amazonensis promastigotes induce and are killed by neutrophil extracellular traps

Anderson B. Guimara˜ es-Costaa, Michelle T. C. Nascimentoa, Giselle S. Fromenta, Rodrigo P. P. Soaresb, Fernanda N. Morgadoc,Fa´ tima Conceic¸a˜ o-Silvac, and Elvira M. Saraivaa,1

aDepartamento de Imunologia, Instituto de Microbiologia Prof. Paulo de Go´es, Universidade Federal do Rio de Janeiro, Bloco I, 21941-902, Rio de Janeiro, Brazil; bLaborato´rio de Entomologia Me´dica, Centro de Pesquisas Rene´Rachou, Fundac¸a˜ o Oswaldo Cruz, 30190-002, Belo Horizonte, Brazil; and cLaborato´rio de Imunoparasitologia, Fundac¸a˜ o Oswaldo Cruz, 21045-900, Rio de Janeiro, Brazil

Edited by Philippa Marrack, National Jewish Medical and Research Center, Denver, CO, and approved March 9, 2009 (received for review January 9, 2009) Neutrophils are short-lived leukocytes that die by apoptosis, ne- are inoculated by the insect into the and local crosis, and NETosis. Upon death by NETosis, neutrophils release inflammation begins. Neutrophils are rapidly recruited to the fibrous traps of DNA, histones, and granule proteins named neu- bite site, are activated, and ingest and kill parasites (11, 12). trophil extracellular traps (NETs), which can kill bacteria and fungi. However, some phagocytosed parasites resist the killing mech- Inoculation of the protozoan Leishmania into the mammalian skin anisms triggered in the phagolysosomes, delaying neutrophil causes local inflammation with neutrophil recruitment. Here, we apoptosis long enough to allow arrival at the investigated the release of NETs by human neutrophils upon their site (11, 13). Interestingly, it has been shown that interaction with Leishmania parasites and NETs’ ability to kill this apoptotic neutrophils harboring intracellular parasites are then protozoan. The NET constituents DNA, elastase, and histones were phagocytosed by (13). detected in traps associated to promastigotes by immunofluores- Although an increasing number of studies have investigated cence. Electron microscopy revealed that Leishmania was ensnared the role of neutrophils in the immune response to Leishmania, by NETs released by neutrophils. Moreover, Leishmania and its the influence of NETs on Leishmania survival is still unknown. surface lipophosphoglycan induced NET release by neutrophils in Induction of NETs and release of antimicrobial components may a parasite number- and dose-dependent manner. Disruption of contribute to the killing of Leishmania parasites before they are NETs by DNase treatment during Leishmania–neutrophil interac- engulfed by professional phagocytes. In the present study, we MEDICAL SCIENCES tion increased parasite survival, evidencing NETs’ leishmanicidal report that L. amazonensis promastigotes and amastigotes are effect. Leishmania killing was also elicited by NET-rich superna- able to induce NET formation in human neutrophils and that tants from phorbol 12-myristate 13-acetate-activated neutrophils. these webs possess leishmanicidal activity. Additionally, Leish- Immunoneutralization of histone during Leishmania–neutrophil mania LPG induces NET formation in neutrophils. Released interaction partially reverted Leishmania killing, and purified his- NETs exert an extracellular leishmanicidal activity caused, at tone killed the parasites. Meshes composed of DNA and elastase least in part, by the histones present in the meshes. Immuno- were evidenced in biopsies of human cutaneous . NET histochemical analysis confirmed the occurrence of these webs in is an innate response that might contribute to diminish parasite lesions from human . burden in the Leishmania inoculation site. Results histone ͉ protozoa ͉ trypanosomatid NETs from Phorbol 12-Myristate 13-Acetate (PMA)-Activated or Naïve Neutrophils Kill Leishmania. Initially, we compared the parasite eutrophils are the most abundant leukocytes in the blood killing abilities of untreated and PMA-activated neutrophils Nand are the first cells recruited to inflamed tissues. They (Fig. S1A). Exposure of promastigotes to PMA-activated neu- have a pivotal role in immunity to infection not only by ingesting trophils resulted in lower promastigote survival relative to and destroying microbes, but also by secreting various cytokines. exposure to untreated neutrophils. PMA-activated neutrophils Although their lifespan is increased in the inflammatory site, killed 53% more promastigotes than naïve cells (Fig. S1A). Next, they are short-lived cells that die by spontaneous apoptosis (1). to distinguish between phagocytic and NET-derived microbici- NETosis is a recently described mechanism of neutrophil death dal activity, PMA-activated neutrophils were treated with the that occurs upon neutrophil activation ultimately leading to the actin inhibitor cytochalasin D (CytD) to block , release of neutrophil extracellular traps (NETs). NETs are webs thereby allowing the exclusive measurement of extracellular composed of chromatin and granular proteins that have been killing. CytD addition increased by 3-fold the parasite survival shown not only to ensnare bacteria and fungi, but also to provide compared with PMA-treated control cells, indicating that a high local concentration of antimicrobial molecules (2–7). phagocytosis contributed to parasite killing by PMA-activated Leishmania protozoa are the etiological agents of leishmaniasis, neutrophils (Fig. 1A). Because NETs are degraded by treatment which include a wide spectrum of disease in humans characterized with DNase (2), this enzyme was added to PMA-treated neu- trophils followed by addition of promastigotes to the cell cul- by lesions in the skin, mucosal surfaces, and visceral organs. tures. DNase treatment protected promastigotes from killing, Worldwide, Ͼ10 million people are affected by this disease that doubling parasite survival compared with those exposed to causes significant morbidity and mortality (8). Leishmania ama- PMA-activated neutrophils (Fig. 1A). Importantly, NET release zonensis is the causative agent of human cutaneous leishmaniasis in the New World, with a high proportion of cases evolving to the

severe anergic diffuse cutaneous leishmaniasis (9). Author contributions: F.C.-S. and E.M.S. designed research; A.B.G.-C., M.T.C.N., G.S.F., and Two major Leishmania developmental stages are recognized: F.N.M. performed research; R.P.P.S. contributed new reagents/analytic tools; A.B.G.-C., amastigotes that live intracellularly in macrophages, and pro- M.T.C.N., and E.M.S. analyzed data; and E.M.S. wrote the paper. mastigotes that multiply inside the insect vector. Promastigotes The authors declare no conflict of interest. of all Leishmania species synthesize lipophosphoglycan (LPG), a This article is a PNAS Direct Submission. glycoconjugate localized over the entire protozoan cell surface, 1To whom correspondence should be addressed. E-mail: [email protected]. which is one of the first parasite molecules to contact the This article contains supporting information online at www.pnas.org/cgi/content/full/ cells (10). Leishmania infection is initiated when promastigotes 0900226106/DCSupplemental.

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0900226106 PNAS Early Edition ͉ 1of6 Downloaded by guest on September 30, 2021 ** A ** B 300 400 *

** 300 200 ** 200

(% of control) 100 (% of control)

100 survival Leishmania Leishmania survival Leishmania

0 0 PMA PMA + CytD PMA + DNase Medium CYTD DNase CYTD + DNase

Neutrophil treatment Neutrophil treatment

Fig. 1. Killing of L. amazonensis by NETs from activated and naïve neutrophils. PMA-activated (A) or naïve (B) neutrophils were treated with CytD or DNase-1 and incubated with promastigotes (1 cell/0.1 parasite ratio) for2hat35°C.Schneider’s complete medium was added to the cultures, and live parasites were counted after 2 days of incubation at 26 °C. Results of at least 7 independent experiments are shown as mean ϩ SEM. PMA raw number: 8.2 ϫ 106 Ϯ 0.7 ϫ 106 promastigotes; medium raw number: 3.2 ϫ 106 Ϯ 0.7 ϫ 106 promastigotes. *, P Ͻ 0.002; **, P Ͻ 0.0001.

was not affected by CytD treatment (Fig. S1B). The number of Because the presence of DNA released from killed parasites in promastigotes cultured in the absence of neutrophils, increased this assay cannot be excluded, we measured the amount of DNA between 70 and 272 times over the initial number. Of note, released by 107 Tween-lysed promastigotes and found 2.5 ␮g/mL promastigote growth was also unaffected by PMA, DNase, DNA released by this procedure. At the 10:1 promastigote/ CytD, or DMSO (CytD’s diluent) at the same concentrations neutrophil ratio, 22 ␮g/mL of DNA was detected (Fig. 4A). used for neutrophil treatments. Based on this value, even if all 107 promastigotes in the inter- We then tested the NET killing characteristics of naïve action medium of our original assay were killed, parasite-derived neutrophils activated only by contact with parasites. We ob- DNA would account for no more than 11% of the total DNA served that parasites were able to induce NETs, and that these content measured. In addition to L. amazonensis, promastigotes structures participated in promastigote killing (Fig. 1B). Parasite of Leishmania chagasi and Leishmania major induced NET survival increased 39% and 60% over controls when neutrophils release by naïve neutrophils (Fig. 4C). were treated with CytD and DNase, respectively. When we simultaneously treated neutrophils with CytD and DNase, pro- Promastigote LPG Induce NETs. To explore parasite molecules that mastigote survival increased 140% compared with control. could participate in the NET induction, neutrophils were incu- bated with different concentrations of LPG, and the DNA Leishmania Promastigotes Are Captured by NETs. Next, we tested content was estimated. Our results showed that LPG induced whether Leishmania promastigotes were captured by NETs NET release by neutrophils in a dose-dependent manner (Fig. formed by PMA-stimulated neutrophils, and webs with trapped 4D). The amount of DNA released by 10 ␮g/mL of LPG was 2.5 promastigotes were seen by DNA staining (DAPI and Sytox times higher than that released by nontreated neutrophils green; Fig. S2). We then evaluated the ability of promastigotes (1.9 ␮g/mL). to directly induce and be ensnared by naïve neutrophil NETs, and promastigotes alone induced NET formation upon interac- The NET Component Histone Kills Leishmania. To find molecules tion with naïve neutrophils (Fig. 2). Because elastase and histone present in NETs that could mediate parasite killing, we added are known NET components (2), we therefore stained NETs antihistone antibody to the CytD-treated neutrophil/Leishmania with DAPI, anti-elastase, and anti-H2A histone antibodies. We cocultures (Fig. 5A). Immune neutralization of histone resulted found promastigotes covered by meshes that were labeled by in a 42% increase in parasite survival relative to non-neutralized DAPI and both antibodies (Fig. 2, Figs. S3–S5, and Movies S1 controls. Addition of the elastase inhibitor methoxysuccinyl- and S2). Controls with secondary antibodies did not label the Ala-Ala-Pro-Val-chloromethylketone did not increase promas- parasites or the neutrophils. Parasites associated with NETs tigote survival, showing that elastase does not participate in the presented, in general, a damaged appearance as seen by the NET-mediated promastigote killing. live/dead assay (Fig. S6). We then tested whether parasite killing depended on NET Analyzing promastigote–NET interactions by scanning elec- integrity by incubating promastigotes with supernatants obtained tron microscopy we visualized widespread NETs with trapped from DNase-treated, PMA-activated neutrophils. Our results promastigotes (Fig. 3 A–C). An intimate contact between pro- showed that even disrupted NETs killed parasites and addition of mastigotes, NETs, and/or neutrophil granules was seen. Inter- antihistone antibody to these supernatants inhibited promastigote estingly, promastigotes trapped by NETs presented a thin, flat death in a concentration-dependent manner (Fig. 5B). Histone and body with protrusions, indicative of cell damage. In contrast, elastase were detected in these supernatants (Fig. S7A). To confirm promastigotes being phagocytosed showed a normal swelling the cytotoxic properties of histones, we treated promastigotes with shape (Fig. 3 D and E). purified H2A histone and found that histone killed 62% of the Further, NET amounts at different Leishmania/neutrophil parasites at 20 ␮g/mL (Fig. 5C). Histone killing effect on promas- ratios were measured by using the DNA picogreen assay, and we tigotes was inhibited by antihistone antibody (Fig. S7B). Likewise found that the extent of NET formation depended on parasite staining by the live/dead method confirmed promastigote death cell ratios (Fig. 4 A and B). The amount of DNA released by the mediated by histone (Fig. S6). 10:1 promastigote/neutrophil or amastigote/neutrophil ratios was 5 and 7 times higher (22 and 14 ␮g/mL) than the DNA Analysis of Human Cutaneous Leishmaniasis Lesion. To confirm the spontaneously released by nonactivated neutrophils (4.4 and 1.8 presence of these webs in vivo, lesion biopsies from patients with ␮g/mL), respectively. DNA was not detected in the supernatant cutaneous leishmaniasis were analyzed (Fig. 6). It has been obtained from control parasites incubated without neutrophils. described that lesions of cutaneous leishmaniasis can present

2of6 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0900226106 Guimara˜es-Costa et al. Downloaded by guest on September 30, 2021 A

B C MEDICAL SCIENCES DE

Fig. 3. Scanning electron microscopy of the interaction between naïve neutrophils and L. amazonensis. Naïve neutrophils were incubated with Fig. 2. Immunostaining of NETs induced by Leishmania.(A–C) Naïve neu- promastigotes (P) for1hat35°C.(A) An extended NET ensnares 3 promas- trophils were incubated with promastigotes (1:5 ratio) for1hat35°C.Cells tigotes. (B) NET fibers and neutrophil granules (arrow) are attached to a were fixed, stained with DAPI (A), antihistone (B), or anti-elastase (C), the last promastigote. (C) NET threads trap a promastigote. (D and E) Neutrophils 2 followed by Texas red-labeled (A) or FITC-labeled (B) secondary antibodies. phagocytosing promastigotes. In A–C, a NET-ensnared promastigote presents Fluorescence staining images merged with differential interference contrast a thin, flat body with protrusions (arrow in C), in contrast to swollen promas- are shown. (D) Overlay of the fluorescence images. Arrow points to NET tigotes (D and E) that were phagocytosed by neutrophils. ensnared promastigote and arrowhead points to promastigote being phago- cytosed. (Bars: 20 ␮m.) that NETs are also induced by the parasites in naïve neutrophils. Moreover, NETs formed under either condition presented leish- areas plenty of cells, whereas cells in other areas of the same manicidal activity, and the amount of NET released directly lesion can be scarce (32). We stained NETs in areas presenting correlated with increasing promastigote and amastigote/ only netosed neutrophils (Fig. 6 A–C), and in areas plenty of neutrophil ratios. NET induction seems to be a general property other cells besides neutrophils (Fig. 6D), to show that NETs occur in both areas of the same lesion. Thus, fluorescence of Leishmania, because different species of this protozoan are pictured in Fig. 6 A–C shows histone and DNA staining in a endowed with the capacity to induce NET. region of matrix deposition, where NET details could be better Released NET seems to attach to the promastigote surface, as visible because no stacked cells were observed. The elastase described for bacteria and fungi (2, 3, 5). Although NET-specific staining was performed in an area containing an abundant recognition sites for different microbes cannot be excluded, it is cellular infiltrate (Fig. 6D), to demonstrate NETs in a panoramic likely that NETs can bind any negatively-charged surface (7). It view of the lesion. Abundant meshes stained for extracellular has already been shown that L. amazonensis promastigotes DNA (Fig. 6A) closely parallels the pattern observed for that of possess a negatively-charged cell surface (14), a feature that is, histones (Fig. 6B), which is confirmed by the overlay of these 2 at least partially caused by the LPG expressed on promastigote stains (Fig. 6C). surface membrane. LPG constitutes one of the first parasite molecules to come in contact with the host and has numerous Discussion phosphodiester bridges that could efficiently bind positively- We report in this study that Leishmania promastigotes are charged ions and proteins (15). Many granule proteins and trapped by NETs released from PMA-activated neutrophils and histones are highly cationic, suggesting that NETs could bind

Guimara˜es-Costa et al. PNAS Early Edition ͉ 3of6 Downloaded by guest on September 30, 2021 A 30 B 20

20 g/ml) g/ml) µ µ 10

10 DNA ( DNA (

0 0 Ctrl 0.1:1 1:1 10:1 Ctrl 0.1:1 1:1 10:1

Promastigote:Neutrophil ratio Amastigote:neutrophil C D * * 8 25 7 * 20 6

15 5 g/ml) g/ml) µ µ 4 10 3 DNA ( DNA 5 DNA ( 2 0 1 CTRL 0.1:1 1:1 0.1:1 1:1 0 L. major L. chagasi crtl 0.1 1.0 10 LPG (µg/ml) Promastigote:Neutrophil ratio

Fig. 4. Leishmania and LPG stimulate NET release from naïve neutrophils. (A and B) Neutrophils were incubated with promastigotes (A) or amastigotes (B)of L. amazonensis at different cell ratios as indicated. Supernatants were recovered after1hat35°CandNETs were quantified. (C) Similar to A and B but using promastigotes of L. major or L. chagasi. Results of at least 3 independent experiments are shown. (A and B) P Ͻ 0.006. (C) *, P Ͻ 0.0001. (D) Neutrophils were incubated at the indicated concentrations of purified LPG for1hat35°C,supernatants were recovered, and NETs were quantified. Results of at least 6 independent experiments are shown. *, P Ͻ 0.05.

ionically to the surface of L. amazonensis. Indeed, we demon- and Toll-like receptors (18). Further studies should be per- strated that NETs associated with Leishmania promastigotes formed to elucidate the mechanism and receptors involved in the contain DNA, histones, and elastase, 3 constituents found in NET release by Leishmania LPG. The higher NET induction by NETs (2). Moreover, our results indicated that purified LPG is promastigotes relative to LPG could be explained by other also able to induce NET release by neutrophils. Several receptors NET-inducing molecules and/or the complex membrane struc- have been implicated in the LPG recognition by macrophages ture of the promastigotes, including the fact that LPG shares (16), among them, some shared by neutrophils, like CR3 (17) domains with other Leishmania membrane molecules, such as

A B * ** C * 400 100 1000 80 300 750 60 200 500 40 * (% of control) of (%

(% of control) of (% 100 (% of control) of (% 250 20

Promastigotes Survival * Promastigotes survival

Promastigotes Survival 0 0 0 e Ctrl Na-acetate 20 H2A 200 H2A n Ctrl 0.65 1.25 2.5 5 to IgG PMA CytD + µ ti-his type g/ml n so anti-histone (µg/ml) PMA a I + + ytD C + A + CytD M MA P P Neutrophil treatment

Fig. 5. Histone kills Leishmania.(A) Killing of Leishmania by NETs is inhibited by antihistone antibody. Neutrophils treated with PMA and CytD were exposed to 5 ␮g/mL of anti-H2A histone or IgG isotypic antibodies, and then incubated with promastigotes (1 cell/0.1 parasite ratio) for2hat35°C.Schneider’s complete medium was added to the cultures and live parasites were counted after 2 days incubation at 26 °C. Results of 4 independent experiments are shown as mean ϩ SEM. PMA raw number: 6.6 ϫ 105 ϩ 0.8 ϫ 105 promastigotes. *, P Ͻ 0.005. (B) Supernatants from activated neutrophils are toxic to Leishmania. Supernatants from neutrophils treated with PMA and DNase were added to 107 promastigotes in the presence or absence (Ctrl) of different concentrations of antihistone antibody. After 30 min, live parasites were counted. Results from 4 independent experiments are shown as mean ϩ SEM. Control raw number: 1.2 ϫ 106 Ϯ 0.1 ϫ 106 promastigotes. P ϭ 0.05. (C) Histone H2A is toxic to Leishmania. Purified histone H2A was added at different concentrations to 5 ϫ 106 promastigotes. After 30-min incubation at 35 °C, live parasites were counted. Sodium acetate buffer (Na-acetate, histone’s diluent), was used as a control. Results from 3 independent experiments are shown as mean ϩ SEM. *, P Ͻ 0.0001.

4of6 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0900226106 Guimara˜es-Costa et al. Downloaded by guest on September 30, 2021 PMA-activated neutrophils were toxic to promastigotes; this A B effect was neutralized by the addition of antihistone antibody. Moreover, purified histone killed promastigotes in a concentra- tion-dependent manner, showing a paramount role of histone in this phenomenon. Toxic properties of histones to microorgan- isms have been recognized for decades, and histones and histone fragments have been suggested to function as part of an ancient innate present in fish, birds, and (22). It has been shown that histones can kill bacteria (2), and toxicity of histone H1 to L. major promastigotes was recently reported (23). Promastigotes treated with H1 histone showed ultrastruc- tural alterations, such as chromatin changes, cytoplasmic vacu- olization, and a rounded appearance (23); this last morphology was also frequently observed in our microscopy studies. Al- though we have demonstrated that histones contribute to NET- mediated promastigote killing, we cannot exclude the possibility C D that other NET components could present similar parasite toxicity. NETs have been visualized in vivo under diverse settings, such as in human appendicitis, lupine Shigella-induced (2), bovine mastitis (24), human preeclampsia (25, 26), and in the blood of Plasmodium falciparum-infected children (27). Based on our findings that these lattices occur in patients’ lesions, and NETs trap and kill Leishmania, we could envisage that NETs could be relevant to control the parasite burden and modulate the immune response through the components released together with the lattices.

It has been described in the murine model of leishmaniasis MEDICAL SCIENCES that neutrophils protected C57BL/6 mice against infection, inducing Leishmania killing by a mechanism that requires mac- rophage activation by neutrophil elastase (28, 29). In our model Fig. 6. Analysis of tissue sections from lesions of cutaneous leishmaniasis. (A and B) Immunofluorescence staining with DAPI (A) reveals nuclear and extra- elastase was not directly involved in parasite toxicity; however, cellular localization of DNA, which largely overlaps with staining for histone in mice, elastase induces leishmanicidal activity of macrophages (B). (C) Overlay of the images in A and B.(D) Staining for elastase reveals through Toll-like receptor 4 (TLR4) activation (28). Thus, in our fibrous extracellular material (arrows). Tissue sections are of 6-␮m thickness. model elastase released by NETosis could indirectly contribute (Bars: 50 ␮M.) for macrophage leishmanicidal activity by a similar pathway of TLR4 activation. Our results pose a major challenge for future investigations of acid phosphatase, proteophosphoglycan, and glycosylinositol- this yet-unrecognized mechanism in leishmaniasis, that is to phospholipids (10), which might contribute to NET formation. elucidate the in vivo relevance of NET formation induced by Previous reports showed that human neutrophils are able to Leishmania. ingest and kill Leishmania (19), but recent studies described that not all ingested promastigotes are destroyed within these cells Materials and Methods (13, 20, 21). Here, we report that a phagocytosis-independent Neutrophil Purification. Human neutrophils from buffy coats of healthy blood killing mechanism, mediated by neutrophil NETs, is effective donors were isolated by density gradient centrifugation (Histopaque; Sigma– against Leishmania promastigotes, and, as reported for bacteria Aldrich) followed by hypotonic lysis of erythrocytes. Purified neutrophils and fungi (2, 3, 5), Leishmania are killed by NETs. Like these (Ն95% of the cells) were resuspended in RPMI medium 1640 (Sigma) and kept previous studies, we detected an increased promastigote survival on ice until use. All procedures were approved by the Institutional Review when NET structure was disrupted. In fact, the extracellular Board for Human Subjects (Hospital Clementino Fraga Filho, Universidade killing seems to be mediated mainly by NETs because promas- Federal do Rio de Janeiro). tigote killing by activated and naïve neutrophils was reduced when NET structure was disrupted by DNase. Parasite survival Parasites. L. amazonensis (WHOM/BR/75/Josefa), L. chagasi (MHOM/BR/2001/ HP-EMO), and L. major V1 (MHOM/IL/80/Friedlin) promastigotes were main- was maximal when neutrophils were simultaneously treated with tained at 26 °C in Schneider‘s Insect medium (Sigma) supplemented with 10% both CytD and DNase. heat-inactivated FCS, 10% human urine, and 40 ␮g/mL of gentamicin. Sta- We detected that neutrophils in vitro captured Leishmania tionary-phase promastigotes were obtained from 5- to 6-day-old cultures, and promastigotes by phagocytosis and NET trapping, and that used throughout. Amastigotes were obtained as described (30). neutrophils formed extensive lattices after interaction with parasites. Importantly, promastigotes associated with NETs pre- Neutrophil Killing Assay. Neutrophils (2 ϫ 106) were incubated with or without sented a damaged appearance (thin, flat body with protrusions, 100 nM PMA (Sigma) for 30 min. CytD (10 ␮g/mL; Sigma), protease-free ethidium homodimer staining), which was not observed in DNase-1 (100 units/mL; Fermentas Life Science), or both were then added to swelling parasites that were being phagocytosed by neutrophils. neutrophils, and cultures were maintained for 20 min, followed by the addi- During our search for factors responsible for NET-mediated tion of L. amazonensis promastigotes. After2hat35°C,FCSandhuman urine (10% each) were added to the preparation, and cultures were incubated for Leishmania killing, we initially observed that addition of anti- 2 days at 26 °C. Parasite growth was assessed by counting live promastigotes histone antibody to the neutrophil–promastigote interaction in a Neubauer chamber. In some experiments, neutrophil and promastigote medium decreased parasite death. Similarly, histone immune cocultures were incubated either with 10 ␮g/mL of the neutrophil elastase neutralization was able to rescue bacteria from killing (2). Of inhibitor methoxysuccinyl-Ala-Ala-Pro-Val-chloromethylketone (Calbio- note, we found that supernatants obtained from DNase-treated chem), 5 ␮g/mL of the antihistone H2A monoclonal antibody (donated by A.

Guimara˜es-Costa et al. PNAS Early Edition ͉ 5of6 Downloaded by guest on September 30, 2021 Zychlinsky, Max Planck Institute, Berlin), or IgG isotype control (Sigma) in the promastigotes in a Neubauer chamber after 30 min at 35 °C. To assure pro- interaction medium. mastigote killing, Schneider’s complete medium was added to the histone- treated parasites followed by culture incubation at 26 °C for 24 h in some Quantification of NET Release. Neutrophils (2 ϫ 106) were incubated with or experiments. Neutrophil–parasite interactions and promastigotes treated without Leishmania at different neutrophil/parasite ratios. After 1 h, restric- with purified histone were stained with the live/dead reagents (20 nM calcein tion enzymes (ECOR1 and HINDIII, 20 units/mL each; BioLabs) were added to AM/2 nM ethidium homodimer; Molecular Probes) according to the manu- cultures, which were then maintained for2hat35°C.NETwasquantified in facturer’s instructions, and preparations were observed unfixed. the culture supernatant by using the Picogreen dsDNA kit (Invitrogen) accord- ing to the manufacturer’s instructions. As a control, DNA released from Tissue Histochemistry. Biopsies from cutaneous leishmaniasis skin lesions were promastigotes was obtained after parasite lysis with 0.1% Tween 20 and obtained from 2 patients with Montenegro skin test positive and diagnoses quantified as above. NET-DNA and parasite-DNA concentrations were calcu- confirmed by parasite culture (32). Lesions were classified as early or late lated by using herring DNA (Sigma) as a standard. LPG from promastigotes was according to their durations, 2 and 5 months after the patients were first extracted as described (31). examined, respectively. Formalin-fixed biopsy fragments were stained with DAPI and antihistone as above. Anti-elastase (Dako Cytomation) was revealed Immunofluorescence. Neutrophils were incubated with or without 100 nM by using peroxidase-labeled secondary antibody (Zymed) with aminoethyl- PMA for 45 min, cultured with promastigotes for 1 h, and fixed with 4% carbazole (AEC kit; Zymed) as the substrate–chromogen system, and the slides ␮ paraformaldehyde. Slides were stained with Sytox Green (0.1 M; Molecular were counterstained with Mayer’s hematoxylin (Dako). Probes), DAPI (10 ␮g/mL; Sigma), anti-elastase (1:1,000; Calbiochem), or anti- histone H2A antibodies, followed by anti-rabbit-FITC (1:100; Vector Labs) or Statistical Analysis. Data were analyzed by GraphPad Prism 4.0 software. Each anti-mouse-Texas red (1:500; Molecular Probes). Confocal images were taken experiment was performed at least 3 times on independent occasions. P Ͻ 0.05 in a Zeiss LSM 510 META; epifluorescence was in a Zeiss Axioplan. was considered significant.

Scanning Electron Microscopy. Neutrophils were adhered on coverslips treated ACKNOWLEDGMENTS. We thank Drs. A. Zychlinsky and D. C. Bou-Habib for with 0.01% polylysine (Sigma). Promastigotes added at a 5:1 ratio were critical reading of this manuscript; the Hemotherapy Service Hospital Clem- incubated for1hat35°C,5%CO2. Cultures were fixed with 2.5% glutaral- entino Fraga Filho for buffy coats; Laborato´rio Ultraestrutura Celular Hertha dehyde in 0.1 M cacodylate buffer, pH 7.2, postfixed with 1% osmium tetrox- Meyer for scanning microscope use; the Program for Technological Develop- ide and 0.8% potassium ferricyanide, and dehydrated with an ascending ment in Tools for Health and Leishmania Type Culture Collection (Fundac¸a˜o ethanol series. After dehydration and critical-point drying, the specimens Oswaldo Cruz) for facilities use and Leishmania provided; Dr. F. Ribeiro-Gomes were coated with carbon and analyzed in a JEOL 1530 scanning electron (Universidade Federal do Rio de Janeiro) and Dr. A. Zychlinsky (Instituto microscope. Oswaldo Cruz, Rio de Janeiro) for reagents; F.L. Jorge (Instituto Nacional de Caˆncer, Rio de Janeiro) for purified amastigotes; H.O. Ferreira for artwork; and B. Pascarelli for confocal analysis. This work was supported by Ministe´rio Leishmanicidal Activity. Promastigotes (107) were incubated with superna- Cieˆncia e Tecnologia/Conselho Nacional de Desenvolvimento Cientı´fico e tants obtained from 100 nM PMA-stimulated neutrophils. Supernatants were Tecnolo´gico/Ministe´rio da Sau´de-Secretaria de Cieˆncia, Tecnologia e Insumos diluted 1:2 in the presence or absence of different concentrations of antihi- Estrate´gicos-Departamento de Cieˆncia e Tecnologia Grant 25/2006 and Fun- stone monoclonal antibody. Additionally, promastigotes were incubated with dac¸a˜ o de Amparo a`Pesquisa do Estado do Rio de Janeiro Grants 110.737/2007 purified histone H2A (from calf thymus, donated by A. Zychlinsky). Control and 111.584/2008. R.P.S. is supported by Conselho Nacional de Desenvolvi- parasites were incubated with RPMI 1640 medium alone or with the histone mento Cientı´ficoe Tecnolo´gico Grant 305008/2007-2 and TDR/World Health diluent acetate buffer. Parasite viability was assessed by counting motile Organization Grant A50880.

1. Nathan C (2006) Neutrophils and immunity: Challenges and opportunities. Nat Rev 20. Laufs H, et al. (2002) Intracellular survival of Leishmania major in neutrophil granulo- Immunol 6:173–182. cytes after uptake in the absence of heat-labile serum factors. Infect Immun 70:826– 2. Brinkmann V, et al (2004) Neutrophil extracellular traps kill bacteria. Science 303:1532– 835. 1535. 21. Gueirard P, Laplante A, Rondeau C, Milon G, Desjardins M (2008) Trafficking of 3. Brinkmann V, Zychlinsky A (2007) Beneficial suicide: Why neutrophils die to make NETs. promastigotes in nonlytic compartments in neutrophils enables Nat Rev Microbiol 5:577–582. the subsequent transfer of parasites to macrophages. Cell Microbiol 10:100–111. 4. Urban CF, Lourido S, Zychlinsky A (2006) How do microbes evade neutrophil killing? 22. Parseghian MH, Luhrs KA (2006) Beyond the walls of the nucleus: The role of histones Cell Microbiol 8:1687–1696. in cellular signaling and innate immunity. Biochem Cell Biol 84:589–604. 5. Urban CF, Reichard U, Brinkmann V, Zychlinsky A (2006) Neutrophil extracellular traps 23. Masina S, Zangger H, Rivier D, Fasel N (2007) Histone H1 regulates chromatin conden- capture and kill Candida albicans yeast and hyphal forms. Cell Microbiol 8:668–676. sation in Leishmania parasites. Exp Parasitol 116:83–87. 6. Urban C, Zychlinsky A (2007) Netting bacteria in sepsis. Nat Med 13:403–404. 24. Lippolis JD, Reinhardt TA, Goff JP, Horst RL (2006) Neutrophil extracellular trap 7. Fuchs TA, et al. (2007) Novel cell death program leads to neutrophil extracellular traps. J Cell Biol 176:231–241. formation by bovine neutrophils is not inhibited by milk. Vet Immunol Immunopathol 8. Mansueto P, et al. (2007) Immunopathology of leishmaniasis: An update. IntJImmu- 113:248–255. nopathol Pharmacol 20:435–445. 25. Gupta AK, Hasler P, Holzgreve W, Gebhardt S, Hahn S (2005) Induction of neutrophil 9. Lainson R, Shaw JJ (1987) Evolution, classification, and geographical distribution. extracellular DNA lattices by placental microparticles and IL-8 and their presence in Leishmaniasis in Biology and Medicine, eds Peters W, Killick-Kendrick R (Academic, preeclampsia. Hum Immunol 66:1146–1154. London), Vol 2, pp 1–120. 26. Gupta AK, Hasler P, Holzgreve W, Hahn S (2007). Neutrophil NETs: A novel contributor 10. Turco SJ, Descoteaux A (1992) Lipophosphoglycan of Leishmania parasites. Annu Rev to preeclampsia-associated placental hypoxia? Semin Immunopathol 29:163–167. Microbiol 46:65–94. 27. Baker VS, et al. (2008) Cytokine-associated neutrophil extracellular traps and antinu- 11. Laskay T, vanZandbergen G, Solbach W (2003) Neutrophil granulocytes Trojan horses clear antibodies in Plasmodium falciparum-infected children under 6 years of age. for Leishmania major and other intracellular microbes? Trends Microbiol 11:210–214. Malar J 7:41–52. 12. Reithinger R, et al. (2007) Cutaneous leishmaniasis. Lancet Infect Dis 7:581–596. 28. Ribeiro-Gomes FL, et al. (2007) Neutrophils activate macrophages for intracellular 13. Aga E, et al. (2002) Inhibition of the spontaneous apoptosis of neutrophil granulocytes killing of Leishmania major through recruitment of TLR4 by neutrophil elastase. by the intracellular parasite Leishmania major. J Immunol 169:898–905. J Immunol 179:3988–3994. 14. Silva-Filho FC, Saraiva EM, Vannier-Santos MA, Souza W (1990) The surface free energy 29. Ribeiro-Gomes FL, et al. (2004) Macrophage interactions with neutrophils regulate of Leishmania amazonensis. Cell Biophys 17:137–151. Leishmania major infection. J Immunol 172:4454–4462. 15. Turco SJ, Spa¨th GF, Beverley SM (2001) Is lipophosphoglycan a virulence factor? A 30. Wanderley JL, Moreira ME, Benjamin A, Bonomo AC, Barcinski MA (2006) Mimicry of surprising diversity between Leishmania species. Trends Parasitol 17:223–226. apoptotic cells by exposing phosphatidylserine participates in the establishment of 16. Dey R, et al. (2007) Functional paradox in host– interaction dictates the fate amastigotes of Leishmania (L) amazonensis in mammalian hosts. J Immunol 176:1834– of parasites. Future Microbiol 2:425–437. 17. VanStrijp JA, Russell DG, Tuomanen E, Brown EJ, Wright SD (1993) Ligand specificity of 1839. purified type three (CD11b/CD18, Mac-1). Indirect effects of an 31. Orlandi PA, Turco SJ (1987) Structure of the lipid moiety of the Leishmania donovani Arg-Gly-Asp sequence. J Immunol 151:3324–3336. lipophosphoglycan. J Biol Chem 262:10384–10391. 18. Tuon FF, et al. (2008) Toll-like receptors and leishmaniasis. Infect Immun 76:866–872. 32. Morgado FN, et al. (2008) Is the in situ inflammatory reaction an important tool to 19. Pearson RD, Steigbigel RT (1981) Phagocytosis and killing of the protozoan Leishmania understand the cellular immune response in American tegumentary leishmaniasis? Br J donovani by human PMN leukocytes. J Immunol 127:1438–1443. Dermatol 158:50–58.

6of6 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0900226106 Guimara˜es-Costa et al. Downloaded by guest on September 30, 2021