Aziridine-2,3-Dicarboxylate-Based Cysteine Cathepsin Inhibitors ...

2 downloads 0 Views 4MB Size Report
Mar 9, 2010 - Both compounds targeted leishmanial cathepsin B-like cysteine cathepsin cysteine proteinase C, as shown by fluorescence proteinase activity ...
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Dec. 2010, p. 5028–5041 0066-4804/10/$12.00 doi:10.1128/AAC.00327-10 Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Vol. 54, No. 12

Aziridine-2,3-Dicarboxylate-Based Cysteine Cathepsin Inhibitors Induce Cell Death in Leishmania major Associated with Accumulation of Debris in Autophagy-Related Lysosome-Like Vacuoles䌤 Uta Schurigt,1 Caroline Schad,2 Christin Glowa,1† Ulrike Baum,1 Katja Thomale,1 Johannes K. Schnitzer,1 Martina Schultheis,1 Norbert Schaschke,3 Tanja Schirmeister,2 and Heidrun Moll1* Institute for Molecular Infection Biology, University of Wu ¨rzburg, 97080 Wu ¨rzburg, Germany1; Institute of Pharmacy and Food Chemistry, University of Wu ¨rzburg, 97074 Wu ¨rzburg, Germany2; and Faculty of Chemistry, Bielefeld University, 33615 Bielefeld, Germany3 Received 9 March 2010/Returned for modification 25 May 2010/Accepted 29 August 2010

The papain-like cysteine cathepsins expressed by Leishmania play a key role in the life cycle of these parasites, turning them into attractive targets for the development of new drugs. We previously demonstrated that two compounds of a series of peptidomimetic aziridine-2,3-dicarboxylate [Azi(OBn)2]-based inhibitors, Boc-(S)-Leu-(R)-Pro-(S,S)-Azi(OBn)2 (compound 13b) and Boc-(R)-Leu-(S)-Pro-(S,S)-Azi(OBn)2 (compound 13e), reduced the growth and viability of Leishmania major and the infection rate of macrophages while not showing cytotoxicity against host cells. In the present study, we characterized the mode of action of inhibitors 13b and 13e in L. major. Both compounds targeted leishmanial cathepsin B-like cysteine cathepsin cysteine proteinase C, as shown by fluorescence proteinase activity assays and active-site labeling with biotin-tagged inhibitors. Furthermore, compounds 13b and 13e were potent inducers of cell death in promastigotes, characterized by cell shrinkage, reduction of mitochondrial transmembrane potential, and increased DNA fragmentation. Transmission electron microscopic studies revealed the enrichment of undigested debris in lysosome-like organelles participating in micro- and macroautophagy-like processes. The release of digestive enzymes into the cytoplasm after rupture of membranes of lysosome-like vacuoles resulted in the significant digestion of intracellular compartments. However, the plasma membrane integrity of compound-treated promastigotes was maintained for several hours. Taken together, our results suggest that the induction of cell death in Leishmania by cysteine cathepsin inhibitors 13b and 13e is different from mammalian apoptosis and is caused by incomplete digestion in autophagy-related lysosome-like vacuoles. costs. Miltefosine, the first drug for effective oral treatment of leishmaniasis, is highly teratogenic. Moreover, the rate of parasite resistance against classical drugs is increasing, and the development of vaccination strategies against leishmaniasis has not yet been successful. Therefore, new leishmanicidal drugs are urgently needed. Many factors contributing to the virulence of Leishmania parasites have been identified. Especially, the cysteine proteinases (CPs) play a key role in the life cycle and infectivity of these parasites, making them attractive targets for the development of new drugs (6, 19, 37). Leishmania expresses numerous CPs belonging to proteinase clans CA, CD, CF, and PC. The papain-like enzymes in family C1 of clan CA (CAC1), also termed cysteine cathepsins, are the best-characterized leishmanial CPs and contain the cathepsin L-like enzymes CPA and CPB, as well as the cathepsin B-like enzyme CPC. Gene deletions of these proteinases in various Leishmania species confirmed the importance of cysteine cathepsins for the life cycle and the virulence of these parasites (1, 3). During the last few decades, a variety of cysteine cathepsin inhibitors have been developed to treat infectious as well as chronic human diseases. One group of irreversible inhibitors comprises molecules containing small rings as electrophilic building blocks. Important electrophilic fragments are threemembered heterocyclic epoxides and aziridines. The epoxysuc-

Leishmaniasis is one of the 13 most important tropical diseases listed by the World Health Organization. The disease is caused by protozoa of the genus Leishmania transmitted to mammalian hosts during a blood meal of infected female sandflies. The clinical outcome of leishmaniasis, ranging from a mild cutaneous form to a potentially lethal visceral form, depends on the infecting Leishmania species and on the type of host immune response. Worldwide, approximately 12 million people suffer from leishmaniasis and an estimated 350 million people are at risk of attracting the disease. Moreover, it is anticipated that the incidence of coinfections with Leishmania and human immunodeficiency virus (HIV) will continue to rise in many countries. At present, only a few drugs are available for the treatment of human leishmaniasis (13). The pentavalent antimony agents (sodium stibogluconate and meglumine antimoniate), amphotericin B, pentamidine, and paromomycin have limitations, including the need for parenteral administration and long courses of treatment, toxic side effects, and high * Corresponding author. Mailing address: Institute for Molecular Infection Biology, University of Wu ¨rzburg, Josef-Schneider-Str. 2/D15, 97080 Wu ¨rzburg, Germany. Phone: 49 (0)931-3182627. Fax: 49 (0)9313182578. E-mail: [email protected]. † Present address: German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. 䌤 Published ahead of print on 20 September 2010. 5028

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

5029

FIG. 1. Schematic representation of the structures of the investigated leishmanicidal aziridine-2,3-dicarboxylate-based cysteine cathepsin inhibitors 13b and 13e as well as those of 13b-B (biotin-tagged derivative of 13b), bADA, E64, E64d, CA074, CA074ME, and E64d-B (biotin-tagged derivative of E64d).

cinyl peptide L-trans-epoxysuccinyl-Leu-4-guanidinobutylamide (E64) was the first CP inhibitor to be discovered, with an epoxide being the essential structural feature (Fig. 1). This natural product was isolated from Aspergillus japonicas in 1978 (10). The broad-spectrum cysteine cathepsin inhibitor E64 selectively inhibited papain-like CPs (clan CA, family C1) and had little or no selectivity for different members of this proteinase family (27). Thus, E64 can be used as a broad-spectrum CAC1 inhibitor which does not discriminate between the CPA, CPB, and CPC enzymes. Furthermore, many derivatives of E64 were developed to achieve selectivity within the papain family. L-trans-Epoxysuccinyl-Ile-Pro-OH propylamide (CA074), one of the first selective inactivators of cathepsin B and cathepsin B-like enzymes (e.g., CPC) (Fig. 1), was developed in 1991 (5, 34). Aziridinyl peptide inhibitors, a second class of cysteine cathepsin inhibitors, are aza analogues of epoxysuccinyl peptide inhibitors. The natural aziridinyl peptide miraziridine A was isolated from the marine sponge Theonella sp. aff. mirabilis (20). Derivatives containing aziridine-2-carboxylic acid and aziridine-2,3-dicarboxylic acid as electrophilic building blocks were synthesized to develop new selective inhibitors (25, 26). In further investigations, aziridine-based inhibitors containing

aziridine-2,3-dicarboxylates [Azi(OBn)2s] showed selective inhibition of cathepsin L-like enzymes, for instance, rhodesain, the major trypanosomal papain-like CP of the parasite Trypanosoma brucei rhodesiense (41), and falcipain-2 and falcipain-3 from the malaria parasite Plasmodium falciparum (29). In addition, aziridine-2,3-dicarboxylate-based inhibitors had antiparasitic activity against Trypanosoma brucei brucei and P. falciparum in vitro (29, 41). Finally, this promising series of peptidomimetic aziridine-2,3-dicarboxylate inhibitors was demonstrated to exert significant antileishmanial activity. Two derivatives of this series, Boc-(S)-Leu-(R)-Pro-(S,S)-Azi(OBn)2 (compound 13b) and Boc-(R)-Leu-(S)-Pro-(S,S)-Azi(OBn)2 (compound 13e) (Fig. 1), reduced the growth and viability of Leishmania major promastigotes and the infection rate of macrophages with L. major amastigotes (23). The peptidomimetic inhibitors 13b and 13e were selectively active against L. major and did not display any cytotoxic effects against macrophages and fibroblasts (23). On the basis of the activities of compounds 13b and 13e against various CAC1 cysteine proteases, it was hypothesized that these inhibitors also target the papain-like cysteine cathepsins CPA, CPB, and CPC of L. major (23). However,

5030

SCHURIGT ET AL.

this inhibition has not yet been demonstrated experimentally. Therefore, the primary aim of the present study was the identification of the target proteinases of compounds 13b and 13e by fluorescence proteinase activity assays and active-site labeling. Furthermore, the mode of cell death induced by these CP inhibitors in L. major promastigotes was elucidated.

MATERIALS AND METHODS Parasites. The cloned virulent L. major isolate MHOM/IL/81/FE/BNI was maintained by passage in BALB/c mice. Promastigotes were grown in vitro in blood agar cultures at 27°C in 5% CO2 with 95% humidity. Fluorescence protease activity assays and active-site labeling. Proteins were isolated from stationary-phase promastigotes harvested from blood agar plates. Parasites were washed twice with phosphate-buffered saline (PBS) and pelleted by centrifugation at 3,000 ⫻ g for 10 min. The pellet was redissolved in acidic sodium acetate buffer, pH 5.5 (200 mM sodium acetate, 1 mM EDTA, 0.05% polyoxyethyleneglycol dodecyl ether [Brij 35], 0.5 mM dithiothreitol [DTT]). Cells were disrupted by repeated freezing in liquid nitrogen and thawing at 37°C (3 cycles), followed by centrifugation at 700 ⫻ g for 15 min at 4°C. Supernatant was transferred to fresh tubes and stored at ⫺20°C until use. Final protein concentrations of these lysates were determined with a bicinchoninic acid protein assay kit (Pierce Biotechnology, Inc., Pittsburgh, PA). Yields were between 0.6 and 0.8 ␮g ␮l⫺1. All inhibitors used for fluorescence proteinase activity assays and active-site labeling were dissolved in dimethyl sulfoxide (DMSO). Aziridine2,3-dicarboxylate CP inhibitors 13b and 13e (Fig. 1) were synthesized as described previously (41). Proteolytic activities were determined by degradation of the fluoropeptide Z-Phe-Arg-4-methyl-coumarin-7-amide (Z-Phe-Arg-AMC; Bachem, Bubendorf, Switzerland), a substrate of the leishmanial cathepsin B-like enzyme (CPC) and cathepsin-L like enzymes (CPA, CPB) (30, 31, 32), using fluorescence proteinase activity assays. These assays were carried out in black 96-well microtiter plates (Nunc GmbH, Langenselbold, Germany). Protein lysate (10 ␮l) was mixed with 83 ␮l acidic sodium acetate buffer for each sample, followed by addition of 1 ␮l DMSO, 1 ␮l 10 mM E64 (Bachem), 1 ␮l 10 mM CA074 (Bachem), 1 ␮l 20 mM compound 13b, or 1 ␮l 20 mM compound 13e. These samples were preincubated for 15 min at room temperature (RT). Thereafter, 1 ␮l DMSO, 1 ␮l 20 mM compound 13b, or 1 ␮l 20 mM compound 13e was added to samples which were subsequently incubated in a second step for a further 15 min at RT. Finally, the proteolytic reaction was initiated by addition of 5 ␮l 500 ␮M Z-Phe-Arg-AMC solution. Proteolytic release of 7-amino-4methyl-coumarin (AMC) was continuously monitored for 45 min by fluorescence spectroscopy at an excitation wavelength of 355 nm and an emission wavelength of 460 nm using a Fluoroskan Ascent fluorescence reader (Thermo Electron Corporation, Langenselbold, Germany). Standard curves were prepared with the fluorochrome AMC (Bachem). Proteinase activities were calculated using the linear range of reaction curves. Active-site labeling (9) was performed using the biotin-tagged aziridine-based cysteine cathepsin inhibitor dibenzyl-1-[(R)-biotin-6-aminohexanoyl-aziridine]-2,3(2S,3S)-dicarboxylate (bADA) (8) and the biotinylated derivative (2S,3S)-dibenzyl 1-{(R)-1-[(S)-4-methyl-2-{5-[(3aS,4S,6aR)-2-oxohexahydro-1H-thieno[3,4-d]imidazol-4-yl]pentanamido}pentanoyl]pyrrolidine-2-carbonyl}aziridine-2,3-dicarboxylate (compound 13b-B) of the leishmanicidal inhibitor compound 13b (Fig. 1). Beyond, the epoxide-based L-trans-epoxysuccinyl-Leu-3-methylbutylamideethyl ester (E64d)-biotinylated derivative (2S,3S)-ethyl 3-{[(S)-4-methyl-1oxo-1-{[4-(6-{5-[(3aS,4S,6aR)-2-oxohexahydro-1H-thieno[3,4-d]imidazol-4-yl] pentanamido}hexanamido)butyl]amino}pentan-2-yl]carbamoyl}oxirane-2-carboxylate (E64d-B) was used for active-site labeling (Fig. 1). Protein lysate (20 ␮l) was mixed with 78 ␮l acidic sodium acetate buffer for each sample. These mixtures were preincubated for 30 min at RT with 1 ␮l DMSO, 1 ␮l 10 mM E64, 1 ␮l 5 mM CA074, 1 ␮l 20 mM compound 13b, or 1 ␮l 20 mM compound 13e. Finally, 1 ␮l 20 mM bADA, 1 ␮l 20 mM compound 13b-B, or 1 ␮l 0.2 mM E64d-B was added to each sample, followed by a second incubation step for 60 min at RT. Control samples without biotin-tagged inhibitors were treated with 1 ␮l DMSO. Reactions were stopped with sodium dodecyl sulfate (SDS) gel loading buffer and denaturation at 95°C for 5 min. Finally, proteins in the lysates were separated by size in a 12% polyacrylamide gel and transferred to a nitrocellulose membrane. After overnight blocking with 10% milk solution at 4°C, the membrane was washed twice with PBS-Tween 20 and incubated for 1 h with streptavidin-horseradish peroxidase (HRP; Pierce Biotechnology, Inc.) at RT. Finally, the membrane was washed three times with PBS-Tween 20 and devel-

ANTIMICROB. AGENTS CHEMOTHER. oped with chemiluminescent HRP substrate Immobilon Western (Millipore, Billerica, MA). Active-site labeling blots were exposed to X-ray films. The intensities of cysteine cathepsin-specific bands were densitometrically analyzed by the ImageJ software, version 1.42q (http://info.nih.go/ij). Additionally, protein lysates from promastigotes (1 ⫻ 107 ml⫺1) cultured for 24 h in RPMI medium with 10% fetal calf serum (FCS) or starved in PBS for 24 h were isolated to measure the cysteine cathepsin activities by fluorescence activity assays as described above. Lysates were incubated before addition of substrate with 1 ␮l DMSO, 1 ␮l 10 mM E64, or 1 ␮l 10 mM CA074 for 15 min. Standard operating procedure (SOP) for inhibitor treatment of L. major promastigotes. Stationary-phase L. major promastigotes harvested from blood agar plates were seeded into 96-well plates at a density of 1 ⫻ 107 ml⫺1 in RPMI medium with 10% FCS (200 ␮l per well) in the absence or presence of 100 ␮M compound 13b or 13e. The compounds were dissolved in DMSO. Control wells lacking compounds were set up with the same final DMSO concentration (0.5%). DMSO- and inhibitor-treated L. major promastigotes were incubated at 27°C in 5% CO2. Giemsa staining. Promastigotes were cultured and treated with 100 ␮M compound 13b or 13e according to SOP for 3 h or 24 h. Thereafter, promastigotes were harvested and pelleted by centrifugation at 3,000 ⫻ g for 2 min. The cell sediment was solved in PBS (1 ⫻ 108 ml⫺1). Ten microliters of this cell suspension was pipetted onto a microscope slide. Air-dried cells were fixed for 5 min with ice-cold 100% methanol. Slides were stained for 5 min in Accustain Giemsa staining solution (Sigma-Aldrich, Taufkirchen, Germany), followed by destaining for 2 min in aqua dest. After the cells were air dried, they were analyzed by transmission light microscopy. Annexin V-PI staining. Double staining with annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) allows the discrimination between apoptotic and necrotic cells. Promastigotes were cultured and treated with 100 ␮M each inhibitor according to the SOP for inhibitor treatment. Control and inhibitor-treated promastigotes were washed twice with annexin V binding buffer (10 mM HEPES, 140 mM NaCl, 3.3 mM CaCl2) and centrifuged at 3,000 ⫻ g for 7 min at RT. The pellets were resuspended in 100 ␮l annexin V binding buffer. Two microliters of 50 ␮g ml⫺1 PI solution and 5 ␮l commercially available annexin V-FITC solution (BD Bioscience, Heidelberg, Germany) were added to each sample. Samples were incubated for 15 min in the dark at RT. Finally, cells were washed with annexin V binding buffer and centrifuged at 3,000 ⫻ g for 7 min. The pellet was redissolved in annexin V binding buffer. Stained samples were immediately analyzed by flow cytometry using a FACSCalibur flow cytometer (BD Bioscience). Results were quantitatively analyzed with WinMDI software, version 2.9 (http://www.facs.scripps.edu./software.html). Detection of DNA fragmentation by TUNEL flow cytometry assay. Detection of DNA fragments was performed by terminal deoxyribonucleotidyltransferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay using an Apo-Direct kit (BD Bioscience, Pharmingen, Heidelberg, Germany). The Apo-Direct kit uses a staining method for labeling DNA breaks with FITC-dUTP to detect apoptotic cells by flow cytometry. Promastigotes were treated with 100 ␮M compound 13b or 13e for 18 h according to SOP. Finally, promastigotes were harvested and fixed for 12 h at 4°C in 4% paraformaldehyde solution and then washed twice in PBS and stored in 70% ethanol for 24 h at ⫺20°C. Further staining was performed as described in the manual for the Apo-Direct kit. Detection of dUTP-positive cells was performed by flow cytometry (FACSCalibur; BD Bioscience). Results were analyzed with WinMDI, version 2.9 (http://www .facs.scripps.edu./software.html). Analysis of mitochondrial transmembrane potential and cell membrane integrity. Cells were cultured according to SOP with 100 ␮M compound 13b or 13e. Control cells were treated with DMSO. Promastigotes were double stained with the mitochondrion-selective dyes MitoTracker red CMXRos (Invitrogen-Molecular Probes, Eugene, OR) and SYTOX green (Invitrogen-Molecular Probes). SYTOX green is a high-affinity nucleic acid stain that easily penetrates cells with compromised plasma membranes yet will not cross the membranes of live cells. This allows discrimination of SYTOX green-negative live cells with intact plasma membranes and SYTOX green-positive dead cells with a loss of cell membrane integrity. For double staining, 2 ␮l per well of 10 ␮M MitoTracker red CMXRos and 2 ␮l per well of 20 ␮M SYTOX green were pipetted directly into the cell culture medium. Promastigotes were stained for 15 min at 27°C. Finally, cells were transferred into fluorescence-activated cell sorter (FACS) tubes and centrifuged at 3,000 ⫻ g for 7 min. The pellets were redissolved in 1 ml FACS buffer (PBS containing 0.1% sodium azide and 2.5% FCS) and immediately analyzed by flow cytometry (FACSCalibur; BD Bioscience). Results were analyzed with WinMDI, version 2.9 (http://www.facs.scripps.edu./software.html).

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

Transmission electron microscopy of promastigotes. Stationary-phase promastigotes were treated with 100 ␮M each inhibitor according to SOP for 30 min, 1 h, 2 h, or 10 h. For detection of lysosome-like vacuoles, promastigotes were directly harvested from blood agar plates, cultivated in RPMI medium with 10% FCS (1 ⫻ 107 ml⫺1), or starved for 24 h in PBS (1 ⫻ 107 ml⫺1). After the promastigotes were harvested, they were contrasted and embedded using the following procedure: (i) fixation of promastigotes for 120 min with 4% paraformaldehyde–1% glutaraldehyde at 4°C; (ii) washing twice for 30 min each time with PBS at RT; (iii) contrasting for 120 min with 1% osmium tetroxide (SigmaAldrich) and 3% potassium ferricyanide (Sigma-Aldrich) in PBS at RT; (iv) contrasting for 30 min with 0.5% uranyl acetate (Sigma-Aldrich) in distilled water at RT; (v) dehydration twice for 15 min each time with 70% ethanol, twice for 15 min each time with 80% ethanol, twice for 15 min each time with 95% ethanol, and twice for 30 min each time with 100% ethanol at RT; (vi) incubation twice for 30 min each time in propylene oxide (Sigma-Aldrich) at RT; (vii) incubation for 60 min in a 1:1 mixture of propylene oxide (SigmaAldrich) and Epon (Sigma-Aldrich) at RT; (viii) incubation for 16 h in Epon; and (ix) polymerization of Epon at 60°C for 48 h. Ultrathin sections were mounted on 300-mesh grids, stained with uranyl acetate and lead citrate, and analyzed with an EM 10 transmission electron microscope (Carl Zeiss AG, Oberkochen, Germany). Transmission electron microscopy of amastigotes. Bone marrow-derived macrophages were generated as described previously (28) and infected with stationary-phase L. major promastigotes directly harvested from blood agar plates at a ratio of 1:15. Cocultures were incubated for 24 h at 37°C. Thereafter, the medium was changed and infected macrophages were treated with 10 ␮M compound 13b for 1 h. For transmission electron microscopic studies, the infected macrophages containing amastigotes were fixed for 45 min with 2.5% glutaraldehyde–50 mM cacodylate (pH 7.2; Sigma-Aldrich) at RT and afterwards contrasted for 2 h at 4°C with 2% OsO4 buffered with 50 mM cacodylate (pH 7.2), washed with distilled water, and incubated overnight at 4°C with 0.5% uranyl acetate in distilled water. The cells were dehydrated, embedded in Epon, sectioned ultrathin, and imaged as described for promastigotes. Determination of IC50s. The drug concentrations inhibiting 50% cell growth or cell survival (IC50s) were determined by Alamar Blue assay, as described previously (23). Stationary-phase L. major promastigotes harvested from blood agar plates were seeded into 96-well plates at a density of 1 ⫻ 107 ml⫺1 or 1 ⫻ 106 ml⫺1 in phenol red-free RPMI medium with 10% FCS, in the absence or presence of increasing concentrations of compounds. IC50s were determined 48 h after addition of Alamar Blue. Amphotericin B (Sigma-Aldrich), pentamidine isothionate (Sigma-Aldrich), 1-hexadecylphosphorylcholine (miltefosine; Cayman Chemical Company, Ann Arbor, MI), and paromomycin sulfate (SigmaAldrich) were used as reference compounds. E64 and CA074, as well as E64d (Bachem) and L-trans-epoxysuccinyl-Ile-Pro-methoxy propylamide (CA074ME; Bachem) (derivatives of E64 and CA074, respectively, with improved cell permeation), were used for IC50 determinations (Fig. 1). Statistical analysis. The data are presented as mean values ⫾ standard errors of the means (SEMs). Differences between groups were analyzed for statistical significance by unpaired Student’s t test or the Mann-Whitney U test using Sigma Plot software, version 9.0 (Systat Software, Inc., San Jose, CA). Differences were considered significant when P was ⱕ0.05.

RESULTS Aziridine-2,3-dicarboxylate-based inhibitors 13b and 13e reduce activity of cathepsin B-like enzyme CPC in protein lysates of L. major. We recently demonstrated that aziridine2,3-dicarboxylate-based CP inhibitors 13b and 13e exhibit highly significant leishmanicidal activity in vitro (23). Both compounds were developed to inhibit parasitic cathepsin Llike enzymes (38, 41). In the present study, we analyzed the ability of aziridine-2,3-dicarboxylate-based CP inhibitors 13b and 13e to affect the cysteine cathepsin activities in L. major promastigotes. The proteolytic activities of the cathepsin Blike and the cathepsin L-like enzymes could be detected by fluorescence proteinase activity assays using the substrate ZPhe-Arg-AMC (Fig. 2A). The detected proteolytic activity was E64 sensitive (inhibition of CPA, CPB, and CPC) and CA074 sensitive (selective inhibition of CPC) (Fig. 2A). However,

5031

minor residual proteinase activity remained after preincubation with E64 (100 ␮M) and demonstrated that Z-Phe-ArgAMC is also cleaved by additional non-papain-like proteinases (Fig. 2A). The leishmanicidal aziridine-2,3-dicarboxylatebased inhibitors 13b (200 ␮M) and 13e (200 ␮M) were also able to inhibit the proteolytic cleavage of substrate Z-Phe-ArgAMC (Fig. 2A). No further reductions of proteinase activity by compounds 13b (200 ␮M) and 13e (200 ␮M) could be detected after preincubation of protein lysates with E64 (100 ␮M) (Fig. 2B). Therefore, the Z-Phe-Arg-AMC-cleaving proteinases inhibited by compounds 13b and 13e were members of the papain-like cysteine cathepsins (clan CA, family C1). Preincubation of lysates with the CPC-specific CA074 (100 ␮M), followed by a second incubation step with compound 13b (200 ␮M) or 13e (200 ␮M), also did not result in an additional reduction of the hydrolytic activity in protein lysates compared to that achieved with the CA074-pretreated DMSO control (Fig. 2C). Therefore, the cathepsin B-like enzyme CPC of L. major is the cysteine cathepsin that is mainly affected by CP inhibitors 13b and 13e. However, whereas nanomolar concentrations (⬍100 nM) of the epoxide-based inhibitors E64 and CA074 were sufficient for the total inhibition of CA074-sensitive cysteine cathepsin-specific activity, concentrations of 200 ␮M compound 13b or 13e were not sufficient to completely inhibit CPC activity (Fig. 2A). This reflects the much lower inactivation rates of aziridine-based inhibitors than epoxidebased inhibitors, which have been shown and analyzed in detail in previous computational and experimental studies (11, 12, 40). The inhibition of the cysteine cathepsin-specific activity by compounds 13b and 13e was also confirmed by active-site labeling using the biotin-tagged inhibitors bADA (200 ␮M) (Fig. 2D, lanes 2 to 6), 13b-B (200 ␮M) (Fig. 2E, lanes 2 to 6), and E64d-B (2 ␮M) (Fig. 2E, lanes 7 and 8). Only the proteolytically active forms and not the proteolytically inactive proforms of cysteine cathepsins could be labeled by these inhibitors. The biotinylated E64d derivative was used to ensure that epoxidebased derivatives and aziridine-2,3-dicarboxylate-based derivatives targeted the identical proteinases. Several cysteine cathepsin-specific bands could be detected with all biotin-tagged inhibitors (Fig. 2D and E). The most prominent three bands labeled by all inhibitors had molecular masses of between 34 and 43 kDa. As recorded in the Gene DB database and the MEROPS database, the cathepsin B-like enzyme CPC of L. major has a putative molecular mass of 37.2 kDa (www.genedb .org and http://merops.sanger.ac.uk). Bands with higher and lower molecular masses detected may be glycosylated isoforms and/or splicing variants of leishmanial cysteine cathepsins. Characterization of these bands must be addressed in further studies. The cysteine cathepsin bands found with all three biotin-tagged inhibitors almost completely disappeared after preincubation of the protein lysates with the broad-spectrum cathepsin inhibitor E64 (100 ␮M) (Fig. 2D, lane 3; Fig. 2E, lane 3). Preincubation with the CPC-specific inhibitor CA074 (50 ␮M), followed by a second incubation with the biotin-tagged inhibitors, did not lead to the total disappearance of all bands but led only to a reduction of the band intensities (Fig. 2D, lane 4; Fig. 2E, lane 4). This demonstrates that CA074, in contrast to E64, does not inactivate all cysteine cathepsins but inactivates only CPC. A similar behav-

5032

SCHURIGT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 2. Fluorescence proteinase activity assays and active-site labeling experiments with biotin-tagged cysteine cathepsin inhibitors. (A to C) For fluorescence proteinase activity assays, protein lysates that had been obtained from stationary-phase promastigotes were preincubated in a first incubation step (1st Inc.) with DMSO, 100 ␮M E64, 100 ␮M CA074, 200 ␮M compound 13b, or 200 ␮M compound 13e. In a second incubation step (2nd Inc.), protein lysates were incubated with either DMSO, 200 ␮M compound 13b, or 200 ␮M compound 13e. Proteinase activities were determined by proteolytic degradation of the fluoropeptide Z-Phe-Arg-AMC. Values represent means ⫾ SEMs of six independent experiments. (D and E) For active-site labeling, protein lysates were preincubated in a first incubation step (1st Inc.) with DMSO, 100 ␮M E64, 50 ␮M CA074, 200 ␮M compound 13b, or 200 ␮M compound 13e. In a second incubation step (2nd Inc.), protein lysates were incubated with the biotin-labeled inhibitor bADA (200 ␮M), 13b-B (200 ␮M), or E64d-B (2 ␮M). Controls lacking biotin-tagged inhibitors were incubated with DMSO in both incubation steps. Arrows indicate cysteine cathepsin-specific proteinase bands between 34 and 43 kDa and at 17 kDa. The intensities of these bands were densitometrically determined (tables below the blots). The values for DMSO-preincubated samples labeled with bADA or 13b-B in the second incubation step were set equal to 100%. *, P ⱕ 0.05; ***, P ⱕ 0.001.

ior was found for preincubation with compounds 13e and 13b, followed by labeling with the biotin-tagged inhibitors. Again, these compounds reduced only the intensities of the bands (Fig. 2D, lanes 5 and 6, and Fig. 2E, lanes 5 and 6, with only one exception for compound 13e, shown in Fig. 2D, lane 6, for bands between 34 and 43 kDa), with compound 13e being less active than compound 13b and with both compounds being less active than CA074. These data are in agreement with those of the fluorescence proteinase assays (Fig. 2A) and again reflect the lower inactivation rates of aziridine-based inhibitors than epoxide-based inhibitors. In summary, these data suggest that both inhibitors compete with biotin-tagged inhibitors bADA, 13b-B, and E64d-B for irreversible binding of cathepsin B-like cysteine cathepsin CPC.

Compounds 13b and 13e are more effective inducers of cell death in L. major promastigotes than epoxide-based inhibitors. Fluorescence proteinase activity assays and active-site labeling of cysteine cathepsins demonstrated that the aziridine2,3-dicarboxylate-based inhibitors 13b and 13e as well as the epoxide-based inhibitors E64 and CA074 targeted cysteine cathepsins in protein lysates of L. major promastigotes. However, the epoxide-based inhibitors were more potent for inhibition of cysteine cathepsins in protein lysates of L. major than the aziridine-2,3-dicarboxylate-based inhibitors, which is in agreement with previous findings concerning inhibitory potencies and inactivation rates of epoxide- and aziridine-based inhibitors (11, 12, 40). We then examined the capacities of both inhibitor families to reduce the growth and viability of L. major

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

TABLE 1. IC50s detected by Alamar Blue assaya Compound

13b 13e E64 E64d CA074 CA074ME Amphotericin B Pentamidine Miltefosine Paromomycin a

IC50 (␮M) for L. major at: 1 ⫻ 10 ml⫺1

1 ⫻ 106 ml⫺1

33.4 ⫾ 2.5 47.0 ⫾ 11.8 ⬎100 ⬎100 ⬎100 ⬎100 2.0 ⫾ 0.7 35.9 ⫾ 4.4 31.9 ⫾ 10.4 ⬎100

28.2 ⫾ 9.5 35.8 ⫾ 7.5 84.2 ⫾ 14.1 46.6 ⫾ 15.8 ⬎100 ⬎100 ND ND ND ND

7

The results are from three independent assays. ND, not determined.

promastigotes in vitro. IC50s were determined by Alamar Blue assay with cysteine cathepsin inhibitors 13b, 13e, E64, E64d, CA074, and CA074ME. E64d (an ethyl ester) and CA074ME (a methyl ester) are derivatives of the free acids E64 and CA074, respectively, which were developed to display increased cell permeation. Furthermore, the IC50s of compounds 13b and 13e were compared to the IC50s of the therapeutically used leishmanicidal agents amphotericin B, miltefosine, pentamidine, and paromomycin. The results of the Alamar Blue assays confirmed our previous finding that compounds 13b and 13e exhibit highly significant antileishmanial activity in vitro (Table 1) (23). The IC50s of compound 13b and compound 13e were comparable to those of miltefosine and pentamidine and lower than the IC50 that of paromomycin (Table 1). Only amphotericin B was more effective against L. major promastigotes (Table 1). The IC50 of compound 13b was lower than that of compound 13e (Table 1). This observation correlated with the inhibition capacities of compounds 13b and 13e detected in cysteine cathepsin activity assays and by active-site labeling (Fig. 2A and D, lanes 5 and 6; Fig. 2E, lanes 5 and 6). Surprisingly, no leishmanicidal activities could be found for the epoxide-based inhibitors E64, E64d, CA074, and CA074ME using the screening SOP with a cell concentration of 1 ⫻ 107 ml⫺1 (Table 1). However, if the cell concentration was reduced 10-fold to 1 ⫻ 106 ml⫺1, the epoxide-based broad-spectrum cysteine cathepsin inhibitors E64 and E64d revealed weak leishmanicidal activities (Table 1). Anyway, the IC50s of E64 and E64d were significantly higher than those of compounds 13b and 13e (Table 1). No leishmanicidal activity could be detected with 1 ⫻ 107 ml⫺1 and 1 ⫻ 106 ml⫺1 for the cathepsin B-specific inhibitors CA074 and CA074ME, although CA074 strongly inhibited the cathepsin B-like activity in protein lysates from L. major promastigotes (Table 1; Fig. 2A; Fig. 2D, lane 4; Fig. 2E, lane 4). These results clearly demonstrated that the leishmanicidal potency of the two aziridine-2,3-dicarboxylate-based cysteine cathepsin inhibitors 13b and 13e is higher than that of the epoxide-based inhibitors E64 and CA074. This may be due to the much higher lipophilicity and cell permeation of compounds 13b and 13e than the free acids E64 and CA074, as well as E64d and CA074ME, which are the ethyl and methyl ester prodrugs of E64 and CA074, respectively. The higher lipophilicity can be attributed to the two hydrophobic benzyl ester moieties of compounds 13b and 13e, since earlier studies showed that the benzyl esters display much

5033

better activity against whole cells than the respective diethyl esters, although the latter are similarly active against isolated cysteine proteases (7, 41). Induction of cell death in compound 13b- and 13e-treated L. major promastigotes is associated with apoptosis-like features. Next, we analyzed the cell death pathway induced by cysteine cathepsin inhibitors 13b and 13e. To this end, several morphological and biochemical features were investigated to distinguish between different forms of cell death (apoptosis, necrosis, and autophagic cell death) (14). Visual examination by light microscopy revealed that treatment with compounds 13b and 13e induced the shrinkage of promastigotes (Fig. 3A, images 2, 3, 5, and 6). Morphological alterations of promastigotes could already be observed after 3 h of incubation with compound 13b or 13e (Fig. 3A, images 2 and 3) and were more pronounced after 24 h (Fig. 3A, images 5 and 6). Promastigotes of the control group cultured for at least 24 h in DMSOcontaining RPMI medium maintained their elongated spindleshaped form (Fig. 3A, image 4). Cell shrinkage is a hallmark of apoptosis-like cell death (14). Therefore, several further features of apoptotic cell death were tested to verify the hypothesis that this is the mechanism underlying the inhibitors’ effect on L. major promastigotes. Double staining of promastigotes with MitoTracker red CMXRos and SYTOX green clearly demonstrated that most compound 13b- and 13e-treated promastigotes first lost their mitochondrial transmembrane potential, followed by a relatively late loss of cell membrane integrity (Fig. 4A to C). A significant reduction of MitoTracker red CMXRos-positive promastigotes could be detected after 1 h of incubation with both inhibitors (Fig. 4B), whereas SYTOX green-positive promastigotes were detected only after 12 h of incubation for compound 13b and 24 h for compound 13e (Fig. 4C). However, the loss of mitochondrial transmembrane potential is associated with both apoptosis- and necrosis-like cell death in mammalian cells (14). Therefore, annexin V/PI double staining was performed (Fig. 3C). This method is generally used to detect apoptosis-like cell death in Leishmania promastigotes after drug treatment. A time course experiment (1 to 24 h after inhibitor treatment) revealed a significant increase of the number of early apoptotic annexin V-positive/PI-negative cells in the inhibitor-treated samples compared to that in the DMSO-treated control sample at 6 h after incubation with compound 13b and at 12 h after incubation with compound 13e (Fig. 3C). After 24 h, about 59% of the compound 13b-treated cells and about 45% of the compound 13e-treated cells were annexin V positive/PI negative (early apoptotic). In addition, lower percentages of annexin V-positive/PI-positive (late apoptotic) and annexin V-negative/PI-positive (early necrotic) promastigotes could be detected after incubation with compound 13b or 13e for 24 h (Fig. 3C). Moreover, TUNEL staining demonstrated an increase of dUTP-FITC-positive cells after 18 h of treatment with compound 13b, thus indicating DNA fragmentation (Fig. 3B). In summary, the results of morphological and flow cytometric analyses suggested that the highest percentage of compound 13b- and 13e-treated promastigotes died through induction of apoptosis-like cell death. Furthermore, the detection of early necrotic promastigotes suggested that the high lipophilicity of aziridine-based CP inhibitors 13b and 13e may also cause an unspecific interaction with plasma membranes.

5034

SCHURIGT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 3. Analysis of cell morphology and results of TUNEL assay and annexin V/PI staining. L. major promastigotes were incubated with 100 ␮M compound 13b or 13e. Control cultures were supplemented with DMSO-containing RPMI medium. (A) Alterations of cell morphology were investigated by Giemsa staining. (B) DNA fragmentation was investigated by TUNEL staining in promastigotes treated with compound 13b. The percentage of dUTP-FITC-positive promastigotes was determined by flow cytometric analysis. The values represent means ⫾ SEMs for three samples. Histograms of representative samples are shown. Marker 1 (M1) indicates the region of dUTP-FITC-positive events. ***, P ⱕ 0.001. (C) Annexin V-FITC/PI staining was performed after 1 to 24 h. Dot plots of representative samples and results of quadrant statistics are shown. The values represent means ⫾ SEMs for three samples. UL, upper left quadrant; UR, upper right quadrant; LL, lower left quadrant; LR, lower right quadrant.

Accumulation of undigested debris in lysosome-like vacuoles in cysteine cathepsin proteinase inhibitor-treated promastigotes. Transmission electron microscopy studies were performed to analyze the ultrastructural alterations during cell

death induced by aziridine-2,3-dicarboxylate-based cysteine cathepsin inhibitors in L. major promastigotes. A short incubation time of 30 min was chosen to detect early ultrastructural alterations after treatment with compound 13b. Interestingly,

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

5035

FIG. 4. Analysis of mitochondrial transmembrane potential and plasma membrane integrity. L. major promastigotes were incubated with 100 ␮M compound 13b or 13e. Control cultures were supplemented with DMSO-containing medium. (A) Parasites were harvested at several time points of incubation (1 to 24 h) and double stained with MitoTracker red CMXRos to monitor the changes in mitochondrial transmembrane potential and with the DNA-binding dye SYTOX green to investigate plasma membrane integrity by flow cytometry. Dot plots of representative samples and results of quadrant statistics are shown. (B and C) Results of single-parameter analyses of MitoTracker red CMXRos-positive (B) and SYTOX green-positive (C) promastigotes. Values represent means ⫾ SEMs for three samples. UL, upper left quadrant; UR, upper right quadrant; LL, lower left quadrant; LR, lower right quadrant; *, P ⱕ 0.05; **, P ⱕ 0.01; ***, P ⱕ 0.001.

an accumulation of electron-dense undigested material in lysosome-like compartments was observed (Fig. 5A, images 2 and 3). This corresponds to the observed inhibition of cysteine cathepsin activity. Furthermore, lysosome-like organelles were

enlarged in compound 13b-treated L. major promastigotes (Fig. 5A, images 2 and 3) compared to the size of control parasites cultured in DMSO-containing medium (Fig. 5A, image 1). An accumulation of several membranous structures,

5036

SCHURIGT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 5. Early ultrastructural alterations and accumulation of debris in lysosome-like vacuoles in compound 13b-treated L. major promastigotes and amastigotes detected by transmission electron microscopy. (A) Promastigotes were incubated with 100 ␮M compound 13b for 30 min (images 2 and 3). Control cultures were supplemented with DMSO-containing RPMI medium (image 1). Chromatin condensation is indicated by arrowheads. (B) Amastigotes residing in macrophages were incubated with 10 ␮M compound 13b for 1 h (images 3 to 6). Control amastigotes were incubated in DMSO-containing medium (images 1 and 2). Asterisks, debris in lysosome-like vacuoles; F, flagellum; FP, flagellar pocket; K, kinetoplast; kDNA, kinetoplast DNA; L, lysosome-like vacuoles; M, mitochondrion; MVB, multivesicular body; N, nucleus.

probably endosomes, to form multivesicular bodies could be detected in some saggital sections of compound 13b-treated L. major promastigotes after 30 min (Fig. 5A, image 3). Additionally, peripheral chromatin condensation along the nuclear membrane, forming crescent structures, a characteristic of apoptosis, was observed after compound treatment (Fig. 5A, images 2 and 3). Mitochondria in inhibitor-treated promastigotes were still intact, electron dense, and indistinguishable from the mitochondria in DMSO-treated control promastigotes in the early phase of cell death induction (Fig. 5A, images 1 to 3). These results suggest that impairment of intralysosomal digestion is responsible for cell death in compound 13b- and 13etreated promastigotes. Dilation of mitochondria and karyolysis followed total destruction of lysosome-like organelles. We next examined later

events in the cell death cascade induced in inhibitor-treated promastigotes. For this purpose, L. major promastigotes were analyzed by transmission electron microscopy 1, 2, and 10 h after treatment with compound 13b. Progressive enlargement of lysosome-like organelles and beginning destruction of their membranes were observed after 1 h of treatment with compound 13b (Fig. 6A and B). Furthermore, nuclei with weakly condensed chromatin were detected (Fig. 6A and B). Interestingly, mitochondria were dilated in promastigotes at this time point (Fig. 6A and B). Mitochondrial swelling was more pronounced in some promastigotes after 2 h of inhibitor treatment (Fig. 6D). The ultrastructural alterations of mitochondria further explain the loss of mitochondrial transmembrane potential detected by flow cytometric analyses during incubation with cysteine cathepsin inhibitors 13b and 13e (Fig. 4A and B).

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

FIG. 6. Ultrastructural alterations detected by transmission electron microscopy in L. major promastigotes treated with compound 13b for 1, 2, or 10 h. (A to D) Destruction of lysosome-like vacuoles and mitochondrial swelling in compound 13b-treated promastigotes after 1 and 2 h of incubation. (E and F) Karyolysis and destruction of intracellular organelles during the late phase of cell death in L. major promastigotes treated with cysteine cathepsin inhibitor for 10 h. Chromatin condensation is indicated by arrowheads. dM, dilated mitochondrion; F, flagellum; kDNA, kinetoplast DNA; L, lysosome-like vacuoles; MVB, multivesicular bodies; N, nucleus.

Moreover, the lysosomal compartment was completely disorganized and the membrane integrity of lysosome-like vacuoles was lost (Fig. 6C and D). Intracellular destruction proceeded with increasing incubation time. Finally, L. major promastigotes treated with compound 13b for 10 h were characterized by heavily destructed nuclei with a damaged nuclear membrane and disappearance of intracellular organelles (Fig. 6E and F). In contrast to destruction of cellular organelles, the cell surface membrane with subpellicular tubules did not appear to be significantly damaged at this time point (Fig. 6E and F). Finally, many inhibitor-treated promastigotes seemed to be totally empty and consisted only of intact plasma membranes and shortened flagella (Fig. 6F). The ultrastructural alterations in promastigotes treated with compound 13e were identical to those in compound 13b-treated promastigotes (data not shown). The progression of intracellular destruction in compound 13e-treated promastigotes was delayed compared to that in compound 13b-treated promastigotes and correlated with IC50s and proteinase inhibition capacities. Control promastigotes incubated for 1, 2, or 10 h in DMSOcontaining medium did not show any significant alterations of intracellular organelles compared to cells directly harvested from blood agar plates (data not shown). In summary, trans-

5037

mission electron microscopic studies for investigation of later events in the cell death cascade demonstrated that compounds 13b and 13e induced the release of digestive enzymes into the cytoplasm after rupture of the membranes of lysosome-like organelles. The data suggest that all intracellular compartments, including the nuclei with DNA, were digested during cell death in cysteine cathepsin inhibitor-treated promastigotes, while the cytoplasmic membrane remained intact. Lysosome-like vacuoles are involved in micro- and macroautophagy-like processes in L. major promastigotes. Transmission electron microscopic studies for detailed characterization of lysosome-like vacuoles in promastigotes directly harvested from blood agar plates (stationary phase) were performed. Roundish lysosome-like vacuoles, as recently described for L. major by Boukai and coworkers (4), could be detected (Fig. 7A). Interestingly, some of these organelles engulfed cytoplasm by invagination in a manner similar to that described for microautophagy (Fig. 7B and C) (35). Some invaginations were characterized by multiple-layered membranous structures characteristic for macroautophagic vacuoles (Fig. 7D) (35). Furthermore, degradative autophagic vacuoles (dAVs; autophagolysosomes) resembling those described for L. mexicana (44) could be detected in L. major promastigotes directly harvested from blood agar cultures (Fig. 7E). The detection frequency of autophagosome- and autophagolysome-like structures was increased after starvation of promastigotes in PBS. Additionally, huge degradative autophagic vacuoles engulfing cytoplasmic material and fusing with lysosome-like structures could be observed in PBS-starved promastigotes (Fig. 7F). This phenotype resembles the macroautophagic vacuoles detected by Waguri and coworkers in pituitary tumor cells (42). The number of autophagy-related lysosome-like vacuoles was significantly lower in promastigotes cultured for 24 h in nutrient-rich medium than in promastigotes starved for 24 h in PBS (Fig. 7G). Furthermore, the intracellular E64- and CA074-sensitive cysteine cathepsin activity was higher in PBS-starved promastigotes than in medium-cultured cells (Fig. 7H). Therefore, our data suggested that lysosome-like vacuoles participate in autophagic degradation and that the debris in lysosome-like vacuoles after inhibition of cysteine cathepsins by compounds 13b and 13e may be derived from impaired autophagic processes. Accumulation of undigested debris in lysosome-like vacuoles and intracellular degradation in compound 13b-treated amastigotes residing in macrophages. Amastigotes residing in macrophages were treated for 1 h with 10 ␮M compound 13b (Fig. 5B, images 3 to 6) to investigate phenotypic alterations in the clinically relevant life stage of L. major. An accumulation of undigested material in lysosome-like vacuoles, similar to the early morphological alteration after 30 min in promastigotes, was detected in most amastigotes (Fig. 5B, images 3 to 5). DMSO-treated amastigotes were used as a control (Fig. 5B, images 1 and 2). Rupture of membranes of lysosome-like vacuoles containing undigested material could be observed (Fig. 5B, image 4). Interestingly, some amastigotes were totally destroyed at this early time point after treatment, resembling the phenotype observed after incubation of promastigotes with compounds 13b and 13e for 10 h (Fig. 5B, image 6). Therefore, our data suggested that inhibition of digestive processes in lysosome-like vacuoles is also responsible for the leishmani-

5038

SCHURIGT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 7. Characterization of lysosome-like vacuoles. (A to F) Stationary-phase promastigotes harvested directly from blood agar plates (A to E) or further starved in PBS (F) were investigated by transmission electron microscopy. Asterisks, microautophagy-like invaginations of membranes of lysosome-like vacuoles; arrow, a macroautophagic-like multiple-layered membranous structure fusing with lysosome-like vacuole; arrowheads, membrane bordering dAVs; L, lysosome-like vacuoles. (G) The numbers of lysosome-like vacuoles were determined by transmission electron microscopy in saggital sections of promastigotes incubated for 24 h in nutrient-rich medium or starved for 24 h in PBS (n ⫽ 25 saggital sections). (H) E64- and CA074-sensitive cysteine cathepsin activities detected by fluorescence protease activity assay in promastigotes cultured for 24 h in medium or starved for 24 h in PBS. For fluorescence proteinase activity assays, protein lysates that had been obtained from medium- and PBS-cultured promastigotes were incubated with DMSO, 100 ␮M E64, or 100 ␮M CA074. Proteinase activities were determined by proteolytic degradation of the fluoropeptide Z-Phe-Arg-AMC. The protease activity in DMSO-incubated lysates obtained from medium-cultured promastigotes was set equal to 100%. Values represent means ⫾ SEMs for four independent experiments. ###, P ⱕ 0.001; n.s., not significant.

cidal activities of the aziridine-based inhibitors against amastigotes. DISCUSSION CPs are essential for the growth, differentiation, and pathogenicity of Leishmania parasites and play a pivotal role in host-parasite interactions (19, 44). Therefore, targeting parasitic CP is a promising strategy to develop new leishmanicidal drugs. We previously reported that the two aziridine-2,3-dicar-

boxylate-based CP inhibitors 13b and 13e display leishmanicidal activities against L. major (23). Both CP inhibitors induced cell death in promastigotes with IC50s of about 40 ␮M and decreased infection rates of macrophages with amastigotes with IC50s of 3 to 5 ␮M (23). In the present study, we characterized the molecular targets of the CP inhibitors 13b and 13e and the mode of cell death induced in L. major by these compounds. The aziridine-2,3-dicarboxylate-based inhibitor series was designed to specifically inhibit mammalian and parasitic ca-

VOL. 54, 2010

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

thepsin L-like enzymes of the papain-like proteinase family (clan CA, family C1) (38, 41). L. major and other Leishmania species express the cathepsin L-like proteinases CPA and CPB as well as the cathepsin B-like enzyme CPC (19). Interestingly, CPC, the cathepsin B-like enzyme characterized by an occluding loop, cleaves only peptide substrates with cathepsin L-like substrate specificity and is thus different from its mammalian homologues (30). In the present study, we demonstrated that compounds 13b and 13e inhibited CA074-sensitive cysteine cathepsin activity in protein lysates of L. major promastigotes. Moreover, active-site labeling studies showed that compounds 13b and 13e have characteristics resembling those of CA074, suggesting that CPC is the main target of these aziridine-based inhibitors. However, inhibition of cathepsin L-like enzymes and their isoforms specifically expressed in amastigotes could be not excluded and may explain the increased susceptibility of amastigotes for compounds 13b and 13e. Comparison of the IC50s of the aziridine-2,3-dicarboxylatebased inhibitors 13b and 13e with the IC50s of the epoxidebased inhibitors E64 and CA074 demonstrated that aziridines are much more efficient in induction of cell death in L. major promastigotes, despite their weaker activity against cathepsins in protein lysates. This is in accordance with the findings of several previous studies (16, 21, 32). No direct inhibitory effect on the growth of L. major promastigotes in vitro could be detected after treatment with epoxide-based inhibitors CA074 (16, 32) and N-{L-3-trans-[2-(pyridin-2-yl)ethylcarbamoyl-oxirane-2-carbonyl]L-phenylalanine} dimethylamide (CLIK-148) (21). The difference in antileishmanial activities between epoxide- and aziridine-based inhibitors may be caused by a much higher lipophilicity and, thus, much better cell permeation of compounds 13b and 13e. The low reactivity toward nucleophilic ring opening by thiolates of aziridine-based inhibitors compared to the reactivity of epoxides is the main reason for their weaker activity against the enzymes in protein lysates (11, 12, 39, 40). Ultrastructural analyses revealed that the early phase of cell death in inhibitor-treated L. major promastigotes was characterized by accumulation of debris in lysosome-like vesicles and eventual rupture of membranes surrounding these organelles. A similar phenotype was also observed in compound 13btreated amastigotes residing in macrophages. Furthermore, an accumulation of endosomes to form multivesicular bodies near the flagellar pocket was detected in inhibitor-treated promastigotes. Similar phenotypes were also described by other groups for L. major and L. mexicana after treatment with cysteine cathepsin inhibitors (31, 44) and resemble alterations seen in lysosomal storage diseases caused by the deficiency or absence of specific acidic hydrolases. Localization of cysteine cathepsins in lysosome-like vacuoles of L. major promastigotes was described by Boukai and coworkers (4). However, the lysosomal compartment is not only the end point of the endocytic pathway; it is also associated with the crucial autodegradative system, the autophagic pathway (35). As reported previously, cysteine cathepsins contribute to autophagic processes in Leishmania spp. (44). In the present study, we observed that lysosome-like organelles engulf cytoplasmic material by microautophagy-like invaginations (35) and fuse with degradative autophagic vacuoles in a macroautophagy-like way (35). Similar macroautophagic structures were recently described in tumor cells (42). Further studies are required to characterize the

5039

origin of these structures in detail and to understand the interplay between endocytic and autophagic pathways. However, nutrient depletion by inhibition of cysteine cathepsin activity in lysosome-like vacuoles involved in endocytic and autophagic pathways in partially starved, stationary-phase L. major promastigotes might be the prime event responsible for cell death induction. Furthermore, we analyzed the mode of cell death induced in promastigotes after treatment with compounds 13b and 13e. Three main forms of cell death described for mammalian cells are apoptosis, necrosis, and autophagic dell death (14). Induction of autophagic cell death in compound 13b- and 13etreated L. major promastigotes could be excluded because no pronounced intracellular vacuolization with intact catabolic digestion was observed but chromatin condensation was observed by transmission electron microscopy (14). Furthermore, the results of flow cytometric analyses as well as the cell morphology studies suggested an apoptosis-like cell death in promastigotes. Several cytoplasmic and nuclear characteristics of apoptosis, including cell shrinkage, long maintenance of plasma membrane integrity, loss of mitochondrial transmembrane potential, and DNA degradation, were observed after treatment of L. major promastigotes with cysteine cathepsin inhibitors 13b and 13e. A high percentage of early apoptotic (annexin V-positive/PI-negative) and late apoptotic (annexin V-positive/PI-positive) promastigotes could be detected by flow cytometry. However, transmission electron microscopy revealed that compound 13b- and 13e-induced cell death was associated with dilation of the single giant mitochondrion and a complete cytoplasmic and lysosomal disorganization, as described for necrosis-like cell death in mammalian cells (14). Maintenance or condensation of cytoplasm and organelles characterizing mammalian apoptosis were not observed (14). Only weak chromatin condensation in the early phase of inhibitor treatment could be detected. Eventually, many inhibitor-treated promastigotes seemed to be totally empty and consisted only of intact plasma membranes and shortened flagella in the late phase of cell death, as detected by transmission electron microscopy. Although the induction of apoptosis-like cell death has frequently been reported for Leishmania after treatment with leishmanicidal drugs (17, 18, 22), the existence of apoptosis in these parasites remains controversial (2, 15, 45). Zangger and colleagues suggested that the programmed cell death mechanism in Leishmania is quite different from the apoptotic cell death in mammals (45). Interestingly, annexin V-positive apoptotic L. major promastigotes in inoculants and in the gastrointestinal tract of sandflies, which contribute to the intraneutrophil survival of nonapoptotic promastigotes, have a round cell morphology, similar to that of promastigotes treated with compounds 13b and 13e, are positive for TUNEL staining, and are also ultrastructurally characterized by swollen mitochondria and cytoplasmic destruction (36, 43). Similar ultrastructural alterations of mitochondria and other organelles have been described after drug treatment of Leishmania promastigotes (2, 24, 33). Van Zandbergen and colleagues observed annexin V-positive promastigotes without any detectable DNA and suggested that loss of DNA is typical for the late phase of apoptotic cell death in Leishmania (36). Taken together, our data are consistent with the hypothesis of Zangger and colleagues that induction of cell death in Leish-

5040

SCHURIGT ET AL.

mania may function via the lysosomes/lysosome-like vacuoles and is characterized by rupture of their membranes and release of several digestive lysosomal enzymes, e.g., proteinases, lipases, and DNases, into the cytoplasm, leading to complete intracellular digestion of parasites surrounded by an intact cellular membrane (45). Zangger et al. postulated that a unique mode of parasitic cell death may have developed in Leishmania to avoid induction of an antileishmanial immune response in the host (45). Thus, the morphological alterations, including the loss of nuclei and other potentially immunogenic intracellular compartments induced by compounds 13b and 13e in promastigotes and amastigotes, seem to be characteristic for the late phase of differentiation-dependent (normal) and drug-induced cell death in L. major. In conclusion, the present study extends our previous findings by documenting that the leishmanicidal aziridine-2,3-dicarboxylate-based CP inhibitors 13b and 13e mainly inhibited the cathepsin B-like enzyme CPC of L. major. Both aziridine2,3-dicarboxylate-based cysteine cathepsin inhibitors were more potent inducers of cell death in L. major promastigotes than epoxide-based cysteine cathepsin inhibitors. The data suggest that the cell death cascade in L. major promastigotes was initialized by interference with the endocytic and autophagic pathways. Our results clearly show that cell death induction in L. major promastigotes by aziridine-2,3-dicarboxylate-based cysteine cathepsin inhibitors is significantly different from mammalian apoptosis, although many characteristics of apoptotic cell death can be detected. ACKNOWLEDGMENTS This work was supported by a grant of the Deutsche Forschungsgemeinschaft (DFG Collaborative Research Center 630, Recognition, Preparation and Functional Analysis of Agents against Infectious Diseases). We thank Bianca Ro ¨ger, Christina de Witt, and Christina Kober for technical assistance. We thank the group of George Krohne (Biocenter of the University of Wu ¨rzburg, Core Unit for Electron Microscopy), especially Daniela Bunsen and Claudia Gehrig. REFERENCES 1. Alexander, J., G. H. Coombs, and J. C. Mottram. 1998. Leishmania mexicana cysteine proteinase-deficient mutants have attenuated virulence for mice and potentiate a Th1 response. J. Immunol. 161:6794–6801. 2. Arnoult, D., K. Akarid, A. Grodet, P. X. Petit, J. Estaquier, and J. C. Ameisen. 2002. On the evolution of programmed cell death: apoptosis of the unicellular eukaryote Leishmania major involves cysteine proteinase activation and mitochondrion permeabilization. Cell Death Differ. 9:65–81. 3. Bart, G., M. J. Frame, R. Carter, G. H. Coombs, and J. C. Mottram. 1997. Cathepsin B-like cysteine proteinase-deficient mutants of Leishmania mexicana. Mol. Biochem. Parasitol. 88:53–61. 4. Boukai, L. K., D. da Costa-Pinto, M. J. Soares, D. McMahon-Pratt, and Y. M. Traub-Cseko. 2000. Trafficking of cysteine proteinase to Leishmania lysosomes: lack of involvement of glycosylation. Mol. Biochem. Parasitol. 107:321–325. 5. Chan, V. J., P. M. Selzer, J. H. McKerrow, and J. A. Sakanari. 1999. Expression and alteration of the S2 subsite of the Leishmania major cathepsin B-like cysteine protease. Biochem. J. 340(Pt. 1):113–117. 6. de Araujo Soares, R. M., A. L. dos Santos, M. C. Bonaldo, A. F. de Andrade, C. S. Alviano, J. Angluster, and S. Goldenberg. 2003. Leishmania (Leishmania) amazonensis: differential expression of proteinases and cell-surface polypeptides in avirulent and virulent promastigotes. Exp. Parasitol. 104: 104–112. 7. Gelhaus, C., R. Vicik, R. Hilgenfeld, C. L. Schmidt, M. Leippe, and T. Schirmeister. 2004. Synthesis and antiplasmodial activity of a cysteine protease-inhibiting biotinylated aziridine-2,3-dicarboxylate. Biol. Chem. 385: 435–438. 8. Gelhaus, C., R. Vicik, T. Schirmeister, and M. Leippe. 2005. Blocking effect of a biotinylated protease inhibitor on the egress of Plasmodium falciparum merozoites from infected red blood cells. Biol. Chem. 386:499–502.

ANTIMICROB. AGENTS CHEMOTHER. 9. Greenbaum, D., K. F. Medzihradszky, A. Burlingame, and M. Bogyo. 2000. Epoxide electrophiles as activity-dependent cysteine protease profiling and discovery tools. Chem. Biol. 7:569–581. 10. Hanada, K., M. Tamai, S. Morimoto, T. Adachi, S. Ohmura, J. Sawada, and I. Tanaka. 1978. Inhibitory activities of E-64 derivatives on papain. Agric. Biol. Chem. 42:537–541. 11. Helten, H., T. Schirmeister, and B. Engels. 2004. Model calculations about the influence of protic environments on the alkylation step of epoxide, aziridine and thiirane based cysteine protease inhibitors. J. Phys. Chem. 108:7691–7701. 12. Helten, H., T. Schirmeister, and B. Engels. 2005. Theoretical studies about the influence of different ring substituents on the nucleophilic ring opening of three-membered heterocycles and possible implications for the mechanisms of cysteine protease inhibitors. J. Org. Chem. 70:233–237. 13. Kedzierski, L., A. Sakthianandeswaren, J. M. Curtis, P. C. Andrews, P. C. Junk, and K. Kedzierska. 2009. Leishmaniasis: current treatment and prospects for new drugs and vaccines. Curr. Med. Chem. 16:599–614. 14. Kroemer, G., L. Galluzzi, P. Vandenabeele, J. Abrams, E. S. Alnemri, E. H. Baehrecke, M. V. Blagosklonny, W. S. El-Deiry, P. Golstein, D. R. Green, M. Hengartner, R. A. Knight, S. Kumar, S. A. Lipton, W. Malorni, G. Nunez, M. E. Peter, J. Tschopp, J. Yuan, M. Piacentini, B. Zhivotovsky, and G. Melino. 2009. Classification of cell death: recommendations of the Nomenclature Committee on Cell Death 2009. Cell Death Differ. 16:3–11. 15. Lee, N., S. Bertholet, A. Debrabant, J. Muller, R. Duncan, and H. L. Nakhasi. 2002. Programmed cell death in the unicellular protozoan parasite Leishmania. Cell Death Differ. 9:53–64. 16. Maekawa, Y., K. Himeno, H. Ishikawa, H. Hisaeda, T. Sakai, T. Dainichi, T. Asao, R. A. Good, and N. Katunuma. 1998. Switch of CD4⫹ T cell differentiation from Th2 to Th1 by treatment with cathepsin B inhibitor in experimental leishmaniasis. J. Immunol. 161:2120–2127. 17. Mehta, A., and C. Shaha. 2004. Apoptotic death in Leishmania donovani promastigotes in response to respiratory chain inhibition: complex II inhibition results in increased pentamidine cytotoxicity. J. Biol. Chem. 279: 11798–11813. 18. Misra, P., T. Khaliq, A. Dixit, S. SenGupta, M. Samant, S. Kumari, A. Kumar, P. K. Kushawaha, H. K. Majumder, A. K. Saxena, T. Narender, and A. Dube. 2008. Antileishmanial activity mediated by apoptosis and structurebased target study of peganine hydrochloride dihydrate: an approach for rational drug design. J. Antimicrob. Chemother. 62:998–1002. 19. Mottram, J. C., G. H. Coombs, and J. Alexander. 2004. Cysteine peptidases as virulence factors of Leishmania. Curr. Opin. Microbiol. 7:375–381. 20. Nakao, Y., M. Fujita, K. Warabi, S. Matsunaga, and N. Fusetani. 2000. Miraziridine A, a novel cysteine protease inhibitor from the marine sponge Theonella aff. mirabilis. J. Am. Chem. Soc. 122:10462–10463. 21. Onishi, K., Y. Li, K. Ishii, H. Hisaeda, L. Tang, X. Duan, T. Dainichi, Y. Maekawa, N. Katunuma, and K. Himeno. 2004. Cathepsin L is crucial for a Th1-type immune response during Leishmania major infection. Microbes Infect. 6:468–474. 22. Paris, C., P. M. Loiseau, C. Bories, and J. Breard. 2004. Miltefosine induces apoptosis-like death in Leishmania donovani promastigotes. Antimicrob. Agents Chemother. 48:852–859. 23. Ponte-Sucre, A., R. Vicik, M. Schultheis, T. Schirmeister, and H. Moll. 2006. Aziridine-2,3-dicarboxylates, peptidomimetic cysteine protease inhibitors with antileishmanial activity. Antimicrob. Agents Chemother. 50:2439–2447. 24. Rodrigues, J. C., M. Attias, C. Rodriguez, J. A. Urbina, and W. Souza. 2002. Ultrastructural and biochemical alterations induced by 22,26-azasterol, a delta(24(25))-sterol methyltransferase inhibitor, on promastigote and amastigote forms of Leishmania amazonensis. Antimicrob. Agents Chemother. 46:487–499. 25. Schirmeister, T. 1999. Inhibition of cysteine proteases by peptides containing aziridine-2,3-dicarboxylic acid building blocks. Biopolymers 51:87–97. 26. Schirmeister, T. 1999. New peptidic cysteine protease inhibitors derived from the electrophilic alpha-amino acid aziridine-2,3-dicarboxylic acid. J. Med. Chem. 42:560–572. 27. Schirmeister, T., and A. Klockow. 2003. Cysteine protease inhibitors containing small rings. Mini Rev. Med. Chem. 3:585–596. 28. Schleicher, U., and C. Bogdan. 2009. Generation, culture and flow-cytometric characterization of primary mouse macrophages. Methods Mol. Biol. 531:203–224. 29. Schulz, F., C. Gelhaus, B. Degel, R. Vicik, S. Heppner, A. Breuning, M. Leippe, J. Gut, P. J. Rosenthal, and T. Schirmeister. 2007. Screening of protease inhibitors as antiplasmodial agents. Part I. Aziridines and epoxides. ChemMedChem 2:1214–1224. 30. Selzer, P. M., X. Chen, V. J. Chan, M. Cheng, G. L. Kenyon, I. D. Kuntz, J. A. Sakanari, F. E. Cohen, and J. H. McKerrow. 1997. Leishmania major: molecular modeling of cysteine proteases and prediction of new nonpeptide inhibitors. Exp. Parasitol. 87:212–221. 31. Selzer, P. M., S. Pingel, I. Hsieh, B. Ugele, V. J. Chan, J. C. Engel, M. Bogyo, D. G. Russell, J. A. Sakanari, and J. H. McKerrow. 1999. Cysteine protease inhibitors as chemotherapy: lessons from a parasite target. Proc. Natl. Acad. Sci. U. S. A. 96:11015–11022. 32. Somanna, A., V. Mundodi, and L. Gedamu. 2002. Functional analysis of

VOL. 54, 2010

33.

34.

35.

36.

37.

38. 39.

LEISHMANICIDAL PROTEINASE INHIBITORS INDUCE CELL DEATH

cathepsin B-like cysteine proteases from Leishmania donovani complex. Evidence for the activation of latent transforming growth factor beta. J. Biol. Chem. 277:25305–25312. Tempone, A. G., A. C. da Silva, C. A. Brandt, F. S. Martinez, S. E. Borborema, M. A. da Silveira, and H. F. de Andrade, Jr. 2005. Synthesis and antileishmanial activities of novel 3-substituted quinolines. Antimicrob. Agents Chemother. 49:1076–1080. Towatari, T., T. Nikawa, M. Murata, C. Yokoo, M. Tamai, K. Hanada, and N. Katunuma. 1991. Novel epoxysuccinyl peptides. A selective inhibitor of cathepsin B, in vivo. FEBS Lett. 280:311–315. Uchiyama, Y., M. Shibata, M. Koike, K. Yoshimura, and M. Sasaki. 2008. Autophagy-physiology and pathophysiology. Histochem. Cell Biol. 129:407– 420. van Zandbergen, G., A. Bollinger, A. Wenzel, S. Kamhawi, R. Voll, M. Klinger, A. Muller, C. Holscher, M. Herrmann, D. Sacks, W. Solbach, and T. Laskay. 2006. Leishmania disease development depends on the presence of apoptotic promastigotes in the virulent inoculum. Proc. Natl. Acad. Sci. U. S. A. 103:13837–13842. Vermelho, A. B., S. G. De Simone, C. Masini d’Avila-Levy, A. L. Souza do Santos, A. C. Nogueira de Melo, F. Paes Silva, E. Pinto da Silva Bon, and M. H. Branquinh. 2007. Trypanosomatidae peptidases: a target for drugs development. Curr. Enzyme Inhibition 3:19–48. Vicik, R., M. Busemann, K. Baumann, and T. Schirmeister. 2006. Inhibitors of cysteine proteases. Curr. Top. Med. Chem. 6:331–353. Vicik, R., M. Busemann, C. Gelhaus, N. Stiefl, J. Scheiber, W. Schmitz, F. Schulz, M. Mladenovic, B. Engels, M. Leippe, K. Baumann, and T.

40.

41.

42.

43.

44.

45.

5041

Schirmeister. 2006. Aziridide-based inhibitors of cathepsin L: synthesis, inhibition activity, and docking studies. ChemMedChem 1:1126–1141. Vicik, R., H. Helten, T. Schirmeister, and B. Engels. 2006. Rational design of aziridine containing cysteine protease inhibitors with improved potency— studies on inhibition mechanism. ChemMedChem 1:1021–1028. Vicik, R., V. Hoerr, M. Glaser, M. Schultheis, E. Hansell, J. H. McKerrow, U. Holzgrabe, C. R. Caffrey, A. Ponte-Sucre, H. Moll, A. Stich, and T. Schirmeister. 2006. Aziridine-2,3-dicarboxylate inhibitors targeting the major cysteine protease of Trypanosoma brucei as lead trypanocidal agents. Bioorg. Med. Chem. Lett. 16:2753–2757. Waguri, S., M. Kohmura, T. Gotow, T. Watanabe, Y. Ohsawa, E. Kominami, and Y. Uchiyama. 1999. The induction of autophagic vacuoles and the unique endocytic compartments, C-shaped multivesicular bodies, in GH4C1 cells after treatment with 17beta-estradiol, insulin and EGF. Arch. Histol. Cytol. 62:423–434. Wanderley, J. L., L. H. Pinto da Silva, P. Deolindo, L. Soong, V. M. Borges, D. B. Prates, A. P. de Souza, A. Barral, J. M. Balanco, M. T. do Nascimento, E. M. Saraiva, and M. A. Barcinski. 2009. Cooperation between apoptotic and viable metacyclics enhances the pathogenesis of Leishmaniasis. PLoS One 4:e5733. Williams, R. A., L. Tetley, J. C. Mottram, and G. H. Coombs. 2006. Cysteine peptidases CPA and CPB are vital for autophagy and differentiation in Leishmania mexicana. Mol. Microbiol. 61:655–674. Zangger, H., J. C. Mottram, and N. Fasel. 2002. Cell death in Leishmania induced by stress and differentiation: programmed cell death or necrosis? Cell Death Differ. 9:1126–1139.