Membrane binding requirements for the cytolytic activity of Leishmania amazonensis leishporin

FEBS Letters 583 (2009) 3209–3214

journal homepage: www.FEBSLetters.org

Membrane binding requirements for the cytolytic activity of Leishmania amazonensis leishporin Thiago Castro-Gomes a, Flávia Regina Almeida-Campos a,1, Carlos Eduardo Calzavara-Silva a,2, Rosiane Aparecida da Silva a,3, Frédéric Frézard b, Maria Fátima Horta a,* a b

Departamento de Bioquímica e Imunologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil Departamento de Fisiologia e Biofísica, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil

a r t i c l e

i n f o

Article history: Received 23 July 2009 Revised 2 September 2009 Accepted 3 September 2009 Available online 6 September 2009 Edited by Michael Ibba Keywords: Leishporin Cytolysin Pore-forming protein Lipid-binding protein Cholesterol Liposome Leishmania

a b s t r a c t To lyse cells, some pore-forming proteins need to bind to receptors on their targets. Studying the binding requirements of Leishmania amazonensis leishporin, we have shown that protease-treated erythrocytes are as sensitive to leishporin-mediated lysis as untreated cells, indicating that protein receptors are dispensable. Similarly, carbohydrate receptors do not seem to be needed, since several sugars do not inhibit leishporin-mediated hemolysis. Conversely, dipalmitoylphosphatidylcholine (DPPC), but not cholesterol, completely inhibits leishporin-mediated lysis. DPPC liposomes, with or without cholesterol, are lysed by leishporin and remove its lytic activity. Our results demonstrate that leishporin is a cholesterol-independent cytolysin that binds directly to phospholipids. Ó 2009 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Protozoa of the genus Leishmania comprise many species that cause leishmaniasis, an endemic disease in several continents. The parasite is spread to mammalian hosts through the bite of a Phlebotomus or Lutzomia sandfly containing the infective metacyclic promastigotes, which end up infecting macrophages. Inside

Abbreviations: CHAPS, 3[(3-cholamidopropyl)dimethyl-ammonium]-2hydroxy-propanesulfonate; DPPC, dipalmitoylphosphatidylcholine; DOPC, dioleylphosphatidylcholine; FBS, fetal bovine serum; GPI, glycosylphosphatidylinositol; H50, inverse of the dilution that caused 50% of hemolysis; HuE, human erythrocytes; mExt, promastigotes membrane detergent-soluble extract; PFP, pore-forming protein(s); PBS, phosphate-buffered saline * Corresponding author. Address: Av. Antônio Carlos, 6627, 31270-901 Belo Horizonte, MG, Brazil. Fax: +55 31 3409 2614. E-mail address: [email protected] (M.F. Horta). 1 Present address: Depto. Ciências Fisiológicas, Instituto de Ciências Biológicas, Universidade Federal do Amazonas, Av. Gal. Rodrigo Octávio Jordão Ramos, 3000, Bairro Coroado I, Manaus, AM 69077-000, Brazil. 2 Present address: Departamento de Imunologia, Laboratório de Imunologia Celular e Molecular, Centro de Pesquisas René Rachou, Fundação Oswaldo Cruz, Av. Augusto de Lima, 1715 – Barro Preto, Belo Horizonte 30190-002, MG, Brazil. 3 Present address: Laboratório de Parasitologia Celular e Molecular, Centro de Pesquisas René Rachou, Fundação Oswaldo Cruz, Av. Augusto de Lima, 1715 – Barro Preto, Belo Horizonte 30190-002, MG, Brazil.

phagolysosomes, they change into amastigotes, proliferate and, ultimately, exit the host cell, to infect healthy ones. The parasite’s life cycle is completed when, during a blood meal, the sandfly ingests amastigotes, which evolve again to metacyclic promastigotes. A crucial but poorly understood step in the pathogenesis of intracellular microorganisms is their exit from host cells. Its consequences are patent both in the amplification of the infection, thus exacerbating the disease and the rate of transmission, and in the outcome of the immune response. Over the last years, we have described and studied a cytolytic activity in Leishmania amazonensis that functions optimally at 37 °C and pH 5.5 (conditions found inside phagolysosome), but also at lower temperatures and/or neutral pH [1,2]. This cytolytic activity is due to pore-formation on target membranes by the action or on the dependence of a protein (or proteins) [2,3] and results from either proteolysis [4] or dissociation of an oligopeptide inhibitor (unpublished result). We have been referring to the component(s) responsible for the pore-formation as leishporin [3–5], although the putative protein(s) remain(s) unidentified. The above features are compatible with a function of lysing membranes from both inside the phagolysosome and, later on, the cytosol. We have then postulated that the exit of Leishmania from macrophages is not simply a consequence of pathogen burden or cell stress, as usually assumed, but could be caused by the action of leishporin [1–6]. In fact, the active egress of

0014-5793/$36.00 Ó 2009 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2009.09.005

3210

T. Castro-Gomes et al. / FEBS Letters 583 (2009) 3209–3214

intracellular pathogens from within the host cell can be orchestrated by the pathogens themselves through a variety of strategies, including pore-formation [7,8]. Pore-forming proteins (PFPs) require binding to specific molecules on their target membranes, which can be lipids, carbohydrates or proteins [9–11]. Here, we have studied the requirements of leishporin to bind to and lyse cells. Using erythrocytes and liposomes as targets, we showed that, to lyse membranes, leishporin bind directly to phospholipids not requiring cholesterol, proteins or carbohydrates. 2. Materials and methods 2.1. Parasites The PH8 (IFLA:PA:67:PH8) strain of Leishmania (Leishmania) amazonensis was used. Promastigotes were grown in Schneider medium (Sigma) with 10% heat-inactivated fetal bovine serum (FBS) (CULTILAB, Campinas, Brazil) and 50 lg/ml gentamycin (Sigma). Four-day cultured parasites were washed with phosphate-buffered saline (PBS) at 1000g, and pellets were kept at 80 °C until required. 2.2. Parasite membrane extract Promastigotes were resuspended in 50 mM borate buffer, pH 7.0, to a density of 2  109 cells/ml and subjected to five cycles of freeze-and-thaw [2]. The extract was centrifuged at 1000g for 10 min to sediment intact cells and nuclei and the supernatant centrifuged for 1 h at 16 000g. The membrane-rich pellet was resuspended in the same buffer containing 0.4% (sub-lytical) CHAPS (3[(3-cholamidopropyl) dimethyl-ammonium]-2-hydroxypropanesulfonate) and kept on ice for 1 h with occasional agitation. The suspension was centrifuged at 100 000g for 1 h, and the supernatant, corresponding to the solubilized membrane molecules, was referred to as promastigotes membrane detergent-soluble extract (mExt). 2.3. Liposomes Liposomes were made of DPPC (La-dipalmitoylphosphatidylcholine) or DOPC (dioleylphosphatidylcholine), with or without cholesterol (10, 15, 20 or 30% w/w). To prepare large unilamellar vesicles, a lipid film was hydrated with PBS containing 30 mM phospholipids. The suspension was subjected to 10 cycles of freeze-and-thaw and extruded through 200-nm pore size polycarbonate membranes. To obtain calcein-containing small unilamellar liposomes, the lipid film was hydrated with a 75 mM calcein solution, pH 7.4, and the suspension was ultrasonicated (Sonics Vibra-Cell, USA). Calcein-loaded liposomes were separated from free calcein by Sephadex G50. 2.4. Hemolytic assay Hemolysis was assessed in human erythrocytes (HuE) [2]. Briefly, 5  106 cells in 200 ll 20 mM acetate buffer, pH 5.5, containing 150 mM NaCl, were incubated in 96 round-bottomed well microplates with 10 ll of 1:2 serially diluted mExt treated or not as described below. After 30 min at 37 °C, microplates were centrifuged for 10 min at 500  g and hemolysis was quantified through the absorbance of the supernatant at 414 nm. The percentage of lysis was in relation to total lysis, obtained by addition of 10 ll of 0.25% Triton X-100 to the same number of HuE. Hemolytic activity was reported as percentage of lysis versus dilution factor or as H50, the inverse of the dilution that caused 50% of hemolysis.

Alternatively, HuE were previously incubated for 1 h at 37 °C with trypsin, PronaseÒ or proteinase K (12.5, 25, 50 or 100 lg/ml) in 10 mM Tris–HCl buffer, pH 7.4, containing 150 mM NaCl and 10 mM CaCl2 and washed three times with PBS. For the competition experiments, mExt was previously incubated for 30 min at 37 °C with fructose, galactose, glucose, lactose, maltose or mannose (12.5, 25, 50 or 100 lM), DPPC or cholesterol (12.5, 25, 50 or 100 mg/ll). 2.5. Removal of the lytic activity from mExt Ten microliters of mExt in a final volume of 200 l were incubated on ice with HuE, Lactobacillus acidophilus or liposomes as follows: (1) HuE: 0.2  107, 1  107 or 2  107, acetate buffer, pH 5.5, 30 min; (2) bacteria: 1  1010/ml (heated at 100 °C, 10 min or treated with 1 mg/ml lysozyme at 37 °C, 30 min), Tris–HCl buffer, pH 8.0, 15 min; (3) liposomes: 5, 15 or 20 ll of suspension, acetate buffer, pH 5.5, 30 s. Cells and liposomes were removed by centrifugation and mExt was assayed for hemolytic activity. All centrifugations in this work were carried out at 4 °C. 2.6. Liposome lysis assay Ten microliters of calcein-loaded liposomes were diluted in 90 ll of acetate buffer, pH 5.5, and incubated at 0 °C or at 37 °C with (1) mExt heated or not at 100 °C or previously incubated with calcein-free liposomes or (2) liposomes previously incubated with mExt. Twenty-microliters aliquots were collected at 5- or 10-min interval and diluted in 2 ml acetate buffer in a quartz cuvette before reading the fluorescence (wavelengths: exciting – 490 nm; emission: 515 nm) (Varian Cary Eclipse Fluorimeter). Lysis was reported as the percentage of total lysis, obtained after addition of 5 ll of 4% CHAPS. All experiments in this work were repeated at least three times and all data correspond to a typical result. 3. Results 3.1. Binding of the leishporin to cell membranes At least two steps are required for pore-formation by mExt on erythrocytes: (1) binding of the cytolysin to the target membrane, which occurs even at 0 °C with no hemolysis and (2) the pore-formation itself that probably consists of oligomerization and insertion of subunits, occurring only at higher temperatures [2,3]. To verify whether HuE remove hemolytic activity from mExt, we incubated both components on ice. As expected, HuE were sedimented without hemolysis and the hemolytic activity of the supernatant was determined. We verified that HuE removes the hemolytic activity of mExt in a dose-dependent manner (Fig. 1A). Similar results were observed when erythrocytes were replaced by prokaryotic cells. Fig. 1B shows that the Gram-positive L. acidophilus partially or completely removes hemolytic activity of mExt if previously heated at 100 °C or treated with lysozyme, respectively. The Gram-negative bacterium Escherichia coli also partially removes hemolytic activity if heated at 100 °C (not shown). This indicates that the elimination or disruption of bacterial cell wall exposes cytolysin-binding sites on the cell, showing that leishporin is able to bind to membranes other than eukaryotic ones. We next sought whether the cytolysin had a particular receptor, investigating the participation of carbohydrates, proteins and lipids. 3.2. Involvement of carbohydrates, proteins and lipids in the binding of leishporin to target membranes To investigate whether hemolysis was dependent on the binding of the cytolysin to proteins, we treated HuE with trypsin, PronaseÒ

3211

T. Castro-Gomes et al. / FEBS Letters 583 (2009) 3209–3214

100

A

120 Erytrhocytes 7 0.2 x 10 7 1 x 10 7 2 x 10

80

0.125

0.25

0.5

(µg/ml)

1

100 80

40

60

20

40

0

20 1

2

4

8

16

0

Trypsin 100

Pronase

Proteinase K

Erythrocytes treatment

B

Bacteria untreated heat-killed lisozyme-treated

80 60 40 20 0 1

2

4

8

16

Fig. 1. Removal of mExt hemolytic activity by erythrocytes or bacteria. Hemolytically active mExt were incubated with HuE (A) or untreated, heat-killed or lisozyme-treated bacteria (B). After sedimentation of cells, the hemolytic activity of mExt was evaluated. Each point represents the mean of percentage of hemolysis from duplicate wells.

Hemolytic Activity (H 50 )

Hemolysis(%)

60

A0

120

B

0

12.5

25

100 (µm)

50

100 80 60 40 20 0

Fruct

Gluc Galact Lact

Malt

Man

mExt treatment 120

C

0

0.25

1

5

10

50

(µg/ml)

100

or proteinase K before the hemolytic assay. We verified that the treatment did not reduce hemolysis, indicating that target cell proteins are not essential for the binding of and lysis mediated by leishporin (Fig. 2A). Trypsin and proteinase K even caused an increase in HuE susceptibility to lysis at the highest concentrations used. The requirement for carbohydrate in leishporin binding and hemolysis was also investigated, using a competition assay. We incubated mExt with fructose, glucose, galactose, lactose, maltose or mannose, at a wide range of concentrations, and evaluated its hemolytic activity. None of the sugars was able to reduce mExt hemolytic activity, indicating that carbohydrates might also not be important for the cytolysin-binding to HuE and hemolysis (Fig. 2B). Since phospholipids and sterols are major structural elements of biological membranes, we used the phospholipid DPPC and the non-phosphorilated lipid, cholesterol, to investigate whether leishporin was binding directly to these molecules. Fig. 2C shows that whereas cholesterol, in concentrations of 0.25–250 lg (higher concentrations not shown), is not able to inhibit leishporin activity, DPPC can totally inhibit hemolysis mediated by mExt in a dosedependent fashion. This result suggested that cholesterol does not participate in the binding of leishporin to HuE and that the most probable binding sites for the cytolysin are the membrane phospholipids. To confirm this assumption, we used liposomes with different compositions in various assays. 3.3. Direct binding of leishporin to lipids We first investigated whether liposomes made of DPPC and cholesterol, similar to HuE or bacteria, were able to pull leishporin

80 60 40 20 0

Cholesterol

DPPC

mExt Treatment Fig. 2. Involvement of erythrocytes surface proteins, carbohydrates or lipids in leishporin-mediated lysis. Hemolytic activity of mExt was evaluated in HuE previously treated with trypsin, PronaseÒ or proteinase K (A) or in untreated cells after being previously incubated with the indicated sugars (B) or lipids (C). Each point represents the mean of H50 from duplicate wells ±S.D.

down. We showed that the supernatant of mExt incubated with liposomes is totally devoid of its original hemolytic activity and the removal of this activity occurs in a dose-dependent manner (Fig. 3). It occurs immediately after the addition of the vesicles since they stay in contact with mExt only the time required for homogenization and centrifugation (around 30 s). Longer incubation periods (up to 5 min) and the replacement of DPPC by DOPC did not affect the binding of the cytolysin to liposomes (not shown). Cholesterol does not seem to have any role in the removal

3212

T. Castro-Gomes et al. / FEBS Letters 583 (2009) 3209–3214

Hemolysis(%)

100

100

Liposomes suspension 5 µl 15 µl 20 µl

80 60

40

20

20

0

0

2

4

8

16

0

of the lytic activity, since cholesterol-free DPPC vesicles were able to remove the hemolytic activity to the same extent as the cholesterol-containing liposomes (not shown). This is in accordance with the fact that mExt-mediated hemolysis is inhibited by DPPC but not by cholesterol (Fig. 2C).

Calcein release (%)

Dilution Factor Fig. 3. Removal of mExt hemolytic activity by liposomes. Hemolytically active mExt were incubated with liposomes made of DPPC and cholesterol. After sedimentation of vesicles, the hemolytic activity of mExt was evaluated. Each point represents the mean of percentage of hemolysis from duplicate wells.

mExt (37ºC) untreated (37ºC) heated at 100ºC (37ºC) liposome-incubated (0ºC) untreated

60

40

1

A

80

100 80

10

20

30

40

50

60

B

Cholesterol (%) 10 15 20 30

60 40 20 0

3.4. Lysis of liposomes To confirm that the only requirement for leishporin-mediated lysis is the presence of phospholipids, we used calcein-loaded DPPC liposomes and evaluated their susceptibility to lysis. We first verified that calcein trapped in DPPC liposomes is released when they are incubated with mExt, demonstrating that vesicles made of a single phospholipid can be lysed by leishporin (Fig. 4A). Lysis does not occur (1) if mExt is heated at 100 °C, (2) if mExt is previously incubated with calcein-free DPPC liposomes, supporting our results that leishporin is removed from mExt by interaction with phospholipids (Fig. 3), (3) at 0 °C, or (4) when liposomes have higher cholesterol content (Fig. 4B). The next experiment was carried out to ensure that liposomes were removing from mExt leishporin itself, and not other component essential for the lytic activity. Liposomes that had been previously incubated with mExt at 0 °C (at which lysis does not occur), extensively washed (still at 0 °C) and had removed the hemolytic activity from mExt were incubated with calcein-loaded liposomes that had no contact with mExt. We found that liposomes that had contact with mExt, in a suspension devoid of any soluble/unbound mExt molecule, produced lysis in the calcein-loaded liposomes (Fig. 4C). This shows that it is the cytolysin that binds to the liposomes and, possibly by liposomes fusion, it is transferred to the calcein-containing vesicles.

4. Discussion To cause cytolysis, PFPs require the binding to target cell receptors. Some PFPs recognize lipids, while other use membrane proteins or carbohydrates as receptors [9–11]. Here, we provide several evidences that the L. amazonensis cytolisin we call leishporin binds directly to membrane phospholipids leading to cytolysis. We had shown that, whereas HuE incubated with mExt at 0 °C are not lysed, hemolysis occurs if these cells are extensively washed (still at 0 °C), to remove unbound/soluble molecules, and brought to 37 °C [2]. This demonstrated that leishporin must bind to the target membrane before making the lytic pore. Here, we

0

50

5

10

15

20

25

30

C

Contact with liposomes: previously incubated with Ext-ms not incubated with Ext-ms

40 30 20 10 0

0

10

20

30

40

50

60

Time (Minutes) Fig. 4. Liposome lysis by leishporin. Calcein-loaded DPPC liposomes without (A and C) or with cholesterol (B) were incubated with native (A and B), heat-inactivated (A) or liposome-incubated mExt (A) or with DPPC liposomes that had been extensively washed after previous incubation with mExt (C) at 0 °C (A) or 37 °C (A, B and C). Lysis was evaluated by the percentage of calcein released at the indicated time.

confirmed these findings by showing that, at 0 °C, HuE (Fig. 1A) or bacteria that had their cell wall eliminated or disrupted (Fig. 1B) are able to remove leishporin-mediated hemolytic activity from mExt. For the bacteria, we assumed that the cell wall blocks leishporin binding and the treatments make plasma membrane accessible to the cytolysin. In fact, intact bacteria remove almost none of the hemolytic activity, are not lysed by leishporin and grow normally in the presence of L. amazonensis mExt (unpublished results). Some cytolysins bind to cells through protein or carbohydrates receptors [10,12–15]. Leishporin does not require surface proteins as a binding site on the erythrocytes membrane, since the treatment of these cells with different proteases did not impair mExtmediated hemolysis (Fig. 2A). On the contrary, the susceptibility to lysis increased, probably by exposing binding sites on the cell surface. HuE carbohydrates also did not seem to be used as recep-

T. Castro-Gomes et al. / FEBS Letters 583 (2009) 3209–3214

tors, since several types of sugars did not compete with binding sites on targets (Fig. 2B). The above findings suggested that leishporin uses lipids as receptor and that they are sufficient for the cytolysin’s binding to membranes and lytic activity. This assumption was corroborated by the findings that (1) DPPC completely inhibits mExt hemolytic activity (Fig. 2C), (2) liposomes made of DPPC (Fig. 3) or DOPC (not shown), similar to HuE or bacteria (Fig. 1A and B), promptly remove hemolytic activity from mExt (suggesting that leishporin is highly hydrophobic) and (3) calcein-loaded DPPC liposomes are lysed by mExt (Fig. 4A). The absence of liposomes lysis by mExt heated at 100 °C or when the assay is carried out at 0 °C confirms the thermolability of the cytolysin and the temperature-dependence of the pore-formation, respectively [1,2]. The fact that vesicles made of a single lipid can be lysed by leishporin, demonstrates that (1) it binds directly to the lipid portion of target membrane, (2) the binding occurs via phospholipids, (3) they are the only required molecules and (4) they are sufficient for the cytolysin’s membrane binding and lytic activity. Other protozoan PFPs, such as amoebapores [16] and naegleriapores [17], also insert to cell lipids without the mediation of classic receptors. Likewise, some bacterial PFPs bind directly to phosphatidylcholine, phosphatidylserine, phosphatidylethanolamine, cholesterol or sphingomyelin [18–21]. Binding to cholesterol is a feature currently used to group bacterial PFPs as a family of proteins, the cholesterol-binding cytolysins [22,23]. We showed that leishporin does not use cholesterol as binding site on target membranes. As opposed to DPPC, cholesterol is unable to inhibit mExt-mediated hemolytic activity (Fig. 2C). It is even dispensable for leishporin binding to the phospholipid and for cytolytic activity, since it does not improve the ability of DPPC liposomes to remove leishporin from mExt (not shown) or the efficiency of the cytolysin to lyse DPPC liposomes (Fig. 4B). The cholesterol-independency is in accordance with leishporin’s ability to bind to bacteria membrane (Fig. 1B). However, cholesterol may be a negative modulator of leishporinmediated lytic activity since it can prevent liposome lysis (Fig. 4B). Since previous data indicated that pore-formation by leishporin occurs by polymerization of monomers [3], it is probable that the higher content in cholesterol impairs either the insertion or the assembly of single subunits into cell membrane. However, in eukaryotic cells, cholesterol, which corresponds to about 25% of their total membrane lipids and is mainly concentrated in lipid rafts, is not an obstacle for leishporin activity. Accordingly, in our experiments, lysis, although reduced, still occurs in liposomes with 20% cholesterol. Because leishporin requires only phospholipids as ligands to cause cytolysis, it can destroy mammalian cell membranes easily and non-specifically, even in the presence of cholesterol. Therefore, susceptibility to lysis by leishporin may be shaped by the lipid composition of the membrane, reflected in the membrane fluidity. Regarding the parasite, its surface lipid composition may protect it from autolysis since, although leishporin can bind to L. amazonensis promastigotes, parasites are resistant to mExtmediated lysis and do not undergo autolysis (unpublished results). In Entamoeba histolytica, a high content of membrane cholesterol (about 40%) protects the protozoan from self-destruction by amoebapores [16]. Likewise, granulysin, a PFP from mammalian cytotoxic T lymphocytes, similarly to what we observed here for leishporin (Figs. 3 and 4B), can bind to cholesterol-containing as well as to cholesterol-free membranes, although, unlike leishporin (1–4 and Fig. 4B), lysis occurs only in the absence of cholesterol [24]. For staphylococcal alpha-toxin, clustering of phosphocholine head groups of sphingomyelin in microdomains containing cholesterol is required, so that oligomerization, a prerequisite for stable attachment of the toxin to the membrane, efficiently occurs.

3213

Outside these clusters, binding to phosphocholine is too transient for monomers to find each other. The principle of membrane targeting in the absence of any genuine, high affinity receptor may also underlie the assembly of other lipid-inserted oligomers [20], and may include leishporin. Regarding the host cell, it is worth mentioning that the parasitophorous vacuoles from macrophages infected with L. amazonensis display high levels of phosphatidylcholine [25]. They must then bind leishporin and be susceptible to permeabilization/lysis, as occurs with listeriolysin O [26], the Legionella pneumophila PFP [27], and the recently described TgPLP1 from Toxoplasma gondii [28]. Another regulation mechanism that may make leisporin-mediated lyses a vectorial event may rely on the leishporin’s need to be proteolitically activated to cause lysis [4], which could be easily fulfilled inside the parasitophorous vacuole. Therefore, our hypothesis that leishporin can disrupt the parasitophorous vacuole membrane and, later on, the plasma membrane [2,4,5], without parasite autolysis, is plausible and already currently accepted [5–8]. It is interesting that DPPC liposomes containing only bound mExt molecules cause the lysis of calcein-loaded liposomes that had no previous contact with the parasite extract (Fig. 4C). Since mExt-free liposomes do not lyse calcein-loaded liposomes, we can infer that leishporin is transferred from the calcein-free to the calcein-loaded liposomes, presumably by fusion of the vesicles. This confirms that the L. amazonensis cytolysin lyse cells using as binding sites only phospholipids, the only requirement on target membranes to form stable pores. This also provides a final demonstration that it is the cytolysin that binds to the phospholipid, ruling out the possibility that liposomes (Fig. 3), HuE or bacteria (Fig. 1) were merely removing from mExt any other component without which the hemolytic activity could not occur. This finding is in agreement with previous results showing that, although HuE incubated with mExt are not lysed at 0 °C, they are indeed lysed at 37 °C after being extensively washed to remove all unbound/soluble molecules [2]. The binding of leishporin to liposomes lead us to the obvious subsequent step, the identification of the protein(s) bound to the vesicles. Preliminary data show that various known and yet unidentified proteins are associated with mExt-treated DPPC liposomes, which we are currently identifying to further determine the identity of the cytolytic pore-forming molecule we call leishporin. Acknowledgments We thank Elimar Faria for technical assistance and the financial support from: UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR), Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), Programa de Apoio a Núcleos de Excelência (PRONEX). T.C.G. and F.R.A.C. were supported by Coordenadoria de Aperfeiçoamento de Pessoal do Ensino Superior (CAPES). C.E.C.S. was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). R.A.S. was a WHO post-doctoral fellow. M.F.H. and F.F. are CNPq research fellows. References [1] Noronha, F.S., Ramalho-Pinto, F.J. and Horta, M.F. (1994) Identification of a putative pore-forming hemolysin active at acid pH in Leishmania amazonensis. Braz. J. Med. Biol. Res. 27, 477–482. [2] Noronha, F.S., Ramalho-Pinto, F.J. and Horta, M.F. (1996) Cytolytic activity in the genus Leishmania: involvement of a putative pore-forming protein. Infect. Immun. 64, 3975–3982. [3] Noronha, F.S., Cruz, J.S., Beirão, P.S. and Horta, M.F. (2000) Macrophage damage by Leishmania amazonensis cytolysin: evidence of pore formation on cell membrane. Infect. Immun. 68, 4578–4584.

3214

T. Castro-Gomes et al. / FEBS Letters 583 (2009) 3209–3214

[4] Almeida-Campos, F.R. and Horta, M.F. (2000) Proteolytic activation of leishporin: evidence that Leishmania amazonensis and Leishmania guyanensis have distinct inactive forms. Mol. Biochem. Parasitol. 111, 363–375. [5] Almeida-Campos, F.R., Noronha, F.S. and Horta, M.F. (2002) The multitalented pore-forming proteins of intracellular pathogens. Microb. Infect. 4, 741–750. [6] Horta, M.F. (1997) Pore-forming proteins in pathogenic protozoan parasites. Trends Microbiol. 5, 363–366. [7] Hybiske, K. and Stephens, R.S. (2008) Exit strategies of intracellular pathogens. Nat. Rev. Microbiol. 6, 99–110. [8] Roiko, M.S. and Carruthers, V.B. (2009) New roles for perforins and proteases in apicomplexan egress. Cell. Microbiol. 11, 1444–1452. [9] Geny, B. and Popoff, M.R. (2006) Bacterial protein toxins and lipids: pore formation or toxin entry into cells. Biol. Cell 98, 667–678. [10] Bravo, A., Gill, S.S. and Soberón, M. (2007) Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon 49, 423–435. [11] Young, J.A. and Collier, R.J. (2007) Anthrax toxin: receptor binding, internalization, pore formation, and translocation. Annu. Rev. Biochem. 76, 243–265. [12] Cortajarena, A.L., Goñi, F.M. and Ostolaza, H. (2001) Glycophorin as a receptor for Escherichia coli alpha-hemolysin in erythrocytes. J. Biol. Chem. 276, 12513– 12519. [13] Gordon, V.M., Nelson, K.L., Buckley, J.T., Stevens, V.L., Tweten, R.K., Elwood, P.C. and Leppla, S.H. (1999) Clostridium septicum alpha toxin uses glycosylphosphatidylinositol-anchored protein receptors. J. Biol. Chem. 274, 27274–27280. [14] Lang, S., Xue, J., Guo, Z. and Palmer, M. (2007) Streptococcus agalactiae CAMP factor binds to GPI-anchored proteins. Med. Microbiol. Immunol. 196, 1–10. [15] Farrand, S., Hotze, E., Friese, P., Hollingshead, S.K., Smith, D.F., Cummings, R.D., Dale, G.L. and Tweten, R.K. (2008) Characterization of a streptococcal cholesterol-dependent cytolysin with a Lewis y and b specific lectin domain. Biochemistry 47, 7097–7107. [16] Andrä, J., Berninghausen, O. and Leippe, M. (2004) Membrane lipid composition protects Entamoeba histolytica from self-destruction by its poreforming toxins. FEBS Lett. 564, 109–115. [17] Young, J.D. and Lowrey, D.M. (1989) Biochemical and functional characterization of a membrane-associated pore-forming protein from the pathogenic ameboflagellate Naegleria fowleri. J. Biol. Chem. 264, 1077–1083.

[18] Potrich, C., Bastiani, H., Colin, D.A., Huck, S., Prévost, G. and Dalla Serra, M. (2009) The influence of membrane lipids in Staphylococcus aureus gamma hemolysins pore formation. J. Membr. Biol. 227, 13–24. [19] Bakrac, B., Gutiérrez-Aguirre, I., Podlesek, Z., Sonnen, A.F., Gilbert, R.J., Macek, P., Lakey, J.H. and Anderluh, G. (2008) Molecular determinants of sphingomyelin specificity of a eukaryotic pore-forming toxin. J. Biol. Chem. 283, 18665–18677. [20] Valeva, A., Hellmann, N., Walev, I., Strand, D., Plate, M., Boukhallouk, F., Brack, A., Hanada, K., Decker, H. and Bhakdi, S. (2006) Evidence that clustered phosphocholine head groups serve as sites for binding and assembly of an oligomeric protein pore. J. Biol. Chem. 281, 26014–26021. [21] Flanagan, J., Tweten, R., Johnson, A. and Heuck, A. (2009) Cholesterol exposure at the membrane surface is necessary and sufficient to trigger perfringolysin O binding. Biochemistry 48, 3977–3987. [22] Alouf, J.E. (2000) Cholesterol-binding cytolytic protein toxins. Int. J. Med. Microbiol. 290, 351–356. [23] Tweten, R.K. (2005) Cholesterol-dependent cytolysins, a family of versatile pore-forming toxins. Infect. Immun. Infect. Immun. 73, 6199– 6209. [24] Barman, H., Walch, M., Latinovic-Golic, S., Dumrese, C., Dolder, M., Groscurth, P. and Ziegler, U. (2006) Cholesterol in negatively charged lipid bilayers modulates the effect of the antimicrobial protein granulysin. J. Membr. Biol. 212, 29–39. [25] Henriques, C., Atella, G.C., Bonilha, V.L. and de Souza, W. (2003) Biochemical analysis of proteins and lipids found in parasitophorous vacuoles containing Leishmania amazonensis. Parasitol. Res. 89, 123–133. [26] Portnoy, D.A., Auerbuch, V. and Glomski, I.J. (2002) The cell biology of Listeria monocytogenes infection: the intersection of bacterial pahogenesis and cellmediated immunity. J. Cell. Biol. 158, 409–414. [27] Alli, A.O., Gao, L.Y., Pedersen, L.L., Zink, S., Radulic, M., Doric, M. and Abu Kwaik, Y. (2000) Temporal pore-formation-mediated egress from macrophages and alveolar epithelial cells by Legionella pneumophila. Infect. Immun. 68, 6431– 6440. [28] Kafsack, B.F., Pena, J.D., Coppens, I., Ravindran, S., Boothroyd, J.C. and Carruthers, V.B. (2009) Rapid membrane disruption by a perforinlike protein facilitates parasite exit from host cells. Science 323, 530– 533.