Hit-to-Lead studies: The discovery of potent adamantane amide P2X 7 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2003.08.034

A Hit-to-Lead optimisation programme was carried out on the adamantane high throughput screening hit 1 resulting in the discovery of a number of potent P2X₇ antagonists.

Full text of DOI:10.1016/j.bmcl.2003.08.034

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4047–4050
Hit-to-Lead Studies: The Discovery of Potent Adamantane Amide
P2X7 Receptor Antagonists
Andrew Baxter,* Janice Bent, Keith Bowers, Martin Braddock, Steve Brough,
Malbinder Fagura, Mandy Lawson, TomMcInally, Mike Mortimore, Mark Robertson,
Richard Weaver and Peter Webborn
AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
Received 17 July 2003; revised 19 August 2003; accepted 19 August 2003
Abstract—A Hit-to-Lead optimisation programme was carried out on the adamantane high throughput screening hit 1 resulting in
the discovery of a number of potent P2X7 antagonists.
# 2003 Elsevier Ltd. All rights reserved.
The use of High Throughput Screening (HTS) is now
widespread in the pharmaceutical industry. There was
an expectation that once a screen was established for a
particular target then potent lead compounds or candi-
date drugs would be found. The reality is often far from
this. Bridging the gap between the end of a HTS and the
start of a full Lead Optimisation (LO) project has been
described as Hit-to-Lead (HtL).1 Hits fromHTS are
profiled and compared to a generic target lead criteria.
The lead target profile used is shown in Figure 1. Lead
series then have a balance of properties — potency and
SAR as well as an encouraging metabolic and selectivity
profile — such that rapid (less than two years) further
optimisation should provide Candidate Drugs (CDs)
suitable for progression into clinical development.
The P2X7 receptor is a ligand-gated ion channel present
on a variety of cell types involved in the inflammatory/
immune process, specifically macrophages, mast cells
and lymphocytes (T and B).2À5 Activation of the P2X7
receptor by the extracellular nucleotide adenosine tri-
phosphate (ATP), leads to the processing and release of
the proinflammatory cytokine interleukin-1b (IL-1b)
from monocytes and macrophages,6 degranulation of
mast cells and the shedding of surface molecules, l-
7
selectin and CD23 fromlymphocytes.
P2X7 receptors
are also located on antigen-presenting cells, keratino-
cytes, salivary acinar cells and hepatocytes. It was
desirable to make compounds effective as P2X7 receptor
antagonists for use in the treatment of chronic inflam-
matory diseases where the P2X7 receptor may mediate
the release of IL-1b. Recently, evidence for the role of
this receptor in disease processes was published; P2X7-
KO mice showing a reduced severity in an anti-collagen
antibody arthritis model.8
Few classes of P2X7 antagonists are known in the lit-
erature. KN-62, a bis-isoquinolinesulphonyltyrosine
derivative, was first described by Gargett and Wiley9
and has an IC50 of 51 nM (see Fig. 2). Subsequently
various groups have explored this area and have found
more potent analogues such as MRS2306 (IC50 40
nM)10 and a phenylpiperazine derivative (IC50 1 nM).11
Whilst being interesting and useful tools to investigate
the biology of P2X7 antagonism, they do not represent
good starting points for the discovery of oral drugs.
These compounds have large molecular weights (>700),
Figure 1. Hit-to-Lead generic lead target profile.
*Corresponding author. Tel.: +44-1509-644772; fax: +44-1509-
645513; e-mail: andrew.jg.baxter@astrazeneca.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.034

4048
A. Baxter et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4047–4050
Table 1. Profile of adamantanes 1–4
Figure 2. KN-62 and analogues.
are very lipophilic (clogP >6) and would be expected to
be metabolically labile by virtue of the essential sulpho-
nate group. The published SAR10 indicates that the
amide, sulphonamide and sulphonate are all essential
for potency and so the transformation of KN-62 analo-
gues into compounds with drug-like properties12 would
be very challenging.
Generic lead criteria
1
2
3
4
P2X7 pA2>7.0
Mr<450
clogP <3.0
6.9
540
6.2
6.4
338
5.2
8.1
304
4.5
NAa
304
5.0
The biological screen for the assessment of P2X7 recep-
tor inhibition was the reduction of plasma membrane
pore formation in a monocytic cell line stimulated with
the synthetic ATP analogue, benzoylbenzoyl adenosine
triphospate (BzATP). This characteristic of P2X7
receptor activation was measured by the entry of the
fluophore, ethidium bromide through the membrane
pores with a consequent increase in total cellular fluo-
rescence.13 A HTS was undertaken using this assay in
96-well plate format and the adamantane 1 was identi-
fied.14 Potency estimates for the compounds were initi-
ally made from pIC50 determinations and then more
robustly with pA2 estimates.
aNA=<50% inhibition at 10 mM.
Commercially available amines and acid chlorides were
reacted to give the corresponding amides (2–6, 8–19,
21–22, 24–33). Borane reduction of the amide 4 gave
amine 7. The anilines 20 and 23 were prepared by iron
ammonium chloride reduction of the corresponding
nitro compounds.
Removal of one of the amide side chains gave 2 which
maintained most of the potency of the bisamide 1 but
now had acceptable molecular weight. At this time the
two mono-chloro amides 3 and 4 were also prepared.
Surprisingly, the 2-chloroamide 3 was significantly more
potent than amides 1 or 2 and the 4-chloro amide 4 was
inactive. It is assumed that the presence of the ortho
substituent causes a twist in the orientation of the ben-
zamide that is necessary for potent P2X7 antagonism
and that substitution in the para position is not allowed.
At this point, variation of the adamantane was under-
taken. Many 2-chlorobenzamides were tested from the
corporate collection, compound purchase and from the
synthesis of combinatorial libraries. No replacements
for adamantane were found (data not shown).
The profile of the adamantane 1 is shown in Table 1
compared with key lead criteria — potency, molecular
weight and lipophilicity. Whilst the potency was accep-
table, the molecular weight and lipophilicity were
appreciably outside the lead criteria limits. Reduction of
these parameters therefore became an essential first step
in assessing if adamantanes were to become lead com-
pounds.
The preparation of compounds in this study was carried
out using standard synthetic methods (Scheme 1).
Scheme 1. (i) ArCOCl, Et3N, CH3CN; (ii) ArNH2, Et3N, CH3CN; (iii) ArCOOH, CDI, DMF; (iv) Fe, NH4Cl.

A. Baxter et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4047–4050
4049
Amide 3 then became the new starting point for further
variation. The nature of the linking group between the
adamantane and 2-chlorophenyl ring was investigated
(Table 2). C- and N-methylation, carbonyl reduction
and methylene removal (5–8) all caused a loss of
potency. Chain extension (9) maintained potency whilst
amide reversal (10) caused some potency loss. At this
point combinatorial libraries were prepared using the
reaction of adamantanemethanamine with a large vari-
ety of acid chlorides or via standard amide coupling
with carboxylic acids. After testing for potency, key
compounds were re-prepared and characterised chemi-
cally and biologically as solid samples (Table 3). Gen-
eral trends indicated that only aromatic amides had
activity as P2X7 antagonists (data not shown). A clear
SAR around the benzamide ring was uncovered — a 2-
substituent is required (3, 12–14 vs 11 and 16) and 4-
substitution is deleterious (2 vs 3). Amongst the
dichlorobenzamides, 2,3 (16) and 2,5 (17) are better
than 2,6 (18) with the dichloro compounds either lack-
ing a 2-substituent — 3,5 (19)— or with a 4-substituent
— 2,4 (2)— being much less potent confirming the con-
clusions drawn from the mono-chloroamides. Extension
of the preferred 2,3- and 2,5-substitution pattern to
other functional groups gave a range of sub-10 nM
antagonists (compounds 16, 17, 20–24).
Even though the reverse amide 10 was less potent than
the benzamide 3, an examination of SAR similarities and
differences between the two series was undertaken. Again
combinatorial libraries were prepared using standard
amide coupling methods between adamantaneacetic acid
and a variety of amines. After potency testing, chemical
and biological characterisation was undertaken on re-
prepared solid samples (Table 4). 2,3- and 2,5-dis-
ubstitution was again found to be the most potent as
was found previously with the benzamides (10 and 25 vs
26–28). Larger potency increases were seen though so
that again sub-10 nM antagonists were found (com-
pounds 27 and 28). In addition a separate sub-set of
anilides was found based on 6,5-ring systems, the inda-
zole 31 has good antagonist potency.
Table 4. P2X7 antagonist potencies—anilides
Ar
P2X7 pA2
10
25
26
27
28
2-ClPh
2-MePh
2-Cl,5-OMePh
2-Me,3-OMePh
2-Me,5-OMePh
6.3 (pIC50
6.8 (pIC50
7.2
8.3
8.0
)
)
Table 2. P2X7 antagonist potencies—linker variation
29
6.5
5.9
30
a
X
P2X7 pA2
31
7.4
3
5
6
7
8
9
10
CH2NHCO
CHMeNHCO
CH2NMeCO
CH2NHCH2
NHCO
8.1
6.4 (pIC50
5.4 (pIC50
5.4
NA
7.8
6.3 (pIC50
)
)
A key part of Hit-to-Lead strategy is to assess and
improve the Drug Metabolism and PharmacoKinetic
(DMPK) properties at an early stage. With a variety of
antagonists with good potency in hand, measurement of
intristic clearance (Clint) in vitro using rat hepatocytes
and human liver microsomes were carried out. This data
should be compared with the lead criteria shown in
Figure 1. Results of some of these experiments are pre-
sented in Table 5. Not surprisingly, the simple lipophilic
benzamides were all rapidly cleared by rat hepatocytes
(3, 16, 17). Confirmation in vivo was carried out for the
dichlorobenzamide 16. Interestingly some of the other
2,3- and 2,5-disubstituted benzamides had better in
vitro rat hepatocyte clearance (20, 22, 23) but, in the
case of the best compound (23 with rat heps Clint 24),
this did not translate into any improvement in vivo. A
similar picture was observed for the simple anilides 26–
28, all having high clearances in vitro. Unexpectedly the
indazole amide 31 has reduced in vitro clearance and
was the first compound to fulfil the in vitro clearance
generic lead criteria. Clearance in vivo for 31 is close to
criteria with volume of distribution and half-life esti-
mate inside limits.
(CH2)2NHCO
CH2CONH
)
aNA, <50% inhibition at 10 mM.
Table 3. P2X7 antagonist potencies—benzamides
R
P2X7 pA2
R
P2X7 pA2
3
2-Cl
H
2-Br
2-OMe
2-Me
3-Cl
2,3-DiCl
2,4-DiCl
8.1
7.1
8.0
7.1
7.9
7.2
8.8
6.4
17
18
19
20
21
22
23
24
2,5-DiCl
2,6-DiCl
3,5-DiCl
2-Cl,3-NH2
2-Cl,5-OMe
2-Cl,5-OH
2-Cl,5-NH2
2-Cl,6-NH2
8.3
7.4
5.4 (pIC50
8.2
8.8
8.0
8.4
8.3
11
12
13
14
15
16
2
)

4050
A. Baxter et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4047–4050
Table 5. P2X7 antagonist potencies and DMPK parameters
Y
Ar subst. Rs
P2X7 pA2
Rat heps Clint
Hu mics Clint
Rat pharmacokinetics iv
Vss
Cl
T1/2
3
NHCO
NHCO
NHCO
NHCO
NHCO
NHCO
NHCO
CONH
CONH
CONH
CONH
2-Cl
2,3-DiCl
2,5-DiCl
8.1
8.8
8.3
8.2
8.0
8.4
8.3
7.2
8.3
8.0
7.4
>100
>100
>100
56
45
24
>100
45
16
17
20
22
23
24
26
27
28
31
33
>100
6.2
3
1.0
2-Cl,3-NH2
2-Cl,5-OH
2-Cl,5-NH2
2-Cl,6-NH2
2-Cl,5-OMe
2-Me,3-OMe
2-Me,5-OMe
5-Indazolyl
88
19
>100
79
>100
>100
21
>100
0.5
1.0
>100
46
5
47
2
From this study the adamantine indazole amide 31 was
identified as a lead compound fulfilling most of the tar-
get lead criteria. Excellent sub-10 nM potency was also
achieved in a number of compounds (e.g., 16). Poor
pharmacokinetic data was a feature of the adamantanes
but this could be ameliorated by structural variation as
in indazole 31. The encouragement obtained from this
clearance data, in vitro and in vivo, suggested that fur-
ther improvements could be made and so this study was
approved as a full LO project.
6. Perregaux, D.; Gabel, C. A. J. Biol. Chem. 1994, 269,
15195.
7. Gu, B.; Bendall, L. J.; Wiley, J. S. Blood 1998, 92, 946.
8. Labasi, J. M.; Petrushova, N.; Donovan, C.; McCurdy, S.;
Lira, P.; Payette, M. M.; Brissette, W.; Wicks, J. R.; Audoly,
L.; Gabel, C. J. Immunol. 2002, 168, 6436.
9. Gargett, C. E.; Wiley, J. S. Br. J. Pharmacol. 1997, 120,
1483.
10. Ravi, R. G.; Kertesy, S. B.; Dubyak, G. R.; Jacobson,
K. A. Drug Dev. Res. 2001, 54, 75.
11. Baraldi, P. G.; Nunez, M. C.; Morelli, A.; Falzoni, S.; Di
Virgilio, F.; Romagnoli, R. J. Med. Chem. 2003, 46, 1318.
12. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P. J. Adv. Drug Deliv. Rev. 1998, 23, 3.
13. For full details of the biological assays used see: Baxter,
A.; Brough, S.; McInally, T.; Mortimore. M.; Cladingboel, D.
Int. Patent Appl. WO 98/02189.
References and Notes
1. Michne, W. F. Pharm. News 1996, 3, 19.
2. North, R. A. Physiol. Rev. 2002, 82, 1013.
3. Burnstock, G. Drug Dev. Res. 2001, 53, 53.
4. Di Virgilio, F.; Vishwanath, V.; Ferrari, D. Handbook Exp.
Pharmacol. 2001, 151, 355.
5. Guile, S. D.; Ince, F.; Ingall, A. H.; Kindon, N. D.; Meghani,
P.; Mortimore, M. P. Prog. Med. Chem. 2001, 38, 115.
14. See accompanying paper for an alternative series of P2X7
antagonists. Alcaraz, L.; Baxter, A.; Bent, J.; Bowers, K.;
Braddock, M.; Cladingboel, D.; Donald, D.; Fagura, M.; Fur-
ber, M.; Laurent, C.; Lawson, M.; Mortimore, M.; McCor-
mick, M.; Roberts, N.; Robertson, M. Bioorg. Med. Chem.
Lett. See preceding article. doi: 10.1016/j.bmcl.2003.08.033.