The molecular mechanism of human hormone-sensitive lipase inhibition by substituted 3-phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones

Article abstract of DOI:10.1016/j.biochi.2011.09.028

Hormone-sensitive lipase (HSL) plays an important role in the mobilization of free fatty acids (FFA) from adipocytes. The inhibition of HSL may offer a pharmacological approach to reduce FFA levels in plasma and diminish peripheral insulin resistance in t

Full text of DOI:10.1016/j.biochi.2011.09.028

Biochimie 94 (2012) 137e145
Contents lists available at SciVerse ScienceDirect
Biochimie
journal homepage: www.elsevier.com/locate/biochi
Research paper
The molecular mechanism of human hormone-sensitive lipase inhibition
by substituted 3-phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones
Yassine Ben Ali a , Robert Verger , Frédéric Carrière , Stefan Petry , Günter Muller ,
,
b
,
c
b
b
d
d
a,b,
*
Abdelkarim Abousalham
Organization and Dynamics of Biological Membranes, UMR 5246 ICBMS, CNRS-Université Claude Bernard Lyon 1, Bâtiment Raulin, 43, boulevard du 11 novembre 1918,
9622 Villeurbanne Cedex, France
CNRS-Aix-Marseille Université, Enzymology at Interfaces and Physiology of Lipolysis, UPR 9025, 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France
Laboratoire de Biochimie et de Génie Enzymatique des Lipases, Ecole Nationale d’Ingénieurs de Sfax, route de Soukra, 3038, Université de Sfax, Tunisia
Sanofi-Aventis Germany, 65926 Frankfurt am Main, Germany
a
6
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 April 2011
Accepted 29 September 2011
Available online 13 October 2011
Hormone-sensitive lipase (HSL) plays an important role in the mobilization of free fatty acids (FFA) from
adipocytes. The inhibition of HSL may offer a pharmacological approach to reduce FFA levels in plasma
and diminish peripheral insulin resistance in type 2 diabetes. In this work, the inhibition of HSL by
substituted 3-phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones has been studied in vitro. 5-methoxy-3-(3-
phenoxyphenyl)-1,3,4-oxadiazol-2(3H)-one (compound 7600) and 5-methoxy-3-(3-methyl-4-pheny-
lacetamidophenyl)-1,3,4-oxadiazol-2(3H)-one (compound 9368) were selected as the most potent HSL
inhibitors. HSL is inhibited after few minutes of incubation with compound 7600, at a molar excess of 20.
This inhibition is reversed in the presence of an emulsion of lipid substrate. The reactivation phenom-
enon is hardly observed when incubating HSL with compound 9368. The molecular mechanism
underlying the reversible inhibition of HSL by compound 7600 was investigated using high performance
liquid chromatography and tandem mass spectrometry. The stoichiometry of the inhibition reaction
revealed that specifically one molecule of inhibitor was bound per enzyme molecule. The inhibition by
compound 7600 involves a nucleophilic attack by the hydroxy group of the catalytic Ser of the enzyme on
the carbon atom of the carbonyl moiety of the oxadiazolone ring of the inhibitor, leading to the formation
of covalent enzymeeinhibitor intermediate. This covalent intermediate is subsequently hydrolyzed,
releasing an oxadiazolone decomposition product, carbon dioxide and the active HSL form. On the basis
of this study, a kinetic model is proposed to describe the inhibition of HSL by compound 7600 in the
aqueous phase as well as its partial reactivation at the lipidewater interface.
Keywords:
Hormone-sensitive lipase
3-Phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones
Inhibition
Reactivation
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction
carbohydrate metabolism. Elevated plasma FFA levels impair
insulin signaling, decrease glucose oxidation and glycogen
Insulin resistance is one parameter of the metabolic syndrome.
This metabolic syndrome is often associated with high levels of free
fatty acid (FFA), in addition to increased levels of triacylglycerols
synthesis [2,3] and results in decreased rates of glucose uptake.
Moreover, FFA may influence both insulin sensitivity in muscle and
insulin secretion in pancreas.
(
[
TAG) and low levels of high-density lipoprotein (HDL)-cholesterol
1]. FFAs stored as TAG in adipose tissue are the main source of
HSL is thought to play an important role in the mobilization of
FFA from the TAG stored in adipocytes, skeletal muscle, and
energy in mammals and influence various aspects of lipid and
pancreatic b-cells (for a review, see ref. [4]). In vivo, HSL is activated
by phosphorylation via cAMP-dependent kinase in response to
various lipolytic hormones such as catecholamines. The functional
significance of HSL in adipose tissue metabolism has been clarified
in studies using HSL null mice [5e9]. TAG lipase activity in white
adipose tissue was reduced by only 40e45% and in brown adipose
tissue was similar to wild-type mice [8]. Additionally, there was
a marked defect or complete absence of catecholamine-stimulated
*
Corresponding author. Organization and Dynamics of Biological Membranes,
UMR 5246 ICBMS, CNRS-Université Claude Bernard Lyon 1, Bâtiment Raulin, 43,
boulevard du 11 novembre 1918, 69622 Villeurbanne Cedex, France. Tel.: þ33 0 4 72
4
4 81 02; fax: þ33 0 4 72 44 79 70.
E-mail address: abousalham@univ-lyon1.fr (A. Abousalham).
URL: http://www.icbms.fr/
0300-9084/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.biochi.2011.09.028

138
Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
glycerol release in adipocytes from HSL null mice, whereas
catecholamine-stimulated FFA release was still observed, but
attenuated [8,9]. This apparent discrepancy in the release of glyc-
erol and FFA from adipocytes of HSL null mice has been clarified by
the observation that the basal- and catecholamine-induced diac-
ylglycerol (DAG) content increased markedly in white and brown
adipose tissue of HSL null mice [5]. Therefore, the study of HSL null
mice appear to substantiate that HSL is the rate limiting-enzyme for
DAG hydrolysis in adipose tissue and is essential for hormone-
stimulated lipolysis. Using chiral-stationary-phase HPLC, we
recently showed that HSL was found to have a pronounced
stereospecificity for the sn-3 position of DAG [10].
2. Materials and methods
2.1. Reagents
Dioleoylglycerol (diolein), tributyroylglycerol (tributyrin),
cholesterol oleate, vinyl butyrate, Nonidet-P, sodium taurodeox-
ycholate (NaTDC), dimethylsulfoxide (DMSO) and bovine serum
albumin (BSA) were purchased from SigmaeAldricheFluka Chimie
(St-Quentin-Fallavier, France). HPLC grade acetonitrile was ob-
tained from Merck (Darmstadt, Germany). All other chemicals and
solvents were of reagent or better quality and were obtained from
local suppliers.
Furthermore, the inactivation of HSL resulted in the loss of more
than 98% of neutral cholesteryl ester hydrolase activity in both male
and female null mice adrenals suggesting that HSL is critically
involved in the intracellular processing and availability of choles-
terol for adrenal steroidogenesis [6,7]. In line with these observa-
tions, we reported previously that HSL activity on cholesteryl esters
was approximately four to five times higher than on long-chain TAG
2
.2. Substituted 3-phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones
Substituted 3-phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones were
synthesized as described previously [23]. Compound 7600 and
compound 9368 (Fig. 1) were selected as the most potent HSL
inhibitors.
[11] and consequently, HSL could be considered more as a choles-
teryl ester hydrolase than as a TAG lipase [11]. An adipose TAG
lipase (ATGL) was identified [12] and proposed to be responsible for
the initial step in TAG catabolism. HSL and ATGL may coordinately
catabolize the TAG stored in mammalian adipose tissues [12].
Another neutral intracellular TAG hydrolase (TGH) also termed
carboxylesterase 3 was shown to be associated with hepatic [13]
and adipocyte lipid droplets [14,15] where it may participate in
TAG turnover and fatty acid efflux. It was reported recently [16] that
2
.3. Proteins
Recombinant human HSL was expressed and purified from
baculovirus-infected insect cells as described previously [24]. The
protein concentrations were determined using Bradford’s proce-
dure [25], with Bio-Rad Dye Reagent and BSA as the standard.
2
.4. Enzyme activity measurements
ꢀ
/ꢀ
the ablation of TGH expression in mice (Tgh ) results in decreased
hydrolase activity accompanied with reduced plasma FFA and
Enzymatic activity was assayed by measuring the FFA released
from mechanically stirred acylglycerol, using 0.1 N NaOH with
a pH-stat (TTT80 radiometer, Copenhagen) adjusted to a fixed end
ꢀ
/ꢀ
glycerol levels. Tgh
mice exhibited improved glucose tolerance
and insulin sensitivity [16].
Targeting FFA metabolism may therefore offer a therapeutic
approach that addresses the fundamental cause of insulin resis-
tance and type 2 diabetes. Thus, the inhibition of HSL offers
a pharmacological approach to reduce FFA levels and diminish
peripheral insulin resistance. Over the last years, various HSL
inhibitors, including natural products (cyclipostins from Strepto-
myces [17] and Eudesmanolides isolated from Iva microcephala nut
CH3
R1
O
N
[
18]) and synthetic products (carbamates [19], pyrrolopyr-
R2
N
azinediones [20] and carbazates [21]) have been described for the
treatment of diabetes and lipid disorders. These reports lack
however, detailed biochemical characterization of HSL inhibition by
these inhibitors.
O
The HSL family is an increasingly large group of proteins
showing structural similarities with the catalytic domain of HSL.
HSL is the only carboxylester hydrolase belonging to this family of
proteins having clearly detectable lipolytic activity on long-chain
TAG, DAG, monoacylglycerol, cholesteryl ester and retinyl ester
substrates. We showed previously [22] that compound 7600
specifically inhibits the HSL family members (HSL, esterase from
the thermophilic eubacterium Alicyclobacillus acidocaldarius (EST2)
and an esterase from hyperthermophilic archaeon Archaeoglobus
fulgidus (AFEST)). Whereas, no significant inhibitory effects were
observed with other lipolytic and non-lipolytic carboxylester
hydrolases that are not members of the HSL family [22]. Surface
enhanced laser desorption ionisation/time of flight (SELDI-TOF)
mass spectrometry analysis of a trypsin digest of AFEST treated or
not with the inhibitor showed the occurrence of an increase in the
molecular masses of catalytic serine-containing peptide, which
is compatible with the formation of a covalent complex with the
inhibitor [22]. The aim of the present study was to investigate the
inhibition properties of compound 7600 on HSL activity, as well as
the molecular mechanism of this inhibition using high performance
liquid chromatography and tandem mass spectrometry (LC-MS).
O
Compound 7600:
R = O
, R = H
2
1
Compound 9368:
O
H C
2
R = CH ,
R = HN
1
3
2
Fig. 1. Compound 7600 and compound 9368 structures.

Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
139
point value. When olive oil or dioleoylglycerol were used as
A 1.2
substrates, gum arabic was used as an emulsifier as previously
1
2
3
ꢁ
described [26]. Each assay was performed in a thermostated (37 C)
vessel containing 0.25 mM TriseHCl buffer and 150 mM NaCl. The
activity of HSL was measured at pH 7.5 in the presence of BSA (2 mM,
1.0
4
final concentration). The specific activities are expressed here in
international units (IU) per milligram of enzyme. One unit corre-
0
0
0
0
.8
.6
.4
.2
0
5
sponds to 1
.5. Inhibition of HSL by compound 7600 and compound 9368
The enzyme was incubated with the inhibitor, in aqueous
mmol of FFA released per minute.
2
Steady
medium for testing direct interaction between the lipase and the
inhibitor in the absence of substrate [27]. Recombinant human HSL
state rate
ꢀ
1
(
0.2 mg ml ; final concentration 2.3
Na HPO buffer, pH 7.5, containing 0.1 M NaCl, 20% glycerol, 3
mercaptoethanol and 0.1% Nonidet-P. Compound 7600 or 9368
10 mM) was pre-dissolved in DMSO for its further dispersion in the
mM) was dissolved in
2
4
mM b-
0
2
4
6
8
10
12
14
(
B
τ
water phase. An aliquot of the above enzyme solution was incu-
bated at 25 C with the inhibitor at an enzyme:inhibitor molar ratio
Time (min)
ꢁ
5
4
3
2
1
varying from 1:100 to 1:1 (final DMSO concentration ranged from 2
to 3%). The residual enzyme activity was then measured after
various incubation times, using emulsions of vinyl butyrate, diolein,
olive oil, cholesterol oleate or tributyrin as substrate (see above).
Control experiments were performed in the absence of inhibitor
and with the same concentration of DMSO. It is worth noticing that
DMSO at a final concentration of less than 5% has no effect on the
enzyme activity.
2.6. Treatment of the kinetic data obtained upon enzyme
reactivation
0
10
20
30
40
50
60
In order to fit the experimental data points with the general
theoretical equation derived from the model proposed by Verger
Incubation period (min)
Fig. 2. (A) Typical kinetic recordings of the hydrolysis of tributyrin emulsion by HSL,
using the pH-stat technique. HSL was incubated for 0 min (curve 1), 3 min (curve 2),
10 min (curve 3), 30 min (curve 4) and 60 min (curve 5) with compound 7600 at
enzyme:inhibitor molar ration of 1:20. At various time intervals, an aliquot of the assay
medium was injected in the pH-stat vessel, after recording the background hydrolysis
et al. [28] we have used this following simplified equation:
ꢀ
ꢁ
C
C
B
ꢀ
t=r
P ¼ Bt þ
s
e
ꢀ 1
(1)
ꢁ
for 2 min. Kinetic assays were performed in a thermostated (37 C) vessel containing
Where P is the product (FFA) concentration (molecule/volume) and
t, time (min). (dP/dt) ¼ C/B when time tends to infinity. C and B are
complex parameters resulting from a combination of various
individual kinetic rate constants (for their exact definition, see
reference [28]).
0.5 ml tributyrin mechanically emulsified in 14.5 ml of 1 mM TriseHCl buffer, 150 mM
NaCl. Kinetic recordings are representative of three independent experiments. The
dashed line corresponds to the asymptote reached under steady state conditions ob-
tained by curve fitting the kinetics using equation (1) (see Materials and methods
section).
of the lag time of HSL activity as a function of its incubation time with compound 7600,
using a tributyrin emulsion. The lag times ( ) were determined as illustrated in Fig. 2
and described in Materials and methods section.
s (lag time) is the intercept with time axis of the dashed line. (B) Variation
Under steady state conditions, C/B is the maximal reaction rate
s
and
s
(lag time) is the intercept with time axis of the asymptote of
are given after curve
equation (1) (see Fig. 2). The values of C/B and
s
fitting of the experimental data with the equation (1), using
KaleidaGraph 3.0 software.
residue was taken up in ethanol (15 mL) and injected into the HPLC
system. The enzymatic conversion of the compound was followed
by LC-MS.
Ò
From equation (1), one can see that theoretically the initial rate
(
at t ¼ 0) is equal to zero. However, in order to estimate experi-
HPLC separation was performed on a Luna C18 column
mentally the initial activity of HSL we measured the slope of the
kinetic recordings 1 min after adding HSL-inhibitor mixture to the
lipase assay.
(100 ꢂ 4.6 mm i.d., 5
mm particle size, Phenomenex, Aschaffenburg,
Germany). The HPLC system was an Agilent 1100 system (Agilent,
Böblingen, Germany). HPLC was performed using the solvent
systems A (acetonitrile) and B (ammonium acetate buffer, pH 2.8) at
a flow rate of 1 ml/min. After equilibration of the column for 2 min
with system A/system B ¼ 50/50 (v/v), the sample was injected and
the chromatography was run for 1 min with system A/system
B ¼ 50/50 (v/v). Then, over a period of 1 min the composition was
changed to system A/system B ¼ 98/2 (v/v). The chromatography
was continued with this composition for 5 min. An API3000 triple-
quadrupole mass spectrometer (AB/MDS Sciex, Toronto, Canada)
was used for detection. A 1/x-weighted linear model was used for
the regression of peak area ratio of compound/internal standard
versus compound concentration.
2.7. Sample preparation for the identification by LC-MS of the
oxadiazolone degradation product
Recombinant human HSL (0.2 mg mlꢀ1; final concentration
2
.3
2 4
mM) was dissolved in Na HPO buffer, pH 7.5, containing 0.1 M
NaCl, 20% glycerol, 3
aliquot of enzyme solution was incubated at 24 C with compound
m
M
b
-mercaptoethanol and 0.1% Nonidet-P. An
ꢁ
2
7600 (200 mM in DMSO). After 1 h, ethyl acetate/H O (200 mL) was
added and the products extracted from the reaction mixture. The
organic layer was separated and the solvent evaporated. The

140
Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
3
. Results
However, on the presence of the inhibitor the lag time increases as
function as the incubation period. When the activity is assayed just
after mixing the enzyme with the inhibitor (t ¼ 0 min), the low
value of the lag time is comparable to those found in the absence of
compound 7600 and could be attributed to the mixing time
required for the injected enzyme sample to be homogeneously
distributed. When the incubation period increases (10, 30 and
60 min), the values of the lag time were higher (a plateau value of
around 4 min) (Fig. 2B).
3
9
.1. Inhibition kinetics of HSL by compound 7600 and compound
368
At various incubation time of HSL with compound 7600
(
HSL:compound 7600 molar ratio, 1:20), aliquots were taken from
the incubation medium and the residual HSL activity was measured
using tributyrin emulsion as substrate. Fig. 2A shows typical kinetic
recordings of the hydrolysis rates of tributyrin emulsion with HSL
after various incubation time with the compound 7600. The same
behavior was observed when incubating HSL with compound 9368
Fig. 3A shows the residual HSL activity measured under steady
state conditions as function of the incubation time of the enzyme
with 20 molar excess of compound 7600. After few seconds of
incubation, approximately 50% of the residual HSL activity was
measured on tributyrin (Fig. 3A) or vinyl butyrate (data not shown)
emulsions. This HSL residual activity further decreases and reaches
a plateau value at approximately 35% after 60 min of incubation.
Control experiments without inhibitor were performed in parallel
to each assay and showed linear kinetics without significant change
in HSL activity. The effects of increasing concentrations of
compound 7600 on the residual steady state HSL activity is shown
in Fig. 3C. The concentration of compound 7600 which reduced the
HSL activity to 50% (IC50) was found to be around 250 nM.
Fig. 3B shows the residual HSL activity measured under steady
state conditions as function of the incubation time of HSL with 20
molar excess of compound 9368. This HSL residual steady state
activity further decreases rapidly and reached approximately 25%
when incubating the HSL and the compound 9368 for a few
seconds and reached a plateau value corresponding approximately
to 90% of inhibition. Fig. 3D shows the effects of increasing
concentrations of compound 9368 on the residual steady state HSL
activity. The IC50 of this inhibitor was found to be around 90 nM.
(
data not shown). The hydrolytic activity gradually increased with
time, however, reaching a steady state regime after a lag period of
a few minutes. The lag time was defined as the point where the
extrapolated steady-state curve intersected with the time axis
(Fig. 2A). The fitting of kinetics according to equation (1) (see
Materials and methods) shows a good adjustment between the
experimental data and the theoretical equation (1) which allows
the measurements of the initial residual activity, the steady state
residual activity and the lag time. The initial residual rates decrease
with increasing incubation times. The HSL activity, however,
progressively increased with time and was partially restored
reaching a steady state after a lag time. When aliquots of inhibited
HSL were added to the lipase assay medium 10 min before adding
tributyrin, similar kinetics recordings (data not shown) as in Fig. 2A
were obtained. This result indicates that the reactivation process of
the inhibited HSL is not the consequence of its dilution into the
aqueous phase but rather to the presence of the substrate.
The reactivation phenomenon of HSL was also seen (data not
shown) when using other known HSL substrates such as diolein, p-
nitrobenzofurazan (NBD)-labeled monoacylglycerol [29] and vinyl
butyrate.
3.2. Products identification of the HSL-catalyzed hydrolysis of
The lag time values, ranging from 1 to around 4 min, were
calculated from the kinetic recordings of the HSL acting on tribu-
tyrin emulsion (Fig. 2B). In the absence of inhibitor compound
compound 7600
The possibility that HSL inhibitors investigated in this study are
substrates of HSL was checked for compound 7600. At various
7600, the kinetic was linear and no lag time was observed.
A
B
1
00
100
75
50
25
0
7
5
2
5
0
5
0
0
20
40
60
0
20
40
60
Incubation time (min)
Incubation time (min)
D
C
100
100
7
5
75
50
5
2
0
5
0
25
0
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Compound 7600 (µM)
Compound 9368 (µM)
Fig. 3. HSL residual activity, measured using the pH-stat technique, under the steady state conditions using a tributyrin emulsion as substrate. (A) and (B), HSL steady state residual
activity measured as function of the incubation time with compound 7600 and compound 9368, respectively, with 20 molar excess of inhibitor. (C) and (D), HSL steady state residual
activity measured as function of compound 7600 and compound 9368 concentrations, respectively, after an incubation period of 60 min. The values of the steady state residual
activity of HSL were determined as described in Materials and methods section. The values are expressed as means ꢃ S.D. (n ꢄ 3 for each activity assay).

Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
141
incubation time of HSL with 20 molar excess of compound 7600,
aliquots were taken from the incubation medium and the resulting
reaction products were analyzed by LC-MS as described in
Materials and methods. As shown in Fig. 4A, liquid chromatography
analysis of HSL-catalyzed hydrolysis of compound 7600 revealed
three peaks having a retention time of 1.63, 1.70 and 1.90 min,
respectively. A control experiment without HSL was performed
metabolite has been resynthesized and co-elutes with the metab-
olite (RT 1.63, data not shown).
In parallel, the HSL activity was measured using the pH-stat
technique. The kinetic of compound 7600 disappearance and the
resulting product appearance are shown in Fig. 5. We notice
a
progressive enzymatic conversion of the inhibitor with
a progressive increase of the reaction product (compound P).
Almost complete conversion of the compound 7600 to compound P
was achieved within 30 min incubation of the enzyme with the
inhibitor (Fig. 5).
(Fig. 4B) and revealed two peaks having a retention time of 1.70 and
1.90 min, respectively. The UV peak (RT 1.70 min) is probably due to
an impurity within the buffer system. Mass spectra were recorded
for the peaks at 1.90 and 1.63 respectively. The mass-spectrum at
1
.90 shows the mass (M þ H 285) corresponding to compound 7600
4. Discussion
(
Fig. 4B, insert), whereas the main peak (M þ H 259) corresponds to
the product of hydrolysis (Fig. 4A, insert). According to the
proposed mechanism of hydrolysis the mass corresponds to methyl
We showed that compound 7600 and compound 9368 from the
oxadiazolone family are potent inhibitors of the recombinant
human HSL ([22] and this study). This inhibition occurs within
2
-(3-phenoxyphenyl)hydrazinecarboxylate (compound P). This
Fig. 4. Identification by LC-MS of the product generated after the hydrolysis of the compound 7600 by HSL. HSL was incubated with compound 7600 (200
of enzyme/compound 7600 was analyzed by LC-MS (A). Control experiment of compound 7600 without HSL is shown in (B). The mass-spectrum at 1.63 (A, insert) shows the mass
M þ H 259.23) corresponding to the product of hydrolysis (compound P), whereas the mass-spectrum at 1.90 (B, insert) shows the mass (M þ H 285.20) corresponding to
compound 7600.
mM in DMSO). An aliquot
(

142
Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
a few minutes both in the absence and presence of detergent such
0
0
0
0
.60
.45
.30
.15
as bile salts in the incubation medium. This result is in line with the
fact that the catalytic serine of HSL is highly reactive and readily
accessible on the contrary to what is observed in many lipases [24].
From the 3D modeling of the HSL and the distribution of the
hydrophobic surfaces (see Fig. 6), HSL share same similarity with
Fusarium solani pisi cutinase [30], a lipase that lacks the peptide
loop (lid) which control the substrate access to the active site.
Compound 7600 efficiently inhibits other HSL family members,
such as EST2 from the thermophilic eubacterium A. acidocaldarius
and AFEST from the hyperthermophilic archaeon A. fulgidus. In
contrast, carboxylester hydrolases that are not belonging to the HSL
family were or poorly inhibited by compound 7600 under the same
experimental conditions [22]. The inhibition of carboxylester
hydrolases belonging to the HSL family resulted in a covalent
modification of the active site serine and in a concomitant increase
in the molecular mass of the protein, corresponding in size to the
expected molecular mass of the inhibitor [22].
Compound 7600
Compound P
0
.00
0
.0
0.5
1.0
1.5
2.0
Time (h)
Fig. 5. Kinetic analysis by LC-MS of HSL-catalyzed compound 7600 disappearance. An
aliquot of enzyme solution was incubated at 24 C with compound 7600 at an enzy-
me:inhibitor molar ratio of 1:1 for 1e24 h. HPLC separation was performed in
Materials and methods section.
When HSL was incubated with compound 7600 and its residual
activity was tested as function of time, two phenomena were
observed: first the initial rate of substrate hydrolysis was reduced
with time, second a partial enzyme reactivation was observed
ꢁ
Fig. 6. The mechanism possibly underlying the inhibition exerted on recombinant human HSL by compound 7600. The acyl enzyme is formed by the nucleophilic attack by the
hydroxy group of the enzyme’s catalytic Ser425 on the carbonyl atom of the carbonyl moiety of the inhibitor’s oxadiazolone ring, leading to the formation of an enzyme-bound
intermediate. In the presence of H O molecule, this intermediate will be hydrolyzed to liberate active HSL and compound P. In the model of the catalytic domain of HSL,
2
Surface distribution of hydrophobic side chain residues (V, A, L, I, W, and F), indicated in yellow, the active site serine is indicated in red [24]. (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of this article.)

Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
143
after a lag phase. These finding were obtained with tributyrin (see
Fig. 2), diolein, vinyl butyrate and cholesterol oleate as substrates
interesterification mechanism involving TAG and characterized by
a kT1 rate constant. Thus the reaction rate increases, due to the
progressive release of an unstable acyl-HSL complex (E*eFFA). This
progressive reactivation process is characterized by the existence of
lag times (see Fig. 2). The fact that the reactivation of HSL does not
reach 100% at the end of the kinetics could be explained by the
existence of the second fraction of the enzyme which is more
(
data not shown). The activity of HSL toward cholesterol esters is 4-
to 5-fold higher than on long-chain TAGs. However, we used in this
study a partly soluble TAG such as tributyrin. We showed previ-
ously (see ref. [11]) that the HSL specific activity on emulsified
tributyrin is comparable to the one on cholesterol oleate
ꢀ1
ꢀ1
(
18 ꢃ 2
m
mol min mg ). Furthermore, from technical point of
2
strongly inhibited by compound 7600 (E*ecompound 7600o ) and
view, the emulsified tributyrin substrate for lipases is very easy to
use within the pH-stat technique. This reactivation phenomenon
indicates that compound 7600 is a transient inhibitor (pseudo-
substrate) of HSL in which the enzyme acylation step is followed by
a rate limiting deacylation reaction step of a fraction of the covalent
HSLecompound 7600 complex during the lipase assay. The
complete recovery of the activity of the inhibited HSL was never
observed. Similar observations have been previously reported for
the inhibition of human pancreatic lipase [31,32] and human car-
boxylester lipase [33] by tetrahydrolistatin, a potent inhibitor of
carboxylester hydrolases. The complete recovery of the original
activity of the enzyme was partially (human pancreatic lipase) or
almost completely (human carboxylester lipase) restored after 24 h
of incubation with tetrahydrolistatin [32,33]. Furthermore, it has
been reported that the inhibition of bovine lipoprotein lipase [34]
or Rhizopus oryzae lipase [35] by tetrahydrolistatin were found to
be reversible in the presence of a lipidewater interface. It has been
which remains stable, at least during the recording kinetic period.
In other words this means that kT1 is much higher than kT2. The co-
existence of two different forms of compound 7600 bound to HSL
could be due to two different orientations of this inhibitor molecule
in the catalytic cavity of the lipase molecule. The observed reac-
tivation phenomenon could be due to the existence of a large
proportion of Eecompound 7600o which can dissociate by an
interesterification mechanism involving TAG and a negligible
proportion of Eecompound 7600o which remain stable during the
recording kinetic period.
It is worth noticing that the reactivation phenomenon is hardly
observed (data not shown) when incubating HSL with compound
9368. These results probably indicate that compound 9368 is
a potent inhibitor of HSL in which the enzyme forms a stable,
covalent and long-lived acyl-enzyme complex. One can hypothe-
sized that the covalent binding of compound 9368 to HSL happens
also in two open conformations: A first fraction of the lipase
molecules will form in the aqueous phase a covalent complex with
the open oxadiazolone ring of compound 9368 in a first confor-
suggested that
complex of a long-lived acyl-enzyme type is formed between the
open -lactone ring of tetrahydrolistatin and the catalytic serine of
pancreatic lipase [31,32,36,37]. This complex was then slowly
a stoichiometric enzymeeinhibitor covalent
b
mation (Eecompound 9368o
form a covalent complex with compound 9368 in a second
conformation (Eecompound 9368o ). The fact that the reactivation
1
) whereas the second fraction will
hydrolyzed, leaving an intact enzyme and the
tetrahydrolistatin in its open (inactive) form.
b
-lactone ring of
2
of HSL was difficult to observe with compound 9368 could be
explained by the existence of a negligible proportion of the
Eecompound 9368o, which can dissociates by an interesterification
mechanism involving TAG and the existence of a large proportion of
the Eecompound 9368o which is strongly inhibited by compound
9368 and which remains stable during our experimental recording
kinetic period.
In our work, the reversible mechanism of HSL inhibition by
compound 7600 was further investigated by analyzing the enzy-
matic hydrolysis of the inhibitor by HSL using LC-MS. The results
show clearly that compound 7600 is a pseudo-substrate of HSL as
attested by the production of compound P (see Figs. 4e6). The
mechanism could involve a nucleophilic attack by the hydroxy
group of the catalytic Ser of the enzyme on the carbon atom of the
carbonyl moiety of the oxadiazolone ring of the inhibitor. This
attack leads to the formation of a stoichiometric enzymeeinhibitor
covalent complex of a “long-lived” acyl-enzyme type is formed
between the open oxadiazolone ring of compound 7600 and the
catalytic serine of the lipase leading to of covalent enzyme-
einhibitor intermediate. The last step is the deacylation reaction in
which the compound P is produced (see Fig. 6).
In light of the above results, one can reasonably hypothesized
that the covalent binding of compound 7600 to HSL happens in two
conformations as depicted in the lower panel of the kinetic model
shown in Fig. 7. A first fraction of the lipase molecules will form in
the aqueous phase a covalent complex with the open oxadiazolone
ring of compound 7600 in a first conformation (Eecompound
Various examples were reported in line with the existence of
two configurations of the ligand molecules in the active site serine
of the lipase. Cygler et al. [38] have reported the three dimensional
(3D) structures of covalent complexes of Candida rugosa lipase with
two enantiomeric (1R and 1S-menthyl hexylphosphonate)
transition-state analogs for the hydrolysis of menthyl esters. The 1R
enantiomer, derived from the fast-reacting enantiomer of menthol,
and the 1S, derived from the slow-reacting enantiomer, were found
to bind to the lipase in two different orientations. Furthermore, in
contrast to the fast-reacting enantiomer, in the lipase/slow-reacting
enantiomer the imidazole ring of the residue His-449 was found to
have rotated, thus disrupting its hydrogen bond with the menthol
oxygen atom, which probably explains the differences in reactivity
between the two enantiomers [38]. Furthermore, Egloff et al. [39]
have reported that the two enantiomers of a C11 alkyl phospho-
nate compound, covalently bound to the active serine residue of
HPL, occupy two different conformations with a different occu-
pancy levels (57% and 43% for R and S conformations, respectively).
As in the case of C. rugosa lipase reported above, in one confor-
mation the methoxy oxygen of the first enantiomer forms
a hydrogen bond with the residue His-263, whereas in the second
enantiomer this hydrogen bond is lacking [39].
7
600o
complex with compound 7600 but in a second conformation
Eecompound 7600o ) (Fig. 7). The formation of a non-covalent
1
) whereas the other fraction will also form a covalent
(
2
HSLecompound 7600c (Compound 7600 in its closed conforma-
tion) is not expected to occur, since results using MALDI-TOF mass
spectrometry have shown the presence of only one form of
HSLecompound 7600 complex [22]. Indeed, if the non-covalent
HSLecompound 7600c complex exists, mass spectra should
contain, in addition to the covalent HSLecompound 7600o
complex, the spectra of pure enzyme [22].
Uppenberg et al. [40] performed structural studies on the lipase
B from Candida antarctica, which showed the presence of a stereo-
specific pocket for secondary alcohols. In their conclusion to this
study, the authors stated that they could not rule out the possibility
that the two enantiomers of the phosphonate inhibitor may bind
competitively to the serine. This would give rise to a mixture of
In the assay system, when the mixture (E, compound 7600c,
Eecompound 7600o
emulsion, the less stable complex (E*ecompound 7600o
cates molecules at the water/lipid interface) dissociates by an
1
and Eecompound 7600o
2
) is added to a TAG
1
) (* indi-

144
Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
Residual HSL activity measurement
TAG emulsion
E*-FFA
DAG +
compound P
DAG +
compound P
DAG
TAG
H O
2
kT1
kT2
TAG
TAG
FFA
E*-compound
E*- compound
E*
7
600o1
7600o2
Sampling at variable incubation period
Aqueous phase
Compound 7600c
E
E-compound
E-compound
7600o2
ka1
ka2
7
600o1
Incubation of HSL with compound 7600
Fig. 7. Kinetic model illustrating the inhibition of HSL by compound 7600 in the aqueous phase and its reactivation at a lipidewater interface. Symbols and abbreviations are as
follows: E, free enzyme (molecule/volume); E*, interfacial enzyme (molecule/surface); E*eFFA, interfacial enzymeefatty acid complex (molecule/surface); compound 7600c, the
closed and reactive compound 7600 in the bulk (molecule/volume); Eecompound 7600o
Eecompound 7600o , form 2 of the covalent enzymeecompound 7600 complex in the bulk (molecule/volume); E*ecompound 7600o
meecompound 7600 complex (molecule/surface); E*ecompound 7600o , interfacial form 2 of the covalent enzymeecompound 7600 complex (molecule/surface). TAG, tri-
acylglycerol at the interface (molecule/surface); DAG, diacylglycerol at the interface (molecule/surface).
1
, form 1 of the covalent enzymeecompound 7600 complex in the bulk (molecule/volume);
2
1
, interfacial form 1 of the covalent enzy-
2
conformations in the acyl and alcohol moieties of the inhibitor
present in the active site pocket. The enzyme’s preference for one of
the two enantiomers is therefore mainly governed by the size of the
two side chains and their ability to make favorable interactions in
the active site of the enzyme. Yapoudjian et al. [41] have resolved
the 3D-structure of an inactive lipase mutant from Thermomyces
lanuginosus lipase co-crystalised with the oleic acid in mixed
micelles with bile salts. In this structure, the oleic acid was found to
bind to the catalytic cavity of the lipase in two different orienta-
tions. In a lipase molecule, oleic acid occupied a conventional sn-1
binding site, and in a second lipase molecule, oleic acid was trapped
of HSL should therefore be of pharmacological interest and could be
used to find means of reducing FFA levels and decreasing peripheral
insulin resistance.
Acknowledgments
This work was supported by the CNRS and by the Agence
Nationale de la Recherche (ANR PCV 2007-184840 PHELIN).
References
ꢁ
in an unexpected fashion, being rotated approximately 180 in
respect to the main sn-1 lipase binding site alkyl chain binding site.
In view of the structural studies mentioned above on the various
ligand configurations present at the active sites of lipases, the
resolution of the 3D-structure of oxadiazolone molecules bound to
HSL would provide direct support for the validity of the kinetic
model presented in Fig. 7.
[1] P. Arner, Insulin resistance in type 2 diabetes: role of fatty acids, Diabetes.
Metab. Res. Rev. 18 (Suppl. 2) (2002) S5eS9.
[
2] Y.T. Kruszynska, D.S. Worrall, J. Ofrecio, J.P. Frias, G. Macaraeg, J.M. Olefsky, Fatty
acid-induced insulin resistance: decreased muscle PI3K activation but
unchanged Akt phosphorylation, J. Clin. Endocrinol. Metab. 87 (2002) 226e234.
3] M. Roden, T.B. Price, G. Perseghin, K.F. Petersen, D.L. Rothman, G.W. Cline,
G.I. Shulman, Mechanism of free fatty acid-induced insulin resistance in
humans, J. Clin. Invest. 97 (1996) 2859e2865.
[
Targeting FFA metabolism may offer a therapeutic approach that
address the fundamental cause of insulin resistance and type 2
diabetes and may thus lead to the development of new therapeutic
methods. The data obtained in this study on the specific inhibition
[4] C. Holm, T. Osterlund, H. Laurell, J.A. Contreras, Molecular mechanisms
regulating hormone-sensitive lipase and lipolysis, Annu. Rev. Nutr. 20 (2000)
365e393.
[5] G. Haemmerle, R. Zimmermann, M. Hayn, C. Theussl, G. Waeg, E. Wagner,
W. Sattler, T.M. Magin, E.F. Wagner, R. Zechner, Hormone-sensitive lipase

Y. Ben Ali et al. / Biochimie 94 (2012) 137e145
145
deficiency in mice causes diglyceride accumulation in adipose tissue, muscle,
and testis, J. Biol. Chem. 277 (2002) 4806e4815.
6] F.B.Kraemer, W.J.Shen, Hormone-sensitivelipase: controlofintracellular tri-(di-
[23] K. Schoenafinger, S. Petry, G. Müller, K.-H. Baringhaus, Preparation of 3-
phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones as hormone-sensitive lipase inhibi-
tors, PCT Int. Appl. (2001) 80 WO 0351841 A2.
[
[
)
acylglycerolandcholesteryl ester hydrolysis, J. Lipid. Res. 43 (2002) 1585e1594.
[24] Y. Ben Ali, H. Chahinian, S. Petry, G. Muller, F. Carriere, R. Verger,
A. Abousalham, Might the kinetic behavior of hormone-sensitive lipase reflect
the absence of the lid domain? Biochemistry 43 (2004) 9298e9306.
[25] M.M. Bradford, A rapid and sensitive method for quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding, Anal.
Biochem. 72 (1976) 248e254.
7] F.B. Kraemer, W.J. Shen, K. Harada, S. Patel, J. Osuga, S. Ishibashi, S. Azhar,
Hormone-sensitive lipase is required for high-density lipoprotein cholesteryl
ester-supported adrenal steroidogenesis, Mol. Endocrinol. 18 (2004) 549e557.
8] J. Osuga, S. Ishibashi, T. Oka, H. Yagyu, R. Tozawa, A. Fujimoto, F. Shionoiri,
N. Yahagi, F.B. Kraemer, O. Tsutsumi, N. Yamada, Targeted disruption of
hormone-sensitive lipase results in male sterility and adipocyte hypertrophy,
but not in obesity, Proc. Natl. Acad. Sci. U S A 97 (2000) 787e792.
[
[26] A. Tiss, F. Carriere, R. Verger, Effects of gum arabic on lipase interfacial binding
and activity, Anal. Biochem. 294 (2001) 36e43.
[
9] S.P. Wang, N. Laurin, J. Himms-Hagen, M.A. Rudnicki, E. Levy, M.F. Robert,
L. Pan, L. Oligny, G.A. Mitchell, The adipose tissue phenotype of hormone-
sensitive lipase deficiency in mice, Obes. Res. 9 (2001) 119e128.
[27] S. Ransac, Y. Gargouri, F. Marguet, G. Buono, C. Beglinger, P. Hildebrand,
H. Lengsfeld, P. Hadvary, R. Verger, Covalent inactivation of lipases, Meth.
Enzymol. 286 (1997) 190e231.
[
[
[
10] J.A. Rodriguez, Y. Ben Ali, S. Abdelkafi, L.D. Mendoza, J. Leclaire, F. Fotiadu, G. Buono,
F. Carriere, A. Abousalham, In vitro stereoselective hydrolysis of diacylglycerols by
hormone-sensitive lipase, Biochim. Biophys. Acta 1801 (2010) 77e83.
11] Y. Ben Ali, F. Carriere, R. Verger, S. Petry, G. Muller, A. Abousalham, Continuous
monitoring of cholesterol oleate hydrolysis by hormone-sensitive lipase and
other cholesterol esterases, J. Lipid. Res. 46 (2005) 994e1000.
12] R. Zimmermann, J.G. Strauss, G. Haemmerle, G. Schoiswohl, R. Birner-
Gruenberger, M. Riederer, A. Lass, G. Neuberger, F. Eisenhaber, A. Hermetter,
R. Zechner, Fat mobilization in adipose tissue is promoted by adipose
triglyceride lipase, Science 306 (2004) 1383e1386.
[28] R. Verger, M.C. Mieras, G.H. de Haas, Action of phospholipase A at interfaces,
J. Biol. Chem. 248 (1973) 4023e4034.
[29] S. Petry, Y. Ben Ali, H. Chahinian, H. Jordan, H. Kleine, G. Muller, F. Carriere,
A. Abousalham, Sensitive assay for hormone-sensitive lipase using NBD-
labeled monoacylglycerol to detect low activities in rat adipocytes, J. Lipid.
Res. 46 (2005) 603e614.
[30] S. Longhi, M. Czjzek, V. Lamzin, A. Nicolas, C. Cambillau, Atomic resolution (1.0
A) crystal structure of Fusarium solani cutinase: stereochemical analysis,
J. Mol. Biol. 268 (1997) 779e799.
[31] A. Tiss, H. Lengsfeld, F. Carrière, R. Verger, Inhibition of human pancreatic
lipase by tetrahydrolipstatin: further kinetic studies showing its reversibility,
J. Mol. Catal. B Enzym. 58 (2009) 41e47.
[32] A. Tiss, N. Miled, R. Verger, Y. Gargouri, A. Abousalham, Digestive lipases
inhibition: an in vitro study. in: G. Müller, S. Petry (Eds.), Lipases and Phos-
pholipases in Drug Development. Wiley-VCH, 2004, pp. 155e193.
[33] B. Borgström, Mode of action of tetrahydrolipstatin: a derivative of the
naturally occurring lipase inhibitor lipstatin, Biochim. Biophys. Acta 962
(1988) 308e316.
[34] A. Lookene, N. Skottova, G. Olivecrona, Interactions of lipoprotein lipase with
the active-site inhibitor tetrahydrolipstatin (Orlistat), Eur. J. Biochem. 222
(1994) 395e403.
[
[
13] R. Lehner, D.E. Vance, Cloning and expression of a cDNA encoding a hepatic
microsomal lipase that mobilizes stored triacylglycerol, Biochem. J. 343 (Pt 1)
(
1999) 1e10.
14] R. Birner-Gruenberger, H. Susani-Etzerodt, M. Waldhuber, G. Riesenhuber,
H. Schmidinger, G. Rechberger, M. Kollroser, J.G. Strauss, A. Lass,
R. Zimmermann, G. Haemmerle, R. Zechner, A. Hermetter, The lipolytic pro-
teome of mouse adipose tissue, Mol. Cell. Proteomics 4 (2005) 1710e1717.
15] V.W. Dolinsky, S. Sipione, R. Lehner, D.E. Vance, The cloning and expression of
a murine triacylglycerol hydrolase cDNA and the structure of its corre-
sponding gene, Biochim. Biophys. Acta 1532 (2001) 162e172.
[
[
16] E. Wei, Y. Ben Ali, J. Lyon, H. Wang, R. Nelson, V.W. Dolinsky, J.R. Dyck,
G. Mitchell, G.S. Korbutt, R. Lehner, Loss of TGH/Ces3 in mice decreases blood
lipids, improves glucose tolerance, and increases energy expenditure, Cell
Metab. 11 (2010) 183e193.
[35] A.P. Potthoff, L. Haalck, F. Spener, Inhibition of lipases from Chromobacterium
viscosum and Rhizopus oryzae by tetrahydrolipstatin, Biochim. Biophys. Acta
1389 (1998) 123e131.
[
17] L. Vertesy, B. Beck, M. Bronstrup, K. Ehrlich, M. Kurz, G. Muller, D. Schummer,
G. Seibert, Cyclipostins, novel hormone-sensitive lipase inhibitors from
Streptomyces sp. DSM 13381. II. Isolation, structure elucidation and biological
properties, J. Antibiot. (Tokyo) 55 (2002) 480e494.
[36] P. Hadvàry, W. Sidler, W. Meister, W. Vetter, H. Wolfer, The lipase inhibitor
tetrahydrolipstatin binds covalently to the putative active site serine of
pancreatic lipase, J. Biol. Chem. 266 (1991) 2021e2027.
[37] Q. Luthi-Peng, H.P. Marki, P. Hadvary, Identification of the active-site serine in
human pancreatic lipase by chemical modification with tetrahydrolipstatin,
FEBS Lett. 299 (1992) 111e115.
[38] M. Cygler, P. Grochulski, R.J. Kazlauskas, J.D. Schrag, F. Bouthillier, B. Rubin,
A.N. Serreqi, A.K. Gupta, A structural basis for the chiral preferences of lipases,
J. Am. Chem. Soc. 116 (1994) 3180e3186.
[39] M.-P. Egloff, F. Marguet, G. Buono, R. Verger, C. Cambillau, H. van Til-
beurgh, The 2.46 Å resolution structure of the pancreatic lipase-colipase
complex inhibited by a C11 alkyl phosphonate, Biochemistry 34 (1995)
2751e2762.
[
[
18] D. Wagle, S.D. Fang, I. Terlecky, J. Resek, J. Egan, Preparation and formulation
of eudesmanolides for use as glucose and lipid lowering compounds, PCT Int.
Appl. (2000) 41 WO 0027388 A1.
19] S. Ebdrup, L.G. Sorensen, O.H. Olsen, P. Jacobsen, Synthesis and structure-
activity relationship for a novel class of potent and selective carbamoyl-
triazole based inhibitors of hormone sensitive lipase, J. Med. Chem. 47
(
2004) 400e410.
[
[
[
20] D.H. Slee, A.S. Bhat, T.N. Nguyen, M. Kish, K. Lundeen, M.J. Newman,
S.J. McConnell, Pyrrolopyrazinedione-based inhibitors of human hormone-
sensitive lipase, J. Med. Chem. 46 (2003) 1120e1122.
21] J.C. de Jong, L.G. Sorensen, H. Tornqvist, P. Jacobsen, Carbazates as potent
inhibitors of hormone-sensitive lipase, Bioorg. Med. Chem. Lett. 14 (2004)
[40] J. Uppenberg, N. Ohrner, M. Norin, K. Hult, G.J. Kleywegt, S. Patkar, V. Waagen,
T. Anthonsen, T.A. Jones, Crystallographic and molecular-modeling studies of
lipase
B from Candida antarctica reveal a stereospecificity pocket for
1741e1744.
secondary alcohols, Biochemistry 34 (1995) 16838e16851.
[41] S. Yapoudjian, M.G. Ivanova, A.M. Brzozowski, S.A. Patkar, J. Vind,
A. Svendsen, R. Verger, Binding of Thermomyces (Humicola) lanuginosa lipase
to the mixed micelles of cis-parinaric acid/NaTDC, Eur. J. Biochem. 269 (2002)
1613e1621.
22] Y. Ben Ali, H. Chahinian, S. Petry, G. Muller, R. Lebrun, R. Verger, F. Carriere,
L. Mandrich, M. Rossi, G. Manco, L. Sarda, A. Abousalham, Use of an inhibitor to
identify members of the hormone-sensitive lipase family, Biochemistry 45
(
2006) 14183e14191.