Infection with Leishmania amazonensis upregulates ... · Infection with Leishmania amazonensis...

Infection with Leishmania amazonensis upregulatespurinergic receptor expression and induces host-cellsusceptibility to UTP-mediated apoptosis

Camila Marques-da-Silva,1 Mariana M. Chaves,1

Suzana Passos Chaves,1,2

Vanessa Ribeiro Figliuolo,1

José Roberto Meyer-Fernandes,7

Suzana Corte-Real,3 Claudiana Lameu,4,5

Henning Ulrich,5 David M. Ojcius,6

Bartira Rossi-Bergmann2 andRobson Coutinho-Silva1*1Laboratory of Immunophysiology and 2Laboratory ofImunopharmacology, Biophysics Institute Carlos ChagasFilho, Federal University of Rio de Janeiro RJ,21941-902, Brazil.3Laboratory of Structural Biology of Oswaldo CruzInstitute – FIOCRUZ, Rio de Janeiro, 21045-900, Brazil.4Center for Applied Toxinology CAT-CEPID, InstitutoButantan, São Paulo, Brazil.5Department of Biochemistry, Chemistry Institute, SãoPaulo University, São Paulo, Brazil.6Health Sciences Research Institute and School ofNatural Sciences, University of California, Merced, CA95343, USA.7Department of Medical Biochemistry, Federal Universityof Rio de Janeiro, RJ, Brazil.

Summary

Nucleotides are released into the extracellularmilieu from infected cells and cells at inflammatorysites. The extracellular nucleotides bind to specificpurinergic (P2) receptors and thereby induce avariety of cellular responses including anti-parasitic effects. Here we investigated whetherextracellular nucleotides affect leishmanial infec-tion in macrophages, and found that UTP reducesstrongly the parasite load in peritoneal macroph-ages. Ultrastructural analysis of infected cellsrevealed that UTP induced morphological damagein the intracellular parasites. Uridine nucleotidesalso induced dose-dependent apoptosis of mac-rophages and production of ROI and RNI only in

infected macrophages. The intracellular calciummeasurements of infected cells showed that theresponse to UTP, but not UDP, increased the sen-sitivity and amplitude of cytosolic Ca2+ changes.Infection of macrophages with Leishmania upregu-lated the expression of P2Y2 and P2Y4 receptormRNA. The data suggest indirectly that Leishma-nia amazonensis infection induces modulation andheteromerization of P2Y receptors on macroph-ages. Thus UTP modulates the host responseagainst L. amazonensis infection. UTP and UTPhomologues should therefore be considered asnovel components of therapeutic strategiesagainst cutaneous leishmaniasis.

Introduction

Leishmania species are kinetoplastid protozoan para-sites that colonize invertebrates and vertebrates and areobligate intracellular parasites of macrophages in theirvertebrate hosts. Leishmania parasites have developeddifferent strategies to adapt and survive in their hosts(Dermine et al., 2005). Thus, Leishmania uses lypophos-phoglycan (LPG) to inhibit the ability of phagosomes tofuse with late endocytic organelles and lysosomes(reviewed Dermine et al., 2005), and it inhibits thehydrolytic activity of lysosomal enzymes, possiblythrough chelation of calcium and inhibition of proteinkinase C (Awasthi et al., 2004). Early studies suggestedthat Leishmania parasites also avoid triggering oxidativeburst by actively inhibiting PKC activation in macroph-ages (Olivier et al., 1992; Moore et al., 1993). Infectiousstages of Leishmania also actively and selectively inhibitIL-12 production, while leaving other pro-inflammatorycytokine response pathways relatively intact (Feng et al.,1999). Once in the macrophage host, Leishmania pro-mastigotes rapidly revert to the amastigote form, andsurvive and multiply inside the macrophages, eventuallyleading to the lysis of the host cells (Awasthi et al.,2004).cmi_1630 1410..1428

Nucleotides released after cell lysis, especially ATP, canmodulate infection by ligating purinergic receptors. Inmammals, there are two families of nucleotide receptors

Received 21 November, 2010; revised 4 April, 2011; accepted 25May, 2011. *For correspondence. E-mail [email protected]; Tel.(+55) 21 2562 6565; Fax (+55) 21 2280 8193.

Cellular Microbiology (2011) 13(9), 1410–1428 doi:10.1111/j.1462-5822.2011.01630.xFirst published online 11 July 2011

© 2011 Blackwell Publishing Ltd

cellular microbiology

(P2). The ligand-gated ion channels P2X receptors (withseven members cloned: P2X1–P2X7) and the G protein-coupled metabotropic P2Y receptors (with eight memberscloned: P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2Y12, P2Y13 andP2Y14) (Burnstock, 2007). ATP binds to the P2X7 receptor,a special member of P2X receptors responsible for induc-ing cell death of intracellular bacterial pathogens such asMycobacterium tuberculosis, Chlamydia trachomatis(Lammas et al., 1997; Coutinho-Silva et al., 2003; 2007)and Toxoplasma gondii (Correa et al., 2010; Jamiesonet al., 2010; Lees et al., 2010), and modulation of celldeath in mice infected with the protozoan parasite, Trypa-nosoma cruzi (Mantuano et al., 2003).

Extracellular ATP and other purine/pyrimidine nucle-otides mediate intracellular signalling via P2 receptorsand induce biological effects in various cell types. Thebroad tissue distribution of the adenine and uridinenucleotide-activated P2 receptors supports the idea thatendogenously released nucleotides act as importantextracellular signalling molecules (Brunschweiger andMuller, 2006; Elliott et al., 2009).

Uridine derivatives have received increasing attentionafter they were shown to activate G protein-coupled P2Ynucleotide receptors with equal or greater efficacy thanadenine nucleotides (Anderson and Parkinson, 1997).Uridine has protective effects during inflammation, damp-ening inflammation under different kinds of cellular stressor pathological conditions (Bours et al., 2006). However,the exact mechanisms by which uridine reduces or altersthe inflammatory response remain unclear (Evaldssonet al., 2007).

In 1997, Lazarowski demonstrated that mechanicalstimulation of cells results in non-lytic release of UTP,and affects autocrine stimulation of P2Y receptors(Lazarowski et al., 1997). The release of UTP is coupledto stimulation of heterologously expressed P2Y4 receptorsin a human astrocytoma cell line. Levels of UTP in tissuesmay increase to 300 nM, which is sufficient to activate P2receptors (Lazarowski and Harden, 1999), and increaseduridine nucleotide release has been detected from vascu-lar endothelial cells under certain pathological conditions(Seifert and Schultz, 1989). In addition, functional assayswith intact cells expressing recombinant P2Y receptorsthat are selectively activated by UTP, UDP and UDP-glucose indicated that these nucleotide species arereleased from resting or mechanically stimulated cells atconcentrations capable of stimulating their respectivereceptors (Lazarowski et al., 1997; 2003; Homolya et al.,2000).

We show here that despite the impressive arsenal ofweapons deployed by Leishmania amazonensis tosubvert host cells, macrophages still manage to eliminatethe infectious parasites by upregulating pyrimidinergicreceptors. This strategy may play a key role in maintaining

protection against long-term parasitism in mammalianhosts.

Results

Parasite load is diminished by uracyl treatment

Nucleotides are known to modulate phagocytosis,promote vesiculation and induce intracellular parasitedeath. Many reports have shown that extracellular ATPcoupling to the P2X7 receptor can diminish parasite sur-vival in macrophages (Fairbairn et al., 2001; Li et al.,2002; Lees et al., 2010; review in Coutinho-Silva et al.,2007). Since macrophages also express several P2receptors that are responsive to uracyl nucleotides, wedecided to characterize the effects of uracyl nucleotideson L. amazonensis-infected macrophages. To analyse theparasitic load, infected macrophages were treated withnucleotides, and parasite growth was quantified by opticalmicroscopy. As Fig. 1 shows, UTP reduced the parasiteload by 75%, compared with untreated cells (Fig. 1A).Compared with UTP, UDP was less effective at reducingthe parasitic load in macrophages. These results led us tohypothesize that UTP may inhibit infection by activatingP2Y receptors. We therefore incubated infected cellswith Suramin or PPADS (two known antagonists of P2receptors) before addition of UTP. Suramin reversed 65%of the anti-parasitic effect of UTP. Similarly, Pyridoxal-phosphate-6-azophenyl-2′,4′-disulfonic acid (PPADS)reversed the anti-parasitic effect by 48%, while PPADSalso inhibited the infection (Fig. 1A). These resultssuggest that UTP mediated its effects by activation of P2receptors.

The UTP-mediated effects on infection started within1 h after nucleotide treatment and became larger after 6 hof treatment (Fig. 1B). In addition, the UTP-mediatedeffect seemed to depend on the intracellular calciumconcentration, since the effects were completely blockedby treatment with the intracellular membrane-permeablecalcium chelator, 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA), but to a much lowerextent by the extracellular calcium chelator, EGTA(Fig. 1C).

UTP treatment results in damage of macrophages andintracellular parasites

Since UTP treatment reduced the parasitic load, wedecided to analyse more carefully the infected macroph-ages and amastigotes by electron microscopy(Fig. 2B–D). After UTP treatment, infected macrophageshad an altered shape, presenting few intact organelles(Fig. 2B) and electron-dense chromatin attached to thenuclear envelope of nuclei (Fig. 2B and C), compared with

UTP-inducing effects in infected macrophages 1411

© 2011 Blackwell Publishing Ltd, Cellular Microbiology, 13, 1410–1428

the structure of uninfected cells that were treated withUTP (Fig. 2A). It was also possible to observe dilatedendoplasmic reticulum (ER) and cytoplasm vacuolizationin infected macrophages that were treated with UTP(Fig. 2D). In addition, it was possible to find damagedparasites with large vacuoles and small electron-densevesicles in the cytoplasm (black arrows), which are notcharacteristic of this amastigote form (Fig. 2C and D). Wedid not observe alterations in parasite kinetoplasts. Wealso found free parasites in our samples, some of whichwere apparently damaged but also some which werehealthy (data not shown).

We frequently observed apoptotic/necrotic macroph-ages, although the majority of the cells did not containvisible parasites in the cytoplasm (not shown).

UTP and UDP induce intermediate reactive oxygen andnitrogen species only in infected cells

Macrophages are phagocytic cells that paradoxically arealso the preferential host cell of Leishmania (Walterset al., 1993). Activation of macrophages is a basicmechanism for elimination of the Leishmania parasites,and the most common pathway mediating parasite lysisor destruction by murine macrophages involves the pro-duction of nitric oxide (NO) from iNOS (Awasthi et al.,2004) and reactive oxygen species (ROS) (Kavoosiet al., 2006). We thus determined whether UTP andUDP stimulate production of reactive oxygen intermedi-ates (ROIs) and ROS by measuring NO, H2O2, O2

- andROS levels. Figure 3A shows NO production induced by10 mg of LPS co-stimulated with ATP, as previouslyreported (Hu et al., 1998). The nucleotides UTP andUDP did not induce any NO in uninfected macrophages(Fig. 3A). Conversely, UTP and UDP induced high levelsof NO production in L. amazonensis-infected macroph-ages (Fig. 3A).

We next measured H2O2 production in infected anduninfected cells (Fig. 3B). Both infected and uninfectedcells produce only a low level of peroxide (Fig. 3B).However, infected cells treated with UTP produced highlevels of peroxide, which was reversed by treatment withthe anti-oxidant, ascorbic acid (A.A.). We also measuredO2

- production in infected and uninfected cells (Fig. 3C).Infected macrophages treated with UTP produced higheramounts of O2

- when compared with uninfected UTP-treated cells (Fig. 3C), although the differences in produc-tion of the short-lived O2

- were not as dramatic as for H2O2

production.As an alternative way to measure total ROS production,

we used the DCFDA fluorescent probe (Fig. 3D). Asshown in Fig. 3D, ATP induced ROS production in unin-fected cells, in agreement with an earlier study (Cruzet al., 2007). In addition, UTP could induce ROS produc-

Fig. 1. UTP reduces parasite load of infected macrophages.Resident macrophages were infected with L. amazonensispromastigotes at an moi of 10:1 for 4 h, and the free parasiteswere then washed off. After 24 h post infection, cells were treatedwith agonists 100 mM UTP (A and B) and UDP (A) or control buffer,in the presence or absence of the antagonists, 100 mM Suraminand 300 mM PPADS (A). After 12 h the cells were fixed andcounted (A and C). In (B) the cells were fixed and counted afterdifferent times post UTP treatment. In (C) the cells were treatedwith 3 mM ATP, 100 mM UTP or control buffer in Ca2+-free solution.At least 250 cells were examined and results were expressed asthe infection index (II), which is obtained by the formula: II = (%infected macrophages) ¥ (amastigotes/infected macrophage)/100.Values represent mean � SEM of four separate experiments.***P < 0.0001, **P < 0.001, *P < 0.05.

1412 C. Marques-da-Silva et al.


tion only in L. amazonensis-infected macrophages. Wetherefore incubated infected cells with Suramin or PPADSbefore addition of UTP or ATP. Suramin and PPADS com-pletely blocked the ROS production induced by UTP, whileboth antagonists partially blocked ATP-induced ROS pro-duction (Fig. S1C). Taken together, these results revealedthat L. amazonensis infection renders infected macroph-ages sensitive to UTP-induced production of ROI andRNI.

L. amazonensis-infected macrophages undergoapoptosis induced by UTP and UDP

In order to survive, Leishmania amastigotes dependon mechanisms that prevent both parasites andinfected host cell from being killed. Extracellular ATP, atmillimolar concentrations, triggers apoptosis followingP2X7 ligation in different cell types such as dendriticcells (Coutinho-Silva et al., 1999) and macrophages(Oshimi et al., 1999). However, little is known about the

ability of UTP to induce cell death (Cavaliere et al.,2005).

Apoptosis was assessed by quantifying hypo-diploidnuclei after 8 h of exposure to nucleotides. As shown inFig. 4A, infected macrophages presented a slight level ofinhibition (7 � 2%) of spontaneous apoptosis, a mecha-nism of parasite survival that was previously reported (Lisiet al., 2005). Infected cells presented a higher level(49 � 6%) of apoptosis after ATP treatment, despite theability of the parasites to inhibit spontaneous apoptosis.Conversely, UTP and UDP did not induce apoptosis inuninfected macrophages, in agreement with previousreports that UTP does not induce apoptosis of uninfectedmacrophages (Coutinho-Silva et al., 2001). In contrast,UTP and UDP induced significant levels of specific apo-ptosis (20 � 1% and 10 � 2% respectively) in infectedmacrophages.

Apoptosis induced by the nucleotides is also dose-dependent. Figure 4B and C shows that uninfected cellsdid not undergo apoptosis at any of the UTP and UDP

Fig. 2. Ultrastructural analysis showing thatUTP induces morphological damage tointracellular Leishmania. Residentmacrophages were infected or mock-infectedwith L. amazonensis promastigotes at an moiof 10:1. After 4 h, free parasites wereremoved. After 24 h of infection, cells weretreated with 100 mm UTP or control buffer,and 8 h later were fixed and processed fortransmission electron microcopy.A. Uninfected macrophage.B–D. Infected and UTP-treated macrophages.N, nucleus; L, L. amazonensis; ER,endoplasmic reticulum; M, mitochondria; PV,parasitophorous vacuole; PL, processedLeishmania; MF, mieline figures; CD, cellulardebris. Black arrows refer to L. amazonensisvacuole, black arrowheads refer tomacrophage vacuoles and white arrowheadsrefer to alterations of the macrophage nucleusalterations. Bar corresponds to 3 mm.



concentrations tested (Fig. 4B and C). UTP (Fig. 4B) andUDP (Fig. 4C) treatment induced apoptosis in infectedmacrophages as a function of concentration, with UTPbeing more effective than UDP. The nucleotide concen-trations were also consistent with concentrations knownto activate different P2Y receptors: UTP at an EC50 of10 mM, and UDP at an EC50 of 20 mM (Del Rey et al.,2006).

We used three additional procedures to characterizethe type of apoptosis induced by the nucleotides: DNAfragmentation was determined by terminal deoxythymi-dine transferase-mediated dUTP nick end-labelling(TUNEL) DNA breaks, which were detected by a fluores-cence microscope; the externalization of phosphaty-dilserine (PS) (an early step of apoptosis) was measuredusing annexin V, which binds to PS; and nuclear con-densation was ascertained by labelling nuclear DNAwith Hoechst and visualizing condensing nuclei under afluorescence microscope.

As shown in Fig. 4D, panel B, ATP induced DNA frag-mentation in infected macrophages, as expected, but sodid UTP (Fig. 4D, panel D). L. amazonensis-infecteduntreated cells and uninfected UTP-treated cells did notshow any DNA fragmentation (Fig. 4D, panels A and C),confirming the results in Fig. 4A.

Using flow cytometry, we evaluated the % of annexinV-positive cells after ATP and UTP treatment in infectedand uninfected cells (Fig. 4E). We observed that theLeishmania-infected cultures were more susceptible toATP-induced apoptosis than uninfected cells, with the per-centage of apoptotic cells rising from about 40% in unin-fected cells to 70% in infected macrophages. UTP also

Fig. 4. UTP and UDP induce cell death of macrophages. Resident macrophages were infected or mock-infected with L. amazonensispromastigotes at an moi of 10:1. After 4 h, free parasites were removed. After 24 h of infection, cells were treated with control buffer, 100 mmUTP, 100 mm UDP, 3 mM ATP (A), increasing concentrations of UTP (B) or UDP (C). After 8 h (hypodiploid cells quantification) and after 6 h(annexin-V staining), cells were incubated with buffer containing ethidium bromide (50 mg ml-1) or AnnexinV-FITC PI binding buffer (E) andanalysed by flow cytometry (A, B, C and E). Data from 10 000 cells were collected on a FACScan and analysed using Cell Quest 3.3software. TUNEL-labelled DNA breaks (D) were analysed in infected macrophages (panel B, D) and uninfected macrophages (panel A, C) thatwere treated with 3 mM ATP (panel B), 100 mm UTP (panel C, D) or control buffer (panels A). Values A–C represent the mean � SEM of fiveindependent experiments performed in triplicate. ***P < 0.0001; **P < 0.001; *P < 0.05.

Fig. 3. Possible role for oxidative burst in UTP-induced damage ofL. amazonensis amastigotes. Resident macrophages were infectedor mock-infected with L. amazonensis promastigotes at an moi of10:1. After 4 h, free parasites were removed. After 24 h of infection,cells were treated with control buffer, 100 mM UTP, 100 mM UDP,3 mM ATP or 3 mM ATP + 1 mg ml-1 LPS. After 8 h, nitric oxideproduction was analysed (A). Alternatively, after 1 h, H2O2 (B), O2

-

(C) and ROS (D) were measured. The colorimetric assay was usedwith Griess reagents (A), phenol red (B), nitroblue tetrazolium (C)and HCFDA fluorescence (D). Infected cells were treated withcontrol buffer, 100 mM UTP or 3 mM ATP, and incubated with buffer,10 mM ascorbic acid (A.A.) or 10 mM N-acetyl-l-cysteine (NAC).Cells were uninfected (white bars) or infected (black bars) with L.amazonensis. Data are expressed as the mean � SEM of four (A)and three (B and C) independent experiments. ***P < 0.0001,*P < 0.05.



induced apoptosis (stimulated externalization of PS) butonly in infected cultures, where about 60% of the cellswere positive for Annexin-V under these experimentalconditions (Fig. 4E).

Finally, Hoechst nuclear labelling showed that ATP, UTPand UDP treatment altered the shape of nuclei in infectedcultures, which assumed the characteristic form of apop-totic nuclei; while uninfected cells treated with the nucle-otides displayed normal nuclei, except in the case of ATPtreatment, which is a known inducer of apoptosis (data notshown).

UTP-induced apoptosis in infected cells iscaspase-dependent

Using a pan-caspase inhibitor (Z-VAD), we observed, byhypodipoid nuclei quantification, that apoptosis induced

by ATP and UTP in infected macrophages is mediatedby caspases (Fig. 5A). In particular, we observedactivated caspase-3 during treatment with both nucle-otides (Fig. 5B). Conversely, caspase activation doesnot seem to be involved with UTP-induced amastigoteelimination, since pre-treatment with Z-VAD did notblock the UTP-induced reduction of the parasite load(Fig. 5C).

L. amazonensis-infected cells undergo apoptosis but notnecrosis induced by UTP and UDP

Extracellular ATP can induce both apoptosis and necrosisin macrophages through activation of the P2X7 receptor(Adinolfi et al., 2005). Since there are no reports describ-ing UTP- or UDP-mediated cell death in macrophages,

Fig. 5. UTP-induced apoptosis but not anti-amastigote activity in infected cells require caspase activation. Peritoneal resident macrophageswere infected with L. amazonensis promastigotes at an moi of 10:1. After 4 h of infection, free parasites were removed by washing. After 24 hof infection, cells were pre-treated with 1 mg ml-1 Z-VAD for 30 min, then treated with control buffer, 100 mM UTP or 3 mM ATP (A and B). After8 h, apoptosis was quantified by cytofluorimetry through hypodiploid nuclei (A) or caspase-3 activation (B). In (C) is shown the quantification ofparasite load. In (D) the supernatant was incubated with LDH kit. LDH release was measured by the colorimetric assay in the supernatantof the different experimental conditions. Data from 10 000 cells were collected on a FACScan and analysed using Cell Quest 3.3 software(A and B), and (C) at least 250 cells were examined and results were expressed as the infection index (II), which is obtained by the formula:II = (% infected macrophages) ¥ (amastigotes/infected macrophage)/100. Values represent mean � SEM of four independent experimentsperformed in triplicate. ***P < 0.0001; **P < 0.001; *P < 0.05.



and our results revealed cell death induced by thesenucleotides only in infected cells, we investigated if UTPand UDP also induce necrosis in infected macrophages.Thus, we performed a lactate dehydrogenase (LDH)release assay after nucleotide treatment of infected cells.As shown in Fig. 5D, ATP triggered LDH release from bothinfected and uninfected cells, as expected; but treatmentwith neither UTP nor UDP induced necrosis. Hence, UTPand UDP stimulate apoptosis of infected macrophages,with little or no contribution from necrosis.

Antagonists of uracyl activated receptors inhibitapoptosis induced by UTP and UDP inL. amazonensis-infected macrophages

Since P2Y receptor antagonists can reverse UTP-dependent inhibition of parasite infection, we verifiedwhether suramin and PPADS can also inhibit apoptosisinduced by UTP and UDP. We found that suramin inhib-ited apoptosis induced by both UDP and UTP in infectedcells (Fig. 6A and B). However, PPADS had no effect onUTP-induced apoptosis (Fig. 6B). UTP-induced apopto-sis started after 5 h of nucleotide treatment and wasmaximal after 6 h (Fig. 6C). However, the pharmacologysuggested that more than one uracyl-activated receptorwas involved. Thus we used two more stable activatorsof P2Y2 and P2Y4 (UTPgammaS) and P2Y6 (UDPbetaS),which have been reported to discriminate between P2receptors. The results show that both agonists canequally reduce apoptosis in infected macrophages(Fig. S1A).

In addition, we verified the dependence of extracellularcalcium for UTP-induced apoptosis. We observed thatspecific apoptosis induced by UTP was completelyblocked when the UTP incubation was done in Ca2+-freesaline buffer in the presence of 1 mM EGTA (Fig. 6D).

Calcium-mediated P2Y signalling

Nucleotides activating P2Y and P2X receptors are knownto promote transient calcium concentration changes inmacrophages (Coutinho-Silva et al., 2005; Del Rey et al.,2006). We therefore analysed whether calcium signallingwas altered after infection with L. amazonensis. We per-formed Ca2+ measurements by loading infected and unin-fected cells with the calcium-sensitive fluorescent dye,FURA-2 AM. As shown in Fig. 7, uninfected macrophagesexhibit a transient Ca2+ response to ATP, but not to sub-sequent treatment with UTP when applied sequentially(Fig. 7A). However, infected macrophages responded toUTP with a transient increase in intracellular Ca2+ concen-tration, when UTP was applied after ATP (Fig. 7B). More-over, when the sequence of nucleotides was reversed,treating first with UTP and then ATP, both infected and

uninfected cells responded with transient Ca2+ peaks(Fig. 7C and D). These results reinforce our interpretationthat there is positive modulation of UTP-activated recep-tors during L. amazonensis infection.

In order to analyse the mechanisms underlying calciumsignalling, we determined whether infection with L. ama-zonensis promotes higher sensitivity to nucleotides withregards to modulation of intracellular calcium concentra-tions. Figure 8 shows that infection promotes higher sen-sitivity to UTP (Fig. 8B), but not to UDP (Fig. 8C) nor ATP(Fig. 8A). We observed that the Ca2+ dose–responsecurves for ATP in infected macrophages were similar touninfected ones, with an EC50 of 1 mM (Fig. 8A), althoughthe curve inclination was slightly changed.

In response to UTP, we observed that the infectionincreased the sensitivity 10-fold, with the EC50 being0.7 mM and 7 mM for infected and uninfected cells respec-tively (Fig. 8B). The intracellular Ca2+ dose–responsescurves for UDP were equivalent for infected and unin-fected cells, displaying an EC50 of 4 mM in both cases(Fig. 8C). In addition, when we compared the amplitude ofUTP- and UDP-induced calcium changes in uninfectedand infected macrophages at saturating nucleotide con-centrations, we observed that the infection modulatesUTP and UDP calcium responses in opposite ways. Whileintracellular Ca2+ responses in response to UTP werehigher in infected cells (Fig. 8D, left graphic), the Ca2+

responses in response to UDP were lower (Fig. 8D, rightgraphic). Collectively, our data obtained from calciummeasurements provide evidence that the infection with L.amazonensis modulates positively UTP-activated recep-tors in macrophages.

Regulation of P2Y2 and P2Y4 gene and proteinexpression in infected macrophages

Since we observed several functions activated by UTP inthe infected macrophages, we hypothesized that amongP2Y receptors, the P2Y2, P2Y4 and P2Y6 receptors maybe modulated during L. amazonensis infection. We there-fore analysed gene expression of P2Y2, P2Y4 and P2Y6

receptors in macrophages. Real-time PCR revealed thatP2Y2 and P2Y4 receptor mRNA transcription levelsincreased 11.7 � 0.7- and 1.4 � 0.1-fold, respectively, ininfected macrophages, compared with gene expressionlevels in control uninfected macrophages (Fig. 9A). P2Y6

receptor expression was not detected in macrophages(not shown). Western blot analysis showed a strongerband compatible with the expected for P2Y4 and P2Y2

receptors in extracts from infected macrophages whencompared with uninfected ones (Fig. 9B). The opticaldensity analysis of bands showed a fivefold and twofoldenhancement of P2Y2 and P2Y4 expression respectively(Fig. 9C).



Fig. 6. Apoptotic response to UTP of infectedcells is time- and calcium-dependent andinhibited by P2Y antagonists. Residentmacrophages were infected or mock-infectedwith L. amazonensis promastigotes at an moiof 10:1. After 4 h, free parasites wereremoved. After 24 h of infection, cells weretreated with control buffer, 100 mM Suramin(A and B), 300 mM PPADS (B), or 1 mMEGTA (D) and/or 100 mM UTP, 100 mM UDP.After 8 h, cells were incubated with buffercontaining ethidium bromide (50 mg ml-1) andanalysed by flow cytometry. At different timesafter UTP treatment, cells were incubated withbuffer containing ethidium bromide(50 mg ml-1) and analysed by flow cytometry.Data are expressed as the mean � SEM offour independent experiments. ***P < 0.0001,**P < 0.001, *P < 0.05.



Discussion

A growing number of studies have demonstrated theimportance of extracellular nucleotide signalling via P2receptors as an important component of the inflammatoryresponse to infection (Fairbairn et al., 2001; Li et al.,2002; Coutinho-Silva et al., 2007; Yilmaz et al., 2008;Lees et al., 2010). In response to the anti-parasitic activityof the nucleotides, some parasites express on their outersurface enzymes that degrade or synthesize nucleotides,in order to facilitate establishment of the infection. Onceinside the host cell, the parasites evade the humoralresponse and attempt to inhibit cellular microbicidalmechanisms such as the oxidative burst, lysosome fusionand host-cell apoptosis (Luder et al., 2001; Coutinho-Silva et al., 2007; Yilmaz et al., 2008).

In the present study, we confirm previous observationsthat macrophages infected with L. amazonensis undergoapoptosis when exposed to ATP following activation ofP2X7 receptors (Chaves et al., 2009). Our current obser-vation that uracyl nucleotides induce apoptosis in mac-rophages from P2X7

-/- mice demonstrate that theapoptosis is independent of P2X7 receptors (Fig. S2). Fur-thermore, a small but reproducible decrease of spontane-ous apoptosis of macrophages from P2X7-deficient micesuggests that spontaneous apoptosis was possibly due torelease of ATP from the same cells or adjacent cells. Weextended those observations and showed that uracylnucleotides can also reduce the parasitic load in infectedcells. Our results suggest that there may be positivemodulation of a UTP-responsive P2Y receptor inLeishmania-infected macrophages. Recently, Osorio yFortéa et al. (2009) have described the genes that aremodulated during L. amazonensis infection of macro-phages. They found that expression of over 1000 geneswas modulated, including the P2X7, P2Y2, P2Y6, P2Y12

and P2Y13 receptors (Fortea et al., 2009).While P2X receptors recruit calcium from the extracel-

lular medium, P2Y receptors mobilize calcium from intra-cellular storage organelles. Our data showing that UTP-induced anti-Leishmania effects were dependent onintracellular calcium mobilization reinforce the hypothesisthat UTP mediates its effects by activation of P2Y recep-tors rather than P2X7 receptors.

There is growing evidence that, by altering the intracel-lular calcium concentration, metabotropic P2Y receptorsmight participate in both growth inhibition and programmedcell death (Fang et al., 1992; Chvatchko et al., 1996;Hopfner et al., 2001; Maaser et al., 2002). P2Y receptorshave been reported to be involved in cell death, sinceSH-SY5Y cells undergo apoptosis mediated by the P2Y4

receptor after exposure to UTP (Cavaliere et al., 2005).Leishmania enhances its own survival by inhibiting

host-cell apoptosis (Ouaissi, 2003; Lisi et al., 2005). We

Fig. 7. UTP modulates Ca2+ changes through P2Y receptors ininfected cells. Resident macrophages were infected ormock-infected with L. amazonensis promastigotes at an moi of 10:1on glass coverslips. After 4 h, free parasites were removed bywashing. After 48 h, cells were incubated with FURA2-AM for40 min, washed and mounted in the perfusion chamber. Cellsreceived in bolus 100 mM ATP, were washed for 200 s, and then100 mM UTP was applied on infected (B) or mock-infected (A) cells.Cells received in bolus 100 mM UTP, were washed for 200 s, andthen 100 mM ATP was applied on infected (D) or uninfected (C)cells. Representative data of a single experiment are shown.Experiments were repeated at least five times with similar results.Bar corresponds to 60 s.



demonstrated that UTP could induce host cell apoptosisthrough caspase-3 activation, thus interfering with Leish-mania’s strategy for survival. Leishmania had been shownto block apoptosis through activation of PI3K/Akt signal-ling (Ruhland et al., 2007), inhibition of effector caspasesand repression of mitochondrial outer membrane perme-abilization (Akarid et al., 2004). Along these lines, UTP-activated P2Y receptors are coupled to phospholipase Cand an increase in diacylglycerol and IP3, followed by anincrease in intracellular calcium release (Brunschweiger

and Muller, 2006). Otherwise our data suggest that ifUTP-induced apoptosis is involved in parasite control invivo, its contribution should be minimal, since the treat-ment with a pan-caspase inhibitor did not revert the leish-manicidal effect of UTP.

Our micrographs revealed that UTP induced somedamage in L. amazonensis even before the host cellsdied. Production of ROS in macrophages is known tomediate parasite killing, and in response, some protozoanparasites such as T. gondii express antioxidant enzymes

Fig. 8. UTP- and UDP-induced Ca2+ mobilization is concentration-dependent. Resident macrophages were infected or mock-infected with L.amazonensis promastigotes at an moi of 10:1 on glass coverslips. After 4 h, free parasites were removed. After 48 h, cells were incubatedwith FURA2-AM for 40 min, washed and mounted in the perfusion chamber. Cells received in bolus increasing concentrations of ATP, UTPand UDP, and calcium dose–response curves were analysed.A. The ATP dose–response curves with EC50 = 1 mM for both infected and uninfected cells.B. The UTP dose–response curves with EC50 = 0.7 mM and 7 mM for infected and uninfected cells.C. The UDP dose–response curves with EC50 = 4 mM for infected and uninfected cells.D. The left graph shows the normalized intracellular calcium amplitude at 10 mM UTP, and the right graph shows normalized intracellularcalcium amplitude at 100 mM UDP. Concentration–response curves were generated by combining data from at least three independentexperiments. ***P < 0.0001.

Fig. 9. P2Y receptor expression levels in uninfected and infected macrophages. Gene expression levels were determined by real-time PCR(A) and protein expression level by Western blot (B and C) as described in Experimental procedures. ANOVA statistical analysis usingBonferroni post-tests (P2Y2, ***P < 00.1; and P2Y4, **P < 0.05) and unpaired t-tests demonstrated significant differences (‘Unpaired t-test’***P < 0.0001 and **P < 0.001).



(Shrestha et al., 2006). Thus UTP-induced parasitedamage and the decrease in the extent of infection mightbe due to ROS and RNI activation. According to ourresults, UTP induced H2O2 and O2

- production, while UTPand UDP induced NO production only in infected cells.The ability of the nucleotides, especially UTP, to modulateiNOS expression reflects a potentially novel function ofnucleotides in host defence, inflammation and cytotoxicresponses (Chen et al., 1998; Chen and Lin, 2000). ROSgenerated locally by cellular metabolism can also modu-late the sensitivity of P2Y receptors in osteoblasts(D’Andrea et al., 2008).

L. amazonensis also interferes with calcium signalling inresponse to UTP, promoting additional responses tonucleotides.

Based on the patterns of intracellular Ca2+ responses tothe P2Y agonists (Marriott et al., 1999) surveyed, treesubtypes (P2Y2, P2Y4 and P2Y6) could be consideredcandidates for mediating uracyl nucleotide effects duringL. amazonensis infection. But none of the P2Y receptorsubtypes can fully account for the present results. Sincemacrophages from P2X7-/- mice do not respond to ATP(Fig. S2), and P2Y2 receptors are equally sensitive forATP and UTP, we can exclude the P2Y2 receptor as themain receptor responsible for the nucleotide effects ininfected cells, at least in terms of apoptosis effects. Thepresence of P2Y6 receptors could account only partiallyfor the present results, since these receptors are moresensitive to UDP than UTP and insensitive to ATP in mice,and our results have shown that UTP is the main agonistwith anti-leishmanicidal activity. Moreover, we could notdetect P2Y6 modulation of mRNA levels during infection,even though P2Y2 and P2Y4 were upregulated. The down-modulation of P2Y6 and upregulation of P2Y2 geneexpression during Leishmania infection have beendescribed previously (Fortea et al., 2009). P2Y4 receptorscould also be considered as a candidate, but the experi-ments using P2 antagonists do not fully support theinvolvement of P2Y4.

An alternative to consider is that the P2Y receptorscould function as heteromers. P2Y receptors in fact canform heteromers with members from the subfamily ofP2Y receptors. All P2Y receptors possess specificdomains involved in oligomerization, and functional het-erogeneity in P2 responses to the same ligand can beexplained by assuming interactions between differentP2Y receptors (D’Ambrosi et al., 2007). Our datashowing that protein expression of both P2Y4 and P2Y2

receptor is upregulated during L. amazonensis infectionreinforce this hypothesis. We therefore propose that apossible heteromeric association between P2Y2 andP2Y4 could take place during Leishmania infection and

be responsible for the new functions described for nucle-otides in infected macrophages.

Clifford and co-workers have proposed that the use ofuridine nucleotides, compared with adenosine nucle-otides, may be an advantage during inflammation (Cliffordet al., 1997). The advantage is due to the fact that uridinenucleotides are not degraded into products such as AMPand adenosine, both of which may inhibit inflammationthrough ligation of adenosine (A2) receptors.

Chemotherapeutic cure of leishmaniasis would requirean effective immune response that activates macroph-ages to produce toxic nitrogen and oxygen intermediatesto kill the amastigotes, and we showed here that UTP caninduce production of both inflammatory mediators. Leish-maniasis its one of the world’s most neglected diseasesand has received little attention from pharmaceutical com-panies, since its treatment is expensive and aggressive.The main currently known anti-leishmanial drugs for usein patients have strong side-effects including being painfuland requiring long-term administration. The challenge isto identify parasite and host molecules that modulatemacrophage function, with the aim of developing newantiparasite drugs and vaccines. We here propose thatUTP should be considered as a possible component ofnew therapies for treatment of cutaneous leishmaniases.

Experimental procedures

Parasites and macrophages

Leishmania amazonensis promastigotes (MHOM/BR/Josefa)were grown in 199 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Gibco BRL) and 5%haemin at 24°C. The parasites were used in the late stationaryphase.

The resident macrophages were obtained from the peritoneumof BALB/c, C57BL6 or P2X7-receptor knockout (P2X7

-/-) miceand allowed to adhere directly onto 24-well plastic dishes(Falcon, Becton Dickinson Labware) or on glass microscopecoverslips at a cell density of 5 ¥ 105 cells per well. After 1 h, cellswere washed gently twice with PBS to remove non-adherentcells. Adherent cells were cultured in cell culture medium (DMEMsupplemented with 10% FBS and 100 units penicillin/streptomycin) at 37°C under a 5% CO2 atmosphere.

Macrophages were infected with late-stationary-phase L. ama-zonensis promastigotes at a ratio of 10:1 (parasite : macroph-age). After 4 h, free parasites were washed off macrophagemonolayers with PBS and cells were cultured in cell culturemedium for 24 h.

Treatment with nucleotides

Treatment of infected or uninfected cells was carried out after aminimum of 24 h of infection with different concentrations ofnucleotides for 30–40 min in PBS at 37°C under a 5% CO2

atmosphere. Supernatants were then removed and replaced with



complete medium for different periods of time indicated in eachexperiment at 37°C under a 5% CO2 atmosphere.

ATP, UTP and UDP, and the inhibitors PPADS and Suramin,were purchased from Sigma (St. Louis, MO). UDPbetaS andUTPgammaS were purchased from TOCRIS (Missouri, USA).The reagents were prepared as stock solutions of 100 mM inPBS (phosphate buffer solution) and stored at -20°C until use.

Anti-amastigote activity

Macrophages were obtained from the peritoneal cavity ofBALB/c mice, and allowed to adhere for 1 h to a 24-well cultureplate at 2 ¥ 106 cells per well. The non-adherent cells wereremoved and after 24 h, the macrophage monolayer wasinfected with 2 ¥ 107 parasites for 4 h at 34°C/4% CO2. Afterremoval of free parasites by washing, the cultures were allowedto rest for 24 h at 37°C/4% CO2, after which they were treatedwith the agonists, UTP or UDP (100 mM), in PBS for 30–40 min;or antagonists, Suramin (100 mM) or PPADS (300 mM), in PBSfor 1 h prior to treatment with agonists. In some experiments,the cells were kept in PBS containing 1 mM EGTA solution for20 min, then the P2Y agonists were added to the same solu-tion. Alternatively, cells were pre-incubated with 1 mg ml-1

Z-VAD (BioVision) for 30 min followed by UTP treatment. Duringall treatments, the cells were incubated at 37°C/5% CO2. Thecells were then washed with PBS and allowed to rest for theindicated periods of time, fixed with 4% paraformaldehyde for10 min, and stained with panotic kit (Laborclin) according to themanufacturer’s instructions. For each coverslip, about 250 cellswere examined at 400 ¥ magnification to count the number ofinfected macrophages and the average number of parasites permacrophage. Individual amastigotes were clearly visible in thecytoplasm of infected macrophages. The results wereexpressed as the infection index (II), which is obtained by theformula: II = (% infected macrophages) ¥ (amastigotes/infectedmacrophage)/100.

Measurement of apoptosis

Measurement of hypo-diploid DNA. Cells with hypo-diploid DNAwere defined as apoptotic. Briefly, uninfected and L.amazonensis-infected macrophages were incubated with 3 mMATP, 100 mM UTP or UDP in PBS for 30 min at 37°C in 5% CO2,and the buffer was then removed and replaced by cell culturemedium. In some experiments, the cells were kept in PBS con-taining 1 mM EGTA for 20 min, and the agonists were then addedin the same solution. In some experiments, cells were pre-incubated with 1 mg ml-1 Z-VAD (BioVision) for 30 min followedby the nucleotide treatment. The cells were kept for 8 h at 37°Cin 5% CO2, and then were detached by scraping at 4°C, washedtwice with PBS, collected and spun for 10 min (200 g), and trans-ferred to a solution containing 50 mg ml-1 ethidium bromide,0.01 g ml-1 sodium citrate and 0.1% Triton X-100. The sampleswere transferred into 12 ¥ 75 mm FALCON 2052 FACS tubes(Becton Dickinson, San Jose, CA). Data from 10 000 cells werecollected on a FACScan flow cytometer (Becton Dickinson) withan argon laser tuned to 488 nm, and analysed using Cell Quest3.3. The events acquired for analysis were gated to eliminate cellaggregates using software WinMDI (Multiple Document InterfaceFlow Cytometry Application 2.8).

Detection of phosphatidylserine externalization on the cellsurface. To detect the early membrane changes of apoptosis(PS externalization on the outside of the cell membrane), weused an AnnexinV-FITC apoptosis staining kit according to themanufacturer’s instructions (R&D Systems, Minneapolis, MN).Uninfected macrophages or 24 h infected macrophages weretreated with 100 mM UTP or 3 mM ATP for 30 min in PBS, thenwashed and allowed to rest in complete medium for 6 h. Cellswere detached from 24-well plates and washed with cold PBSand incubated with binding buffer containing 0.25 mg ml-1

AnnexinV-FITC and 5 mg ml-1 propidium iodide (PI) for 15 min atroom temperature. At least 10 000 cells were analysed by flowcytometry, and apoptotic cells were defined as FITC-positive andPI-negative cells.

TUNEL assay. Macrophages (1 ¥ 105 cells) were plated in roundglass coverslips and infected with 1 ¥ 106 parasites for 4 h at34°C/4% CO2. After removal of free parasites by washing, thecultures were allowed to rest for 24 h at 37°C/4% CO2. Cells werethen treated or not with 100 mM UTP and 3 mM ATP for 30 min inPBS, then washed and were fixed 8 h later with 4% paraformal-dehyde and permeabilized with 1% saponin. TUNEL staining wasperformed using an in situ cell death detection kit (Boehringer-Mannheim; Meylan, France) according to the manufacturer’s rec-ommendations. Fixed cells were labelled with FITC-dUTP usingterminal deoxythymidine transferase at 37°C for 60 min. Cover-slips were washed three times with PBS. FITC-stained cells wereanalysed and images acquired using Image Pro Plus (Olympus)software in a Zeiss Axioplan Inverted microscope equipped withrhodamine (Zeiss BP 546/FT 580/LP 590) and fluorescein (ZeissBP 450-490/FT 510/LP 520) filters.

Caspase-3 activation. To measure caspase-3 activation, weused the CaspGLOW Fluorescein Active Caspase-3 staining kitaccording to the manufacturer’s instruction (BioVision MountainView, CA). Briefly, after 24 h of infection, macrophages werepre-treated or not with Z-VAD for 30 min, then treated with100 mM UTP or 3 mM ATP for 30 min in PBS, then washed andallowed to rest in complete medium for 8 h. Cells were detachedfrom 24-well plates and washed with cold PBS and incubatedwith FITC-DEVD-FMK and incubated for 30 min at 37°C incuba-tor with 5% CO2. At least 10 000 cells were analysed by flowcytometry using the FL1 filter.

Electron microscopy

Uninfected and L. amazonensis-infected cells were incubatedwith 100 mM UTP for 30 min and 8 h later were fixed in 2.5%glutaraldehyde in 0.1 M sodium cacodylate buffer containing3% sucrose and 3 mM CaCl2, pH 7.4, for 60 min at 4°C. The cellswere then washed three times in the same buffer, scraped off,transferred to an Eppendorf tube and pelleted by centrifugation.The pellets were post-fixed in 1% osmium tetroxide in 0.1 Msodium cacodylate buffer with 1.6% potassium ferrocyanate for1 h at 4°C, dehydrated sequentially in acetone and embedded inPolyBed 812. Ultrathin sections (90 nm) were cut using a Reichertultramicrotome OmU3, collected on copper grids, stained withaqueous uranyl acetate followed by lead citrate, and examined inan EM10C Zeiss transmission electron microscope. Electronmicrographs were digitalized employing an 8-bit grey scale at aresolution of 5.5 nm per pixel in an EPSON series Scanjet 3970.



Intracellular calcium measurements

Infected cells were grown on glass coverslips for 48 h. The cellswere then loaded with 2.5 mM Fura-2-AM (Molecular Probes) for40 min at room temperature in culture medium. The coverslipswere washed in PBS and mounted in a three-compartmentsuperfusion chamber attached to the stage of a NIKONDIAPHOT 300 TMD inverted microscope whose base wasformed by a coverslip containing the cells. The central chambercontaining the cells had a volume of 200 ml, and was perfusedwith Ca2+-containing saline (PBS supplemented with 1 mMCaCl2) at room temperature at a rate of 1 ml min-1. The intracel-lular calcium concentration of groups of 15–30 cells was moni-tored continuously with the use of a fluorescence photometer(Photon Technology; Princeton, NJ). Fura-2 was excited alter-nately at 340 and 380 nm, and the emission at 510 nm wasmeasured. The ratio measurement, which is proportional to theintracellular calcium concentration, was determined every100 ms. The nucleotides were injected in bolus, allowing thedrugs to persist in the presence of the cells for 1 min, at roomtemperature.

Lactate dehydrogenase assay (LDH)

To evaluate the sensitivity of infected or uninfected peritonealmacrophages to lysis by ATP, UTP and UDP, cells were infectedas described above. After 48 h, cells were treated with 100 mMUTP or UDP or 3 mM ATP for 8 and 24 h in DMEM plus 5% ofHIFCS at 37°C/5% CO2. The supernatant was collected, and theactivity of the enzyme LDH was dosed by a colorimetric assay ina spectrophotometer at 510 nm in a 96-well microplate. Resultswere expressed as optical density (OD).

Production of NO

The supernatant obtained from treated or infected cells wasutilized to evaluate NO production, by measuring the nitrite con-centration by a colorimetric assay in a spectrophotometer, usingthe Griess reaction. Briefly, 100 ml of the culture supernatant wasreacted with 100 ml of Griess reagent (1% sulfanilamide, 0.1%naphthylethylene diamine dihydrocloride and 2.5% H3PO4) for10 min at room temperature. The absorbance was measured at570 nm. The nitrite concentration was calculated using a stan-dard curve of sodium nitrite. Results were expressed as theconcentration of nitrite in mM. All tests were performed at least intriplicate.

Generation of H2O2

The procedure was the same as described by Pick and Mizel(1981), in which the ability of haem peroxidase to catalyse hydro-gen peroxide is dependent on dimerization of substituted phe-nolic compounds. This assay is based on the horseradishperoxidase (HRPO)-dependent conversion of phenol red into acoloured compound by H2O2.

Macrophages were plated and infected or mock-infected in96-well plates, as described. The cells were then treated with100 mM UTP, 100 mM UDP and 3 mM ATP and/or 10 mM ascorbicacid for 30 min. The cells were then incubated with phenol redsolution (140 mM NaCl, 10 mM Potassium phosphate buffer,

5 mM glucose, 0.28 mM phenol red, 8.5 UI ml-1 HRPO) for 1 h at37°C. The H2O2 was measured in a spectrophotometer at 630 nm(Power Wave XS, Biotek).

Superoxide production

Superoxide production was estimated by the nitroblue tetrazo-lium (NBT) reduction assay. Macrophages were plated andinfected or mock-infected as described, in 96-well plates. Thecells were then treated with 100 mM UTP or 3 mM ATP. Macroph-ages suspended in 0.45 ml of PBS were incubated for 1 h at37°C to which were added 0.03 ml of phorbol myristyl acetate(PMA, 5 mM final concentration) and 0.1% NBT. The reaction wasstopped by adding 0.45 ml of acetic acid. The mixture was thencentrifuged for 30 s at 1000 g. Reduction of NBT results in theformation of blue formazan which was detected spectrophoto-metrically (560 nm). The results are expressed as OD (arbitraryscale).

ROS measurement

The dye, 2,7-dichlorodihydrofluorescein diacetate (DCFH2DA), isone of the most widely used probes for direct measurement of theredox state of a cell (Soh, 2006). ROS production can be mea-sured as an increase in fluorescence at 530 nm when the sampleis excited at 485 nm. Peritoneal macrophages were cultured andinfected or mock-infected as mentioned previously. After 30 h ofinfection, cells were pre-incubated or not with 10 mM N-acetyl-L-cysteine (NAC) (antioxidant) or control buffer for 1 h. The plateswere then incubated with DCFH2DA (20 mg ml-1) for 30 min in37°C, 5% CO2, in a humidified incubator to allow loading of thedye. After 30 min, the cells were exposed to nucleotides for10 min at 37°C and the conversion of non-fluorescent DCFH2DAinto the highly fluorescent DCF was measured using a flowcytometer (FACScan, Becton Dickinson, USA).

Real-time PCR

Total RNA was extracted from 24 h infected or uninfected mac-rophages using Trizol™ (Life Technologies, Gaithersburg, MD).Every sample was further treated with Amplification GradeDNase I (Sigma Chemical, Poole, UK). Reverse transcription forcDNA synthesis was carried out with a thermal cycler (AppliedBiosystems, Foster City, CA) using the SuperScript III First-Strand synthesis system according to the manufacturer’s proto-col (Invitrogen, Carlsbad, CA). The expression levels of selectedmRNAs were measured by real-time PCR using the ABI StepOne Plus Instrument (Applied Biosystems, Foster City, CA). ThePCR was performed in 25 ml of buffer containing 1 ml of cDNA,SYBR Green Master Mix (Applied Biosystems, Foster City, CA)and 5 pmol of sequence-specific primers for P2Y2, P2Y4 andP2Y6 coding sequences (Lugo-Garcia et al., 2007) (Table 1).Thermal cycling conditions consisted of a pre-incubation step for2 min at 50°C, then denaturation for 10 min at 95°C followed by40 cycles for denaturation for 15 s at 95°C and annealing/extension for 1 min at 60°C. Gene expression changes ininfected macrophages were normalized with glyceraldehyde3-phosphate dehydrogenase gene expression as internal control.The data were compared with gene expression levels of unin-fected macrophages. The comparative 2-DDCT method was used



for relative quantification of gene expression as described previ-ously (Schmittgen and Livak, 2008).

Western blot assay

Macrophages (2 ¥ 106) were infected with 2 ¥ 107 L. amazonen-sis promastigotes for 4 h, as described. After washing awaynon-internalized parasites, cells were cultured for a further 48 h inDMEM plus 10% FBS. The medium was removed and the cellmonolayers were incubated for 30 min in 700 ml of ice-cold lysisbuffer containing protease inhibitors (20 mM Tris-HCl, 1 mMEDTA, 1% Triton X-100, 1 mM PMSF, 10 mg ml-1 aprotinin,10 mg ml-1 leupeptin, 10 mg ml-1 pepstatin-A, 1 mM PMSFpH 8.0). The cell lysates were collected and centrifuged at 2000 gfor 10 min. Uninfected macrophages (2 ¥ 106) from C57/BL6mice and BALB/c mice were treated in the same way. The pelletswere suspended in PBS containing 2% (v/v) b-mercaptoethanolas reducing agent, and the cell extracts (50 mg of protein perline) were separated by electrophoresis in 10% polyacrylamidegels. Blotting to nitrocellulose membranes was carried out usingTriseglycine transfer buffer with 20% (v/v) methanol. The mem-branes were pre-incubated with 5% (w/v) non-fat milk in Tris-buffered saline containing 0.2% (v/v) Tween-20 prior to additionof 1:200 anti-P2Y2R or anti-P2Y4R IgG (Alomone) or 1:1000anti-actin IgG (Sigma). After washing, secondary peroxidase-conjugated anti-rabbit IgG (Sigma) was added for 1 h at roomtemperature. Immunostained bands were visualized using theECL detection system (Amersham Biosciences) and acquiredusing Storm 840 imaging system (Amersham Biosciences) inblue fluorescence mode. The optical density of the revealedbands was determined by Scion Image software (Scion Corpo-ration, Maryland, USA). Quantification results are expressed inpercentage considering actin bands 100% of optical density.

Statistics

Experiments were repeated at least three times and probability(P) was calculated using a Student’s t-test. In real-time PCRexperiments were analysed by ANOVA statistical analysis usingBonferroni post-tests.

Acknowledgements

The authors would like to thank Vandir Costa and Pryscila Bragada Silva for technical support and Prof. Cerli Gatass for providingHCFDA. We are grateful to Dr James Mobley (PGRD, Pfizer,Groton, CT) for generously providing the P2X7R-/- mice. Thiswork was supported by funds from the Conselho Nacional de

Desenvolvimento Cientifico e Tecnológico do Brasil (CNPq), Pro-grama de Núcleos de Excelência (PRONEX), Fundação deAmparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), andInstituto Nacional de Ciência e Tecnologia para Pesquisa Trans-lacional em Saúde e Ambiente na Região Amazônica (INPeTAm/UFRJ).

References

Adinolfi, E., Pizzirani C., Idzko, M., Panther, E., Norgauer, J.,Di Virgilio, F., and Ferrari, D. (2005) P2X7 receptor: deathor life? Purinergic Signal 1: 219–227.

Akarid, K., Arnoult, D., Micic-Polianski, J., Sif, J., Estaquier,J., and Ameisen, J.C. (2004) Leishmania major-mediatedprevention of programmed cell death induction in infectedmacrophages is associated with the repression of mito-chondrial release of cytochrome c. J Leukoc Biol 76:95–103.

Anderson, C.M., and Parkinson, F.E. (1997) Potential signal-ling roles for UTP and UDP: sources, regulation andrelease of uracil nucleotides. Trends Pharmacol Sci 18:387–392.

Awasthi, A., Mathur, R.K., and Saha, B. (2004) Immuneresponse to Leishmania infection. Indian J Med Res 119:238–258.

Bours, M.J., Swennen, E.L., Di Virgilio, F., Cronstein, B.N.,and Dagnelie, P.C. (2006) Adenosine 5′-triphosphateand adenosine as endogenous signaling molecules inimmunity and inflammation. Pharmacol Ther 112: 358–404.

Brunschweiger, A., and Muller, C.E. (2006) P2 receptorsactivated by uracil nucleotides – an update. Curr MedChem 13: 289–312.

Burnstock, G. (2007) Purine and pyrimidine receptors. CellMol Life Sci 64: 1471–1483.

Cavaliere, F., Nestola, V., Amadio, S., D’Ambrosi, N., Ange-lini, D.F., Sancesario, G., et al. (2005) The metabotropicP2Y4 receptor participates in the commitment to differen-tiation and cell death of human neuroblastoma SH-SY5Ycells. Neurobiol Dis 18: 100–109.

Chaves, S.P., Torres-Santos, E.C., Marques, C., Figliuolo,V.R., Persechini, P.M., Coutinho-Silva, R., et al. (2009)Modulation of P2X(7) purinergic receptor in macrophagesby Leishmania amazonensis and its role in parasite elimi-nation. Microbes Infect 11: 842–849.

Chen, B.C., and Lin, W.W. (2000) Pyrimidinoceptor potentia-tion of macrophage PGE(2) release involved in theinduction of nitric oxide synthase. Br J Pharmacol 130:777–786.

Table 1. Primer sequences used for real-time PCR amplifications.

Genes

Primers

Forward (5′–3′) Reverse (5′–3′)

P2Y2 TGCTGGGTCTGCTTTTTGCT ATCGGAAGGAGTAATAGAP2Y4 TCGATTTGCAAGCCTTCTCT CCATAGGAGACCAGGGTGATP2Y6 TGCTGCTACCCCCAGTTTAC TGGCATAGAAGAGGAAGCGTGAPDH TGGCCTCCAAGGAGTAAGAAA GGCCTCTCTCTTCCTCTCAGTATC

Sequences for P2Y6 receptor cDNA amplification were as described by Lugo-Garcia et al. (2007).



Chen, B.C., Chou, C.F., and Lin, W.W. (1998)Pyrimidinoceptor-mediated potentiation of inducible nitric-oxide synthase induction in J774 macrophages. Role ofintracellular calcium. J Biol Chem 273: 29754–29763.

Chvatchko, Y., Valera, S., Aubry, J.P., Renno, T., Buell, G.,and Bonnefoy, J.Y. (1996) The involvement of an ATP-gated ion channel, P2X1 in thymocyte apoptosis. Immunity5: 275–283.

Clifford, E.E., Martin, K.A., Dalal, P., Thomas, R., andDubyak, G.R. (1997) Stage-specific expression of P2Yreceptors, ecto-apyrase, and ecto-5′-nucleotidase inmyeloid leukocytes. Am J Physiol 273: C973–C987.

Correa, G., Marques da, S.C., de Abreu Moreira-Souza, A.C.,Vommaro, R.C., and Coutinho-Silva, R. (2010) Activationof the P2X(7) receptor triggers the elimination of Toxo-plasma gondii tachyzoites from infected macrophages.Microbes Infect 12: 497–504.

Coutinho-Silva, R., Persechini, P.M., Bisaggio, R.C.,Perfettini, J.-L., Sá-Neto, A.C.T., Kanellopoulos, J.M., et al.(1999) P2Z/P2X7 receptor-dependent apoptosis of den-dritic cells. Am J Physiol 276: C1139–C1147.

Coutinho-Silva, R., Perfettini, J.-L., Persechini, P.M.,Dautry-Varsat, A., and Ojcius, D.M. (2001) Modulation ofP2Z/P2X7 receptor activity in macrophages infected withChlamydia psittaci. Am J Physiol 280: C81–C89.

Coutinho-Silva, R., Stahl, L., Raymond, M.-N., Jungas, T.,Verbeke, P., Burnstock, G., et al. (2003) Inhibition ofchlamydial infectious activity due to P2X7R-dependentphospholipase D activation. Immunity 19: 403–412.

Coutinho-Silva, R., Ojcius, D.M., Gorecki, D.C., Persechini,P.M., Bisaggio, R.C., Mendes, A.N., et al. (2005) MultipleP2X and P2Y receptor subtypes in mouse J774, spleenand peritoneal macrophages. Biochem Pharmacol 69:641–655.

Coutinho-Silva, R., da Cruz, C.M., Persechini, P.M., andOjcius, D.M. (2007) The role of P2 receptors in controllinginfections by intracellular pathogens. Purinergic Signal 3:83–90.

Cruz, C.M., Rinna, A., Forman, H.J., Ventura, A.L.,Persechini, P.M., and Ojcius, D.M. (2007) ATP activates areactive oxygen species-dependent oxidative stressresponse and secretion of proinflammatory cytokines inmacrophages. J Biol Chem 282: 2871–2879.

D’Ambrosi, N., Iafrate, M., Saba, E., Rosa, P., and Volonte, C.(2007) Comparative analysis of P2Y4 and P2Y6 receptorarchitecture in native and transfected neuronal systems.Biochim Biophys Acta 1768: 1592–1599.

D’Andrea, P., Romanello, M., Bicego, M., Steinberg, T.H., andTell, G. (2008) H(2)O(2) modulates purinergic-dependentcalcium signalling in osteoblast-like cells. Cell Calcium 43:457–468.

Del Rey, A., Renigunta, V., Dalpke, A.H., Leipziger, J., Matos,J.E., Robaye, B., et al. (2006) Knock-out mice reveal thecontributions of P2Y and P2X receptors to nucleotide-induced Ca2+ signaling in macrophages. J Biol Chem 281:35147–35155.

Dermine, J.F., Goyette, G., Houde, M., Turco, S.J., andDesjardins, M. (2005) Leishmania donovani lipophospho-glycan disrupts phagosome microdomains in J774 mac-rophages. Cell Microbiol 7: 1263–1270.

Elliott, M.R., Chekeni, F.B., Trampont, P.C., Lazarowski, E.R.,

Kadl, A., Walk, S.F., et al. (2009) Nucleotides released byapoptotic cells act as a find-me signal to promote phago-cytic clearance. Nature 461: 282–286.

Evaldsson, C., Ryden, I., and Uppugunduri, S. (2007) Anti-inflammatory effects of exogenous uridine in an animalmodel of lung inflammation. Int Immunopharmacol 7:1025–1032.

Fairbairn, I.P., Stober, C.B., Kumararatne, D.S., andLammas, D.A. (2001) ATP-mediated killing of intracellularmycobacteria by macrophages is a P2X(7)-dependentprocess inducing bacterial death by phagosome–lysosomefusion. J Immunol 167: 3300–3307.

Fang, W.-G., Pirnia, F., Bang, Y.-J., Myers, C.E., and Trepel,J.B. (1992) P2-purinergic receptor agonists inhibit thegrowth of androgen-independent prostate carcinoma cells.J Clin Invest 89: 191–196.

Feng, G.J., Goodridge, H.S., Harnett, M.M., Wei, X.Q.,Nikolaev, A.V., Higson, A.P., and Liew, F.Y. (1999) Extra-cellular signal-related kinase (ERK) and p38 mitogen-activated protein (MAP) kinases differentially regulate thelipopolysaccharide-mediated induction of inducible nitricoxide synthase and IL-12 in macrophages: Leishmaniaphosphoglycans subvert macrophage IL-12 production bytargeting ERK MAP kinase. J Immunol 163: 6403–6412.

Fortea, J., de La, L.E., Regnault, B., Coppee, J.Y., Milon, G.,Lang, T., et al. (2009) Transcriptional signatures of BALB/cmouse macrophages housing multiplying Leishmania ama-zonensis amastigotes. BMC Genomics 10: 119.

Homolya, L., Steinberg, T.H., and Boucher, R.C. (2000) Cellto cell communication in response to mechanical stress viabilateral release of ATP and UTP in polarized epithelia.J Cell Biol 150: 1349–1360.

Hopfner, M., Maaser, K., Barthel, B., von Lampe, B., Hanski,C., Riecken, E.O., et al. (2001) Growth inhibition and apo-ptosis induced by P2Y2 receptors in human colorectal car-cinoma cells: involvement of intracellular calcium andcyclic adenosine monophosphate. Int J Colorectal Dis 16:154–166.

Hu, Y., Fisette, P.L., Denlinger, L.C., Guadarrama, A.G.,Sommer, J.A., Proctor, R.A., and Bertics, P.J. (1998) Puri-nergic receptor modulation of lipopolysaccharide signalingand inducible nitric-oxide synthase expression in RAW264.7 macrophages. J Biol Chem 273: 27170–27175.

Jamieson, S.E., Peixoto-Rangel, A.L., Hargrave, A.C.,Roubaix, L.A., Mui, E.J., Boulter, N.R., et al. (2010) Evi-dence for associations between the purinergic receptorP2X(7) (P2RX7) and toxoplasmosis. Genes Immun 11:374–383.

Kavoosi, G., Ardestani, S.K., Kariminia, A., Abolhassani, M.,and Turco, S.J. (2006) Leishmania major: reactive oxygenspecies and interferon gamma induction by soluble lipo-phosphoglycan of stationary phase promastigotes. ExpParasitol 114: 323–328.

Lammas, D.A., Stober, C., Harvey, C.J., Kendrick, N., Pan-chalingam, S., and Kumararatne, D.S. (1997) ATP-inducedkilling of mycobacteria by human macrophages is medi-ated by purinergic P2Z(P2X7) receptors. Immunity 7: 433–444.

Lazarowski, E.R., and Harden, T.K. (1999) Quantitation ofextracellular UTP using a sensitive enzymatic assay. Br JPharmacol 127: 1272–1278.



Lazarowski, E.R., Homolya, L., Boucher, R.C., and Harden,T.K. (1997) Direct demonstration of mechanically inducedrelease of cellular UTP and its implication for uridine nucle-otide receptor activation. J Biol Chem 272: 24348–24354.

Lazarowski, E.R., Boucher, R.C., and Harden, T.K. (2003)Mechanisms of release of nucleotides and integration oftheir action as P2X- and P2Y-receptor activating mol-ecules. Mol Pharmacol 64: 785–795.

Lees, M.P., Fuller, S.J., McLeod, R., Boulter, N.R., Miller,C.M., Zakrzewski, A.M., et al. (2010) P2X7 receptor-mediated killing of an intracellular parasite, Toxoplasmagondii, by human and murine macrophages. J Immunol184: 7040–7046.

Li, C.M., Campbell, S.J., Kumararatne, D.S., Bellamy, R.,Ruwende, C., McAdam, K.P., et al. (2002) Association of apolymorphism in the P2X7 gene with tuberculosis in aGambian population. J Infect Dis 186: 1458–1462.

Lisi, S., Sisto, M., Acquafredda, A., Spinelli, R., Schiavone,M., Mitolo, V., et al. (2005) Infection with Leishmania infan-tum inhibits actinomycin D-induced apoptosis of humanmonocytic cell line U-937. J Eukaryot Microbiol 52: 211–217.

Luder, C.G., Gross, U., and Lopes, M.F. (2001) Intracellularprotozoan parasites and apoptosis: diverse strategies tomodulate parasite–host interactions. Trends Parasitol 17:480–486.

Lugo-Garcia, L., Filhol, R., Lajoix, A.D., Gross, R., Petit, P.,and Vignon, J. (2007) Expression of purinergic P2Y recep-tor subtypes by INS-1 insulinoma beta-cells: a molecularand binding characterization. Eur J Pharmacol 568: 54–60.

Maaser, K., Hopfner, M., Kap, H., Sutter, A.P., Barthel, B.,von Lampe, B., et al. (2002) Extracellular nucleotidesinhibit growth of human oesophageal cancer cells viaP2Y(2)-receptors. Br J Cancer 86: 636–644.

Mantuano, M.B., Henriques-Pons, A., Araujo-Jorge, T.C., DiVirgilio, F., Coutinho-Silva, R., and Persechini, P.M. (2003)Extracellular ATP induces cell death in CD4+/CD8+ double-positive thymocytes in mice infected with Trypanosomacruzi. Microbes Infect 5: 1363–1371.

Marriott, I., Inscho, E.W., and Bost, K.L. (1999) Extracellularuridine nucleotides initiate cytokine production by murinedendritic cells. Cell Immunol 195: 147–156.

Moore, K.J., Labrecque, S., and Matlashewski, G. (1993)Alteration of Leishmania donovani infection levels by selec-tive impairment of macrophage signal transduction.J Immunol 150: 4457–4465.

Olivier, M., Brownsey, R.W., and Reiner, N.E. (1992) Defec-tive stimulus–response coupling in human monocytesinfected with Leishmania donovani is associated withaltered activation and translocation of protein kinase C.Proc Natl Acad Sci USA 89: 7481–7485.

Oshimi, Y., Miyazaki, S., and Oda, S. (1999) ATP-inducedCa2+ response mediated by P2U and P2Y purinoceptors inhuman macrophages: signalling from dying cells to mac-rophages. Immunology 98: 220–227.

Osorio y Fortéa, J., de La Llave, E., Regnault, B., Coppée,J.Y., Milon, G., Lang, T., and Prina, E. (2009) Transcrip-tional signatures of BALB/c mouse macrophages housingmultiplying Leishmania amazonensis amastigotes. BMCGenomics 20: 110–119.

Ouaissi, A. (2003) Apoptosis-like death in trypanosomatids:search for putative pathways and genes involved. Kineto-plastid Biol Dis 2: 5.

Pick, E., and Mizel, D. (1981) Rapid microassays for themeasurement of superoxide and hydrogen peroxide pro-duction by macrophages in culture using an automaticenzyme immunoassay reader. J Immunol Methods 46:211–226.

Ruhland, A., Leal, N., and Kima, P.E. (2007) Leishmaniapromastigotes activate PI3K/Akt signalling to confer hostcell resistance to apoptosis. Cell Microbiol 9: 84–96.

Schmittgen, T.D., and Livak, K.J. (2008) Analyzing real-timePCR data by the comparative C(T) method. Nat Protoc 3:1101–1108.

Seifert, R., and Schultz, G. (1989) Involvement of pyrimidi-noceptors in the regulation of cell functions by uridineand by uracil nucleotides. Trends Pharmacol Sci 10: 365–369.

Shrestha, S.P., Tomita, T., Weiss, L.M., and Orlofsky, A.(2006) Proliferation of Toxoplasma gondii in inflammatorymacrophages in vivo is associated with diminished oxygenradical production in the host cell. Int J Parasitol 36: 433–441.

Soh, N. (2006) Recent advances in fluorescent probes for thedetection of reactive oxygen species. Anal Bioanal Chem386: 532–543.

Walters, L.L., Irons, K.P., Chaplin, G., and Tesh, R.B. (1993)Life cycle of Leishmania major (Kinetoplastida: Trypanoso-matidae) in the neotropical sand fly Lutzomyia longipalpis(Diptera: Psychodidae). J Med Entomol 30: 699–718.

Yilmaz, O., Yao, L., Maeda, K., Rose, T.M., Lewis, E.L.,Duman, M., et al. (2008) ATP scavenging by the intracel-lular pathogen Porphyromonas gingivalis inhibits P2X7-mediated host-cell apoptosis. Cell Microbiol 10: 863–875.

Supporting information

Additional Supporting Information may be found in the onlineversion of this article:

Fig. S1. Effects of P2Y receptors agonists and antagonists onparasite load, apoptosis and ROS production. Resident mac-rophages were infected with L. amazonensis promastigotes atan moi of 10:1 for 4 h, and free parasites were then washed off.After 24 h post infection, cells were treated with the agonists,100 mM UTP and UDP or 50 mM UTPgS and UDPbS, or controlbuffer. After 12 h the cells were fixed and parasite load counted(A); or after 8 h, cells were incubated with buffer containingethidium bromide (50 mg ml-1) and analysed by flow cytometry(B). Alternatively, cells were treated with 100 mM UTP or 3 mMATP in the presence or absence of the antagonists, 100 mMSuramin and 300 mM PPADS. After 1 h, ROS (C) were mea-sured by HCFDA fluorescence (C). In (A) at least 250 cellswere examined and results were expressed as the infectionindex (II), which is obtained by the formula: II = (% infectedmacrophages) ¥ (amastigotes/infected macrophage)/100. In (B)and (C), data from 10 000 cells were collected on a FACScanand analysed using Cell Quest 3.3 software. Values representmean � SEM of two (A), three (B) or four (C) separated experi-ments. ***P < 0.0001, **P < 0.001, *P < 0.05.



Fig. S2. Knockout mice reveal that uracyl-induced apoptosis isindependent of the P2X7 receptor. Peritoneal resident macroph-ages from P2X7+/+ (A) and P2X7-/- (B) mice were infected ormock-infected with L. amazonensis promastigotes at a moi of10:1. After 4 h of infection, free parasites were removed bywashing. After 24 h of infection, cells were treated with controlbuffer, 100 mM UTP, 100 mM UDP or 3 mM ATP. After 8 h, apop-tosis was quantified by cytofluorimetry, measuring hypodiploidnuclei. Data from 10 000 cells were collected on a FACScan andanalysed using Cell Quest 3.3 software. Cells were uninfected(white bars) or infected (black bars) with L. amazonensis. Valuesrepresent mean � SEM of four independent experiments per-formed in triplicate ***P < 0.0001; **P < 0.001; *P < 0.05. Usingmacrophages from C57BL6 mice, we observed that infectedmacrophages undergo apoptosis induced by ATP (30 � 2%),UTP (34 � 2%) and UDP (19 � 1%) (A). These results show thatthe genetic background does not alter the sensitivity of the cells

to nucleotide-mediated apoptosis. In addition, we observed thatin macrophages from P2X7

-/- mice, spontaneous apoptosis wasslightly lower than in macrophages from P2X7

+/+ mice (3 � 2%),as seen in (B). The apoptosis induced by ATP was abolished inboth infected and uninfected macrophages from P2X7

-/- mice (B),suggesting that extracellular ATP was acting mainly throughP2X7. In contrast, UTP and UDP induced apoptosis in 22 � 1%and 16 � 1%, respectively, only in infected macrophages(A) from either wild-type or P2X7-deficient mice (B). This indicatesthat uracyl nucleotides induce apoptosis via activation of amechanism that is independent of the P2X7 receptor.

Please note: Wiley-Blackwell are not responsible for thecontent or functionality of any supporting materials suppliedby the authors. Any queries (other than missing material)should be directed to the corresponding author for thearticle.



Infection with Leishmania amazonensis upregulates ... · Infection with Leishmania amazonensis...

Documents

Transcript of Infection with Leishmania amazonensis upregulates ... · Infection with Leishmania amazonensis...