Design, synthesis, and evaluation of 4,6-diaminonicotinamide derivatives as novel and potent immunomodulators targeting JAK3

Article abstract of DOI:10.1016/j.bmc.2016.08.007

In organ transplantation, T cell-mediated immune responses play a key role in the rejection of allografts. Janus kinase 3 (JAK3) is specifically expressed in hematopoietic cells and associated with regulation of T cell development via interleukin-2 signaling pathway. Here, we designed novel 4,6-diaminonicotinamide derivatives as immunomodulators targeting JAK3 for prevention of transplant rejection. Our optimization of C4- and C6-substituents and docking calculations to JAK3 protein confirmed that the 4,6-diaminonicotinamide scaffold resulted in potent inhibition of JAK3. We also investigated avoidance of human ether-a-go-go related gene (hERG) inhibitory activity. Selected compound 28 in combination with tacrolimus prevented allograft rejection in a rat heterotopic cardiac transplantation model.

Full text of DOI:10.1016/j.bmc.2016.08.007

Accepted Manuscript
Design, synthesis, and evaluation of 4,6-diaminonicotinamide derivatives as
novel and potent immunomodulators targeting JAK3
Yutaka Nakajima, Naohiro Aoyama, Fumie Takahashi, Hiroshi Sasaki, Keiko
Hatanaka, Ayako Moritomo, Masamichi Inami, Misato Ito, Koji Nakamura,
Fumihiro Nakamori, Takayuki Inoue, Shohei Shirakami
PII:
S0968-0896(16)30600-9
DOI:
http://dx.doi.org/10.1016/j.bmc.2016.08.007
BMC 13190
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
1 July 2016
5 August 2016
6 August 2016
Please cite this article as: Nakajima, Y., Aoyama, N., Takahashi, F., Sasaki, H., Hatanaka, K., Moritomo, A., Inami,
M., Ito, M., Nakamura, K., Nakamori, F., Inoue, T., Shirakami, S., Design, synthesis, and evaluation of 4,6-
diaminonicotinamide derivatives as novel and potent immunomodulators targeting JAK3, Bioorganic & Medicinal
Chemistry (2016), doi: http://dx.doi.org/10.1016/j.bmc.2016.08.007
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 evaluation of 4,6-diaminonicotinamide derivatives as novel and
potent immunomodulators targeting JAK3
*
Yutaka Nakajima, Naohiro Aoyama, Fumie Takahashi, Hiroshi Sasaki, Keiko Hatanaka, Ayako
Moritomo, Masamichi Inami, Misato Ito, Koji Nakamura, Fumihiro Nakamori, Takayuki Inoue
*
and Shohei Shirakami
Drug Discovery Research, Astellas Pharma Inc., 21, Miyukigaoka, Tsukuba, Ibaraki 305-8585,
Japan
*
Corresponding authors:
Tel.: +81 (0)29-829-6185
fax: +81 (0)29 852 5387
E-mail address: yutaka.nakajima@astellas.com (Y. Nakajima,); shohei.shirakami@astellas.com
(
S. Shirakami).
1

Abstract
In organ transplantation, T cell-mediated immune responses play a key role in the rejection of
allografts. Janus kinase 3 (JAK3) is specifically expressed in hematopoietic cells and associated
with regulation of T cell development via interleukin-2 signaling pathway. Here, we designed
novel 4,6-diaminonicotinamide derivatives as immunomodulators targeting JAK3 for prevention
of transplant rejection. Our optimization of C4- and C6-substituents and docking calculations to
JAK3 protein confirmed that the 4,6-diaminonicotinamide scaffold resulted in potent inhibition
of JAK3. We also investigated avoidance of human ether-a-go-go related gene (hERG)
inhibitory activity. Selected compound 28 in combination with tacrolimus prevented allograft
rejection in a rat heterotopic cardiac transplantation model.
Keywords: JAK3 (Janus kinase 3), immunomodulator, transplant rejection, hERG (human ether-
a-go-go related gene)
2

1
. Introduction
Prevention of transplant rejection is urgently required for patients with end-stage organ
1
failure. Transplant rejection is triggered by T cell recognition of allografts, resulting in
1
production of interleukin-2 (IL-2) and further T cell differentiation and proliferation. The
immune responses injure allografts, resulting in graft loss. In current immunotherapy, calcineurin
inhibitors (CNIs) such as tacrolimus and cyclosporin A are used to treat transplant rejection.
CNIs inhibit IL-2 production and the following T cell activation, and are therefore highly
effective in preventing acute rejection. However, use of CNIs require constant monitoring of
2
-4
drug levels to manage the related nephrotoxic and neurotoxic side effects. Co-administration
with a drug working via a distinct mechanism of action will likely reduce the drug levels
required and side effects of CNIs.
Janus kinases (JAKs) are a family of cytoplasmic protein tyrosine kinases consisting of
JAK1, JAK2, JAK3, and TYK2, each of which are associated with corresponding cytokine
5
-9
receptors to mediate signal transduction. JAK3 is involved in the signaling pathways of IL-2,
IL-4, IL-7, IL-9, IL-15, and IL-21. IL-2 stimuli induce JAK3 activation following the
phosphorylation of signal transducers and activators of transcription protein 5 (STAT5) in the
cytoplasm. Phosphorylated STAT5 then undergoes dimerization and translocation to the nucleus,
where it activates the transcription of specific genes related to the differentiation, proliferation,
and survival of T cells. In contrast to the ubiquitous expression of other JAKs throughout the
1
0,11
body, JAK3 is predominantly expressed in hematopoietic cells.
The phenotype of JAK3
knockout mice is defined by immunodeficiency with a marked reduction in the numbers of
1
2,13
functional T cells.
JAK3 inhibitors might therefore be novel and safe immunomodulators for
the prevention of organ transplant rejection. Regarding other JAKs, JAK1 is associated with IL-2
3

receptors and the regulation of T cell function in concert with JAK3, and is also involved in the
5
-9
signaling pathways of IL-6 and interferon (IFN)-γ in inflammatory responses.
JAK2
participates in the differentiation and proliferation of erythrocytes, neutrophils, and thrombocytes
by mediating the signaling of hematopoietic growth factors such as erythropoietin, colony-
5
-9
stimulating factor, and thrombopoietin.
A number of JAK inhibitors have been studied,
is approved to use for the treatment of rheumatoid arthritis (RA). This compound exhibited in
1
4-18
and tofacitinib (compound 1, Figure 1)
1
9-21
vivo efficacy in various animal models of organ transplantation, RA, and psoriasis.
In our
laboratory, we identified a series of 1H-pyrrolo[2,3-b]pyridine-5-carboxamide derivatives to
2
2,23
show JAK3 inhibitory activity.
Docking calculations to JAK3 indicated that the 1H-
pyrrolo[2,3-b]pyridine-5-carboxamide scaffold interacts with the hinge region in the ATP-
binding site and that the C4-amino substituent interacts with the hydrophobic cavity. The
chemical modification of C4-substituents of lead compound 2 was investigated to enhance the
affinity of the hydrophobic cavity, and compound 3 was found to have increased JAK3
inhibitory activity over 2 (Figure 2). However, 3 exhibits undesirable human ether-a-go-go
related gene (hERG) inhibitory activity associated with cardiotoxicity, such as QT interval
2
4
prolongation. Our optimization study of the C4-substituent led to the identification of
compound 4, which exhibited potent JAK3 inhibitory activity and low hERG inhibitory activity
(
Figure 2).
4

1
Figure 1. Chemical structure of tofacitinib.
4
R =
2
: JAK3, IC50 = 5.1 nM
3 : JAK3, IC50 = 1.3 nM
rat CLint : 204 mL/min/kg
hERG, IC50 = 13 M
4 : JAK3, IC50 = 0.30 nM
rat CLint : 27.5 mL/min/kg
hERG, IC50 >100 M
rat CLint : >1000 mL/min/kg
Figure 2. Profiles of 1H-pyrrolo[2,3-b]pyridine-5-carboxamide derivatives.
As mentioned above, our previous study revealed that conversion of the C4-substituent was
important for interaction with the hydrophobic cavity and increase of JAK3 inhibitory activity.
On the other hand, the binding to the hinge region is generally critical for kinase inhibitory
activity, and modification of the scaffold moiety around hinge region may increase the affinity to
JAK3. Although many other JAK inhibitors have bicyclic fused heteroaryl scaffold comprised of
1
6-18
5
- and 6-membered rings (e.g., pyrrolopyrimidine of 1),
little studies are known about effect
of modification of the 5-membered ring moiety on JAK3 inhibitory activity. We therefore
designed and focused on a novel single pyridine ring of 4,6-diaminonicotinamide scaffold
5

mimicking the 1H-pyrrolo[2,3-b]pyridine-5-carboxamide. In our previous study, the 1H-
pyrrolo[2,3-b]pyridine-5-carboxamide moiety occupied the ATP-binding site to form a hydrogen
bond with Glu903 and Leu905 in the hinge region, with the N1-hydrogen atom serving as a
2
2
hydrogen donor and N7-nitrogen atom as an acceptor. As shown in Figure 3, a 4,6-
diaminonicotinamide scaffold was designed to retain two hydrogen bond interactions with the
hinge region, as this interaction is considered important for JAK3 inhibitory activity. In addition,
as the bicyclic aromaticity of the 1H-pyrrolo[2,3-b]pyridine ring contributed to CH-π interaction
with the Leu828 and Val836, introduction of a phenyl ring in C6-position of the 4,6-
diaminonicotinamide was considered effective for holding the scaffold around the hinge region.
Leu905
Glu903
Hydrogen
bond
interaction
6
4
CH-
interaction
Leu828, Val836
Figure 3. Design of 4,6-diaminonicotinamide derivatives.
Here, we describe our structure-activity relationship (SAR) study of the 4,6-
diaminonicotinamide derivatives for JAK3 inhibitory activity. In addition, reduction of hERG
inhibition was investigated via modulation of molecular lipophilicity and basicity. Further, the in
2
5
vivo efficacy of this series of compounds on a rat heterotopic cardiac transplant model was also
investigated.
6

2
. Chemistry
Scheme 1 shows the general synthesis of 4,6-diaminonicotinamide derivatives. Ethyl 4,6-
dichloronicotinate (5) was reacted with (1S,2R)-2-methylcyclohexanamine, and nucleophilic
substitution gave compound 6. The ester group of 6 was hydrolyzed to give 7, and successive
condensation with aqueous ammonia gave carboxamide 8. Treatment of 8 with aniline under a
Pd-catalyzed condition gave the 4,6-diaminonicotinamide derivative 9.
a
d
5
6 : R = CO Et
9
2
b
7
8
: R = CO H
2
c
: R = C(O)NH2
Scheme 1. Reagents and conditions: (a) (1S,2R)-2-methylcyclohexanamine hydrochloride,
DIPEA, n-BuOH, 120 °C; (b) 2 M NaOH, dioxane, 110 °C; (c) CDI, DMF, room temperature,
then 28% NH
4
OH, room temperature; (d) aniline, Pd
2
(dba) , 2-dicyclohexylphosphino-2',4',6'-
3
triisopropyl-1,1'-biphenyl, t-BuONa, t-BuOH, microwave, 130 °C.
Scheme 2 shows synthesis of the derivatives having N-cyanopyridyl-4-aminopiperidine in
C4-position. Similar to the synthesis of 8, compound 12 was prepared using 5 and N-Boc-4-
aminopiperidine. The Pd-catalyzed coupling reaction of 12 with a variety of arylamines gave
compound 13a-i. The Boc group of 13a-i was deprotected under acidic condition, followed by
reaction with 2-chloro-5-cyanopyridine to give compound 14a-i.
7

a
d
5
1
1
1
0 : R = CO Et
13a-i
2
b
c
1 : R = CO H
2
2 : R = C(O)NH2
1
3a, 14a : R =
13f, 14f : R =
e, f
1
1
3b, 14b : R =
3c, 14c : R =
13g, 14g : R =
13h, 14h : R =
1
4a-i
1
1
3d, 14d : R =
3e, 14e : R =
13i, 14i : R =
Scheme 2. Reagents and conditions: (a) N-Boc-4-aminopiperidine, DIPEA, n-BuOH, 120 °C; (b)
M NaOH, dioxane, 110 °C; (c) CDI, DMF, room temperature, then 28% NH OH, room
temperature; (d) amines, Pd (dba) , 2-dicyclohexylphosphino-2',4',6'-triisopropyl-1,1'-biphenyl,
t-BuONa (or K CO ), t-BuOH, microwave, 130-140 °C; (e) 4 M HCl, dioxane, room
temperature; (f) 2-chloro-5-cyanopyridine, K
50 °C).
2
4
2
3
2
3
2
CO
3
, DMSO, 100 °C (or Et
3
N, NMP, microwave,
1
8

Scheme 3 shows the synthesis of benzylamine derivatives. Coupling reaction of 17 with
aniline gave 18, and nucleophilic substitution of 17 with N-methylaniline gave 19 and with 3,4-
dimethoxybenzylamine gave 20. Deprotection of the 3,4-dimethoxybenzyl group of 20 by TFA
gave compound 21.
1
1
8 : R =
9 : R =
a
d, e, or f
5
1
1
1
5 : R = CO Et
2
b
c
20 : R =
6 : R = CO H
2
g
7 : R = C(O)NH2
2
1 : R =
Scheme 3. Reagents and conditions: (a) benzylamine, DIPEA, i-PrOH, 80 °C; (b) 6 M NaOH,
EtOH, 70 °C; (c) CDI, DMF, room temperature, then 28% NH OH, room temperature; (d)
aniline, Pd (dba) , 2-dicyclohexylphosphino-2',4',6'-triisopropyl-1,1'-biphenyl, t-BuONa, t-
4
2
3
BuOH, NMP, dioxane, microwave, 140 °C (for 18); (e) N-methylaniline, tetraethylammonium
chloride, NMP, 180 °C (for 19); (f) 3,4-dimethoxybenzylamine, DIPEA, NMP, microwave,
1
80 °C (for 20); (g) TFA, microwave, 100 °C.
Scheme 4 shows the synthesis of compound 27. Compound 23 was prepared by reacting 5
2
3
with chiral 4-amino-3-fluoropiperidine 22, which was synthesized as previously described. The
ester group of 23 was converted to carboxamide to give compound 25, followed by a coupling
9

reaction with 2-methylpyrimidine-4-amine to give 26. After deprotection of the Boc group, 26
was reacted with 2-chloro-5-cyanopyridine to give 27.
a
5
2
2
2
2
2
3 : R = CO Et
2
b
c
4 : R = CO H
2
5 : R = C(O)NH2
d
e, f
2
6
2
7
Scheme 4. Reagents and conditions: (a) DIPEA, DMF, 130 °C; (b) 2 M NaOH, dioxane, 110 °C;
c) CDI, DMF, room temperature, then 28% NH OH, room temperature; (d) 2-methylpyrimidin-
-amine, Pd (dba) , 2-dicyclohexylphosphino-2',4',6'-triisopropyl-1,1'-biphenyl, t-BuONa, t-
(
4
4
2
3
BuOH, 90 °C; (e) 4 M HCl, dioxane, room temperature; (f) 2-chloro-5-cyanopyridine, Et
NMP, 100 °C.
3
N,
Scheme 5 shows the synthesis of compound 28. After deprotection of the Boc group under
acidic condition, 13g was reacted with 4-fluorobenzonitrile to give 28.
1
0

a, b
1
3g
2
8
Scheme 5. Reagents and conditions: (a) 4 M HCl, dioxane, room temperature; (b) 4-
fluorobenzonitrile, K CO , DMSO, microwave, 140 °C.
2
3
1
1

3
. Results and discussion
We evaluated human JAK3 inhibitory activity of synthesized compounds, and JAK1
inhibitory activity was also evaluated, as JAK3 and JAK1 collaboratively regulate the IL-2
signaling pathway for T cell-mediated immune responses. JAK2 inhibitory activity was then
investigated, as JAK2 inhibition might be related to adverse hematopoietic effects such as
2
6
anemia. To determine the in vivo effect, a rat heterotopic cardiac transplantation model was
investigated using ACI rats as donors and Lewis rats as recipients, and median survival times
(
MST) of grafts were analyzed. In the model, we evaluated immunomodulating effect to prevent
transplant rejection via JAK3 pathway and following IL-2 mediated T cell development.
Table 1 shows the fundamental SAR study results for the 4,6-diaminonicotinamide
derivatives. In the C4-position, a methylcyclohexylamine corresponding to the C4-substituent of
2
and an N-cyanopyridyl-4-aminopiperidine corresponding to that of 3 were introduced to
interact with the hydrophobic cavity. The resulting respective compounds 9 and 14a exhibited
moderate JAK3 inhibitory activity (IC50 = 37 and 18 nM, respectively). These results suggest
that chemical modification of C4-position of the 4,6-diaminonicotinamide derivatives maintains
JAK3 inhibitory activity. Conversion to a simple aliphatic group such as a benzylamine retained
moderate JAK3 inhibitory activity (18; IC50 = 19 nM). To then validate the effect of the C6-
substituent on JAK3 inhibitory activity, modification of the C6-position of 18 was investigated.
N-methylation and deletion of the phenyl group resulted in loss of JAK3 inhibitory activity,
indicating that hydrogen bond donor and aromaticity around the hinge region were critical for
JAK3 inhibitory activity (19 and 21; IC50 >1000 nM, respectively).
1
2

Table 1
SARs of 4,6-diaminonicotinamide derivatives
a
Compound
Structure
JAK3 IC50 (nM)
9
37
18
1
1
1
4a
8
19
>1000
9
(
40% inh.@ 1 μM)
1
3

>1000
2
2
1
(
41% inh.@1 μM)
5.1
a
IC50 values are mean of duplicate experiments.
1
4

Optimization of the C6-substituent of 4,6-diaminonicotinamide derivatives is shown in
Table 2. Regarding the C4-substituent, the N-cyanopyridyl-4-aminopiperidine group was
selected as 3 showed preferable pharmacokinetic (PK) profiles such as metabolic stability in our
2
3
previous study. Compared to the 1H-pyrrolo[2,3-b]pyridine-5-carboxamide derivative 3, the 6-
anilino-nicotinamide derivative 14a have more flexible structure around the hinge region of
JAK3 due to the rotation of C6-substituent. We considered that mimicking the planar bicyclic
scaffold of 3 was effective for JAK3 inhibitory activity, because the aromaticity of 1H-
pyrrolo[2,3-b]pyridine contributed to CH-πinteraction with the Leu828 and Val836 (Figure 3).
Therefore, conversion of the aniline group of 14a to a 2-aminopyridine in the C6-position was
attempted to maintain the scaffold in a planar orientation by reducing steric hindrance (Figure 4).
steric
favorable for
hindrance
planar structure
Figure 4. Modification of C6-substituent.
In addition, a polar functional group such as pyridine in the C6-position might be more
favorable in the hydrophilic hinge region. Results showed that 14b achieved a 40-fold increase in
JAK3 inhibitory activity (IC50 = 0.46 nM) over 14a. Regarding the potency of JAK3 inhibitory
activity, the 4,6-diaminonicotinamide scaffold was superior to the 1H-pyrrolo[2,3-b]pyridine (3;
IC50 = 1.3 nM vs. 14b; IC50 = 0.46 nM). Further, 14b maintained moderate selectivity for JAK3
over JAK1 (6-fold selectivity) and JAK2 (11-fold selectivity). The potent JAK3 inhibitory
1
5

activity of 14b suggested that the 4,6-diaminonicotinamide was a potential scaffold to form
suitable binding to the hinge region.
However, despite the promising profile of 14b as a JAK3 inhibitor, this compound also
inhibited hERG with an IC50 value of 0.36 μM. As binding to the hERG channel is related to
2
3,24
hydrophobic interactions
, the pyridine in the C6-position of 14b was therefore converted to
another hetero aromatic ring to reduce molecular lipophilicity. Results showed that 6-
pyrazylamine analogue 14c and 6-pyrimidylamine analogue 14d maintained potent JAK3
inhibitory activity (IC50 = 0.72 and 0.61 nM, respectively). Substitution of pyrazine and
pyrimidine was then investigated, and introduction of a methyl group to the pyrazine (14e and
1
4f) was tolerable for JAK3 inhibitory activity (IC50 = 1.0 and 1.1 nM, respectively). In the case
of methylation of the pyrimidine, 14g exhibited increased JAK3 inhibitory activity over 14d
IC50 = 0.36 nM). Further, introduction of a bicyclic hetero ring in the C6-position was examined,
(
and benzoxazole analogues 14h and 14i showed moderate JAK3 inhibitory activity (IC50 = 22
and 25 nM, respectively). Regarding hERG inhibition of the 4,6-diaminonicotinamide
derivatives, the 6-pyrimidylamine analogues 14d and 14g with lower CLogP value (= 4.0 and 4.3,
respectively) tended to have reduced hERG inhibitory activity (IC50 = 17 and 8.1 μM,
respectively) compared to 14b while retaining potent JAK3 inhibitory activity.
1
6

Table 2
SARs of N-cyanopyridyl-4-aminopiperidine derivatives
6
4
3
JAK3
JAK1
JAK2
hERG
c
Compound
R
CLogP
5.1
a
a
a
b
IC50 (nM) IC50 (nM) IC50 (nM) IC50 (μM)
d
d
1
1
1
1
4a
4b
4c
4d
18
NT
2.9
2.5
3.5
190
4.9
4.1
5.5
NT
0.46
0.72
0.61
0.36
7.7
17
4.7
4.2
4.0
1
4e
1.0
1.9
4.0
5.4
4.7
1
1
4f
1.1
6.7
5.4
5.3
3.2
4.8
8.1
4.7
4.3
4g
0.36
1
7

d
d
1
4h
4i
22
NT
400
6.6
NT
4.7
d
1
3
25
NT
16
130
18
4.7
3.4
1.3
13.4
a
IC50 values are mean of duplicate experiments.
b
Inhibitory activity in Rb efflux assay (n = 2).
c
CLogP values calculated using ACD/Percepta 14.0.0.
d
NT = not tested.
1
8

Table 3 shows modification of the C4-substituent of the 6-pyrimidylaminonicotinamide
derivatives. Findings from our previous study suggested that the molecular basicity around the
N-cyanopyridyl-4-aminopiperidine may have been involved in hERG inhibitory activity, and
2
3
decreasing the basicity in the C4-substituent was effective for reduction of this activity. The
introduction of a fluoro group to the piperidine of 14g was therefore investigated. The basicity of
the 3-fluoropiperidine is considered to be reduced by electron withdrawing effect due to the
2
7,28
fluorine atom,
which effectively reduces hERG inhibitory as shown in the 1H-pyrrolo[2,3-
b]pyridine-5-carboxamide derivatives (4; hERG IC50 >100 μM). As expected, the 3-
fluoropiperidine analogue 27 showed substantially decreased hERG inhibitory activity (IC50
>
100 μM) while the JAK3 inhibitory activity was maintained (IC50 = 0.39 nM). The pyridine of
4g was then converted to a phenyl ring to remove the basicity of pyridine, and 28 exhibited a
large decrease in hERG inhibitory activity (IC50 >100 μM) with potent JAK3 inhibitory activity
IC50 = 0.80 nM). Despite an increase in molecular lipophilicity, conversion of the pyridyl
1
(
piperidine structure to the phenyl piperidine in the C4-substituent was mainly effective to reduce
hERG inhibitory activity. JAK3 selectivity of 28 toward JAK1 and JAK2 was slightly increased
compared to 27. In addition, 28 exhibited weak inhibitory activity of 19 representative kinases in
a kinase panel assay (see Supplementary data). In a series of the 4,6-diaminonicotinamide
derivatives, 28 was identified as a potent JAK3 inhibitor without hERG inhibitory activity.
1
9

Table 3
SARs of 6-pyrimidylaminonicotinamide derivatives
4
JAK3
JAK1
JAK2
hERG
c
Compound R1
R2
CLogP
4.3
a
a
a
b
IC50 (nM) IC50 (nM) IC50 (nM) IC50 (μM)
1
2
4g
7
H
F
0.36
0.39
5.4
3.2
1.1
8.1
0.24
>100
4.0
(
3S, 4R)
2
4
8
H
0.80
0.30
4.2
4.1
3.5
3.2
>100
>100
4.9
3.1
a
IC50 values are mean of duplicate experiments.
b
Inhibitory activity in Rb efflux assay (n = 2).
c
CLogP values calculated using ACD/Percepta 14.0.0.
2
0

To clarify the binding mode of the 4,6-diaminonicotinamide derivatives to JAK3, docking
calculations were conducted for 28. Figure 5 shows superposition of binding mode of 28 and 4.
Docking calculations were performed for the crystal structure of 1 bound to JAK3 (PDB code:
2
9
3
LXK ) using the XP mode in Glide (version 5.8; Schrödinger, LLC, New York, NY, 2012).
Around the hinge region, the 4,6-diaminonicotinamide scaffold of 28 maintained a planar
conformation, and the N1-nitrogen group formed a hydrogen bond with Leu905 via a hydrogen
acceptor while the C6-amino group formed one via a hydrogen donor. The C3-carbamoyl group
of 28 was located in the back of the hinge region and formed an additional hydrogen bond
interaction with Glu903. Further, the C3-carbamoyl group interacted with the gatekeeper amino
acid Met902. Interestingly, the position of the carbamoyl group of 28 was different from that of 4,
in spite of structural similarity between the 4,6-diaminonicotinamide scaffold and the 1H-
pyrrolo[2,3-b]pyridine-5-carboxamide. Although the pyridine ring of 28 was located in the
position of the corresponding pyridine ring of 4, the C3-carbamoyl group of 28 was directed to
narrow space in the back side, occupied by the pyrrole ring in the case of 4. The different
orientation of C3-carbamoyl group of 28 is likely due to the spatial capacity in the back of the
hinge region. The pyrrolopyridine ring of 4 was acceptable in the narrow area, while the
pyrimidinylaminopyridine moiety of 28 was not allowed. It was suggested that the bulky C6-
substituent of 4,6-diaminonicotinamide scaffold of 28 was directed to solvent exposure space,
and the small C3-carbamoyl group was turned to the back side to form hydrogen bond
interaction with Glu903 and Met902. The novel hydrogen bond of C3-carbamoyl group of 28
may retain the 4,6-diaminonicotinamide scaffold around the hinge region and contribute to
potent JAK3 inhibitory activity. Each of the aromaticity of pyridine and pyrimidine ring of 28
enabled a CH-π interaction with Leu828. The SAR results in Table 2 suggested that introduction
2
1

of a heteroaryl ring such as pyridine, pyrazine, and pyrimidine in C6-position was favorable for
the planar conformation to form the CH-π interaction, leading to potent JAK3 inhibitory activity.
Regarding the C4-substituent, the N-cyanophenyl-4-aminopiperidine group of 28 occupied the
hydrophobic cavity. In the back of the hydrophobic cavity, the cyanophenyl group formed a CH-
π interaction with Gly831 and a hydrogen bond with Arg953. These findings indicate that 28 was
effectively positioned in the JAK3 binding pocket and achieved high affinity to JAK3.
Figure 5. Predicted binding mode of compound 28 (blue) and 4 (pink) to human JAK3.
In vitro profiles of the selected compound 28 are shown in Table 4. The metabolic stability
of 28 in rat liver microsomes was acceptable, with an in vitro intrinsic clearance (CL ) value of
int
1
46 mL/min/kg, and no species difference was shown between rats and humans (human CLint =
2
2

1
09 mL/min/kg). In a parallel artificial membrane permeability assay (PAMPA), 28 exhibited a
-6
high effective permeability value (P >30 × 10 cm/sec). Table 5 shows PK parameters of 28
e
after oral (p.o.) and intravenous (i.v.) administration, and the oral bioavailability was 23.6%,
suggesting moderate oral exposure in rats.
Table 4
In vitro profiles of compound 28
Metabolic stability
a
Membrane permeability
CLint (mL/min/kg)
b
-6
P
e
(× 10 cm/sec)
Rat Human
1
46
109
>30
a
In vitro metabolism with liver microsomes in the presence of NADPH-generating system (n =2).
b
TM
PAMPA Evolution (pION Inc., USA), donor buffer pH: 6.5 (n = 3).
Table 5
Pharmacokinetic parameters after single administration of 28 to rats
i.v.
p.o.
(
1 mg/kg, n=3)
(1 mg/kg, n=3)
a
AUC0-24
t
1/2
V
dss
CLtot
C
max
t
max
AUC0-24
F
Species
Rat
(
ng·h/mL)
(h)
3.4
(L/kg) (mL/min/kg)
(ng/mL)
52.9
(h)
0.4
(ng·h/mL)
278
(%)
1177
3.5
14.3
23.6
a
F = bioavailability.
2
3

Selected compound 28 exhibited potent JAK3 inhibitory activity and acceptable oral
exposure, and therefore 28 was evaluated in an ACI-to-Lewis rat heterotopic cardiac transplant
model to validate the in vivo efficacy as an immunomodulator. In the model, allografts were
ultimately rejected by approximately 5 days after transplantation in untreated rats. In addition,
tacrolimus monotherapy at a dose of 0.02 mg/kg via intramuscular injection exhibited only
2
5
marginal efficacy in preventing allograft rejection (MST = 6.5 days). As shown in Table 6,
combination therapy of 28 with tacrolimus prevented allograft rejection and prolonged graft
survival (MST = 22.5 days). This result demonstrates the additive effect of JAK3 inhibitors on
current CNI-based immunosuppressive therapy, and 28 might therefore be an effective
immunomodulator for preventing transplant rejection.
Table 6
a
Effect of compound 28 on graft survival in a rat cardiac transplantation model
Median survival times
Treatment
(
days)
2
8, 10 mg/kg (p.o.)
2
2.5
b
with tacrolimus, 0.02 mg/kg (i.m.)
a
ACI rat hearts were heterotopically transplanted into Lewis rats (n = 4).
b
2
8 and tacrolimus were administered from day of transplantation for 14 days or by day of graft
rejection; i.m., intramuscular.
2
4

4
. Conclusion
We designed 4,6-diaminonicotinamide derivatives to modify interaction with hinge region, and
conducted an SAR study of JAK3 inhibitory activity. In the C6-position of the 4,6-
diaminonicotinamide derivatives, the introduction of a heteroaryl ring such as pyridine,
pyrazine, and pyrimidine was important for achieving potent JAK3 inhibitory activity, while N-
cyanopyridyl-4-aminopiperidine was acceptable as a C4-substituent (14b, 14c, and 14d).
Although a number of 4,6-diaminonicotinamide derivatives exhibited undesirable hERG
inhibitory activity, decreasing the molecular lipophilicity in the C6-position and modulating the
molecular basicity in the C4-position effectively reduced this inhibitory activity (27 and 28). A
docking calculation of 28 to JAK3 indicated the predicted binding mode around the hinge
region and hydrophobic cavity, suggesting that the 4,6-diaminonicotinamide derivative formed
not only conventional interaction of our previous series of compounds, but also novel hydrogen
bond interaction of C3-carbamoyl group and CH-π interaction of aryl ring around the hinge
region. In a rat heterotopic cardiac transplantation model, the combination of 28 and tacrolimus
resulted in prolonged graft survival. These findings suggest that the 4,6-diaminonicotinamide
derivatives are novel immunomodulators targeting JAK3 for treating organ transplant rejection.
2
5

5
5
. Experimental Section
.1. Chemistry
1
H NMR spectra were recorded on a Bruker Biospin AV400 or AV400M spectrometer.
Chemical shifts are expressed in δ units using tetramethylsilane as an internal standard (NMR
peak description: s = singlet; d = doublet; t = triplet; q = quartet; m = multiplet; and br = broad
peak). Mass spectra (MS) were recorded on a Hitachi LC/3DQMS M8000 or an Agilent HP1100
LC/MCD spectrometer. High-resolution mass spectroscopy (HRMS) spectra were recorded on a
Waters LCT Premier XE. Column chromatography was conducted using silica gel 60N (Kanto
TM
Chemical, 63-210 μm) or a Yamazen HI-FLASH
Column. The purity of all compounds
screened in biological assays was given by HPLC analysis with a ZORBAX Eclipse Plus C18
column (detection at 254 nm). Elemental analysis values were recorded on a YANACO JM-10
and were within 0.4% of the theoretical values calculated for C, H, and N.
5
.1.1. Ethyl 6-chloro-4-{[(1S,2R)-2-methylcyclohexyl]amino}nicotinate (6)
1S,2R)-2-Methylcyclohexanamine hydrochloride (510 mg, 3.41 mmol) and DIPEA (0.59
(
mL, 3.41 mmol) were added to a solution of ethyl 4,6-dichloronicotinate (5) (500 mg, 2.27
mmol) in n-BuOH (5 mL), and the mixture was stirred at 120 °C for 3 h. After adding (1S,2R)-2-
methylcyclohexanamine hydrochloride (102 mg, 0.68 mmol), the mixture was stirred at the same
temperature for additional 1 h. After cooling to room temperature, the mixture was extracted
with EtOAc, and washed with H
2
O. The organic layer was separated, dried over MgSO , and
4
concentrated under reduced pressure. The residue was purified by silica gel column
chromatography (n-hexane/EtOAc = 90/10 to 80/20) to give the title compound (660 mg, 98%).
1
H NMR (400 MHz, DMSO-d ) δ: 0.85 (3H, d, J = 6.8 Hz), 1.17-1.39 (8H, m), 1.43-1.70 (3H,
6
2
6

m), 1.82-1.93 (1H, m), 3.83-3.89 (1H, m), 4.32 (2H, q, J = 7.2 Hz), 6.91 (1H, s), 8.37 (1H, d, J =
+
8
.8 Hz), 8.54 (1H, s). MS (ESI) m/z: 297 (M + H) .
5
.1.2. 6-Chloro-4-{[(1S,2R)-2-methylcyclohexyl]amino}nicotinic acid (7)
NaOH aqueous solution (2 M; 4.45 mL, 8.89 mmol) was added to a solution of 6 (660 mg,
.22 mmol) in dioxane (6.6 mL), and the mixture was stirred at 110 °C for 2 h. After cooling to
2
4
°C, the mixture was acidified with 1 M HCl aqueous solution. The precipitate was filtrated and
1
washed with H
2
O to give the title compound (348 mg, 58%). H NMR (400 MHz, DMSO-d
6
) δ:
0
8
.85 (3H, d, J = 7.2 Hz), 1.21-1.69 (8H, m), 1.82-1.93 (1H, m), 3.78-3.85 (1H, m), 6.85 (1H, s),
+
.50 (1H, s), 8.63 (1H, d, J = 8.8 Hz), 13.38 (1H, brs). MS (ESI) m/z: 269 (M + H) .
5
.1.3. 6-Chloro-4-{[(1S,2R)-2-methylcyclohexyl]amino}nicotinamide (8)
Carbonyldiimidazole (CDI) (463 mg, 2.86 mmol) was added to a solution of 7 (384 mg,
1
.43 mmol) in DMF (3.84 mL), and the mixture was stirred at room temperature for 1 h. After
the addition of 28% NH
room temperature for 1 h. H
washed with H
DMSO-d
4
OH aqueous solution (0.89 mL, 7.15 mmol), the mixture was stirred at
2
O was added to the mixture, and the precipitate was filtrated and
1
2
O and IPE to give the title compound (350 mg, 91%). H NMR (400 MHz,
6
) δ: 0.83 (3H, d, J = 6.8 Hz), 1.20-1.67 (8H, m), 1.79-1.89 (1H, m), 3.69-3.76 (1H, m),
6
2
.73 (1H, s), 7.46 (1H, br), 8.09 (1H, br), 8.41 (1H, s), 9.22 (1H, d, J = 8.8 Hz). MS (ESI) m/z:
+
68 (M + H) .
5
.1.4. 6-Anilino-4-{[(1S,2R)-2-methylcyclohexyl]amino}nicotinamide (9)
2
7

In the vessel of a microwave reactor, 8 (30 mg, 0.112 mmol), aniline (31.3 mg, 0.336 mmol),
(dba) (5.1 mg, 5.6 μmol), 2-dicyclohexylphosphino-2',4',6'-triisopropyl-1,1'-biphenyl (2.7
Pd
2
3
mg, 5.7 μmol), and t-BuONa (16.2 mg, 0.169 mmol) were suspended in t-BuOH (0.75 mL). The
vessel was then sealed, and the mixture was reacted at 130 °C for 30 min under microwave
irradiation. After cooling to room temperature, the mixture was extracted with CHCl
3
/MeOH
, and
(
4/1), and washed with brine. The organic layer was separated, dried over MgSO
4
concentrated under reduced pressure. The residue was purified by NH silica gel column
1
chromatography (CHCl
NMR (400 MHz, DMSO-d
.87-1.94 (1H, m), 3.44-3.49 (1H, m), 5.96 (1H, s), 6.84-6.88 (1H, m), 6.99 (1H, br), 7.19-7.25
2H, m), 7.57-7.60 (2H, m), 7.74 (1H, br), 8.36 (1H, s), 8.86 (1H, s), 8.88 (1H, d, J = 8.2 Hz).
3
/MeOH = 100/0 to 90/10) to give the title compound (20 mg, 55%). H
6
) δ: 0.87 (3H, d, J = 7.0 Hz), 1.30-1.61 (7H, m), 1.67-1.73 (1H, m),
1
(
+
+
MS (ESI) m/z: 325 (M + H) . HRMS (M + H) calcd for C19
H
24
4
N O: 324.1950. Found: 325.2031.
5
.1.5. Ethyl 4-{[1-(tert-butoxycarbonyl)piperidin-4-yl]amino}-6-chloronicotinate (10)
The title compound was prepared from 5 and N-Boc-4-aminopiperidine in accordance with
1
the procedure for preparing 6 in 99% yield. H NMR (400 MHz, DMSO-d
6
) δ: 1.28-1.43 (14H,
m), 1.85-1.93 (2H, m), 2.88-3.06 (2H, m), 3.74-3.90 (3H, m), 4.29 (2H, q, J = 7.2 Hz), 6.99 (1H,
+
s), 8.05 (1H, d, J = 8.0 Hz), 8.54 (1H, s). MS (ESI) m/z: 384 (M + H) .
5
.1.6. 4-{[1-(tert-Butoxycarbonyl)piperidin-4-yl]amino}-6-chloronicotinic acid (11)
The title compound was prepared from 10 in accordance with the procedure for preparing 7
1
in 88% yield. H NMR (400 MHz, DMSO-d
6
) δ: 1.24-1.45 (11H, m), 1.86-1.94 (2H, m), 2.88-
2
8

3
.07 (2H, m), 3.72-3.90 (3H, m), 6.94 (1H, s), 8.31 (1H, d, J = 8.0 Hz), 8.51 (1H, s), 13.39 (1H,
+
br). MS (ESI) m/z: 356 (M + H) .
5
.1.7. tert-Butyl 4-[(5-carbamoyl-2-chloropyridin-4-yl)amino]piperidine-1-carboxylate (12)
The title compound was prepared from 11 in accordance with the procedure for preparing 8
1
in 90% yield. H NMR (400 MHz, DMSO-d
6
) δ: 1.21-1.32 (2H, m), 1.40 (9H, s), 1.81-1.92 (2H,
m), 2.88-3.08 (2H, m), 3.64-3.74 (1H, m), 3.76-3.86 (2H, m), 6.82 (1H, s), 7.48 (1H, br), 8.09
+
(
1H, br), 8.41 (1H, s), 8.86 (1H, d, J = 8.0 Hz). MS (ESI) m/z: 377 (M + Na) .
5
.1.8. tert-Butyl 4-[(2-anilino-5-carbamoylpyridin-4-yl)amino]piperidine-1-carboxylate
(
13a)
The title compound was prepared from 12 and aniline in accordance with the procedure for
1
preparing 9 in 48% yield. H NMR (400 MHz, DMSO-d
6
) δ: 1.22-1.37 (2H, m), 1.41 (9H, s),
1
6
8
.89-1.98 (2H, m), 2.91-3.10 (2H, m), 3.40-3.49 (1H, m), 3.76-3.87 (2H, m), 5.99 (1H, s), 6.84-
.89 (1H, m), 7.03 (1H, br), 7.21-7.26 (2H, m), 7.59-7.63 (2H, m), 7.74 (1H, br), 8.37 (1H, s),
+
.65 (1H, d, J = 7.2 Hz), 8.91 (1H, brs). MS (ESI) m/z: 412 (M + H) .
5
.1.9. tert-Butyl 4-{[5-carbamoyl-2-(pyridin-2-ylamino)pyridin-4-yl]amino}piperidine-1-
carboxylate (13b)
In the vessel of a microwave reactor, 12 (400 mg, 1.13 mmol), 2-aminopyridine (318 mg,
3
.38 mmol), Pd
biphenyl (26.9 mg, 56.4 μmol), and K
mL). The vessel was then sealed, and the mixture was reacted at 130 °C for 1 h under microwave
2
(dba)
3
(51.6 mg, 56.4 μmol), 2-dicyclohexylphosphino-2',4',6'-triisopropyl-1,1'-
2
CO (312 mg, 2.25 mmol) were suspended in t-BuOH (3
3
2
9