Tài liệu Báo cáo khoa học: Multi-targeted activity of maslinic acid as an antimalarial natural compound pdf

Bautista1,3, Amalia Diez1,3and Antonio Puyet1,3 1 Departamento de Bioquı´mica y Biologı´a Molecular IV, Facultad de Veterinaria, Universidad Complutense de Madrid, Spain 2 Chemogenomics

natural compound

Carlos Moneriz1,4, Jordi Mestres2, Jose´ M Bautista1,3, Amalia Diez1,3and Antonio Puyet1,3

1 Departamento de Bioquı´mica y Biologı´a Molecular IV, Facultad de Veterinaria, Universidad Complutense de Madrid, Spain

2 Chemogenomics Laboratory, Research Unit on Biomedical Informatics (GRIB), Institut Municipal d’Investigacio´ Me`dica and

Universitat Pompeu Fabra, Barcelona, Spain

3 Instituto de Investigacio´n del Hospital 12 de Octubre, Universidad Complutense de Madrid, Spain

4 Departamento de Bioquı´mica, Facultad de Medicina, Universidad de Cartagena, Colombia

Keywords

apicomplexa; merozoite surface protein;

metalloprotease inhibition; PfSUB1;

phospholipase; plasmodium

Correspondence

A Puyet, Departamento de Bioquı´mica y

Biologı´a Molecular IV, Facultad de

Veterinaria, Universidad Complutense de

Madrid, E28040 Madrid, Spain

Fax: +34 913 943 824

Tel: +34 913 943 827

E-mail: apuyet@vet.ucm.es

(Received 21 April 2011, revised 14 June

2011, accepted 17 June 2011)

doi:10.1111/j.1742-4658.2011.08220.x

Most drugs against malaria that are available or under development target

a single process of the parasite infective cycle, favouring the appearance of resistant mutants which are easily spread in areas under chemotherapeutic treatments Maslinic acid (MA) is a low toxic natural pentacyclic triterpene for which a wide variety of biological and therapeutic activities have been reported Previous work revealed that Plasmodium falciparum erythrocytic cultures were inhibited by MA, which was able to hinder the maturation from ring to schizont stage and, as a consequence, prevent the release of merozoites and the subsequent invasion We show here that MA effectively inhibits the proteolytic processing of the merozoite surface protein com-plex, probably by inhibition of PfSUB1 In addition, MA was also found

to inhibit metalloproteases of the M16 family by a non-chelating mecha-nism, suggesting the possible hindrance of plasmodial metalloproteases belonging to that family, such as falcilysin and apicoplast peptide-process-ing proteases Finally, in silico target screenpeptide-process-ing was used to search for other potential binding targets that may have remained undetected Among the targets identified, the method recovered two for which experimental activity could be confirmed, and suggested several putative new targets to which

MA could have affinity One of these unreported targets, phospholipase A2, was shown to be partially inhibited by MA These results suggest that

MA may behave as a multi-targeted drug against the intra-erythrocytic cycle of Plasmodium, providing a new tool to investigate the synergistic effect of inhibiting several unrelated processes with a single compound,

a new concept in antimalarial research

Introduction

As long as effective vaccines against malaria remain

unavailable, the search for new antimalarial drugs is

still required because of the incomplete protection

obtained with the present therapeutic methods and the

emergence of resistant strains in endemic regions Most

present and prospective drugs against Plasmodium

falciparum, the causative agent of the most virulent form of human malaria, have been designed to inter-fere with essential processes at the blood stage of the parasite [1], which accounts for the main clinical symp-toms of disease Despite the wide variety of potential targets identified in the intra-erythrocytic cycle of

Abbreviations

IC 50 , half maximal inhibitory concentration; MA, maslinic acid; MSP, merozoite surface protein; PLA2, phospholipase A2; RBCs, red blood cells; SERA, serine repeat antigen.

P falciparum [2] only a few drugs have found

applica-tion as therapeutic agents, like those interfering with

hemozoin polymerization in the vacuole (chloroquine,

quinine, mefloquine and other alkaloids), the

dihydro-folate pathway (pyrimethamine, sulphadoxine,

progua-nil), the mitochondrial electron transport chain

(atovaquone) or triggering of oxidative stress

(artemis-in and derivatives, primaqu(artemis-ine)

Maslinic acid (MA) is a natural pentacyclic

triter-pene found in the olive fruit [3] Different activities

have been reported for MA in a variety of biological

systems In addition to reported antioxidant [4,5],

va-sorelaxation [6] and anti-tumoural [7,8] activities, MA

has been shown to specifically inhibit glycogen

phos-phorylase [9,10] and protein tyrosine phosphatase 1B

[11], exerting anti-diabetic action Inhibition of HIV

protease has also been reported for MA as well as

other structurally related compounds like ursolic,

epi-pomolic and tormentic acids [12] MA appears to

dis-play antiparasitic activities in apicomplexa [13], further

demonstrated in Toxoplasma gondii cultures, where the

likely inhibition of proteolytic activity leads to a

reduc-tion in gliding motility and ultra structural alterareduc-tions

of the parasite [14]

Previous work from our laboratory showed that

MA inhibits the progress of the intra-eythrocytic stages

of P falciparum, both in vitro [15] and in vivo [16]

Depending on the timing and extent of treatment,

parasites cultured in the presence of MA display

accumulation of ring, trophozoite or schizont

intra-erythrocytic forms At low MA doses, the inhibition is

reversible, as removal of MA from the cultures relieves

this hindrance, allowing further maturation of the

parasite The use of parasitostatic drugs has not been

investigated as a possible alternative or complement to

current drug therapies Parasitostatic drugs may

enhance the host immune response by delaying the

infection progress and thus facilitating the presentation

of plasmodial antigens during the first infective stages,

therefore favouring the development of the acquired

immune response [16–18]

The actual target of MA on P falciparum remains

to be investigated MA does not hinder the formation

of hematin [15], discarding a possible interference

with the formation of hemozoin Among the above

mentioned previously identified biological processes

affected by MA, the inhibition of proteases and⁄ or

protein tyrosine phosphatases appears, a priori, as

potential targets for this compound in Plasmodium

While little is known on protein tyrosine phosphatase

activities in P falciparum, extensive work has been

devoted to finding and using specific inhibitors of

Plas-modiumproteases The main source of amino acids for

plasmodial protein synthesis derives from haemoglobin degradation in the food vacuole by a process which involves several proteases: plasmepsins, falcipains and falcilysins Plasmepsin II inhibitors have been devel-oped based on the structure of the available inhibitors

of cathepsin D [19], a lysosomal protease of mamma-lian cells, and chalcones and phenothiazines have been assayed as inhibitors of falcipain-2 [20,21] Inhibition

of the metalloproteases falcilysin and neutral amino-peptidases acting at the terminal stages of haemoglobin degradation are also considered as potential antimalar-ial targets [22,23] Other proteases related to the matu-ration of parasites and the invasive process are also investigated as possible targets for specific antimalarial drugs: inhibition of merozoite surface protein (MSP1) processing protease (PfSUB1) has been shown to reduce erythrocyte invasion [24], while proteases involved in merozoite egress of the red blood cell, like the serine repeat antigen (SERA) family also regulated

by PfSUB1 proteolytic activiy [25], have been shown

to be required for the late-stage development of para-sites [26]

Computational ligand-based approaches to predict the potential affinity of compounds have been devel-oped in the last years These methods allow the virtual screening of proteins showing the potential to bind a given compound among large chemical collections This methodology was recently applied to analyse the polypharmacology of drugs [27,28], to design chemical libraries directed to protein families [29] and to analyse the chemogenomic space of cardiovascular diseases [30] In this report, a comprehensive analysis of poten-tial inhibitory activities of MA on P falciparum has been carried out, focusing first on the putative protease inhibition Furthermore a chemogenomic-based screen-ing usscreen-ing MA as ligand to predict its most probable targets was performed, searching for enzymatic activi-ties and protein binding structures which could eventu-ally reveal new plasmodial target molecules for this triterpene and novel strategies in malaria therapy

Results and Discussion

Inhibition of proteases by MA

It has been previously proposed that MA may inhibit the activity of proteases of T gondii [14] and HIV [12,31] To ascertain the inhibition range of MA on the different P falciparum protease classes, in vitro enzymatic assays were performed encompassing cyste-ine, aspartic, serine and metalloproteases The results, shown in Table 1, indicate that MA is a strong inhibi-tor of the metalloprotease thermolysin, showing also

low half maximal inhibitory concentration values

(IC50) for serine and cysteine proteases Remarkably,

no inhibition was observed on the aspartic protease

pepsin However, a possible specific inhibition of

plas-modial aspartic proteases could not be discarded, as

strong inhibition by MA on the HIV protease, which

belongs to the aspartic protease catalytic class, was

previously reported [12] Accordingly, an additional

inhibition assay was performed using P falciparum

protein extracts and including cathepsin D (an aspartic

protease) and the aspartic protease inhibitor pepstatin

A as controls The results (Table 2) showed that MA

does not inhibit cathepsin D nor the aspartic protease

activity in plasmodial protein extracts A limited

inhi-bition by MA was observed in extracts obtained from

leukocytes In contrast, the protease inhibitor pepstatin

A showed strong inhibitory activity on all samples

These results can be explained assuming that MA may

behave as a specific inhibitor of HIV protease, or the

protease class A2 to which the HIV protease belongs,

showing no activity on class A1 proteases (pepsin,

cathepsin D) Remarkably, the only aspartic proteases

predicted from comparative genomic analysis in P

fal-ciparum belong to the A1 class [32], thus explaining

the lack of inhibition observed in the protein extracts

These results support the data reported on T gondii infections [14], and suggest that the parasitostatic effect

of MA on P falciparum infected erythrocytes may be mediated by the inhibition of one or more proteases, probably corresponding to metalloproteases, serine and⁄ or cysteine proteases, which are required to reach the schizont stage

The proteolytic hydrolysis of haemoglobin as a source of amino acids constitutes one of the essential processes which take place along the intra-erythrocytic stage of the parasite Degradation of haemoglobin is performed in the food vacuole by the combined action

of aspartic proteases (plasmepsins I, II, IV and histoas-partic protease), cysteine proteinases (falcipains) and a metalloprotease (falcilysin) [33] The resulting small peptides are reduced to dipeptides by aminopeptidases [23] which may be further hydrolysed to free amino acids outside the digestive vacuole [34,35] It has been shown that cysteine protease inhibitors, such as vinyl sulfones, reduce the initial cleavage of globin peptides

in the trophozoite vacuole [36,37] This effect has been explained either by the direct inhibition of falcipain [38] or by the indirect effect on the functionality of the vacuole as a result of the accumulation of partially hydrolysed peptides, leading to the accumulation of uncleaved globin [39] The possible effect, either direct

or indirect, of MA on globin hydrolysis was tested by incubating synchronized ring-stage parasites with MA

or leupeptin, a cysteine protease inhibitor, and visuali-zation of the globin band by SDS⁄ PAGE The results did not show the characteristic accumulation of globin

in MA-treated cultures (Fig 1), indicating that the ini-tial hydrolysis of haemoglobin is not inhibited by MA

In addition, the morphology of infected erythrocytes incubated in the presence of MA is visibly different from leupeptin-treated cultures As can be seen in Fig 1B, the food vacuoles of parasites incubated 24 h with leupeptin were abnormally dark-stained due to the blockage in globin hydrolysis, while MA-treated cultures showed abnormal trophozoite morphology due to the growth arrest, but no accumulation of glo-bin or vacuolization These results confirm that MA does not hinder the initial processing of globin and, in consequence, it is unlikely that falcipains are targeted

by the drug

Effect of MA on the activity of P falciparum MSP

As shown in Table 1, MA inhibits subtilisin with an

IC50in the range of 50 lm Subtilisin is a serine prote-ase of the S8 family closely related to the subtilprote-ases 1 (PfSUB1), 2 (PfSUB2) and 3 (PfSUB3) reported in

P falciparum [32,40] These proteins play an essential

Table 1 Effect of MA on different representative proteases All

tests were performed at a fixed concentration of enzyme and

selected according to the detection limit The MA concentration

range was 1–400 l M

Enzyme

Protease

class

Enzyme (mUÆmL)1)

Detection limit (mUÆmL)1) IC50MA (l M )

Table 2 Effect of MA on aspartic protease activity in P falciparum

protein extracts MA and pepstatin A were tested at 300 l M

Enzyme or total

role in the erythrocyte invasion by the parasite

mero-zoite through a mechanism involving the discharge of

PfSUB1 into the parasitophorous vacuole and the

pro-teolytic activation of SERA proteases, which are

required for merozoite egress [41,42] An additional

role in the maturation of MSPs (MSP1, MSP6 and

MSP7) has also been recently reported for PfSUB1

[24] MSPs are involved in the merozoite invasion of

erythrocytes [43] PfSUB1 function is complemented

by the reported activity of PfSUB2, which performs a

secondary extracellular processing step on the MSP

complex [44] Due to their similarity with subtilisin,

these subtilases may be expected to be inhibited by

MA To verify this hypothesis, parasite proteins were

analysed by western blot with mouse anti-P falciparum

MSP1 Incubation of synchronized cultures at schizont

stage with MA for 12 h led to the inhibition of MSP1

processing, revealed by the detection of the 195 kDa

band, which was not detectable in untreated cultures

at the same cycle time (Fig 2A) Furthermore, the

morphology of 12 h cultures treated with MA showed

a delay in the maturation of schizonts, which can be

associated with the inhibition of the MSP1 processing

(Fig 2B) These results indicate that at least PfSUB1

is inhibited by MA in erythrocyte cultures,

corroborat-ing that this compound behaves as a serine protease

inhibitor Remarkably, the only specific inhibitor of

PfSUB1 reported before is also a natural product,

MRT12113 [24] (Fig 2C), showing few structural

simi-larities with MA A comparative study on both

mole-cules might help in the design of simpler structures

behaving as specific inhibitors of this protease

Noteworthy, it is well established that antibodies against different regions of the MSP1 complex are present in populations showing a level of immunity to

P falciparum malaria [45], and Plasmodium yoelii poly-morphic variant MSP7-3 has been used to immunize mice against blood stage infection [46] The immuniza-tion of ICR mice treated with MA after a primary infection observed in our laboratory [16] could then be related to the inhibition of MSP processing reported here, as the prolonged exposure of unprocessed MSP complex would allow the selection of specific neutraliz-ing antibodies able to bind to released merozoites, thus hindering invasion of new red blood cells (RBCs)

Chelation-independent protease inhibition by MA

As shown in Table 1, MA is a potent inhibitor of thermolysin, a bacterial zinc metalloprotease belonging

to the M16 family [47] Non-specific inhibition of metalloprotease activity can readily be achieved by chelating agents that bind to metal cations required in the active site of the enzyme The observed inhibition of

MA on thermolysin and PfSUB1, a calcium-dependent serine protease, might then also be explained if MA behaves as chelating agent on divalent cations To test this possibility, a colorimetric chelation assay was car-ried out using zinc as divalent cation As shown in the Fig 3, no significant chelation capacity was detected for MA, even at higher concentrations than those used

in the treatments This result shows that MA inhibits metalloprotease and PfSUB1 activities by a specific, non-chelating mechanism and also reinforces the

250 98 50

22

6

Leupeptin (24 h) Start rings (0 h) Untreated (24 h)

Maslinic acid (24 h)

Fig 1 MA-treated cultures do not accumulate undegraded haemoglobin Synchronized ring-stage P falciparum was cultured in the pres-ence of 100 l M MA or 100 l M leupeptin for 18 h, followed by extraction of proteins and parasite visualization in thin blood smears (A) Coo-massie Blue stained 15% SDS ⁄ PAGE of total protein from infected RBC after 18 h of culture, which corresponds to the trophozoite stage Lane 1, untreated control parasites; lane 2, culture incubated with leupeptin; lane 3, parasites incubated with MA; lane 4, human haemoglo-bin standard 14 kDa bands correspond to undegraded glohaemoglo-bin monomers Parasites incubated with the cysteine proteinase inhibitor leupeptin accumulated undegraded globin, while no differences with the untreated control were observed in MA-treated cultures (B) Morphological changes in infected RBC in drug-treated cultures: aliquots of the cultures described above before and after the addition of inhibitor were obtained at 18 h and stained with Wright’s The food vacuoles of parasites incubated with leupeptin were abnormally dark-staining due to a block in globin hydrolysis Control parasites matured to trophozoite state Parasites incubated with MA generated abnormal trophozoites, but

no accumulation of haemoglobin was observed.

possible inhibition exerted by MA on plasmodial key

metalloprotease activities required for the maturation

of the trophozoite MA might also specifically inhibit

metalloproteases of the thermolysin class M16 Several

candidates of this family, playing important roles in the

parasite erythrocytic cycle, have been identified by data

mining of P falciparum proteome [32]: falcilysin, in its

dual role as haemoglobin peptidase and transit peptide

processing activity in the apicoplast [48]; the

mito-chondrial processing peptidases (PFE1155c, PFI1625c),

or insulysin and pitrilysin [32], possibly involved in the

processing of apicoplast protein leader sequence

Inhibition of any of these activities could contribute to

MA interference in the maturation of the parasite

Polypharmacology of MA

The observed inhibition on PfSUB1 may contribute to

the arrest of P falciparum infective cycle detected in

MA-treated cultures [15] However, the morphology of

the blocked parasite cannot be completely explained

by inhibition of MSP1 processing MSP1 is synthesized

from the onset of schizogony and is processed by

PfSUB1 at the time of merozoite egress from the

infected erythrocyte It has been previously shown that

incubation of synchronic cultures with a highly specific

inhibitor of PfSUB1 produced no apparent effect on

pre-schizont stages, but rather a very specific inhibition

of schizont rupture and reduced invasion of the

released merozoites, which can be revealed by accu-mulation of merozoite parasites in the cultures [41]

In contrast, cultures treated with MA display an increased fraction of ring, trophozoite or schizont stages [15], suggesting an additional inhibitory effect early in the intra-erythrocytic cycle The probable inhi-bition of plasmodial metalloproteases by MA opens up the possibility of a multi-targeted drug, interfering with different parasite processes and leading to a blockage

of parasite maturation in the RBC from early ring to schizont stages

To further investigate the extent of possible multi-targeted inhibitory activities of MA, a computational screening was carried out to identify potential binding targets for MA Here, in silico target screening was used to test MA against ligand-based models derived for 4500 proteins Table 3 compiles the list of six pro-teins identified by this method as putative targets for

MA Two of the proteins retrieved correspond to previously reported targets for MA, namely protein tyrosine phosphatase and glycogen phosphorilase, providing support for the validity of the approach

To the best of our knowledge, the other four proteins have never been suggested as possible targets for MA Nevertheless, there is compelling evidence of the connection between these targets and malaria Phos-pholipase A2 (PLA2) was detected in P falciparum-infected human erythrocytes and found to be inhibited

by the antimalarial drugs chloroquine, quinine and

207 78 53 35 19 6

114

A Control (0 h)

B

Control (12 h)

MA (12 h)

Start schizonts (0 h)

kDa

O

O

O HO

OH HO OH

HO OH

OH OH

HO HO

CH3

H3C

CH3 CH3

CH3

H3C CH3

COOH

H H H

C

Fig 2 Inhibition of MSP1 primary processing by MA Highly synchronized early schizonts of P falciparum 3D7 ( 36 h post-invasion) cul-tured for 12 h in the presence of MA (100 l M ) were used to obtain an extract of parasite proteins and Wright’s stained thin smears for mon-itorization of the infective stage (A) Western blot of the protein extract using the MSP1 antibody as probe The unprocessed MSP1 ( 195 kDa) remains present in cultures treated with MA, while no detectable band was observed in the untreated control (B) Morphology

of the infected erythrocytes before MA incubation (0 h) and after 12 h incubation in the presence or absence of MA 100 l M A delay in the maturation of schizonts is observed, which may be related to inhibition of MSP1 processing (C) Structures of MA and the reported PfSUB1 inhibitor MTR12113.

artetether at concentrations that cause 50% inhibition

at millimolar concentrations of those drugs [49]

Although the other three possible targets may not be

involved in the effect of MA on the intra-erythrocytic

cycle, they could be relevant to other aspects of

malaria therapy The polymorphic cytochrome P450

(CYP) isoform 2C8 has been reported to be actively

involved in drug efficacy due to its capacity to

metab-olize antimalarial drugs in humans [50] On the other

hand, selective and irreversible inhibitors of mosquito

acetylcholinesterases for controlling malaria and other

mosquito-borne diseases have recently been described

[51] Finally, even though the serotonin 5-HT2B

receptor subtype has not yet been specifically related

to malaria, there are reports linking serotonin

recep-tors in general as potential targets mediating

differen-tial chemical phenotypes in P falciparum [52] Given

the high levels of cross-pharmacology among amine

G-protein-coupled receptors [29], if some of them

have been linked already to malaria, the remaining

members of this subfamily could be relevant to

malaria as well

Among the four novel putative targets identified,

MA was tested on PLA2, since it is the target showing

the highest predicted affinity The results obtained are collected in Table 4 As can be observed, MA inhibits PLA2 in a dose-dependent manner up to 25% at

400 lm, which may be comparable to the 50% inhibi-tion of PLA2 reported for other antimalarial drugs at millimolar concentrations [49] Accordingly, PLA2 may indeed be considered a new target for MA The inhibition of plasmodial PLA2, although incomplete, can be related to the lipid metabolism and membrane dynamics, contributing to the overall effect of this compound on parasite maturation when combined with the observed PfSUB1 and metalloprotease inhibi-tion

It is worth stressing that in silico target screening of

MA did not point to any protease or peptidase as potential target for MA Considering the increasing body of evidence of protease inhibition activity of

MA, gathered in this and previous reports, the fact that MA was found outside the current chemical cov-erage for those targets may suggest a non-standard binding mechanism to these proteins Most natural or designed protease inhibitors mimic the structure of a flexible peptidic molecule to bind to the active site This approach is based on the observation that prote-ases frequently bind their inhibitors⁄ substrates in extended or b-strand conformation, requiring a linear arrangement of the substrate backbone [53] The pen-tacyclic triterpene structure of MA (Fig 2C), however, does not fit with that principle and may represent a new kind of protease-binding molecule displaying novel specific inhibition activities Two lines of evi-dence support such a target-specific notion: the reported inhibition of HIV protease [12,54] which is not extended to other aspartic proteases, and the fact that MA is present in a variety of food products, showing no toxic effects by inhibition of human prote-ases Should these arguments be confirmed experimen-tally, MA would be a valuable lead molecule for development of specific drugs against apicomplexa parasites

Concentration (m M )

0

20

40

60

80

100

50

Fig 3 Zinc chelating assay for MA Percentage of zinc chelation

detected using Eriochrome Black T as an indicator of

non-com-plexed zinc cations The assay was carried out by adding different

amounts of MA (black bars) or EDTA (grey bars) to a solution

con-taining 32 l M Zn 2+ Results are expressed as 100 · A 610 nm

(sam-ple) ⁄ A 610 nm (control without zinc).

Table 3 Results of the chemogenomic screening of proteins with

high probability of binding to MA.

Enzyme

MA predicted affinity (l M )

Table 4 Inhibition of PLA2 activity by MA MA and 3,4-dichloroiso-coumarin (DIC) were tested at 100 and 400 l M The compounds were incubated with the enzyme control and schizonts protein for

15 min and 2 h respectively Results are expressed as the percent-age of inhibition compared with the control with no inhibitor added Enzyme or total

protein extract Inhibitor

Inhibition at

100 l M (%)

Inhibition at

400 l M (%)

Materials and methods

Drugs and inhibitors

MA was kindly provided by Dr Garcı´a-Granados from the

University of Granada, Spain Leupeptin, pepstatin A and

3,4-dichloroisocoumarin were purchased from

Sigma-Aldrich (St Louis, MO, USA) All drugs were dissolved in

100% dimethylsulfoxide prior to assay

In vitro cultures of P falciparum

P falciparum strain 3D7 (clone MRA-102) was provided

by The Malaria Research and Reference Reagent

Resource Center (MR4, deposited by DJ Carucci)

Ery-throcytes (RBC) were obtained from type A+ healthy

human local donors and collected in tubes with

citrate-phosphate-dextrose anticoagulant (Vacuette Greiner

Bio-One GmbH, Kremsmu¨nster, Austria) The culture

med-ium consisted of standard RPMI 1640 (Sigma-Aldrich)

supplemented with 0.5% Albumax I (Life Technologies,

Paisley, UK), 100 lm hypoxanthine (Sigma-Aldrich),

25 mm HEPES (Sigma-Aldrich), 12.5 lgÆmL)1 gentamicine

(Sigma-Aldrich) and 25 mm NaHCO3 (Sigma-Aldrich)

Each culture was started by mixing uninfected and

infected erythrocytes to achieve a 1% haematocrit and

incubated in 5% CO2 at 37C in tissue culture flasks

(Iwaki Asahi Glass, Tokyo, Japan) The progress of

growth in the culture was determined by microscopy in

thin blood smears stained with Wright’s eosin methylene

blue solution (Merck, Darmstadt, Germany), using the

freely available plasmoscore software [56] to monitor the

parasitaemia A detailed description of the culture and

synchronization methods used has been published

previ-ously [57]

P falciparum protein extracts

Proteins from parasite extracts were obtained from a 25-mL

culture of infected RBCs The cultures were harvested

and the cells resuspended in 1 volume of saponin 0.2%

in NaCl⁄ Pi (PBS) 1· and vortexed gently for 5 s to lyse

the RBC membranes The released parasites were pelleted

at 10 000 g for 10 min and washed three times in cold

PBS The pellets were solubilized in 50 mm Tris⁄ HCl pH

8, 50 mm NaCl, 3% Chaps, 0.5% MEGA 10 and gently

mixed at 4C for 15 min followed by three freeze–thaw

cycles After centrifugation (13 000 g, 10 min, 4C) the

supernatant was collected and referred to as

para-site extract Approximately 10 lg of total protein

supernatant was boiled in electrophoresis sample buffer

[5% (wt⁄ vol) SDS, 62.5 mm Tris ⁄ HCI, 5% (vol ⁄ vol)

2-mercaptoethanol, pH 6.8] and separated on 15%

SDS⁄ PAGE

Protease inhibition assays The protease activities were conducted in vitro using the internally quenched fluorogenic peptide substrate (EnzChek Protease Assay Kit-E33758; Invitrogen) 50 lL of different protease dilutions, with or without MA at different concen-trations, were put into separate wells of a microplate After the addition of 50 lL of substrate working solution (1.57 mg

in 50% dimethylsulfoxide, 10 mm Tris⁄ HCl, pH 7.8) the plate was incubated at room temperature, protected from light, for 60 min The fluorescence intensity was measured at

485 nm excitation and 528 nm emission using a Perkin Elmer LS-50B luminescence spectrophotometer Background fluo-rescence was subtracted from the inset data Plots of percent-age control activity versus concentration of inhibitor were used to determine the concentrations that inhibited 50% of protease activity The enzymes evaluated, all purchased from Sigma-Aldrich, were as follows: subtilisin A (EC 3.4.21.62.) from Bacillus licheniformis (serine protease), papain (EC 3.4.22.2) from Carica papaya (cysteine protease), pepsin (EC 3.4.23.1) from porcine gastric mucosa (aspartic protease) and thermolysin (EC 3.4.24.27) from Bacillus thermoproteolyti-cus rokko (metalloprotease) Assays were performed in

10 mm Tris⁄ HCl (pH 7.8) except for pepsin, which was assayed in 10 mm HCl (pH 1.8) Dilutions of papain were made from a buffer containing 30 mm l-cysteine

The effect of MA on the activity of aspartic proteases present in parasite protein extracts was assayed spectroflu-orometrically with the internally quenched fluorescent substrate MCA-G-K-P-I-L-F-F-R-L-K(DNP)-D-Arg-NH2

trifluoroacetate salt (Sigma-Aldrich), which is not initially fluorescent due to quenching of the 7-methoxycoumarin-4-acetyl (MCA) group by the 2,4-dintrophenyl group (DNP) For the above, 15 lL of protein extract or the control enzyme (cathepsin D aspartic protease from bovine spleen provided by Sigma-Aldrich) were disposed in a 96-well mi-croplate, with or without MA (10 ll) 20 lL of reaction buffer (0.5 m sodium acetate at pH 5.0) were added and the volume completed to 98 lL with distilled water All sam-ples were incubated at 37C for 10 min to allow the inhibi-tion of the enzyme After this time, 2 lL of the substrate were added and the samples were incubated at 37C for

30 min Parasite protein extracts were also incubated with the aspartic protease inhibitor pepstatin A as a control Finally, the fluorescence signal was measured at 323 nm excitation and 398 nm emission using a Perkin Elmer LS-50B luminescence spectrophotometer

Plasmodial MSP processing assay

To analyse the possible inhibitory activity of MA on the pro-cessing of the MSP1, synchronous cultures of 3D7 schizonts (10% parasitaemia) were supplemented with either MA (100 lm) or dimethylsulfoxide only and cultured for 8–12 h

to allow schizont rupture and merozoite invasion The

para-site pellet was resuspended in PBS containing a complete

Protease Inhibitor Cocktail (Roche, Basel, Switzerland)

After incubation on ice for 10 min, proteins from parasite

extracts were obtained as described above Parasite proteins

were analysed by western blot with mouse anti-P falciparum

MSP1 (AbD Serotec, Oxford, UK), followed by horseradish

peroxide conjugated anti-mouse IgG (GE Healthcare,

Wau-keha, WI, USA) at a 1 : 5000 dilution [24] Detection was

performed using the Super Signal chemiluminescent substrate

(Thermo Fisher Scientific, Rockford, IL, USA) and exposure

to X-ray film Finally, aliquots from cultures grown were also

examined microscopically

Haemoglobin degradation assay

To assess the effect of MA on the haemoglobin

accumula-tion in trophozoites, synchronized ring-stage parasites at

10% parasitaemia were incubated at 37C in microtitre

plate cultures with MA (100 lm) or the cysteine proteinase

inhibitor leupeptin (100 lm) as a control After 18 h of

incubation, Wright-stained smears were prepared from

cul-tures, and the parasites were evaluated for the presence of a

marked food vacuole abnormality that has been correlated

with a block in haemoglobin degradation [37,58] To assess

haemoglobin accumulation, parasites cultured with

inhibi-tors as indicated were collected after 18 h incubation, and

proteins from parasite extracts were obtained as described

above, solubilized in electrophoresis sample buffer and

elec-trophoresed through 15% SDS⁄ PAGE [36,37] Proteins

were identified by staining the gel with Coomassie Blue

In silico target screening

Our in silico target screening approach relies on the

assump-tion that the set of bioactive ligands collected for a given

tar-get provides a complementary description of the tartar-get from

a ligand perspective In order to be able to process this

infor-mation efficiently, chemical structures need to be encoded

using some sort of mathematical descriptors In this work,

three types of two-dimensional descriptors were used, namely

phrag, fpd and shed, each one of them characterizing

chem-ical structures with a different degree of fuzziness and thus

complementing each other in terms of structural

congenerici-ty and hopping abilities [59] Then, the probabilicongenerici-ty of any

molecule interacting with a particular target is assumed to be

related to the degree of similarity relative to the set of known

bioactive ligands for that target A detailed description of the

methodology used can be found elsewhere [60]

Phospholipase inhibition assays

The effect of MA on PLA2 (EC 3.1.1.4) activity was

determined by using a secretory PLA2 assay kit (Cayman

Chemical, Ann Arbor, MI, USA) on 96-well microplates Each well contained 10 lL of dinitrobenzoic acid (10 mm 3,5-dinitrobenzoic acid in 0.4 m Tris⁄ HCl, pH 8.0), 10 lL of parasite protein extract or 10 lL of PLA2 bee venom sup-plied by the kit as positive control, and 5 lL of the corre-sponding inhibitor: MA or 3,4-dichloroisocoumarin dissolved in dimethylsulfoxide The reactions were initiated

by adding 200 lL substrate solution [1.66 mm 1,2-bis(hepta-noylthio)glycerophosphocholine] to all the wells and the plate was shaken carefully Absorbances were monitored at

414 nm using a plate reader every minute for controls and every hour for parasite samples Absorbance data were expressed as a percentage of inhibitory activity compared with control without inhibitor Assays were performed in

25 mm Tris⁄ HCl, pH 7.5, containing 10 mm CaCl2, 100 mm KCl and 0.3 mm Triton X-100

Chelating activity assay The metal chelating activity was measured using the com-plexometric indicator Eriochrome Black T (Sigma-Aldrich) Samples were prepared by mixing 100 lL of 100 lm ZnSO4.H2O, 200 lL of 0.3 m Na2CO3pH 10, 10 lL of com-pound (MA or EDTA as control) and 6 lL of Eriochrome Black T (5 mgÆmL)1in ethanol, pH 10) The pH of the Erio-chrome Black T solution was adjusted by adding buffer solu-tion dropwise until the colour changed from purple to blue

200 lL of each sample were transferred to a 96-well micro-plate and the absorbance was measured at 610 nm Data were expressed as a percentage of the increase in absorbance caused by the removal of zinc cations due to chelating activ-ity compared with a control without zinc EDTA was used as

a positive control for chelating activity

Acknowledgements

This work was supported by grants from the Spanish Ministry of Education and Science (BIO2007-67885 and BIO2010-17039) and the Research Teams Consoli-dation Programme of the UCM, Research Group

920267 and the Iberoamerican CYTED network No 210RT0398 CM is supported by the Universidad de Cartagena (Colombia) and the Alban Programme of High Level Scholarships for Latin America, fellowship E07D400516CO We are grateful to Dr Andre´s Gar-cı´a-Granados for providing MA and helpful discus-sions We thank Susana Pe´rez-Benavente for excellent technical help

Competing interests

The use of MA as an antiparasitic agent is protected

by a patent owned by the University of Granada (date

of filing, 29 March 2007; Patent Number WO⁄ 2007 ⁄

034009) The authors declare no competing financial

interests

References

1 Schlitzer M (2008) Antimalarial drugs – what is in use

and what is in the pipeline Arch Pharm 341, 149–163

2 Wiesner J, Ortmann R, Jomaa H & Schlitzer M (2003)

New antimalarial drugs Angew Chem Int Ed 42, 5274–

5293

3 Garcı´a-Granados A, Martı´nez A, Parra A & Rivas F

(1998) Process for the industrial recovery of oleanoic and

maslinic acids contained in the olive milling subproducts

Patent number WO⁄ 1998 ⁄ 004331 University of

Gra-nada, Spain

4 Martin AM, Vazquez RDP, Fernandez-Arche A &

Ruiz-Gutierrez V (2006) Suppressive effect of maslinic

acid from pomace olive oil on oxidative stress and

cyto-kine production in stimulated murine macrophages

Free Radic Res 40, 295–302

5 Montilla MP, Agil A, Navarro MC, Jimenez MI,

Gar-cia-Granados A, Parra A & Cabo MM (2003)

Antioxi-dant activity of maslinic acid, a triterpene derivative

obtained from Olea europaea Planta Med 69, 472–474

6 Rodriguez-Rodriguez R, Perona JS, Herrera MD &

Ruiz-Gutierrez V (2006) Triterpenic compounds from

‘orujo’ olive oil elicit vasorelaxation in aorta from

spon-taneously hypertensive rats J Agric Food Chem 54,

2096–2102

7 Juan ME, Wenzel U, Ruiz-Gutierrez V, Planas JM &

Daniel H (2005) Maslinic acid, a natural compound

from olives, induces apoptosis in HT-29 human colon

cancer cell line FASEB J 19, A996

8 Reyes FJ, Centelles JJ, Lupianez JA & Cascante M

(2006) (2 alpha,3 beta)-2,3-dihydroxyolean-12-en-28-oic

acid, a new natural triterpene from Olea europea,

induces caspase dependent apoptosis selectively in colon

adenocarcinoma cells FEBS Lett 580, 6302–6310

9 Liu J, Sun H, Duan W, Mu D & Zhang L (2007)

Maslinic acid reduces blood glucose in KK-A(y) mice

Biol Pharm Bull 30, 2075–2078

10 Wen XA, Sun HB, Liu J, Cheng KG, Zhang P, Zhang

LY, Hao J, Zhang LY, Ni PZ, Zographos SE et al

(2008) Naturally occurring pentacyclic triterpenes as

inhibitors of glycogen phosphorylase: synthesis,

struc-ture–activity relationships, and X-ray crystallographic

studies J Med Chem 51, 3540–3554

11 Qiu W-W, Shen Q, Yang F, Wang B, Zou H, Li J-Y,

Li J & Tang J (2009) Synthesis and biological

evaluation of heterocyclic ring-substituted maslinic acid

derivatives as novel inhibitors of protein tyrosine

phosphatase 1B Bioorg Med Chem Lett 19, 6618–6622

12 Xu HX, Zeng FQ, Wan M & Sim KY (1996) Anti-HIV

triterpene acids from Geum japonicum J Nat Prod 59,

643–645

13 Garcı´a-Granados A (2007) Use of maslinic acid as an antiparasitic agent against phylum apicomplexa protozo-ans Patent number WO⁄ 2007 ⁄ 034009 Universidad de Granada, Spain

14 De Pablos LM, Gonzalez G, Rodrigues R, Granados

AG, Parra A & Osuna A (2010) Action of a pentacyclic triterpenoid, maslinic acid, against Toxoplasma gondii

J Nat Prod 73, 831–834

15 Moneriz C, Marı´n-Garcı´a P, Garcı´a-Granados A, Bautista JM, Diez A & Puyet A (2011) Parasitostatic effect of maslinic acid I Growth arrest of Plasmodium falciparumintraerythrocytic stages Malar J 10, 82

16 Moneriz C, Marı´n-Garcı´a P, Bautista JM, Diez A & Puyet A (2011) Parasitostatic effect of maslinic acid II Survival increase and immune protection in lethal Plasmodium yoelii-infected mice Malar J 10, 103

17 Stevenson MM & Riley EM (2004) Innate immunity to malaria Nat Rev Immunol 4, 169–180

18 Sauerwein RW, Bijker EM & Richie TL (2010) Empow-ering malaria vaccination by drug administration Curr Opin Immunol 22, 367–373

19 Carroll CD, Patel H, Johnson TO, Guo T, Orlowski M,

He ZM, Cavallaro CL, Guo J, Oksman A, Gluzman IY

et al.(1998) Identification of potent inhibitors of Plas-modium falciparumplasmepsin II from an encoded sta-tine combinatorial library Bioorg Med Chem Lett 8, 2315–2320

20 Li RS, Kenyon GL, Cohen FE, Chen XW, Gong BQ, Do-minguez JN, Davidson E, Kurzban G, Miller RE, Nuzum

EO et al (1995) In vitro antimalarial activity of chalcones and their derivatives J Med Chem 38, 5031–5037

21 Dominguez JN, Lopez S, Charris J, Iarruso L, Lobo G, Semenov A, Olson JE & Rosenthal PJ (1997) Synthesis and antimalarial effects of phenothiazine inhibitors of a Plasmodium falciparumcysteine protease J Med Chem

40, 2726–2732

22 Kitjaroentham A, Suthiphongchai T & Wilairat P (2006) Effect of metalloprotease inhibitors on invasion

of red blood cell by Plasmodium falciparum Acta Trop

97, 5–9

23 Skinner-Adams TS, Stack CM, Trenholme KR, Brown

CL, Grembecka J, Lowther J, Mucha A, Drag M, Kafarski P, McGowan S et al (2010) Plasmodium falciparumneutral aminopeptidases: new targets for anti-malarials Trends Biochem Sci 35, 53–61

24 Koussis K, Withers-Martinez C, Yeoh S, Child M, Hackett F, Knuepfer E, Juliano L, Woehlbier U, Buj-ard H & Blackman MJ (2009) A multifunctional serine protease primes the malaria parasite for red blood cell invasion EMBO J 28, 725–735

25 Arastu-Kapur S, Ponder EL, Fonovic UP, Yeoh S, Yuan F, Fonovic M, Grainger M, Phillips CI, Powers

JC & Bogyo M (2008) Identification of proteases that regulate erythrocyte rupture by the malaria parasite Plasmodium falciparum Nat Chem Biol 4, 203–213

26 Fairlie WD, Spurck TP, McCoubrie JE, Gilson PR,

Miller SK, McFadden GI, Malby R, Crabb BS &

Hodder AN (2008) Inhibition of malaria parasite

development by a cyclic peptide that targets the vital

parasite protein SERA5 Infect Immun 76, 4332–4344

27 Mestres J & Gregori-Puigjane E (2009) Conciliating

binding efficiency and polypharmacology Trends

Phar-macol Sci 30, 470–474

28 Mestres J, Gregori-Puigjane E, Valverde S & Sole RV

(2008) Data completeness – the Achilles heel of

drug-target networks Nat Biotechnol 26, 983–984

29 Gregori-Puigjane E & Mestres J (2008) A ligand-based

approach to mining the chemogenomic space of drugs

Comb Chem High Throughput Screen 11, 669–676

30 Cases M & Mestres J (2009) A chemogenomic approach

to drug discovery: focus on cardiovascular diseases

Drug Discov Today 14, 479–485

31 Wang SM, Milne GWA, Yan XJ, Posey IJ, Nicklaus

MC, Graham L & Rice WG (1996) Discovery of novel,

non-peptide HIV-1 protease inhibitors by

pharmaco-phore searching J Med Chem 39, 2047–2054

32 Wu YM, Wang XY, Liu X & Wang YF (2003)

Data-mining approaches reveal hidden families of proteases

in the genome of malaria parasite Genome Res 13,

601–616

33 Goldberg DE (2005) Hemoglobin degradation Curr

Top Microbiol Immunol 295, 275–291

34 Dalal S & Klemba M (2007) Roles for two

aminopep-tidases in vacuolar hemoglobin catabolism in

Plasmo-dium falciparum J Biol Chem 282, 35978–35987

35 Azimzadeh O, Sow C, Geze M, Nyalwidhe J & Florent

I (2010) Plasmodium falciparum PfA-M1

aminopepti-dase is trafficked via the parasitophorous vacuole and

marginally delivered to the food vacuole Malar J 9,

189

36 Rosenthal PJ, Olson JE, Lee GK, Palmer JT, Klaus JL

& Rasnick D (1996) Antimalarial effects of vinyl

sul-fone cysteine proteinase inhibitors Antimicrob Agents

Chemother 40, 1600–1603

37 Rosenthal PJ, Wollish WS, Palmer JT & Rasnick D

(1991) Antimalarial effects of peptide inhibitors of a

Plasmodium falciparumcysteine proteinase J Clin Invest

88, 1467–1472

38 deDominguez NDG & Rosenthal PJ (1996) Cysteine

proteinase inhibitors block early steps in hemoglobin

degradation by cultured malaria parasites Blood 87,

4448–4454

39 Francis SE, Gluzman IY, Oksman A, Banerjee D &

Goldberg DE (1996) Characterization of native

falci-pain, an enzyme involved in Plasmodium falciparum

hemoglobin degradation Mol Biochem Parasitol 83,

189–200

40 Withers-Martinez C, Jean L & Blackman MJ (2004)

Subtilisin-like proteases of the malaria parasite Mol

Microbiol 53, 55–63

41 Yeoh S, O’Donnell RA, Koussis K, Dluzewski AR, Ansell KH, Osborne SA, Hackett F, Withers-Martinez

C, Mitchell GH, Bannister LH et al (2007) Subcellular discharge of a serine protease mediates release of invasive malaria parasites from host erythrocytes Cell

131, 1072–1083

42 Blackman MJ (2008) Malarial proteases and host cell egress: an ‘emerging’ cascade Cell Microbiol 10, 1925– 1934

43 Blackman MJ, Heidrich HG, Donachie S, McBride JS

& Holder AA (1990) A single fragment of a malaria merozoite surface protein remains on the parasite dur-ing red-cell invasion and is the target of invasion-inhib-iting antibodies J Exp Med 172, 379–382

44 Hackett F, Sajid M, Withers-Martinez C, Grainger M

& Blackman MJ (1999) PfSUB-2: a second subtilisin-like protein in Plasmodium falciparum merozoites Mol Biochem Parasitol 103, 183–195

45 Fowkes FJI, Richards JS, Simpson JA & Beeson JG (2010) The relationship between anti-merozoite antibod-ies and incidence of Plasmodium falciparum malaria: a systematic review and meta-analysis Plos Med 7, e1000218

46 Mello K, Daly TM, Long CA, Burns JM & Bergman

LW (2004) Members of the merozoite surface protein 7 family with similar expression patterns differ in ability

to protect against Plasmodium yoelii malaria Infect Immun 72, 1010–1018

47 Rawlings ND, Barrett AJ & Bateman A (2010) MER-OPS: the peptidase database Nucleic Acids Res 38, D227–D233

48 Ponpuak M, Klemba M, Park M, Gluzman IY,

Lamp-pa GK & Goldberg DE (2007) A role for falcilysin in transit peptide degradation in the Plasmodium falcipa-rumapicoplast Mol Microbiol 63, 314–334

49 Zidovetzki R, Sherman IW & Obrien L (1993) Inhibi-tion of Plasmodium falciparum phospholipase A2 by chloroquin, quinine and arteether J Parasitol 79, 565– 570

50 Gill JP & Berglund EG (2007) CYP2C8 and antimalaria drug efficacy Pharmacogenomics 8, 187–198

51 Pang YP, Ekstrom F, Polsinelli GA, Gao Y, Rana S, Hua DH, Andersson B, Andersson PO, Peng L, Singh

SK et al (2009) Selective and irreversible inhibitors of mosquito acetylcholinesterases for controlling malaria and other mosquito-borne diseases PLoS ONE 4, e6851

52 Yuan J, Johnson RL, Huang RL, Wichterman J, Jiang

HY, Hayton K, Fidock DA, Wellems TE, Inglese J, Austin CP et al (2009) Genetic mapping of targets mediating differential chemical phenotypes in Plasmo-dium falciparum Nat Chem Biol 5, 765–771

53 Fairlie DP, Tyndall JDA, Reid RC, Wong AK, Abben-ante G, Scanlon MJ, March DR, Bergman DA, Chai CLL & Burkett BA (2000) Conformational selection of