1R)-4t) in a two-step sequence. Compound (1S)-
8q was synthesized from (1S)-4q, which was obtained from a
commercial source. Absolute configuration of (1S)-4t was
confirmed by vibrational circular dichroism (VCD) spectrum
analysis of its synthetic intermediate (1S)-5-bromo-8-methoxy-1-
phenyl-1,2,3,4-tetrahydroisoquinoline ((1S)-22).7 Preparations of
the chloroacetamides (3b–3e) and tetrahydroisoquinolines (4f–
4p, and 4r, 4s, (1S)-4t, and (1R)-4t) are described in Supporting
information.
Replacement of cyclohexyl moiety in compound
1 with
dimethyl group (5) resulted in an approximately 2-fold decrease
in N-type calcium channel blockade. Transformation to hydroxy-
butylamine (6 and 7) and hydroxypropylamine (8a and 9) led to
2- to 2.5-fold decreases in potency for N-type calcium channel
inhibition compared with compound 1. The hydrophobic analog
10 showed an equivalent effect on N-type calcium channel inhibi-
tion compared with the hydrophilic analogs 6 and 9. Another
hydrophobic analog 11 showed a slight decrease in potency for
N-type calcium channels compared with compound 1. Taken
together, these results demonstrated that there is no correlation
between inhibitory activity against N-type calcium channels and
the lipophilicity of this part. Compound 5 lacked an antinociceptive
effect in the formalin test in mice. Compounds 6 and 7 both
showed only weak antinociceptive activity. Compound 8a showed
an adequate antinociceptive effect, although compound 9 did not
produce any antinociceptive effect. The antinociceptive activity of
compound 10 was comparable to that of compound 8a.
Compound 11 showed slightly weak antinociceptive activity com-
pared with compound 8a and 10. Regarding CYP2D6 inhibition
liability, the relationship with the lipophilicity of the compound
was obscure as well as our previous study.5b Transformations
into hydrophilic groups (6, 8a, and 9) afforded small
3. Results and discussion
Inhibitory activity against N-type calcium channels was evalu-
ated using the fluorometric imaging plate reader (FLIPR) calcium
imaging assay in IMR-32 human neuroblastoma cells.
Experiments were conducted in the presence of nitrendipine to
block endogenous L-type calcium channels. The nociceptive
responses to chemical stimuli were evaluated in the second phase

4640
T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
6
6
7
R1
R1
N
N
7
Cl
a
R2
O
O
5-7, 8a, 9-11 (from 3a)
8b-8e (from 3b-3e)
3a: R1 = H
N
H
N
8d: R1 = 6-CN, R2 =
5: R1 = H , R2 =
6: R1 = H , R2 =
7: R1 = H , R2 =
H
3b: R1 = 7-Et
3c: R1 = 7-iPr
3d: R1 = 6-CN
3e: R1 = 7-CN
OH
OH
OH
N
N
H
8e: R1 = 7-CN, R2 =
H
OH
OH
OH
N
9: R1 = H , R2 =
N
H
H
OH
HO
N
H
8a: R1 = H, R2 =
10: R1 = H , R2 =
N
H
HO
8b: R1 = 7-Et, R2 =
8c: R1 = 7-iPr, R2 =
N
H
OH
OH
11: R1 = H , R2 =
N
N
H
H
5
5
6
6
b, a
R2
R2
1
NH
1
N
7
7
N
H
8
8
R1
R1
O
OH
4f: R1 = cyclohexyl, R2 = 5-F
4g: R1 = cyclohexyl, R2 = 5-MeO
4h: R1 = cyclohexyl, R2 = 6-F
4i: R1 = cyclohexyl, R2 = 6-Me
4j:R1 = cyclohexyl, R2 = 6-MeO
4k: R1 = cyclohexyl, R2 = 6-CONH2
8f: R1 = cyclohexyl, R2 = 5-F
8g: R1 = cyclohexyl, R2 = 5-MeO
8h: R1 = cyclohexyl, R2 = 6-F
8i: R1 = cyclohexyl, R2 = 6-Me
8j: R1 = cyclohexyl, R2 = 6-MeO
8k: R1 = cyclohexyl, R2 = 6-CONH2
4l: R1 = cyclohexyl, R2 = 7-F
4m:R1 = cyclohexyl, R2 = 7-Me
4n: R1 = cyclohexyl, R2 = 7-MeO
4o: R1 = cyclohexyl, R2 = 8-F
4p: R1 = cyclohexyl, R2 = 8-MeO
(1S)-4q: R1 = (1S)-phenyl, R2 = H
8l: R1 = cyclohexyl, R2 = 7-F
8m: R1 = cyclohexyl, R2 = 7-Me
8n: R1 = cyclohexyl, R2 = 7-MeO
8o: R1 = cyclohexyl, R2 = 8-F
8p: R1 = cyclohexyl, R2 = 8-MeO
(1S)-8q: R1 = (1S)-phenyl, R2 = H
8r: R1 = phenyl, R2 = 8-F
2
4r: R1 = phenyl, R = 8-F
2
4s: R1 = phenyl, R = 8-Me
8s: R1 = phenyl, R2 = 8-Me
(1S)-4t: R1 = (1S)-phenyl, R2 = 8-MeO
(1R)-4t: R1 = (1R)-phenyl, R2 = 8-MeO
(1S)-8t: R1 = (1S)-phenyl, R2 = 8-MeO
(1R)-8t: R1 = (1R)-phenyl, R2 = 8-MeO
Scheme 1. Reagents and conditions: (a) amine, K2CO3, MeCN, 60 °C, 3 h; (b) chloroacetyl chloride, sat. NaHCO3 aq, AcOEt, rt, 1 h.
improvements in CYP2D6 liability compared with compound 1.
Highly hydrophobic compounds 10 and 11 were potent CYP2D6
inhibitors. In contrast, CYP3A4 inhibition liability correlated with
the lipophilicity of a compound to some extent. The less lipophilic
compounds 6 and 7 (clogD7.4 = 2.60) showed an improvement in
5-Fluorine substituted analog 8f exhibited favorable potency for
N-type calcium channel inhibition. Introduction of the 5-methoxy
(8g) to compound 8a did not affect the enhancement of potency for
N-type calcium channel inhibition. Both compounds 8f and 8g
resulted in deterioration in the TDI profile of CYP3A4. Among com-
pounds substituted at the C-6 position of tetrahydroisoquinoline
(8h, 8i, 8j, 8d, and 8k), compound 8h (R = 6-F) showed relatively
favorable inhibitory activity against N-type calcium channels.
Introduction of hydrophilic 6-substituents, such as cyano (8d)
and carbamoyl (8k) groups, resulted in a decrease or complete loss
of potency for N-type calcium channels. Compound 8h (R = 6-F)
and 8j (R = 6-MeO) showed favorable CYP3A4 inhibition liability
as well as compound 8a. Incorporation of substituents (8l, 8m,
8b, 8c, 8n, and 8e) into tetrahydroisoquinoline at the C-7 position
resulted in relatively preferable N-type calcium channel inhibition,
except for the fluorine (8l) and cyano (8e) groups. In particular,
compound 8b (R = 7-ethyl) showed optimal potency in this study.
However, 7-substituted compounds (8b, 8c, 8l, and 8n), particu-
larly the lipophilic alkyl substituted analogs of 8b and 8c, showed
a deterioration of TDI profile of CYP3A4 compared with compound
8a (clogD7.4 = 3.63, 3.29 vs 2.53). Inhibitory activity of compound
8p (R = 8-methoxy) against N-type calcium channels was more
favorable than that of compound 8o (R = 8-fluorine). Regarding
potency for N-type calcium channel inhibition, apparent
the TDI liability of CYP3A4 compared with compound
1
(clogD7.4 = 3.48). The much less lipophilic analogs 8a and 9
(clogD7.4 = 2.53) exhibited favorable CYP3A4 inhibition profiles
compared with compounds 6 and 7 (clogD7.4 = 2.60). Hydrophobic
compounds 10 and 11, whose log D7.4 value were around 3.8,
showed marked deterioration in CYP3A4 inhibition profiles
compared with compound 1. Overall, compound 8a, which has a
(2R)-hydroxypropylamine group, exhibited a favorable profile. In
addition, the biologically active isomer (1S)-8a exhibited desirable
in vivo efficacy compared with diastereomixture of 8a, although
the other in vitro profiles of (1S)-8a are almost equivalent to those
of 8a. de Groot et al. provided knowledge-based general strategies
to address CYP2D6 inhibition liability.9b According to their
suggestion, small modifications to the periphery of the molecule,
such as the exchange of substituents on the benzene moiety,
could have a significant impact on reducing CYP2D6 inhibition
liability. Thus, we next investigated the effect of substituents on
the benzene ring in tetrahydroisoquinoline of compound 8a to
address CYP2D6 inhibition liability (Table 3).

T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
4641
Table 2
Effect of 1-cyclohexyltetrahydroisoquinoline derivatives on N-type calcium channel potency, CYP 2D6 and 3A4 inhibitory activities
CYP3A4b
Preincubation time
N
0 min
30 min
Formalin test % inhibitiona
CYP2D6 IC50
(lM)
clogD7.4
c
O
N-type FLIPR IC50
(l
M)
Compound no.
OH
N
H
1
0.77
62
0.78
86%
61%
3.48
N
H
5
1.7
1.5
1.9
2.0
1.5
7.7
19
27
39
0
N.T.
1.0
N.T.
87%
81%
92%
96%
N.T.
88%
83%
95%
91%
2.94
2.60
2.60
2.53
2.53
OH
N
H
6
OH
N
H
7
<0.3
1.1
OH
N
H
8a
9
OH
OH
N
H
1.6
HO
10
11
1.4
1.1
40
32
0.068
0.044
65%
69%
33%
46%
3.77
3.80
N
H
HO
N
H
N
N
H
1.7
55
1.4
105%
87%
2.53
O
OH
(1S)-8a
a
b
c
At 100 mg/kg p.o.
Residual activities (%) of HLM were evaluated using midazolam as a probe substrate. Details are described in the experimental section.
clogD values at pH 7.4 were calculated with ACD/PhysChem Batch (version 12.01).
structure–activity relationships (SARs) were not observed.
However, compounds which exhibited potent inhibitory activity
against N-type calcium channels (IC50 <2 lM) tended to possess
to (1S)-8a with decreased lipophilicity (clogD7.4 value (1S)-8q:
1.80 vs (1S)-8a: 2.53). Introduction of a fluorine group (8r)
showed preferable potency for N-type calcium channel inhibition
in spite of an exceptionally low lipophilic property. Inhibitory
activity of the methyl group (8s) and S-isomer of the methoxy
group ((1S)-8t) exhibited comparable potency for N-type calcium
channel inhibition to 8p and was superior to that of R-isomer
(1R)-8t. As expected, these compounds (8r, 8s, (1S)-8t and (1R)-
8t) exhibited desirable CYP3A4 inhibition profiles due to the
weaker lipophilic properties than 8p (log D7.4 values: 1.33–2.35
vs 3.07). Regarding CYP2D6 inhibition, introduction of methyl
group 8s resulted in over 2-fold attenuation compared with
compound (1S)-8q. Moreover, (1S)-8t completely eliminated
CYP2D6 inhibition liability. These results are consistent with our
expectation and indicated that the attenuation of CYP2D6
inhibition liability resulted from conformational changes,
induced by steric repulsion between the 8-substituted group and
1-phenyl moiety of tetrahydroisoquinoline (1S)-8t.
relatively high lipophilicity (clogD7.4 value >2.3 or more).
Introduction of an 8-methoxy group (8p) led to a deterioration in
CYP3A4 inhibition liability due to lipophilicity (clogD7.4 = 3.07).
Regarding CYP2D6 inhibition liability, no apparent improvement
was observed in evaluated compounds, except for compounds 8n
(R = 7-methoxy) and 8p (R = 8-methoxy). Introduction of a 7-meth-
oxy group (8n) only led to a greater than 3-fold attenuation of
CYP2D6 inhibition compared with 8a. Notably, introduction of an
8-methoxy group (8p) led to significant attenuation of CYP2D6
inhibition liability. The marked improvement in CYP2D6 inhibition
liability was the first observation of our study.5b
In the course of the discovery of (R)-2, 1-phenyl (S)-12, which
like 1-cyclohexyl (S)-1 had a bulky substituent at the C-1 position
of the tetrahydroisoquinoline, was synthesized and found to exhi-
bit much less lipophilicity than (S)-1 (clogD7.4 value (S)-12: 2.75 vs
(S)-1: 3.48) with an equivalent in vitro profile to (S)-1 (Table 4).5b
Given our results so far, transformation of the 1-cyclohexyl group
of 8p to the 1-phenyl group could improve CYP3A4 profile with
retention of favorable CYP2D6 inhibition liability due to its less
lipophilic nature and bulkiness. Therefore, we investigated the
effect on incorporation of substituents into the C-8 position of
1-phenyl tetrahydroisoquinoline (Table 5). As expected,
compound (1S)-8q exhibited an almost equivalent in vitro profile
Based on the X-ray crystal structure of human CYP2D6 bound
with
10-{2-[(2R)-1-methylpiperidin-2-yl]ethyl}-2-(methylsul-
fanyl)-10H-phenothiazine (RTZ),10 which is known as a non-heme
binder, the interaction of (1S)-8q with CYP2D6 was examined to
obtain a better understanding of why (1S)-8t led to a significant
decrease in CYP2D6 inhibition liability. Molecular docking was
performed between (1S)-8q and CYP2D6, on the assumption that
the binding site of (1S)-8q was around RTZ, as indicated by their

4642
T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
of Asp301, and hydrogen bonding (CH...O interaction) between the
carbonyl group of (1S)-8q and an alpha proton of Asp301. Second,
the overall (1S)-8q molecule showed potential hydrophobic inter-
Table 3
Inhibitory effect of tetrahydroisoquinoline substitutions on N-type calcium channels,
CYP 2D6 and 3A4
actions interpreted as donor-p, p–p and van der Waals interactions
5
6
with Phe120, Leu121, Leu213, Glu216, Leu248, Ala300, Asp301, Ser304 and
Phe483 of CYP2D6 (Fig. 1b).
R
N
N
H
7
8
Whereas the phenyl ring of the tetrahydroisoquinoline (1S)-8q
was tightly stacked by Phe120, Glu216 and Phe483, a relatively large
pocket was spread out flat within the same plane as the aromatic
ring, which could accommodate various substituents on tetrahy-
droisoquinoline. These observations imply that incorporation of
substituents has no or little effect on the impairment of interac-
tions with CYP2D6, which might explain why our attempt using
a knowledge-based approach did not work, except for introduction
of the 8-methoxy group of 8t.
Interestingly, the 1-phenyl group was packed into the flat
pocket formed by Ser304 and Leu213, where the pocket was rela-
tively narrow in the vertical direction of the 1-phenyl ring, indicat-
ing that the relative arrangement between tetrahydroisoquinoline
and the 1-phenyl group was critical for binding to CYP2D6.
Dihedral angle of (1S)-8q between tetrahydroisoquinoline and
the 1-phenyl group was 114.8° in this model.
In addition, the dihedral angle of 8-methoxytetrahydroisoqino-
line between tetrahydroisoquinoline and the 1-phenyl group in the
most stable conformation was significantly different from that of
tetrahydroisoquinoline, depending on the presence or absence of
the 8-methoxy moiety (145.8° vs 130.2°).11 Based on these results,
we speculate that introduction of an 8-methoxy group to (1S)-8q
led to a change in the relative arrangement between tetrahydroiso-
quinoline and the 1-phenyl group by steric repulsion with the
1-phenyl group, and that (1S)-8t was probably unable to fit into
the CYP2D6 enzyme and thus exhibited a significant decrease in
activity against CYP2D6.
O
OH
CYP3A4a
Preincubation
time
clogD7.4
b
R
N-type FLIPR
CYP2D6
IC50
(compound no.)
IC50
(l
M)
(l
M)
0 min
30 min
H(8a)
5-F (8f)
5-MeO (8g)
6-F (8h)
2.0
1.2
2.0
1.6
2.4
1.9
3.1
>10
2.3
1.7
0.89
1.3
1.6
2.6
2.5
1.4
1.1
92%
95%
107%
92%
N.T.
93%
N.T.
N.T.
109%
N.T.
85%
88%
97%
N.T.
N.T.
85%
95%
49%
63%
91%
N.T.
97%
N.T.
N.T.
80%
N.T.
65%
43%
83%
N.T.
N.T.
58%
2.53
2.34
3.08
2.71
2.96
2.72
2.05
1.31
2.39
3.12
3.63
3.29
2.99
2.30
2.05
3.07
<0.39
<0.78
1.6
N.T.
<0.78
N.T.
N.T.
1.3
6-Me (8i)
6-MeO (8j)
6-CN (8d)
6-CONH2 (8k)
7-F (8l)
7-Me (8m)
7-Et (8b)
7-iPr (8c)
7-MeO (8n)
7-CN (8e)
8-F (8o)
N.T.
1.3
1.9
3.7
N.T.
N.T.
30.2
8-MeO (8p)
a
Residual activities (%) of HLM were evaluated using midazolam as a probe
substrate. Details are described in the experimental section.
b
clogD values at pH 7.4 were calculated with ACD/PhysChem Batch (version
12.01).
3D-pharmacophore resemblance. The binding modes of (1S)-8q
and RTZ in the CYP2D6 crystal structure are illustrated in Fig. 1a.
In this model, the two aromatic parts of (1S)-8q overlap with the
phenothiazine ring of RTZ, and the basic part of both inhibitors
are shown to interact with Asp301, which is a key acidic residue
which interacts with basic CYP2D6 inhibitors. Given that (1S)-8q
interacts with CYP2D6 in a manner similar to RTZ, as the model
strongly suggested, the potency of (1S)-8q for CYP2D6 inhibition
Several knowledge-based strategies, such as modifying the
periphery of the molecule or reducing the overall lipophilicity of
the compound, have been proposed to reduce CYP2D6 inhibition
liability and many successful studies based on these strategies
have been reported.8,9 However, in cases where knowledge-based
approaches do not work or are not suitable to striking a balance
with bioactivity against the target molecule or ADMET, consider-
able effort is required, such as in our study due to the large and
flexible nature of the CYP2D6 binding pocket. Despite having been
wise after the event, our result indicated that molecular docking
simulation could guide many aspects of drug design. At present,
we still lack understanding of CYP2D6 inhibition. For example, only
one study has described the X-ray crystal structure of human
(IC50 = 2.2 lM) could be explained by the following two reasons.
First, three critical electrostatic interactions were observed in the
model, namely hydrogen bonding between the hydroxypropyl
group of (1S)-8q and a hydroxyl group in Ser304, salt bridge
between the amino group of (1S)-8q and a carboxylic acid residue
Table 4
In vitro pharmacological properties of (S)-1 and (S)-12
CYP3A4a
clogD7.4
b
Compound
N-type FLIPR IC50
(lM)
CYP2D6 IC50
(lM)
Preincubation time
0 min
30 min
OH
OH
N
N
H
1.0
1.0
0.60
87%
89%
67%
78%
3.48
2.75
O
(S)-1
N
N
H
<0.39
O
(S)-12
a
Residual activities (%) of HLM were evaluated using midazolam as a probe substrate. Details are described in the experimental section.
clogD values at pH 7.4 were calculated with ACD/PhysChem Batch (version 12.01).
b

T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
4643
Table 5
Inhibitory activities on N-type calcium channels, CYP 2D6 and 3A4
N
1
N
H
R
O
OH
CYP3A4a
clogD7.4
b
R (Compound No.)
N-type FLIPR IC50
(lM)
CYP2D6 IC50
(
lM)
Preincubation time
0 min
30 min
N
N
H
2.4
2.2
98%
94%
1.80
O
OH
(1S)-8q
R = F (8r)
R = Me (8s)
1.0
1.3
1.3
5.6
103%
100%
95%
89%
1.33
2.14
N
N
H
1.4
3.3
>50
94%
90%
91%
2.35
2.35
O
O
O
OH
(1S)-8t
N
N
H
17.9
110%
O
OH
(1R)-8t
a
Residual activities (%) of HLM were evaluated using midazolam as a probe substrate. Details are described in the experimental section.
clogD values at pH 7.4 were calculated with ACD/PhysChem Batch (version 12.01).
b
Figure 1a. Docking mode of (1S)-8q and crystal structure of RTZ in complex with
human CYP2D6. Only the hetero molecules in the binding site are shown for clarity
with the atoms colored as nitrogen—blue, oxygen—red, Fe—light blue, and carbon—
white or green. The wall of the binding site is represented as a blue dotted surface
and waters kept in the docking study are shown as red spheres. White carbon: RTZ
and Heme in the crystal structure of human CYP2D6 (PDB ID: 3TBG). Green carbon:
Docked conformation of (1S)-8q. (1S)-8q and RTZ occupied the same site, with
aromatic and basic features in common position.
Figure 1b. Observed interactions in the docking model of (1S)-8q with the binding
site residues of human CYP2D6. (1S)-8q is depicted as ball-and-stick model with the
same color scheme as Figure 1a. Three hydrogen bonds (light blue rod) and many
hydrophobic interactions (green: donor-p, orange: p–p, and yellow: van der Waals,
indicated by Scorpion) are observed between (1S)-8q and CYP2D6, which support
its potent inhibition of CYP2D6. Both Figure 1a and b were created using the
Molecular Operating Environment (MOE 2013).
CYP2D6 bound with a non-heme binder. Given the large and flex-
ible nature of CYP2D6, however, we speculate that several other
sites able to bind with other CYP2D6 inhibitors are present. Use
of structural information to attenuate CYP2D6 inhibition liability
in the drug discovery process requires more information of the
X-ray crystal structure of CYP2D6 bound with non-heme binders,
and considerable additional investigation is accordingly needed.8,9
Encouraged by the result, further characterizations of compound
(1S)-8t were conducted (Table 6). Regarding inhibitory activities
against the three other major CYP450 isoforms (CYP1A2, 2C9 and
2C19), compound (1S)-8t exhibited no inhibitory activities.
Moreover, compound (1S)-8t showed weak inhibitory activities
against both L-type calcium channels (IC50 >10
lM) and hERG
potassium channels (IC50 >100 M). The human liver microsomal
l
(HLM) stability of compound (1S)-8t (CL 45.9 mL/min/kg)
exhibited approximately 5-fold better metabolic stability than
compound (S)-1 (CL 223.4 mL/min/kg). Together, these results
prompted us to investigate its in vivo efficacy.

4644
T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
Table 6
Inhibitory activities of (1S)-8t against CYP isoforms, hERG and L-type calcium channel
Compound
CYP1A2 IC50
(lM)
CYP2C9 IC50
(l
M)
CYP2C19 IC50
(lM)
hERG IC50
>100
(
lM)
L-type FLIPR IC50 (lM)
N
N
H
>50
>50
>50
>10
O
O
OH
(1S)-8t
1.2
1.0
0.8
0.6
0.4
0.2
0.0
standard. Abbreviations used for the signal patterns are as follows:
s, singlet; d, doublet; t, triplet; q, quartet; sep., septet; m, multi-
plet; and br, broad. Mass spectra were recorded on a JEOL LX-
2000 or Waters ZQ-2000 mass spectrometer. Elemental analysis
was conducted using a Yanaco MT-5 microanalyzer. HPLC analysis
was performed using a Daicel OD-H or AD-RH chiral column on a
Hitachi HPLC system (L-7000 series) equipped with a UV source
***
###
(210 or 230 nm). Specific rotations were measured using
HORIBA SEPA-300 polarimeter.
a
Normal Vehicle
0.3
1
3
10
(1S)-8t (mg/kg, po)
5.1.2. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-[(2-
hydroxy-2-methylpropyl)amino]ethanone fumarate (5)
Figure 2. Effect of compound (1S)-8t on mechanical allodynia in SNL model rats.
Each bar represents the mean SEM withdrawal threshold of the hindpaw (n = 8).
###P <0.001, statistically significant compared with normal group (Student’s t-test).
***P <0.001, statistically significant compared with the vehicle-treated group
(Dunnett’s test). Closed columns, operated side paw in the drug-treated and
vehicle-treated group. Open column, non-operated side paw in the vehicle-treated
group.
To a solution of compound 3a (200 mg) in acetonitrile (3 mL)
was added potassium carbonate (474 mg) and 1-amino-2-
methyl-2-propanol (244 mg). The reaction mixture was stirred
for 3 hours at 60 °C. The mixture was diluted with chloroform
and water, filtered through a phase separation filter and the filtrate
was concentrated in vacuo. The residue was purified by silica gel
column chromatography, and eluted with chloroform-methanol
to give the desired product as a colorless gum (198 mg). The gum
was dissolved in 2-propanol (2 mL), and fumaric acid (67 mg)
was added to the solution. The mixture was stirred at room tem-
perature for 3 hours. The precipitate was filtered off to obtain the
title compound 5 (185 mg, 59%) as a colorless solid. 1H NMR
(DMSO-d6): This compound exists as a pair of rotamers at room
temperature. d 7.09–7.24 (4H, m), 7.11 (fumaric acid 2H, s), 7.13
(major isomer 1H, d, J = 9.2 Hz), 4.43 (minor isomer 1H, d,
J = 9.4 Hz), 4.16–4.27 (minor isomer 1H, m), 3.59–3.69 (2H, m),
3.48–3.57 (1H, m), 3.22–3.32 (minor isomer 1H, m), 2.78–2.98
(2H, m), 1.46–1.78 (6H, m), 0.96–1.21 (11H, m); FAB MS m/z:
345 [M+H]+.
Orally administrated (1S)-8t exhibited antinociceptive activity
(77% inhibition) in the second phase of the formalin test in mice
at 100 mg/kg. Moreover, the analgesic effect of (1S)-8t was evalu-
ated using the L5/L6 SNL model of neuropathic pain in rats.12 (1S)-
8t dose-dependently reduced mechanical allodynia with oral
administration from 0.3 to 10 mg/kg. Oral administration of (1S)-
8t improved mechanical allodynia in the rat SNL model with an
ED50 value of 2.8 mg/kg (Fig. 2).
4. Conclusion
In lead optimization efforts from (S)-1, we identified (1S)-8t as a
novel orally active small-molecule N-type calcium channel blocker
without CYP inhibition liability. Based on the previous results,
improvement in CYP3A4 inhibition profile was achieved by reduc-
ing the lipophilicity of (S)-1. Incorporation of a methoxy group into
the C-8 position of tetrahydroisoquinoline ((1S)-8t) resulted in the
5.1.3. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2R)-2-hydroxybutyl]amino}ethanone oxalate (6)
Compound 6 was synthesized using a similar protocol to that
for compound 5. Compound 6 was obtained in 50% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair
of rotamers at room temperature. d 7.13–7.26 (4H, m), 5.11 (1H,
d, J = 9.4 Hz), 4.27–4.34 (minor isomer 1H, m), 3.51–4.26 (5H, m),
3.30–3.41 (minor isomer 1H, m), 2.72–3.03 (4H, m), 1.29–1.78
(7H, m), 0.96–1.22 (4H, m), 0.87 (3H, t, J = 7.4 Hz); FAB MS m/z:
345 [M+H]+.
loss of potency for CYP2D6 inhibition (IC50 >50 lM). A docking
study suggested that a change in the relative arrangement between
tetrahydroisoquinoline and the 1-phenyl group by installation of
8-MeO group disfavored binding to CYP2D6 enzyme. Oral adminis-
tration of (1S)-8t exhibited efficacy in the rat SNL model of neuro-
pathic pain, with an ED50 value of 2.8 mg/kg.
5. Experimental section
5.1. Chemistry
5.1.4. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2S)-2-hydroxybutyl]amino}ethanone oxalate (7)
Compound 7 was synthesized using a similar protocol to that
for compound 5. Compound 7 was obtained in 44% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair
of rotamers at room temperature. d 7.08–7.28 (4H, m), 7.11 (1H,
d, J = 9.3 Hz), 4.28–4.35 (minor isomer 1H, m), 4.09–4.25 (1H, m),
3.96–4.07 (1H, m), 3.52–3.76 (3H, m), 3.28–3.39 (1H, m), 2.70–
3.05 (4H, m), 1.28–1.79 (7H, m), 0.95–1.22 (4H, m), 0.86 (3H, t,
J = 7.4 Hz); FAB MS m/z: 345 [M+H]+.
5.1.1. General
All reactions were carried out using commercially available
reagents and solvents without further purification. Column chro-
matography was performed using a Shoko Scientific SI series silica
gel cartridge on Shoko Scientific Purif-a
2. 1H NMR spectra were
recorded on a JNM-EX400 spectrometer. Chemical shifts are
expressed in d units (ppm) using tetramethylsilane as an internal

T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
4645
5.1.5. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8a)
isomer 1H, br s), 7.38–7.48 (1H, m), 5.20 (major isomer 1H, d,
J = 9.4 Hz), 4.40–4.48 (minor isomer 1H, m), 4.20–4.29 (1H, m),
4.15 (major isomer 1H, d, J = 16.1 Hz), 3.90–4.09 (2H, m), 3.56–
3.73 (major isomer 2H, m), 3.30–3.39 (minor isomer 1H, m), 3.07
(2H, t, J = 6.1 Hz), 2.85–3.00 (1H, m), 2.68–2.82 (1H, m), 1.37–
1.79 (6H, m), 0.92–1.25 (8H, m); FAB MS m/z: 356 [M+H]+.
Compound 8a was synthesized using a similar protocol to that
for compound 5. Compound 8a was obtained in 68% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.10–7.26 (4H, m), 5.01 (1H, d,
J = 9.4 Hz), 4.28–4.34 (minor isomer 1H, m), 3.23–4.24 (5H, m),
2.69–3.04 (4H, m), 1.45–1.79 (6H, m), 0.93–1.27 (8H, m); FAB MS
m/z: 331 [M+H]+. Anal. Calcd for C20H30N2O2ꢀC2H2O4.0.8H2O: C,
60.76; H, 7.79; N, 6.44. Found: C, 60.73; H, 7.56; N, 6.41.
5.1.11. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2S)-2-hydroxypropyl]amino}ethanone oxalate (9)
Compound 9 was synthesized using a similar protocol to that
for compound 5. Compound 9 was obtained in 66% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair
of rotamers at room temperature. d 7.09–7.27 (4H, m), 5.11 (major
isomer 1H, d, J = 9.4 Hz) 3.25–4.36 (5H, m), 2.67–3.08 (4H, m),
1.45–1.78 (6H, m), 0.91–1.26 (8H, m); FAB MS m/z: 331 [M+H]+.
5.1.6. 1-[(1S)-1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl]-2-
{[(2R)-2-hydroxypropyl]amino}ethanone fumarate ((1S)-8a)
Compound (1S)-8a was synthesized using a similar protocol to
that for compound 8a, except that (1S)-cyclohexyl-1,2,3,4-tetrahy-
droisoquinoline13 was used as a starting material. Compound
(1S)-8a was obtained in 37% yield as a colorless solid. 1H NMR
(DMSO-d6): d 7.07–7.25 (4H, m), 6.53 (fumaric acid 2H, s), 5.12
(1H, d, J = 9.3 Hz), 2.29–3.93 (9H, m), 1.42–1.78 (6H, m), 0.89–
1.29 (8H, m); FAB MS m/z: 331 [M+H]+. Anal. Calcd for
5.1.12. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2S)-1-hydroxy-3,3-dimethylbutan-2-yl]amino}ethanone
oxalate (10)
Compound 10 was synthesized using a similar protocol to that
for compound 5. Compound 10 was obtained in 68% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.09–7.26 (4H, m), 5.12 (major
isomer 1H, d, J = 9.0 Hz), 3.53–4.45 (7H, m), 2.85–3.07 (2H, m),
1.45–1.79 (6H, m), 0.90–1.24 (14H, m); FAB MS m/z: 373 [M+H]+.
C
20H30N2O2ꢀC4H4O4.0.1H2O: C, 64.29; H, 7.69; N, 6.25. Found: C,
20
64.24; H, 7.69; N, 6.21. [
a]
ꢁ30.7° (c = 0.1, MeOH).
D
5.1.7. 1-(1-Cyclohexyl-7-ethyl-3,4-dihydroisoquinolin-2(1H)-
yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8b)
Compound 8b was synthesized using a similar protocol to that
for compound 5. Compound 8b was obtained in 67% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.17 (1H, d, J = 7.8 Hz), 7.03–7.09
(1H, m), 6.97 (1H, m), 5.07 (major isomer 1H, d, J = 9.4 Hz), 4.25–
4.31 (minor isomer 1H, m), 4.09–4.22 (1H, m), 3.91–4.05 (2H, m),
3.61–3.70 (major isomer 1H, m), 3.50–3.60 (major isomer 1H, m),
3.28–3.37 (minor isomer 1H, m), 2.70–2.99 (4H, m), 2.51–2.61
(2H, m), 1.48–1.78 (6H, m), 0.95–1.21 (11H, m); ESI MS m/z: 359
[M+H]+.
5.1.13. 1-(1-Cyclohexyl-3,4-dihydroisoquinolin-2(1H)-yl)-2-
{[(2S)-1-hydroxy-4-methylpentan-2-yl]amino}ethanone
oxalate (11)
Compound 11 was synthesized using a similar protocol to that
for compound 5. Compound 11 was obtained in 68% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.10–7.26 (4H, m), 5.12 (major
isomer 1H, dd, J = 1.6, 9.4 Hz), 4.41 (minor isomer 1H, d,
J = 9.3 Hz), 2.79–4.31 (9H, m), 1.31–1.79 (9H, m), 0.95–1.26 (5H,
m), 0.76–0.93 (6H, m); FAB MS m/z: 373 [M+H]+.
5.1.8. 1-(1-Cyclohexyl-7-isopropyl-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8c)
Compound 8c was synthesized using a similar protocol to that
for compound 5. Compound 8c was obtained in 69% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.06–7.15 (2H, m), 6.96–7.02 (1H,
m), 5.07 (1H, d, J = 9.5 Hz), 3.89–4.33 (4H, m), 3.48–3.70 (2H, m),
3.26–3.38 (minor isomer 1H, m), 2.69–2.99 (6H, m), 1.47–1.81
(8H, m), 0.95–1.23 (18H, m); FAB MS m/z: 373 [M+H]+.
5.1.14. 1-(1-Cyclohexyl-5-fluoro-3,4-dihydroisoquinolin-2(1H)-
yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8f)
To a solution of compound 4f (1.30 g) in ethyl acetate (20 mL)
and saturated aqueous sodium bicarbonate (25 mL) was added a
solution of chloroacetyl chloride (0.75 mL) in ethyl acetate (5 mL)
at 4 °C, and the reaction mixture was stirred at room temperature
for 1 hour. The mixture was diluted with water, extracted with
ethyl acetate, dried over magnesium sulfate and concentrated in
vacuo to give 2-chloro-1-(1-cyclohexyl-5-fluoro-3,4-dihydroiso-
quinolin-2(1H)-yl)ethanone (971 mg 65%) as a colorless solid.
FAB MS m/z 310 [M+H]+. To a solution of compound 2-chloro-1-
(1-cyclohexyl-5-fluoro-3,4-dihydroisoquinolin-2(1H)-yl)ethanone
(290 mg) in acetonitrile (10 mL) was added potassium carbonate
(736 mg), (2R)-1-amino-propanol (300 mg) and N,N,N-tributyl-1-
butanaminium iodide (36 mg). The reaction mixture was stirred
for 7 hours at 70 °C. The mixture was diluted with ethyl acetate
and water, extracted with ethyl acetate, washed with brine, dried
over magnesium sulfate and then concentrated in vacuo. The resi-
due was purified by silica gel column chromatography, eluting
with chloroform–methanol to give the desired product as a color-
less oil (270 mg). The oil was dissolved in a solution of 2-propanol
and ether, and oxalic acid (70 mg) was added to the solution. The
precipitate was filtered off to obtain the title compound 8f
(290 mg, 68%) as a colorless solid. 1H NMR (DMSO-d6): This com-
pound exists as a pair of rotamers at room temperature. d 7.17–
7.27 (1H, m), 6.98–7.14 (2H, m), 5.18 (major isomer 1H, d,
J = 9.3 Hz), 4.36–4.46 (minor isomer 1H, m), 4.18 (major isomer
1H, dd, J = 2.5, 13.6 Hz), 3.91–4.09 (2H, m), 3.67–3.77 (major iso-
mer 1H, m), 3.54–3.64 (major isomer 1H, m), 3.22–3.32 (minor
5.1.9. 1-Cyclohexyl-2-{N-[(2R)-2-hydroxypropyl]glycyl}-1,2,3,4-
tetrahydroisoquinoline-6-carbonitrile oxalate (8d)
Compound 8d was synthesized using a similar protocol to that
for compound 5. Compound 8d was obtained in 56% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.74 (1H, br s), 7.61–7.71 (1H, m),
7.36–7.44 (1H, m), 5.22 (major isomer 1H, d, J = 9.2 Hz), 4.44–4.53
(minor isomer 1H, m), 4.20–4.30 (minor isomer 1H, m), 4.16
(major isomer 1H, d, J = 16.2 Hz), 3.90–4.09 (2H, m), 3.56–3.73
(major isomer 2H, m), 3.29–3.38 (minor isomer 1H, m), 2.86–
3.07 (3H, m), 2.69–2.82 (1H, m), 1.38–1.78 (6H, m), 0.89–1.25
(8H, m); FAB MS m/z: 356 [M+H]+.
5.1.10. 1-Cyclohexyl-2-{N-[(2R)-2-hydroxypropyl]glycyl}-
1,2,3,4-tetrahydroisoquinoline-7-carbonitrile oxalate (8e)
Compound 8e was synthesized using a similar protocol to that
for compound 5. Compound 8e was obtained in 60% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.66–7.75 (2H, m), 7.63 (minor

4646
T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
isomer 1H, m), 2.65–3.05 (4H, m), 1.48–1.79 (6H, m), 0.92–1.23
(1H, m), 2.84–3.07 (3H, m), 2.64–2.81 (1H, m), 1.44–1.79 (6H,
(8H, m); FAB MS m/z: 349 [M+H]+.
m), 0.94–1.21 (8H, m); FAB MS m/z: 361 [M+H]+.
5.1.20. 1-(1-Cyclohexyl-7-fluoro-3,4-dihydroisoquinolin-2(1H)-
yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8l)
Compound 8l was synthesized using a similar protocol to that
for compound 8f. Compound 8l was obtained in 97% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.18–7.31 (1H, m), 6.99–7.13 (2H,
m), 5.15 (major isomer 1H, d, J = 9.2 Hz), 4.33–4.41 (minor isomer
1H, m), 4.09–4.24 (1H, m), 3.90–4.08 (2H, m), 3.53–3.71 (major
isomer 2H, m), 3.28–3.38 (minor isomer 1H, m), 2.69–3.00 (4H,
m), 1.43–1.78 (6H, m), 0.94–1.25 (8H, m); FAB MS m/z: 349
[M+H]+.
5.1.15. 1-(1-Cyclohexyl-5-methoxy-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8g)
Compound 8g was synthesized using a similar protocol to that
for compound 8f. Compound 8g was obtained in 20% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.33–7.40 (1H, m), 7.14 (1H, t,
J = 7.9 Hz), 6.84–6.91 (1H, m), 6.76 (1H, d, J = 7.7 Hz), 5.09 (major
isomer 1H, d, J = 9.3 Hz), 4.32–4.41 (minor isomer 1H, m), 4.26–
4.31 (minor isomer 1H, m), 4.17 (1H, d, J = 16.1 Hz), 3.89–4.08
(3H, m), 3.51–3.71 (2H, m), 3.18–3.28 (minor isomer 1H, m), 3.78
(3H, s), 2.85–2.98 (1H, m), 2.65–2.81 (2H, m), 1.50–1.79 (6H, m),
0.91–1.22 (8H, m); FAB MS m/z: 361 [M+H]+.
5.1.21. 1-(1-Cyclohexyl-7-methyl-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8m)
5.1.16. 1-(1-Cyclohexyl-6-fluoro-3,4-dihydroisoquinolin-2(1H)-
yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8h)
Compound 8h was synthesized using a similar protocol to that
for compound 8f. Compound 8h was obtained in 68% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.16–7.25 (1H, m), 6.96–7.13 (2H,
m), 5.12 (major isomer 1H, d, J = 9.4 Hz), 4.31–4.40 (minor isomer
1H, m), 4.10–4.26 (1H, m), 3.90–4.07 (2H, m), 3.51–3.69 (major
isomer 2H, m), 3.26–3.37 (minor isomer 1H, m), 2.69–3.06 (4H,
m), 1.42–1.79 (6H, m), 0.90–1.23 (8H, m); FAB MS m/z: 349
[M+H]+.
Compound 8m was synthesized using a similar protocol to that
for compound 8f. Compound 8m was obtained in 91% yield as a
colorless solid. 1H NMR (DMSO-d6): This compound exists as a pair
of rotamers at room temperature. d 7.09 (1H, d, J = 7.8 Hz), 7.04
(1H, d, J = 7.7 Hz), 6.96 (1H, s), 5.05 (major isomer 1H, d,
J = 9.3 Hz), 4.23–4.29 (minor isomer 1H, m), 4.10–4.21 (1H, m),
3.91–4.06 (2H, m), 3.60–3.69 (major isomer 1H, m), 3.49–3.59
(major isomer 1H, m), 3.28–3.37 (minor isomer 1H, m), 2.70–
2.99 (4H, m), 2.28 (minor isomer 3H, s), 2.27 (major isomer 3H,
s), 1.47–1.77 (6H, m), 0.94–1.22 (8H, m); FAB MS m/z: 345 [M+H]+.
5.1.22. 1-(1-Cyclohexyl-7-methoxy-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8n)
5.1.17. 1-(1-Cyclohexyl-6-methyl-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8i)
Compound 8n was synthesized using a similar protocol to that
for compound 8f. Compound 8n was obtained in 64% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.06–7.15 (1H, m), 6.77–6.85 (1H,
m), 6.71–6.76 (1H, m), 5.09 (major isomer 1H, d, J = 9.2 Hz), 4.27–
4.35 (minor isomer 1H, m), 3.87–4.19 (2H, m), 3.73 (minor isomer
3H, s), 3.72 (major isomer 3H, s), 3.49–3.68 (2H, m), 3.28–3.37
(minor isomer 1H, m), 2.68–2.95 (4H, m), 1.48–1.77 (6H, m),
0.96–1.20 (8H, m); FAB MS m/z: 361 [M+H]+.
Compound 8i was synthesized using a similar protocol to that
for compound 8f. Compound 8i was obtained in 94% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 6.94–7.06 (3H, m), 5.06 (major
isomer 1H, d, J = 9.3 Hz), 4.09–4.31 (1H, m), 3.90–4.06 (2H, m),
3.49–3.70 (major isomer 2H, m), 3.26–3.40 (minor isomer 1H,
m), 2.68–3.00 (4H, m), 2.26 (3H, s), 1.47–1.81 (6H, m), 0.92–1.23
(8H, m); FAB MS m/z: 345 [M+H]+.
5.1.18. 1-(1-Cyclohexyl-6-methoxy-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8j)
5.1.23. 1-(1-Cyclohexyl-8-fluoro-3,4-dihydroisoquinolin-2(1H)-
yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8o)
Compound 8o was synthesized using a similar protocol to that
for compound 8f. Compound 8o was obtained in 63% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.23–7.32 (1H, m), 7.02–7.15 (2H,
m), 5.47 (major isomer 1H, d, J = 10.0 Hz), 4.51 (minor isomer 1H,
d, J = 8.7 Hz), 3.91–4.22 (3H, m), 3.69–3.79 (major isomer 1H, m),
3.49–3.60 (major isomer 1H, m), 3.36–3.45 (minor isomer 1H, s),
2.86–3.14 (3H, m), 2.69–2.83 (1H, m), 1.52–1.83 (5H, m), 1.30–
1.43 (1H, m), 0.97–1.24 (8H, m); ESI MS m/z: 349 [M+H]+.
Compound 8j was synthesized using a similar protocol to that
for compound 8f. Compound 8j was obtained in 68% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.01–7.09 (1H, m), 6.77–6.81 (1H,
m), 6.70–6.76 (1H, m), 5.05 (major isomer 1H, d, J = 9.3 Hz), 4.27–
4.33 (minor isomer 1H, m), 4.15–4.24 (minor isomer 1H, m), 3.94
(minor isomer 1H, d, J = 7.0 Hz), 3.73–3.92 (2H, m), 3.72 (3H, s),
3.52–3.66 (2H, m), 3.22–3.30 (minor isomer 1H, m), 2.57–2.98
(4H, m), 1.47–1.75 (6H, m), 0.95–1.17 (8H, m); FAB MS m/z: 361
[M+H]+.
5.1.24. 1-(1-Cyclohexyl-8-methoxy-3,4-dihydroisoquinolin-
2(1H)-yl)-2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate
(8p)
5.1.19. 1-Cyclohexyl-2-{N-[(2R)-2-hydroxypropyl]glycyl}-
1,2,3,4-tetrahydroisoquinoline-6-carboxamide oxalate (8k)
Compound 8k was synthesized using a similar protocol to that
for compound 8f. Compound 8k was obtained in 32% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.92 (1H, br s), 7.73 (1H, br s),
7.64–7.71 (1H, m), 7.33 (1H, br s), 7.20–7.27 (1H, m), 5.17 (major
isomer 1H, d, J = 9.2 Hz), 4.35–4.44 (minor isomer 1H, m), 4.20–
4.31 (minor isomer 1H, m), 4.15 (major isomer 1H, d, J = 16.1 Hz),
3.89–4.07 (2H, m), 3.55–3.73 (major isomer 2H, m), 3.30–3.42
Compound 8p was synthesized using a similar protocol to that
for compound 8f. Compound 8p was obtained in 59% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.15–7.24 (1H, m), 6.78–6.91 (2H,
m), 5.61 (major isomer 1H, d, J = 10.0 Hz), 5.60 (minor isomer 1H,
d, J = 9.8 Hz), 3.83–4.18 (3H, m), 3.80 (minor isomer 3H, s), 3.77
(major isomer 3H, s), 3.68–3.76 (1H, m), 3.37–3.48 (1H, m), 2.70–
3.12 (4H, m), 1.51–1.77 (5H, m), 1.25–1.35 (1H, m), 0.96–1.21
(8H, m); FAB MS m/z: 361 [M+H]+.

T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
4647
5.1.25. 2-{[(2R)-2-Hydroxypropyl]amino}-1-[(1S)-1-phenyl-3,4-
dihydroisoquinolin-2(1H)-yl]ethanone oxalate ((1S)-8q)
isomer 3H, s), 3.70 (major isomer 3H, s), 3.58–3.66 (1H, m),
3.29–3.49 (1H, m), 2.90–3.02 (1H, m), 2.61–2.78 (3H, m), 1.01–
1.10 (3H, m); FAB MS m/z: 355 [M+H]+. Anal. Calcd for
Compound (1S)-8q was synthesized using a similar protocol to
that for compound 8f. Compound (1S)-8q was obtained in 45%
yield as a colorless solid. 1H NMR (DMSO-d6): This compound
exists as a pair of rotamers at room temperature. d 7.15–7.40
(9H, m), 6.66 (major isomer 1H, s), 6.17 (minor isomer 1H, br s),
4.38 (minor isomer 1H, d, J = 16.1 Hz), 4.22 (major isomer 1H, d,
J = 16.2 Hz), 3.89–4.16 (2H, m), 3.60–3.70 (1H, m), 3.47–3.58 (1H,
m), 3.35–3.45 (1H, m), 2.93–3.06 (2H, m), 2.72–2.91 (2H, m),
1.04–1.14 (3H, m); FAB MS m/z: 325 [M+H]+. Anal. Calcd for
C
21H26N2O3ꢀC4H4O4: C, 63.82; H, 6.43; N, 5.95. Found: C, 63.64;
26
H, 6.42; N, 5.88. [
a]
–153.2° (c = 0.1, MeOH).
D
5.2. VCD analysis of (1S)-5-bromo-8-methoxy-1-phenyl-1,2,3,4-
tetrahydroisoquinoline (1S)-22
An absolute configuration of compound (1S)-22 was assigned using
vibrational dichroism (VCD) spectra.7 See Supporting information.
C
20H24N2O2ꢀC2H2O4: C, 63.76; H, 6.32; N, 6.76. Found: C, 63.69;
26
H, 6.33; N, 6.71. [
a
]
+115.2° (c = 0.1, MeOH).
D
5.3. Docking study of potent CYP2D6 inhibitor (1S)-8q
5.1.26. 1-(8-Fluoro-1-phenyl-3,4-dihydroisoquinolin-2(1H)-yl)-
2-{[(2R)-2-hydroxypropyl]amino}ethanone oxalate (8r)
Compound 8r was synthesized using a similar protocol to that
for compound 8f. Compound 8r was obtained in 48% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.26–7.43 (4H, m), 7.07–7.23 (4H,
m), 6.87 (major isomer 1H, br s), 6.17 (minor isomer 1H, d,
J = 6.8 Hz), 4.42–4.51 (minor isomer 1H, m), 4.07–4.31 (2H, m),
3.93–4.04 (1H, m), 3.63–3.73 (1H, m), 3.30–3.42 (1H, m), 2.74–
3.15 (4H, m), 1.10 (major isomer 3H, d, J = 6.2 Hz), 1.04 (minor iso-
mer 3H, d, J = 6.2 Hz).
For docking simulation, the crystal structure of human
cytochrome CYP2D6 in complex with 10-{2-[(2R)-1-methylpiperi-
din-2-yl]ethyl}-2-(methylsulfanyl)-10H-phenothiazine (RTZ, repre-
sented by residue name in the Protein Data Bank record 3TBG)
was processed using the Protein Preparation Wizard of Maestro
9.7 (Schrodinger, Inc., New York, 2014) with default parameters.
As the crystal structure contains four CYP2D6 molecules, we first
removed discarded molecules for simulation and only one set of
receptor, ligands and waters belonging to chain A were kept. The
hydrogens were then added to them and energetically optimized
using Prime. All docking processes were performed using Glide
on Maestro 9.7. Given that the molecular size and pharmacophore
positioning of RTZ are similar to that of (1S)-8q, grids defining the
binding site were generated around one of two RTZ molecules
observed in the crystal structure (RTZ1 in 3TBG).10 All of the water
molecules, including three blocking the route to HEME (HOH 614,
615 and 628 in 3TBG), were kept as part of the receptor during grid
generation. This was intended to prevent (1S)-8q from positioning
nearby heme. Docking calculation was performed with the default
parameter of Glide SP protocol, without any intentional restraint,
and the best glidescore mode was selected as the CYP2D6 binding
model of (1S)-8q. To understand why (1S)-8q has potency to
CYP2D6, putative interactions between them were visualized using
MOE 2013 (Chemical Computing Group Inc., Quebec, 2014) and
Scorpion (Desert Scientific Software, Australia, 2014).
5.1.27. 2-{[(2R)-2-Hydroxypropyl]amino}-1-(8-methyl-1-
phenyl-3,4-dihydroisoquinolin-2(1H)-yl)ethanone oxalate (8s)
Compound 8s was synthesized using a similar protocol to that
for compound 8f. Compound 8s was obtained in 44% yield as a col-
orless solid. 1H NMR (DMSO-d6): This compound exists as a pair of
rotamers at room temperature. d 7.19–4.40 (4H, m), 7.01–7.17 (4H,
m), 6.80 (major isomer 1H, br s), 5.96 (minor isomer 1H, br s), 4.41
(minor isomer 1H, br s), 3.92–4.22 (3H, m), 3.56–3.68 (1H, m),
3.28–3.39 (1H, m), 2.90–3.09 (2H, m), 2.70–2.89 (2H, m), 2.10
(minor isomer 3H, d, J = 5.1 Hz), 2.03 (major isomer 3H, d,
J = 3.4 Hz), 1.11 (3H, d, J = 6.2 Hz); ESI MS m/z: 339 [M+H]+.
5.1.28. 2-{[(2R)-2-Hydroxypropyl]amino}-1-[(1S)-8-methoxy-1-
phenyl-3,4-dihydroisoquinolin-2(1H)-yl]ethanone oxalate
((1S)-8t)
Compound (1S)-8t was synthesized using a similar protocol to
that for compound 8f. Compound (1S)-8t was obtained in 85% yield
as a colorless solid. 1H NMR (DMSO-d6): This compound exists as a
pair of rotamers at room temperature. d 7.21–7.36 (4H, m), 7.14
(minor isomer 2H, d, J = 7.2 Hz), 7.08 (major isomer 2H, d,
J = 7.1 Hz), 6.91–6.98 (1H, m), 6.84–6.90 (2H, m), 6.06 (minor iso-
mer 1H, br s), 4.03–4.34 (2H, m), 3.93–4.03 (1H, m), 3.72 (minor
isomer 3H, s), 3.71 (major isomer 3H, s), 3.53–3.63 (1H, m),
3.32–3.43 (1H, m), 2.91–3.05 (2H, m), 2.69–2.84 (2H, m), 1.07–
1.13 (3H, m), 1.04 (1H, d, J = 6.2 Hz); FAB MS m/z: 355 [M+H]+.
5.4. Pharmacology
5.4.1. Inhibitory activity against N-type calcium channels was
evaluated using an in vitro FLIPR assay in IMR-32 human
neuroblastoma cells
IC50 values were determined in duplicate in one experiment.
Details of experimental procedures were as previously described.5a
For more advanced compounds, inhibitory activities were
determined in several times of assays to obtain stable IC50 values
with high confidence. These data will be published elsewhere.
Anal. Calcd for C21H26N2O3ꢀC2H2O4: C, 62.15; H, 6.35; N, 6.30.
26
Found: C, 61.92; H, 6.39; N, 6.22. [
a]
+97.8° (c = 0.1, MeOH).
D
5.4.2. Inhibitory activity against L-type calcium channels was
evaluated using an in vitro FLIPR assay in IMR-32 human
neuroblastoma cells
5.1.29. 2-{[(2R)-2-Hydroxypropyl]amino}-1-[(1R)-8-methoxy-1-
phenyl-3,4-dihydroisoquinolin-2(1H)-yl]ethanone fumarate
((1R)-8t)
Compound (1R)-8t was synthesized using a similar protocol to
that for compound 8f. Compound (1R)-8t was obtained in 41%
yield as a colorless solid. 1H NMR (DMSO-d6): This compound
exists as a pair of rotamers at room temperature. d 7.19–7.34
(4H, m), 7.13 (minor isomer 2H, d, J = 7.2 Hz), 7.06 (major isomer
2H, d, J = 7.3 Hz), 6.82–6.97 (3H, m), 6.52 (fumaric acid 2H, s),
6.17 (minor isomer 1H, br s), 3.75–3.97 (3H, m), 3.73 (minor
The inhibitory activity of (1S)-8t against L-type calcium chan-
nels was evaluated as previously described for the evaluation of
inhibitory activity against N-type calcium channels5a using
conotoxin MVIIA instead of nitrendipine.
x-
5.4.3. Evaluation of hERG inhibition using Rb Efflux assay
The inhibitory activity of (1S)-8t for hERG was evaluated using
Rb efflux assay. Details of experimental procedures were as previ-
ously described.5a

4648
T. Ogiyama et al. / Bioorg. Med. Chem. 23 (2015) 4638–4648
5.4.4. Measurement of CYP inhibition
Acknowledgements
The inhibitory activities of test compounds for CYP1A2, 2C9,
2C19 and 2D6 were determined using a fluorescence-based assay.
Details of experimental procedures have been described previ-
ously.5a For CYP3A4 inhibition assay, midazolam was used as a
probe substrate to monitor changes in CYP3A4 activity during
exposure to each test compound. Details of experimental
procedures have also been described previously.5a Residual
metabolic activities for reversible (Eq. 1) and time-dependent
(Eq. 2) inhibition were calculated using the following equations,
respectively:
We wish to acknowledge Dr. Hiroshi Izumi (The National
Institute of Advanced Industrial Science and Technology) for his
help in the VCD analysis of (1S)-22. We thank Dr. Tatsuya Niimi
for his valuable comments regarding the structural basis of drug
binding to CYP2D6 enzyme. We also thank Drs. Kenichi Onda and
Susumu Watanuki for their helpful advice and careful reading of
the manuscript. In addition, we thank the staff of the Division of
Analytical Science Laboratories for conducting the elemental anal-
yses and spectral measurements.
ꢀ
% Residual Activity ¼ Activitycompound;0 min Activityvehicle;0 min ꢂ 100
Supplementary data
ð1Þ
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2015.05.053.
À
ꢀ
Áꢀ
% Residual Activity ¼ Activitycompound;30 min Activityvehicle;30 min
À
ꢀ
Á
Activitycompound;0 min Activityvehicle;0 min ꢂ 100
ð2Þ
References and notes
1. (a) McGivern, J. G.; McDonough, S. I. Curr. Drug Targets: CNS Neurol. Disord.
2004, 3, 457; (b) McGivern, J. G. Drug Discovery Today 2006, 11, 245.
2. (a) Miljanich, G. P. Curr. Med. Chem. 2004, 11, 3029; (b) Schmidtkco, A.; Lötsch,
J.; Freynhagen, R.; Geisslinger, G. Lancet 2010, 375, 1569.
3. (a) Yamamoto, T.; Takahara, A. Curr. Top. Med. Chem. 2009, 9, 377; (b) Barrow, J.
C.; Duffy, J. L. Annu. Rep. Med. Chem. 2010, 45, 2; (c) Lee, M. S. Prog. Med. Chem.
2014, 53, 147.
where Activitycompound, 0 min denotes activity obtained in the pres-
ence of compound and without pre-incubation, Activityvehicle, 0 min
denotes activity obtained in the absence of compound and without
pre-incubation, Activitycompound, 30 min denotes activity obtained in
the presence of compound and with pre-incubation, and
4. (a) Beebe, X.; Yeung, C. M.; Darczak, D.; Shekhar, S.; Vortherms, T. A.; Miller, L.;
Milicic, I.; Swensen, A. M.; Zhu, C. Z.; Banfor, P.; Wetter, J. M.; Marsh, K. C.;
Jarvis, M. F.; Scott, V. E.; Schrimpf, M. R.; Lee, C.-H. Bioorg. Med. Chem. Lett. 2013,
23, 4857; (b) Shao, P. P. ACS Med. Chem. Lett. 2013, 4, 1064; (c) Winters, M. P.;
Subashinghe, N.; Wall, M.; Beck, E.; Brandt, M. R.; Finley, M. F. A.; Liu, Y.; Lubin,
M. L.; Neeper, M. P.; Qin, N.; Flores, C. M.; Sui, Z. Bioorg. Med. Chem. Lett. 2014,
24, 2053; (d) Winters, M. P.; Subashinghe, N.; Wall, M.; Beck, E.; Brandt, M. R.;
Finley, M. F. A.; Liu, Y.; Lubin, M. L.; Neeper, M. P.; Qin, N.; Flores, C. M.; Sui, Z.
Bioorg. Med. Chem. Lett. 2014, 24, 2057.
5. (a) Ogiyama, T.; Inoue, M.; Honda, S.; Yamada, H.; Watanabe, T.; Gotoh, T.; Kiso,
T.; Koakutsu, A.; Kakimoto, S.; Shishikura, J. Bioorg. Med. Chem 2014, 22, 6899;
(b) Ogiyama, T.; Yonezawa, K.; Inoue, M.; Watanabe, T.; Sugano, Y.; Gotoh, T.;
Kiso, T.; Koakutsu, A.; Kakimoto, S.; Shishikura, J. Bioorg. Med. Chem. 2015.
accepted for publication.
6. (a) Zhou, S.-F.; Liu, J.-P.; Lai, X.-S. Curr. Med. Chem. 2009, 16, 2661; (b) Zhou, S.-
F. Curr. Pharm. Des. 2008, 14, 990.
7. (a) Stephens, P. J.; Devlin, F. J.; Pan, J.-J. Chirality 2008, 20, 643; (b) Nafie, L. A.;
Freedman, T. B. Vibrational Optical Activity Theory. In Circular Dichroism:
Principles and Applications; Berova, N., Nakanishi, K., Woody, R. W., Eds., 2nd
ed.; Wiley-VCH: New York, 2000; p 97.
8. (a) Brändén, G.; Sjögren, T.; Schnecke, V.; Xue, Y. Drug Discovery Today 2014, 19,
905; (b) Kumar, S.; Sharma, R.; Roychowdhury, A. Curr. Med. Chem. 2012, 19,
3605; (c) Stoll, F.; Göller, A.; Hillisch, A. Drug Discovery Today 2011, 16, 530; (d)
Moroy, G.; Martiny, V. Y.; Vayer, P.; Villoutreix, B. O.; Miteva, M. A. Drug
Discovery Today 2012, 17, 44; (e) Güner, O.; Bowen, J. P. Curr. Top. Med. Chem.
2013, 13, 1327.
9. (a) Bourdonnec, B.; Leister, L. Curr. Med. Chem. 2009, 16, 3093; (b) de Groot, M.
J.; Wakenhut, F.; Whitlock, G.; Hyland, R. Drug Discovery Today 2009, 14, 964.
10. The crystal structure of human cytochrome CYP2D6 in complex with RTZ was
registered by Wang et al. and available in RCSB protein data bank as PDB ID of
3TBG (http://www.rcsb.org/pdb/home/home.do).
Activityvehicle,
compound and with pre-incubation.
denotes activity obtained in the absence of
30 min
5.4.5. Animal experiments
Male ddY mice or Sprague-Dawley rats (SLC, Hamamatsu,
Japan) were used for in vivo experiments. Animals were group-
housed and kept under a 12-hour light/dark cycle (lights on from
7:30 AM to 7:30 PM) with free access to food and water. All animal
experimental procedures were approved by the Committee for
Animal Experiments of Astellas Pharma Inc. and conformed to
the International Guiding Principles for Biomedical Research
Involving Animals (CIOMS) and Guidelines for Proper Conduct of
Animal Experiments (Science Council of Japan, 2006). All efforts
were made to minimize the number of animals used and their
suffering.
We examined the antinociceptive effect of selected compounds
in the mice formalin test previously published by Hunskaar et al.14
with slight modifications. Details of the experimental procedures
have been described previously.5b In addition, the efficacy of
compound (1S)-8t in neuropathic pain was evaluated in the rat
SNL model reported by Chung et al.12 Evaluated compounds were
suspended and diluted in 0.5% methylcellulose solution. The dose
of drug is shown as the free form. Details of the experimental pro-
cedures have been described previously.15
11. Conformational
analyses
of
8-methoxytetrahydroisoquinoline
and
tetrahydroisoquinoline were conducted with Conformation Import module in
MOE with MMFF94x force field. Potential energy of each conformation was
calculated using PM6 method implemented in MOPAC 2012. The energetically
lowest conformation was selected as the most stable one of 8-
methoxytetrahydroisoquinoline and tetrahydroisoquinoline. The dihedral
angles between tetrahydroisoquinoline and the 1-phenyl group were
measured in MOE 2013 (Chemical Computing Group Inc., Quebec, 2014). The
dihedral angle is represented by the angles between the plane defined by
NH
β
(C
aꢁCbꢁCc) and the plane (CbꢁCcꢁCd). Each carbon atom (a, b, c, and d) was
α
assigned as depicted in Figure 3.
R
δ
12. Kim, S. H.; Chung, J. M. Pain 1992, 50, 355.
13. Li, C.; Xiao, J. J. Am. Chem. Soc. 2008, 130, 13208.
γ
R= H, MeO
14. Hunskaar, S.; Fasmer, O. B.; Hole, K. J. Neurosci. Meth. 1985, 14, 69.
15. Watabiki, T.; Kiso, T.; Kuramochi, T.; Yonezawa, K.; Tsuji, N.; Kohara, A.;
Kakimito, S.; Aoki, T.; Matsuoka, N. J. Pharmacol. Exp. Ther. 2011, 336, 743.
Figure 3. The dihedral angles of tetrahydroisoquinoline derivatives.