Toxoplasma gondii Infection Promotes Epithelial Barrier Dysfunction of Caco-2 Cells

Abstract

After oral infection, Toxoplasma gondii invades intestinal cells, induces breakdown of intestinal physiology and barrier functions, and causes intestinal pathology in some animal species. Although parasites’ invasion into host cells is a known phenomenon, the effects of T. gondii infection in the intestinal barrier are still not well established. To evaluate morphological and physiological modifications on the colorectal adenocarcinoma-derived Caco-2 cell line during T. gondii infection, microvilli, tight junction integrity, and transepithelial electrical resistance (TEER) were investigated under infection. It was observed that the dextran uptake (endocytosis) and distribution were smaller in infected than in noninfected Caco-2 cells. The infection leads to the partial loss of microvilli at the cell surface. Claudin-1, zonula occludens-1 (ZO-1), and occludin expressions were colocalized by immunofluorescence and presented discontinuous net patterns in infected cells. Immunoblotting analysis at 24 hr postinfection revealed decreasing expression of occludin and ZO-1 proteins, whereas claudin-1 presented similar expression level compared with noninfected cells. T. gondii decreased TEER in Caco-2 cells 24 hr after infection. Our results suggest that T. gondii infection may lead to the loss of integrity of intestinal mucosa, resulting in impaired barrier function.

Introduction

Toxoplasma gondii, a zoonotic Apicomplexa phylum parasite, is an obligate intracellular pathogen responsible for 10% to 50% of human infections worldwide.¹ T. gondii first invades the enterocytes and disseminates throughout the body.² The small intestinal ileum, jejunum, and duodenum offer an optimal environment for parasite replication and burden.^3,4 As a consequence, transepithelial migration of immunological cells, preferentially neutrophils and also inflammatory monocytes, represents an important mechanism for parasite spreading.⁵ Moreover, oral T. gondii infection results in intestinal tissue damage associated with necrosis in certain lineages of mice, such as C57BL/6, that die within 13 days of peroral infection with 100 cysts of ME-49 strain.⁶ T. gondii induces ileal inflammation that is accompanied by substantial mucosal barrier defects resulting in bacterial translocation.⁷

To understand how T. gondii replicates at the first week after infection,⁴ it is demonstrated that T. gondii in the gut may spread from the intestinal lumen. Naïve tissues from major histocompatibility complex class II (MHC-II) green fluorescent protein (GFP) reporter mice on a C57BL/6 background showed an intact villi monolayer. By contrast, villi from MHC-II GFP infected mice presented varied structural architectures from normal pattern to areas with indiscernible epithelial cells at 3 days postinfection. These differences were observed in addition to decreased stability of infected tissue.⁴

It is known that parasite requires intercellular adhesion molecule 1 (ICAM-1) protein to succeed in their invasion along with parasite microneme protein-2 (MIC-2). During the process of transmigration, T. gondii does not alter the integrity of host cell barrier suggesting that the parasite enters the cell via the paracellular pathway. Occludin distribution and transepithelial resistance are conserved at the moment of invasion in mammalian cell line.^8,9 However, epithelial cells derived from the crypts of Lieberkühn of the murine small intestinal epithelium infected with T. gondii showed a redistribution of occludin protein from the basolateral domain to apical plasma membrane to cytosol at 24 hr postinfection.⁹ These data lead us to ask whether claudin-1, zonula occludens-1 (ZO-1), and occludin molecules could be involved in the epithelial damage after T. gondii invasion as key players to lead cell injury.

The human colon adenocarcinoma, Caco-2 cell line, has been used to understand cancer mechanism, cell polarity, endocytosis process, and cellular differentiation because it is able to develop in a differentiated monolayer in three-dimensional (3D) culture.^10–14 Apical polarity and junctional complex are features of Caco-2 cells after 20 days of culture, which is characteristic of human enterocytes.^15,16 Therefore, Caco-2 cells seem to be a good model to explore how T. gondii infection impacts on epithelial integrity and polarity of human enterocytes.

Here, we focus to investigate whether T. gondii is involved in damage of cellular barrier integrity and redistribution of claudin-1, ZO-1, and occludin proteins and actin filaments in intestinal tissue after 24 hr postinfection. We found that Caco-2 cells lost polarity as a consequence of the decreased transepithelial resistance; defects of claudin-1, ZO-1, and occludin distribution; and leakage of dextran fluid-phase endocytosis. Moreover, cellular brush border is less developed in Caco-2 infected cells in comparison with in noninfected cells. In addition, T. gondii infection caused modification in the actin filament distribution. All these results suggest that T. gondii could initiate human cell injury through cell polarity loss and microvilli disruption.

Materials and Methods

Parasite and Cell Culture

Tachyzoites of T. gondii 2F1 strain (β-Gal), which expresses cytoplasmic β-galactosidase constitutively and is derived from RH strain, was a gift from Dr. Vern Carruthers, Medicine School of Michigan University (USA). The parasites were maintained in HeLa cells cultivated in RPMI and 2% fetal bovine serum (FBS; Cultilab, Campinas, São Paulo, Brazil), and 100 U/ml penicillin and 100 µg/ml streptomycin (Sigma, St. Louis, MO), and incubated at 37C, 5% CO₂, and 95% humidity.

Caco-2 cell line (Banco de Células do Rio de Janeiro [BCRJ]: CR059) was maintained in 25-cm² flasks containing Dulbecco’s modified Eagle’s medium (DMEM; Cultilab) supplemented with 20% FBS, 2-mM l-glutamine, and 100 U/ml penicillin and 100 µg/ml streptomycin (Sigma) at 37C in 5% CO₂ and 95% humidity. Caco-2 cells were used after 20 days when reached a confluent, polarized, and differentiated state.

T. gondii Infection Assay

Before infection of Caco-2 cells, FBS was reduced to 2%, and cells were maintained in this condition for 24 hr. Next, cells were infected with T. gondii at 5:1 (parasite:cell), followed by 3-hr incubation at 37C. Then, cells were washed with fresh DMEM and again incubated for 24 hr at 37C.

Uptake of Dextran in Caco-2 Cells

Caco-2 cells were cultivated in 13-mm coverslips in 24-well plates at a density of 5.0 × 10⁵ cells/coverslip. We used a proportion of five parasites/cell for infection procedures as mentioned above. After infection period, cells were incubated at 37C with FITC-Dextran (4 kDa; Sigma) diluted in DMEM without serum. Internalization of the tracer was tested in three concentrations of the tracer including control (0.2, 1.0, and 5.0 mg/ml). For additional experiments, the 1.0 mg/ml was used. Next, coverslips were washed three times with fresh phosphate-buffered saline (PBS), fixed with 4% formalin for 30 min at dark, and washed again three times with PBS. Finally, coverslips were mounted with 10 µl of SlowFade Gold antifade reagent (Invitrogen Life Technologies Co., Grand Island, NY), sealed with a transparent glaze, and observed in the confocal laser scanning microscopy (CLSM; Zeiss LSM 510; Carl Zeiss, Jena, Germany).

Epithelial Paracellular Permeability Analysis by Quantification of FITC-Dextran in T. gondii–Infected Caco-2 Cells

For evaluation of the permeability of the tracer FITC-Dextran that passed through the paracellular barrier in Caco-2 cells, we used a transwell system (Corning Costar Corp., New York, NY) by adding to the upper compartment the Caco-2 cells in a density of 2.0 × 10⁵ cells per well that were cultivated for 20 days. After this period, cells were infected for 24 hr with T. gondii in a proportion of five parasites per cell. To evaluate the tracer in Caco-2 cells, the cells were incubated with fresh DMEM medium without serum for 30 min, and then FITC-Dextran was added at a concentration of 1.0 mg/ml in the apical compartment of the transwell for 1 hr. Medium was collected from the upper and lower compartments, and the FITC-Dextran was measured in a fluorimeter (VersaMax ELISA Microplate Reader; Molecular Devices, Sunnyvale, CA) at 483 nm of excitation and 525 nm of emission. All assays were performed in triplicate in two different assays. The values of FITC-Dextran were calculated from a standard curve obtained with known concentrations.

Immunofluorescence for Detection of Claudin-1, ZO-1, Occludin, and Actin Proteins in Caco-2 T. gondii–Infected Cells

Caco-2 cells were cultivated for 20 days in coverslips in a 24-well plate at a density of 5.0 × 10⁵ cells/coverslip. Infection procedures were the same as described above. The coverslips were washed three times with PBS, and the cells were permeabilized with saponin for 15 min at room temperature. Nonspecific sites were blocked for 30 min at room temperature with block solution (2% BSA and 0.01% saponin diluted in PBS). Next, coverslips were incubated with rabbit anti-occludin (1:100; Life Technologies, Eugene, OR), Alexa Fluor 488 mouse monoclonal anti-ZO-1 (1:25; Life Technologies), polyclonal rabbit anti-claudin-1 (1:50; Thermo Fisher Scientific, Waltham, MA), and A3A4 monoclonal anti-SAG1¹⁷ (1:300) antibodies diluted in block solution for 1 hr at room temperature. Coverslips were washed six times with block solution, 5 min each, and corresponding secondary antibodies, CF 488A–conjugated goat anti-rabbit IgG (Sigma), Alexa Fluor 564 rabbit anti-mouse IgG (Life Technologies), Alexa Fluor 633–conjugated goat anti-mouse IgG (Life Technologies), and Alexa Fluor 564–conjugated phalloidin (Life Technologies; which has a high affinity for actin filaments), were diluted in PBS (1:300), added, and incubated for 1 hr at room temperature in the dark. Finally, coverslips were washed (PBS), fixed with 4% formalin for 5 min, washed again (PBS), and mounted with 10-µl SlowFade Gold antifade reagent (Invitrogen Life Technologies Co.). Images were obtained using CLSM (Zeiss LSM 510; Carl Zeiss).

Electrical Resistance Measurements

Cells were seeded on transwell inserts with a pore size of 0.4 µm and growth area of 0.3 cm² (Corning Costar Corp.) at a density of 3.0 × 10⁴ cells/insert and cultured for 20 days. The use of the cell culture inserts allows access to both basolateral and apical sides of the cell, which represents the circulatory and luminal poles of the intestinal epithelium.¹⁶ The transepithelial electrical resistance (TEER) of Caco-2 cells was monitored using an EVOM2 voltmeter (World Precision Instruments, FL), before infection and 20 and 24 hr after infection with T. gondii.

Immunoblotting

After 24-hr infection with T. gondii, Caco-2 cells were washed twice with PBS, sonicated (SONOPULS mini20; Bandelin, Berlin, Germany) for 30 sec in lysis buffer (10 µg/ml aprotinin, 100 µg/ml leupeptin, 1-mM benzamidine, 1.6-mM PMSF, 5-mM EDTA, 0.01% Triton X-100), and stored at −20C until used. Proteins were separated by SDS-PAGE and electrophoretically transferred onto a polyvinylidene difluoride membrane (GE Healthcare, Little Chalfont, UK). The membrane was blocked with washing buffer (20-mM Tris, 0.15-M NaCl, 5-mM EDTA, 0.5% Tween 20) containing 5% skim milk and incubated for 2 hr at room temperature. The primary antibodies for claudin-1 (Thermo Fisher Scientific), ZO-1 (Life Technologies), occludin (Life Technologies), and β-actin (Santa Cruz Biotechnology, Dallas, TX) were diluted in washing buffer containing 1% BSA at a concentration of 1:250, 1:500, 1:100, and 1:1000, respectively, and incubated overnight at 4C with gentle agitation. After that, four washing steps of 5 min each were done with washing buffer. Horseradish peroxidase secondary antibodies (goat polyclonal anti-mouse IgG and goat polyclonal anti-rabbit IgG, both from Sigma) diluted in washing buffer with 1% BSA were added at a concentration of 1:2000 and 1:4000, respectively. After incubation for 1 hr at room temperature, the membrane was then rinsed again and finally reacted using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific). Protein bands were visualized using a ChemiDoc MP system (Bio-Rad Laboratories, Montreal, Québec, Canada) and quantified using ImageJ software.

Statistical Analysis

The data were analyzed using GraphPad Prism 5.0 software package (GraphPad Software, Inc., San Diego, CA). Data were expressed as the mean ± standard error of the mean of experimental groups. Differences were considered statistically significant when p<0.05.

Results

Epithelial barrier integrity of Caco-2 cells is altered by T. gondii infection

As the natural route of T. gondii infection is the intestine, it was verified the parasite interaction with the intestinal epithelia. At first, it was analyzed the endocytosis of 4-kDa Dextran labeled fluorescently (FITC-Dextran). The determination of optimal dextran concentration for Caco-2 endocytosis assays was performed using three FITC-Dextran doses: 0.2, 1.0, and 5.0 mg/ml. At 0.2 mg/ml dextran concentration, it was observed few and poorly defined intracellular vesicles, whereas, an excessive concentration of tracer was observed at 5.0 mg/ml dose. In contrast, vesicles formation with better definition was observed at 1.0 mg/ml of dextran (Supplementary Fig. S1). Thus, this concentration was chosen for additional experiments.

After determination of optimum FITC-Dextran concentration, the effect of T. gondii on FITC-Dextran uptake in polarized Caco-2 cells was investigated 24 hr after parasite inoculation (Fig. 1A). FITC-Dextran internalization was examined by confocal microscopy at 60 min of uptake in Caco-2 cells. By examining the time of fluid-phase marker endocytosis, it was observed a decrease in the number and size of FITC-labeled vesicles in infected Caco-2 cells compared with noninfected cells (Fig. 1A). The interactive 3D surface plots on the right panels, which translate the luminescence of the image in Fig. 1A, on the left, show higher peaks of dextran labeling in noninfected cells (Fig. 1A). T. gondii parasites at 24 hr postinfection drastically decreased the entrance of FITC-Dextran at the apical surface on Caco-2 cells, evidenced by reduced number of vesicles detected by confocal microscopy.

Figure 1.

Toxoplasma gondii infection affects endocytosis and paracellular permeability of FITC-Dextran in Caco-2 cells. (A) Caco-2 cells were incubated with Dulbecco’s modified Eagle’s medium and 20% fetal bovine serum (FBS) for 20 days. Fluorescence assay was performed incubating Caco-2 cells with 1.0 mg/ml of 4-kDa FITC-Dextran for 1 hr, after 24 hr of T. gondii infection. Cells were stained for nuclei with TO-PRO-3 (blue; Life Technologies) and FITC-Dextran (green). The histograms on the right panels represent the FITC-Dextran fluorescence intensity. (B) The T. gondii infection decreases paracellular permeability of FITC-Dextran in Caco-2 cell monolayers. The apical-to-basolateral flux of FITC-Dextran through Caco-2 monolayers cultured in transwell system was monitored for 1 hr after dextran incubation in the presence or absence of tachyzoites. Result represents the mean ± standard error of the mean of three independent experiments. ***p<0.0001. Scale bar is 10 µm.

To better understand the smaller dextran internalization in infected Caco-2 cells, we evaluated alterations in the barrier function 24 hr after infection with T. gondii tachyzoites. For this purpose, Caco-2 cells were plated on the transwell system and Dextran was added in the upper compartment of the chamber. After 1-hr incubation with FITC-Dextran, infected Caco-2 monolayer showed increased permeability to FITC-Dextran (0.01258 ± 0.0009746 mg/ml) from the apical to the basolateral compartment compared with uninfected cells (0.008476 ± 0.001350 mg/ml) (Fig. 1B). These results suggest that after 24 hr of T. gondii infection, the epithelial barrier of Caco-2 enterocytes was altered showing increased paracellular permeability.

Effect of T. gondii Tachyzoites on Transepithelial Electrical Resistance

Based on the augmentation of FITC-Dextran at the basolateral compartment of cells after 24 hr of infection, the effects of T. gondii parasites were evaluated on the maintenance of transepithelial resistance. To assess the electrical resistance, Caco-2 cells were plated on the transwell system and incubated for 20 days to reach high confluence and maturation. Furthermore, transepithelial resistance was measured before and after infection with T. gondii. Our results revealed decreased electric resistance at 20 and 24 hr postinfection (Fig. 2). Twenty hours postinfection presented a decrease of 20% of TEER, whereas the TEER dropped to 50% at 24 hr postinfection compared with noninfected cells (Fig. 2), suggesting that T. gondii tachyzoites adversely affected monolayer barrier function integrity.

Figure 2.

Toxoplasma gondii tachyzoites decrease transepithelial electrical resistance (TEER) of Caco-2 monolayers. Caco-2 cells were cultured in the transwell system for 20 days in Dulbecco’s modified Eagle’s medium with 20% fetal bovine serum (FBS) and then were infected with T. gondii. The TEER was measured after 20 and 24 hr of infection. White column represents TEER of the insert with medium. Gray columns represent TEER of Caco-2 cells without tachyzoites. Black columns represent TEER of Caco-2 cells after 20 and 24 hr of infection. Data are expressed as the mean ± standard error of the mean for triplicate samples. ***p<0.0001.

Loss of Actin Architecture Upon T. gondii Infection

As we detected that the paracellular barrier and TEER were altered by T. gondii infection, we asked whether the parasite could affect Caco-2 apical domain. To observe the actin filaments of the microvilli structure, cells were stained with Alexa Fluor 594–conjugated phalloidin (Fig. 3). T. gondii infection altered the actin structure, evidenced by the loss of microvilli and cytoskeleton organization in comparison with noninfected cells, as shown in the apical and basolateral images (Fig. 3A). The ZEN 2012 SP2 (Carl Zeiss) software was used to analyze confocal microscope images, allowing a relative quantification of actin filaments based on fluorescence transmitted by Alexa Fluor 594 (Fig. 3B). Full image projection of Caco-2 cells monolayer revealed a high density of actin filaments without T. gondii infection. Instead, full image projection of Caco-2 cells infected with T. gondii indicated a lack of actin net microvilli (Fig. 3B). High fluorescence intensity in noninfected cells was observed in contrast with infected ones using relative quantification based on 3D surface plots. Considering the images obtained, a relevant alteration on the actin cytoskeleton was observed 24 hr postinfection in intestinal cells, where microvilli of the brush border formed by actin filaments at the apical domain were less developed or absent compared with noninfected cells. Moreover, at the basolateral domain, the actin filaments seem to be thinner than noninfected cells.

Figure 3.

Loss of actin filaments in the presence of Toxoplasma gondii infection. Caco-2 cells were cultured in coverslips for 20 days in Dulbecco’s modified Eagle’s medium supplemented with 20% of fetal bovine serum (FBS). The cells were infected with T. gondii, and 24 hr later, the actin cytoskeleton was analyzed using Alexa Fluor 594–conjugated phalloidin (which has a high affinity for actin filaments). The coverslips were observed in a confocal microscope. (A) Confocal microscopy images of microvilli and actin cytoskeleton of Caco-2 cells after 24 hr of T. gondii infection. At the apical surface, on the left and upper panel, phalloidin was detected in developed microvilli structures on noninfected Caco-2 monolayer (arrows). At right side, poorly developed microvilli on the apical surface of infected Caco-2 cells can be observed by phalloidin detection (arrows). At the basolateral surface, the actin filament net was detected by phalloidin. On left panel (below), a thick net of actin filament is revealed, whereas a weak net is detected in the infected cells (right panel, below). (B) Optical x–y sections from full projection of the epithelium are presented by captured images of phalloidin and T. gondii detection using SAG-1 antibody (top side). The histograms show the relative quantification analysis of phalloidin–actin detection (bottom side). The relative quantification of the full projection images confirms the decrease of actin filament on infected cells. Scale bar is 10 µm.

T. gondii–Infected Caco-2 Cells Present Disorganization and Redistribution of Tight Junction Proteins, Claudin-1, Occludin, and ZO-1

The tight junction (TJ) is formed by a protein complex that is widely used as a marker of maturation and cell polarization.¹⁸ Our data showed that the cytoskeleton of the infected cells has been compromised by the presence of the parasite; thus, we postulated that the integrity of the TJs in the apical membrane could have been altered. The integrity of the TJs in Caco-2 cells infected with T. gondii was examined by confocal microscopy through analyses of immunostaining of the integral membrane proteins, claudin-1, ZO-1, and occludin (Fig. 4). The proteins were observed as a continuous and organized net pattern surrounding the plasma membrane constituting a cell–cell junction (Fig. 4A, D, and G). After 24 hr of T. gondii infection, Caco-2 cells presented a depletion of claudin-1, ZO-1, and occludin in several sites with a discontinued net pattern compared with noninfected ones (Fig. 4J, M, and P). Parasites were observed inside cells nearby TJ proteins and also in some areas colocalized with the discontinuous net pattern of TJs (Fig. 4L, O, and R). In addition, cytoplasmic accumulation of TJ proteins in some regions of infected cells was also observed (Fig. 4J, M, and P). To analyze whether the levels of TJ proteins were altered in infected cells, the protein expression was detected by immunoblotting analysis in Caco-2 infected and noninfected cells (Fig. 4S). It was observed that at 24 hr of T. gondii infection, the expression levels of occludin and ZO-1 were decreased; however, claudin-1 expression level was similar to noninfected cells (Fig. 4T). These data suggest that the TJ’s integrity has been altered with a redistribution of the claudin-1, occludin, and ZO-1 24 hr after infection.

Figure 4.

Expressions of claudin-1, occludin, and zonula occludens-1 (ZO-1) are altered in Caco-2 cells 24 hr after Toxoplasma gondii infection. Caco-2 cells were cultured in coverslips for 20 days in Dulbecco’s modified Eagle’s medium supplemented with 20% of fetal bovine serum (FBS). The cells were infected with T. gondii for 24 hr and immunostained with specific antibodies to detect claudin-1, ZO-1, and occludin. Continuous net pattern of claudin-1, ZO-1, and occludin in Caco-2 cells (panels A, C, D, F, G, I) is observed after 20 days of culture. Discontinuous and disorganized net patterns of claudin-1, ZO-1, and occludin (panels J, L, M, O, P, R) in infected cells are indicated with asterisks. Immunofluorescence assays for claudin-1, ZO-1, and occludin were performed using rabbit polyclonal anti-claudin-1 (green), mouse monoclonal anti-ZO-1 (green), and rabbit anti-occludin (green). Panels A to C, D to F, and G to I represent claudin-1, ZO-1, and occludin immunostaining, respectively. B, E, and H images represent negative SAG-1 immunolocalization in noninfected cells. C, F, and I images are a merge of images from noninfected cells. Panels J, L; M, O; and P, R represent claudin-1, ZO-1, and occludin immunostaining, respectively, in infected cells. K, N, and Q images show T. gondii parasites (red) that were labeled with mouse monoclonal anti-SAG-1 antibody, and L, O, and R merged pictures show SAG-1 immunolocalization within Caco-2 cells. The immunoblot analysis was done to measure the claudin-1, ZO-1, and occludin proteins expression in infected and noninfected cells (S). Whole cell lyses were separated on SDS-PAGE gel and immunoblotted with antibodies against the corresponding tight junction proteins, ZO-1, occludin, and claudin-1, followed by chemiluminescence detection. Protein bands were visualized using a ChemiDoc MP system. Quantification of immunoblotting by using β-actin as an internal control in ImageJ software (T). Scale bar is 10 µm.

Discussion

Recently, it has been shown that T. gondii (RH) tachyzoites infect and transmigrate between adjacent intestinal epithelial cells in polarized monolayers without altering barrier integrity.⁹ However, after oral ingestion of T. gondii, C57BL/6 mice and other animal species develop intestinal pathology.¹⁹ Thus, we became interested in investigating whether the parasite would interfere in the intestinal barrier of the enterocytes after invasion. The landmark of epithelial cell differentiation is the maturation of the apical domain, which includes specialized absorptive epithelia, maturation of microvilli into the brush border as well as high endocytic activity.²⁰ Endocytosis encompasses several mechanisms by which cells internalize extracellular macromolecules or particles.²¹ In the present study, we have observed decreased number and size of FITC-labeled vesicles in infected Caco-2 cells 24 hr after parasite inoculation, suggesting that the parasite altered the endocytosis mechanisms of this cell lineage. In addition, we have also detected higher concentrations of the 4-kDa FITC-Dextran at the basolateral compartment, indicating paracellular permeability and the breakage of the intestinal barrier.

Endocytosis of fluid-phase markers such as FITC-Dextran is related to actin cytoskeleton.²² Our results showed that T. gondii infection altered the actin structure of Caco-2 cells 24 hr after infection. Although other studies have implicated actin in endocytosis at the apical cell surface,^23–25 it was demonstrated that polymerization of actin enhances the endocytosis of fluid-phase markers such as FITC-Dextran from the basolateral membrane of polarized cells.²⁶ It has been previously shown that Toxoplasma tachyzoites secret Toxofilin, an actin-binding protein which upregulates actin turnover, increasing depolymerization of the filaments on the area of invasion, facilitating T. gondii internalization.²⁷

It was previously shown that another protozoan parasite, Giardia intestinalis, which colonizes and causes intestinal disease, also induces disorganization of the apical F-actin cytoskeleton in Caco-2 monolayer in an adhesion-dependent manner, leading to the impaired distribution of functional brush border–associated proteins.²⁸ The binary toxin, Clostridium difficile transferase, induces actin depolymerization that alter the microtubule and endoplasmic reticulum organization, increasing pathogen adherence in Caco-2 cell line.²⁹ The present investigation demonstrated that Toxoplasma tachyzoites appeared to have an effect on the arrangement of actin cytoskeleton of cells, and a direct consequence over FITC-Dextran endocytosis.

TJ proteins have been widely studied and considered promising targets to understand parasite invasion and infection.^30–32 TJs act as a semipermeable barrier (or gate) to the paracellular transport of ions, solutes, and water, as well as cells, and are considered to function as a fence that divides apical and basolateral domains of plasma membranes.³³ Several proteins, such as occludin, claudins, and ZOs, are associated and maintain the structure of the TJ.^34–36 Within 1 hr of peroral Toxoplasma infection, there was a striking redistribution of the TJ protein, occludin to the apical surface of epithelial intestinal cells in mice, which was characterized by intense and almost continuous staining of occludin in the apical surface of each villus.³⁷ T. gondii tachyzoites invade small intestinal epithelial cell monolayers via the paracellular pathway in a process that appears to involve interactions with occludin, as specific small interfering RNAs significantly reduced the ability of T. gondii to penetrate between and infect epithelial cells.⁹ In the present study, we have observed that Caco-2 cells presented discontinuous circumferential TJ proteins staining 24 hr after T. gondii infection. Thus, after invasion, the parasite was able to disturb the TJ proteins, inducing a decrease in occludin and ZO-1 expression, and redistribution of claudin-1, occludin, and ZO-1 in the cytoplasm of intestinal epithelial cells. In accordance, Salmonella typhi and Shigella infections lead to claudin-1, ZO-1 and occludin redistribution, occludin phosphorylation and disengagement from the TJ complex.^38,39 Also, Yersinia enterocolitica induces epithelial barrier dysfunction through regional TJ changes.⁴⁰ These findings suggest that TJ proteins are frequent targets of intestinal pathogens in the process of invasion and infection. It is known that NF-κB enhances the permeability of T84 or Caco-2 cells, decreasing the levels of TJ proteins and disturbance in TJ localization.^41,42 However, the mechanism(s) used by T. gondii to disrupt TJs in Caco-2 cells has yet to be determined.

The findings that endocytosis, actin cytoskeleton, and TJ proteins redistribution were altered by T. gondii suggest that the mucosal barrier integrity is affected by Toxoplasma tachyzoites. To measure the integrity of TJ dynamics in a cell culture model, TEER is largely accepted as a quantitative method for this purpose.⁴³ Our results demonstrated that T. gondii infection decreased the TEER of Caco-2 cells resulting in loss of enterocytes’ polarity. Intestinal pathogenic bacteria and protozoan, such as enterotoxigenic Escherichia coli, Salmonella typhimurium, and Giardia intestinalis, also decrease TEER using human enterocyte-like Caco-2 cells as experimental model.^44,45 Consistent with our results, virulence factors of Listeria monocytogenes, such as the pore-forming toxin listeriolysin O, induced a significant drop in TEER in a dose-dependent manner in the Caco-2 monolayer.⁴⁶

The integrity of the intestinal epithelial cell layer is crucial for morphological function of the intestinal tissue, and cell polarity, microvilli, and actin cytoskeleton play an essential role on cell activity.⁴⁷ The present study used an in vitro model to investigate how T. gondii explores Caco-2 enterocytes. No longer than 24 hr after infection, the parasite is able to disturb endocytosis and cell permeability. Furthermore, our results showed for the first time that microvilli of Caco-2 cells were disrupted by T. gondii. Moreover, parasitized Caco-2 cells presented disruption of cell polarity and loss of transepithelial resistance, explaining the intestinal tissue injuries caused by T. gondii.

In conclusion, we have identified that T. gondii infection has an impact on the intestinal epithelial barrier that involves impaired endocytosis, disrupted actin cytoskeleton organization, and redistribution of TJ proteins leading to discontinuous net patterns, resulting in structural defects of the epithelial cell monolayer. Our findings highlight the consequences of T. gondii infection on the host intestinal microenvironment.