2-Aminohydroxamic acid derivatives as inhibitors of Bacillus cereus phosphatidylcholine preferred phospholipase C PC-PLCBc

Article abstract of DOI:10.1016/j.bmc.2010.10.031

Phosphatidylcholine preferring phospholipase C (PC-PLC) is an important enzyme that plays a key role in a variety of cellular events and lipid homoeostases. Bacillus cereus phospholipase C (PC-PLC_Bc) has antigenic similarity with the elusive mammalian PC-PLC, which has not thus far been isolated and purified. Therefore the discovery of inhibitors of PC-PLC _Bc is of current interest. Here, we describe the synthesis and biological evaluation of a new type of compounds inhibiting PC-PLC_Bc. These compounds have been designed by evolution of previously described 2-aminohydroxamic acid PC-PLC_Bc inhibitors that block the enzyme by coordination of the zinc active site atoms present in PC-PLC_Bc [Gonzalez-Roura, A.; Navarro, I.; Delgado, A.; Llebaria, A.; Casas, J. Angew. Chem. Int. Ed. 2004, 43, 862]. The new compounds maintain the zinc coordinating groups and possess an extra trimethylammonium function, linked to the hydroxyamide nitrogen by an alkyl chain, which is expected to mimic the trimethylammonium group of the phosphatidylcholine PC-PLC_Bc substrates. Some of the compounds described inhibit the enzyme with IC ₅₀'s in the low micromolar range. Unexpectedly, the most potent inhibitors found are those that possess a trimethylammonium group but have chemically blocked the zinc coordinating functionalities. The results obtained suggest that PC-PLC_Bc inhibition is not due to the interaction of compounds with the phospholipase catalytic zinc atoms, but rather results from the inhibitor cationic group recognition by the PC-PLC_Bc amino acids involved in choline lipid binding.

Full text of DOI:10.1016/j.bmc.2010.10.031

Bioorganic & Medicinal Chemistry 18 (2010) 8549–8555
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
2-Aminohydroxamic acid derivatives as inhibitors of Bacillus cereus
phosphatidylcholine preferred phospholipase C PC-PLCBc
Patricia González-Bulnes a, , Albert González-Roura a, , Daniel Canals a, Antonio Delgado a,b, Josefina Casas a,
Amadeu Llebaria a,
⇑
a Research Unit on BioActive Molecules (RUBAM), Department of Biomedicinal Chemistry, Institute of Advance Chemistry of Catalonia (IQAC-CSIC), Jordi Girona 18-26,
08034 Barcelona, Spain
b Universitat de Barcelona, Facultat de Farmàcia, Unitat de Química Farmacèutica (Associada al CSIC), Avda. Joan XXIII, s/n, 08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Phosphatidylcholine preferring phospholipase C (PC-PLC) is an important enzyme that plays a key role in
a variety of cellular events and lipid homoeostases. Bacillus cereus phospholipase C (PC-PLCBc) has anti-
genic similarity with the elusive mammalian PC-PLC, which has not thus far been isolated and purified.
Therefore the discovery of inhibitors of PC-PLCBc is of current interest. Here, we describe the synthesis
and biological evaluation of a new type of compounds inhibiting PC-PLCBc. These compounds have been
designed by evolution of previously described 2-aminohydroxamic acid PC-PLCBc inhibitors that block the
enzyme by coordination of the zinc active site atoms present in PC-PLCBc [Gonzalez-Roura, A.; Navarro, I.;
Delgado, A.; Llebaria, A.; Casas, J. Angew. Chem. Int. Ed. 2004, 43, 862]. The new compounds maintain the
zinc coordinating groups and possess an extra trimethylammonium function, linked to the hydroxyamide
nitrogen by an alkyl chain, which is expected to mimic the trimethylammonium group of the phospha-
tidylcholine PC-PLCBc substrates. Some of the compounds described inhibit the enzyme with IC50’s in the
low micromolar range. Unexpectedly, the most potent inhibitors found are those that possess a trimethy-
lammonium group but have chemically blocked the zinc coordinating functionalities. The results
obtained suggest that PC-PLCBc inhibition is not due to the interaction of compounds with the phospho-
lipase catalytic zinc atoms, but rather results from the inhibitor cationic group recognition by the PC-
PLCBc amino acids involved in choline lipid binding.
Received 27 April 2010
Revised 14 September 2010
Accepted 12 October 2010
Available online 20 October 2010
Keywords:
Phospholipase C
Bacillus cereus
2-Aminohydroxamic acids
Trimethylammonium
Choline lipids
Inhibitors
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
PLCBc) had an antigenic similarity with mammalian PC-PLC21
prompted the use of this well known bacterial PLC, as a model
Phospholipases C (PLC) constitute a class of lipid phosphohy-
drolase enzymes that are involved in the hydrolysis of the phos-
phodiester bond in phospholipids to provide a lipidic alcohol and
a phosphorylated polar head group. Phospholipases play a key role
in different biological processes, such as cell membrane homoeos-
tases,1 digestion,2,3 inflammation,4–6 infection7,8 and signal trans-
duction.9–11 In mammals the products released by the action of
phosphatidylcholine-specific phospholipase C (PC-PLC), diacylglyc-
erol and phosphocholine, are involved in cell function and signal-
ling.12–16 Just to cite a few examples, PC-PLC is involved in
hepatocarcinogenesis,17 ovarian tumour progression18 or leukae-
mia.19 In contrast to phosphatidylinositol phospholipases C,20
mammalian PC-PLCs are poorly characterised, and no eukaryotic
PC-PLC has thus far been isolated and purified. The discovery that
Bacillus cereus phosphatidylcholine-specific phospholipase C (PC-
for its mammalian counterpart. Therefore, the discovery of inhibi-
tors of PC-PLCBc opens up a way to the development of mammalian
PC-PLC inhibitors and their use as pharmacological tools for the
study of this enzyme.
PC-PLCBc is a small, monomeric enzyme, with three Zn2+ atoms
in the active site likely to be involved in the binding to the sub-
strate22–24 and essential for the enzymatic activity and protein
conformational stability.25 Chemical modifications of a specific
carboxyl group or histidine, lysine and arginine residues result in
enzyme inactivation,26–30 this indicating they are essential resi-
dues for catalytic activity. Comparison of the structures of PC-PLCBc
variants with that of the wild-type enzyme suggests that minor
changes in steric and electronic properties in the binding site of
PC-PLCBc are responsible for significant changes in substrate
selectivity.31
To date, several inhibitors of PC-PLCBc have been identified.
Among them, D609, occupies a prominent place. This potent com-
pound inhibits the activity of PC-PLCBc while not affecting the
activities of phospholipase D and phosphatidylinositol-specific
⇑
Corresponding author. Tel.: +34 934 006 108; fax: +34 932 045 904.
E-mail address: amadeu.llebaria@cid.csic.es (A. Llebaria).
These authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.10.031

8550
P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
phospholipase C. The synthesis in our group of all four diastereo-
mers of D609 and the lack of significant differences in their inhib-
itory activity suggests an absence of diastereochemical control of
the enzyme by xanthate inhibitors.32 Besides its PC-PLCBc inhibi-
tory activity, D609 has been reported as inhibitor of other enzymes
such as sphingomyelin synthase, acidic sphingomyelinase and a
group IV cytosolic phospholipase A2.33 As a consequence, the
effects of this inhibitor on PC-PLCBc are far from being selective.
Moreover, D609 is a potent antioxidant34 and is able to modulate
sphingolipid transport.35 With this plethora of options, the identi-
fication of cellular targets involved in D609 activities remains ob-
scure. Zinc-chelating molecules like univalent anions,36 Tris
buffer23 or cyclic N,N0-hydroxyureas37 have been found to inhibit
PC-PLCBc as well. Their interaction with the three metal ions in
the protein molecule22,23,36 prevents enzymatic activity, a fact that
confirms the importance of the metal ions in the catalytic mecha-
nism. Phosphate and phosphocholine analogues constitute another
group of PC-PLCBc inhibitors; all of them are compounds able to
prevent the catalytic activity by interaction with the active site
of the enzyme.30,38–40
The N-substituted hydroxylamines to be coupled with the ami-
no acids were obtained following the synthetic pathway depicted
in Scheme 2. O-Benzylhydroxylamine was reacted with di-tert-
butyldicarbonate in alkaline medium to give the N-Boc protected
derivative 8. Treatment of compound 8 with NaH, followed by
reaction of the resulting anion with 1,2-dichloroethane or 1,2-
dichloropropane resulted in nitrogen alkylation to give carbamates
9a and 9b. The next step was the SN2 substitution of the terminal
chlorine with dimethylamine as nucleophile in a reaction that fur-
nished diamines 10a and 10b. Deprotection of compounds 10a and
10b under acidic conditions gave the target O-benzyl protected
hydroxylamines 11a and 11b.
Compounds 11a and 11b were reacted with the required amino
acids using EDC/HOBt standard coupling conditions in moderate
yields. For compounds with a short alkyl tether chain in the choline
group, the tertiary amine was converted into a quaternary ammo-
nium salt by treatment with MeI. Subsequent deprotection of the
hydroxyl and amino groups gave
a-aminohydroxamic acids 24–
27 (Scheme 3).
The synthesis of the second group of analogues, those with a
longer alkyl tether chain in the choline group, was carried out as
described in Scheme 4. Only 2-aminooctanoic acid and to 2-amin-
ododecanoic acid derivatives were prepared in this series, since
biological studies performed on compounds 16–27 showed a clear
correlation between the alkyl chain length and cytotoxicity in
mammalian cells (Table 3). Moreover, to determine the importance
of the presence of a quaternary ammonium group for the inhibi-
tory activity, two different classes of compounds were synthesized,
one having a dimethylamino substituent in the polar head (com-
pounds 28–33) and the other having a trimethylammonium group
(compounds 34–39). The synthetic approach, depicted in Scheme 4,
was similar to the above described, involving the coupling of N-Boc
amino acids 1 and 5 to diamine 11b, to give intermediates 28 and
29, followed by amine methylation, when appropriate, and final N-
Boc and O-benzyl deprotection steps.
The strategy followed by our group to design PC-PLCBc inhibi-
tors was based on the structural similarities between PC-PLCBc
and other zinc enzymes.41 The comparison of the structures Strep-
tomyces griseus and Aeromonas proteolytica aminopeptidases with
that of PC-PLCBc showed a notable structural similarity, especially
in the relative disposition of catalytic zinc atoms involved in the
hydrolytic reactions promoted by these enzymes. Starting from
known inhibitors of the aforementioned aminopeptidases, we de-
signed and synthesized a small series of
a-aminohydroxamic acids
and -aminophosphonic acids and studied their activity as phos-
a
pholipase inhibitors, to know if the similarities in the tridimen-
sional active sites of the enzymes could also be extended to
these inhibitors. This approach was successful, and a-aminohydr-
oxamic acids, initially reported as aminopeptidase inhibitors, were
found to be also potent PC-PLCBc inhibitors.41
In this article we describe the modifications introduced in the 2-
aminohydroxamic scaffold to improve the inhibitor properties on
PC-PLCBc. The X-ray analysis of the complex of PC-PLCBc and a phos-
phonate inhibitor showed that the trimethylammonium group,
corresponding to the phosphonate polar head, interacts with cer-
tain active site amino acid residues that are important for an effec-
tive binding with the protein.23 With the aim of achieving a
stronger and more selective inhibitor–enzyme interaction, a cho-
line-like group has been added to the hydroxyamide nitrogen of
Whereas the removal of N-Boc groups under acidic conditions
in the 2-aminohydroxamic derivatives was successful, the O-ben-
zyl deprotection proved problematic and required considerable
experimentation. For compounds having a dimethylamino substi-
tuent, the hydrogenolysis worked nicely with Pd/C catalyst at room
temperature and 1 atm H2. In contrast, debenzylation of com-
pounds having quaternary trimethylammonium groups under the
above conditions failed. We attempted first to deprotect the N-
Boc derivatives, but the hydrogenolysis to obtain the N-Boc 2-
aminohydroxamic acids failed in all cases with different palladium
catalysts (Pd/C, Pd(OH)2/C, Pd/BaSO4) obtained from different com-
mercial sources and different batches. Several reaction tempera-
tures (from rt to 80 °C) and hydrogen pressure (1–3 atm) and
different reaction times (up to 5 days) were attempted. In general,
no reaction was observed at low temperature and/or pressure con-
ditions. When forcing hydrogenation conditions we observed mix-
tures of the starting material, accompanied by variable amounts of
the desired N-hydroxyamide and the amide resulting from reduc-
tion of the N–O bond. These types of compounds predominate
when reaction times were extended to complete the consumption
of the O-benzyl starting materials. This side reaction is described in
literature45, in a related hydrogenolysis of a O-benzyl hydroxamate
in cyclic 3-amino-1-hydroxypyrrolidin-2-one systems. In this kind
of molecules a fine tuning of the reactivity of the substrate and cat-
alyst is required for successful hydrogenolysis of the benzyl group
without overreduction of the hydroxyamide to amide. These prece-
dents prompted us to try the debenzylation of the corresponding
the above
a-aminohydroxamic acids. Since a-aminohydroxamic
PC-PLCBc inhibition was found to be non-enantioselective,41 we
decided to employ racemic amino acids in this study. Further anal-
ysis of the binding model proposed by Martin et al.30 for the en-
zyme–substrate interaction suggested that the binding of the
inhibitor to the enzyme active site might be modulated by the
length of the alkyl tether chain between the trimethylammonium
group and the hydroxyamide nitrogen of the polar head (Fig. 1).
So, two series of analogues, containing two or three methylene
units, were obtained (Fig. 1) and tested as inhibitors of PC-PLCBc
.
2. Results
2.1. Synthesis of compounds
The target molecules were obtained coupling an N-Boc pro-
tected n-alkyl-2-amino acid with a N-alkylated hydroxylamine.
Amino acid 1 was obtained by N-Boc protection of commercial 2-
aminooctanoic acid, whereas longer alkyl amino acids 2–4 were
synthesized following classical procedures described in the litera-
ture,42–44 followed by amine protection by treatment with di-
tert-butyldicarbonate to give intermediates 5–7 (Scheme 1).
free
removal.
This strategy proved effective and compounds 20 and 21
(Scheme 3) were debenzylated with 10% Pd/BaSO4 catalyst
a-ammonium compounds arising from previous N-Boc

P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
8551
Figure 1. Binding model for enzyme–substrate interaction [adapted from Martin and co-worker52] and for enzyme–2-aminohydroxamic inhibitor interaction.
Scheme 1. 2-Amino acid synthesis. Reagents, conditions and yields: (a) (tBuOCO)2O, TEA/CH2Cl2, 80%; (b) Br(CH2)9CH3 (for 2), Br(CH2)13CH3 (for 3), Br(CH2)17CH3 (for 4)
NaOEt/EtOH, reflux; then HCl, reflux; (c) (tBuOCO)2O, NaOH/dioxane (59% for 5, 49% for 6, 33% for 7).
Scheme 2. Synthesis of compounds 11a and 11b. Reagents, conditions and yields: (a) NaHCO3, (tBuOCO)2O, H2O/CH2Cl2; (b) n = 1: 1,2-dichloroethane, NaH/DMF, 77% global
yield; n = 2: 1,3-dichloropropane, NaH/DMF, 70%; (c) dimethylamine, NaI, EtOH, 80 °C, n = 1: 78%; n = 2: 74%; (d) HCl, MeOH, n = 1: 99%; n = 2: 100%.
(3 atm H2, room temperature, 24 h) to give excellent yields of the
corresponding hydroxamic acids 24 and 25. However, for the long-
er alkyl derivatives 22 and 23 this catalyst was not operative and
the more active Pearlman’s catalyst was required to obtain the cor-
responding hydroxamic acids 26 and 27 in acceptable to good
yields. For the series having three methylene units in the polar
head, the O-benzyl derivatives were deprotected in a very similar
way (Scheme 4). As in the shorter derivatives, the reactivity of

8552
P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
Scheme 3. Synthesis of compounds 20–23. Reagents, conditions and yields: (a) EDC/HOBt, TEA/CH2Cl2, 60 °C, 60 h; (b) EDC/HOBt, TEA/THF, 60 °C, 60 h (43% for 12, 31% for 13,
30% for 14, 31% for 15); (c) MeI, CH2Cl2, 24 h, rt (99% for 16, 99% for 17, 99% for 18, 100% for 19); (d) HCl, MeOH, 2 h, 0 °C (99% for 20, 99% for 21, 99% for 22, 100% for 23); (e)
10% Pd/BaSO4, 3 atm H2, MeOH, 24 h, rt (99% for 24, 98% for 25); (f) 20% Pd(OH)2/C, 3 atm H2, MeOH, 24 h, rt (55% for 26, 88% for 27).
Scheme 4. Synthesis of compounds 28–39. Reagents, conditions and yields: (a) EDC, HOBt, TEA/THF, 60 h, 60 °C [45% for 28, 56% for 29]; (b) HCl, MeOH, 2 h, 0 °C (96% for 30,
99% for 31); (c) H2, Pd/C, 1 atm, 24 h (99% for 32, 88% for 33); (d) MeI, CH2Cl2, 18 h, rt (99% for 34, 99% for 35); (e) HCl, MeOH, 2 h, 0 °C (99% for 36, 99% for 37); (f) H2, Pd(OH)2,
3 atm, 24 h (40% for 38, 52% for 39).
the dimethylamino derivatives was higher than that of the trime-
thylammonium compounds. Thus, 30 and 31 were converted into
hydroxamic acids 32 and 33 with Pd/C and 1 atm of H2 in excellent
yields. Trimethylammonium compounds 36 and 37 were

P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
8553
unreactive under these conditions but could be deprotected under
3 atm of H2 using Pd(OH)2/C as catalyst, to give 38 and 39 in 40%
and 52% yields, respectively. In contrast with the related com-
pounds with two methylene units, the use of Pd/BaSO4 as catalyst
for O-debenzylation in 36 and 37 was ineffective.
modes of the new inhibitors, some Ki values were determined.
Compound 25, with a trimethylammonium group and a two meth-
ylene bridge, compounds 35 and 39, with a trimethylammonium
group and a three methylene bridge and compound 29, with a di-
methyl amino group and a three methylene bridge were chosen.
The results obtained showed that there was not a common pat-
2.2. Inhibition of PC-PLCBc
tern of inhibition. Thus, compound 25 (Ki, 12
an uncompetitive inhibitor whereas compounds 35 (Ki, 2.5
and 39 (Ki, 26 M) showed a mixed type inhibition and compound
29 (Ki, 2.5 M) was a non-competitive inhibitor. Analysis of the re-
l
M) appeared to be
lM)
The PC-PLCBc activity of final products and intermediates was
measured according to the method described by Hergenrother
and Martin46 which is based on the quantification of phosphate
ions arising from alkaline phosphatase (AP) catalysed hydrolysis
of phosphorylcholine produced by PC-PLCBc. Given that AP is a
Zn-phosphoesterase and to avoid possible interferences in the as-
say, we studied the effect of the tested compounds on AP, conclud-
ing that its activity was not affected at 0.5 mM (data not shown).
Since PC-PLCBc exhibits greatly increased activity on a micellar
environment,47 the assay is best conducted below the critical mi-
celle concentration (CMC) of amphiphilic molecules present. The
substrate used in this case was 1,2-di-n-hexanoyl-sn-glycero-3-
phosphocholine, at 2 mM final concentration, well below its CMC
l
l
sults obtained indicates that, for aminohydroxamic compounds 25
and 29 that have a trimethylammonium group, the introduction of
an extra methylene between the trimethylammonium group and
the hydroxyamide nitrogen caused a change in the inhibition type.
Whereas 25 was an uncompetitive inhibitor, compound 39 at low
substrate concentrations was also uncompetitive, while at high
substrate concentrations they behaves as a competitive inhibitor.
A similar mixed inhibition mode was obtained for compound 35,
which has the amino and hydroxamate functionalities blocked.
The related compound 29 which posses a dimethylamino group
was found to be a PC-PLCBc non-competitive inhibitor.
of 11.1 mM. We also calculated CMC values for the
a
-aminohydr-
To rule out that PC-PLCBc inhibition was originated from protein
denaturalisation by the inhibitors, compound 18 and octadecyl-2-
aminohydroxamic acid, one of our previously described PC-PLCBc
inhibitor lacking the choline-like group,41 were independently
incubated in the presence of PC-PLCBc at a concentration in which
the enzymatic inhibition was around 80% of the untreated enzyme.
After 10 min, the medium was diluted to reach an inactive inhibi-
tor concentration and the enzymatic activity was determined. In
both cases the enzymatic activity was around 100% (data not
shown). This result clearly indicates that these compounds do
not affect irreversibly the enzyme activity, excluding an unspecific
inhibition due to protein denaturation by the inhibitors.
oxamic acid derivatives, to ensure that the enzymatic assay was
performed under non-micellar conditions. As expected, CMC val-
ues decreased as long chain lipophilicity increased (Tables 1 and
2). As a general trend, for the same R substituent, CMC values
increased as protective groups were removed and dimethylammo-
nium derivatives exhibited higher CMC than the corresponding
trimethylammonium analogues.
As shown in Table 1, the compounds obtained from 2-aminooc-
tanoic acid were weak inhibitors of PC-PLCBc, since three out of the
six compounds resulted to be inactive and the rest exhibited IC50’s
around 50–100 lM. The inhibitory activities found for the remain-
der members of the trimethylammonium derivative families ran-
ged the low micromolar IC50’s and exhibited two remarkable
trends. Inhibitory activity increased roughly with the alkyl chain
length and, for a given alkylamino acid derivative, the compounds
having N-Boc and O-benzyl protecting groups were among the
more active ones. This unexpected result implies that the zinc
coordinating functionalities are not required for inhibition. Similar
results were obtained for compounds with a three methylene
bridge (n = 2) and a dimethylamino in the polar head. A higher
inhibitory activity was observed for compounds with a longer alkyl
chain (Table 2). Comparison of the results depicted in Tables 1 and
2 show very similar activities for inhibitors with a two (n = 1) and
three (n = 2) methylene bridges.
Next, we checked the effect of the compounds on cell growth
using the MTT test. Regarding trimethylammonium derivatives
(Table 3), none of the compounds with a 6-carbon alkyl chain
was found out to be toxic. Longer alkyl chains were more toxic
and their toxicity decreased as the protective groups were re-
moved (see LD50 values for 18, 22 and 26, for example). Dimethyl-
amino compounds (Table 4) exhibited lower LD50 values than the
corresponding trimethylammonium analogues.
3. Discussion
Our initial objective was to improve the PC-PLCBc inhibitory
activity of a family of new a-aminohydroxamic PC-PLCBc inhibitors
we had previously disclosed41 by adding to these structures a cho-
line like group. These inhibitors were found among compounds
We previously had reported that
a-aminohydroxamic acids
41
competitively inhibited PC-PLCBc
.
To know about the inhibition
Table 1
Inhibition (IC50 values,
lM) of PC-PLCBc by 2-aminohydroxamic acid derivatives
R
n
Compd
CMC
IC50
Compd
CMC
IC50
Compd
CMC
IC50
500
a
a
b
a
n-C6H13
1
2
1
2
1
1
16
34
17
35
18
19
—
56
78
5
4
2
20
36
21
37
22
23
—
—
24
38
25
39
26
27
—
—
—
a
b
221
210
100
10
50
4
—
a
a
a
n-C10H21
—
—
40
58
2
40
141
90
10
97
120
50
a
n-C14H29
n-C18H37
—
75
3
4
4
a
CMC value higher than 500
l
M.
M.
b
IC50 value higher than 500
l

8554
P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
Table 2
Inhibition (IC50 values,
l
M) of PC-PLCBc by 2-aminohydroxamic acid derivatives with a dimethylamino group
R
Compd
CMC
IC50
Compd
CMC
IC50
Compd
CMC
IC50
b
a
b
C6H13
C10H21
28
29
75
10
77
8
30
31
158
8
—
32
33
—
—
32
79
100
a
CMC value higher than 500
l
M.
M.
b
IC50 value higher than 500
l
Table 3
Cytotoxicity of 2-aminohydroxamic acid derivatives
R
n
Compd
LD50
n
Compd
LD50
n
Compd
LD50
a
a
a
n-C6H13
1
2
1
2
1
1
16
34
17
35
18
19
—
—
1
2
1
2
1
1
20
36
21
37
22
23
—
—
1
2
1
2
1
1
24
38
25
39
26
27
—
—
—
—
a
a
a
a
n-C10H21
28.2
17.0
27.6
25.1
76.1
43.4
75.3
29.0
a
n-C14H29
n-C18H37
100
52.7
a
LD50 value higher than 100 lM.
Table 4
Cytotoxicity of 2-aminohydroxamic acid derivatives
R
Compd
LD50
Compd
LD50
Compd
LD50
a
a
C6H13
C10H21
28
29
21
6
30
31
—
7
32
33
—
40
a
LD50 value higher than 100 lM.
that inhibited enzymes with active sites having structural similar-
ity with PC-PLCBc. Inhibitor modifications, derived from the analy-
sis of the complex of PC-PLCBc and a phosphonate inhibitor,48 and a
binding model of the hydroxamic compounds led us to introduce
alkylamine substituents in the hydroxyamide nitrogen to mimic
the choline present in the phosphatidylcholine PC-PLCBc substrate.
The series obtained contained an ethyltrimethylammonium group
and resulted in compounds 16–27. Study of their activity on PC-
PLCBc showed that they were potent inhibitors with IC50’s in the
philic chain were essential for potent inhibitory activity. However,
we found that final compounds having the zinc coordinating func-
tionalities (the amino group and/or the N-hydroxyl group), such as
24, 25, 32, 33 and 38, were less active than their N-Boc and/or O-
benzyl synthetic precursors, while others, like compounds 26, 27
and 39, had activities similar but not higher than those of protected
synthetic intermediates. Rather unexpectedly, the inhibitor zinc
coordination, initially thought to be essential for the inhibitory
activity, can be avoided provided that a charged trimethylammoni-
um (or even dimethylammonium) group is present in the lipophilic
chain presumably to mimic the choline group of PC-PLCBc sub-
strates. To confirm this, we tested hexadecyltrimethylammonium
chloride (cetrimide) in the phospholipase assay and we confirmed
that this was certainly a potent PC-PLCBc inhibitor with a IC50 of
low micromolar range, similar to the
a-aminohydroxamic inhibi-
tors. The importance of the interaction between the trimethylam-
monium group and certain residues of the enzyme, led us to
further modify the structure of the inhibitors by an elongation of
the polar head alkyl tether chain. This second modification pro-
duced compounds 28–39, another series having several inhibitors
with IC50 in the low micromolar range and some inactive
compounds.
4 lM. However, cetrimide toxicity (LD50 1 lM) precludes its use
in cellular studies of PC-PLCBc inhibition. This result confirms the
importance of the charged ammonium salt for enzyme inhibition
in the compounds studied, in accordance with the important role
played by choline in the recognition of phospholipid substrates
by the enzyme.31 As for the importance of hydrophobic interac-
tions, this had already been reported49,50 and we had observed a
The results obtained gave us some insights about the structural
features that favour inhibition of PC-PLCBc. From our initial work
on xanthate32 and
a
-aminohydroxamic41 PC-PLCBc inhibitors we
concluded that the presence of a zinc binding group and a lipho-

P. González-Bulnes et al. / Bioorg. Med. Chem. 18 (2010) 8549–8555
8555
similar trend in xanthate PC-PLCBc inhibitors,32 where the binding
of the compounds to the active site depends on the presence of the
zinc coordinating dithiocarbonate group, with the present hydro-
phobic moiety modulating its inhibitory activity.
3. Watkins, J. B. Pediatrics 1985, 75, 151.
4. Oestvang, J.; Johansen, B. Biochim. Biophys. Acta 2006, 1761, 1309.
5. Triggiani, M.; Granata, F.; Frattini, A.; Marone, G. Biochim. Biophys. Acta 2006,
1761, 1289.
6. Zhang, L.; Zhao, J.; Su, L.; Huang, B.; Wang, L.; Su, H.; Zhang, Y.; Zhang, S.; Miao,
J. Arterioscler. Thromb. Vasc. Biol. 2010, 30, 411.
The increase of the alkyl tether length from two to three carbon
atoms did not result in a substantially higher inhibitory potency.
Compounds from the first and second series exhibited similar
IC50 values. Overall, it seems that inhibitory activity increases with
the lipopholicity of the molecule that can be the result of either a
longer amino acidic alkyl aminohydroxamic functionalities. The
compounds having dimethylamino groups in place of trimethy-
lammonium substituents display inhibitory activities in the same
order of magnitude but systematically less potent than their trime-
thylammonium counterparts. Four compounds, representative of
the different inhibitor types studied, were selected for kinetic stud-
ies and we found that none of them was a pure PC-PLCBc compet-
itive inhibitor. Compound 25 was an uncompetitive inhibitor,
compound 29 turned out to inhibit the enzyme in a non-competi-
tive way and compounds 35 and 39 were mixed inhibitors. The
lack of interaction of the compounds with the active site of the en-
zyme would explain why the zinc coordinating moiety is not
essential for inhibitory activity. This behaviour is different from
that of other PC-PLCBc competitive inhibitors synthesized in our
group, such as alpha-aminohydroxamic acids or xanthathes, where
a low inhibitory potency was found when the zinc coordinating
groups were chemically blocked. Although these results cannot
be generalised to all members of the different chemical families
studied, we think that they are compatible with inhibitor binding
to PC-PLCBc protein sites other than the zinc catalytic centre, a fact
that might be related to the known interfacial activation of this en-
zyme.51 The cytotoxicity exhibited for long chain analogues and
the low inhibitory activity elicited by n-hexylamino acid deriva-
tives, points out to n-decanoyl derivatives as the best candidates
to develop future work on in vivo studies of this new family of
PC-PLC inhibitors.
7. Alberti-Segui, C.; Goeden, K. R.; Higgins, D. E. Cell. Microbiol. 2007, 9, 179.
8. Ghelardi, E.; Celandroni, F.; Salvetti, S.; Fiscarelli, E.; Senesi, S. Microbes Infect.
2007, 9, 591.
9. Becker, K. P.; Hannun, Y. A. Cell. Mol. Life Sci. 2005, 62, 1448.
10. Mateos, M. V.; Uranga, R. M.; Salvador, G. A.; Giusto, N. M. Lipids 2006, 41, 273.
11. Wang, H.; Kazanietz, M. G. Biochem. Soc. Trans. 2006, 34, 855.
12. Bell, R. M. Cell 1986, 45, 631.
13. Exton, J. H. J. Biol. Chem. 1990, 265, 1.
14. Nishizuka, Y. Nature 1988, 334, 661.
15. Singer, S. J.; Nicolson, G. L. Science 1972, 175, 720.
16. Wakelam, M. J. O. Biochim. Biophys. Acta 1998, 1436, 117.
17. Xingzhong, W. U.; Lu, H.; Zhou, L. A. N.; Huang, Y.; Chen, H. Cell Biol. Int. 1997,
21, 375.
18. Iorio, E.; Mezzanzanica, D.; Alberti, P.; Spadaro, F.; Ramoni, C.; D’Ascenzo, S.;
Millimaggi, D.; Pavan, A.; Dolo, V.; Canevari, S.; Podo, F. Cancer Res. 2005, 65,
9369.
19. Plo, I.; Lautier, D.; Levade, T.; Sekouri, H.; Jaffrezou, J. P.; Laurent, G.; Bettaieb, A.
Biochem. J. 2000, 351, 459.
20. Suh, P. G.; Park, J. I.; Manzoli, L.; Cocco, L.; Peak, J. C.; Katan, M.; Fukami, K.;
Kataoka, T.; Yun, S.; Ryu, S. H. BMB Rep. 2008, 41, 415.
21. Clark, M. A.; Shorr, R. G. L.; Bomalaski, J. S. Biochem. Biophys. Res. Commun.
1986, 140, 114.
22. Hansen, S.; Hansen, L. K.; Hough, E. J. Mol. Biol. 1992, 225, 543.
23. Hansen, S.; Hough, E.; Svensson, L. A.; Wong, Y.-L.; Martin, S. F. J. Mol. Biol.
1993, 234, 179.
24. Hough, E.; Hansen, L. K.; Birknes, B.; Jynge, K.; Hansen, S.; Hordvik, A.; Little, C.;
Dodson, E.; Derewenda, Z. Nature 1989, 338, 357.
25. Little, C.; Otnass, A. B. Biochim. Biophys. Acta 1975, 391, 326.
26. Aurebekk, B.; Clive, L. Int. J. Biochem. 1977, 8, 757.
27. Aurebekk, B.; Little, C. Biochem. J. 1977, 161, 159.
28. Little, C. Biochem. J. 1977, 167, 399.
29. Little, C.; Aurebekk, B. Acta Chem. Scand. B 1977, 31, 273.
30. Martin, S. F.; Wong, Y.-L.; Wagman, A. S. J. Org. Chem. 1994, 59, 4821.
31. Benfield, A. P.; Goodey, N. M.; Phillips, L. T.; Martin, S. F. Arch. Biochem. Biophys.
2007, 460, 41.
32. Gonzalez-Roura, A.; Casas, J.; Llebaria, A. Lipids 2002, 37, 401.
33. Kang, M. S.; Jung, S. Y.; Jung, K. M.; Kim, S. K.; Ahn, K. H.; Kim, D. K. Mol. Cells
2008, 26.
34. Lauderback, C. M.; Drake, J.; Zhou, D.; Hackett, J. M.; Castegna, A.; Kanski, J.;
Tsoras, M.; Varadarajan, S.; Butterfield, D. A. Free Radic. Res. 2003, 37, 355.
35. Luberto, C.; Hannun, Y. A. J. Biol. Chem. 1998, 273, 14550.
36. Aakre, S. E.; Little, C. Biochem. J. 1982, 203, 799.
37. Martin, S. F.; Follows, B. C.; Hergenrother, P. J.; Franklin, C. L. J. Org. Chem. 2000,
65, 4509.
4. Experimental
38. Franklin, C. L.; Li, H.; Martin, S. F. J. Org. Chem. 2003, 68, 7298.
39. Roberts, M. F.; Wu, Y.; Zhou, C.; Geng, D.; Tan, C. Adv. Enzyme Regul. 1996, 36,
57.
Detailed experimental procedures and spectral data for the syn-
thesized compounds are provided in Supplementary data.
40. Tan, C. A.; Roberts, M. F. Biochim. Biophys. Acta 1996, 1298, 58.
41. Gonzalez-Roura, A.; Navarro, I.; Delgado, A.; Llebaria, A.; Casas, J. Angew. Chem.,
Int. Ed. 2004, 43, 862.
Acknowledgements
42. Albertson, N. F. J. Am. Chem. Soc. 1946, 68, 450.
43. Madsen, U.; Brauner-Osborne, H.; Frydenvang, K.; Hvene, L.; Johansen, T. N.;
Nielsen, B.; Sanchez, C.; Stensbol, T. B.; Bischoff, F.; Krogsgaard-Larsen, P. J.
Med. Chem. 2001, 44, 1051.
Partial financial support from the ‘Ministerio de Ciencia e Inno-
vación’, Spain (Project CTQ2008-01426/BQU) and ‘Generalitat de
Catalunya’ (Grant 2009SGR-1072) is acknowledged.
44. Mori, K.; Otsuka, T. Tetrahedron 1985, 41, 547.
45. Nikam, S. S.; Komberg, B. E.; Johnson, D. R.; Doherty, A. M. Tetrahedron Lett.
1995, 36, 197.
46. Hergenrother, P. J.; Martin, S. F. Anal. Biochem. 1997, 251, 45.
47. el-Sayed, M. Y.; DeBose, C. D.; Coury, L. A.; Roberts, M. F. Biochim. Biophys. Acta
1985, 837, 325.
48. Martin, S. F.; Hergenrother, P. J. Bioorg. Med. Chem. Lett. 1998, 8, 593.
49. Antikainen, N. M.; Monzingo, A. F.; Franklin, C. L.; Robertus, J. D.; Martin, S. F.
Arch. Biochem. Biophys. 2003, 417, 81.
50. Martin, S. F.; Follows, B. C.; Hergenrother, P. J.; Trotter, B. K. Biochemistry 2000,
39, 3410.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.10.031.
References and notes
51. Singh, J.; Ranganathan, R.; Hadju, J. J. Phys. Chem. B 2008, 112, 16741.
52. Martin, S. F.; Hergenrother, P. J. Biochemistry 1999, 38, 4403.
1. Scott, J. A. J. Theor. Biol. 1984, 111, 659.
2. Borgstrom, B. Gastroenterology 1980, 78, 954.