The Anti-Apoptotic Role of COX-2 during In Vitro Infection of Human Intestinal Cell Line by Giardia duodenalis and the Potential Regulators

ABSTRACT

The protozoan parasite Giardia duodenalis inhabits the upper small intestine of mammals, including humans, and causes a disease known as giardiasis, which can lead to diarrhea, abdominal cramps, and bloating. G. duodenalis was known as a causative factor of intestinal epithelial cell (IEC) apoptosis. Cyclooxygenase-2 (COX-2) has been identified as an influencing factor of pathogen infection by participating in immune response, while its role in host defense against Giardia infection is not clear. Here, we initially observed the involvement of COX-2 in the regulation of Giardia-induced IEC apoptosis. Inhibition of COX-2 activity could promote Giardia-induced reduction of IEC viability, increase of reactive oxygen species (ROS) production, and decrease of nitric oxide (NO) release, which would exacerbate IEC apoptosis. In addition, during Giardia-IEC interactions, COX-2 inhibition was able to accelerate caspase-3 activation and poly(ADP-ribose) polymerase (PARP) cleavage and inhibit the expressions of some anti-apoptotic proteins like cIAP-2 and survivin. In contrast, COX-2 overexpression could reduce Giardia-induced IEC apoptosis. We further investigated the regulatory mechanisms affecting COX-2 expression in terms of anti-apoptosis. The results showed that p38/ERK/AKT/NF-κB signaling could regulate COX-2-mediated ROS/NO production and anti-IEC apoptosis during Giardia infection. We also found that COX-2-mediated anti-IEC apoptosis induced by Giardia was related to Toll-like receptor 4 (TLR4)-dependent activation of p38-NF-κB signaling. Collectively, this study identified COX-2 as a promoter for apoptotic resistance during Giardia-IEC interactions and determined the potential regulators, furthering our knowledge of anti-Giardia host defense mechanisms.

INTRODUCTION

Giardia duodenalis is a binucleated, flagellated protozoan parasite that lives and replicates in the small intestine of different vertebrate hosts and causes giardiasis, which is known as one of the most common gastrointestinal diseases around the world (1–3). Since intestinal epithelial cells (IECs) are renewed every 3 to 5 days, the noninvasive Giardia trophozoites must frequently detach and adhere to IECs to avoid being eliminated by intestinal tract movement (4). The engagement of both innate and adaptive cellular immune responses is necessary for efficient clearance of Giardia (4).

Studies have shown that the physiological flux of reactive oxygen species (ROS) mediates cell migration, proliferation, differentiation, polarization, and cell death (5), and increased oxidative stress often leads to early senescence, inflammation, and apoptosis (6). Nitric oxide (NO) is a free radical synthesized from l-arginine by NO synthases (7). IECs and immune cells synthesize NO presenting immunomodulatory and cytotoxic activities (4). Giardia inhibits IEC NO production because the parasite competes with IECs for the energy source, arginine (8), and low concentrations of arginine can induce IEC apoptosis (9). NO has also been shown to function as an inducer for intestinal peristalsis, promoting the discharge of Giardia from the intestine (10, 11). Toll-like receptor (TLR) activation is important for microbial infection and inflammation occurrence, and TLR is a critical linker between innate and adaptive immunity (12). TLR-dependent immune responses are commonly relevant to the activation of the mitogen-activated protein kinase (MAPK) and NF-κB pathways (13). MAPKs are important signal transducing serine/threonine-specific protein kinases participating in a great variety of cellular processes, including cell survival and programmed cell death (14, 15). The AKT signaling pathway is involved in inhibiting cell apoptosis and stimulating cell proliferation following the activation of protein kinase B, a serine/threonine kinase (16). NF-κB is a family of transcription factors that modulate diverse biological processes, including cell survival and apoptosis (17, 18). It has been reported that, during Giardia infection, NF-κB signaling or TLR2-dependent MAPK/AKT signaling regulates the secretion of proinflammatory cytokines interleukin-12 (IL-12), necrosis factor alpha (TNF-α), and IL-6 in mouse macrophages and potentially controls the severity of giardiasis (19). IECs are well known as the first line of defense against enteric pathogens, whereas whether Giardia adhesion can trigger TLR/MAPK/AKT/NF-κB signaling in IECs and the potential relationship of this process with the regulation of ROS/NO production and downstream cellular apoptosis are not clear.

Cyclooxygenase (COX), known as prostaglandin (PG) G/H synthases, is a bifunctional enzyme that catalyzes the committed step in the conversion of arachidonic acid to PGs (20). There are two isoforms for COX, a constitutive form COX-1 and an inducible form COX-2 (20). The former is present in most cell types and has a role in maintaining normal physiological functions, while COX-2 mostly involves inflammation in macrophages, fibroblasts, and endothelial cells (21). As noted before, COX-2 was reported to be able to modulate proinflammatory immune response to help the host defend against Toxoplasma gondii infection (22). COX-2 is upregulated in mycobacteria-infected macrophages to enhance host defense through regulating adaptive immunity and restoring the mitochondrial inner membrane (23, 24). However, it is still unclear whether COX-2 plays a role in host defense against microbial infections through modulating cell apoptosis, especially noninvasive Giardia. IEC apoptosis probably represents a key pathogenic factor for giardiasis as described (4). It is, therefore, worthwhile to examine the anti-apoptotic role of COX-2 during Giardia infection. In addition, as mentioned earlier, ROS and NO are important signals for maintaining the survival and homeostasis of cells and toxic species responsible for cellular injury (5, 9), while little is known about the interplay between COX-2 and ROS/NO, notably in the context of microbial infections. Giardia-influenced changes in ROS/NO levels have been linked to IEC apoptosis (9, 25), whereas whether COX-2 functions as a regulator in this process needs further in-depth investigation. In addition, it is also worth studying the correlation between TLR/MAPK/AKT/NF-κB and COX-2.

Thus far, cellular responses to Giardia infection are still largely unknown, notably those involving IECs. The present study was conducted to determine the potential anti-apoptotic function of COX-2 in IECs exposed to Giardia trophozoites; investigate the probable signaling mechanisms regulating COX-2 expression, ROS/NO production, and IEC apoptosis; and try to reveal the interrelation.

RESULTS

COX-2-regulated anti-apoptosis during Giardia-IEC interactions.

We initially explored if Giardia could induce COX-2 upregulation in Caco-2 and HT-29 human intestinal cell lines. IECs were challenged with Giardia trophozoites at a ratio of 10 parasites/cell as described (25–27), and the same applies hereinafter, although other ratios (5 or 20 parasites/cell) have been previously applied (9, 28). We observed that the expression of COX-2 was significantly upregulated at both mRNA and protein levels within hours, notably at 6 h (Caco-2) and 9 h (HT-29) (Fig. 1a). In the CCK-8 assays, the COX-2 inhibitor NS398 reduced the viability of Caco-2 and HT-29 cells exposed to Giardia for 3 h and 6 h (Fig. 1b). It is clear that cells must keep the levels of ROS and NO in check for survival, growth, and apoptosis (5, 9). Here, we investigated if COX-2 could regulate IEC ROS/NO production during in vitro infection of Caco-2 and HT-29 cells by Giardia. As shown in Fig. 1c, Giardia exposure increased the cellular ROS levels at 3 h and 6 h, and COX-2 inhibition promoted this process. In contrast, in Fig. 1d, the release of NO from IECs decreased gradually following Giardia exposure, and COX-2 inhibition augmented the reduction of NO release. The data indicated that COX-2 could regulate the levels of ROS and NO during Giardia-IEC interactions.

FIG 1.

COX-2-regulated anti-apoptosis during Giardia-IEC interactions. Caco-2 and HT-29 cells were exposed to Giardia trophozoites for the indicated time periods. (a to d) Caco-2 and HT-29 cells were used. (a) Upon exposure, the mRNA and protein levels of COX-2 were determined by qPCR and Western blot analyses. (b) Inhibition of COX-2 by NS398 promoted Giardia-induced decrease of cell viability as assessed by the CCK-8 assay. (c) COX-2 inhibition promoted Giardia-induced ROS increase as examined by fluorescence microscope (scale bar = 1,000 μm) and microplate reader. (d) COX-2 inhibition promoted Giardia-induced decrease of NO production as examined by using a microplate reader. (e to j) Caco-2 cells were used. (e) COX-2 inhibition augmented Giardia-induced apoptosis as assessed by AO/EB staining analysis (scale bar = 1,000 μm). (f) At 6 h after exposure, COX-2 overexpression reduced Giardia-induced apoptosis as assessed by AO/EB staining analysis (scale bar = 1,000 μm). (g) At 6 h after exposure, COX-2 inhibition promoted Giardia-induced cleavage of CASP3 and PARP as assessed by Western blot analysis. (h and i) At 6 h after exposure, COX-2 inhibition promoted Giardia-induced downregulation of anti-apoptotic proteins cIAP-2 and survivin as assessed by Western blot and gray value analyses. (j) At 6 h after exposure, COX-2 overexpression blocked Giardia-induced cleavage of CASP3 and PARP and downregulation of anti-apoptotic proteins cIAP-2 and survivin, as assessed by Western blot analysis. All experiments were repeated at least three times. (a to d and i) The values are expressed as the mean ± SD (*, P < 0.05; **, P < 0.01). (a, c, e to h, and j) Representative images are presented. Gl, Giardia.

Acridine orange (AO)/ethidium bromide (EB) double staining was performed to assess the potential regulatory role of COX-2 in Giardia-induced IEC apoptosis. As shown in Fig. 1e, upon Giardia exposure, more apoptotic cells stained orange were seen in NS398-pretreated 3-h and 6-h groups. In contrast, apoptosis induced by a 6-h Giardia exposure was blunted by preadministration of rebamipide to overexpress COX-2 level in IECs (Fig. 1f). Therefore, COX-2 potentially operated as an anti-apoptotic regulator during Giardia infection. We further examined the effects of COX-2 on the expression levels of apoptosis-related proteins. As expected, exposure of IECs to Giardia for 6 h induced the cleavage of caspase-3 (CASP3) and PARP, and preincubation of cells with NS398 before exposure enhanced the cleavage of CASP3 and PARP (Fig. 1g). We also found that NS398 pretreatment could promote the downregulation of the inhibitors of apoptosis proteins cIAP-2 and survivin induced by a 6-h Giardia exposure (Fig. 1h and i). By contrast, COX-2 overexpression by its agonist rebamipide had remarkable inhibitory effect on IEC apoptosis induced by a 6-h Giardia exposure (Fig. 1j). Taken together, Giardia exposure upregulated COX-2 expression in IECs, and it played a role in modulating cell survival and anti-apoptotic process.

TLR4/MAPK/AKT signaling regulated COX-2-mediated anti-apoptosis.

As mentioned earlier, MAPK (p38/ERK) and AKT signaling pathways affect multiple cellular processes, including cell proliferation and apoptosis (16, 29). TLR4 signaling plays key roles in the innate immune response to microbial infection and involves apoptotic cell death (30). We first studied the involvement of TLR4/MAPK/AKT signaling in regulating COX-2 expression. The changes in the phosphorylation levels of MAPK/AKT-related proteins were first evaluated. There was a significant increase in the phosphorylation of p38, ERK1/2, and AKT within hours after Giardia exposure (Fig. 2a). p38 inhibitor SB202190, ERK1/2 inhibitor SCH772984, and AKT inhibitor MK-2206 2HCl were able to block the upregulation of COX-2 induced by a 6-h Giardia exposure, revealing an association between p38/ERK/AKT signaling and COX-2 (Fig. 2b). TLR4 inhibition by its inhibitor TAK-242 significantly inhibited both the p38 phosphorylation and COX-2 expression in IECs exposed to Giardia for 6 h, while there were no observed changes for ERK and AKT (Fig. 2c). The data indicated a role of TLR4-dependent p38 signaling in the regulation of COX-2 expression.

FIG 2.

TLR4/MAPK/AKT signaling regulated COX-2-mediated anti-apoptosis. Caco-2 cells were exposed to Giardia trophozoites for the indicated time periods. (a) Giardia exposure activated p38, ERK1/2, and AKT signaling as assessed by Western blot analysis. (b to g) Prior to a 6-h exposure to Giardia, Caco-2 cells were incubated with or without p38 inhibitor SB202190, ERK1/2 inhibitor SCH772984, AKT inhibitor MK-2206 2HCl, and TLR4 inhibitor TAK-242 for 1 h. Mock groups were included. (b) p38/ERK/AKT inhibition suppressed Giardia-induced COX-2 up-expression as assessed by Western blot analysis. (c) TLR4 inhibition suppressed Giardia-induced p38/COX-2 up-expression as measured by Western blot analysis. (d and e) TLR4/p38/ERK/AKT inhibition affected the levels of ROS and NO production after Giardia exposure. (f) TLR4/p38/ERK/AKT inhibition promoted Giardia-induced IEC apoptosis as assessed by AO/EB staining analysis (scale bar = 1,000 μm). (g and h) TLR4/p38/ERK/AKT inhibition promoted Giardia-induced cleavage of CASP3 and PARP and downregulation of anti-apoptotic proteins cIAP-2 and survivin as assessed by Western blot detection and gray value analysis. All experiments were repeated at least three times. (d, e, and h) The values are expressed as the mean ± SD (*, P < 0.05; **, P < 0.01). (a, b, c, f, and g) Representative pictures are shown.

We then assessed the potential influence of TLR4/MAPK/AKT signaling on ROS/NO generation and the downstream apoptotic cascades. Inhibition of those signaling proteins led to a significant increase in ROS production and a significant decrease in NO production in IECs exposed to Giardia for 6 h, except for the effect of TLR4 inhibition on ROS production (Fig. 2d and e). As reflected in AO/EB staining analysis, at 6 h after Giardia exposure, more obvious apoptosis-inducing effects were observed when TLR4, p38, ERK1/2, and AKT were inhibited (Fig. 2f). In addition, inhibition of TLR4, p38, ERK1/2, and AKT significantly increased the levels of CAPS3 and PARP cleavage and repressed the expressions of cIAP-2 and survivin in IECs after 6 h of exposure to Giardia (Fig. 2g and h). In the context of Giardia-IEC interactions, considering the formerly established close links between TLR4/MAPK/AKT signaling and COX-2 and between COX-2 and ROS/NO/apoptosis, it can be inferred that COX-2-mediated ROS/NO regulation and anti-IEC apoptosis was possibly controlled by p38/ERK/AKT signaling or TLR4-dependent p38 signaling.

TLR4-dependent p38-NF-κB signaling regulated COX-2-mediated anti-apoptosis.

NF-κB could modulate expression of a diverse array of genes involved in different biological processes (31), while its association with Giardia-induced changes in COX-2 expression, ROS/NO production, and IEC apoptosis is uncertain. Immunofluorescence and Western blot analyses showed that nuclear NF-κB p65 was significantly increased after exposure of IECs to Giardia for 3 h and 6 h (Fig. 3a and b). We also observed that COX-2 up-expression in IECs caused by a 6-h Giardia exposure could be blocked by NF-κB p65 inhibition with its inhibitor JSH-23 (Fig. 3c). At 6 h after Giardia exposure, NF-κB p65 inhibition could result in a significant increase/decline in ROS/NO production in IECs (Fig. 3d and e) and also enhance IEC apoptosis as reflected by both AO/EB staining and Western blot analyses (Fig. 3f and g). The activation of MAPK and NF-κB has been mostly studied in the context of TLR agonists (13); therefore, we assayed if NF-κB p65 nuclear translocation was associated with TLR4-dependent p38 signaling. As expected, NF-κB p65 nuclear translocation in IECs caused by a 6-h exposure to Giardia could be attenuated by both TLR4 and p38 inhibition (Fig. 3h and i), and this was confirmed in the following immunofluorescence analysis (Fig. 3j). Combined with our former findings, we could clearly figure out that COX-2-mediated anti-IEC apoptosis during Giardia infection is possibly dependent on TLR4-p38 signaling-controlled NF-κB nuclear translocation.

FIG 3.

TLR4-dependent p38-NF-κB signaling regulated COX-2-mediated anti-apoptosis. Caco-2 cells were exposed to Giardia trophozoites for the indicated time periods. (a and b) Giardia-induced NF-κB p65 nuclear translocation in Caco-2 cells examined by fluorescence microscope (scale bar = 40 μm) and Western blot analysis. (c to j) Prior to a 6-h exposure to Giardia, Caco-2 cells were incubated with or without NF-κB p65 inhibitor JSH-23, p38 inhibitor SB202190, and TLR4 inhibitor TAK-242 for 1 h. Mock groups were included. (c) NF-κB p65 inhibition suppressed Giardia-induced COX-2 up-expression as assessed by Western blot analysis. (d and e) NF-κB p65 inhibition affected the levels of ROS and NO production after Giardia exposure. (f) NF-κB p65 inhibition promoted Giardia-induced IEC apoptosis as assessed by AO/EB staining analysis (scale bar = 1,000 μm). (g) NF-κB p65 inhibition promoted Giardia-induced cleavage of CASP3 and PARP and downregulation of anti-apoptotic proteins cIAP-2 and survivin as assessed by Western blot analysis. (h to j) p38/TLR4 inhibition attenuated Giardia-induced NF-κB p65 nuclear translocation as assessed by Western blot and indirect immunofluorescence (scale bar = 40 μm) analyses. All experiments were repeated at least three times. (d and e) The values are expressed as the mean ± SD (**, P < 0.01). (a to c and f to j) Representative pictures are shown.

DISCUSSION

G. duodenalis is an important zoonotic pathogen distributed globally and leads to an estimated 280 million cases of human giardiasis each year (32), while the mechanism for host defense against Giardia infection remains largely unexplored. Our study reported the anti-apoptotic role of COX-2 during IEC interactions with Giardia and identified the possible regulators (Fig. 4). Mechanistically, we first found that COX-2 involved the process of attenuating Giardia-induced IEC apoptosis. The anti-apoptotic process mediated by COX-2 was confirmed to be regulated by MAPK/AKT signaling or by TLR4-dependent p38-NF-κB signaling. The Giardia-induced ROS increase and NO decrease have been identified as IEC apoptosis stimulators (9, 25). Here, we provide new findings that the levels of ROS/NO could be influenced by MAPK/AKT/NF-κB signaling-controlled COX-2 regulation.

FIG 4.

Schematic diagram illustrating the signaling pathways involved in regulating the levels of ROS and NO and the COX-2-mediated anti-apoptosis during Giardia-IEC interactions.

To date, the role of COX-2 in resistance to IEC apoptosis caused by microbial infections remains elusive. Giardia infection was known to disrupt the intestinal epithelial barrier and induce apoptosis (33). Our previous studies have reported that Giardia could induce TNFR1-mediated extrinsic and ROS-mediated mitochondrial pathways of apoptosis in IECs (25, 26). It has also been indicated that sodium-dependent glucose cotransporter-1-dependent glucose uptake protects human IECs against Giardia-induced apoptosis (34). This study took a step forward and reported an important function of COX-2 in regulating anti-IEC apoptosis and maintaining cell viability during in vitro Giardia infection. It has been well recognized that, Giardia adheres to and interacts with IECs, producing merely a slight inflammatory response (4). In this case, IEC apoptosis caused by noninvasive Giardia attacks may be a key pathogenic factor for giardiasis as noted (4). Our previous study indicated that Giardia can induce oxidative stress-mediated IEC apoptosis (25). Given this circumstance, COX-2-mediated anti-apoptotic effect may be partially due to the anti-ROS function of COX-2 confirmed in this study. NO released from infected cells was previously understood to inhibit the process of Giardia trophozoite proliferation and excystation (8). A decline in NO release from IECs due to the competitive uptake of arginine by Giardia is indicative of the occurrence of apoptosis as previously described (9). This study identified that COX-2 exerted a function in maintaining the level of NO release, suggesting a role of COX-2 in helping promote cell survival. Taken together, it is not difficult to draw the conclusion that COX-2 up-expression in IECs responding to Giardia infection is vital to facilitating anti-apoptosis and maintaining cell survival.

TLR-mediated immune responses represent the first line of host defense, which help remove infectious disease threats and restore immune homeostasis (35). Dysregulation of the TLR system may be related to various immunopathological manifestations during microbial/parasitic infections (35). It has been reported that TLR4^−/− mice produce less COX-2 and are more susceptible to bowel disease (36). We found that COX-2-mediated anti-apoptotic function could be influenced by TLR4 inhibition. As indicated before, MAPK and AKT could be activated in Giardia-infected mouse macrophages, affecting the release of some inflammatory factors (19). Here, we demonstrated the novel interplay between p38/ERK/AKT signaling and COX-2-mediated anti-apoptosis in Giardia-IEC interactions. It has been shown that Propionibacterium acnes can induce rat intervertebral disc degeneration by promoting COX-2/PGE2 expression via activation of NF-κB (37). Here, the cross talk between NF-κB and COX-2 regulated by TLR4-dependent activation of p38 signaling was proven to influence Giardia-induced apoptotic outcomes. It is of interest to note that a 1-h exposure to Giardia inhibits the activation of NF-κB in macrophages as described recently (38), which is contradictory to our finding that a 3-h or prolonged exposure promoted the activation of NF-κB in a TLR4-dependent manner in Caco-2 cells. This might be due to macrophages and Caco-2 cells responding differently to Giardia WB trophozoites, as well as the difference in the duration of parasite exposure, which is worthy of being studied further. Moreover, the detailed factors affecting Giardia-induced ERK/AKT activation in IECs need to be further investigated. Here, we also displayed the connections between MAPK/AKT/NF-κB signaling and COX-2-regulated ROS/NO, expanding the related mechanisms.

Apoptosis is a crucial modulator of defense-associated metabolism in microbial infections; after successful replication, intracellular microbial pathogens induce apoptosis in order to facilitate their escape and reinfection of host cells (39). Parasite-induced apoptosis may be conducive to eliminating parasitic infections, while some protozoan parasites can cause certain diseases via inducing apoptosis by direct and indirect mechanisms (40). Apoptosis in response to noninvasive Giardia infection may disrupt intestinal epithelial barrier function by increasing IEC permeability and participate in the development of giardiasis (41, 42). Anti-apoptotic treatment is considered a potent strategy for mitigation of giardiasis without harming other tissues (42). In spite of those advances, the clinical significance of COX-2 in treatment of giardiasis needs further investigations. In conclusion, the present study indicated a potential new rescue mechanism against Giardia-induced IEC apoptosis, the TLR4/MAPK/NF-κB-dependent COX-2 regulation, providing new insights into the interaction mechanisms between Giardia and IECs and the pathogenesis of giardiasis, as well as providing important implications for the development of new treatment strategies for giardiasis.

MATERIALS AND METHODS

Cell culture.

The human colon adenocarcinoma cell lines Caco-2 and HT29 that closely resemble normal human small IECs (43, 44), were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and applied to interact with Giardia trophozoites. Caco-2 cells were cultured in high-glucose Dulbecco modified Eagle medium (DMEM) (HyClone, Logan, UT, USA) containing 10% fetal bovine serum (FBS), 1% MEM nonessential amino acids (NEAA), and 1% penicillin/streptomycin. By contrast, HT29 cells were cultured in DMEM/F12 (HyClone, Logan, UT, USA) containing 10% FBS. The two cell lines were incubated at 37°C and 5% CO₂ and subcultured every 3 to 4 days at 80% to 90% confluence.

Parasite culture.

G. duodenalis WB isolate typed as assemblage A (ATCC 30957; ATCC, Manassas, VA, USA) was used in the present study. Trophozoites were axenically cultivated at 37°C in 15-mL conical bottomed tubes in modified TYI-S-33 medium containing 10% FBS and 0.1% bovine bile supplemented with 0.1% gentamicin and 1% penicillin/streptomycin (45). Cultures were harvested by chilling on ice for 15 min. Detached trophozoites were centrifuged, washed with phosphate-buffered saline (PBS), resuspended in cell culture medium, counted using a hematocytometer, and used to treat cells at a ratio of 10 parasites/cell. Prior to exposure, an endotoxin enzyme-linked immunosorbent assay (ELISA) kit (Meimian Biotech, Yancheng, China) was applied to ensure that the washed parasites and resuspension medium are free of endotoxins that may be responsible for the observed TLR activation.

qPCR analysis.

Cells were exposed to Giardia for 0, 3, 6, and 9 h. Total RNA was extracted from cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). cDNA was synthesized from total RNA (1 μg) using a HiScript II 1st Strand cDNA synthesis kit (Vazyme, Nanjing, China). We used primer sets 5′-ATCATTCACCAGGCAAATTGC-3′ (F) and 5′-GGCTTCAGCATAAAGCGTTTG-3′ (R) for COX-2, and 5′-GAAGGTGAAGGTCGGAGTC-3′ (F) and 5′-GAAGATGGTGATGGGATTTC-3′ (R) for glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Quantitative PCR (qPCR) was performed using a SYBR green PCR Master Mix kit (Vazyme, Nanjing, China) on an LC480 LightCycler system (Roche, Indianapolis, IN, USA). Relative mRNA expression levels of COX-2 were calculated according to the 2^−ΔΔCT method normalized using GAPDH.

Western blot analysis.

Cells were treated for the indicated periods of time and washed with ice-cold PBS to remove parasites. Total cellular proteins were extracted using radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime, Shanghai, China) supplemented with 1% phenylmethylsulfonyl fluoride (PMSF) (Beyotime, Shanghai, China). Protein concentration was quantified using an enhanced BCA Protein assay kit (Beyotime, Shanghai, China). Protein expression levels were assessed by Western blot analysis. In brief, proteins were separated by 12% SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes by electroblotting. Membranes were blocked with 5% skim milk in phosphate-buffered saline with Tween 20 (PBST) at room temperature (RT) for 2 h and then incubated at 4°C for 12 h with the appropriate primary antibodies (1:1,000 dilution in PBST) against COX-2, β-actin, pro-/cL-CASP3, pro-/cL-PARP, XIAP, cIAP-1/2, survivin, p38, p-p38, ERK1/2, p-ERK1/2, AKT, p-AKT, and NF-κB p65. Primary antibodies were obtained from two commercial sources (ABclonal, Wuhan, China; Bioss, Beijing, China). Membranes were washed three times for about 1 h in PBST and further incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (1:5,000 dilution in PBST; ABMART, Shanghai, China) at RT for 1 h. Proteins were visualized by use of chemiluminescence detection (Syngene, Cambridge, UK). The band intensity of indicated protein was measured with NIH Image J software.

Cell viability assay.

Cells were plated in 96-well plates at a seeding density of 2 × 10⁴ cells/well in complete medium, incubated for 12 h, and then replaced with FBS-free medium incubated for 3 h. Cells were exposed to parasites for 0, 3, and 6 h, and negative control wells containing just DMEM and trophozoites in DMEM were included. Cell viability was measured by a CCK-8 assay (Apexbio, Houston, TX, USA). The absorbance was measured at 450 nm wavelength.

ROS/NO detection.

For ROS detection, cells were seeded in 24-well plates (1 × 10⁵ cells/well) and 96-well plates (1 × 10⁴ cells/well) as required by the detection methods that we used below. After 12 h of incubation, cells were exposed to parasites for the indicated time periods and then washed with ice-cold PBS. The intracellular ROS levels were measured using an oxidation-sensitive fluorescent probe 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) (Beyotime, Shanghai, China). The intensity of DCF fluorescence was measured using a Lionheart FX Automated Microscope (BioTek, Winooski, VT, USA) and a Fluostar Omega microplate reader (BMG, Ortenberg, Germany). For NO detection, cells were seeded in 96-well plates (1 × 10⁴ cells/well) and incubated for 12 h. After exposure to parasites for the indicated time periods, NO production represented by nitrite concentration in supernatants of cultured cells was assayed with Griess reaction using a NO assay kit (Beyotime, Shanghai, China). The absorbance was measured at 540 nm wavelength.

AO/EB assay.

Cells were seeded in 24-well plates at a density of 1 × 10⁵ cells/well and incubated for 12 h. Cells were exposed to parasites for the indicated time periods. At the indicated time, the medium was removed, and the cells were washed with ice-cold PBS. Cell apoptosis induced by Giardia was evaluated by dual staining with combined fluorescent dyes AO and EB (BestBio, Shanghai, China) and observed using a Lionheart FX Automated Microscope. Cells stained green are regarded as normal cells, while those stained orange are apoptotic cells.

Immunofluorescence assay.

Cells in 24-well plates (1 × 10⁵ cells/well) were exposed to parasites for the indicated time periods. Cells were washed with ice-cold PBS, 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 1% bovine serum albumin (BSA) in PBS for 1 h at RT. Cells were incubated with anti-NF-κB p65 antibody (dilution, 1:200) with 1% BSA in PBST at 4°C overnight and then fluorescein isothiocyanate (FITC)-AffiniPure goat anti-rabbit IgG (H+L) (dilution, 1:200; Jackson, West Grove, USA) at 37°C for 1 h in the dark. Cell nucleus was stained with DAPI (1 μg/mL; AlphaBio, Tianjin, China). Fluorescent images were captured using a Lionheart FX automated microscope.

Protein inhibition.

We used COX-2 inhibitor NS398 (50 μM in use), p38 inhibitor SB202190 (10 μM), ERK1/2 inhibitor SCH772984 (10 μM), AKT inhibitor MK-2206 2HCl (50 μM), and NF-κB inhibitor JSH-23 (50 μM) (Selleck Chemicals, Houston, TX, USA), and TLR4 inhibitor TAK-242 (10 μM; APEXBIO, Houston, TX, USA) in inhibition analyses. All inhibitors were applied 1 h before exposure. A 0.1% dimethyl sulfoxide (DMSO) solution was applied as the solvent control.

COX-2 overexpression.

Overexpression of COX-2 level was performed using rebamipide (MCE, Shanghai, China) at the concentration of 10 nM. The agonist was applied 1 h before exposure.

Statistical analysis.

Statistical analyses were performed using the GraphPad Prism 7.0 software. All experiments were performed at least three times, and the results are presented as means ± standard deviation (SD). The statistical significance of the differences was assessed using Student's t test for comparing two groups or one-way analysis of variance (ANOVA) for comparing three or more groups. P values less than 0.05 were considered to be statistically significant (*, P < 0.05; **, P < 0.01).