The pathogenic responses elicited during exposure of human intestinal cell line with Giardia duodenalis excretory-secretory products and the potential attributed endocytosis mechanism

Abstract

Giardia duodenalis, an important zoonotic protozoan parasite, adheres to host intestinal epithelial cells (IECs) via the ventral disc and causes giardiasis characterized mainly by diarrhea. To date, it remains elusive how excretory-secretory products (ESPs) of Giardia enter IECs and how the cells respond to the entry. Herein, we initially demonstrated that ESPs evoked IEC endocytosis in vitro. We indicated that ESPs contributed vitally in triggering intrinsic apoptosis, pro-inflammatory responses, tight junction (TJ) protein expressional changes, and autophagy in IECs. Endocytosis was further proven to be implicated in those ESPs-triggered IEC responses. Ten predicted virulent excretory-secretory proteins of G. duodenalis were investigated for their capability to activate clathrin/caveolin-mediated endocytosis (CME/CavME) in IECs. Pyridoxamine 5'-phosphate oxidase (PNPO) was confirmed to be an important contributor. PNPO was subsequently verified as a vital promoter in the induction of giardiasis-related IEC apoptosis, inflammation, and TJ protein downregulation. Most importantly, this process seemed to be involved majorly in PNPO-evoked CME pathway, rather than CavME. Collectively, this study identified Giardia ESPs, notably PNPO, as potentially important pathogenic factors during noninvasive infection. It was also noteworthy that ESPs-evoked endocytosis might play a role in triggering giardiasis-inducing cellular regulation. These findings would deepen our understanding about the role of ESPs, notably PNPO, in the pathogenesis of giardiasis and the potential attributed endocytosis mechanism.

Introduction

Giardia duodenalis is a well-known flagellated diarrhea-causing protozoan parasite infecting humans and a great variety of nonhuman animals such as livestock, wildlife, birds, and among others [1]. The parasite is transmitted predominantly by fecal-oral route through contaminated food and water; G. duodenalis infection is one of the most common causes of diarrhea and thus is of socioeconomic importance across the world [2]. The life cycle of Giardia has two distinct stages: a disease-causing vegetative stage, trophozoite, and an environmentally resistant and infective stage, cyst. After cyst ingestion and passage through stomach, excystation occurs in the proximal small intestine. Giardia trophozoites colonize the duodenum and replicate via asexual binary fission [3]. Giardia induces an illness known as giardiasis characterized by diarrhea, abdominal cramps, bloating, weight loss, and malabsorption [4]. Thus far, the pathogenesis of giardiasis and the related regulatory network remain unknown, which seem to be related to host cell apoptosis, inflammation, pyroptosis, nitric oxide (NO) decrease, tight junction (TJ) disruption, and barrier dysfunction [5–12]. It is imperative to elucidate the mechanisms and the putative trigger or virulence factors that are behind giardiasis development.

Endocytosis occurs at the cell surface and involves internalization of the plasma membrane (PM) and the following vesicle separation, which is essential for the transport of various cargo molecules (mainly transmembrane proteins and their extracellular ligands) from the extracellular space into the cytoplasm [13]. These cargoes are associated with an assortment of physiological processes such as nutrient uptake, cell signaling, and developmental regulation [13]. There are two broad categories of endocytosis: phagocytosis typically specific to mammalian immune cells and pinocytosis commonly present in most cell types. Pinocytosis occurs by at least four basic mechanisms that differ in the composition of the coat, the size of the detached vesicles, and the fate of internalized particles: macropinocytosis, clathrin/caveolin-independent endocytosis, and clathrin- and caveolin-mediated endocytosis (CME/CavME) [14]. CME, the main endocytic pathway in mammalian cells, functions vitally in the uptake of transmembrane receptors and transporters, remodeling of PM composition in response to environmental changes, and regulation of cell surface signaling [15]. Caveolin-1, one of the three mammalian caveolin proteins enriched in caveolae (an important component of membrane invaginations with an enrichment of signaling molecules and membrane transporters), is believed to be necessary for caveolar biogenesis [16]. Several obligate intracellular protozoan parasites are known to internalize host cells through endocytic pathways, such as Leishmania donovani [17], Toxoplasma gondii [18], and Trypanosoma cruzi [19, 20]. The invasion into and internalization by host cells are potentially important steps in the evolution of parasitism, which presents at least two advantages: protection against the host immune defenses and access to a microenvironment rich in metabolic products [21]. The excretory-secretory products (ESPs) of some important helminths like Opisthorchis viverrini [22] and Echinococcus granulosus [23] are able to utilize multiple mammalian cell endocytic pathways to become internalized. Endocytosis is also inseparably linked with some signal transduction processes such as apoptosis [24], inflammation [25], TJ destruction [26], and autophagy [27], among others. For instance, the ESPs of O. viverrini endocytosed by human cholangiocytes are capable of inducing cell proliferation and IL6 production [22]. In spite of those advances, such information is still lacking for noninvasive parasite, Giardia.

It has been revealed that Giardia releases ESPs that can be bound to intestinal epithelial cells (IECs) and partially internalized, which have been predicted as an integral part of Giardia virulence factors responsible for the induction of giardiasis [28]. Nevertheless, the mechanistic association between ESPs and giardiasis is still far away from being fully understood, especially the endocytic pathways by which ESPs internalize into IECs and their potential impacts on cellular regulation. Among Giardia-secreted proteins, pyridoxamine 5'-phosphate oxidase (PNPO) is most abundant and is also significantly upregulated during Giardia-IEC interactions, followed by tenascins (Tena), an extracellular nuclease, and cathepsin B (GCATB) [29, 30]. PNPO has been identified as a flavin mononucleotide-dependent enzyme associated with the fix of molecular oxygen [31]. It is still uncertain whether Giardia-induced ROS-mediated intrinsic apoptosis [9] gets involved in the internalization of ESPs into IECs. The interactions between Giardia and macrophages cause an upregulation of pro-inflammatory cytokines such as TNF-α, IL-6, IL-18 and IL-1β and trigger ROS-dependent NLRP3-mediated pyroptosis [8, 32], which may be attributed to internalization of Giardia extracellular vesicles (EVs) into macrophages and regulation of host innate immunity [33]. However, IL-10, an anti-inflammatory regulator vital for intestinal homeostasis, appears to be effective in coordinating anti-Giardia host defense [34]. In spite of this, it remains to be elucidated whether the Giardia-triggered pro-inflammatory responses are also associated with ESP endocytosis in IECs, which represent not only immune cells but also non-immune cells directly responding to external agents. While being a noninvasive protozoan, Giardia not only induces IEC apoptosis as noted earlier, but also disrupts TJ integrity and enhances intestinal permeability [35]. CME is known in epithelial cells as a major mechanism for TJ complex dissociation, and CavME in endothelial cells likewise [26]. Several Giardia-secreted cysteine proteases, the important component of ESPs, are characterized as virulence factors responsible for TJ disruption and immune modulation [7], while it remains unexplored whether this process is implicated in IEC endocytic mechanisms. Numerous membrane sources involve autophagosome formation as described [36]. Endocytosis is critical for autophagosome formation by membrane vesicle transport of autophagy mediators [27], while the correlation between Giardia ESP internalization and IEC autophagy is not known.

This study aims to elucidate the potential effects of ESPs, notably PNPO, on giardiasis-related IEC responses such as apoptosis, inflammation, and TJ disruption, as well as to reveal the potential underlying transport mechanisms by which those pathogenic responses were triggered.

Materials and methods

Cell culture

The human colon adenocarcinoma cell line Caco-2 that closely resembles normal human small IECs [37, 38], was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were maintained in high-glucose DMEM (Hyclone, Logan, USA) supplemented with 10% FBS, 1% MEM NEAA, 1% GlutaMAX, and 1% penicillin/streptomycin. Cells were incubated at 37 °C in an atmosphere of 5% CO₂, and subcultured every three to four days at 80 to 90% confluence prior to being seeded. Medium was changed every other day. Cells were seeded in 6-well (1 × 10⁶ cells/well), 12-well (5 × 10⁵ cells/well), 24-well (2 × 10⁵ cells/well), and 96-well (1 × 10⁴ cells/well) plates. All experiments were performed two to three days post-seeding when the cell culture was 80% to 90% confluent, with the exception of those regarding assessment of TJ protein levels which were performed two to three days post-confluence on fully differentiated, confluent monolayers.

Parasite culture

G. duodenalis WB isolate (ATCC 30957, Manassas, USA) typed as assemblage A was used in the present study. Parasites were grown in filter sterilized modified TYI-S-33 medium with 10% FBS and 0.1% bovine bile supplemented with 0.1% gentamycin and 1% penicillin/streptomycin at 37 °C in microaerophilic conditions and passaged when ~80% confluent [39].

ESP isolation and exposure

Cultures grown to log phase were harvested by chilling on ice for 15 min. Detached trophozoites were centrifuged at 2000 rpm for 10 min at room temperature (RT) and washed three times with PBS. The remaining sediment was suspended in Hanks saline solution and incubated for 2 h at 37 °C, at mild agitation. Parasites were then centrifuged at 3000 rpm for 5 min at 4 °C, and then the supernatant was collected and again centrifuged at 5000 rpm for 10 min at 4°C. The final supernatants were decanted, filtered through 0.22 μm membranes, and used as ESPs. Protein concentration (ESPs) was quantified by an enhanced BCA Protein Assay Kit (Beyotime, Shanghai, China). Prior to exposure, an endotoxin ELISA kit (Meimian Biotech, Yancheng, China) was applied to ensure that the ESP solution was free of endotoxins that would influence the observed results. Caco-2 cells were exposed to ESPs at the concentrations and for the time periods as indicated. In time-course experiments, ESPs were added at different time points and cells harvested together. Before further analysis, cells were rinsed thrice with PBS to remove non-interacted ESPs.

qPCR analysis

Total RNA was isolated from the harvested cells cultured in 12-well plates using Trizol total RNA isolation reagent (Invitrogen, Carlsbad, USA). cDNA was synthesized with the HiScript II 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). qPCR was performed using the SYBR-Green PCR Master Mix Kit (Vazyme, Nanjing, China) on an LC480 Lightcycler system (Roche, Indianapolis, USA). Primer pairs used in qPCR analysis can be found in Additional file 1. The 2^−ΔΔCt method was used to calculate relative changes in mRNA expression.

Western blot analysis

Total cellular proteins were extracted from the harvested cells cultured in 6-well plates using the commercial RIPA lysis buffer (Beyotime, Shanghai, China) supplemented with 1% PMSF (Beyotime, Shanghai, China). Protein concentration was quantified by the BCA assay mentioned earlier. Protein expression levels were assessed by western blot analysis. In brief, extracted proteins were separated on 12% SDS-PAGE gels and transferred to PVDF membranes. Membranes were blocked with 5% skim milk in PBST for 2 h at RT, and incubated with the appropriate primary antibodies (1:1000 dilution in PBST) against clathrin, caveolin-1, β-actin, pro-/cl-caspase (CASP)-3, pro-/cl-CASP-9, pro-/cl-PARP, Bax, Bcl-2, IL-18, pro-/mature IL-1β, TNF-α, NLRP3, claudin-1, claudin-4, occludin, ZO-1, LC3 and p62 overnight at 4 °C. The primary antibodies were obtained from four commercial sources (Affinity Biosciences, Changzhou, China; ABclonal, Wuhan, China; ABMART, Shanghai, China; Bioss, Beijing, China). After five washes with PBST, the membranes were probed with HRP-conjugated secondary antibody (1:5000 dilution in PBST; ABMART, Shanghai, China) for 1 h at RT. Proteins were visualized by chemiluminescence (Syngene, Cambridge, UK). The intensity of the protein bands was quantified by NIH Image J software (NIH, Bethesda, USA).

Preparation of anti-ESP polyclonal antibody (PAB)

PAB specific to Giardia ESPs was raised according to the following procedures. New Zealand white rabbits were subcutaneously immunized three times at 2-week interval, with the first immunization performed using 0.2 to 0.5 mg of ESPs in 0.5 mL of Freund’s complete adjuvant and the subsequent immunizations using 0.2 to 0.5 mg of ESPs in 0.5 mL of Freund’s incomplete adjuvant. On day 0, a serum sample was collected from the rabbit before immunization and this pre-immunized serum was used as a negative control. Antisera were recovered from blood obtained by cardiac exsanguinations on day 52. Binding specificity of anti-ESP PAB was tested by ELISA as shown in Additional file 2.

Immunofluorescence assay

Cells grown on cover slips in 24-well plates were fixed with 4% paraformaldehyde in PBS for 30 min at RT and permeabilized with 0.25% Triton X-100 in PBS for 10 min at RT. Nonspecific binding sites were blocked by incubation in 2% BSA in PBS for 30 min at RT. Cells were incubated with the obtained anti-ESP PAB in PBST at a concentration of 5 µg/mL overnight at 4 °C and then FITC-AffiniPure goat anti-rabbit IgG (H + L) (1:200 dilution; Jackson, West Grove, USA) in the dark for 1 h at 37 °C. Cell nucleus was stained with DAPI (1 μg/mL; Alphabio, Tianjin, China). Fluorescent image was taken by a Lionheart FX Automated Microscope (BioTek, Winooski, USA).

Co-immunoprecipitation (co-IP) analysis

Cells cultured in 6-well plates were treated with the lysis buffer as noted earlier. Cell lysates were incubated with protein A/G magnetic beads (Bimake, Houston, USA) that bind the obtained anti-ESP PAB overnight at 4 °C. The immunoprecipitates were collected using a magnetic rack, and the sediments were resuspended in lysis buffer and boiled for 8 min. The eluted proteins were analyzed by western blotting with the anti-clathrin and -caveolin-1 antibodies (both diluted 1:1000 in PBST).

Acridine orange (AO)/ethidium bromide (EB) staining

Cells grown on cover slips in 24-well plates were evaluated for ESPs/PNPO-induced apoptosis. Cell apoptosis was detected by dual staining with combined fluorescent dyes AO and EB (BestBio, Shanghai, China) and observed using a Lionheart FX Automated Microscope. With this dye combination, apoptotic cells were shown in orange and viable cells in green.

Protein inhibition

We used dynamin inhibitor dynasore (100 μM in use; Abmole, Houston, USA), clathrin inhibitor chlorpromazine (CPZ) (10 μM; Abmole, Houston, USA), and caveolin-1 inhibitor genistein (100 μM; Selleck, Shanghai, China) in inhibition analyses. All inhibitors were dissolved in 0.1% DMSO, applied 1 h, and washed with PBS thrice for drug removal before exposure.

Recombinant protein and exposure

Ten Giardia excretory-secretory (ES) proteins were investigated for their potential to activate endocytosis in Caco-2 cells, including PNPO (GL50803_5810), peptidyl-prolyl cis-trans isomerase B (PPIB) (GL50803_17163) and Tena-2/3 (GL50803_10330/16833) as before described [8], and GCATB-1/2/3/4/6/8 (GL50803_17516/16468/14019/16779/16160/29304). The latter six proteins were expressed in a prokaryotic expression system similar to the formers as described [8]. In brief, the amplified genes encoding GCATB-1/2/3/4/6/8 were cloned into pCold I vector (TaKaRa, Ohtsu, Japan). The resulting plasmids were transformed into Escherichia coli strain BL21 (DE3) cells. Protein expression was induced with 1 mM isopropyl β-d-thiogalactoside (Solarbio, Beijing, China) for 20 h at 16 °C. Following induction, cells were harvested and lysed. The precipitated proteins and supernatants were subjected to SDS-PAGE analysis. The recombinant His6-tagged PNPO and PPIB were expressed as soluble proteins in E. coli, while the expressed tenascins and GCATB-1/2/3/4/6/8 were present in the form of inclusion bodies (Additional files 3 and 4). The inclusion bodies were then isolated and subjected to denaturation and renaturation. Before exposure, all the ten recombinant proteins were purified with a HisTrap HP nickel column (SMART, Changzhou, China). Before cell treatment, the concentration of the purified proteins in PBS was quantified using the BCA assay and the protein solution was tested to be free of endotoxins using the assay mentioned earlier. Cells were exposed to the recombinant proteins at the concentrations and for the time periods as indicated. In time-course experiments, proteins were added at different time points and cells harvested together. Before further analysis, non-interacted proteins were removed by PBS washes.

Statistical analysis

Statistical analyses were conducted using the GraphPad Prism 7.0 program. Data from triplicate wells (or more) from a representative of at least three independent experiments are presented as means ± SD. The statistical significance of the differences was assessed using Student’s t-test and one-way ANOVA. P-values < 0.05 were considered statistically significant (* p < 0.05, ** p < 0.01).

Results

Giardia ESP exposure involved the endocytic activation

It is well known that CME and CavME are the two major endocytic pathways regulated by dynamin [16], while it is quite difficult to explore the links between ESP exposure and activation of endocytic pathways due to the complexity of ESP component. Therefore, for this part of the study, we explored the involvement of the multiprotein complexes in endocytosis initiation, not the detailed endocytic pathways. We first screened for the appropriate ESP concentration used for cell challenge. As shown in Figure 1A, the mRNA levels of dynamin, clathrin and caveolin-1 were significantly increased upon exposure of IECs with Giardia ESPs at a concentration of 20 μg/mL. The same concentration was used for ESP or rESP exposure hereinafter. The elevated levels of clathrin and caveolin-1 were also observed in western blot analysis (Figure 1B). The Giardia ESPs were shown to be bonded to Caco-2 cells at 3 h after exposure (Figure 1C), and co-IP analysis revealed binding between ESPs and clathrin/caveolin-1 (Figure 1D), implying the involvement of ESPs in endocytic activation.

Figure 1. Giardia ESP exposure involved the endocytic activation.

Unless otherwise specified, Caco-2 cells were exposed to Giardia ESPs for the indicated time periods. (A) Upon exposure of IECs with the amount of ESPs of 20, 40 and 100 μg/mL, the mRNA levels of dynamin, clathrin and caveolin-1 were assessed by qPCR analysis. The comparisons were made with the first group of the graph. (B-D) Cells were exposed to ESPs at the concentration of 20 μg/mL. (B) The expression levels of clathrin and caveolin-1 were determined by western blot analysis. (C) Anti-ESP PAB was used to indicate the bonding between ESPs and IECs after a 3-h exposure (scale bar = 20 μm). (D) Binding between ESPs and clathrin/caveolin-1 was assessed by co-IP analysis. All experiments were repeated four times. Data from triplicate wells from a representative of four independent experiments are presented as means ± SD. * p < 0.05, ** p < 0.01.

Giardia ESPs triggered varying cellular responses of IECs

The role of ESPs from parasites in modulating cellular responses and balancing host immunity has been well documented over the years [40], while this is not the case for Giardia. Here we explored how IECs respond to Giardia ESP exposure. Giardia ESP exposure was shown to induce Caco-2 cell apoptosis via intrinsic pathway by increasing the ratio of Bax to Bcl-2 and the levels of cleaved apoptotic initiator CASP-9 and effector CASP-3 in a time-dependent fashion (Figure 2A and 2B). This was confirmed by AO/EB double staining, with increasing numbers of apoptotic cells stained orange seen following ESP exposure (Figure 2C). It is therefore inferred that the IEC mitochondrial apoptosis activated by Giardia trophozoite exposure [9] is probably attributed to the interactions of ESPs with IECs. Upregulation of some inflammatory mediators such as COX-2, NLRP3, TNF-α, IL-6, IL-18 and IL-1β has been described in Giardia trophozoite-exposed macrophages [8, 32]. Herein, IECs, the first defending line of host preventing from enteric pathogen infection, expressed a significant increase in the expressions of intracellular NLRP3 and TNF-α and released IL-18 and IL-1β in a time-dependent manner in response to ESP exposure, extending the host defense mechanism against Giardia (Figure 2D and 2E). It is also noteworthy that IECs showed strikingly decreased levels of TJ proteins following ESP exposure, typically claudin-1, claudin-4, occludin and ZO-1 (Figure 2F to 2H). It has been indicated that, during autophagosome formation, LC3-I is conjugated to phosphatidylethanolamine to form LC3-II, which then binds to the expanding phagophore membranes to facilitate autophagosome formation [41]. Activation of autophagy is typically assessed by an elevated ratio of LC3-II to LC3-I and a decreased level of p62, a selective substrate of autophagy [42]. In this study, exposure of IECs with Giardia ESPs was able to increase the LC3-II/LC3-I ratio and decrease the level of p62 (Figure 2I and 2J), agreeing with the findings when Giardia trophozoites acted as a stimulus to IECs [43]. The collective data suggested that ESPs tended to be important virulence factors of Giardia infections during interactions with IECs.

Figure 2. Giardia ESPs triggered varying cellular responses of IECs.

Caco-2 cells were exposed to Giardia ESPs at the concentration of 20 μg/mL for the indicated time periods. (A and B) ESP exposure promoted cleavage of CASP-9, CASP-3 and PARP as well as elevation of Bax to Bcl-2 ratio in IECs as determined by western blot and gray value analyses. (B) The comparisons were made with the first group of the graph. (C) ESPs-induced IEC apoptosis was assessed by AO/EB staining (scale bar = 100 μm). (D and E) The protein levels of intracellular NLRP3 and TNF-α and released IL-18 and IL-1β were determined by western blot and gray value analyses. (E) The comparisons were made with the first group of the graph. (F) The mRNA levels of TJ proteins were determined by qPCR analysis. The comparisons were made with the first group of the graph. (G and H) The protein levels of claudin-1, claudin-4, occludin and ZO-1 was determined by western blot and gray value analyses. (H) The comparisons were made with the first group of the graph. (I and J) The protein levels of LC3-I, LC3-II and p62 were determined by western blot and gray value analyses. (J) The comparisons were made with the first group of the graph. All experiments were repeated three times. Data from triplicate wells from a representative of three independent experiments are presented as means ± SD. * p < 0.05, ** p < 0.01.

Endocytosis involved ESPs-triggered cellular responses

We also investigated the potential link between endocytosis and ESPs-triggered cellular responses. As shown in Figure 3A, inhibition of endocytosis by the use of dynamin inhibitor dynasore could effectively block the interactions between ESPs and IECs. Cellular responses to Giardia ESPs could be derived from extracellular and intracellular sources [8]. Herein, we tested our hypothesis that ESPs-evoked endocytosis involved the induction of those IEC responses we mentioned earlier. ESPs-activated IEC apoptosis could be blunted after inhibition of ESP endocytosis when dynasore was applied, as confirmed by western blotting showing reduced cleavage of CASP-9, CASP-3 and PARP and decreased ratio of Bax to Bcl-2 (Figure 3B and 3C) and AO/EB staining showing a less obvious apoptosis-inducing effect (Figure 3D). Blocking of ESP endocytosis by dynasore significantly downregulated the expressions of inflammatory molecules including intracellular NLRP3 and TNF-α and released IL-18 and IL-1β (Figure 3E). Inhibition of IEC endocytosis by dynasore attenuated ESPs-induced downregulation of claudin-1, claudin-4, occludin and ZO-1 (Figure 3F). Upon application of dynasore, the autophagy markers LC3-II and p62 underwent respective changes, with a decreased ratio of LC3-II to LC3-I and an increased level of p62 observed (Figure 3G and 3H). Collectively, ESPs-evoked endocytosis seemed to play an important part in contributing to the cellular responses involving apoptosis, inflammation, TJ, and autophagy.

Figure 3. Endocytosis involved ESPs-triggered cellular responses.

Dynasore was dissolved in 0.1% DMSO for use and shown as “dynasore + DMSO” panel here. Caco-2 cells were exposed to Giardia ESPs at the concentration of 20 μg/mL for 3 h. (A) Dynamin inhibition by dynasore blocked the interactions between ESPs and IECs as assessed by immunofluorescence analysis (scale bar = 20 μm). (B and C) Dynamin inhibition attenuated ESPs-induced cleavage of CASP-3, CASP-9 and PARP and elevation of Bax to Bcl-2 ratio as assessed by western blot and gray value analyses. (C) “Control” was compared to “ESPs”, and “ESPs” was compared to “ESPs + dynasore + DMSO”. (D) Dynamin inhibition suppressed ESPs-induced IEC apoptosis as assessed by AO/EB staining (scale bar = 100 μm). (E) Dynamin inhibition suppressed ESPs-induced upregulation of intracellular NLRP3 and TNF-α and released IL-1β and IL-18 as assessed by western blotting. (F) Dynamin inhibition protected claudin-1, claudin-4, occludin and ZO-1 expressions from being influenced by ESP exposure. (G and H) Dynamin inhibition blocked ESPs-induced elevation of LC3-II to LC3-I ratio and reversed ESPs-induced downregulation of p62 as assessed by western blot and gray value analyses. (H) “Control” was compared to “ESPs”, and “ESPs” was compared to “ESPs + dynasore + DMSO”. All experiments were repeated three times. Data from triplicate wells from a representative of three independent experiments are presented as means ± SD. * p < 0.05, ** p < 0.01.

PNPO evoked CME/CavME-mediated endocytosis

The most abundant Giardia-secreted proteins, PNPO, PPIB, tenascins and GCATB, have been predicted as important virulence factors [28–30], while it is uncertain whether they act as potential triggers for activation of IEC endocytosis. Ten recombinant proteins obtained by prokaryotic expression and purification were evaluated for their potential to activate IEC endocytic signaling, four (PNPO, Tena-2, Tena-3 and PPIB) of which were from our former study [8] and the others, Giardia GCATB-1/2/3/4/6/8, were generated here. Among the candidates, exposure of IECs with recombinant PNPO (rPNPO) for 2 hours (applied hereafter) led to a dramatic increase in the expressions of clathrin and caveolin-1 at both mRNA and protein levels (Figure 4A to 4C). Co-IP assay confirmed the binding between rPNPO and clathrin/caveolin-1 during in vitro interactions, notably at 2 h after exposure (Figure 4D). Then, the potential endocytic pathways evoked by rPNPO were determined by immunofluorescence staining. Blocking of CME by CPZ and CavME by genistein could inhibit PNPO-evoked endocytosis (Figure 4E). The data implied the involvement of Giardia-secreted PNPO in evoking CME and CavME-mediated endocytosis.

Figure 4. PNPO evoked CME/CavME-mediated endocytosis.

Genistein or CPZ was dissolved in 0.1% DMSO for use and directly shown as “Genistein” or “CPZ” panel here. Unless otherwise specified, Caco-2 cells were exposed to recombinant Giardia ES proteins at the concentration of 20 μg/mL for the indicated time periods. (A) Upon exposure of IECs to the indicated recombinant proteins, the mRNA levels of clathrin and caveolin-1 were assessed by qPCR analysis. The comparisons were made with the first group of the graph. (B and C) After a 2-h exposure of IECs to rPNPO, rTena-3, rPPIB and rGCATB-1, the protein levels of clathrin and caveolin-1 were assessed by western blot and grey value analysis. (C) The comparisons were made with the first group of the graph. (D) Binding between PNPO and clathrin/caveolin-1 was assessed by co-IP analysis. (E) Inhibitions of CME by CPZ and CavME by genistein blocked PNPO endocytosis (scale bar = 20 μm). All experiments were repeated four times. Data from triplicate wells from a representative of four independent experiments are presented as means ± SD. * p < 0.05, ** p < 0.01.

PNPO-triggered IEC responses depended on specific endocytic pathway

We explored the potential for biological link between PNPO-triggered IEC responses and specific endocytic pathway. Inhibition of CME by CPZ rather than CavME by genistein showed a striking suppression effect on PNPO-elevated levels of CASP-9/CAPS-3/PARP cleavage and ratio of Bax to Bcl-2, suggesting a specific link of CME to intrinsic apoptosis initiation (Figure 5A and 5B). This was confirmed by AO/EB staining analysis (Figure 5C). Likewise, inhibition of CME by CPZ rather than CavME by genistein significantly repressed PNPO-induced upregulated levels of intracellular NLRP3 and TNF-α and released IL-18 and IL-1β, indicating that CME was associated with PNPO-initiated IEC pro-inflammatory responses (Figure 5D). PNPO could also decrease the levels of TJ proteins claudin-1, claudin-4, occludin and ZO-1 at 2 h after exposure (Figure 5E). One striking observation was that application of CPZ showed an alleviative effect on PNPO-decreased levels of claudin-4, occludin and ZO-1, and genistein on occludin likewise (Figure 5E). However, no such effects were observed for claudin-1 no matter CPZ or genistein was applied (Figure 5E). This implied that PNPO-evoked CME and CavME might play a role specifically in regulation of certain TJ proteins. Notwithstanding, in contrast to a 3-h ESP exposure, a 2-h PNPO exposure could not initiate autophagy in Caco-2 cells, and this is the case when preadministration of CME or CavME was performed (Figure 5F and 5G). Collectively, PNPO appeared to be an important virulence factor as judged by its capability in initiating giardiasis-related IEC apoptosis, pro-inflammatory responses, and reduction of TJ protein levels, which seemed to be regulated majorly by CME pathway.

Figure 5. PNPO-triggered IEC responses depended on specific endocytic pathway.

Genistein or CPZ was dissolved in 0.1% DMSO for use and directly shown as “Genistein” or “CPZ” panel here. Caco-2 cells were exposed to rPNPO at the concentration of 20 μg/mL for 2 h. (A and B) Blocking of CME by CPZ rather than CavME by genistein showed a striking suppression effect on PNPO-elevated levels of CASP-9/CAPS-3/PARP cleavage and ratio of Bax to Bcl-2 as assessed by western blot and gray value analyses. (B) “Control” was compared to “PNPO”, and “PNPO” was compared to “PNPO + CPZ”. (C) Inhibition of CME by CPZ blocked PNPO-induced IEC apoptosis as measured by AO/EB staining (scale bar = 100 μm). (D) Blocking of CME by CPZ rather than CavME by genistein showed a striking suppression effect on PNPO-elevated levels of intracellular NLRP3 and TNF-α and released IL-1β and IL-18 as assessed by western blot analysis. (E) PNPO-decreased expression of TJ proteins involved specific endocytic pathways as assessed by western blot analysis. (F and G) Autophagy seemed to be not activated by a 2-h PNPO exposure as assessed by western blot analysis. (G) “Control” was compared to “PNPO”, and “PNPO” was compared to “PNPO + CPZ”. All experiments were repeated three times. Data from triplicate wells from a representative of three independent experiments are presented as means ± SD. ** p < 0.01.

Discussion

G. duodenalis has long featured as an important diarrhea-causing microbial pathogen responsible for the majority of parasitic gastroenteritis in humans, notably children, across the world [44]. Thus far, it remains largely unexplored about the pathogenesis of this ubiquitous extracellular parasite. In this study, as illustrated in the left panel of Figure 6, Giardia ESPs were shown to activate endocytosis. ESPs-evoked endocytosis was likely to act as an inducer for intrinsic apoptosis, pro-inflammatory responses, TJ protein downregulation, and autophagy in IECs. In contrast, as illustrated in the right panel of Figure 6, PNPO, a key virulence factor out of ESPs, appeared to activate CME and CavME in IECs. PNPO-evoked CME rather than CavME seemed to make an outstanding contribution to the giardiasis-related IEC apoptosis, inflammation, and TJ protein downregulation.

Figure 6.

Schematic diagram illustrating the potential endocytic mechanisms evoked by Giardia ESPs, notably PNPO, as well as the potential effects of this process on the modulation of some critical cellular responses against the extracellular infections caused by Giardia.

Our knowledge about trophozoite pathogenesis during the early stage of Giardia infection is not fully studied [45–47]. IECs become damaged or disrupted in response to strong trophozoite attachment, while there appear to be no cell invasion, secreted toxins, and overt inflammation [10]. Giardia-IEC interaction model with Caco-2 cells is dedicated to advancing our understanding of the interplay between the host and the parasite and the development of giardiasis [28, 29, 48]. ESPs released by Giardia have been predicted to be the predominant factor of giardiasis [28–30, 49], while the underlying cellular and molecular mechanisms are poorly understood. IEC apoptosis probably represents a key pathogenic factor for giardiasis as noted [5, 50, 51]. IECs exposed to Giardia trophozoites undergo apoptosis as described [9, 52–55], while little is known about the causative factors. Giardipain-1 has been identified as an important Giardia virulence factor involved in IEC apoptosis [56]. In this study, we showed the capability of Giardia ESPs and PNPO in inducing IEC apoptosis through a CASP-9/-3-regulated intrinsic pathway in vitro. The attachment of trophozoites to IECs and the interaction are known as a potential cause of host intestinal inflammation related to the pathogenesis of giardiasis, although minimal [3, 5]. Nevertheless, interestingly, Giardia infection and Giardia cathepsin B cysteine proteases have been shown to attenuate the accumulation of polymorphonuclear leukocytes or neutrophils and the host inflammatory response in the intestine [57]. TNF-α has been recognized as an important pro-inflammatory cytokine in response to Giardia infection [32, 58, 59]. NLRP3 inflammasome is vital in host defense against infectious pathogens due to subsequent IL-1β (a potent pro-inflammatory mediator implicated in both innate and adaptive immune responses) and IL-18 (an inducer of sequential activation of NK cells and Tc cells) secretion [60]. NLRP3 inflammasome is activated in macrophages in response to the intracellular infections by protozoan parasites Plasmodium, Leishmania, Trypanosoma, Toxoplasma, and Entamoeba [61], and recently, it has also been shown to be active during in vitro Giardia/rPPIB-macrophage interactions [8]. Those studies concerning Giardia-triggered host inflammatory responses were mostly conducted by virtue of an in vivo animal model or an in vitro macrophage-like cell line. This study took a step forward and confirmed the efficacy of Giardia ESPs/PNPO in activating immune-defense-related markers TNF-α, NLRP3, IL-1β and IL-18 in IECs, which would extend our understanding of the functional role of IECs that represent the first and direct line of defense against Giardia infection.

TJ is known as a major epithelial barrier maintained by TJ proteins such as ZO-1, occludin and claudins, which restricts paracellular permeability and balances the targeted transport and the exclusion of unexpected pathogens [62, 63]. Intestinal epithelial barrier (IEB) integrity functions vitally in host defense against noninvasive Giardia, and IEB disruption is an important part of pathogenesis of giardiasis [5]. Giardia trophozoites are known to disrupt the TJ complexes of IECs during extracellular infection, thereby increasing epithelial permeability, damaging IEB function, and influencing nutrient absorption [35, 64–66]. Giardia-expressed cysteine proteases and GCATB-like protease giardipain-1 are recognized as vital virulence factors in relation to destruction of IEC junctional complexes and barrier function, which involves immune modulation and pathogenesis of giardiasis [7, 56]. Giardia-induced apoptosis is shown to be another important contributor to TJ barrier disruption via increasing IEC permeability, involving the development of giardiasis [50, 51]. Here Giardia ESPs and PNPO were proven as potential inducers of TJ protein downregulation in IECs, such as claudin-1, claudin-4, occludin and ZO-1. However, in any case, the in vitro interaction system used in this study could not reflect the complexity of interactions from neighboring cells of different types and functions. A functional barrier assessment based on organoid-derived monolayers [67] and/or an accessible laboratory animal model is required to confirm the observations here. Intracellular infections caused by the protozoan parasites Leishmania and Cryptosporidium parvum have been documented as a cause of autophagy in immune cells and IECs [68–70], while this is not case for the extracellular parasite, G. duodenalis. Our one recent study has demonstrated that Giardia trophozoites were able to induce IEC autophagy in vitro, participating in TJ/NO regulation [43]. Herein, Giardia secreta were identified as the potential causative factor.

Endocytosis is involved in a variety of physiologic and pathophysiologic processes, such as cell growth and differentiation, cell chemotaxis, and immune responses [71], while its role in noninvasive Giardia infection remains uncertain. Here we initially observed the possible association of Giardia ESP exposure with endocytosis initiation. We further determined that endocytosis was associated with ESPs-induced IEC apoptosis, enhancement of pro-inflammatory cytokines, downregulation of TJ proteins, and autophagy. It has been shown that O. viverrini ESPs were internalized preferentially by liver cell line rather than intestinal epithelial cell line, and cellular uptake of ESPs through CME and CavME possibly drives cell proliferation and IL-6 secretion [22]. The internalization of Giardia EVs into primary mouse peritoneal macrophages has been linked to NLRP3 activation [33]. T. gondii ESPs enter cells through CME and micropinocytosis, downregulating the expressions of TJ proteins including claudin-1, occludin and ZO-1 [72]. Despite these findings, it remains largely unknown about how parasite ESPs internalize into host cells and modulate pathological or protective responses. It is of interest to note here that the PNPO-evoked CME might play an important role in regulation of giardiais-related IEC apoptosis, pro-inflammatory responses, and TJ protein levels. Nevertheless, more reliable and quantifiable high-resolution protein localization and some functional data with regard to IEC TJ barrier are needed to clarify the involvement of specific Giardia ESPs and the associated endocytic pathways in the pathogenesis of giardiasis.

In conclusion, this study identified Giardia ESPs, especially PNPO, as potentially important virulence factors involved in the development of giardiasis, and revealed the association of specific endocytic pathways with this process by regulating multiple pathogenesis-related factors, such as apoptosis, inflamation, and TJ integrity. The data generated here would provide new insights into the interaction mechanisms between Giardia and IECs and the pathogenesis of giardiasis.

Abbreviations

AO/EB

acridine orange/ethidium bromide

CASP

caspase

CavME

caveolin-mediated endocytosis

CME

clathrin-mediated endocytosis

Co-IP

co-immunoprecipitation

CPZ

chlorpromazine

ES

excretory-secretory

ESPs

excretory-secretory products

EVs

extracellular vesicles

GCATB

Giardia cathepsin B

IEB

intestinal epithelial barrier

IECs

intestinal epithelial cells

NO

nitric oxide

PAB

polyclonal antibody

PM

plasma membrane

PNPO

pyridoxamine 5'-phosphate oxidase

PPIB

peptidyl-prolyl cis-transisomerase B

rPNPO

recombinant PNPO

RT

room temperature

Tena

tenascins

TJ

tight junction