The discovery and preclinical characterization of 6-chloro-N-(2-(4,4-difluoropiperidin-1-yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-carboxamide based P2X7 antagonists

Article abstract of DOI:10.1016/j.bmcl.2016.08.029

The synthesis, SAR and preclinical characterization of a series of 6-chloro-N-(2-(4,4-difluoropiperidin-1-yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-carboxamide based P2X7 antagonists is described herein. The lead compounds are potent inhibitors in Ca²⁺flux and whole blood IL-1β P2X7 release assays at both human and mouse isoforms. Compound 1e showed a robust reduction of IL-1β release in a mouse ex vivo model with a 50?mg/kg oral dose. Evaluation of compound 1e in the mouse SNI tactile allodynia, carrageenan-induced paw edema or CIA models resulted in no analgesic or anti-inflammatory effects.

Full text of DOI:10.1016/j.bmcl.2016.08.029

Accepted Manuscript
The discovery and preclinical characterization of 6-chloro-N-(2-(4,4-difluoro-
piperidin-1-yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-car-
boxamide based P2X7 antagonists
Jason C. Rech, Anindya Bhattacharya, Bryan J. Branstetter, Christopher J. Love,
Joseph E. Leenaerts, Ludwig P. Cooymans, William A. Eckert, Hong Ao, Qi
Wang, Sandra R. Chaplan, Alan D. Wickenden, Alec D. Lebsack, J. Guy
Breitenbucher
PII:
S0960-894X(16)30850-2
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.08.029
BMCL 24155
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
21 July 2016
9 August 2016
10 August 2016
Please cite this article as: Rech, J.C., Bhattacharya, A., Branstetter, B.J., Love, C.J., Leenaerts, J.E., Cooymans,
L.P., Eckert, W.A., Ao, H., Wang, Q., Chaplan, S.R., Wickenden, A.D., Lebsack, A.D., Guy Breitenbucher, J., The
discovery
and
preclinical
characterization
of
6-chloro-N-(2-(4,4-difluoropiperidin-1-yl)-2-(2-
(
trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-carboxamide based P2X7 antagonists, Bioorganic & Medicinal
Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.08.029
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.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.els evier.com
The discovery and preclinical characterization of 6-chloro-N-(2-(4,4-
difluoropiperidin-1-yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-
carboxamide based P2X7 antagonists
a∗
Jason C. Rech , Anindya Bhattacharya , Bryan J. Branstetter , Christopher J. Love , Joseph E. Leenaerts ,
a
a†
b
b
b
Ludwig P. Cooymans ,William A. Eckert , Hong Ao , Qi Wang , Sandra R. Chaplan , Alan D. Wickenden ,
a
a
a
a
a
a
Alec D. Lebsack , J. Guy Breitenbucher
a†
a
Janssen Pharmaceutical Research & Development L.L.C., 3210 Merryfield Row, San Diego, CA 92121, United States
Janssen Pharmaceutical Research & Development L.L.C., Turnhoutseweg 30, Beerse, Belgium
b
ARTICLE INFO
ABSTRACT
Article history:
Received
The synthesis, SAR and preclinical characterization of a series of 6-chloro-N-(2-(4,4-
difluoropiperidin-1-yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethyl)quinoline-5-carboxamide
2
+
Revised
based P2X7 antagonists is described herein. The lead compounds are potent inhibitors in Ca
Accepted
flux and whole blood IL-1β P2X7 release assays at both human and mouse isoforms.
Compound 1e showed a robust reduction of IL-1β release in a mouse ex vivo model with a 50
mg/kg oral dose. Evaluation of compound 1e in the mouse SNI tactile allodynia, carrageenan-
induced paw edema or CIA models resulted in no analgesic or anti-inflammatory effects.
Available online
Keywords:
P2X7
2
009 Elsevier Ltd. All rights reserved.
IL-1β
Pain
Inflammation
The P2X7 receptor is an ATP-gated non-selective ion channel
belonging to the P2X superfamily (P2X1-7) of purinoreceptors.
It is abundantly expressed on monocytes, macrophages and
lymphocytes in the periphery and on astrocytes and microglia in
the CNS. In vitro activation of the P2X7 receptor by elevated
levels of ATP, or benzoylbenzoyl-ATP (BzATP), results in
cation flux and the release of proinflammatory cytokines
interleukin-1β (IL-1β) and interleukin-18 (IL-18) in cells primed
robust target engagement and were not advanced into further
6,7
development.
outcomes has been reviewed elsewhere.
A detailed review of P2X7 clinical candidate
10
1
,2
with lipopolysaccharide (LPS).
The P2X7 receptor has been implicated in a number of disease
states including chronic pain, inflammation, rheumatoid arthritis,
Crohn’s disease, mood disorders, multiple sclerosis, Alzheimer’s,
3
and Huntington’s disease. As a result many groups in both
industry and academia have dedicated significant research efforts
toward the development of P2X7 antagonists. Most of the early
work focused on the discovery of P2X7 antagonists for treating
diseases related to pain and/or inflammation. These efforts
4
,5
6
resulted in at least four compounds, GSK1482160 , AZD9056 ,
7
,8
9
CE-224535 , and EVT 401 (structure not disclosed), that
progressed into clinical trials (Figure 1). To our knowledge these
compounds showed no or limited efficacy in the clinic despite
Figure 1: P2X7 clinical antagonists
—
——
Corresponding author.
∗
E-mail address: jrech2@its.jnj.com (J.C. Rech)
present address: Dart NeuroScience, L.L.C., 12278 Scripps Summit Drive, San Diego, CA 92131
†

Despite recent setbacks in the clinic, efforts to unlock the
therapeutic potential of the P2X7 receptor are ongoing with new
antagonists continuing to appear in the literature. Described
herein is the discovery and profiling of a series of peripherally
restricted P2X7 antagonists with potency at both human and
mouse isoforms and robust in vitro and ex vivo reduction of IL-
1
β release. A general structure of our antagonist series is shown
in Figure 2. Rodent P2X7 pharmacology was an important
criterion for our program as at least two compounds that failed in
the clinic had weak rodent activity and as such were most likely
6
,7
never evaluated in rodent models of pain or inflammation.
The series of P2X7 antagonists described in the following text
was identified through an HTS screen followed by a brief
medicinal chemistry effort. For convenience these compounds
were generated and tested as racemic mixtures as the identity of
the stereocenter in Figure 2 has no bearing on the physical
properties or potency. Our work on this series led us to explore
all possible points of diversity. The most fruitful changes were
shown to occur at the 2-position of the quinoline moiety (Figure
2
). Compounds generated as part this effort are the focus of this
article. A complete list of the compounds generated in the
course of our work on this series can be found in our recent
11
disclosure.
Figure 2: P2X7 antagonist general structure
The 2-position of the quinoline moiety was shown to be
highly flexible allowing for the incorporation of a variety of
functional groups (Figure 2). Appending small alkyl, alkoxy,
ether, or amine moieties via an amine linkage ultimately provided
the most favorable combination of potency and drug-like
properties. A subset of compounds generated for this series is
a
+2
IC50s were determined using a FLIPR Ca flux assay. The values are the
average of at least three experiments, unless otherwise noted. h denotes
human and m denotes mouse.
2+
shown in Table 1 along with their potency in a P2X7 Ca flux
assay.
Compounds with (R)-pyrrolidin-3-ol (1e) and morpholine
1h) on the 2-position of the quinoline moiety provided the most
The compounds detailed in Table 1 were prepared using the
reaction sequences shown in Schemes 1-3. The synthesis of
these antagonists was highly convergent, scalable and allowed
for diversification in the ultimate transformation of the synthesis.
The requisite 2,6-dichloroquinoline-5-carboxylic acid (4, Scheme
(
favorable combination of potency at both the human and mouse
isoforms of the P2X7 receptor. Interestingly, (S)-pyrroldin-3-ol
1f showed a 4 fold reduction in potency when compared to 1e,
indicating the absolute stereochemistry of the pyrrolidin-3-ol
influences antagonist potency. Cyclopropylamine 1c was also
1
dichloroquinoline (2) with AlCl and bromine to provide 2,6-
) was prepared by the treatment of commercially available 2,6-
3
2+
shown to inhibit Ca flux at both human and mouse isoforms.
The 2-aminoquinoline 1a was a highly potent antagonist of
human P2X7 but was significantly less potent at the mouse
receptor. Aside from (R)-pyrrolidin-3-ol 1e, incorporating other
aminoalcohols (1d and 1g) and an N-methylpiperazine, 1i,
resulted in antagonists with modest potency.
dichloro-5-bromoquinoline (3). Treatment of resulting 2,6-
dichloro-5-bromoquinoline (3) with isopropyl magnesium
chloride followed by carbon dioxide completed the synthesis of
1
2
2
,6-dichloroquinoline-5-carboxylic acid 4.
Table 1: IC50 data for compounds 1a-i
3 2
Scheme 1: Synthesis of compounds 1a-1i: a. AlCl , Br , 83%; b. iPrMgCl,
CO , THF, 0 °C ꢀ rt, 74%.
2

Synthesis
trifluoromethyl)pyrimidin-5-yl)ethanamine (8) commenced from
of
2-(4,4-difluoropiperidin-1-yl)-2-(2-
flux which can be attributed to the inclusion of plasma protein
inherent in the undiluted whole blood. Compound 1e was most
effective at attenuating the release of IL-1β in both human and
mouse isoforms (Table 2). Both compounds 1c and 1h were
shown to have moderate human and mouse whole blood
potencies.
(
commercially available methyl-2-(trifluoromethyl)pyrimidine-5-
carboxylate (5) which was reduced to the aldehyde (6) by
treatment with DIBAL. The aldehyde was subjected to Strecker
reaction conditions to provide nitrile 7 in good yield. The
reduction of nitrile 7 to the requisite 2-(4,4-difluoropiperidin-1-
yl)-2-(2-(trifluoromethyl)pyrimidin-5-yl)ethanamine (8) was
accomplished in good yield by treatment with Raney Nickel
‡
Table 2: P2X7 whole blood assay data
2
under an H atmosphere.
MeO2C
OHC
N
a.
N
b.
N
CF3
N
N
CF3
5
6
CF3
N
CF3
†
The values are the mean of at least two experiments, unless otherwise
NC
N
N
noted. * Single data point. ‡ Mouse blood obtained for male C57/bl6 mice
from Charles River Laboratories
H2N
c.
N
N
The reasonable mouse whole blood potency of compound 1e
prompted additional profiling to determine its utility as a tool
compound to explore mouse ex vivo pharmacodynamic and in
vivo efficacy models. A mouse tissue distribution study showed
compound 1e is orally bioavailable and restricted to the periphery
F
F
F
F
7
8
Scheme 2: Synthesis of compounds 1a-1i: a. DIBAL, -78 °C, DCM; b. 4,4-
difluoropiperidine hydrochloride, Me SiCN, acetic acid, 45% over 2 steps; c.
Raney Nickel, 7N NH in methanol, H (45 psi), MeOH, 98%.
3
3
2
(
Table 3). A 50 mg/kg oral dose resulted in high plasma
concentrations (6415 ng/mL) and negligible compound
concentrations (21 ng/mL) in the brain. The high plasma
concentrations and lack of CNS exposure upon oral dosing
coupled with its reasonable mouse whole blood potency
suggested compound 1e could serve as an adequate tool
compound to explore the role of peripherally localized P2X7
receptors in pain and inflammation.
Coupling of the 2,6-dichloroquinoline-5-carboxylic acid
and 2-(4,4-difluoropiperidin-1-yl)-2-(2-
trifluoromethyl)pyrimidin-5-yl)ethanamine (8) occurred in the
of (benzotriazol-1-yloxy)tris(dimethylamino)-
phosphonium hexafluorophosphate (BOP) to provide compound
. The completion of the synthesis was accomplished through a
(
(
4)
presence
9
SNAr on the 2-chloro moiety under standard SNAr conditions
employing the desired amine.
Table 3: Mouse tissue distribution for 1e
Plasmab
(ng/mL)
Brainb
(ng/mL)
Compound
PO Dose
50.0 mg/kga
6416
21
1
e
a
b
vehicle = 30% SBECD 5 mL/kg, p.o., n = 5, data collected at 4 hours
Compound 1e was taken into a mouse P2X7 ex vivo IL-1β
reduction assay where it was shown to attenuate the release of IL-
1
β dose dependently (5, 15 and 50 mg/kg, oral dose) with a
robust reduction observed at the 50 mg/kg dose (Figure 2). Two
hours post dose blood was collected and IL-1β release was
measured in response to LPS/BzATP treatment. In addition to
the robust attenuation of IL-1β at the 50 mg/kg dose, a modest
reduction was also observed at the 15 mg/kg dose. No significant
reduction in IL-1β was observed at the 2 mg/kg dose.
Figure 2: Mouse ex vivo IL-1β Reduction
Scheme 3: Synthesis of Compounds 1a-1i: a. BOP, NEt
amine, NEt , MeCN, 90 °C, 15-95%.
3
, DCM, 85%; b.
3
2
+
Compounds showing the most favorable potency in the Ca
flux assay were further profiled in mouse and human P2X7
whole blood assays. The whole blood assays are conducted in
undiluted whole blood to best mimic an in vivo environment and
measure the attenuation of P2X7 mediated IL-1β release initiated
1
3
by LPS challenge and subsequent stimulation by BzATP. The
whole blood assay results are shown in Table 2.
A rightward shift in potency was observed for all compounds
2
tested in the whole blood P2X7 assays with respect to the Ca
+

Figure 4: Duration of action for compound 1e
Male C57/bl6 mice obtained from Charles River Laboratories were
employed in this study. vehicle = 30% SBECD 5 mL/kg, p.o., n = 8, data
collected at 2 hours post dose
The robust reduction in IL-1β observed in the mouse ex vivo
model at the 2 hour time point with a 50 mg/kg dose prompted
the progression of compound 1e to a mouse spared nerve injury
Male C57/bl6 mice obtained from Charles River Laboratories were employed
in this study. *BLQ = below detectable levels
1
4
(
SNI) model of tactile allodynia (Figure 3). The time frame of
Compound 1e was evaluated in the mouse CIA at a 50 mg/kg
b.i.d. oral dose. The results of this study are shown in Figure 5.
Animals treated with collagen and CFA for 21 days developed a
significant and pronounced arthritic like condition which was
reversed with dexamethasone and methotrexate. Administration
of compound 1e commenced on day 21 and continued through
day 34. Arthritic scores were measured on day 35. The arthritic
scores of animals treated with compound 1e did not differentiate
from those of vehicle.
the SNI model was designed to correspond to the ex vivo IL-1β
reduction model to ensure target coverage for a majority of the
experimental paradigm.
Paw withdrawal thresholds were
measured before and after induction of nerve injury and again
after compound 1e was administered orally. Measurements were
recorded every hour for 4 hours and at no point were significant
analgesic effects observed. Similarly compound 1e produced no
anti-inflammatory or analgesic effects in the carrageenan-induced
model of paw edema at the same oral dose of 50 mg/kg (data not
shown).
*
Figure 5: Mouse collagen-induced arthritis model
Figure 3: Mouse SNI tactile allodynia
One-Way ANOVA overall, ****p<0.0001; Bonferonni post-tests: * = p
vehicle = 30% SBECD 5 mL/kg, p.o., n = 8; 1e 50 mg/kg, p.o, n = 8.
Male C57/bl6 mice obtained from Charles River Laboratories were employed
in this study.
<
0.05, *** = p<0.001, ****p<0.0001; naïve n = 5, veh. n = 11, Dex n = 11,
MTX n = 12, 2e n = 12; dosing vehicle is 30 % SBE 5 mL/kg, p.o. b.i.d.; Dex
dexamethasone 0.3 mg/kg, p.o. q.d.; MTX = methotrexate 30 mg/kg, i.p. 2x
=
per week; 2e 50 mg/kg, p.o. b.i.d. *Female DBA mice obtained from Charles
River laboratories were employed in this study.
In addition the robust attenuation of IL-1β observed at the 2
hour time point, a 50 mg/kg oral dose of compound 1e was also
shown to sustain IL-1β reduction over an 8 hour period (Figure
We have identified a novel series of P2X7 antagonists with
potency in both calcium flux and whole blood assays at both
human and mouse isoforms. The lead compound of this series,
4
). At the 24 hour time point compound 1e was no longer
detectable and hence no attenuation of IL-1β release was
observed. Therefore b.i.d. dosing of compound 1e was
1
e, is a good in vivo tool compound to address the role of
employed to maintain sustained levels of IL-1β reduction over 24
hours making compound 1e amenable to evaluation in the mouse
peripheral P2X7 in animal models of inflammation and pain; as
such it was shown to significantly attenuate the release of IL-1β
with a 50 mg/kg dose for up to 8 hours. Surprisingly, even
though 1e suppressed IL-1β release, it was ineffective in
providing an analgesic or anti-inflammatory response in the
mouse SNI model and in the carrageena-induced model of paw
edema after a single oral 50 mg/kg dose. In addition, compound
1
5
collagen-induced arthritis (CIA) model.
We postulated the
sustained attenuation of IL-1β release under disease state
conditions may be required to see the therapeutic effect of a
P2X7 antagonist.
1e did not reduce the arthritic scores in the mouse CIA model
after 14 day continuous 50 mg/kg b.i.d. dosing. Combined these

results indicate that blockade of peripheral P2X7 mediated IL-1β
release is not sufficient for efficacy in the animal models of pain
and inflammation.
References:
1
2.
.
North RA.; Jarvis MF. Mol Pharmacol 2013, 67, 759-69.
Volonte C.; Apolloni S.: Skaper SD.; Burnstock G. CNS
Neurol Disord Drug Targets 2012, 11, 705-21.
3.
Romagnoli, R.; Baraldi, PG.; Cruz-Lopez, O.; Lopez-
Cara, C.; Preti, D.;Borea PA; Gessi, S. Expert Opin.
Ther. Targets 2008, 12, 647-61.
4.
Ali, Z.; Laurijssens, B.; Ostenfeld, T.; McHugh, S.;
Stylianou, A.; Scott-Stevens, P.; Hosking, L.; Dewit, O.;
Richardson, JC.; Chen, C. Br. J. Clin. Pharmacol. 2013,
7
5, 197-207.
5.
Abdi, MH.; Beswick, PJ.; Billington, A.; Chambers, LJ.;
Charlton, SD.; Collins, SD; Collis, KL.; Dean, DK.;
Fonfria, E.; Gleave, RJ.; Lejeune, CL.; Livermore, DG.;
Medhurst, SJ.; Michel, AD.; Moses, AP.; Page, L.; Patel,
S.; Roman, SA.; Senger, S.; Slingsby, B.; Steadman, JG.;
Stevens AJ.; Walter, DS. Bioorg. Med. Chem. Lett. 2010,
2
0, 5080-84.
6
.
Keystone, EC.; Wang, MM.; Layton M.; Hollis, S.;
McInnes IB. Ann. Rheum. Dis. 2012, 71, 1630-35.
Eser, A.; Colombel, JF.; Rutgeerts, P.; Vermeire, S.;
Vogelsang H.; Braddock, M.; Persson, T.; Reinisch, W.
Inflamm. Bowel Dis. 2015, 21, 2247-53.
7.
8.
Stock, TC.; Bloom, BJ.; Wei, N.; Ishaq, S.; Park, W.;
Wang, X.; Gupta, P.; Mebus, CA. J. Rheumatol. 2012,
39, 720-27.
Evotec.
9.
Evotec
development
partnerships,
http://evotec.com/article/en/Alliances-Projects/EVT401-
P2X7-Antagonist/67/4?selected_category_id=6.
1
0. Rech, Jason C; Bhattacharya, Anindya; Letavic, Michael
A; Savall, Brad, M. Bioorg. Med. Chem. Lett. 2016, 26,
838-3845.
3
1
1. Love, Christopher J; Leenaerts, Joseph E; Cooymans,
Ludwic P; Lebsack, Alec D; Branstetter, Bryan J;
Gleason, Elizabeth A; Venable, Jenifer D; Weiner,
Danielle; Smith, Deborah M; Breitenbucher James G.
PCT Int. Appl. 2009 WO 2009132000 A1.
12. Ford, Rhonan; Thompson, Toby; Willis, Paul, PCT Int.
Appl. 2005 WO 2005009968 A1.
1
3. Lord, B; Alusio L; Shoblock, J R; Neff, R A; Varlinskaya
E I; Ceusters, M; Lovenberg T W; Carruthers N;
Bonaventure, P; Letavic M A; Deak T; Wilhelmus D;
Bhattacharya A. J Pharmacol Exp Ther 2014, 351, 628-
6
41.
1
4. Decosterd, I and Woolf C J. Pain, 2000, 87, 149-158.
5. Courtenay JS, Dallman MJ, Dayan AD, Martin A,
Mosedale B. Nature, 1980, 283, 666-668.
1

Graphical Abstract
To create your abstract, type over the instructions in the
template box below.
Type the title of your article here
Leave this area blank for abstract info.
Jason C. Rech, Anindya Bhattacharya, Bryan J. Branstetter, Christopher J. Love, Joseph E. Leenaerts, Ludwig
P. Cooymans,William A. Eckert, Hong Aa, Qi Wang, Sandra R. Chaplan, Alan D. Wickenden, Alec D.
Lebsack, J. Guy Breitenbucher
Fonts or abstract dimensions should not be changed or altered.