Synthesis and activity of N-cyanoguanidine-piperazine P2X7 antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.06.055

A novel series of cyanoguanidine-piperazine P2X₇ antagonists were identified and structure-activity relationship (SAR) studies described. Compounds were assayed for activity at human and rat P2X₇ receptors in addition to their ability to inhibit IL-1β release from stimulated human whole blood cultures. Compound 27 possesses potent activity (0.12 μM) in this latter assay and demonstrates moderate clearance in-vivo.

Full text of DOI:10.1016/j.bmcl.2008.06.055

Bioorganic & Medicinal Chemistry Letters 18 (2008) 3848–3851
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and activity of N-cyanoguanidine-piperazine P2X7 antagonists
,
*
Patrick Betschmann, Brian Bettencourt, Diana Donnelly-Roberts, Michael Friedman , Jonathan George,
Gavin Hirst ,à, Nathan Josephsohn, Donald Konopacki, Biqin Li, John Maull, Michael J. Morytko,
Nigel StJohn Moore, Marian Namovic, Paul Rafferty, Jose-Andres Salmeron-Garcia,
Edit Tarcsa, Lu Wang, Kevin Woller
Abbott Bioresearch Center, Department of Medicinal Chemistry, 381 Plantation Street, Worcester, MA 01605, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of cyanoguanidine-piperazine P2X7 antagonists were identified and structure–activity rela-
tionship (SAR) studies described. Compounds were assayed for activity at human and rat P2X7 receptors
in addition to their ability to inhibit IL-1b release from stimulated human whole blood cultures. Com-
Received 13 May 2008
Revised 16 June 2008
Accepted 17 June 2008
Available online 20 June 2008
pound 27 possesses potent activity (0.12
lM) in this latter assay and demonstrates moderate clearance
in-vivo.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
P2X7
Antagonists
Cyanoguanidine
Piperazine
The P2X7 receptor is an ATP-activated ligand-gated ion channel
that represents a promising new therapeutic target with potential
applications in the treatment of pain and inflammation.1–5 Expres-
sion of the P2X7 receptor is primarily on cells that regulate the im-
mune system such as mast cells, macrophages, and lymphocytes. A
clear role has been defined for ATP agonism of the P2X7 receptor
leading to caspase-1 activation, a protease required for the pro-
cessing and release of the pro-inflammatory cytokines IL-1b and
IL-18.1,5,6 Macrophages from P2X7 knock-out (KO) mice show an
impaired ability to release IL-1b in response to stimulation with
lipopolysaccaride (LPS) and ATP both in-vitro and in-vivo.7 Small
molecule P2X7 antagonists inhibit IL-1b and IL-18 release in-vitro
and/or in-vivo.8–12 A study using P2X7R knockout mice established
that joint inflammation was substantially decreased in the anti-
collagen Ab-induced mouse model of arthritis.13 The receptor is
also expressed in the central nervous system on microglia14 and
astrocytes15 and this suggests a role for the development and
maintenance of neuropathic pain as evidenced in P2X7 knock-out
mice that showed protection from inflammatory pain and partial
nerve ligation-induced neuropathic pain.16 These data are in agree-
ment with the efficacy profile of P2X7 antagonists in models of
chronic inflammatory and neuropathic pain.8–10,12,17
The role of the P2X7 receptor in the processing and release of
both IL-1b and IL-18 represents an attractive approach to inhibit-
ing levels of these cytokines to provide a potential treatment for
inflammatory conditions such as rheumatoid arthritis. We recently
published initial structure–activity relationships of a novel series
of cyanoguanidine-piperazine P2X7 antagonists exemplified by
the amides 1 and 2 with sub-100-nM potency against the human
P2X7 receptor.18 A general limitation of these amides was the
low potency observed in LPS/ATP stimulated human whole blood
cultures (data not shown) that precluded further advancement.
This letter describes significant further optimization of this novel
series of cyanoguanidine-piperazine P2X7 antagonists resulting
in urea-linked compounds with less than 250 nM potency in hu-
man whole blood.
O
OMe
N
H
N
N
Ar
Ph
OMe
N
NC
1, Ar = o-Tolyl (hP2X7 IC50 = 94 nM)
2, Ar = 5-Quinolinyl (hP2X7 IC50 = 59 nM)
* Corresponding author.
E-mail address: michael.friedman@abbott.com (M. Friedman).
These authors contributed equally to this work.
Present address: SGX Pharmaceuticals, 10505 Roselle Street, San Diego, CA 92121,
The amides in this letter were prepared using the general syn-
thetic methods shown in Scheme 1 and illustrated for the synthesis
of target molecule 1. Reaction of the o-tolyl isothiocyanate I with
à
USA.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.06.055

P. Betschmann et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3848–3851
3849
Table 2
Ph
Na+
S-
H
N
Cl
i
ii
NH
NCS
O
+
N
HN
N
N
NC
H
H
N
N
I
II
III
Ph
X
Ph
Ph
O
OMe
OMe
a
a
X
rP2X7IC50
(
lM)
hP2X7IC50 (lM)
NH
N
iii
H
N
H
N
N
N
9
10
11
CHNO2
NCONH2
O
3.11
10.3
0.62
0.31
6.4
0.78
N
N
NC
NC
IV
1
aValues are means of 2–3 experiments. Compounds tested at the recombinant
human and rat P2X7 receptors as described.9
Scheme 1. Reagents and conditions: (i) NaHNCN, DMF; (ii) EDC, DMF; (iii) EDC,
DMF, 3,4-dimethoxybenzoic acid.
if further improvements in activity could be attained. The results
of this exercise can be illustrated with just three examples (Table
2) as all analogs exhibited a reduction in potency compared to
the cyanoguanidine.
Having established that compound 5 represented our lead com-
pound, we sought to profile it more widely. To study the effect of
P2X7, antagonists on ICE-dependent cytokines, compound 5 was
tested for its ability to inhibit both IL-1b and IL-18 in LPS/ATP stim-
ulated human whole blood cultures. In accord with its mechanism
of action, compound 5 inhibited the secretion of both cytokines
with similar concentration–response relationships at an IC50 of
sodium cyanamide gave the thiocyanoimidate intermediate II. In-
situ coupling with the phenyl piperazine III constructed the cyano-
guanidine IV at the less sterically congested nitrogen. Standard
amide-coupling protocols provided the final product 1 by reaction
at the more hindered nitrogen. Alternatively, ureas such as 4 and 5
were accessed by reacting intermediates such as IV with the appro-
priate isocyanate. To prepare analogs shown in Table 4 with the R2
group proximal to the cyanoguanidine and distal to the amide or
urea (such as 13), the sequence of coupling reactions was reversed
making the cyanoguanidine as the last step.
In-vitro P2X7 activity was assessed using the recombinant rat
and human receptors. Antagonist potencies were determined by
measuring the inhibition of Ca2+ flux with a fluorometric imaging
plate reader (FLIPR) using Fluo-4 as the dye and 30-O-(4-ben-
zoylbenzoyl) ATP (BzATP) as the agonist.9 Selected compounds
were subsequently profiled for their ability to inhibit IL-1b release
from stimulated human whole blood cultures.
1.19 and 1.26 lM, respectively (data not shown).
Target validation studies required a compound with cellular
activity married with the ability to achieve good oral exposure.
Profiling of compound 5 in liver microsomal preparations (rat
and human) indicated a high rate of oxidative metabolism, an
observation reinforced by the high clearance and short half-life
of compound 5 after iv administration in rats shown in Table 3.
Combination of these data and the whole blood potency suggested
that analogs with an improved profile were needed.
Our initial focus was to identify groups that would impart
greater human whole blood potency by exploring further cyano-
guanidine substitutions (R1 in Table 4). In contrast to the data in
the recombinant human receptor, human whole blood data for
the inhibition of IL-1b indicate that the phenyl group and the iso-
propyl group are not interchangeable in the 2- or 3-position. For
example, in human whole blood, the 2-isopropyl is preferred over
the 3-isopropyl as seen for the quinolines 14 and 15with the for-
mer being fourfold more potent. In addition, there is a profound ef-
fect in this assay for the 3-phenyl group in indole 17 compared to
its 2-isopropyl analog 16 with the latter being greater than 30-fold
more potent. The reasons for this discrimination are currently un-
clear. One common feature of the three most potent compounds in
the whole blood assay (14, 23, and 24) is that each contains a basic
residue at R1.
We initially probed the requirement for R1 as shown in Table 1.
The tolylcyanoguanidine substitution was selected for these inves-
tigations based on the potency of compound 1. Although the sul-
fonamide 7 provided a moderate level of potency at hP2X7,
greater than 8-fold less activity was observed at the rat receptor.
Replacing the amide with a urea (4 and 5) provided a useful phar-
macophore as these had comparable potency to 1, both at hP2X7
and at rP2X7. Methylation of the urea provided compound 6 that
exhibited eroded potency compared to its progenitor 5. Analysis
of these data is consistent with a preference for a hydrogen-bond
donor in this region of the molecule.
With the promising activity of urea 5, additional exploration
was conducted on the cyanoguanidine replacements to ascertain
Table 1
R1
Our attention turned to the 5-quinolyl analog, compound 14, as
this represented a 10-fold increase in potency compared to com-
pound 5 in the key human whole blood assay. Unfortunately 14
suffered from microsomal instability (43% parent remaining after
25 min in rat liver microsomes) and high in-vivo clearance in rats
(3.2 l/h/kg). In mouse liver microsomes, the primary metabolic
route is oxidation of the quinoline to the N-oxide with no impact
on the cyanoguanidine. We were also able to ascertain that the
quinoline N-oxide was the major metabolite formed in-vivo.
N
H
N
N
Ph
N
NC
a
R1
rP2X7IC50
(l
M)
hP2X7aIC50
(lM)
1
3
4
5
6
7
8
–CO-(3,4-dimethoxyphenyl)
0.29
0.14
0.178
0.095
0.29
4.06
>10
0.094
0.63
0.025
0.132
1.03
–COCH2ꢀ(3,4-dimethoxyphenyl)
–CONH-(3,4-dimethoxyphenyl)
–CONH-(4-chlorophenyl)
–CONMe-(4-chlorophenyl)
–SO2-(4-chlorophenyl)
Table 3
0.54
>10
Rat pharmacokinetic parameters at 5 mg/kg iv for compound 5
–CH2ꢀ(4-chlorophenyl)
Vss (l/kg)
5.3
Clp (l/h/kg)
4.1
t1/2 (h)
aValues are means of 2–3 experiments. Compounds tested at the recombinant
human and rat P2X7 receptors as described.9.
0.9

3850
P. Betschmann et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3848–3851
Table 4
O
R2
3
2
N
NH
H
N
N
R1
N
R3
NC
a
a
R1
R2
R3
rP2X7IC50
(l
M)
hP2X7IC50
(
lM)
Whole Blooda
(lM)
5
12
13
14
15
16
17
18
19
20
21
22
23
24
o-Tolyl
o-Tolyl
o-Tolyl
5-Quinolinyl
5-Quinolinyl
4-Indolyl
4-Indolyl
4-Indazolyl
4-Benzofuranyl
2-(CH2NMe2)Ph
1-N-Methyl-4-benzimidazoyl
3-Pyridyl
5-Tetrahydro quinolinyl
5-Tetrahydro iso-quinolinyl
3-Ph
2-iPr
3-iPr
2-iPr
3-iPr
2-iPr
3-Ph
2-iPr
2-iPr
3-Ph
3-Ph
3-Ph
2-iPr
2-iPr
p-Cl
p-Cl
p-Cl
p-Cl
p-Cl
p-Cl
p-Cl
m-F
m-F
p-Cl
p-Cl
p-Cl
m-F
p-Cl
0.095
0.125
0.11
0.057
0.093
0.11
0.056
NT
0.132
0.06
0.112
0.04
0.067
0.045
0.13
0.25
24.9
0.48
1.66
1.33
0.06
1.19
NT
0.98
0.121
0.88
0.47
12.2
1.59
NT
NT
NT
NT
0.46
0.21
NT
1.07
5.19
2.04
NT
0.18
0.11
aValues are means of 2–3 experiments. NT, not tested. Compounds tested at the recombinant human receptors as described.9
Table 5
and a half-life in excess of 4 h in rats. Unfortunately compound
27 had a low oral bioavailability of 5% in rats, a finding that is likely
due to solubility limited absorption as it exhibited a low aqueous
Cl
O
solubility of 19 lg/mL. This is further supported by the moderate
N
N
H
H
N
clearance value that reduces the likelihood of a major first-pass ef-
fect leading to low systemic concentrations. Compound 27 was
found to be selective for P2X7 over other P2 receptors, as evidenced
N
N
N
R1
NC
by the lack of activity at P2X2/3, P2X3, P2X4, and P2Y2 at 10 lM.
In summary, a novel series of cyanoguanidine-piperazines were
discovered with potent activity at P2X7 and in a human whole
blood assay measuring the inhibition of the pro-inflammatory
cytokine IL-1b. It was found from these studies that, with suitable
structural modifications, compounds with increased whole blood
potency and moderate in-vivo clearance could be identified. Most
analogs displayed approximately 3-5-fold greater potency for
hP2X7 over rP2X7; however, compound 27 was highly potent at
both species. These findings indicate that substantial flexibility
around the pharmacophore of 1 and 27 exists to incorporate struc-
tural changes with retention of potent P2X7 antagonism. Addi-
tional studies describing structural modifications around 27 to
improve the solubility and oral bioavailability in this series will
be the subject of future reports.
R1
Whole blooda
(
lM)
Microsomal stabilityb
(% parent remaining)
14
25
26
27
28
H
2-Cl
2-F
2-Me
8-Me
0.121
0.87
2.64
0.12
3.6
43
NT
60
74
64
NT, not tested.
a
Values are means of 2–3 experiments.
Rat microsomes after 1 h incubation.
b
This route of metabolism prompted an effort to block this path-
way as shown in Table 5. Incorporation of flanking groups at either
the 2- or 8-position of the quinoline provided compounds with an
increased stability in microsomal preparations. All but one of these
compounds suffered from a decrease in potency in human whole
blood (4-30-fold) compared to quinoline 14. In contrast to its 8-
methyl counterpart 28, the regioisomeric 2-methyl analog 27
maintained potency in this assay and was also equally potent for
References and notes
1. DiVirgilio, F.; Vishwanath, V.; Ferrari, D. Handb. Exp. Pharmacol. 2001, 151, 355.
2. Baraldi, P. G.; DiVirgilio, F.; Romagnoli, R. Curr. Top. Med. Chem. 2004, 4, 1707.
3. Romagnoli, R.; Baraldi, P. G.; DiVirgilio, F. Expert Opin. Ther. Patents 2005, 15,
271.
4. Gunosewoyo, H.; Coster, M. J.; Kassiou, M. Curr. Med. Chem. 2007, 14, 1505.
5. Donnelly-Roberts, D. L.; Jarvis, M. F. Br. J. Pharmacol. 2007, 151, 571.
6. Mehta, V. B.; Hart, J.; Wewers, M. D. J. Biol. Chem. 2001, 276, 3820.
7. Solle, M.; Labasi, J.; Perregaux, D. G.; Stam, E.; Petrushova, N.; Koller, B. H.;
Griffiths, R. J.; Gabel, C. A. J. Biol. Chem. 2001, 276, 125.
8. Nelson, D. W.; Gregg, R. J.; Kort, M. E.; Perez-Medrano, A.; Voight, E. A.; Wang,
Y.; Grayson, G.; Namovic, M. T.; Donnelly-Roberts, D. L.; Niforatos, W.; Honore,
P.; Jarvis, M. F.; Faltynek, C. R.; Carroll, W. A. J. Med. Chem. 2006, 49, 3659.
9. Honore, P. M.; Donnelly-Roberts, D.; Namovic, M.; Hsieh, G.; Zhu, C.;
Mikusa, J.; Hernandez, G.; Zhong, C.; Gauvin, D.; Chandran, P.; Harris, R.;
Perez-Medrano, A.; Carroll, W.; Marsh, K.; Sullivan, J.; Faltynek, C.; Jarvis,
M. F. J. Pharmacol. Exp. Ther. 2006, 319, 1376.
10. Carroll, W. A.; Kalvin, D. M.; Perez Medrano, A.; Florjancic, A. S.; Wang, Y.;
Donnelly-Roberts, D. L.; Namovic, M. T.; Grayson, G.; Honore, P.; Jarvis, M. F.
Bioorg. Med. Chem. Lett. 2007, 17, 4044.
both the rat P2X7 (0.151 lM) and the human P2X7 (0.093 lM). It
is also worthwhile noting that both enantiomers of the 2-methyl
quinoline 27 were profiled in both rat and human P2X7 receptor
assays and were indistinguishable from the racemate (data not
shown). In accord with its increased microsomal stability,
quinoline 27 exhibited improved intravenous pharmacokinetic
parameters to compound 14 (Table 6), with reduced clearance
Table 6
Rat pharmacokinetic parameters at 5 mg/kg iv for compound 27
Vss (l/kg)
11.8
Clp (l/h/kg)
1.4
t1/2 (h)
11. Furber, M.; Alcaraz, L.; Bent, J. E.; Beyerbach, A.; Bowers, K.; Braddock, M.;
Caffrey, M. V.; Cladingboel, D.; Collington, J.; Donald, D. K.; Fagura, M.; Ince, F.;
4.4

P. Betschmann et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3848–3851
3851
Kinchin, E. C.; Laurent, C.; Lawson, M.; Luker, T. J.; Mortimore, M. M. P.; Pimm,
A. D.; Riley, R. J.; Roberts, N.; Robertson, M.; Theaker, J.; Thorne, P. V.; Weaver,
R.; Webborn, P.; Willis, P. J. Med. Chem. 2007, 50, 5882.
15. Irnich, D.; Burgstahler, R.; Grafe, P. Drug Dev. Res. 2001, 52, 83.
16. Chessell, I. P.; Hatcher, J. P.; Bountra, C.; Michel, A. D.; Hughes, J. P.; Green, P.;
Egerton, J.; Murfin, M.; Richardson, J.; Peck, W. L.; Grahames, C. B. A.; Casula, M.
A.; Yiangou, Y.; Birch, R.; Anand, P.; Buell, G. N. Pain 2005, 114, 386.
17. Perez-Medrano, A.; Donnelly-Roberts, D. L.; Honore, P.; Hsieh, G.; Namovic, M.
T.; Peddi, S.; Wang, Y.; Faltynek, C. R.; Jarvis, M. F.; Carroll, W. A. J. Med. Chem.,
accepted for publication.
18. Morytko, M. J.; Betschmann, P.; Woller, K.; Ericsson, A.; Chen, H.; Donnelly-
Roberts, D. L.; Namovic, M. T.; Jarvis, M. J.; Carroll, W. A.; Rafferty, P. Biorg Med.
Chem. Lett. 2008, 18, 2093.
12. Nelson, D. W.; Sarris, K.; Kalvin, D. M.; Namovic, M. T.; Grayson, G.; Donnelly-
Roberts, D. L.; Harris, R.; Honore, P.; Jarvis, M. F.; Faltynek, C. R.; Carroll, W. A. J.
Med. Chem., accepted for publication.
13. Labasi, J. M.; Petrushova, N.; Donovan, C.; McCurdy, S.; Lira, P.; Payette, M. M.;
Brissette, W.; Wicks, J. R.; Audoly, L.; Gabel, C. A. J. Immunol. 2002, 168, 6436.
14. Ferrari, D.; Villalba, M.; Chiozzi, P.; Falzoni, S.; Ricciarrdi-Castagnoli, P.;
DiVirgilio, F. J. Immunol. 1996, 156, 1531.