Synthesis and SAR of novel 4,5-diarylimidazolines as potent P2X7 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.10.052

The synthesis and SAR of a series of potent P2X₇ receptor antagonists is described. The synthesis featured high-throughput solution- and solid-phase techniques including a microwave induced imidazoline formation. A series of 4,5-diarylimidazoline libraries were prepared using high-throughput solid-phase and microwave techniques. The compounds were evaluated as P2X ₇ antagonists and their SAR is described.

Full text of DOI:10.1016/j.bmcl.2004.10.052

Bioorganic & Medicinal Chemistry Letters 15 (2005) 435–438
Synthesis and SAR of novel 4,5-diarylimidazolines
as potent P2X7 receptor antagonists
Gregory H. Merriman,* Liang Ma, Patrick Shum, Daniel McGarry, Frank Volz,
Jeffrey S. Sabol, Alexandre Gross, Zhicheng Zhao, David Rampe, Lin Wang,
Friederike Wirtz-Brugger, Bruce A. Harris and Douglas Macdonald
Aventis Inc., 1041 Route 202-206, Bridgewater, New Jersey, NJ 08807-0800, USA
Received 9 August 2004; accepted 20 October 2004
Abstract—Aseries of 4,5-diarylimidazoline libraries were prepared using high-throughput solid-phase and microwave techniques.
The compounds were evaluated as P2X7 antagonists and their SAR is described.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
As a result of a high-throughput screening effort we
identified 1 as a potent new P2X7 antagonist lead. Here
we describe our efforts to rapidly explore the SAR of a
series of diarylimidazoline analogs of 1 using a high-
throughput medicinal chemistry approach.
The P2X7 receptor has received considerable attention
over the past few years as a therapeutic target.1,2 The
P2X7 receptor is a ligand-gated ion channel that is
present on a variety of cell types, such as, mast cells,
macrophages and lymphocytes, all thought to be
involved in inflammation and autoimmune processes.3
Activation of the P2X7 receptor by adenosine triphos-
phate (ATP) leads to release of the pro-inflammatory
cytokines interleukin-1b (IL-1b) and interleukin-18
(IL-18), as well as giant cell formation, including macro-
phages and microglial cells and also leads to mast cell
degranulation.4,5 As such, it is anticipated that a P2X7
antagonist will show anti-inflammatory activity and
may be of therapeutic use for diseases with this compo-
nent such as rheumatoid arthritis, AlzheimerÕs disease
and stroke.
2. Chemistry
The design and synthesis of a focused 4,5-diaryl-imidazo-
line library based on 1 employed a combination of
solid phase and microwave techniques. Initially we
envisioned a simple condensation of a 1,2-diaryl-1,2-
ethanediamine scaffold 2 with a series of carboxylic acid
building blocks to afford the desired cyclic imidazoline
targets 7 directly (Scheme 1). Unfortunately attempts
to prepare the imidazolines in one step under a variety
of dehydrative conditions failed to give the desired tar-
gets, usually affording a complex mixture of compounds
that was difficult to separate. As an alternate approach
we explored the stepwise formation of target imidazo-
lines using solid-phase and microwave techniques high-
lighted in Scheme 1. Thus, coupling of the
ethanediamine scaffold 2 with p-nitrophenyl-carbonate
resin 3 afforded the carbamate bound amine 4 in quanti-
tative yield.6 Aseries of carboxylic acids, chosen to ex-
plore the chain length, electronic and steric requirements
of the pendent aromatic ring, were then coupled using
a standard amide coupling with EDC in dimethyl-form-
amide to afford 5 in quantitative yield.7 Attempts to
cleave and cyclize (5 ! 7) in one pot using TFAor other
H
N
N
1
P2X7 IC50 = 0.08 µM
*
Corresponding author. Fax: +1 908 231 4774; e-mail: gregory.
merriman@aventis.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.052

436
G. H. Merriman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 435–438
O
O
NH2
NH2
O
O
+
a
O
N
H
NH2
NO2
2
3
4
b
O
H
N
c
O
N
O
d
NH
H2N
O
R
N
NH
6
R
R
Libraries 7-8
5
Scheme 1. Reagents and conditions: (a) CH2Cl2, rt; (b) RCO2H, EDC, DMF; (c) 50% TFA/CH2Cl2; (d) TMS-polyphosphate, CH2Cl2, microwave
140ꢁC, 8min.
strong organic acids with heat failed to give 7, affording
only the uncyclized amine cleavage product 6. Since 6
could be isolated in high yield and purity we focused
on cyclizing 6 ! 7 off resin. Our initial attempts to
cyclize amides 6 using polymer bound Burgess reagent
failed to afford 7 under a variety of conditions. We then
turned our attention to the use of trimethylsilyl
polyphosphate (TMS-PP), a reagent previously shown
to be effective in the preparation of 1,4,5,6-tetrahydro-
pyrimidines.8 Thus, treatment of 6 with TMS-PP in
dichloromethane under microwave irradiation at
140ꢁC (ꢀ12–15bar) afforded the desired imidazolines
in good overall yield.9 Two focused libraries 7 and 8
were prepared using the aforementioned route to
explore the SAR.
Benzyl substituted analogs also maintained activity as
shown with compound 7g. Notably, there is no apparent
enantiomeric preference for the branched substitutent,
as both enantiomers of 7g were prepared and were
found to be equipotent.12
It was anticipated that substitution of the pendent aryl
ring would serve to maintain or improve potency as well
as potentially offer compounds that would have reduced
metabolic liability. Based, in part, on the aforemen-
tioned P2X7 antagonists (7) and on the relative avail-
ability of requisite carboxylic acid building blocks,
additional focused libraries of phenethyl and benzyl
analogs with substitution on the aryl ring (8) were pre-
pared and the activities of a representative group are
highlighted in Table 2.
3. Results
Electron withdrawing and small neutral groups were
generally well-tolerated on both the phenethyl (n = 2)
and benzyl (n = 1) analogs. However, electron-donating
groups were much less active as exemplified by 8k,l,m
and y. Substitution of the ortho and meta aryl positions
of the phenethyl group gave analogs that tended to be
more potent in general than the corresponding para-
substituted analogs (8c,f,h,j). On the other hand
ortho-substituted benzyl analogs (8n,r,s,w) were less po-
tent than corresponding meta- and para-substituted
analogs.
Target compound P2X7 IC50s were assessed using a cell-
ular (U373 cells stably expressing human P2X7) YO-
PRO-1 dye uptake fluorescence assay.10,11 To assess
the appropriate chain length, branching and configura-
tional constraints of the alkyl-aromatic R group in the
2-position of the imidazoline, a small group of com-
pounds (7a–k) was prepared (Table 1). In general, activ-
ity was maintained in compounds with chain lengths
1–4, with a drop in potency once the 3 carbon chain is
reached (7i). Compounds with aryl moieties directly at-
tached to the imidazoline scaffold were inactive as exem-
plified by 7h. Branching with small alkyl groups on the
chain was also well tolerated within the series. Alkyl-
branched phenethyl substituted compounds 7a–e all
maintained activity, however, 7f, a conformationally re-
stricted trans-cyclopropane analog, was inactive at the
concentration tested.
4. Conclusion
In conclusion we have prepared and evaluated a new
series of imidazolines as P2X7 antagonists. The
chemistry used to prepare this series included a
combined solid- and solution-phase approach that was
quite general and amenable to library synthesis.
The key step in the synthesis was a microwave cycliza-
tion using TMS-PP to afford the desired imidazoline
targets.
The most potent compounds in the group were
methyl and gem-dimethyl substituted 7a and 7b,
respectively.

G. H. Merriman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 435–438
437
Table 1.
Table 2.
H
N
H
N
( )n
R
N
N
X
8
7
Y
Entry
R
P2X7IC50 (lM)a
Entry
n
X
Y
P2X7 IC50 (lM)a
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
H
2-F
H
0.08
0.16
1
0.08
0.03
8a
8b
8c
8d
8e
8f
H
3-F
4-F
H
0.04
0.22
H
7a
7b
2-Cl
3-Cl
4-Cl
3-F
H
0.07
0.12
>1.00b
0.12
1.40
H
H
0.01
8g
8h
8i
4-F
4-Cl
H
3-Cl
2-Me
4-Me
2-OMe
4-OMe
3-OMe
2-F
0.07
2.19
8j
H
8k
8l
H
0.83
7c
7d
0.32
0.57
H
>3.00b
>3.00b
0.58
8m
8n
8o
8p
8q
8r
8s
8t
4-OMe
H
3-F
4-F
H
0.08
0.08
H
3-F
2-F
4-F
3-F
H
0.03
0.93
7e
7f
0.05
2-Cl
3-Cl
4-Cl
4-CF3
2-Me
4-Me
2-OMe
0.16
0.03
H
>1.00b
8u
8v
8w
8x
8y
H
0.27
0.40
H
H
0.32
0.08
>3.00b
7g
0.11
H
3-OMe
a Values are means of experiments performed in duplicate; one, two or
three times.
7h
7i
7j
3.80
0.54
0.24
b Indicates the highest concentration tested resulting in no appreciable
inhibition.
References and notes
1. (a) Di Virgilio, F. Immunol. Today 1995, 524; (b) Supr-
enant, A.; Rassendren, F.; Kawashima, E.; North, R. A.;
Buell, G. Science 1996, 272, 5262, 735.
7k
0.61
2. Several groups have recently reported on P2X7 antago-
nists: (a) Baxter, A.; Bent, J.; Bowers, K.; Braddock, M.;
Brough, S.; Fagura, M.; Lawson, M.; McInally, T.;
Mortimore, M.; Robertson, M.; Weaver, R.; Webborn,
P. Bioorg. Med. Chem. Lett. 2003, 13(22), 4047; (b)
Alcaraz, L.; Baxter, A.; Bent, J.; Braddock, M.; Clad-
ingboel, D.; Donald, D.; Fagura, M.; Laurent, C.;
Lawson, M.; Mortimore, M.; McCormick, M.; Roberts,
N.; Robertson, M. Bioorg. Med. Chem. Lett. 2003, 13(22),
4043; (c) Baraldi, P. G.; Nunez, M. C.; Morelli, A.;
Falzoni, S.; Di Virgilio, F.; Romagnoli, R. J. Med. Chem.
2003, 46, 1318; (d) Baraldi, P. G.; Romagnoli, R.; Tabrizi,
M. A.; Falzoni, S.; Di Virgilio, F. Bioorg. Med. Chem.
Lett. 2000, 681; (e) Ravi, R. G.; Kertsey, S. B.; Dubyak,
G. R.; Jacobson, K. A. Drug Dev. Res. 2001, 54(2), 75;
Chessell, I. P.; Michel, A. D.; Humphrey, P. P. A. Brit. J.
Pharmacol. 1998, 124(6), 1314.
a Values are means of experiments performed in duplicate, one or two
times.
b Indicates the highest concentration tested resulting in no appreciable
inhibition.
The initial investigation of SAR to explore the appropri-
ate chain length, branching and aryl substitution at the
2-position of the imidazoline were described. Several
new potent P2X7 antagonists were found offering a vari-
ety of substitution patterns that will be used to help fur-
ther optimize this series.
We are currently preparing additional libraries using the
aforementioned chemistry to investigate the SAR of
other areas of the scaffold to modulate the physical
chemical characteristics and pharmacokinetic profiles
of compounds in this series.
3. (a) Collo, G.; Neidhart, E.; Kawashima, E.; Kosco-
Vilbois, M.; North, R. A.; Buell, G. Neuropharmacology
1997, 36(9), 1277; (b) Di Virgilio, F.; Sanz, J. M.; Chiozzi,
P.; Falzoni, S. Progr. Brain Res. 1999, 120, 355.

438
G. H. Merriman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 435–438
4. Perregaux, D. G.; McNiff, P.; Laliberte, R.; Conklyn, M.;
Gabel, A. G. J. Immunol. 2000, 165, 4615.
5. Solle, M.; Labasi, J.; Perregaux, D. G.; Stam, E.;
Petrushova, N.; Koller, B. H.; Griffiths, R. J.; Gabel, C.
A. J. Bio. Chem. 2001, 276, 1, 125.
then purified using preparative reverse phase HPLC to
afford 66mg (41%) of the imidazoline 8i. 1H NMR
(CDCl3, 300MHz) d 2.38 (s, 3H, CH3), 2.82 (t, 2H,
CH2Ph), 3.15 (t, 2H, CH2C@N), 5.25 (s, 2H, N–CH–Ph),
6.8 (m, 4H, ArH), 7.0 (m, 6H, ArH), 7.15 (d, 2H, ArH),
7.25 (d, 2H, ArH); LC/MS M+1 342.
6. Resin preparation: To a suspension of 10.0g (13.2mmol) of
polystyrene-4-nitrophenylcarbonate resin (commercially
available Advanced ChemTech 1.32mmol/g loading) in
150mL of dichloromethane was added 8.41g (39.6mmol)
of 1,2-diphenylethylenediamine. The mixture was shaken
at room temperature for 12h then filtered and washed with
4 · 500mL of dichloromethane and dried in a vacuum
oven. Loading was quantitative by UV determination (p-
nitrophenol k 320nm).
7. Rich, D. H.; Singh, J. In The Peptides: Analysis, Synthesis,
Biology; Academic: New York, 1979; Vol. 1, p 241.
8. Perillo, I. A.; Garcia, M. B.; Niemevz, F.; Orelli, L. R. J.
Heterocycl. Chem. 1999, 36, 105.
9. General example of EDC coupling and TMS-PP Micro-
wave cyclization: 4,5-Diphenyl-2-(2-p-tolyl-ethyl)-4,5-dihy-
dro-1H-imidazole 8i. To 300mg (0.40mmol) of the
aforementioned resin suspended in 6mL of DMF was
added 0.15g (0.78mmol) of 1-(3-dimethylaminopropyl)-3-
ethylcarbodiimide (EDC) followed by 0.13g (0.78mmol)
of 3-p-tolyl-propionic acid. The mixture was shaken for
12h and the resin was filtered and washed sequentially
with 3 · 10mL of DMF, 3 · 10mL of methanol and
3 · 10mL of dichloromethane. The resin was then treated
with 4mL of 50% trifluoroacetic acid in dichloromethane
and stirred for 2h. The resin was filtered and washed with
dichloromethane. The filtrate was concentrated in vacuo
to give 135mg (100%) of the amine as a yellow oil.
Asolution of the aforementioned residue in 2mL of
dichloromethane was added 1mL of TMS-PP in dichlo-
romethane solution (25g TMS-PP/100mL). The resulting
solution was warmed in the microwave (Smith Synthe-
sizerTM—Personal Chemistry) at 140ꢁC for 8min final
pressure ꢀ15bar. The work-up was performed using the
automated AllexTM liquid–liquid extraction station as
follows: The sample was washed with 2 · 3mL of 2N
NaOH (aq) and 2 · 3mL of brine and then dried over
Na2SO4. The organic phase was concentrated in vacuo
10. P2X7 receptor-mediated dye uptake assay: Bz-ATP
(20- & 30-O-(4-benzoylbenzoyl)-adenosine-50-triphosphate
(Sigma Chemical, St. Louis, MO)) activation of the P2X7
receptor was measured using YO-PRO-1 iodide (491/509)
dye (Molecular Probes, Eugene, OR) uptake in U373 cells
stably expressing the human P2X7 receptor with a method
modified from the procedure described by Virginio, C.;
MacKenzie, A.; North, R. A.; Surprenant, A. J. Physiol.
1999, 519.2, 335. This dye uptake assay was used to screen
for P2X7 receptor antagonists by measuring inhibition of
dye uptake. Briefly, cells were plated and grown overnight
on collagen coated 96-well plates at a density of 35,000
cells/well. The following day, culture media was replaced
with Mg2+ and Ca2+-free HankÕs balanced salt solution in
the presence of increasing concentrations of test comp-
ounds in duplicate and the cells were then incubated for
20min at 37ꢁC. Next, 5mM YO-PRO-1 iodide dye and
300mM Bz-ATP were added to the cells sequentially and
then incubated for 1.5h at 37ꢁC. Dye uptake through
P2X7 was measured by fluorescence YO-PRO-1 nucleic
acid staining using a CytoFluor Series 4000 fluorescence
plate reader (PerSeptive Biosystems, Framingham, MA)
with an excitation filter of 485/20 and an emission filter of
530/25. Fluorescence in the presence of the test comp-
ounds was compared to that in the absence of test
substances (control) and IC50s were calculated using XLfit
Version 2.0.9 (ID Business Solutions, Ltd, Surrey, UK)
and nonlinear sigmoidal curve fit.
11. ATP-stimulated YO-PRO uptake in U373MG cells was
abolished by oxidized ATP see: Rampe, D.; Wang, L.;
Ringheim, G. E. J. Neuroimmunol. 2004, 147, 56.
12. Only one enantiomer of 7a (that shown in Table 1) was
prepared in this set of compounds, however, given that 7b
is also quite potent it is anticipated that the opposite
enantiomer of 7a would also be equipotent. Compound 7d
was tested only as a racemic mixture.