Novel P2X7 receptor antagonists.

Article abstract of DOI:10.1016/j.bmcl.2003.08.033

The synthesis and pharmacological evaluation of a new series of potent P2X(7) receptor antagonists is disclosed. The compounds inhibit BzATP-mediated pore formation in THP-1 cells. The distribution of the P2X(7) receptor in inflammatory cells, most notabl

Full text of DOI:10.1016/j.bmcl.2003.08.033

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4043–4046
Novel P2X7 Receptor Antagonists
L. Alcaraz,a,* A. Baxter,a J. Bent,a K. Bowers,b M. Braddock,b D. Cladingboel,a
D. Donald,a M. Fagura,b M. Furber,a C. Laurent,a M. Lawson,b M. Mortimore,a
M. McCormick,b N. Robertsa and M. Robertsonb
aDepartment of Medicinal Chemistry, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
bDepartment of Discovery Biosciences, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
Received 17 July 2003; revised 18 August 2003; accepted 19 August 2003
Abstract—The synthesis and pharmacological evaluation of a new series of potent P2X7 receptor antagonists is disclosed. The
compounds inhibit BzATP-mediated pore formation in THP-1 cells. The distribution of the P2X7 receptor in inflammatory cells,
most notably the macrophage, mast cell and lymphocyte, suggests that P2X7 antagonists have a significant role to play in the
treatment of inflammatory disease.
# 2003 Elsevier Ltd. All rights reserved.
The P2X7 receptor is a ligand-gated ion channel acti-
vated by adenosine 50-triphosphate (ATP). Activation
leads to a rapid non-selective influx of cations and, with
prolonged exposure to ATP, the opening of large
plasma membrane pores.1 The P2X7 receptor is found
on a wide range of inflammatory and immunomodula-
tory cells, most notably the monocyte/macrophage line-
age, mast cells and lymphocytes.2 Its most notable
biological function is the processing and release of
interleukin-1b (IL-1b) from monocytes, macrophages
and microglial cells, and as such it is implicated in a
number of autoimmune and inflammatory diseases.3
Mice genetically modified to have the P2X7 receptor
‘knocked-out’ have been reported to show a reduced
severity of arthritis in an anti-collagen antibody arthritis
model.4 An antagonist of the P2X7 receptor could
therefore provide a new treatment for inflammatory
diseases.
analogue (benzoylbenzoyl adenosine triphospate-
BzATP). This characteristic of P2X7 receptor activation
was measured by the entry of a fluorescent DNA probe
(ethidium bromide) through the membrane pores with a
consequent increase in total cellular fluorescence.5 The
agonist potency order, BzATP>ATPꢀ2-methyl-
thioATP was also characteristic of the P2X7 receptor. A
HTS was undertaken using this assay in 96-well plate
format.6 Estimates of potency of the compounds were
made initially from pIC50 determinations and then more
robustly from pA2 estimates.
Though the initial hits were weakly active (e.g., 1,7
pA2=5.0), a clear SAR was observed and it soon became
evident that 4-pyridyl was the preferred left-hand side
heterocycle and that a 30-substituent on the biphenyl
portion was beneficial for activity (e.g., 2, pA2=6.9).
Our initial lead compounds came through high-
throughput screening of our company compound collec-
tion against a functional readout of the P2X7 receptor
expressed endogenously on the human pre-monocytic
cell line THP-1. The biological screen for the assessment
of P2X7 receptor inhibition was the reduction of plasma
membrane pore formation induced by a synthetic ATP
Very few examples of P2X7 receptor antagonists have
been described in the literature.8 Herein, we report further
structure–activity relationship studies carried out around
this novel series of antagonist 2 in which replacements for
*Corresponding author. Tel.: +44-1509-644542; e-mail: lilian.
alcaraz@astrazeneca.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.033

4044
L. Alcaraz et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4043–4046
the succinimide and the 4-pyridyl moieties have been
investigated.
The synthesis of simple acyclic analogues was carried
out as outlined in Scheme 1. Intermediate alcohol 4 was
obtained by treatment of epoxyether 3 with lithiated
4-methyl pyridine.
Scheme 1. Reagents and conditions: (a) 4-methylpyrine, n-BuLi,
ꢁ78 ꢂC to rt (64%); (b) phthalimide, Ph3P, DEAD, THF (98%);
(c) 40% aq MeNH2, MeOH, 90 ꢂC; (d) Dess–Martin periodinane,
DCM, rt; (e) MeNH2, NaBH(OAc)3, THF, AcOH (60%); (f) (3-
Scheme 2. Reagents and conditions: (a) (3-nitrophenyl)boronic acid,
Pd(Ph3P)4, N aCO3, PhMe/EtOH, 80 ꢂC (64%); (b) Ph3P, DEAD,
2
THF (29–65%); (c) 4-chlorobutanoyl chloride, NEt3, DCM, rt; then
t-BuOK, THF, rt (44%); (d) 5-chlorovaleryl chloride, NEt3, DCM,
rt; then t-BuOK, THF, rt (56%); (e) 3-chloropropane sulfo-
nylchloride, NEt3, DCM; then EtONa, EtOH, Á; (f) 2-chloroethyl
chloroformate, MeCN, rt, then NaH, THF/DMF, 70 ꢂC (73%); (g)
phosgene, PhMe, rt; 3-chloropropanol, rt; then NaH, DMF, 70 ꢂC
(33%).
nitrophenyl)boronic acid, Pd(Ph3P)4, N aCO3, EtOH, 80 ꢂC (5a: 72%,
2
5b: 91%); (g) acetic anhydride, rt (6a: 94%, 6d: 83%); (h) methyl
chloroformate, DCM, rt (6b: 72%, 6c: 66%).
A Mitsonobu reaction using phthalimide then depro-
tection of the primary amine or, oxidation of the sec-
ondary alcohol to the corresponding ketone and
reductive amination using methylamine, was followed
by a Suzuki coupling with (3-nitrophenyl)boronic acid
to yield amines 5a and 5b, respectively. The desired
acetamides 6a and 6d were obtained by reaction of 5a
with acetic anhydride and carbamates 6b and 6c by
reaction with methylchloroformate.
The preparation of cyclic analogues 8(a–b and h–m) was
performed starting from intermediate 4 which under-
went Suzuki coupling with (3-nitrophenyl)boronic acid
and Mitsunubo reaction using a suitable cyclic nucleo-
phile (Scheme 2). The synthesis of 8c–g was achieved
from 5a by a two-steps-procedure in which the first step
was conversion to an acyclic amide, sulfonamide or
carbamate bearing a leaving group on the end of the
side chain and the second step was a base-mediated
intramolecular cyclisation.
4-Substituted benzamides analogues of 8a, 8b and 8h
were synthesised as shown in Scheme 3. Aldehyde 10,
derived from a Heck reaction between aryl iodide 9 and
allyl alcohol, was converted to the hydroxyketone 11 by
means of an acyloin reaction.9 The primary alcohol was
activated as the mesylate and then displaced with
30-nitrobiphenyl-4-ol. The resulting ketone was reduced
to the corresponding alcohol 12. A Mitsunobu reaction
was performed to introduce the cyclic sustituent and the
t-butyl esters were hydrolysed using formic acid to give
acids 13a, 14a and 15a. Finally, the amides 13, 14 and
15b–d were generated via the acyl chloride.
Scheme 3. Reagents and conditions: (a) allyl alcohol, Pd(OAc)2,
NaHCO3, (n-Bu)4NBr, PhMe, Á (83%); (b) HCHO, cat. N-ethyl
benzothioazolium bromide, EtOH, Á (45%); (c) MsCl, NEt3, DCM, rt
(88%); (d) 30-nitrobiphenyl-4-ol, Cs2CO3, MeCN,
Á
(e) NaBH4, MeOH, rt (79%); (f) Ph3P, DEAD, THF, rt; (g) HOOCH,
(57%);
rt; (h) (COCl)2, DCM; (i) amine, i-Pr2EtN, DCM, rt.

L. Alcaraz et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4043–4046
Table 2. Inhibition results for compounds 9–115
4045
Initial SAR-studies around 2 focused on an exami-
nation of the effects on P2X7 inhibition induced by
changes in the cyclic imide portion. As shown in Table
1, simple replacements such as amine 5a, acyclic amides
6a and 6d or acyclic carbamates 6b and 6c gave inactive
compounds. It would seem that a cyclic substituent is
required for potency and the results in Table 1 illustrate
tight SARs around the cyclic imide portion of the
molecule. Moving from a five- 8a to a six-membered
imide 8b afforded an increase in activity. Deletion of one
of the carbonyl groups dramatically decreased activity
with a loss of ꢃ100-fold 8c and 8d and the cyclic sufo-
namide analogue 8e was poor. Some activity was
regained with cyclic carbamates 8f and 8g. Consistent
with the ring-size effect observed, an increase in potency
was seen upon moving to the larger thiazolidine-2,4-
dione 8h. The hydantoin 8i and, to a lesser extent, the
dihydro-uracil 8k were stable replacements for the imide
group, although substitution by methyl groups on the
hydantoin moiety abolished all activity 8l–m. These
results demonstrate that this portion of the molecule
interacts with a very sterically demanding environment.
Compd
Activity pIC50
Compd
Activity pIC50
13b
13c
13d
14b
14c
6.8
6.8
6.8
6.6
6.6
14d
15b
15c
15d
6.7
6.9
7.1
7.3
imide system with pyrrolidine amide 15b being of com-
parable activity to the corresponding 4-pyridyl analogues.
Typical data from the functional screen are shown in
Figure 1.
BzATP concentration–effect (E/[A]) curves, 0.3–300
mM, were generated in the absence and presence of
compound 8h, 30, 100 and 300 nM. Increasing con-
centrations of compound 8h displaced the location of
the BzATP E/[A] to the right progressively, consistent
with the action of a competitive antagonist, although
there was some depression of the curve with 300 nM of
the compound. Clarke analysis of these data gave a pA2
estimate of 7.7ꢄ0.1 (n=4).
Table 1. Inhibiton results for compounds 5a, 6 and 85
Compd
Activity pIC50
Compd
Activity pIC50
5a
6a
6b
6c
6d
8a
8b
8c
8d
<5
<5
<5
<5
<5
6.9a
7.5a
<5
5.2
8e
8f
8g
8h
8i
8j
8k
8l
<5
6.4a
6.2
7.7a
6.7
<5
6.2
5
8m
<5
apA2 value.
Even though the succinimide and thiazolidine-2,4-dione
groups were suitable substituents for activity, they were
metabolically and chemically fragile, being cleaved or
hydrolysed readily by nucleophiles. However, even if
less active, the carbamates 8f and 8g turned out to be
more stable metabolically and also more stable chemi-
cally to inorganic and organic nucleophiles, having the
advantage that it allowed a relatively simple exploration
of the biphenyl portion of the molecule through Suzuki
coupling on a late stage intermediate and using a paral-
lel approach. Unfortunately, this exercise failed to
afford replacements displaying any advantages over the
3-nitrobiphenyl moiety.
Figure 1. Effects of compound 8h on the human P2X7 receptor. P2X7
receptor activation was assessed by a fluorescent measurement of
BzATP-induced ethidium bromide uptake in THP-1 cells using a
standard 96-well-plate fluorimeter. Experiments were performed in a
potassium (140 mM), EDTA (1 mM), HEPES (20 mM)-buffered salt
solution containing ethidium bromide (0.1 mM) designed to optimize
the assay system for P2X7 receptor activation. Symbols represent the
following, *: vehicle control, *: 30 nM, ~: 100 nM and ~: 300 nM.
We have discovered a novel series of cyclic imides that
are potent P2X7 receptor antagonists. SARs around this
series show that it is possible to change the metaboli-
cally labile thiazolidine-2,4-dione for a stable cyclic car-
bamate group. We have also found that a 30-nitro
substituent on the biphenyl portion is optimal for
activity and that a 4-substituted benzamide group is a
suitable replacement for the 4-pyridyl moiety. The
accepted link between antagonism of the P2X7 receptor
and inhibition of ATP-induced release of IL-1b means
that P2X7 antagonists may have a significant role to
play in the treatment of inflammatory diseases. Unfor-
tunately, although we were able to find a stable cyclic
carbamate replacement for the imide group, the reduc-
tion in activity at the P2X7 receptor made this series of
compounds less attractive for further exploration.
Amongst many 4-pyridyl replacements that we explored
by rapid parallel synthesis, obvious analogues such as
4-pyrimidine, 4-pyridazine, 4-pyridin-2(1H)-one or
4-quinoline failed to yield encouraging levels of activity.
However, the 4-substituted benzamide group turned out
to be more promising and was studied further. Data for
these compounds 13–15 is reported in Table 2. In the
thiazolidine-2,4-dione series 13b–d, the SAR is flat and
results in a 10-fold drop of activity compared to the
4-pyridyl. Interestingly, the succimide series 14a–d
showed similar level of potency as the thiazolidine-
2,4-dione. Finally, the glutarimides 15a–d gave the
expected increase with respect to the five-membered

4046
L. Alcaraz et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4043–4046
5 mM, ethidium bromide 0.1 mM and HEPES 20 mM (H 7.4).
Acknowledgements
BzATP (10 mL, pre-diluted in buffer from a 10 mM stock) and
either vehicle (10% (v/v) DMSO in HEPES-buffered salt
solution, 10 mL) or the appropriate concentration of com-
pound (10 mL in 10% (v/v) DMSO in HEPES-buffered salt
solution) was added to the wells of an assay plate (96-well
non-sterile white, Costar, Corning, Kent, UK). Cell suspen-
sion (80 mL) was then added to each well and the plates were
incubated at room temperature for 90 min, after which time
the plates were read in a plate fluorimeter (excitation 520 nm;
emission 595 nm). For pIC50 determinations cells were chal-
lenged with 10 mM BzATP (a just-maximal concentration) in
the absence and presence of compound and data were
expressed as percentage inhibition of the BzATP response.
For pA2 determinations concentration–effect (E/[A]) curves
for BzATP were constructed using 1/4-log unit increments
(0.3–300 mM) in the presence and absence of compound.
Affinity estimates (pA2) were determined from the relative
pA50s of the curves.
The authors would like to thank Mark Dixon and
Richard Lewis for NMR support, Hema Pancholi for
MS support, Richard Weaver and Peter Webborn for
pharmacokinetic studies.
References and Notes
1. (a) North, R. A. Physiol. Rev. 2002, 82, 1013. (b) Guile,
S. D.; Ince, F.; Ingall, A. H.; Kindon, N. D.; Meghani, P.;
Mortimore, M. P. Prog. Med. Chem. 2001, 38, 115.
2. (a) Burnstock, G. Drug Dev. Res. 2001, 53, 53. (b) Di Vir-
gilio, F.; Vishwanath, V.; Ferrari, D. Handbook Exp. Phar-
macol. 2001, 151, 355. (c) Gu, B.; Bendall, L. J.; Wiley, J. S.
Blood 1998, 92, 946.
3. (a) MacKenzie, A.; Wilson, H. L.; Kiss-Toth, E.; Dower,
S. K.; North, R. A. Immunity 2001, 15, 825. (b) Perregaux, D.;
Gabel, C. A. J. Biol. Chem. 1994, 269, 15195.
4. Labasi, J. M.; Petrushova, N.; Donovan, C.; McCurdy, S.;
Lira, P.; Payette, M. M.; Brissette, W.; Wicks, J. R.; Audoly,
L.; Gabel, C. J. Immunology 2002, 168, 6436.
6. For a second series emerging from this HTS, see: Baxter,
A.; Bent, J.; Bowers, K.; Brough, S.; Fagura, M.; Lawson, M.;
McInally, T.; Mortimore, M.; Robertson, M.; Weaver, R.;
Webborn, P. Bioorg. Med. Chem. Lett. See following article.
doi:10.1016/j.bmcl.2003.08.034.
5. Biological materials and methods. General procedures. Cell
cultures. THP-1 cells were grown in roller flask bottles in
RPMI 1640 media, containing heat-inactivated foetal bovine
serum (10% v/v) and l-glutamine (2 mM), at 37 ꢂC, 5% CO2,
95% humidity. Cultures were split 1 in 3 when a density of
1.5ꢅ106 cells/mL was reached (every 3 days). Plasma mem-
brane pore formation. BzATP-induced changes in plasma
membrane pore formation were assessed by the entry of the
DNA-binding dye ethidium bromide (Sigma Chemical Co.,
Poole, UK). Prior to use in an experiment, cells were trans-
ferred to centrifuge tubes and centrifuged at 700g for 10 min.
The cells were re-suspended at 2.5ꢅ106 cells/mL in HEPES-
buffered salt solution that contained potassium chloride 140
mM, ethylene diamine tetraacetic acid (EDTA) 1 mM, glucose
7. Baxter, A.; Cheshire, D.; McInally, T.; Mortimore, M.;
Cladingboel, D. PCT Int. Appl. WO 9929661, 1999; Chem.
Abstr. 1999, 131, 44809.
8. (a) Baraldi, P. G.; Del Carmen Nunez, M.; Morelli, A.;
Falzoni, S.; Di Virgilio, F.; Romagnoli, R. J. Med. Chem.
2003, 46, 1318. (b) Baraldi, P. G.; Nunez, M. C.; Morelli, A.;
Falzoni, S.; Di Virgilio, F.; Romagnoli, R. Arzn-Forsch. 2002,
52, 273. (c) Ravi, R. G.; Kertesy, S. B.; Dubyak, G. R.;
Jacobson, K. A. Drug Dev. Res. 2001, 54, 75. (d) Baraldi,
P. G.; Romagnoli, R.; Tabrizi, M. A.; Falzoni, S.; Di Virgilio,
F. Bioorg. Med. Chem. Lett. 2000, 10, 681.
9. (a) Matsumoto, T.; Ohishi, M.; Inoue, S. J. Org. Chem.
1985, 50, 603. (b) Taylor, E. C.; Patel, H. H.; Jun, J.-G. J. Org.
Chem. 1995, 60, 6684.