Synthesis and enzyme inhibitory activities of a series of lipidic diamine and aminoalcohol derivatives on cytosolic and secretory phospholipases A2

Article abstract of DOI:10.1016/S0960-894X(99)00680-0

We have synthesised some lipidic diamines and aminoalcohols and examined their behaviour as inhibitors of secretory and cytosolic PLA₂. Some structure-activity relationships considerations have been deduced. Compound 14 was a potent and selective inhibitor of cPLA₂ and compound 4 showed a dual inhibitory profile against both types of PLA₂ while no cytotoxicity at 10 μM on human neutrophils or on murine macrophage line was observed for both. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(99)00680-0

Bioorganic & Medicinal Chemistry Letters 10 (2000) 285±288
Synthesis and Enzyme Inhibitory Activities of a Series of Lipidic
Diamine and Aminoalcohol Derivatives on Cytosolic and Secretory
Phospholipases A2
Rut Lucas, a Amalia Ubeda, a Miguel Paya, a,* Mario Alves, b
Esther del Olmo, b Jose L. Lopez b and Arturo San Feliciano b
a
Â
Departmento de Farmacologõa, Univ. de Valencia, Av. V. Andres Estelles s/n, 46100, Burjassot, Valencia, Spain
Â
Departmento de Quõmica Farmaceutica, Facultad de Farmacia, 37007, Salamanca, Spain
b
Â
Â
Received 16 September 1999; accepted 6 December 1999
AbstractÐWe have synthesised some lipidic diamines and aminoalcohols and examined their behaviour as inhibitors of secretory
and cytosolic PLA2. Some structure±activity relationships considerations have been deduced. Compound 14 was a potent and
selective inhibitor of cPLA2 and compound 4 showed a dual inhibitory pro®le against both types of PLA2 while no cytotoxicity at
10 mM on human neutrophils or on murine macrophage line was observed for both. # 2000 Elsevier Science Ltd. All rights
reserved.
There is evidence that phospholipases A2 (PLA2) play
an important pathophysiological role in various in¯am-
matory diseases, such as septic shock,1 adult respiratory
distress syndrome,2 arthritis3 and acute pancreatitis.4
PLA2 catalyzes the hydrolysis of a fatty acyl ester bond
at the sn-2 position of a glycerophospholipid, being the
enzyme responsible for arachidonic acid release for the
biosynthesis of the eicosanoids, including prostacyclin,
thromboxanes, and other prostaglandins as well as
leukotrienes.5 The other product regulated by PLA2
enzymes is the platelet-activating factor (PAF), which is
also a potent cellular mediator.6 It is known that inhi-
bition of PLA2 can modulate the production of di€erent
in¯ammatory mediators. However, it is not clear which
forms of PLA2 should be inhibited, because PLA2 really
comprises a rapidly growing superfamily of enzymes
like other signal transduction proteins. In this regard,
distinct forms of PLA2 have been characterised.5 Two
main groups have been reported, the secretory PLA2
(sPLA2 groups I, II, III, V, IX and X with a relatively
small molecular weight), and the cytosolic PLA2 (cPLA2
groups IV, VI, VII and VIII with a higher molecular
weight). They also di€er in Ca2+ requirements and
molecular characteristics.
We report the synthesis and the in vitro pharmacologi-
cal evaluation of a series of diamine and aminoalcohol
derivatives sharing a long alkyl chain as the common
feature. Some of these compounds were tested for their
inhibitory e€ect on sPLA2, belonging to the groups I
(Naja naja venom and porcine pancreatic enzymes), II
(human synovial recombinant enzyme) and III (bee
venom enzyme) as well as on cPLA2 from macrophage
line RAW 264.7 (group IV), using manoalide7 and
palmitoyl tri¯uoromethyl ketone (PTK) as reference
inhibitors.
Chemistry
Compounds were designed as not too close structural
analogues of the phospholipidic substrate of PLAs. The
aliphatic chain (R=C14H29), was included to mimic the
sn-1 chain of the substrate and its size was selected to
provide the molecules with an adequate lipophilic
character, which would be modulated by substituents at
positions sn-2 and sn-3. The absence of any hydro-
lysable function at the chain would prevent the unde-
sirable deactivation by chemical or enzymatic factors.
Main functional groups attached to positions 1 and 2 of
the aliphatic long-chain were separated di€erently by
increasing the number of ethylene groups between them
in order to de®ne the best distance for calcium coordi-
nation in the enzyme. Compounds selected for testing
*Corresponding author. Tel.: +34-96-386-4946; +34-96-386-4943;
e-mail: miguel.paya@uv.es
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(99)00680-0

286
R. Lucas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 285±288
were synthesised, among others, as depicted in Scheme
1. The Boc-aminoacid 1, prepared from diethyl 2-acet-
amidomalonate, as reported by Gibbons et al.,8 was
transformed into the Boc-aminoalcohol 2 by the Koko-
tos procedure.9 2, through mesylation, followed by
substitution with sodium azide and reduction with
NaBH4/MeOH, in presence of Pd-C,10 led to the Boc-
diamine 3, which on alkylation/acylation followed (or
not) by Boc-deprotection, yielded the diamine deriva-
tives 4±6 (or 7), respectively.
which were further treated with the appropriate acyl
chlorides to obtain the amidoesters 19, 20 and 21.
Complete details on the synthetic procedures, product
characterisation and yields will be published elsewhere.
New studies relating the resolution of racemic samples
and/or the synthesis of enantiopure compounds will be
done.
Biological Evaluation
The Boc-aminoalcohol 2 was benzylated to obtain the
Boc-aminoether 8, which was Boc-deprotected to give
compound 9. The benzylic aminoether 9, treated with
1.2 and 2.2 equivalents of EtBr, gave 10 and 11, respec-
tively. Acylation of 9 with succinic, maleic and glutaric
anhydrides led to the amidoethers 12, 13 and 14 respec-
tively. Hydrogenolytic debenzylation of 10 and 14, gave
15 and 16 respectively. Alkylation of 9 with ethyl
bromoacetate, followed by debenzylation and hydro-
lysis with 10% KOH/MeOH, yielded the glycine deri-
vative 17.
sPLA2 was assayed by using autoclaved [3H]oleate-
labeled membranes.12 Naja naja venom enzyme, porcine
pancreatic, human recombinant synovial enzyme, and
bee venom enzyme were used. cPLA2 activity was mea-
sured as the release of radiolabeled arachidonic acid13
using cytosolic fractions of murine macrophage cell line
RAW 264.7 as the source of enzyme. The mitochondrial
dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) to formazan14 was
used to assess the possible cytotoxic e€ects of test com-
pounds on human neutrophils as well as on murine
macrophage line.
The Boc-aminoalcohol 2 was treated with heptanoyl
chloride and palmitoyl chloride to give the esters 18a
and 18b, respectively. The attempted Boc-deprotection
of these compounds with TMSCl/PhOH11 in HCCl3,
followed by puri®cation on silica gel, led to the intra-
molecularly transamidated compounds 18c and 18d,
Results and Discussion
Inhibition of speci®c PLA2s constitutes a potentially use-
ful approach for treating a great variety of in¯ammatory
Scheme 1. Synthesis of lipidic diamine and aminoalcohol derivatives. i: a) EtOOCCl/N-methylmorpholine/THF; b) NaBH4/MeOH. ii) MsCl/Et3N/
CH2Cl2. iii. NaN3/DMF. iv. Pd-C/HCCl3, NaBH4/MeOH. v: EtBr (or EtOOCCH2Br)/Et3N/DMF. vi: glutaric anhydride/EtOAc. vii: HCl/THF,
Argon. viii: BnCl/NaH/DMF. ix: EtBr (1.2 or 2.2 eq.)/Et3N/DMF. x: succinic (maleic, glutaric) anh./EtOAc. xi: EtOOCCH2Br/Et3N. xii: H2/Pd-C/
AcOH. xiii: 10% KOH/MmeOH. xiv: n-C6H13COCl (n-C15H31COCl)/Et2O. xv: TMSCl/PhOH/HCCl3.

R. Lucas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 285±288
287
Table 1. E€ect of lipidic diamine and aminoalcohol derivatives on cytosolic PLA2 activitya
Compound
R1
R2
cPLA2%I (100 mM)b
IC50c (mM)
4
5
6
7
-NEt2
-N(CH2COOEt)2
-NHCO(CH2)3COOH
-NHCH2COOEt
-OBn
-NH2
-NH2
-NH2
-NHBoc
-NHEt
-NEt2
90.6‹5.7**d
43.9‹10.9**
0.0‹0.0
0.0‹0.0
0.0‹0.0
1.9‹1.9
72.0‹3.5**
60.7‹6.1**
84.8‹9.6**
25.2‹3.8**
0.0‹0.0
25.0‹3.7**
2.3‹1.2
0.5‹0.5
11.9
NDe
ND
ND
ND
ND
62.8
69.5
10.5
ND
ND
ND
ND
ND
ND
11.7
10
11
12
13
14
15
16
17
19
20
21
PTK
-OBn
-OBn
-OBn
-OBn
-OH
-OH
-OH
-NHCO(CH2)2COOH
-NHCOCHˆCHCOOH(E)
-NHCO(CH2)3COOH
-NHEt
-NHCO(CH2)3COOH
-NHCH2COOH
-NHCO(CH2)5CH3
-NHCO(CH2)13CH3
-NHCO(CH2)13CH3
-OCO(CH2)5CH3
-OCO(CH2)5CH3
-OCO(CH2)13CH3
5.1‹3.3
67.7‹3.1**
CH3(CH2)14COCF3
aMean ‹SEM (n=6).
b%I=% inhibitory e€ect of compounds at 100 mM.
cIC50 values were determined for those compounds that reached 50% of inhibition at 100 mM.
d** P<0.01.
eND=not determined.
Table 2. E€ect of compounds 4 and 15 on di€erent secretory PLA2 activitiesa
Compound
Naja naja venom
Pancreas
%I (10 mM)
Human synovial
Bee venom
c
%I (10 mM)b
IC50
IC50
%I (10 mM)
IC50
%I (10 mM)
IC50
4
15
51.5‹3.0**e
62.6‹3.2**
17.0‹1.7*d
9.5
6.2
NDf
68.5‹4.9**
81.9‹5.1**
32.3‹2.7**
3.6
1.4
ND
82.7‹1.6**
80.0‹1.3**
93.2‹0.2**
5.6
3.7
3.9
57.6‹3.9**
63.5‹3.6**
62.5‹3.8**
11.0
6.1
7.5
Manoalide
aMean‹SEM (n=6).
b%I=% inhibitory e€ect of compounds at 10 mM.
cIC50 values (mM) were determined for those compounds that reached 50% of inhibition at 10 mM.
d*P<0.05.
e**P<0.01.
fND=not determined.
disorders. Unfortunately, no potent and absolutely
type-speci®c PLA2 inhibitors are widely available to
investigators. Non-speci®c covalent-modifying PLA2
drugs such as manoalide7 have received much attention
as sPLA2 inhibitors. On the other hand, owing to the
central role of cPLA2 in arachidonic acid signalling,
design of cPLA2 inhibitors has recently been an area of
great interest. In this group, palmitoyl tri¯uoromethyl
ketone (PTK) is the most well known cPLA2 inhibitor.15
Some long-chain 3-amino-1,2-diols have been reported
to exert anti-in¯ammatory properties16 but they have
not been evaluated against di€erent PLA2s. Herein we
report some structure±activity relationship considera-
tions that can be deduced from the in vitro pharmaco-
logical study of these new diamine and aminoalcohol
derivatives (Tables 1 and 2). Several compounds of
these series displayed potent inhibitory activities and, in
some cases, even higher than those of reference com-
pounds PTK (for cPLA2) and manoalide (for sPLA2).
In addition, none of them presents any sign of cyto-
toxicity at 10 mM (data not shown).
Looking at data in Table 1, particular stucture±activity
relationships observations can be made: (a) both series
have representative compounds able to inhibit exten-
sively cPLA2 at the 100 mM level, the diamine 4 and the
benzylated amidoacid 14 being the most potent for each
series; (b) compounds in which both diamino or ami-
noalcohol groups were acylated/alkylated were inactive
(7) or practically inactive (19, 20, 21), probably due to
their excessive lipophilicity or low solubility; (c) com-
pounds having both extreme functions free of any alkyl
or acyl substituent and separated by eight bonds (6 and
16) were inactive, while if the distance was only ®ve
bonds (17), a 25% inhibition was observed, thus indi-
cating a fair structure±activity dependence; (d) com-
pounds having only one of the extreme functions
blocked by alkylation or acylation are good inhibitors,

288
R. Lucas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 285±288
independently of the position, 1 or 2, of the blocked
function in the long chain, thus informing about the
minor relevance of the location of the long chain in the
molecule. A relevant comparison illustrating the above
statements involves the closely related compounds 14
and 16, for which the presence or absence of the benzyl
moiety is de®nitive for the activity. This behaviour
seems parallel to that of compounds 4 or 5 in relation
with the aminoacid 6 within the diamine family. (e)
Simple diethylation of the amino group at C-1 (4),
seems to be the most bene®cial transformation of dia-
mines for activity, because the introduction of larger
substituents (bis-ethoxycarbonylmethyl, glycinyl ester
residue, 5) at this part of the molecule decreases the
potency substantially.
secretory PLA2. Based on these results, two compounds
have been selected for extensive preclinical evaluation.
Acknowledgements
This work was supported by grants SAF98-0119 and
PB-93-0608 from C.I.C.Y.T. and D.G.Y.C.Y.T., Spain.
R. L. was the recipient of a scholarship from Spanish
Ministerio de Educacion y Cultura. The collaborative
work was performed under the auspices of the CYTED
Program (Programa Iberoamericano de Ciencia y Tec-
nologõa para el Desarrollo), subprogram X.
On the other hand, the presence of a larger acyl side
chain having an ionisable free carboxylic group at its
end (6), provoked the complete disappearance of the
inhibitory activity. (f) Relating to the most potent group
of aminoalcohol derivatives (R1=-OBn), it seems clear
that there exists an optimum size, of ®ve or more carbon
atoms length, for the acyl chain and that the introduction
of a trans double bond decreases the inhibitory potency.
References
1. Vadas, P. Biochim. Biophys. Acta 1997, 1346, 193.
2. Kaiser, E.; Chiba, P.; Zaky, K. Clin. Biochem. 1990, 23, 349.
3. Bomalaski, J. S.; Clark, M. A. Arthritis Rheum. 1993, 36,
190.
4. Buchler, M.; Malfertheiner, P.; Schadlich, H.; Nevalainen,
T. J.; Friess, H.; Berger, H. G. Gastroenterology 1989, 97,
1521.
Evaluation of the diamine 4 and the aminoalcohol 15 in
their ability to inhibit four di€erent classes of secretory
PLA2 (Table 2) showed that the former compound dis-
played slightly better IC50 values than that found for
cPLA2 but did not show any appreciable selectivity. In
contrast, in the case of compound 15, the inhibitory
potency for sPLA2 was much stronger (IC50=1.4±6.2
mM) than for cPLA2 ( 200 mM), thus revealing a fair
selectivity, ranging between 30 to 140 times, for the
preferential inhibition of the sPLA2s.
5. Balsinde, J.; Balboa, M. A.; Insel, A.; Dennis, E. A. Annu.
Rev. Pharmacol. Toxicol. 1999, 39, 175.
6. Chao, W.; Olson, M. S. Biochem J. 1993, 292, 617.
7. Soriente, A.; De Rosa, M.; Scettri, A.; Sodano, G.; Ter-
encio, M. C.; Paya, M.; Alcaraz, M. J. Curr. Med. Chem. 1999,
6, 415.
8. Gibbons, W. A., Hughes, R. A., Charalanibous, M., Chris-
todoulou, M., Szeto, A., Aulabauh, A. E., Mascagni, P.; Toth,
I. Liebigs Ann. Chem. 1990, 1175.
9. Kokotos, G. Synthesis 1990, 190.
10. Constantinou-Kokotou, V.; Kokotos, G.; Roussakis, C.
Anticancer Res. 1998, 18, 3439.
Evaluations will be extended to other members of the
series for de®ning more precisely the new series of com-
pounds to be prepared and tested for optimising the
inhibitory response. Furthermore, taking into account
that all the compounds tested were racemic mixtures,
stronger and/or more selective inhibitions can be expec-
ted for one of the enantiomers of each pair, and addi-
tional work in this sense will be done. In conclusion, we
have developed a novel series of lipidic diamine and
aminoalcohol derivatives with a PLA2 inhibitor pro®le,
showing potency and selectivity towards cytosolic and
11. Kaiser, E.; Tam, J. P.; Kubiak, T. M.; Merri®eld, R. B.
Tetrahedron Lett. 1988, 29, 303.
12. Franson, R.; Patriarca, P.; Elsbach, P. J. Lipid Res. 1974,
15, 380.
13. Clark, J. D.; Milona, N.; Knopf, J. L. Proc. Natl. Acad.
Sci. USA 1990, 87, 7708.
14. Gross, S. S.; Levi, R. J. Biol. Chem. 1992, 267, 25722.
15. Ackermann, E. J.; Conde-Frieboes, K.; Dennis, E. A. J.
Biol. Chem. 1995, 270, 445.
16. Padron, J. M.; Martin, V. S.; Hadjipavlou-Litina, D.;
Noula, C.; Constantinou-Kokotou, V.; Peters, G. J.; Kokotos,
G. Bioorg. Med. Chem. Lett. 1999, 9, 821.