Autophagy Activated by Peroxiredoxin of Entamoeba histolytica

Autophagy, an evolutionarily conserved mechanism to remove redundant or dangerous cellular components, plays an important role in innate immunity and defense against pathogens, which, in turn, can regulate autophagy to establish infection within a host. However, for Entamoeba histolytica, an intestinal protozoan parasite causing human amoebic colitis, the interaction with the host cell autophagy mechanism has not been investigated. In this study, we found that E. histolytica peroxiredoxin (Prx), an antioxidant enzyme critical for parasite survival during the invasion of host tissues, could activate autophagy in macrophages. The formation of autophagosomes in macrophages treated with recombinant Prx of E. histolytica for 24 h was revealed by immunofluorescence and immunoblotting in RAW264.7 cells and in mice. Prx was cytotoxic for RAW264.7 macrophages after 48-h treatment, which was partly attributed to autophagy-dependent cell death. RNA interference experiments revealed that Prx induced autophagy mostly through the toll-like receptor 4 (TLR4)–TIR domain-containing adaptor-inducing interferon (TRIF) pathway. The C-terminal part of Prx comprising 100 amino acids was the key functional domain to activate autophagy. These results indicated that Prx of E. histolytica could induce autophagy and cytotoxic effects in macrophages, revealing a new pathogenic mechanism activated by E. histolytica in host cells.


Introduction
Entamoeba histolytica is a protozoan parasite that causes human amoebic colitis and amoebic liver abscess (ALA); it infects 50 million people annually, causing 40,000-100,000 deaths [1]. People are infected by ingesting food and water contaminated by amoeba cysts. Approximately 90% of infected individuals are asymptomatic but in 10%, amoeba trophozoites, driven by unknown stimuli, can penetrate the mucosal barrier of the colon and invade the intestinal lamina propria, leading to amoebic diarrhea and colitis, or even ALA if trophozoites disseminate through the portal circulation [2]. Peroxiredoxins (Prxs) are an evolutionarily conserved family of ubiquitously expressed antioxidant enzymes, which can effectively reduce peroxides, including hydrogen peroxide (H 2 O 2 ) and peroxynitrite (ONOO − ) [3]. As a facultative anaerobic organism, E. histolytica requires high amounts of Prx to resist oxidative damage during invasion of host tissues and organs [4,5]. It was shown that ALA was alleviated by Prx downregulation [6], which implicates Prx in the pathogenesis of E. histolytica infection. There are more than 20 different transcripts of Prx in E. histolytica. In a previous study, we cloned and expressed a Prx of E. histolytica (XP_648522.1), which represents a typical 2-Cys Prx containing two catalytic cysteine residues in the active sites [7]. Considering that human Prx-1, which is also a typical 2-cys-Prx, has been shown to be secreted and to bind toll-like receptor 4 (TLR4) on macrophages to promote the inflammatory response [8], it is important to find out whether the Prx of E. histolytica can act as a pathogen-associated molecular pattern (PAMP) motif.
Autophagy is a housekeeping process necessary to remove damaged or redundant cellular components; it is executed through the function of autophagosomes and lysosomes and is essential for the maintenance of the metabolic balance in eukaryotic cells during nutritional restriction, infection, or other physiological/pathological conditions [9]. There are three forms of autophagy: microautophagy, chaperone-mediated autophagy, and macroautophagy [10]. Macroautophagy (hereinafter referred to as autophagy) is an evolutionarily conserved process involved in the stress response, during which it eliminates redundant or potentially dangerous cytosolic entities such as damaged mitochondria or invading pathogens [10,11]. Furthermore, it has been found that autophagy plays an important role in innate immunity, a process activated by TLRs and other pattern recognition receptors to eliminate pathogens [12].
Several protozoan parasites have been shown to interact with the autophagy machinery in host cells [9,13]. Thus, it has been reported that the number of light chain 3 (LC3)-positive phagophores in host cells gradually increases after infection with Trypanosoma cruzi, indicating a specific molecular exchange between parasites and host cells [14]. In the process of Leishmania donovani infection, macrophage autophagy was inhibited at the early stage but was activated 24 h after invasion, which proved to be beneficial for parasite survival [15]. However, the relationship between E. histolytica and macrophage autophagy has rarely been reported [16].
The aim of this study was to investigate whether E. histolytica Prx (Eh-Prx) is involved in the regulation of autophagy in macrophages. The results indicate that the recombinant protein (Eh-rPrx) could induce macrophage autophagy accompanied by autophagy-dependent cell death (ADCD) and that the C-terminal part of Eh-rPrx containing 100 residues is the principal domain responsible for autophagy activation. This study reveals, for the first time, that Eh-Prx is capable of inducing autophagy in immune cells, which is important for further understanding of E. histolytica pathogenesis and the role of autophagy in infection.

Ethics Statement
All animal experiments in our study were conducted in strict accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals (1988.11.1) and were approved by the Institutional Animal Care and Use Committee (IACUC) (Permit Numbers: 20160225-097). All efforts were made in our study to minimize the suffering of animals.

Expression of Recombinant Proteins
Trophozoites of E. histolytica in the logarithmic growth phase were collected. Total RNA was obtained from trophozoites by the RNeasy Maxi Kit (QIAGEN, Hilden, Germany, #75162).
Complementary DNA was then acquired from the total RNA above by the PrimeScript™ 1st Strand cDNA Synthesis Kit (TaKaRa, Kusatsu, Japan, # 6110A) and used as a template to generate Prx fragments by PCR. Whole-length Prx (XP_648522.1) was amplified using the following primers: sense-CCCATATGTCTTGCAATCAACAAAAAGAGT and antisense-CCGGATCCTTTTAATGTGCTGTTAAATATT. Three fragments, each encoding a 100-residue part of Prx, were obtained using the following primer pairs: N-terminal sense-CCCATATGTCTTGCAATCAACAAAAAGAGT and antisense-CCGGATCCTTTTATTGTCCTGCAAGTTCACTAT; middle sense-CCCATATGAAGTTGACATTCCCATTAGTATCA and antisense-CCGGATCCTTTTAATGTGCTGTTAAATATT; C-terminal sense-CCCATATGGGAAAATATGTTGTATTGTTGTTT and antisense-CCGGATCCTTTTAATCATCAATGATGACATATC. The protocol for plasmid construction and protein expression was published previously [7]. His-Bind Resin (Novagen, San Diego, CA, USA, #69670) was used to purify the recombinant proteins, which were dialyzed against Tris-HCl (20 mM, pH 8.0), followed by filtration via a 0.22-µm membrane, and tested for endotoxin levels using the Toxin Sensor TM Gel Clot Endotoxin Assay Kit (Genscript, Nanjing, China, L00351). The endotoxin levels of the recombinant proteins correspond to the national standard of the People's Republic of China for medical products (GB/t14233.2-2005). After counterstaining with 0.5 µg/mL 4 ,6-diamidino-2-phenylindole (DAPI), cells were preserved at 4 • C in a dark place temporarily or immediately observed under a laser confocal microscope (SP8, Leica, Wetzlar, Germany).

Construction of a Mouse Model of Peritonitis
Female C57 BL/6 mice (10-12 weeks old) were purchased from Shanghai Slake Laboratory Animal Company. Peritonitis was induced by intraperitoneal injection of Eh-rPrx (100 µg) in 500 µL Tris-HCl (pH 8.0); the control group received the same volume of Tris-HCl (pH 8.0). After 12 h of treatment, mice were injected with excess of 1% sodium pentobarbital and blood was collected from the heart. Mice were then intraperitoneally injected with 5 mL of cold PBS and massaged gently for 5 min and peritoneal cells were then collected as peritoneal macrophages.

Analysis of Cell Morphology by Differential Interference Contrast Microscopy
RAW264.7 cells were collected in the logarithmic growth phase and placed in 4-chamber glass bottom 35-mm dishes (In Vitro Scientific, Sunnyvale, CA, USA, D35C4-20-1-N) with 2.5 × 10 5 cells/chamber. After the cells adhered and fully stretched, serum-free medium containing Eh-rPrx was added and real-time changes in cell morphology were immediately observed under a differential interference microscope (Carl Zeiss Meditec, Jena, Germany).

Quantification of Autophagy by High-Content Screening Analysis
RAW264.7 cells were harvested in the logarithmic growth phase and placed in 96-well plates (10 5 cells/well). RAW264.7 cells were pre-treated or not with TLR4 inhibitor TAK242 (1 µM; APExBIO, A3850), incubated with recombinant proteins (5 µg/mL) for 24 h, and stained with monodansylcadaverine (MDC) (100 µM; Sigma-Aldrich, #30432) at 37 • C in the dark for 10 min. The cells were washed gently with warm Ca/Mg-containing Hank's Balanced Salt Solution (Gibco, #14025-092) and counterstained with Hoechst 33342 (1 µM) at 37 • C in the dark for 5 min. After repeated washing with the same buffer, cells kept in warm buffer were immediately observed and quantified using the Operetta High-content Screening system (Perkin-Elmer, Hopkinton, MA, USA).

Statistical Analysis
The data are presented as the mean ± standard error of the mean (SEM). GraphPad Prism 5 (GraphPad Software, Version 5.01, USA) was used for statistical analysis. The difference between treatment groups was analyzed by Student's t-test and a p-value < 0.05 was considered statistically significant.

Autophagosome Formation Was Induced by Eh-rPrx
Maturation of LC3 to the conjugated LC3-II form at the expansion stage of the phagophore is one of the markers of autophagosome formation [10]. After 24-h treatment with Eh-rPrx, aggregation of large LC3 particles in RAW264.7 cells could be observed by laser confocal microscopy, indicating the formation of autophagosomes ( Figure 1A). In order to investigate the function of Eh-rPrx in vivo, a mouse peritonitis model was established by intraperitoneal injection of Eh-rPrx. Western blotting analysis of peritoneal macrophages after 12 h revealed that the ratio of LC3-II to LC3-I in the Eh-rPrx group was significantly increased compared with the control group (p < 0.05, Figure 1B), thus confirming autophagosome formation. These results suggested that Eh-rPrx could induce autophagy in macrophages. Mice were intraperitoneally injected with Eh-rPrx (100 µg) and peritoneal macrophages were analyzed after 12 h for LC3 expression by Western blotting. A representative image is shown. The data are expressed as the mean ± standard error of the mean (SEM) (n = 6 or 11); ** p < 0.01 by Student's t-test.

Eh-rPrx Activated Autophagy through the TLR4-TRIF Pathway
Quantitative assessment of Eh-rPrx-induced mean fluorescence intensity (MFI) of MDC by high-content screening analysis did not reveal any difference between the treatment and control groups; however, intracellular aggregation of MDC-labelled vesicles was observed in the treatment group ( Figure 3A). Measurement of the fluorescent spot area revealed that it was larger in the treatment group than in the control group, suggesting the formation of autophagosomes, which could be significantly antagonized by a TLR4 inhibitor (p < 0.001) ( Figure 3A). To further verify the molecular mechanisms of autophagy induction by Eh-rPrx, we performed RNA interference experiments using TLR4-, MyD88-, and TRIF-specific siRNAs. The results showed that autophagy was activated by Eh-rPrx, mostly through the TLR4-TRIF pathway; the TLR4-MyD88 pathway implicated in NLR family pyrin domain containing 3 (NLRP3) inflammasome activation [18] could also play a role but the effect was not statistically significant ( Figure 3B). Eh-rPrx activated autophagy via the toll-like receptor 4 (TLR4)-TRIF signaling pathway. (A) Eh-rPrx induced autophagosome formation through TLR4. RAW264.7 cells were pre-treated with the TLR4 inhibitor TAK242, incubated with Eh-rPrx for 24 h, and stained with monodansylcadaverine (MDC) (green) and Hoechst 33342 (blue). Tris-HCl (pH 8.0) was used for the negative control. The spot area of MDC fluorescence was quantitatively assessed by high-content screening analysis (n = 5). Scale bar: 100 µm. (B) Autophagy induction by Eh-rPrx occurred through the TLR4-TRIF signaling pathway. RAW264.7 cells were pre-treated with TLR4-, MyD88-, or TRIF-specific siRNAs for 24 h, treated with Eh-rPrx (5 µg/mL) for 24 h, and analyzed for the expression of Beclin-1 and LC3 by Western blotting (n = 3). Tris-HCl (pH 8.0) was used for the negative control. The data are expressed as the mean ± SEM; * p < 0.05 and *** p < 0.001 by Student's t-test; ns, not significant. The Eh-rPrx C-terminal was the principal domain responsible for autophagy activation. (A) The C-terminal of Eh-rPrx induced autophagosome formation. RAW264.7 cells were incubated with 5 µg/mL of full-length Eh-rPrx or its N-terminal, middle, or C-terminal fragments for 24 h and then stained with MDC (green) and Hoechst 33342 (blue). The spot area of MDC fluorescence was quantified using high-content screening analysis (n = 5). Scale bar: 100 µm. (B) Eh-rPrx C-terminal fragment activated autophagy. RAW264.7 cells were treated as in (A) and analyzed for the expression of Beclin-1 and LC3 by Western blotting (n = 3). (C) Autophagy activation by Eh-rPrx was inhibited by mAb 4G6. RAW264.7 cells were pre-treated with different concentrations of mAb 4G6, incubated with Eh-rPrx (5 µg/mL) for 24 h, and analyzed for the protein expression of Beclin-1 and LC3 by Western blotting (n = 3). (D) Eh-rPrx C-terminal fragment contained the recognition site for mAb 4G6. The reaction of full-length Eh-rPrx or its N-terminal, middle, or C-terminal fragments with mAb 4G6 was analyzed by ELISA. The data are expressed as the mean ± SEM; * p < 0.05 and ** p < 0.01 by Student's t-test; ns, not significant.

Eh-rPrx Activated Autophagy through Its C-Terminal Domain
Our previous results showed that the N-terminal of Eh-rPrx (XP_648522.1) was cysteine-rich and was longer than that of non-pathogenic Entamoeba moshkovskii Prx [7]. To determine whether the N-terminal of Eh-rPrx was required for autophagy activation, we cloned, expressed, and purified three recombinant fragments of Eh-rPrx, each comprising 100 residues of the N-terminal, middle, and C-terminal domains, respectively. Among them, the Eh-rPrx C-terminal fragment demonstrated a stronger effect on autophagy induction ( Figure 4A,B). In addition, Eh-rPrx-activated autophagy was reduced by mAb 4G6 (Figure 4C), whose recognition site was confirmed by ELISA to be located at the Eh-rPrx C-terminal ( Figure 4D). These results suggested that the 100-residue C-terminal domain of Eh-rPrx was crucial for autophagy activation.

Discussion
Host innate immunity acts against E. histolytica infection as a double-edged sword. Inducible nitric oxide synthase (iNOS) plays an important role in macrophage-mediated killing of E. histolytica and mice lacking iNOS are prone to ALA and hepatocyte apoptosis [2]. However, the formation of ALA is not directly caused by amoebae but is rather related to the innate immune response of the host [19]. Therefore, it is important to study the interaction between E. histolytica and macrophages.
Prx can be abundantly produced following parasitic invasion of tissues and organs [20]. The interaction between Prxs of various parasites and macrophages has been reported in several studies. Thus, the Prx of Plasmodium berghei could act as a PAMP binding to TLR4 on the macrophage surface, which promoted inflammation [21], whereas the Prx of Schistosoma mansoni and Fasciola hepatica activated Th2 cells and macrophages [22]. In this study, we showed that the Prx of E. histolytica could also act as a PAMP that activated macrophage autophagy through the TLR4-TRIF pathway.
Microtubule-associated protein 1A/1B-LC3 is a key biomarker of autophagy in mammalian cells. During autophagy, ATG4, ATG7, and ATG3 cooperate to cleave the precursor of the LC3-like protein into a mature form [10]. Cytosolic LC3-I is conjugated with phosphatidylethanolamine to form LC3-II, which is recruited to and integrated into the autophagosome membrane; autophagosomes then fuse with lysosomes to form autolysosomes [23]. There is no "gold standard" for measuring autophagy [24,25], which is traditionally evaluated based on LC3 expression determined by immunoblotting and fluorescence microscopy [26]. Immunoblotting is used to detect the transformation of LC3 from Type I to Type II, which is related to the number of autophagosomes [27], whereas immunofluorescence can identify cells undergoing autophagy by labeling LC3 spots [23]. In this study, we observed that treatment with Eh-rPrx stimulated the aggregation of LC3 spots in RAW264.7 cells, indicating the formation of autophagosomes ( Figure 1A), and significantly increased the LC3-II/LC3-I ratio in mouse peritoneal macrophages, suggesting the upregulation of autophagosome production in vivo ( Figure 1B). The complex containing Beclin-1 and VPS34, VPS15, AMBRA1, and/or UVRAG has Class III phosphoinositide 3-kinase activity, which can produce phosphatidylinositol (3,4,5)-trisphosphate to trigger autophagosome nucleation [10]. Therefore, some studies have shown that activation of autophagy is accompanied by the upregulation of Beclin-1 [28,29]. Consistent with these results, we also observed an increase in Beclin-1 expression in RAW264.7 cells after treatment with Eh-rPrx ( Figure 3B).
Cell death can be classified into three types based on the molecular mechanism and cell morphology: Type 1 or apoptosis, Type 2 or autophagy, and Type 3 or necrosis [30]. Among these, autophagy is characterized by extensive intracellular vacuolization [30], which is consistent with our results (Figure 2A). Some studies have found that the loss of Bcl-2 (a Beclin-1 inhibitor) can lead to excessive autophagy, resulting in cell death [31]. Although cell death is usually accompanied by autophagy, ADCD can be considered only if it is suppressed by autophagy inhibitors (such as 3-methyladenine and wortmannin) or genetic ablation of essential autophagy genes [32]. In our study, the reduced viability of Eh-rPrx-treated RAW264.7 macrophages could be partly attributed to ADCD, as it was sensitive to wortmannin ( Figure 2B).
It has been shown that autophagy is the downstream effect of TLR signaling [12]. In our study, the TLR4 inhibitor TAK242 significantly reduced the area of autophagy spots, indicating that Eh-rPrx is a TLR4 ligand ( Figure 3A), which is consistent with similar results for P. berghei Prx [21] and human Prx-1 [8]. Autophagy induced by Eh-rPrx was shown to occur mainly through the TLR4-TRIF pathway ( Figure 3B). The TLR4-MyD88 pathway, which is involved in activation of the NLRP3 inflammasome related to autophagy [18], may also contribute to Eh-rPrx-dependent stimulation of autophagy in macrophages, although the results did not reach statistical significance. The 100-residue C-terminal domain of Eh-rPrx was the key site responsible for Eh-rPrx binding to macrophage TLR4 and appeared to be also the recognition site of the Prx-specific monoclonal antibody (Figure 4).
In conclusion, this is the first study to show that Prx of E. histolytica activates autophagy in macrophages through the TLR4-TRIF pathway. The pathogenicity of E. histolytica is known to depend on its contact with macrophages; however, the results of this study show that E. histolytica is capable of inducing autophagy and cell death in a parasite-macrophage contact-independent manner through secretion of Prx. Our study provides further insights into the molecular mechanism underlying E. histolytica pathogenicity, which could aid in the identification of potential drug targets in the amoeba.

Conflicts of Interest:
The authors declare no conflict of interest.