at SciVerse ScienceDirect

Microbial Pathogenesis 52 (2012) 69e76

Contents lists available

Microbial Pathogenesis

journal homepage: www.elsevier .com/locate/micpath

Anoikis potential of Entameba histolytica secretory cysteine proteases:Evidence of contact independent host cell death

Sudeep Kumar 1, Rajdeep Banerjee 1, Nilay Nandi 1, Abul Hasan Sardar 1, Pradeep Das*

Department of Molecular Biology, Rajendra Memorial Research Institute of Medical Sciences, Agamkuan, Patna, Bihar 800007, India

a r t i c l e i n f o

Article history:Received 5 August 2010Received in revised form11 October 2011Accepted 11 October 2011Available online 20 October 2011

Keywords:AnoikisE. histolyticaMDA-MB-231JNKCysteine proteasesApoptosis

* Corresponding author. Tel.: þ91 612 263 6651;E-mail addresses: skb.111@gmail.com (S. K

(R. Banerjee), nilaynandi@gmail.com (N. Nandi), ab(A.H. Sardar), drpradeep.das@gmail.com (P. Das).

1 Tel.: þ91 612 263 6651.

0882-4010/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.micpath.2011.10.005

a b s t r a c t

Mammalian epithelial, endothelial and various other cell types, upon their detachment from theextracellular matrix (ECM) undergo a specialized kind of apoptosis, known as anoikis. Entameba histo-lytica cysteine proteases have been implicated in degradation of the host ECM, which may induce anoikisin host cells. To explore this hypothesis, supernatant obtained from 2 h in-vitro cultivation of E. histolytica(SRP), was used as a source of cysteine proteases. MDA-MB-231 (human mammary epithelial adeno-carcinoma) cells were treated with SRP and their detachment and apoptosis was evaluated. 25 mg/ml(with respect to protein concentration), SRP was found to be the optimal concentration to dislodge over98% MDA-MB-231 cells from monolayer in 20 min. The detachment was followed by apoptosis of at least41.2% cells, characterized by caspase-3 dependent inter-nucleosomal DNA fragmentation. The SRP-induced apoptosis was associated exclusively with the detached fraction. Moreover, detachmentpreceded apoptosis. E-64 (a cysteine protease inhibitor) abolished the SRP-induced detachment as wellas inter-nucleosomal DNA fragmentation. Interestingly, SRP induced a 3.21 fold increase in the JNKactivity, whilst SP600125 (a JNK inhibitor) blocked the SRP-induced inter-nucleosomal DNA fragmen-tation. Thus, it was concluded that spontaneously released cysteine proteases of E. histolytica can induceJNK dependent anoikis of MDA-MB-231 cells, which may be implicated in contact independent host celldeath during amebiasis.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Amebiasis is one of the most aggressive and is considered to bethe third leading cause of death amongst parasitic diseases, sur-passed only by malaria and Schistosomiasis [1]. Invasive amebiasisis responsible for enteric syndromes ranging from dysentery tocolitis, which is highly lethal. Sometimes, active trophozoitesinvade the portal circulation, through the bowl wall and dissemi-nate systemically, reaching the liver to cause hepatic amebiasis.Following the depression of the protective mucus blanket,trophozoites attach to the cells of the inter-glandular epitheliumand with the aid of proteolytic enzymes, especially cysteineproteases, degrade elastin, collagen and fibronectin; the compo-nents of extracellular matrix (ECM). This invasion is followed bylateral extension of the destructive lesions in the sub-mucosal layer,

fax: þ91 612 263 4379.umar), araj.bng@gmail.comulhasansardar82@gmail.com

All rights reserved.

creating classical flask shaped ulcers [reviewed in [2e4]]. Despitevarious studies on pathogenesis of amebiasis, themechanism of celldeath at the site of invasion remains illusive.

Earlier investigations provide evidence of apoptotic death ofhost cells by Entameba histolytica [5e8]. However, most of themshow that the apoptosis of the host cells depends on contact withthe Entameba trophozoites [5,7,9]. Interestingly, all the studiesshowing contact-dependent apoptosis were conducted in myeloidor lymphoid cell types [9e11]. As per the current knowledge aboutmammalian cell death, the mechanism of cell death varies with celltype. Therefore, the contact-dependency of the death of themyeloid/lymphoid cells induced by E. histolytica may not beapplicable to the other cell types, particularly intestinal and hepaticcells, the most commonly affected organs in invasive amebiasis.Hepatocyte and enterocyte apoptosis by E. histolytica is very welldocumented [6,12,13]. Further, the hepatocyte apoptosis has beenimplicated in amebic liver abscess (ALA). Interestingly, amebiccysteine proteases have been implicated in ALA too [12].So apparently it seems that there is a link between cysteineprotease mediated tissue damage, host-cell apoptosis and amebicliver abscess and/or colonic ulcer formation.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e7670

The intestinal epithelial cells grow in an adhesion dependentmanner, where they require attachment with ECM for their growthand survival [14]. The degradation of ECM-Cell attachment causesa specialized type of apoptosis, known as anoikis [15]. ECMdegradation is a well known phenomenon in amebiasis [16], whichmay render cells devoid of contact with the ECM. So, it washypothesized that anoikis may be involved in amebiasis. Therefore,the major objective of the present investigation was to explore theanoikis potential of E. histolytica cysteine proteases. As the cysteineproteases are known to be secreted spontaneously during in-vitrocultivation of E. histolytica [17], the supernatant of 2 h in-vitroculture of E. histolytica was used as a source of cytsteine protease.On the other hand, the MDA-MB-231 (human mammary epithelialcell adenocrcinoma) was used as anoikis susceptible cell line [18,19]while Jurkat (which grows in adhesion independent manner) wasused as an anoikis resistant cell line.

2. Results

2.1. SRP causes detachment of MDA-MB-231 monolayer

To evaluate the morphological changes of the MDA-MB-231monolayer inflicted by SRP, the monolayer was treated withdifferent doses of SRP for various time points andwas photographedwith a digital camera mounted on an inverted microscope. Thedetachment effect was found to be dose dependent. Upon 20 minincubation with SRP, 5 mg or 10 mg resulted in detachment of <10%cellswhereas 30% and98% cellsweredetachedwith 20 mg, and 25 mgrespectively (Fig. 1A and B). Similarly the duration of treatment alsoaffected the detachment of cells. At 20min of treatment with 25 mg/ml of SRP>98% cellswere detached,whereas 18% and56% cellsweredetached at 5 min and 15 min respectively (Fig. 1A and B).

2.2. SRP induces apoptosis of adhesion dependent MDA-MB-231but not of Jurkat

Since adhesion dependent cells undergo apoptosis afterdetachment, we wanted to explore the fate of the cells which showdetachment by SRP treatment. Therefore, the MDA-MB-231monolayer was treated with 25 mg SRP for indicated time(Fig. 2A), after which all the cells (attached and detached) werepooled and their DNAwas isolated. The DNA from each sample wasseparated on 1.5% agarose gel. Intenucleosomal DNA fragmentswere seen in samples treated with SRP for 8 h or more. Themaximum inter-nucleosomal DNA fragmentation was seen at 24 hof SRP treatment, which did not increase further until 36 h oftreatment (Fig. 2A). Caspase-3 activity was found to be significantlyhigher in SRP treated samples compared to control after 6 h ofincubation. However, no significant increase in caspase-3 activitywas found at or before 4 h of treatment (Fig. 2B). Since the cellulardetachment is complete within 20min, this observation shows thatthere is a delay between detachment and augmentation ofapoptosis. This correlates well with the fact that DNA degradationoccurred only after 8 h of SRP treatment (Fig. 2B). The cell cycleanalysis revealed that until 8 h of SRP treatment there is a slightincrease in the sub-G0/G1 population (3.53%) compared to control(0.41%), whereas there is a significant increase in the same at 12 h(14.2%) and 24 h (41.27%) (Fig. 2C). The, z-VAD-FMK (a pan caspaseinhibitor) completely abolished the SRP-induced inter-nucleosomalDNA fragmentation (Fig. 2D). On the other hand, Jurkat cells treatedwith SRP in a similar way to the MDA-MB-231 cell line did notshow any inter-nucleosomal DNA fragmentation, even after 36 h oftreatment (Fig. 2F).

To know whether the SRP-induced inter-nucleosomal DNAfragmentation was associated with the detached or attached cells,

a suboptimal dose (5 mg) of SRP was selected, which detached only40% cells of the monolayer, upon 24 h treatment. The attached anddetached fractions were subjected to the DNA laddering assay,which revealed inter-nucleosomal DNA fragments in the detachedfraction while there was none in the attached fraction (Fig. 2E).

2.3. Re-seeding of SRP treated MDA-MB-231

Previous experiments demonstrated that apoptosis is associatedwith detached cells. Because the cells undergoing apoptosis alsoexhibit detachment, we wanted to know whether the detachmentinduced by SRP is the cause or the effect of apoptosis, i.e. whetherdetachment is an upstream or downstream event to apoptosis. Ifdetached living cells are reseeded, they are expected to exhibit nosigns of apoptosis. Therefore, MDA-MB-231 cells were reseededafter incubationwith SRP for various times. Inter-nucleosomal DNAfragmentation assay occurs only after 8 h of SRP treatment and ismaximal at 24 h (Fig. 2A). Therefore, inter-nucleosomal DNA frag-mentation was evaluated after 24 h of SRP treatment. Samples thatwere treated with SRP for less than 1 h before reseeding did notshow any inter-nucleosomal DNA fragments. On the other hand,samples that were treated with SRP for 2 h or more showed inter-nucleosomal DNA fragmentation (Fig. 3).

Interestingly, cells reseeded after 15e60 min of SRP treatmentcompletely attached to the surface of the tissue culture flask (datanot shown), which would be expected from the living cells.However, the cells remain afloat in media for 2e3 h before theyattached completely (data not shown). Thus, the total time forwhich the cells remained under suspension was 15 min þ 2e3 h,30 min þ 2e3 h and likewise. In this experiment, the cells requiredat least 2 h of SRP treatment for apoptosis (Fig. 3), thereby stayingunder suspension for at least 4e5 h before undergoing apoptosis.

2.4. Involvement of cysteine proteases in SRP-induced apoptosis

In this investigation detachment was found to be involved inSRP-induced apoptosis. Since cysteine proteases are secretedduring the in-vitro cultivation of Entameba [17], which have beenimplicated in monolayer detachment [20], we evaluated theinvolvement of cysteine proteases in SRP-induced detachment andapoptosis. The MDA-MB-231 monolayer was treated with SRP(25 mg/ml) in the presence or absence of E-64 (a cysteine proteaseinhibitor). E-64 þ SRP treated samples showed normal monolayerwhereas SRP treated samples showed detached cells within5e20 min of SRP treatment (Fig. 4A). Moreover, E-64 þ SRP treatedsamples showed no inter-nucleosomal DNA fragmentation, whichwas clearly observed in SRP treated samples (Fig. 4B). Further, thecell cycle analysis showed 5% apoptotic cells in samples treatedwith E-64 þ SRP, whereas 38% apoptotic cells were associated withthe SRP treated samples (Fig. 4C).

2.5. Involvement of JNK in SRP-induced apoptosis

JNK activity has been implicated in detachment and anoikis [21].Since we had found the SRP induced apoptosis involved detach-ment, we wanted to elucidate whether JNK activity is involved inSRP-induced apoptosis. The active JNK phosphorylates c-Jun,thereby increasing the phospho-c-Jun level. SRP treated samplesshowed 3 fold more phospho-c-Jun compared to control (withoutSRP treatment) (Fig. 5A). Moreover, SRP þ SP600125 (a JNK inhib-itor) treated samples showed no inter-nucleosomal DNAfragments (Fig. 5B). The cell cycle analysis showed 8% apoptoticcells associated with SRPþ SP600125 treated samples whereas 33%apoptotic cells associated with control (only SRP treated samples)(Fig. 5C).

Fig. 1. Detachment of MDA-MB-231: (A) shows the micrograph of SRP treated MDA-MB-231, wherein MDA-MB-231 monolayer was treated with indicated doses of SRP forindicated time before taking the image with a digital camera, mounted on an inverted microscope. (B) shows the % of detached cells as counted with haemocytometer after SRPtreatment with indicated doses of SRP for indicated time. % detached cells show fraction of cells detached among the total cells (Attached þ Detached) per sample. The error barshows SE of mean of 3 individual experiments.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e76 71

Fig. 2. SRP induces apoptosis of MDA-MB-231: (A) shows the agarose gel electrophoresis of DNA isolated fromMDA-MB-231 after treatment with SRP (25 mg/ml) for indicated time.(B) shows the caspase-3 activity in terms of fluorescence intensity of AMC (7-amino-4methylcoumarin) obtained after reacting Ac-DEVD-AMC with MDA-MB-231 lysates, whereinMDA-MB-231 monolayer was treated/untreated with SRP for indicated time. The Y error bar shows SE of mean of three individual experiments. (C) shows the % of cells in sub G0/G1phase obtained by cell cycle analysis of MDA-MB-231 which were treated with 25 mg/ml SRP for indicated time. 24 h EtOH indicates that MDA-MB-231 was treated with 5% Ethanolfor 24 h. The cells were stained with propidium iodide after treatment with SRP or Ethanol before cell cycle analysis. The G0/G1, G2 and S phases were distinguished and % of cells insub G0/G1 was recorded, while at least 30000 events were analyzed by FC-500 (Beckman Coulter). The Y error bars indicate the SE of mean of three individual experiments. (D)shows the agarose gel electrophoresis of DNA of MDA-MB-231 isolated after treatment with indicated reagents. (E) shows the agarose gel electrophoresis of DNA isolated fromattached and detached fractions of SRP treated MDA-MB-231 cells. (F) shows the agarose gel electrophoresis of DNA isolated from SRP untreated or treated Jurkat for indicated time.Negative and positive sign shows the absence or presence of SRP. 24 h indicates that Jurkat was treated with Fas-L for 24 h.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e7672

3. Discussion

The present investigation demonstrates the capability of SRP ofE. histolytica to induce anoikis in MDA-MB-231 cells. SRP treatmentcaused detachment of MDA-MB-231 cells (Fig. 1A and B) in a timeand dose dependent manner and induced apoptosis in 41% of MDA-

MB-231 cells (Fig. 2C). On the other hand, Jurkat cells, which growin suspension, were found to be resistant to the SRP-inducedapoptosis (Fig. 2F). This SRP-induced apoptosis was characterizedby inter-nucleosomal DNA fragmentation (Fig. 2A) and caspase-3activation (Fig. 2B). Furthermore, the inter-nucleosomal DNA frag-mentation caused by SRP was abolished by z-VAD-fmk (Fig. 2E).

Fig. 3. Effect of reseeding of MDA-MB-231, detached by SRP treatment: MDA-MB-231was treated with SRP for indicated time, thereupon was suspended in serum and SRPfree DMEM and was incubated for the duration so as to complete 24 h of incubation.After completion of incubation DNA was isolated and separated on 1.5% agarose gel.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e76 73

In this investigation, the SRP-induced inter-nucleosomal DNAfragmentation was found to be associated with the detached frac-tion of the monolayer (Fig. 2E). Moreover, the SRP-induceddetachment preceded apoptosis while SRP mediated detachedcells remained alive for at least 4e5 h in suspension (Fig. 3) andshowed normal growth pattern upon reseeding (data not shown).Taken together it was concluded that SRP induces detachmentmediated apoptosis (i.e. anoikis).

Previous investigations have shown that E. histolytica lysate orthe cell culture filtrate was incapable of inducing cell death of hostcells [22,23]. These observations may account for the peculiarity ofthe anoikis mechanism, which is known to occur in variousanchorage dependent primary cell cultures e.g. endothelial cells[24], hepatocytes [25] and enterocytes [27] as well as in trans-formed cell types like MDCK [27] and FSK-7 [28]. However, celltypes likemacrophages and lymphocytes, which aremigratory cellsand grow independent of cell-ECM or cellecell contact, are resis-tant to anoikis. There are other cell types which show differentialbehavior to different detachment inducing factors e.g. CHOundergoes apoptosis by plasmin induced detachment [29] whereasdoes not undergo apoptosis upon detachment induced byE. histolytica trophozoite lysate/culture filtrate [22]. Thus, theapoptosis susceptibility of a cell type depends both on its origin andthe type of apoptosis stimulus. This notion may be applied toexplain the apoptosis resistance of BHK, CHO, Jurkat and othermonocyte derived cells by Entameba cell lysates or culture filtrateor purified cysteine proteases. In this investigation we found thatSRP, a cell free culture supernatant, is capable of inducing apoptosis,which may be implicated in contact independent host cell death byEntameba.

Amebiasis has been found to be associated with the apoptosis ofhepatocytes [6] and enterocytes [13], while apoptosis has beenimplicated in ALA formation [12]. In a mouse model of ALA,

inhibition of cysteine protease activity reduced the ALA [30]. Earlierit was noted that E. histolytica cysteine proteases have beenimplicated in monolayer detachment [30], invasion through thelamina propria [31] and degradation of colonic mucin [32].The detachment caused by E. histolytica cysteine protease isconsistent with observations in this investigation, as monolayerdetachment was found to be dependent on the cysteine proteaseactivity of SRP. The proteases which specifically degrade the ECMhave been implicated in tissue remodeling and apoptosis [33].Moreover, there are evidences on the ability of proteases to induceapoptosis e.g. Granzyme B, mast cell chymase [34] and Jararhagin(a protease from snake venom) [35]. In the present study, it wasfound that, the pattern of apoptosis induced by SRP correlates withtrypsin mediated apoptosis of HUVEC [36], which further supportsthe role of cysteine proteases in SRP mediated apoptosis. However,because of the complexity of SRP, the possibility of other SRP factorsbeing activated to induce apoptosis via cysteine protease activitymay not be ruled out.

The JNK pathway (which is also known as stress activatedprotein kinase pathway), is known to respond to extracellularstimuli that induce apoptosis eg. tumor necrosis factor a, ceramide,granzyme b, interleukin 1, UV light and g radiation [37,38]. JNK isknown to be rapidly activated by the disruption of cell matrixalterations in MDCK cells [21]. In this investigation, treatment withSRP induced JNK activity, which was found to be involved inapoptosis (Fig. 5).

In conclusion, this investigation shows that secretory cysteineproteases of Entameba are capable of inducing anoikis in MDA-MB-231 cells. While, the anoikis may be implicated in the pathogenicmechanism of amebiasis, a model of pathogenesis of amebiasis maybe proposed involving cysteine protease mediated degradation ofECM-Cell and cellecell attachment, thereby inducing anoikis ofhepatocytes and enterocytes which could lead to liver abscess andcolonic ulcer formation. However, further investigations areneeded to unravel the role of anoikis in amebiasis.

4. Methods

4.1. Cell culture

MDA-MB-231 (a human mammary epithelial adenocarcinomacell line, obtained from National Center for Cell Sciences, Pune,India) was maintained with DMEM (Invitrogen, NY, USA) supple-mented with 10% FBS (Invitrogen, NY, USA) & 2 mM L-Glutamine at37 �C and 5% CO2. All the experiments involving MDA-MB-231 cellswere done with 2 � 105 cells/cm2 cellular density unless statedotherwise. Since serum components have been known to delayanoikis [39], all the experiments involving detachment mediatedapoptosis were done in serum free medium.

Jurkat (a Human lymphoma derived cell line, obtained fromNCCS, Pune, India), was maintained with RPMI-1640 (Invitrogen,NY, USA) supplemented with 10% FBS at 37 �C and 5% CO2. Allexperiments with Jurkat cells were done at 1�105 cells/ml density.

E. histolytica strain HM1: IMSS was maintained with TYI-S-33medium supplemented with 10% bovine serum and incubation at35.5 �C [40].

4.2. Isolation of SRP

48 h old E. histolytica trophozoites (HM1: IMSS) in their expo-nential growth phase were chilled in a 4 �C ice bath for 10 min,harvested and washed twice with serum free TYI-S-33 medium.The cells were resuspended in serum free TYI-S-33 medium ata density of 2� 106 cells/ml and were incubated at 35.5 �C. After 2 hof incubation, the contents in the culture flask were centrifuged at

Fig. 4. Involvement of cysteine proteates: (A) shows the micrograph of MDA-MB-231 monolayer, treated as indicated. (B) shows the agarose gel electrophoresis of DNA obtainedfrom MDA-MB-231 after treatment with indicated reagents for 24 h. (C) shows the % MDA-MB-231 in sub G0/G1 region as obtained from cell cycle analysis after treatment withindicated reagents for 24 h wherein positive and negative sign shows the presence or absence of respective reagents. The G0/G1, G2 and S phases were distinguished and % of cells insub G0/G1 was recorded, while at least 30000 events were analyzed by FC-500 (Beckman Coulter). The Y error bar shows the SE of mean of three individual experiments.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e7674

400 g and the supernatant was collected, recentrifuged at 5000 gfor 10 min and the resultant supernatant was collected in freshtubes. The supernatant was then concentrated 10 times and dia-lyzed simultaneously with PBS (pH 7.4) in cold condition byultrafiltraion device with 10 mm pore size (Centricon YM 10, Mil-lipore, USA). This product was called spontaneously releasedproduct (SRP). The total protein concentration of the SRP wasmeasured by Bradford’s method. The SRP was stored in 100 mlaliquots at �70 �C until use. The cells collected from centrifugationat previous step was analyzed for viability by trypan blue dyeexclusion method where �95% cells were found to be alive

(data not shown) confirming that the supernatant fractionscollected as SRP contains the actual secretory products and not thelysed/dead product of Entameba trophozoites.

4.3. Microscopy

To assess the morphological changes inflicted by SRP on MDA-MB-231 cells, the monolayer was trypsinized and the cells wereallowed to grow in 6 well tissue culture plates with 3 mlDMEMþ10% FBS at a density of 2 � 105 cells/cm2. After overnightincubation the monolayer was washed twice with serum free

Fig. 5. Involvement of JNK: (A) shows the western blot of protein isolated from MDA-MB-231, which were untreated or treated with SRP. The blot was reacted with phospho-c-Junand developed with BCIP/NBT. Normalization was done with total c-Jun. (B) shows the agarose gel electrophoresis of DNA of MDA-MB-231 cells, which were treated with indicatedreagents for 24 h. (C) shows the % cells in sub G0/G1 phase as obtained by cells cell cycle analysis of MDA-MB-231, which was treated with indicated reagents for 24 h, whereinpositive and negative sign shows the presence or absence of respective reagents. The G0/G1, G2 and S phases were distinguished and % of cells in sub G0/G1 was recorded, while atleast 30,000 events were analyzed by FC-500 (Beckman Coulter). The Y error bar shows SE of mean of three individual experiments.

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e76 75

DMEM before treatment with 0, 5, 10, 20, 25 and 50 mg/ml of SRPseparately. Photographswere taken by a digital cameramounted onan inverted microscope after completion of 0, 5, 15 and 20 min oftreatment for each dose.

To see the effect of cysteine protease inhibitors on the detach-ment of MDA-MB-231 cells, the monolayer was washed twice withserum free DMEM and treated with 25 mg/ml SRP (found to beoptimum in detaching cells) in the presence or absence of 50 mM E-64 (optimum for blocking the detachment) and photographs weretaken as described above after 0, 5, 15 and 20 min of treatment.

4.4. DNA laddering assay

MDA-MB-231 monolayer was trypsinized and grown in 25 cm2

tissue culture flasks at a cellular density of 2 � 105 cells/cm2. Afterovernight incubation at 37 �C in a 5% CO2 incubator, the cells werewashed twice with serum free DMEM and flooded with 5 ml serumfree DMEM. The cells were then treated with 25 mg/ml SRP for 2, 4,6, 8, 12, 24 and 36 h. After completion of each treatment, as all thecells were found to be detached and floating in the media, theywere collected by centrifugation. The cell pellet was washed withPBS (pH 7.2) and 5 � 106 cells from each sample were lysed in100 mM NaCl, 10 mM EDTA, 25 mM TriseHCl (pH 8.0), 1% SDS and500 mg/ml proteinase K. Lysates were extracted with Phenol:Chloroform (1:1). The DNA was precipitated by ethanol, treatedwith RNAse and separated in 1.5% agarose gel containing ethidiumbromide. The photograph of the gel was captured by Gel Docu-mentation system (Bio-Rad) under UV light.

For treatment with z-VAD-FMK, E-64 or SP600125, MDA-MB-231 monolayer was trypsinized and cells were allowed to grow in25 cm2

flasks at a density of 2 � 105 cells/ml. After overnightincubation cells were washed twice with serum free DMEM andtreated with Z-VAD-FMK (40 mM), E-64 (50 mM) or SP600125(20 mM) separately, 1 h prior to SRP treatment, while SRP was givenat 25 mg/ml (with respect to protein) concentration for 24 h. Thecells were kept at 5% CO2 and 37 �C during incubation, thereafterthey were harvested and DNA laddering assay was done asdescribed above.

4.5. Reseeding

For reseeding experiments, the MDA-MB-231 monolayer wastrypsinized and grown in 25 cm2 tissue culture flasks at a density of2�105 cells/cm2. After overnight incubation, the cells werewashed

twice with serum free DMEM and flooded with 5 ml fresh serumfree DMEM before treatment with 25 mg/ml (with respect toprotein) SRP for 15 min, 30 min, 1 h, 2 h, 4 h, 6 h and 8 h. Aftercompletion of incubation, cells were collected by centrifugation,washed twice with serum free DMEM and re-plated in fresh 25 cm2

flasks with 5 ml serum free DMEM. The cells were incubatedfurther to total 24 h, which included the time incubation before andafter reseeding. The cells were kept at 37 �C with 5% CO2 duringincubation. After completion of incubation all the cells includingattached and detached were harvested, pooled and DNA ladderingassay was done as described above.

4.6. Caspase-3 assay

Caspase-3 assay was done according to the manufacturer’sprotocol with fluorogenic caspase-3 assay kit (BD Biosciences, USA).The kit consists of a fluorogenic substrate Ac-DEVD-AMC, which hascaspase-3 specific recognition and cleavage site (DEVD). Uponreaction with active caspase-3, AMC is liberated, which has excita-tion maxima at 380 nm and emission maxima between 430 and460 nm. Briefly, the monolayer was prepared as described for theDNA laddering assay. After twowasheswith serum freeDMEM, cellswere treated with 25 mg/ml SRP for 30 min, 1 h, 2 h, 4 h, 6 h, 8 h and12 h, wherein the cells were submerged in 5 ml serum free DMEM.After completion of incubation, cells were harvested and lysed withlysis buffer provided with the kit. The lysate was mixed with fluo-rogenic substrate and incubated for 1 h at 37 �C and thefluorescencewas measured with 380 nm excitation and 430e460 nm emissionwavelength with a spectropfluorometer (LS-55, Perkin Elmer, USA).The data obtained as arbitrary fluorescence units afterflouorometryof samples, was presented as bar graph for each time point. Each barshows mean � SE of three individual experiments.

4.7. Cell cycle analysis

The preparation of cells and treatment was the same to that ofDNA laddering assay except the time of incubation which was fol-lowed as 8 h, 12 h and 24 h. After indicated time of incubation, cellswere harvested by centrifugation and washed with PBS (pH 7.2).The cell pellets were resuspended in 200 ml PBS (pH 7.2) and 70%ethanol was added slowly while vortexing, and were kept on ice forat least 30 min. The cells were diluted to 2 � 106 cells/ml in PBScontaining 50 mg/ml propidium iodide and 40 mg/ml RNAse andincubated for 1 h at 37 �C before analysis with flow cytometer (FC-

S. Kumar et al. / Microbial Pathogenesis 52 (2012) 69e7676

500, Beckman Coulter, Brea, CA, USA) using FC-500 software.At least 30,000 events were included in the analysis. The cellulardebries were avoided by setting the threshold of PI fluroscence at25. Three regions corresponding to G0/G1, S and G2 phases weredistinguished and the sub G0/G1 regionwas evaluated as apoptoticpopulation. Three individual experiments were done for eachsample and data was represented as mean � SE of three individualexperiments.

For inhibition assays, cells were treated with 25 mg/ml SRP inpresence or absence of E-64 (50 mM) or SP600125 (20 mM) sepa-rately. The cells were incubated for 24 h before cell cycle analysis asdescribed above. Three individual experiments were done for eachsample and data were presented as mean � SE.

4.8. Western blotting

MDA-MB-231 monolayer was trypsinized and grown on 25 cm2

tissue culture flask at a density of 2 � 105 cells/cm2. After overnightincubation cells were washed twice with serum free DMEM andtreated with 25 mg/ml SRP for 1 h. Thereafter, the cells were har-vested and total protein was isolated and quantified. Subsequently60 mg sample was separated with SDS-Polyacrylamide gel electro-phoresis and transferred to nitrocellulose membrane, which wasfollowed by blocking with 5% BSA and treatment with antiphospho-c-Jun primary antibody (Cell Signalling Technologies,Boston, MA, USA) diluted to 1:1000. Alkaline phosphatase conju-gated secondary antibody was applied at 1: 5000 dilution. Dilutionof primary and secondary antibodywas donewith 5% BSA. The blot,reacted with primary and secondary antibody, was developed withBCIP/NBT. Normalization was done using antibody which recog-nizes phosphorylated as well as nonphosphorylated c-Jun.

Acknowledgments

We duly acknowledge Council of Scientific and IndustrialResearch, New Delhi, India for providing fellowship to Mr. SudeepKumar during this study.

References

[1] WHO/PAHO/UNESCO report. A consultation with experts on amebiasis.Mexico City, Mexico 28-29 January, 1997. Epidemiol Bull 1997;18:13e4.

[2] Stanley Jr SL. Amebiasis. Lancet 2003;361:1025e34.[3] Espinosa-Cantellano M, MartinezePalomo A. Pathogenesis of intestinal

amebiasis: from molecules to disease. Clin Micobiol Rev 2000;13:318e31.[4] Que X, Reed S. cysteine proteinases and the pathogenesis of amebiasis. Clin

Microbiol Rev 2000;13:196e206.[5] Ragland BD, Ashley LS, Vaux DL, Petri Jr WA. Entameba histolytica: target cells

killed by trophozoites undergo DNA fragmentation which is not blocked byBcl-2. Exp Parasitol 1994;79:460e7.

[6] Seydel KB, Stanley Jr SL. Entameba histolytica induces host cell death in amebicliver abscess by a non-Fas-dependent, non-tumor necrosis factor alpha-dependent pathway of apoptosis. Infect Immun 1998;66:2980e3.

[7] HustonCD,HouptER,MannBJ,HahnCS, Petri JrWA.Caspase3-dependent killingof host cells by the parasite Entameba histolytica. Cell Microbiol 2000;2:617e25.

[8] Teixeira JE, Mann BJ. Entameba histolytica induced dephosphorylation in hostcells. Infect Immun 2002;70:1816e23.

[9] Huston CD, Boettner DR, Miller-Sims V, Petri Jr WA. Apoptotic killing andhagocytosis of host cells by the parasite Entameba histolytica. Infect Immun2003;71:964e72.

[10] Singh D, Naik SR, Naik S. Role of cysteine proteinase of Entameba histolytica intarget cell death. Parasitology 2004;129:127e35.

[11] Sim S, Kim KA, Yong TS, Park SJ, Im KI, Shin MH, et al. Ultrastructuralobservation of human neutrophils during apoptotic cell death triggered byEntameba histolytica. Korean J Parasitol 2004;42:205e8.

[12] Yan L, Stanley Jr SL. Blockade of caspases inhibits amebic liver abscessformation in a mouse model of disease. Infect Immun 2001;69:7911e4.

[13] Becker SM, Cho KN, Guo X, Fendig K, Oosman MN, Whitehead R, et al.Epithelial cell apoptosis facilitates Entameba histolytica infection in the gut.Am J Pathol 2010;176:1316e22.

[14] Streuli CH, Gilmore AP. Adhesion-mediated signaling in the regulation ofmammary epithelial cell survival. J Mammary Gland Biol Neoplasia 1999;4:183e91.

[15] Gilmore AP. Anoikis. Cell Death Differ 2005;12:1473e7.[16] Schulte W, Scholze H. Action of the major protease from Entameba histolytica

on proteins of the extracellular matrix. J Protozool 1989;36:538e43.[17] Leippe M, Sievertsen HJ, Tannich E, Horstmann RD. Spontaneous release of

cysteine proteinases but not of pore-forming peptides by viable Entamebahistolytica. Parasitology 1995;111:569e74.

[18] Deschesnes RG, Patenaude A, Rousseau JLC, Fortin JS, Ricard C, Cote MF, et al.Microtubule-destabilizing agents induce focal adhesion structure disorgani-zation and anoikis in cancer cells. J Pharmacol Exp Ther 2007;320:853e64.

[19] Yuan M, Tomlinson V, Lara R, Holliday D, Chelala C, Harada T, et al. Yes-associated protein (YAP) functions as a tumor suppressor in breast. Cell DeathDiffer 2008;15:1752e9.

[20] Feingold C, Bracha R, Wexler A, Mirelman D. Isolation, purification, and partialcharacterization of an enterotoxin from extracts of Entameba histolyticatrophozoites. Infect Immun 1985;48:211e8.

[21] Frisch SM, Kristiana V, Keleita D, Sicks S. A role for jun N terminal kinase inanoikis; suppression by bcl2 and Crm A. J Cell Biol 1996;135:1377e82.

[22] Ravdin JI, Croft BY, Guerrant RL. Cytopathogenic mechanisms of Entamebahistolytica. J Exp Med 1980;152:377e90.

[23] Keene WE, Petitt MG, Allen S, McKerrow JH. The major neutral proteinase ofEntameba histolytica. J Exp Med 1986;163:536e49.

[24] Li AE, Ito H, Rovira II, Kim KS, Takeda K, Yu ZY, et al. A role for reactive oxygenspecies in endothelial cell anoikis. Circ Res 1999;85:304e10.

[25] Smets FN, Chen Y, Wang LJ, Soriano HE. Loss of cell anchorage triggersapoptosis (anoikis) in primary mouse hepatocytes. Mol Genet Metab 2002;75:344e52.

[27] Frisch SM, Francis H. Disruption of epithelial cell-matrix interactions inducesapoptosis. J Cell Biol 1994;124:619e26.

[28] Valentijn AJ, Gilmore AP. Translocation of Full-length bid to mitochondriaduring anoikis. J Biol Chem 2004;279:32848e57.

[29] Rossignol P, Ho-Tin-Noe B, Vranckx R, Bouton MC, Meilhac O, Lijnen HR, et al.Protease Nexin-1 inhibits plasminogen activation-induced apoptosis ofadherent cells. J Biol Chem 2004;279:10346e56.

[30] Ankri S, Stolarsky T, Bracha R, Padilla-Vaca F, Mirelman D. Antisense inhibitionof expression of cysteine proteinases affects Entameba histolytica-inducedformation of liver abscess in hamsters. Infect Immun 1999;67:421e2.

[31] Bansal D, Ave P, Kerneis S, Frileux P, Boché O, Baglin AC, et al. An ex-vivohuman intestinal model to study Entameba histolytica pathogenesis. PLoSNegl Trop Dis 2009;17:e551.

[32] Moncada D, Keller K, Ankri S, Mirelman D, Chadee K. Antisense inhibition ofEntameba histolytica cysteine proteases inhibits colonic mucus degradation.Gastroenterol 2006;130:721e30.

[33] Dickinson DP. Cysteine peptidases of mammals: their biological roles andpotential effects in the oral cavity and other in health and disease. Crit RevOral Biol Med 2002;13:238e75.

[34] Choy JC, Hung VH, Hunter AL, Cheung PK, Motyka B, Goping IS, et al. Gran-zyme B induces smooth muscle cell apoptosis in the absence of perforin;involvement of extracellular matrix degradation. Arterioscler Thromb VascBiol 2004;24:2245e50.

[35] Tanjoni R, Weinlich MS, Della-Casa PB, Clissa RF, Saldanha-Gama MS. Jar-arhagin; A snake venom metalloproteinase induces a specialized form ofapoptosis (anoikis) selective to endothelial cells. Apoptosis 2005;10:851e61.

[36] Hasmim M, Vassalli G, Alghisi GC, Bamat J, Ponsonnet L, Bieler G, et al.Expressed isolated integrin beta1 subunit cytodomain induces endothelial celldeath secondary to detachment. Thromb Haemost 2005;94:1060e70.

[37] Davis RJ. MAPKs: new JNK expands the group. TIBS (Trends Biochem Sci)1994;19:470e4.

[38] Derijard B, HIbi M, Wu I, Barrett T, Su B, Deng T, et al. JNK1; a protein kinasestimulated by UV light and H-Ras that binds and phosphorylated the c-junactivation domain. Cell 1994;76:1025e37.

[39] Ordonez C, Screaton RA, Ilantzis C, Stanners CP. Human carcinoembryonicantigen functions as a general inhibitor of anoikis. Cancer Res 2000;60:3419e24.

[40] Diamond LS, Harlow DR, Cunnick CC. A newmedium for the axenic cultivationof Entameba histolytica and other Entameba. Trans R Soc Trop Med Hyg 1978;72:431e2.

Further reading

Lugo-Martínez VH, Petit CS, Fouquet S, Beyec JL, Chambaz J, Pincorn-Raymond M,et al. Epidermal growth factor receptor is involved in enterocyte anoikis throughthe dismantling of E-cadherin-mediated junctions. Am J Physiol Gastrointest LiverPhysiol 2009;296:G235e44.