Design, synthesis, and biological evaluation of novel biphenyl-4-carboxamide derivatives as orally available TRPV1 antagonists

Article abstract of DOI:10.1016/j.bmc.2018.06.001

A new series of transient receptor potential vanilloid type 1 (TRPV1) antagonists were designed and synthesized from N-(3-hydroxyphenyl)-2-(piperidin-1-ylmethyl)biphenyl-4-carboxamide hydrochloride (8). SAR studies identified (R)-N-(1-methyl-2-oxo-1,2,3,4-tetrahydro-7-quinolyl)-2-[(2-methylpyrrolidin-1-yl)methyl]biphenyl-4-carboxamide hydrochloride (ASP8370, 7), as a compound with high aqueous solubility, satisfactory stability in human liver microsomes, and reduced CYP3A4 inhibition. ASP8370 was selected as a clinical development candidate with significant ameliorative effects on neuropathic pain. SAR studies also revealed the structural mechanisms underlying the switching between TRPV1 antagonism and agonism.

Full text of DOI:10.1016/j.bmc.2018.06.001

Accepted Manuscript
Design, synthesis, and biological evaluation of novel biphenyl-4-carboxamide
derivatives as orally available TRPV1 antagonists
Hiromasa Oka, Koichi Yonezawa, Akio Kamikawa, Kazuhiro Ikegai, Norio
Asai, Shohei Shirakami, Satoshi Miyamoto, Toshihiro Watanabe, Tetsuo Kiso,
Yukihiro Takemoto, Seiji Tamura, Takahiro Kuramochi
PII:
S0968-0896(18)30537-6
https://doi.org/10.1016/j.bmc.2018.06.001
BMC 14393
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
20 March 2018
1 June 2018
1 June 2018
Please cite this article as: Oka, H., Yonezawa, K., Kamikawa, A., Ikegai, K., Asai, N., Shirakami, S., Miyamoto,
S., Watanabe, T., Kiso, T., Takemoto, Y., Tamura, S., Kuramochi, T., Design, synthesis, and biological evaluation
of novel biphenyl-4-carboxamide derivatives as orally available TRPV1 antagonists, Bioorganic & Medicinal
Chemistry (2018), doi: https://doi.org/10.1016/j.bmc.2018.06.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, synthesis, and biological evaluation of novel biphenyl-4-carboxamide
derivatives as orally available TRPV1 antagonists
Hiromasa Oka*, Koichi Yonezawa, Akio Kamikawa, Kazuhiro Ikegai, Norio Asai,
†
Shohei Shirakami, Satoshi Miyamoto , Toshihiro Watanabe, Tetsuo Kiso, Yukihiro
Takemoto, Seiji Tamura, and Takahiro Kuramochi
Drug Discovery Research, Astellas Pharma Inc., 21 Miyukigaoka, Tsukuba, Ibaraki
3
05-8585, Japan
*
Corresponding author:
Tel: +81-29-863-6699, Fax: +81-29-856-8560
E-mail: hiromasa.oka@astellas.com
†
Present address:
Life Science Department, NARD Institute Ltd., 5-4-1 Minatojima Minamimachi
Chuo-ku, Kobe 650-0047, Japan
1

Abstract
A new series of transient receptor potential vanilloid type 1 (TRPV1) antagonists were
designed
and
synthesized
from
N-(3-hydroxyphenyl)-2-(piperidin-1-ylmethyl)biphenyl-4-carboxamide hydrochloride
(
(
8).
SAR
studies
identified
R)-N-(1-methyl-2-oxo-1,2,3,4-tetrahydro-7-quinolyl)-2-[(2-methylpyrrolidin-1-yl)meth
yl]biphenyl-4-carboxamide hydrochloride (ASP8370, 7), as a compound with high
aqueous solubility, satisfactory stability in human liver microsomes, and reduced
CYP3A4 inhibition. ASP8370 was selected as a clinical development candidate with
significant ameliorative effects on neuropathic pain. SAR studies also revealed the
structural mechanisms underlying the switching between TRPV1 antagonism and
agonism.
2

Graphical Abstract
HCl
HCl
8
ASP8370 (7)
TRPV1 IC : 0.53 M
TRPV1 IC : 0.51 M
50
50
aqueous solubility (pH 6.8): 2.2 M
weak agonist
aqueous solubility (pH 6.8): > 100 M
antagonist
3

Keywords
TRPV1 antagonist, Neuropathic pain, Agonist, Biphenyl-4-carboxamide, Oral
absorption
4

1
. Introduction
TRPV1 is a non-selective cation channel with high Ca permeability that is
2
+
predominantly expressed in both central and peripheral terminals of non-myelinated
1
C-fibers, where it plays an important role in pain signal transmission. In addition to
exogenous agonists such as capsaicin, TRPV1 is activated by a number of functionally
unrelated stimuli, such as protons, heat, and some endogenous vanilloids produced by
2
inflamed tissues. The expression of TRPV1 is up-regulated in non-myelinated C-fibers
3
-5
as well as A-fibers under certain nerve injury and inflammatory disease conditions.
Capsaicin, a well-known TRPV1 agonist, is clinically effective for treating a variety
of neuropathic pain conditions, with its analgesic actions are thought to result from the
inhibition of TRPV1 function by desensitization. Interestingly, in addition to TRPV1
6
agonists, TRPV1 antagonists also show analgesic potency. For example, BCTC (1) and
capsazepine (2), well-known TRPV1 antagonists, exhibit efficacy against both
inflammatory and neuropathic pain in animal models (Figure 1). Therefore, inhibition of
TRPV1 function may be an effective approach to the treatment of pain.
7
A large number of TRPV1 antagonists have been discovered to date, with some
8
9
having reached clinical development, including AMG-517 (3), ABT-102 (4), and
1
0
SB-705498 (5). A phase I clinical trial of 3 demonstrated dose-limiting hyperthermia
5

8
in both healthy volunteers and patients with postoperative dental pain. Compound 4
9
also induced hyperthermia. Interestingly, the hyperthermic potential of TRPV1
antagonists varies among compounds. It is well established that the TRPV1 channel can
be activated through three main mechanisms, namely by chemical ligands such as
1
1
capsaicin, heat, and acidic pH. Recent reports have revealed that, whereas blockade of
proton-induced activation of TRPV1 causes hyperthermia, inhibition of ligand- or
1
2,13
heat-induced pathways has no effect on body temperature.
For example, a
modality-selective TRPV1 antagonist, JYL1421 (6), which selectively inhibits
ligand-induced TRPV1 activation, is recognized as being a potential hyperthermia-free
1
2
agent.
In 2011, a study by our colleagues Watabiki et al., reported the biological
characterization of ASP8370 (7), a novel TRPV1 antagonist with oral absorbability and
1
4
brain penetrability. This compound ameliorates neuropathic pain without causing
hyperthermia at analgesic doses.
Our early exploratory approach identified a biphenyl-4-carboxamide 8 as a TRPV1
binder with an IC50 value of 0.53 μM, but this lead compound showed poor aqueous
solubility (2.2 μM at pH 6.8). Here, we describe the optimization pathway used to
generate 7 from 8 (Figure 2). In addition, we examined the structure activity
6

relationships (SARs) of TRPV1 antagonistic activity, as well as other issues including
water solubility, human cytochrome P450 3A4 (CYP3A4) inhibitory activity, and
metabolic stability in human liver microsomes (HLMs). We also examined the
structural mechanism underlying TRPV1 agonism and antagonism.
7

1
2
3
BCTC
Capsazepine
AMG-517
4
5
6
ABT-102
SB-705498
JYL1421
Figure 1. Selected TRPV1 antagonists.
HCl
HCl
ASP8370 (7)
TRPV1 IC : 0.51 M
8
TRPV1 IC : 0.53 M
50
50
aqueous solubility (pH 6.8): 2.2 M
aqueous solubility (pH 6.8): > 100 M
Figure 2. Compounds 8 and 7 (ASP8370).
8

2
. Chemistry
Biphenyl-4-carboxamide derivatives were prepared as illustrated in Scheme 1.
Suzuki-Miyaura coupling of 9 and phenylboronic acid yielded 10, which was converted
to benzyl bromide 11. Nucleophilic substitution of the bromo group with corresponding
secondary amines gave 12a–g, which were hydrolyzed to yield carboxylic acid
intermediates 13a–g. Condensation of 13a–g with anilines, which attached to various
substituent groups or condensed with various five-, six-, or seven-membered-rings,
resulted in target biphenyl-4-carboxamide derivatives 7, 8, and 14–36.
e, f or g
a
c
9
10 (Y = H)
1 (Y = Br)
12a–g (R = Et)
13a–g (R = H)
7, 8, 14–36
b
d
1
Scheme 1. Reagents and conditions: (a) PhB(OH)
toluene, reflux (99%); (b) NBS, AIBN, CCl
33%‒quant); (d) aqueous NaOH, EtOH; (e) WSCD-HCl, ArNH
HBTU, ArNH , N-methylmorpholine, DCE (17‒57%); (g) SOCl
ArNH , DCE (5‒77%).
2
, Pd(PPh
3
)
4
, aqueous Na
2
CO ,
3
1
2
4
, reflux (quant); (c) R R NH, K
2
CO , DMF
3
(
2
, DMF (13%); (f)
or (COCl) , Et N,
2
2
2
3
2
9

3
. Results and Discussion
The potency of synthesized compounds against TRPV1 receptor was determined using
1
4
two consecutive in vitro assays : (1) a binding assay to evaluate the affinity of test
compounds to the TRPV1 receptor and (2) a functional assay to estimate their agonistic
activity. Binding studies were conducted with tritium-labeled resiniferatoxin (RTX) and
human TRPV1-HEK293 cell membranes, using a method reported by Szallasi et al.
1
5
2+
with slight modifications. TRPV1 agonistic activity was assessed using a Ca influx
2
+
assay that measured intracellular Ca levels in CHO cells that stably expressed human
TRPV1. We reported the maximal agonistic activity of the test compounds, which was
2
+
defined as the ratio of maximal Ca levels induced by the compound (%) to maximal
2
+
Ca
levels induced by 1 μM of capsaicin, which was regarded as 100%.
Time-dependent CYP3A4 inhibitory activity was defined as the relative activity of
midazolam with and without 30 min of pre-incubation in the presence of a test
compound (5 μM).
During our early TRPV1 research program, we identified compound 8 as a lead which
bound to TRPV1 receptor with an IC50 value of 0.53 μM. However, the aqueous
solubility of 8 was very poor (2.2 μM at pH 6.8). High aqueous solubility is widely
1
6
recognized as a requirement for good in vivo bioavailability. In addition, it was
1
0

identified that compound 8 showed weak TRPV1 agonistic activity (44%), making it
unsuitable for expected mechanism of action to obtain therapeutic efficiency. Therefore,
we defined our synthetic strategy to reduce TRPV1 agonistic activity below 5%.
Furthermore, we added the evaluation of time-dependent CYP3A4 inhibitory potency
and microsomal stability to in vitro screening cascade from the beginning of the
program. Inhibition of CYP3A4 should be avoided to reduce the risk of clinically
1
7
undesirable event caused by severe drug-drug interactions, and good microsomal
stability is known to be important to achieve a satisfactory in vivo efficacy.
Given the detailed SAR study by Walpole et al. concerning the correlation between
TRPV1 agonism/antagonism switching and phenol scaffold in a series of thiourea
1
8
derivatives, we hypothesized that the phenolic proton in 8 might be important for
agonistic activity. Therefore, we tested whether the addition of anilide (ArNH) subunit
except 3-hydroxyaniline, could reduce TRPV1 agonistic activity. The results are
summarized in Table 1. Monocyclic derivatives 14, 15, and 16, which possess
methoxyphenyl, dimethylaminophenyl, dimethylaminopyridyl groups respectively,
exhibited reduced TRPV1 agonistic activity and improved aqueous solubility compared
to 8. Nevertheless, all three compounds had high human liver microsomal intrinsic
clearance (CLint > 450 mL/min/kg) which means that none of them exhibited
1
1

satisfactory microsomal stability.
We subsequently examined 6,5- and 6,6-fused bicyclic amide derivatives (compounds
1
7–25). Interestingly, whereas all of 6,5-fused bicyclic compounds except 21, behaved
as weak or even full TRPV1 agonists (compounds 17–20), 6,6-fused derivatives showed
markedly suppressed agonistic activity (compounds 22–25). Compounds 21–23
exhibited reduced TRPV1 agonistic activity with strong binding affinity, however these
three compounds also showed time-dependent CYP3A4 inhibitory potency (38‒75%)
1
9
along with increased lipophilicity (ACD logP 4.60–5.20). On the other hand,
1
H-quinolin-2-one derivative 24, which was less lipophilic (ACD logP 4.40) than
compounds 21–23, showed weaker time-dependent CYP3A4 inhibitory activity.
Compound 24 did not demonstrate only satisfactory TRPV1 affinity, reduced agonistic
activity, and weak time-dependent CYP3A4 inhibitory activity, but also good
microsomal stability (CLint = 96 mL/min/kg). However, its aqueous solubility was not
satisfactory (11 μM). Intriguingly, conversion into 3,4-dihydro-1H-quinolin-2-one
substructure (25) demonstrated much higher aqueous solubility (94 μM) than 24 with
preferable TRPV1 binding affinity and low agonistic activity. More flexible
substructure in 25 seemingly brought the improved aqueous solubility. Ring expansion
of piperidin-2-one in 25 to azepan-2-one substantially reduced TRPV1 binding affinity
1
2

(
26).
1
3

Table 1
Effect of anilide subunit on TRPV1 binding affinity, agonistic TRPV1 activity, aqueous solubility,
time-dependent CYP3A4 inhibitory activity, and human liver microsomal intrinsic clearance (CLint
)
1
4

TRPV1
agonistic
activity
TRPV1
IC50
Solubility
pH 6.8
(μM)
Time-dependent
CLint
(mL/min/kg)
f
Compd
Ar
CYP3A4 inhibitory
ACD logP
e
(
μM)
activity (%)
(
%)
a
d
d
8
0.53
0.35
0.70
0.32
0.48
0.47
0.58
0.14
0.073
0.15
0.023
0.098
0.095
1.8
44
4.4
7.3
0.2
52
2.2
98
NT
NT
4.45
4.96
5.25
4.55
3.97
4.41
4.30
4.29
4.60
5.20
5.15
4.40
4.48
4.97
a
1
1
4
5
78
67
89
461
>1000
542
a
87
1
6
>100
b
b
a
d
d
d
1
1
1
2
2
2
2
2
2
2
7
NT
NT
NT
d
d
d
8
58
NT
NT
NT
d
d
d
9
0
1
105
57
NT
NT
NT
a
a
d
d
d
NT
NT
NT
1.3
3.8
0.2
2.3
0.7
>100
2.5
38
75
52
86
96
239
540
287
96
b
c
2
3
4
5
6
>100
11
a
a
a
94
96
d
d
d
d
NT
NT
NT
NT
a
b
c
d
e
Hydrochloride salt.
Dihydrochloride salt.
Dihydrobromide salt.
Not tested.
Time-dependent CYP3A4 inhibitory activity was defined as the relative activity of midazolam with
and without 30 min of pre-incubation (see Section 5.3).
f
19
ACD logP was calculated using ACD/Percepta.
1
5

Table 2
Optimization of anilide subunit
TRPV1
agonistic
activity
TRPV1
IC50
Solubility
pH 6.8
(μM)
Time-dependent
CLint
(mL/min/kg)
e
Compd
Ar
CYP3A4 inhibitory
ACD logP
d
(
μM)
activity (%)
(
%)
a
2
2
2
2
3
3
5
7
8
9
0
0.095
0.7
0.6
1.3
1.6
33
94
>100
89
96
106
114
94
96
4.48
4.35
4.88
5.41
3.65
5.28
a
a
a
a
0.072
0.086
0.82
130
221
496
257
25
0.20
>100
102
b
c
c
c
1
0.32
2.6
NT
NT
NT
a
b
c
d
Hydrochloride salt.
Oxalate.
Not tested.
Time-dependent CYP3A4 inhibitory activity was defined as the relative activity of midazolam with
and without 30 min of pre-incubation (see Section 5.3).
e
19
ACD logP was calculated using ACD/Percepta.
1
6

Since Compound 25 showed preferable in vitro properties, further optimization of
anilide subunit had been conducted (Table 2). N-Methylated compound 27 exhibited
even better aqueous solubility than 25; however, introduction of a bulkier group on the
nitrogen atom reduced TRPV1 potency, solubility, and metabolic stability in HLMs
along with increased lipophilicity (28 and 29). Conversion of the piperidin-2-one
scaffold in 25 into a corresponding morpholinone or N-methylpiperidine also reduced
TRPV1 binding potency (30 and 31). Compound 30 behaved as a weak TRPV1 agonist
(
33%). Introduction of an oxygen atom into the para-position of anilinic nitrogen, as in
compound 30 or 19, tended to enhance TRPV1 agonistic activity.
Although our results so far did not provide a clear indication of which structural
component was key to the TRPV1 agonist/antagonist activity of the compound, the
property of the ArNH subunit appeared to be important (Tables 1 and 2). We
hypothesized that two different elements may affect the agonist/antagonist switching.
The first one is the existence of a dissociative proton, which is included in the anilide
1
8
subunit, evokes TRPV1 agonism. For example, compounds 8 and 17–20, which
possess dissociative protons exhibit agonistic potency for TRPV1. On the other hand,
compounds 14–16 and 21–23, which have no dissociative protons in the anilide subunit,
do not behave as TRPV1 agonists. According to the molecular dynamics simulation
1
7

study by using homology modeling, TRPV1 agonists and antagonists shared the same
binding pocket and similar binding interactions, however, it is indicated that agonists
tend to induce more molecular flexibility than antagonists and form hydrophilic
2
0
interactions with additional residues (Tyr554 and Arg557) of TRPV1. We suppose the
active proton in the anilide subunit may play an essential role to form such hydrophilic
interactions to express agonistic activity.
The second element is steric bulkiness of the anilide subunit. Our results showed that
6
,6-fused bicyclic derivatives (22–25, 27–29 and 31) had drastically reduced TRPV1
agonistic activity compared to 6,5-fused compounds (17–20). Although compounds 24
and 25 have dissociative protons in the anilide subunit as similar as 20, they did not
exhibit TRPV1 agonistic activity. It is likely that the bulkiness of the 6,6-fused bicyclic
group prevented the molecule from binding to the agonist recognition site of TRPV1.
The morpholinone substructure in compound 30 was considered to be less bulky than 25
because of the lack of benzylic protons at the para position of anilide nitrogen and
could not sufficiently inhibit the hydrophilic interaction to achieve agonistic activity,
which made 30 a weak agonist.
We identified compound 27, which exhibited the most desirable in vitro properties
(
strong affinity with TRPV1, low agonistic activity, high aqueous solubility (>100 μM),
1
8

reduced time-dependent CYP3A4 inhibitory activity, and good stability in HLMs),
through optimizing the ArNH subunit. For further optimization, we attempted to modify
the terminal phenyl ring in the biphenyl substructure, however all the attempts to
convert this phenyl ring into other substructure caused deterioration of in vitro
properties (data not shown). Therefore, we decided to modify the 2-(piperidinyl)methyl
unit attached to the biphenyl ring in compound 27 to seek an appropriate substituent
group (Table 3). Conversion of the piperidine ring to morpholine retained TRPV1
antagonistic activity but produced undesirable time-dependent CYP3A4 inhibition
despite its low lipophilicity (32). Thiomorpholine analog 33 showed reduced
time-dependent CYP3A4 inhibitory activity but it was not metabolically stable in HLMs
(
CLint = 716 mL/min/kg). Pyrrolidine analog 34 exhibited reduced TRPV1 antagonistic
activity, which was regained in (S)-2-methyl pyrrolidine analog 7. The R-enantiomer 35
was less potent than the S-isomer 7. Introduction of a methoxymethyl group instead of a
methyl at the 2-position (compound 36) produced undesirable time-dependent CYP3A4
inhibitory potency and destabilization in HLMs (CLint = 962 mL/min/kg). In the amine
modification, there was no clear correlation between lipophilicity of the compounds and
time-dependent CYP3A4 inhibition. Introduction of an oxygen atom in the amine
substructure was considered to lead to time-dependent CYP3A4 inhibition (compounds
1
9

3
2 and 36).
2
0

Table 3
Effect of aliphatic amino group modification on in vitro potency and ADME properties
TRPV1
agonistic
activity
TRPV1
IC50
Solubility
pH 6.8
(μM)
Time-dependent
CLint
(mL/min/kg)
f
Compd
R
CYP3A4 inhibitory
ACD logP
d
(
μM)
activity (%)
(
%)
a
2
3
3
3
7
2
3
4
0.072
0.15
0.029
1.3
0.6
1.5
0.2
2.0
0.1
0.1
0.2
>100
13
106
65
131
285
716
4.35
2.77
3.78
3.79
4.28
4.28
3.44
a
a
a
<1
114
c
c
c
NT
NT
NT
a
e
e
7
0.51
0.89
0.15
>100
>100
8.5
92
123
a
c
3
3
5
NT
114
962
b
6
34
a
b
c
d
Hydrochloride salt.
Oxalate.
Not tested.
Time-dependent CYP3A4 inhibitory activity was defined as the relative activity of midazolam with
and without 30 min of pre-incubation (see Section 5.3).
e
Data was obtained for salt-free form.
f
19
ACD logP was calculated using ACD/Percepta.
2
1

Both compounds 27 and 7 had satisfactory in vitro properties and compound 7 exhibited higher
efficacy in our in vivo screening assay (data not shown), although the TRPV1 binding affinity of 7 was
7
-fold weaker than that of 27. Therefore, we chose compound 7 as an orally available candidate drug
for development (ASP8370). Its oral administration was shown to have significant effects on
2
1
mechanical allodynia in a spinal nerve ligation (Chung) rat model without causing hyperthermia at
analgesic doses. In addition, 7 functions by inhibiting capsaicin-induced stimulation rather than
1
4
proton-evoked activation of TRPV1. We speculate that under acidic conditions, the amino group in 7
might be protonated to reduce hydrophobic interactions between 7 and TRPV1, leading to suppressed
1
4
its antagonistic activity. The biological evaluation of 7 has been reported in detail elsewhere.
2
2

4
. Conclusion
Here, we established detailed SARs of TRPV1 binding affinity, TRPV1 agonistic activity, aqueous
solubility, time-dependent CYP3A4 inhibition potency, and metabolic stability in HLMs beginning
from compound 8, which we identified through our early research program on TRPV1. Our study
revealed the structural relationship between the amide group and agonistic interaction with the TRPV1
receptor. Introduction of 6,6-membered heterocycles eliminated the undesirable agonistic activity. Our
SAR results further suggested that the presence of an active proton on the atom at the para-position of
the anilide nitrogen and the bulkiness of the ArNH scaffold influenced the agonistic/antagonistic
property of the compound. Finally, introduction of a methyl group at the 2-position in the pyrrolidine
ring successfully leaded to the development of compound 7 as a potent and orally available TRPV1
antagonist with high aqueous solubility, weak time-dependent CYP3A4 inhibition, and good stability
in HLMs.
2
3

5
5
. Experimental
.1. Chemistry
1
H NMR spectra were recorded on a Varian VNS-400, JEOL JNM-LA300, JNM-LA400, or
JNM-AL400 and the chemical shifts were expressed in δ (ppm) values with tetramethylsilane as an
internal reference (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, dd = doublet of
doublets, and br = broad peak). Mass spectra (MS) were recorded on JEOL JMS-LX2000, Waters
Alliance HT-ZQ or ZQ2000. High-resolution mass spectra (HRMS) were recorded on Thermo
Scientific Exactive Plus. Elemental analyses were performed using Yanaco MT-6 (C, H, N) and
Dionex ICS-3000 (halogen, S) and were within ±0.4% of theoretical values. All reactions were carried
out using commercially available reagents and solvents without further purification unless otherwise
22
noted. Compound 9 was synthesized according to a previously reported method. The following
abbreviations are used: AIBN, 2,2’-azobisisobutyronitrile; DCE, 1,2-dichloroethane; DMF,
N,N-dimethylformamide; DMSO, dimethylsulfoxide; EtOAc, ethyl acetate; EtOH, ethanol; HBTU,
O-benzotriazol-1-yl-N,N,N',N'-tetramethyluronium hexafluorophosphate; MeOH, methanol; NBS,
N-bromosuccimide; WSCD, 1-ethyl-3-(3’-dimethylaminopropyl) carbodiimide.
5
.1.1. Ethyl 2-methylbiphenyl-4-carboxylate (10)
A mixture of water (150 mL), sodium carbonate (55.9 g, 527 mmol) and phenylboronic acid (38.6 g,
3
17 mmol) was added
9
(51.4 g, 211 mmol) dissolved in toluene (400 mL)
2
4

Tetrakis(triphenylphosphine)palladium(0) (4.0 g, 3.5 mmol) was added and the mixture was heated
under reflux for 2 h. After the reaction solution was cooled to room temperature, insoluble materials
were removed by filtration through a pad of Celite, water was added to the filtrate and the organic
layer was extracted with toluene. After the organic layer was dried over MgSO , the solvent was
4
removed by evaporation and the resulting residue was purified by silica gel column chromatography
1
(
hexane/EtOAc) to obtain 10 as a colorless oil (50.4 g, 99% yield). H NMR (DMSO-d
6
) δ 1.34 (3H, t,
J = 7.2 Hz), 2.29 (3H, s), 4.33 (2H, q, J = 7.2 Hz), 7.32–7.53 (6H, m), 7.81–7.87 (1H, m), 7.90 (1H,
+
m); MS (FAB) m/z 241 [M+H] .
5
.1.2. Ethyl 2-(bromomethyl)biphenyl-4-carboxylate (11)
Compound 10 (10 g, 42 mmol) was dissolved in carbon tetrachloride (130 mL), and the mixture was
heated under reflux. NBS (7.78 g, 43.7 mmol) and AIBN (136 mg, 0.828 mmol) were added, the
reaction solution was heated under reflux for 1.5 h and the reaction solution was cooled to room
temperature. The precipitate was removed by filtration, water was added to the filtrate, and the organic
layer was extracted with carbon tetrachloride. After the resulting organic layer was dried over MgSO ,
4
1
the solvent was removed by evaporation to give 11 as a colorless solid (13.6 g, quantitative). H NMR
(
(
DMSO-d
6
) δ 1.35 (3H, t, J = 7.2 Hz), 4.36 (2H, q, J = 7.2 Hz), 4.68 (2H, s), 7.32–7.60 (6H, m), 7.96
+
1H, dd, J = 1.6, 8.0 Hz), 8.18 (1H, d, J = 1.6 Hz); MS (FAB) m/z 319 [M+H] .
5
.1.3. Ethyl 2-(piperidin-1-ylmethyl)biphenyl-4-carboxylate (12a)
Compound 11 (13.6 g, 42.6 mmol) was dissolved in DMF (50 mL), and a mixture of piperidine (5.3
g, 63 mmol), potassium carbonate (9.2 g, 67 mmol), and DMF (50 mL) was added, and the solution
was stirred for 3 h at room temperature. Water was added to the reaction solution and the mixture was
2
5

extracted with EtOAc. The resulting organic layer was dried over MgSO , and the solvent was
4
removed by evaporation. The obtained residue was purified by silica gel column chromatography
1
(
(
(
(
CHCl
3
/MeOH/28% aqueous NH
3
) to yield 12a as a pale yellow oil (12.6 g, 91% yield). H NMR
DMSO-d
6
) δ 1.34 (3H, t, J = 7.2 Hz), 1.30–1.48 (6H, m), 2.13–2.27 (4H, m), 3.27–3.39 (2H, m), 4.35
2H, q, J = 7.2 Hz), 7.33–7.50 (6H, m), 7.89 (1H, dd, J = 1.6, 8.0 Hz), 8.12 (1H, d, J = 1.6 Hz); MS
+
ESI) m/z 324 [M+H] .
5
.1.4. Ethyl 2-(morpholin-4-ylmethyl)biphenyl-4-carboxylate (12b)
Compound 12b was prepared from 11 and morpholine as a yellow solid (77% yield) in a manner
1
similar to that described for compound 12a. H NMR (DMSO-d
6
) δ 1.34 (3H, t, J = 7.1 Hz), 2.17–2.32
(
4H, m), 3.42 (2H, s), 3.47–3.57 (4H, m), 4.35 (2H, q, J = 7.1 Hz), 7.34–7.53 (6H, m), 7.91 (1H, d, J =
+
7
.9 Hz), 8.13 (1H, s); MS (FAB) m/z 326 [M+H] .
5
.1.5. Ethyl 2-(thiomorpholin-4-ylmethyl)biphenyl-4-carboxylate (12c)
Compound 12c was prepared from 11 and thiomorpholine in a manner similar to that described for
+
compound 12a (33% yield). MS (ESI) m/z 342 [M+H] .
5
.1.6. Ethyl 2-(pyrrolidin-1-ylmethyl)biphenyl-4-carboxylate (12d)
Compound 12d was prepared from 11 and pyrrolidine as a colorless solid (71% yield) in a manner
1
similar to that described for compound 12a. H NMR (DMSO-d
6
) δ 1.34 (3H, t, J = 7.0 Hz), 1.61–1.70
(
4H, m), 2.27–2.40 (4H, m), 3.54 (2H, s), 4.35 (2H, q, J = 7.0 Hz), 7.32–7.53 (6H, m), 7.89 (1H, dd, J
+
=
1.3, 8.0 Hz), 8.14 (1H, d, J = 1.3 Hz); MS (FAB) m/z 310 [M+H] .
5
.1.7. Ethyl (R)-2-[(2-methylpyrrolidin-1-yl)methyl]biphenyl-4-carboxylate (12e)
2
6

Compound 12e was prepared from 11 and (R)-2-methylpyrrolidine as a pale yellow oil (85% yield)
1
in a manner similar to that described for compound 12a. H NMR (DMSO-d
6
) δ 0.92 (3H, d, J = 6.0
Hz), 1.22–1.42 (1H, m), 1.34 (3H, t, J = 7.0 Hz), 1.46–1.64 (2H, m), 1.73–1.95 (2H, m), 2.17–2.35
1H, m), 2.65–2.79 (1H, m), 3.10 (1H, d, J = 13.6 Hz), 3.95 (1H, d, J = 13.6 Hz), 4.34 (2H, q, J = 7.0
(
Hz), 7.32–7.49 (6H, m), 7.88 (1H, dd, J = 1.7, 8.0 Hz), 8.12 (1H, d, J = 1.7 Hz); MS (FAB) m/z 324
+
[
M+H] .
5
.1.8. Ethyl (S)-2-[(2-methylpyrrolidin-1-yl)methyl]biphenyl-4-carboxylate (12f)
Compound 12f was prepared from 11 and (S)-2-methylpyrrolidine as a brown oil (85% yield) in a
1
manner similar to that described for compound 12a. H NMR (DMSO-d
6
) δ 0.92 (3H, d, J = 6.0 Hz),
1
2
7
.22–1.42 (1H, m), 1.34 (3H, t, J = 7.0 Hz), 1.46–1.64 (2H, m), 1.73–1.95 (2H, m), 2.17–2.35 (1H, m),
.65–2.79 (1H, m), 3.10 (1H, d, J = 13.6 Hz), 3.95 (1H, d, J = 13.6 Hz), 4.34 (2H, q, J = 7.0 Hz),
.32–7.49 (6H, m), 7.88 (1H, dd, J = 1.7, 8.0 Hz), 8.12 (1H, d, J = 1.7 Hz); MS (FAB) m/z 324
+
[
M+H] .
5
.1.9. Ethyl (S)-2-{[2-(methoxymethyl)pyrrolidin-1-yl]methyl}biphenyl-4-carboxylate (12g)
Compound 12g was prepared from 11 and (S)-2-(methoxymethyl)pyrrolidine as a yellow oil
1
(
(
quantitative) in a manner similar to that described for compound 12a. H NMR (DMSO-d
6
) δ 1.34
3H, t, J = 7.1 Hz), 1.38–1.64 (3H, m), 1.70–1.85 (1H, m), 1.91–2.05 (1H, m), 2.47–2.60 (1H, m),
2
.64–2.76 (1H, m), 3.01–3.36 (6H, m), 4.06 (1H, d, J = 13.6 Hz), 4.35 (2H, q, J = 7.1 Hz), 7.29–7.52
+
(
6H, m), 7.88 (1H, dd, J = 1.5, 7.9 Hz), 8.14 (1H, d, J = 1.5 Hz); MS (FAB) m/z 354 [M+H] .
5
.1.10. 2-(Piperidin-1-ylmethyl)biphenyl-4-carboxylic acid (13a)
2
7

Compound 12a (11 g, 34 mmol) was dissolved in EtOH (150 mL), and 1 M aqueous sodium
hydroxide (51 mL, 51 mmol) was added under cooling in an ice-water bath, then the mixture was
stirred for 10 h at room temperature. The mixture was cooled again in an ice-water bath, 1 M aqueous
hydrochloric acid (51 mL, 51 mmol) was added to the reaction solution and the solvent was removed
by evaporation to obtain 12.6 g of a pale pink solid which was a mixture of 13a and 1.5 molar
equivalents of sodium chloride. The mixture was used for the next reaction without further
+
purification. MS (FAB) m/z 296 [M+H] .
5
.1.11. 2-(Morpholin-4-ylmethyl)biphenyl-4-carboxylic acid (13b)
Compound 13b was prepared from 12b in a manner similar to that described for compound 13a. MS
+
(
FAB) m/z 298 [M+H] .
5
.1.12. 2-(Thiomorpholin-4-ylmethyl)biphenyl-4-carboxylic acid (13c)
Compound 13c was prepared from 12c in a manner similar to that described for compound 13a. This
compound was used for the next reaction without further structural confirmation.
5
.1.13. 2-(Pyrrolidin-1-ylmethyl)biphenyl-4-carboxylic acid (13d)
Compound 13d was prepared from 12d in a manner similar to that described for compound 13a. MS
+
(
FAB) m/z 282 [M+H] .
5
.1.14. (R)-2-[(2-Methylpyrrolidin-1-yl)methyl]biphenyl-4-carboxylic acid (13e)
Compound 13e was prepared from 12e in a manner similar to that described for compound 13a. MS
+
(
FAB) m/z 296 [M+H] .
5
.1.15. (S)-2-[(2-Methylpyrrolidin-1-yl)methyl]biphenyl-4-carboxylic acid (13f)
2
8

Compound 13f was prepared from 12f in a manner similar to that described for compound 13a. MS
+
(
FAB) m/z 296 [M+H] .
5
.1.16. (S)-2-{[(2-(Methoxymethyl)pyrrolidin-1-yl]methyl}biphenyl-4-carboxylic acid (13g)
Compound 13g was prepared from 12g in a manner similar to that described for compound 13a. MS
+
(
FAB) m/z 326 [M+H] .
5
.1.17. N-(3-Hydroxyphenyl)-2-(piperidin-1-ylmethyl)biphenyl-4-carboxamide hydrochloride (8)
A mixture of 3-aminophenol (240 mg, 2.20 mmol) and 13a (885 mg, mixture with 1.5 equivalents of
sodium chloride, 2.30 mmol of 13a) was suspended in DMF. WSCD-HCl (662 mg, 3.45 mmol) was
added at room temperature and the mixture was stirred for 18 h. EtOAc and 1 M aqueous hydrochloric
acid were added to the reaction solution and NaHCO
basic. The aqueous layer was extracted with EtOAc. The resulting organic layer was dried over
MgSO and the solvent was removed by evaporation. The obtained colorless solid was dissolved in
MeOH, then a 4 M hydrogen chloride solution in EtOAc (1 mL) was added, followed by removal of
3
was added until the aqueous layer was rendered
4
the solvent by evaporation. EtOH and Et
2
O were added to the obtained residue, which was solidified
1
to give 8 as a colorless solid. (121 mg, 13% yield). H NMR (DMSO-d
6
) δ 1.19–1.34 (1H, m),
1
.52–1.63 (3H, m), 1.73–1.88 (2H, m), 2.53–2.64 (2H, m), 3.18–3.24 (2H, m), 4.37 (2H, d, J = 5.4
Hz), 6.50–6.55 (1H, m), 7.13 (1H, t, J = 8.1 Hz), 7.29–7.33 (1H, m), 7.37–7.42 (2H, m), 7.44–7.56
(
(
5H, m), 8.02 (1H, dd, J = 1.7, 8.1 Hz), 8.67 (1H, d, J = 1.7 Hz), 9.45 (1H, s), 10.19 (1H, br s), 10.44
+
1H, s); HRMS (ESI) Calcd for C25
H
27
N
2
O
2
[M–HCl+H] : 387.2067. Found: 387.2057; Anal. Calcd
for C25
H
26
N
2
2
O ·HCl: C, 70.99; H, 6.43; N, 6.62; Cl, 8.38. Found: C, 70.71; H, 6.50; N, 6.58; Cl, 8.24.
5
.1.18. N-(3-Methoxyphenyl)-2-(piperidin-1-ylmethyl)biphenyl-4-carboxamide hydrochloride
(
14)
2
9