Synthesis and evaluation of substrate-mimicking cytosolic phospholipase A2 inhibitors - Reducing the lipophilicity of the arachidonyl chain isostere

Article abstract of DOI:10.1016/j.bmcl.2004.05.024

The high lipophilicity of a series of cytosolic phospholipase A₂ inhibitors has been reduced by the modification of a decyloxyphenyl chain designed to mimic the arachidonyl group of the natural substrate. These changes have resulted in an improvement in the whole cell potency of the inhibitors.

Full text of DOI:10.1016/j.bmcl.2004.05.024

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3645–3649
Synthesis and evaluation of substrate-mimicking cytosolic
phospholipase A2 inhibitors––reducing the lipophilicity
of the arachidonyl chain isostere
Iain Walters,* Colin Bennion, Stephen Connolly, Pamela J. Croshaw, Kim Hardy,
Paul Hartopp, Clive G. Jackson, Sarah J. King, Louise Lawrence, Antonio Mete,
David Murray, David H. Robinson, Linda Stein, Edward Wells and W. John Withnall
Departments of Medicinal Chemistry, Molecular Biology, and Discovery BioScience,
AstraZeneca R&D Charnwood, Bakewell Road, Loughborough, Leicestershire LE11 5RH, United Kingdom
Received 28 November 2003; revised 10 May 2004; accepted 13 May 2004
Available online
Abstract—The high lipophilicity of a series of cytosolic phospholipase A2 inhibitors has been reduced by the modification of a
decyloxyphenyl chain designed to mimic the arachidonyl group of the natural substrate. These changes have resulted in an
improvement in the whole cell potency of the inhibitors.
Ó 2004 Elsevier Ltd. All rights reserved.
Cytosolic phospholipase A2 (cPLA2) is the cellular
enzyme believed to be responsible for cleavage of ara-
chidonic acid from the sn-2 position of membrane
phospholipids.1–4 This process is thought to be the
principal source of free arachidonic acid within acti-
vated cells.5 Since arachidonic acid is the raw material
for the cellular production of a variety of other pro-
inflammatory mediators such as prostaglandins and
leukotrienes,6;7 inhibition of cPLA2 activity represents
an attractive approach to the control of inflamma-
tion.8–10
O
O
O
OH
X
O
1a R = CH3(CH2)9O
1b R = Ph(CH2)5O
1c R = Ph(CH2)5S
Figure 1. Structure of cPLA2 inhibitors.
We recently described a series of potent cPLA2 inhibi-
tors such as 1a–c (Fig. 1) designed to mimic the binding
of the natural substrate.11 These inhibitors comprise
three core elements: (i) a terminal carboxylic acid, which
functions as a phosphate isostere, (ii) an electrophilic
ketone, which serves as a serine trap, (iii) an extended
alkoxyphenyl chain, which mimics the arachidonyl
group of the substrate. This moiety is the major con-
tributor to the very high lipophilicity of these com-
pounds, and the present communication describes our
efforts to moderate this undesirable property and so
discover more ‘drug-like’ inhibitors.
Since we were interested in modifying the long alkyl
chain of these cPLA2 inhibitors, it was desirable to
employ a general synthesis which enabled this group to
be introduced at a late stage. The two principal routes
used to prepare these compounds are outlined in
Schemes 1 and 2. The key epichlorohydrin-derived
epoxide intermediate 3 may be reacted with a suitable 4-
functionalised phenol as shown in Scheme 1, to generate
secondary alcohols such as 4 or 5. These may then be
attached to a modified alkyl chain by either alkylation of
the phenolic hydroxyl of 4, followed by oxidation and
ester hydrolysis to give 12 and 17–21 (Table 1), or
Keywords: cPLA2 inhibition; Arachidonyl chain mimic.
* Corresponding author. Tel.: +44-1509-644001; fax: +44-1509-6455-
71; e-mail: iain.walters@astrazeneca.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.05.024

3646
HO
I. Walters et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3645–3649
in Scheme 2. The resultant phenols 9, 10 and 11 can be
O
a,b
O
subsequently combined with epoxide 3 to furnish the
inhibitor products 13–16 (Table 1) and 24–30 (Table 2)
after oxidation and ester hydrolysis.
OH
OH
OtBu
3
2
O
c,d
O
The activities of the compounds described herein were
compared in two biological screens, a ‘monomeric’ assay
based on the cPLA2-mediated hydrolysis of a simple,
soluble synthetic substrate,12 and a cellular screen based
on the stimulated production of arachidonic acid by
HL60 cells.11 Our initial attempts sought to reduce the
log P of these inhibitors by introducing a polar func-
tionality such as an ether oxygen as in compound 12 or
an amide linkage (13) within the body of the long alkyl
chain. The poor enzyme inhibition shown by these two
compounds (Table 1) was shared by many other similar
analogues, and so we turned our attention instead to
varying the end of the lipophilic chain. Previous SAR11
had shown that a phenylpentyl group as in 1b or 1c
maintained most of the potency of the decyl chain of 1a,
so a number of analogues containing a terminal het-
erocycle were prepared (14–20). Three of these ana-
logues (14, 15 and 17) showed some promise with near
micromolar whole cell potencies and reduced lipophi-
licities (Table 1). Another attractive approach entailed
replacing the decyloxy chain with a substituted benzyl-
oxy group, and the best of these, the 3,4-dichlorobenzyl
derivative 21, was equipotent with its decyloxy analogue
1a in the whole cell assay despite its markedly lower
c log P (Table 1).
e,d
O
O
OtBu
4
HO
O
OH
O
O
f,g,h
OtBu
HO
5
O
O
i,g,h
12,17-21
22,23
Scheme 1. Reagents and conditions: (a) tBuOH, DCC, DMAP, rt,
69%; (b) epichlorohydrin, Cs2CO3, MeCN, reflux, 76%; (c) 4-benzyl-
oxyphenol, DABCO, DMF, 120 °C, 61%; (d) H2 (4 bar), Pd/C, EtOAc,
rt, 95%; (e) benzyl 4-hydroxybenzoate, DABCO, DMF, 120 °C, 50%;
(f) alkyl halide, Cs2CO3, MeCN, rt; (g) Ac2O, DMSO, rt; (h) TFA,
DCM, rt; (i) 5-phenyl-1-pentanol or 4-phenylbutylamine, DCC, DCM,
rt.
HO2C
HO2C
OH
OH
OH
NC
HS
8
6
7
Further success was achieved by altering the function-
ality tethering the lipophilic chain to the phenoxy group
believed to mimic the D-5,6 double bond of arachidonic
acid.11 Several different polar functional groups were
investigated as alternatives to ether and sulfide linkages
(Table 2), with phthalimide 24 and oxadiazole 25 both
giving rise to submicromolar whole cell potencies.
Finally, the best features of these different strategies
were combined to give the dichlorophenyl analogues 28–
30 (Table 2). The two oxadiazoles 28 and 29 showed
pleasing activities in the monomeric assay, and the di-
chlorobenzylphthalimide 30 was micromolar in the
whole cell assay.
a,(b)
c,d
e
O
N
OH
OH
OH
R
N
O
R
R
S
N
O
9
10
(O)n
11
f,g,h
f,g,h
f,g,h
The ‘intrinsic’ SAR of the compounds described herein
in the soluble assay is largely lipophilicity-driven with a
component of ketone electrophilicity also playing a part.
Thus the most polar compounds (13, 18, 20, 23) all show
very poor activity in the soluble assay whereas the most
lipophilic (14, 17, 22, 28) are all highly active. The
influence of ketone electrophilicity may be seen in the
10-fold higher activity of 22 over 1b: although these
compounds share the phenylpentyl group and have
similar c log P’s, the linking ester group of 22 is likely to
render its ketone serine trap more electrophilic than the
corresponding ether-linked ketone of 1b.14 This soluble
assay SAR is consistent with an inhibition mechanism
which involves the lipophilic groups binding in the
arachidonyl-binding pocket of cPLA2, presumably a
highly lipophilic domain, and the ketone functionality
binding the serine nucleophile of the enzyme as a
hemiacetal. The SAR of the whole cell assay is more
O
O
O
Ar
OH
13-16, 24-30
O
Scheme 2. Reagents and conditions: (a) Alkyl halide, Cs2CO3, MeCN,
rt; (b) mCPBA, DCM, rt (for 26 and 27); (c) NH2OHÆHCl, Na2CO3,
EtOH, H2O, 70 °C, 33%; (d) acid chloride, pyridine, reflux; (e) amine,
DCC, THF, reflux; (f) 3, DABCO, DMF, 120 °C; (g) Ac2O, DMSO, rt;
(h) TFA, DCM, rt.
condensation of an alcohol or amine with the carboxyl
group of 5, giving rise to 22 and 23 after oxidation and
hydrolysis (Table 2). Alternatively, the modified alkyl
chain can be attached first to a suitably functionalised
phenol such as 6, 7 or 8 by alkylation or condensation
with appropriate electrophiles or nucleophiles as shown

I. Walters et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3645–3649
Table 1. cPLA2 inhibition and c log P of compounds 1a–c, 12–21
3647
O
O
O
OH
R
O
Compound
R
Soluble assay IC50 (lMa)
HL60 cell assay IC50 (lMb)
cLogPc
7.8
1a
1b
0.03
0.46
2.8
2.6
O
O
S
6.6
1c
12
13
0.08
1.7
5.8
7.2
5.5
4.6
NTd
NTd
O
O
N
34
S
O
O
S
S
14
0.08
1.0
6.3
S
N
15
16
0.76
12
1.1
5.9
6.1
S
O
NTd
N
S
S
N
O
17
0.14
1.1
6.5
O
O
O
18
19
20
5.2
0.59
63
NTd
4.8
4.8
5.8
4.7
O
N
O
O
S
NTd
N
Cl
Cl
O
21
0.25
2.0
5.8
a Inhibition of recombinant cPLA2 measured against a soluble substrate. IC50 values are the means of at least two independent observations. Errors
are within ꢀ20%.
b Inhibition of 3H-arachidonic acid production by stimulated HL60 cells.
c Lipophilicity as calculated by Daylight.13
d Not tested.
Table 2. cPLA2 inhibition and c log P of compounds 22–30
O
X
O
OH
O
Compound
X
Soluble assay IC50 (lMa)
HL60 cell assay IC50 (lMb)
cLogPc
O
22
0.05
1.3
6.8
O
O
(continued on next page)

3648
I. Walters et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3645–3649
Table 2 (continued)
Compound
X
Soluble assay IC50 (lMa)
HL60 cell assay IC50 (lMb)
cLogPc
O
H
N
23
20
NTd
5.1
O
O
O
24
25
26
27
0.18
0.27
1.0
0.76
5.8
6.2
5.5
5.2
N
O
O
0.76
5.2
N
N
O
O
O
S
O O
1.8
18
S
O
O
O
Cl
Cl
28
29
30
N
0.04
0.09
0.56
4.7
2.7
1.0
6.7
5.9
5.5
Cl
Cl
N
N
O
N
Cl
O
O
Cl
O
N
O
a Inhibition of recombinant cPLA2 measured against a soluble substrate. IC50 values are the means of at least two independent observations. Errors
are within 20%.
b Inhibition of 3H-arachidonic acid production by stimulated HL60 cells.
c Lipophilicity as calculated by Daylight.13
d Not tested.
complex: superimposed on the intrinsic lipophilicity
requirement of the enzyme is a tendency for the more
120
lipophilic inhibitors to partition into the cell membranes
of the assay, thus lowering their effective concentration
100
and therefore their apparent activity in the cellular as-
say. Consequently, the ratio of whole cell IC50 to soluble
assay IC50 tends to decrease as the inhibitor c log P is
lowered (Fig. 2), a trend which accounts for the
improvement in cellular potency on going from 28 to 29
to 30 in counterpoint to the accompanying decrease in
soluble assay activity. Clearly an understanding of these
complex, overlaying SAR’s, driven by both true enzyme
binding and lipophilic partitioning, is necessary to de-
sign potent cPLA2 inhibitors of lower lipophilicity.
80
60
40
20
0
5
6
7
8
cLogP
In conclusion, an investigation into the SAR of the
lipophilic alkyl chain of the cPLA2 inhibitors 1 has been
carried out, with the identification of several more ‘drug-
like’ analogues such as 30 which display improved whole
cell potencies against cPLA2 along with lower c log P’s.
Further exploration of the electrophilic ketone serine
Figure 2. Relationship of ratio of HL60 cell assay IC50/soluble assay
IC50 with inhibitor c log P.
trap and the carboxylic acid phosphate mimic will be the
subject of future communications.

I. Walters et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3645–3649
3649
8. Lehr, M. Drugs Future 2000, 25, 823–832.
9. Connolly, S.; Robinson, D. H. Exp. Opin. Ther. Pat. 1995,
5, 673–683.
10. Lehr, M. Exp. Opin. Ther. Pat. 2001, 11, 1123–
1136.
11. Connolly, S.; Bennion, C.; Botterell, S.; Croshaw, P. J.;
Hallam, C.; Hardy, K.; Hartopp, P.; Jackson, C. G.; King,
S. J.; Lawrence, L.; Mete, A.; Murray, D.; Robinson,
D. H.; Smith, G. M.; Stein, L.; Walters, I.; Wells, E.;
Withnall, W. J. J. Med. Chem. 2002, 45, 1348–1362.
12. Huang, Z.; Laliberte, F.; Tremblay, N. M.; Weech, P. K.;
Street, I. P. Anal. Biochem. 1994, 222, 110–115.
13. Daylight, version 3.64; Daylight Chemical Information
Systems, Inc.: Irvine, CA 92715.
References and notes
1. Diez, E.; Louis-Flamberg, P.; Hall, R. H.; Mayer, R. J.
J. Biol. Chem. 1992, 267, 18342–18348.
2. Clark, J. D.; Lin, L.-L.; Kriz, R. W.; Ramesha, C. S.;
Sultzman, L. A.; Lin, A. Y.; Milona, N.; Knopf, J. L. Cell
1991, 65, 1043–1051.
3. Diez, E.; Chilton, F. H.; Stroup, G.; Mayer, R. J.;
Winkler, J. D.; Fonteh, A. N. Biochem. J. 1994, 301, 721–
726.
4. Clark, J. D.; Schievella, A. R.; Nalefski, E. A.; Lin, L.-L.
J. Lipid Mediators Cell Signaling 1995, 12, 83–117.
5. Dennis, E. A. Drug Dev. Res. 1987, 10, 205–220.
6. Moncada, S.; Higgs, E. A. Ann. N. Y. Acad. Sci. 1988, 522,
454–463.
14. As evidenced by the rp values for ether (ꢁ0.28) and ester
(þ0.44) functional groups: March, J. Advanced Organic
Chemistry, 4th ed.; Wiley, 1992, Chapter 9.
7. Samuelsson, B.; Dahlen, S.-E.; Lindgren, J. A.; Rouzer,
C. A.; Serhan, C. N. Science 1987, 237, 1171–1176.