Interaction of Cryptosporidium hominis and Cryptosporidium parvum with Primary Human and Bovine Intestinal Cells
http://www.100md.com
感染与免疫杂志 2006年第1期
The Children's Research Centre, Our Lady's Hospital for Sick Children, Crumlin, Dublin 12, Ireland
UCD School of Medicine and Medical Science
UCD School of Agriculture, Food Science and Veterinary Medicine
UCD Conway Institute of Biomolecular and Biomedical Science, University College Dublin, Dublin 4, Ireland
ABSTRACT
Cryptosporidiosis in humans is caused by the zoonotic pathogen Cryptosporidium parvum and the anthroponotic pathogen Cryptosporidium hominis. To what extent the recently recognized C. hominis species differs from C. parvum is unknown. In this study we compared the mechanisms of C. parvum and C. hominis invasion using a primary cell model of infection. Cultured primary bovine and human epithelial intestinal cells were infected with C. parvum or C. hominis. The effects of the carbohydrate lectin galactose-N-acetylgalactosamine (Gal/GalNAc) and inhibitors of cytoskeletal function and signal transduction mechanisms on entry of the parasites into host cells were tested. HCT-8 cells (human ileocecal adenocarcinoma cells) were used for the purpose of comparison. Pretreatment of parasites with Gal/GalNAc inhibited entry of C. parvum into HCT-8 cells and primary bovine cells but had no effect on entry of either C. parvum or C. hominis into primary human cells or on entry of C. hominis into HCT-8 cells. Both Cryptosporidium species entered primary cells by a protein kinase C (PKC)- and actin-dependent mechanism. Staurosporine, in particular, attenuated infection, likely through a combination of PKC inhibition and induction of apoptosis. Diversity in the mechanisms used by Cryptosporidium species to infect cells of different origins has important implications for understanding the relevance of in vitro studies of Cryptosporidium pathogenesis.
INTRODUCTION
Cryptosporidium is an obligate enteric parasite of the phylum Apicomplexa and an important cause of diarrheal disease worldwide. The infection is transmitted by fecal, food, and waterborne routes and is initiated when sporozoites released from oocysts in the intestinal tract attach to and invade mucosal epithelial cells. In susceptible immunocompetent humans (mainly children), infection leads to a self-limiting diarrhea. In immunocompromised patients, especially AIDS patients and those with congenital immune deficiencies, chronic diarrhea frequently develops and leads to considerable fluid and weight loss and even death. In addition to the upper intestine, the parasite can also infect the stomach, pancreas, liver, and bile ducts of immunocompromised hosts. Although promising chemotherapeutic treatments against cryptosporidiosis are beginning to appear (22), elimination of the parasite is extremely difficult and recurrence is common.
The genus Cryptosporidium consists of various different species and genotypes that infect a wide range of hosts. Originally, almost all isolates from human patients were assigned to Cryptosporidium parvum. Within C. parvum, two genotypes were distinguished: the "human genotype," type I, and the "cattle genotype," type II, both of which are capable of initiating human infection. The two genotypes differ significantly in their host range and genetics (24, 31) and maintain separate reproductive cycles (27). They are now widely recognized as two separate species, Cryptosporidium hominis (formerly known as type I) and C. parvum (formerly known as type II). So far, natural infections of C. hominis have only been reported from humans, although experimental infections have been carried out in neonatal pigs (21). In contrast, C. parvum infects most, if not all, mammals, including humans, and is a major pathogen of calves. Our knowledge of how these species differ in relation to host range, infectivity, or pathogenicity is only beginning to evolve.
Generally, humans are infected with C. hominis in an anthroponotic cycle and with C. parvum in a zoonotic cycle. Most work to date on the biology and pathogenicity of Cryptosporidium in relation to human infection has been done using C. parvum. Therefore, compared to C. parvum little is known of the biology of invasion of the human-restricted C. hominis. In a recent study we compared the pathogenesis of C. hominis and C. parvum using both HCT-8 cells and primary culture of both bovine and human intestinal cells. C. hominis infection of HCT-8 cells differed from C. parvum infection (17). Entry of C. hominis into HCT-8 cells was less efficient than C. parvum and showed a distinct focal pattern of infection in the monolayer, whereas C. parvum infection of HCT-8 cells was evenly distributed throughout the monolayer. In contrast to HCT-8 cells, there was no difference between the two species of Cryptosporidium during infection of primary human intestinal cells. However, only C. parvum was capable of infecting primary bovine cells. These data suggested that the species restriction of C. hominis is due to host tissue tropism of the infecting isolate and that different adhesin-receptor interactions may promote entry of these organisms in cells of different origins.
Identification of the parasite and host molecules that mediate the initial host-parasite interactions during host-cell invasion is crucial for designing preventive and interventional strategies to combat cryptosporidiosis. A number of sporozoite glycoproteins are known to act as mediators of C. parvum attachment in vitro including a galactose-N-acetylgalactosamine (Gal/GalNAc)-specific sporozoite epitope (7, 18, 19, 26). Studies have shown that sporozoite motility and invasion are dependent on parasite and host cytoskeletal elements (7, 14, 15). In the present study we have investigated the role of a Gal/GalNAc-specific lectin carbohydrate interaction in mediating colonization of primary human cells by C. hominis and C. parvum and colonization of primary bovine cells by C. parvum. In addition, we have studied the cytoskeletal and signal transduction events that occur following invasion of primary human and bovine intestinal cells by C. hominis and C. parvum.
MATERIALS AND METHODS
C. parvum and C. hominis. C. parvum oocysts (Iowa strain) were obtained from a commercial source (Pleasant Hill Farms, Troy, ID). This strain was originally isolated from a calf by Harley Moon. It has been passaged through calves and purified from the fecal material by ether extraction, followed by a one-step sucrose gradient.
C. hominis TU502 oocysts were a gift from Donna Akiyoshi, Tufts University School of Veterinary Medicine. Boston, Mass. C. hominis strain TU502 is a well-characterized isolate that has been passaged through gnotobiotic piglets (2), and its genome has been sequenced (32). Before host cells were infected, oocysts were decontaminated. Briefly, they were washed twice with distilled water and then incubated with freshly prepared 10% (vol/vol) Clorox bleach (Sigma-Aldrich) for 10 min on ice. Following two further washes with ice-cold distilled water and one with RPMI 1640 medium (Bio-Whittaker), oocysts were resuspended at a concentration of 2 x 105 oocysts/ml and then used to infect cell monolayers.
Cell culture. Human ileocecal adenocarcinoma cells (HCT-8) (ATCC CCL 244) were obtained from the American Type Culture Collection. Cells were maintained in 75-cm2 tissue culture flasks in RPMI 1640 medium and 10% (vol/vol) fetal bovine serum (Sigma). The cells were grown as adherent monolayers at 37°C in a 5% CO2-95% air humidified incubator. Twenty-four hours prior to infection, the monolayers were trypsinized with trypsin-EDTA (Bio-Whittaker) for 15 min at 37°C. The cells were grown on 25-mm Thermonox coverslips in six-well Costar tissue culture plates (Gibco) for 24 h, at which time they were 80 to 85% confluent.
Isolation and culture of primary human and bovine intestinal epithelial cells. Human and bovine intestinal epithelial cells were isolated as described previously (17). Following approval by the ethics committee at the hospital and parental consent, human small-bowel biopsy tissue was obtained from children undergoing endoscopy at Our Lady's Hospital for Sick Children. Only children undergoing endoscopy for clinical reasons were recruited (e.g., investigation of abdominal pain, gastroesophageal reflux, and failure to thrive). Only grossly normal tissue was sampled. Bovine small-duodenal tissue was removed from cattle under 30 months of age immediately after slaughter at a local abattoir. Cells from both tissues were then isolated using similar procedures. Briefly, the tissue was washed with Hank's balanced salt solution (Bio-Whittaker) with Ca2+ or Mg2+ containing 0.1 mM EDTA (Sigma) and 0.1 mM dithiothreitol (Sigma) at 37°C for 10 min with vigorous shaking. The crypts and cells were isolated using 0.05% (wt/vol) collagenase (Sigma). The isolated cells and crypts were kept on ice until ready for use. A total of 500 μl was plated in 24-well Costar culture plates on 13-mm plastic coverslips using Dulbecco's modified Eagle's medium-Ham's F-12 with 10% fetal bovine serum, 8 μg/ml insulin, 10 μg/ml gentamicin, 50 μg/ml hydrocortisone, 100 μg/ml streptomycin, 100 U/ml penicillin, and 2.5 μg/ml amphotericin B. Cells were grown for 24 to 48 h at 37°C in a 5% CO2-95% air humidified incubator, at which time isolated cells and crypts attached to the coverslips. Cells grew as isolated epithelial colonies and propagated by growing out from the crypts. The epithelial origin of the cultured cells was demonstrated by immunofluorescence staining with a monoclonal anti-pan cytokeratin (Sigma) as previously described (17).
Infection of HCT-8 cells and primary human and bovine intestinal cells. Maintenance medium was removed from the cells, and 2 x 105 Cryptosporidium oocysts in 2 ml of growth medium were added to the cells. The plates were placed at 37°C in a 5% CO2-95% air humidified incubator. After incubation for 3 h, the infected cells were washed with phosphate-buffered saline (PBS) to remove unexcysted oocysts, empty oocyst walls, and toxic materials that may have been liberated from the oocysts. Then, 2 ml of growth medium containing 100 U of penicillin/ml and 100 μg of streptomycin/ml was added. Primary cells were infected for 24 h, and HCT-8 cells were infected for up to 72 h. During this time period the cells remained viable and did not appear to be damaged by infection.
Lectin VVL staining. Parasites were detected in the monolayers using lectin VVL (a plant lectin from Vicia villosa) staining, as previously described (16). Monolayers were fixed with 4% formaldehyde in PBS for 30 min, permeabilized with 1% Triton X-100 in PBS for 10 min, and blocked with 0.5% (wt/vol) bovine serum albumin for a further 10 min. They were then incubated with 1 μg/ml of conjugated lectin VVL-biotin (B1235; Vector Laboratories) in 0.5% (wt/vol) bovine serum albumin followed by 1 μg/ml of conjugated fluorophore streptavidin-CY3 (Sigma). The coverslips were mounted using fluorescence mounting medium (Dako) and examined by fluorescence microscopy using a 50x water immersion lens. The numbers of infected cells in 10 high-power fields were counted and the percent rate of infection ([number of infected cells/total number of cells] x 100) was calculated. All experiments were performed in triplicate and expressed as the mean percentage of cells infected ± the standard deviation of the mean. Uninfected monolayers were used as a control.
Colocalization of host cell actin and Cryptosporidium. In order to stain cells for both actin and parasite, cells were stained with lectin VVL as described above with a slight modification. After permeabilization with Triton X-100, cells were incubated with 5 μg/ml phalloidin conjugated to fluorescein isothiocyanate (FITC) (Sigma) together with 1 μg/ml lectin VVL for 30 min. Slides were examined using a laser scanning confocal microscope (Bio-Rad MRC 1024).
Effect of Gal/GalNAc on C. hominis and C. parvum entry of HCT-8 cells and primary human and bovine intestinal cells. The effect of Gal/GalNAc and bovine submaxillary mucin (BSM; BSM contains Gal/GalNAc) on entry of sporozoites of C. hominis and C. parvum to HCT-8 cells and primary human and bovine intestinal cells was tested. Sporozoites were prepared by excystation of oocysts using a previously described method (28). Oocysts were suspended in 1 ml of PBS and incubated for 60 min at 37°C using a water bath. The suspension was pelleted by centrifugation at 5,000 x g for 5 min at 4°C using a microcentrifuge (Sigma). The supernatant was discarded, and the pellet was resuspended in 1 ml of isotonic Percoll solution and centrifuged at 5,000 x g for 5 min at 4°C. The pellet contained some sporozoites and intact oocysts. Immediately above the pellet was a buffy coat consisting of highly enriched sporozoites, while oocyst walls lay on top of the Percoll. By using a micropipette tip, the sporozoites were aspirated into a 1.5-ml Eppendorf tube. They were resuspended in four times their volume with PBS and were spun at 5,000 x g for 5 min at 4°C. Sporozoites were then incubated with 10 μM and 5 μM Gal/GalNAc and 0.1 mg/ml and 0.01 mg/ml BSM for 1 hour at 37°C. The sporozoites were then washed twice with PBS and used to infect HCT-8 cells and primary human and bovine intestinal cells. Infected HCT-8 cells and primary human and bovine intestinal cells were stained with lectin VVL as described above, and the number of infected cells in a total of 250 cells was counted and expressed as the infection rate. Untreated sporozoites were used as controls.
Effect of inhibitors of cytoskeletal components on C. parvum and C. hominis entry of HCT-8 cells and primary human and bovine intestinal cells. Stock solutions of 1 mg/ml of cytochalasin D and 5 mg/ml of cytochalasin B (Sigma) (actin inhibitors) were prepared and diluted to their final concentrations using culture medium. Three different experiments were performed. In the first set of experiments, cells were incubated for 1 h in an assay medium containing either 1 μg/ml cytochalasin D or 10 μg/ml cytochalasin B (7) for 1 h at 37°C (30). The treated cells were washed twice with PBS and infected with 2 x 105 C. parvum oocysts for either 48 h or 22 h. In a second set of experiments, cells were infected as previously described, but the cytoskeletal inhibitors were present throughout the infection period. Finally, in order to assess the effect of cytoskeletal inhibition on the parasite, oocysts were incubated with 1 μg/ml cytochalasin D or 10 μg/ml cytochalasin B for 15 min at 37°C. Oocysts were then washed twice with PBS before being used to infect monolayers. Untreated cells or oocysts were used as controls.
Effect of signal transduction inhibitors on C. hominis and C. parvum invasion of HCT-8 cells and primary human and bovine intestinal cells. A number of signal transduction inhibitors were used to assess the importance of various host signal transduction pathways during invasion of C. hominis and C. parvum into host cells. Stock solutions of different inhibitors (Table 1) were prepared according to the manufacturers' recommendations and diluted to their final concentrations using culture medium. The final concentrations tested were the concentrations recommended by the manufacturers for inhibition of cellular function. Cells were incubated with the different inhibitors for 1 h. The cells were then washed twice with PBS and were infected with oocysts for up to 48 h. Untreated monolayers were used as controls.
Statistical analysis. The above results were expressed as the mean percentage of cells infected ± standard deviation of triplicate experiments. Means were compared using a nonparametric Mann-Whitney U test. A P value of <0.05 was considered statistically significant.
Annexin V and PI staining. Annexin V and propidium iodide (PI) staining was used to detect apoptotic cells. Cells were treated with different concentrations of staurosporine for 1 h, trypsinized with trypsin-EDTA from the bottom of 24-well plates, washed twice in ice-cold PBS, and resuspended in binding buffer (10 mM HEPES-NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) at a concentration of 106 cells/ml. One hundred microliters of cells was transferred to a FACScan tube to which were added 10 μl of FITC-conjugated annexin V solution (10 μg/ml) (IQ Corp., Groningen, The Netherlands) and 10 μl of PI (50 μg/ml). Annexin V is a phosphatidylserine binding protein that detects phosphatidylserine on the surface of cells undergoing apoptosis, whereas PI associates with nuclear DNA, indicating necrotic cell death. Cells were vortexed gently and then incubated at room temperature for 15 min in the dark. Then, 400 μl of binding buffer was added to the cells, and they were analyzed by flow cytometry within 1 h of staining. The data from 10,000 cells were collected and analyzed using LYSIS II software (Becton Dickinson). The signals for green (FL1; annexin V) and orange fluorescence (FL2; PI) were measured by logarithmic amplification.
RESULTS
Role of Gal/GalNAc-specific lectin carbohydrate interaction in mediating attachment of C. hominis and C. parvum to HCT-8 cells and primary human and bovine intestinal cells. In a recent study (17) we have shown that restriction of C. hominis to human hosts reflects species-specific tissue tropism of this species. To investigate the basis for this tropism, the role of a Gal/GalNAc-specific lectin-carbohydrate interaction in mediating attachment to host cells was examined. Pretreatment of sporozoites with both BSM and Gal/GalNAc itself reduced invasion of primary bovine cells and HCT-8 cells by C. parvum but did not affect entry into primary human cells. In addition, pretreatment of C. hominis sporozoites with both BSM and Gal/GalNAc did not affect entry into either HCT-8 cells or into primary human cells (Fig. 1). These data suggest that C. parvum enters HCT-8 cells and bovine cells using a Gal/GalNAc-specific lectin-carbohydrate interaction. However, entry of C. parvum and C. hominis into primary human cells is Gal/GalNAc independent. In addition, C. hominis enters HCT-8 cells via a Gal/GalNAc-independent mechanism.
Role of cytoskeletal components during infection of primary intestinal cells with C. parvum and C. hominis. Previous studies using cultured cell lines have demonstrated that host actin polymerization occurs at the host-parasite interface during C. parvum (12) and C. hominis (17) invasion and that inhibition of polymerization inhibits parasite invasion. To investigate whether host or parasite actin are involved during entry of C. hominis and C. parvum into primary human and bovine intestinal epithelial cells, cells or parasites were treated with actin inhibitors prior to infection. Pretreatment of host cells with cytochalasin D and cytochalasin B significantly reduced invasion of C. parvum into primary human and bovine cells. The inhibitory effect of cytochalasin D and cytochalasin B on C. parvum invasion into primary human cells and bovine cells was more pronounced when both inhibitors were left throughout the infection period (Fig. 2). Similarly, pretreatment of host cells with cytochalasin D and cytochalasin B significantly reduced invasion of C. hominis into primary human cells (Fig. 2). Treatment of cells with cytochalasins did not affect cell viability or induce any gross morphological changes in the cells. However, pretreatment of oocysts of both species with cytochalasin D or cytochalasin B for 15 min prior to infection did not affect invasion (results not shown). Double staining of C. parvum- or C. hominis-infected primary human cells with phalloidin conjugated to FITC and lectin VVL demonstrated host actin accumulation in association with internalized sporozoites (Fig. 3).
Signal transduction pathways involved during infection of HCT-8 cells by C. parvum and C. hominis. To assess the signal transduction pathways involved during infection of HCT-8 cells by C. parvum and C. hominis, cells were treated with a variety of protein kinase inhibitors for 1 h prior to infection. Exposure of cells to staurosporine (a general protein kinase inhibitor that inhibits tyrosine kinase, protein kinase C [PKC] and serine/threonine kinases) completely abolished entry of C. parvum and C. hominis into HCT-8 cells (Fig. 4). Two PKC inhibitors, calphostin C and G6976, had an inhibitory effect on invasion of C. hominis. In addition the PKC inhibitors chlerythrine chloride and G6976 and the serine/threonine kinase inhibitor KN-93 had inhibitory effects on C. parvum invasion of cells (Fig. 5). However, it was noteworthy that none of these inhibitors reduced C. parvum or C. hominis invasion of HCT-8 cells to the levels seen following staurosporine treatment. Moreover, combining PKC inhibitors with genistein or with serine/threonine kinase inhibitors could not reproduce the complete inhibition seen with staurosporine (Tables 2 and 3).These data suggested that staurosporine was affecting invasion by an additional or alternative mechanism. As C. parvum has been shown not to invade apoptotic cells (8, 29) and staurosporine is a known apoptosis-inducing agent (10), we examined if it could induce apoptosis in HCT-8 cells. The number of apoptotic cells increased from 19.5% for control (untreated cells) to a maximum of 67.4% as the concentration of staurosporine was increased to 2.5 μM (Fig. 6).
Signal transduction pathways involved during infection of primary intestinal cells by C. parvum and C. hominis. Primary cells were more resistant to the effect of staurosporine than HCT-8 cells. A concentration of 5 μM was required to completely abolish invasion of all primary cells by either C. parvum or C. hominis. Staurosporine at a concentration of 2.5 μM reduced C. parvum infection of primary human cells from 97.3% ± 0.2% to 60.4% ± 0.4% and of bovine cells from 93.6% ± 0.2% to 51.3% ± 0.1%, while C. hominis entry into primary human cells was reduced from 95.2% ± 0.2% to 60.0% ± 0.2%. This inhibitory effect was mimicked by the PKC inhibitor G6976 that reduced C. parvum infection of human cells from 97.4% ± 0.2% to 49.6% ± 1.6% and of bovine cells from 92.7% ± 0.7% to 48.3% ± 1.5%, while C. hominis infection of human cells was reduced from 95.2% ± 0.2% for control cells to 60.0% ± 0.2% (Fig. 7). None of the other tyrosine phosphorylation inhibitors, PKC inhibitors, serine/threonine kinase inhibitors, wortmannin, or suramin sodium salt had any effect on entry of C. parvum into primary human or bovine cells or on entry of C. hominis into primary human cells (data not shown).
DISCUSSION
In this study a primary cell model was used to study the interactions of C. parvum and C. hominis with host primary intestinal epithelial cells. Primary intestinal cells have a number of advantages over immortalized tissue culture cell lines for the study of host-parasite interactions in vitro. They may more accurately reflect in vivo conditions than immortalized cells. In particular, primary cells allow direct and meaningful examination ex vivo of species tropism and the importance of specific-receptor ligand interaction, as they are likely to occur in vivo.
In order to investigate the basis for species tropism, the mechanisms by which C. parvum and C. hominis enter human and bovine cells were examined. It has been shown previously that Gal/GalNAc-specific lectin-carbohydrate interactions play a role in mediating attachment of C. parvum to intestinal and biliary epithelial cells (7). Therefore, we wanted to investigate whether Gal/GalNAc and BSM could be used to block the ligand on the parasite surface prior to infection of primary human and bovine intestinal cells. The results obtained showed that C. parvum entry into primary bovine and HCT-8 cells is dependent on the Gal/GalNAc-specific lectin-carbohydrate interaction. However, neither C. parvum nor C. hominis entry into human cells involved such an interaction. In addition, C. hominis entry into HCT-8 cells was Gal/GalNAc independent. These data clearly underline the differences between the invasion mechanisms used by the two species of parasite and the differences between the different cell types. These results also provide an explanation for the findings of our earlier study (17), where we showed that C. hominis did not invade primary bovine cells and that there was a different pattern of infection when HCT-8 cells were used (infection localized to certain cells) compared to when primary human cells were used (infection dispersed evenly throughout the monolayer). These results suggest that C. hominis may be using a pathway distinct from that of C. parvum for infection. Alternatively, both C. parvum and C. hominis may have evolved a specialized mechanism of host-parasite interaction specific for infecting human intestinal cells. Comparison of the genomes of C. parvum and C. hominis showed that the two genomes are very similar, exhibiting only 3 to 5% sequence divergence, with no large insertions, deletions, or rearrangements evident (32). Extensive arrays of potentially variant surface proteins were not observed in either the C. hominis or the C. parvum genome (1, 32). The identification of the surface molecules which interact with human cells and/or the receptor(s) on human cells for C. parvum and C. hominis could potentially be a significant development in the search for therapeutics to specifically treat infections.
Studies have shown that C. parvum induces host cell actin rearrangement upon invasion of different cell lines and that this process is necessary for invasion to occur (6, 9, 12, 13).We have previously shown that this also occurs when C. hominis enters HCT-8 cells (17). Our demonstration of the colocalization of C. parvum and C. hominis with host actin using primary human intestinal cells corroborates the importance of this cellular event previously observed in immortalized cells during infection with Cryptosporidium. Inhibition of entry of both species to primary cells by the cytoskeletal inhibitors cytochalasin B and cytochalasin D underlines the importance of actin rearrangement for successful invasion of the host cell.
A previous study has shown that treatment of primary bovine fallopian tube epithelial cells with genistein, staurosporine, suramin sodium salt, or wortmannin inhibited entry of C. parvum sporozoites (14). C. parvum sporozoites were shown to induce tyrosine phosphorylation of cortactin in a human bile duct epithelial cell line, and inhibition of c-Src, a host protein tyrosine kinase inhibitor, inhibited invasion (5). In contrast, Elliot and Clark infected HCT-8 cells with C. parvum oocysts and were unable to demonstrate tyrosine phosphorylation at the site of developing trophozoites and merozoites (12). We found that only staurosporine had an effect on entry of C. parvum or C. hominis oocysts into HCT-8 cells and into primary bovine or human intestinal epithelial cells. The effect of staurosporine was shown to be mediated in primary cells through inhibition of a PKC signaling pathway. However, the effect of staurosporine on HCT-8 cells could only be partly mimicked by a PKC inhibitor. A plausible mechanism to explain the potency of staurosporine compared with other PKC inhibitors is the induction of apoptosis (10). However, whether apoptosis or some other mechanism underlies these different effects remains to be proven. These results highlight the importance of the type of host cell used to study the invasion mechanisms of Cryptosporidium species and suggest that some of the conflicting data in the literature about the role of different signaling pathways in the invasion process may be due to the use of a variety of host cells for infection studies. The use of sporozoites in some assays and oocysts in others to initiate infections could possibly also explain some conflicting results. Some inhibitors may be more potent when sporozoites are used as they can adhere to and invade host cells faster than oocysts, which need to excyst prior to infection. In the time that it takes the oocysts to excyst, it is possible that the host cells may recover from the effect of the inhibitor.
Our results clearly point toward a role for the classical Ca2+-dependent PKC isoenzyme PKC and/or PKC in the invasion process by C. parvum of primary human and primary bovine intestinal cells and C. hominis entry into primary human intestinal cells. PKC is a ubiquitous phospholipid-dependent serine/threonine kinase involved in major signaling events that regulate a wide variety of biological responses to stimuli (11). It has been shown that the discharge of C. parvum sporozoite apical organelle contents is dependent on both intracellular Ca2+ and the cytoskeleton and is required for host cell invasion (9). We have also found that chelation of intracellular Ca2+ inhibits entry of C. parvum oocysts into HCT-8 cells (unpublished observation). Interestingly, PKC has been shown to play an important role in the invasion of human brain microvascular endothelial cells by Escherichia coli. Like Cryptosporidium, E. coli induces host actin condensation at the site of infection when it invades host cells. Immunocytochemical studies indicated that activated PKC is associated with actin condensation beneath the bacterial entry site (25). Further studies are required to investigate whether a similar situation occurs upon invasion of host cells by Cryptosporidium species. The PKC inhibitor used in this study, G6976, has been shown to promote the formation of tight junctions in urinary bladder carcinoma cells. Specifically, G6976 was shown to induce the formation of adherens and desmosomal cell-cell junctions (20). Disruption of tight junctions is known to contribute to intestinal disorders caused by enteric pathogens (4). Infection of bovine and human epithelial cell lines with Cryptosporidium andersoni disrupted tight junctional zonula occludens 1 (3). PKC expression has been correlated with tight junctional leakiness in renal epithelial cells (23). It is plausible that the induction of tight junctions in host cells and inhibition of disruption of those junctions is a mechanism that could prevent entry of C. parvum and C. hominis.
This study demonstrates the importance of the model system used to study Cryptosporidium invasion of host cells and suggests that the use of primary cells, although technically more difficult to work with, may provide a more biologically relevant model system than conventional cell lines. There is a need now to identify the potential receptor-ligand interactions that promote C. hominis invasion of human intestinal epithelial cells. The identification of such receptors and further characterization of the signaling pathways used by C. parvum and C. hominis to enter primary human cells could lead to the development of new therapeutic agents for the treatment of infection.
ACKNOWLEDGMENTS
This work was supported by grants from the Health Research Board, Ireland, and The Children's Medical and Research Foundation, Crumlin, Dublin 12, Ireland.
REFERENCES
1. Abrahamsen, M. S., T. J. Templeton, S. Enomoto, J. E. Abrahante, G. Zhu, C. A. Lancto, M. Deng, C. Liu, G. Widmer, S. Tzipori, G. A. Buck, P. Xu, A. T. Bankier, P. H. Dear, B. A. Konfortov, H. F. Spriggs, L. Iyer, V. Anantharaman, L. Aravind, and V. Kapur. 2004. Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 304:441-445.
2. Akiyoshi, D. E., X. Feng, M. A. Buckholt, G. Widmer, and S. Tzipori. 2002. Genetic analysis of a Cryptosporidium parvum human genotype 1 isolate passaged through different host species. Infect. Immun. 70:5670-5675.
3. Buret, A. G., A. C. Chin, and K. G. Scott. 2003. Infection of human and bovine epithelial cells with Cryptosporidium andersoni induces apoptosis and disrupts tight junctional ZO-1: effects of epidermal growth factor. Int. J. Parasitol. 33:1363-1371.
4. Chen, M. L., C. Pothoulakis, and J. T. LaMont. 2002. Protein kinase C signaling regulates ZO-1 translocation and increased paracellular flux of T84 colonocytes exposed to Clostridium difficile toxin A. J. Biol. Chem. 277:4247-4254.
5. Chen, X. M., B. Q. Huang, P. L. Splinter, H. Cao, G. Zhu, M. A. McNiven, and N. F. LaRusso. 2003. Cryptosporidium parvum invasion of biliary epithelia requires host cell tyrosine phosphorylation of cortactin via c-Src. Gastroenterology 125:216-228.
6. Chen, X. M., B. Q. Huang, P. L. Splinter, J. D. Orth, D. D. Billadeau, M. A. McNiven, and N. F. LaRusso. 2004. Cdc42 and the actin-related protein/neural Wiskott-Aldrich syndrome protein network mediate cellular invasion by Cryptosporidium parvum. Infect. Immun. 72:3011-3021.
7. Chen, X. M., and N. F. LaRusso. 2000. Mechanisms of attachment and internalization of Cryptosporidium parvum to biliary and intestinal epithelial cells. Gastroenterology 118:368-379.
8. Chen, X. M., S. A. Levine, P. L. Splinter, P. S. Tietz, A. L. Ganong, C. Jobin, G. J. Gores, C. V. Paya, and N. F. LaRusso. 2001. Cryptosporidium parvum activates nuclear factor kappaB in biliary epithelia preventing epithelial cell apoptosis. Gastroenterology 120:1774-1783.
9. Chen, X. M., S. P. O'Hara, B. Q. Huang, J. B. Nelson, J. J. Lin, G. Zhu, H. D. Ward, and N. F. LaRusso. 2004. Apical organelle discharge by Cryptosporidium parvum is temperature, cytoskeleton, and intracellular calcium dependent and required for host cell invasion. Infect. Immun. 72:6806-6816.
10. Couldwell, W. T., D. R. Hinton, S. He, T. C. Chen, I. Sebat, M. H. Weiss, and R. E. Law. 1994. Protein kinase C inhibitors induce apoptosis in human malignant glioma cell lines. FEBS Lett. 345:43-46.
11. Dempsey, E. C., A. C. Newton, D. Mochly-Rosen, A. P. Fields, M. E. Reyland, P. A. Insel, and R. O. Messing. 2000. Protein kinase C isozymes and the regulation of diverse cell responses. Am. J. Physiol. Lung Cell Mol. Physiol. 279:L429-L438.
12. Elliott, D. A., and D. P. Clark. 2000. Cryptosporidium parvum induces host cell actin accumulation at the host-parasite interface. Infect. Immun. 68:2315-2322.
13. Elliott, D. A., D. J. Coleman, M. A. Lane, R. C. May, L. M. Machesky, and D. P. Clark. 2001. Cryptosporidium parvum infection requires host cell actin polymerization. Infect. Immun. 69:5940-5942.
14. Forney, J. R., D. B. DeWald, S. Yang, C. A. Speer, and M. C. Healey. 1999. A role for host phosphoinositide 3-kinase and cytoskeletal remodeling during Cryptosporidium parvum infection. Infect. Immun. 67:844-852.
15. Forney, J. R., D. K. Vaughan, S. Yang, and M. C. Healey. 1998. Actin-dependent motility in Cryptosporidium parvum sporozoites. J. Parasitol. 84:908-913.
16. Gut, J., and R. G. Nelson. 1999. Cryptosporidium parvum: synchronized encystation in vitro and evaluation of sporozoite infectivity with a new lectin-based assay. J. Eukaryot. Microbiol. 46:56S-57S.
17. Hashim, A., M. Clyne, G. Mulcahy, D. Akiyoshi, R. Chalmers, and B. Bourke. 2004. Host cell tropism underlies species restriction of human and bovine Cryptosporidium parvum genotypes. Infect. Immun. 72:6125-6131.
18. Joe, A., D. H. Hamer, M. A. Kelley, M. E. Pereira, G. T. Keusch, S. Tzipori, and H. D. Ward. 1994. Role of a Gal/GalNAc-specific sporozoite surface lectin in Cryptosporidium parvum-host cell interaction. J. Eukaryot. Microbiol. 41:44S.
19. Joe, A., R. Verdon, S. Tzipori, G. T. Keusch, and H. D. Ward. 1998. Attachment of Cryptosporidium parvum sporozoites to human intestinal epithelial cells. Infect. Immun. 66:3429-3432.
20. Koivunen, J., V. Aaltonen, S. Koskela, P. Lehenkari, M. Laato, and J. Peltonen. 2004. Protein kinase C alpha/beta inhibitor Go6976 promotes formation of cell junctions and inhibits invasion of urinary bladder carcinoma cells. Cancer Res. 64:5693-5701.
21. Morgan-Ryan, U. M., A. Fall, L. A. Ward, N. Hijjawi, I. Sulaiman, R. Fayer, R. C. Thompson, M. Olson, A. Lal, and L. Xiao. 2002. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporidiidae) from Homo sapiens. J. Eukaryot. Microbiol. 49:433-440.
22. Rossignol, J. F., A. Ayoub, and M. S. Ayers. 2001. Treatment of diarrhea caused by Cryptosporidium parvum: a prospective randomized, double-blind, placebo-controlled study of Nitazoxanide. J. Infect. Dis. 184:103-106.
23. Rosson, D., T. G. O'Brien, J. A. Kampherstein, Z. Szallasi, K. Bogi, P. M. Blumberg, and J. M. Mullin. 1997. Protein kinase C-alpha activity modulates transepithelial permeability and cell junctions in the LLC-PK1 epithelial cell line. J. Biol. Chem. 272:14950-14953.
24. Spano, F., L. Putignani, J. McLauchlin, D. P. Casemore, and A. Crisanti. 1997. PCR-RFLP analysis of the Cryptosporidium oocyst wall protein (COWP) gene discriminates between C. wrairi and C. parvum, and between C. parvum isolates of human and animal origin. FEMS Microbiol. Lett. 150:209-217.
25. Sukumaran, S. K., and N. V. Prasadarao. 2002. Regulation of protein kinase C in Escherichia coli K1 invasion of human brain microvascular endothelial cells. J. Biol. Chem. 277:12253-12262.
26. Thea, D. M., M. E. Pereira, D. Kotler, C. R. Sterling, and G. T. Keusch. 1992. Identification and partial purification of a lectin on the surface of the sporozoite of Cryptosporidium parvum. J. Parasitol. 78:886-893.
27. Tzipori, S., and H. Ward. 2002. Cryptosporidiosis: biology, pathogenesis and disease. Microbes Infect. 4:1047-1058.
28. Upton, S. 1997. In vitro cultivation, p. 181-207. In R. Fayer (ed.), Cryptosporidium and cryptosporidiosis. CRC Press, Boca Raton, Fla.
29. Widmer, G., E. A. Corey, B. Stein, J. K. Griffiths, and S. Tzipori. 2000. Host cell apoptosis impairs Cryptosporidium parvum development in vitro. J. Parasitol. 86:922-928.
30. Wiest, P. M., J. H. Johnson, and T. P. Flanigan. 1993. Microtubule inhibitors block Cryptosporidium parvum infection of a human enterocyte cell line. Infect. Immun. 61:4888-4890.
31. Xiao, L., U. M. Morgan, J. Limor, A. Escalante, M. Arrowood, W. Shulaw, R. C. Thompson, R. Fayer, and A. A. Lal. 1999. Genetic diversity within Cryptosporidium parvum and related Cryptosporidium species. Appl. Environ. Microbiol. 65:3386-3391.
32. Xu, P., G. Widmer, Y. Wang, L. S. Ozaki, J. M. Alves, M. G. Serrano, D. Puiu, P. Manque, D. Akiyoshi, A. J. Mackey, W. R. Pearson, P. H. Dear, A. T. Bankier, D. L. Peterson, M. S. Abrahamsen, V. Kapur, S. Tzipori, and G. A. Buck. 2004. The genome of Cryptosporidium hominis. Nature 431:1107-1112.(Amna Hashim, Grace Mulcah)
UCD School of Medicine and Medical Science
UCD School of Agriculture, Food Science and Veterinary Medicine
UCD Conway Institute of Biomolecular and Biomedical Science, University College Dublin, Dublin 4, Ireland
ABSTRACT
Cryptosporidiosis in humans is caused by the zoonotic pathogen Cryptosporidium parvum and the anthroponotic pathogen Cryptosporidium hominis. To what extent the recently recognized C. hominis species differs from C. parvum is unknown. In this study we compared the mechanisms of C. parvum and C. hominis invasion using a primary cell model of infection. Cultured primary bovine and human epithelial intestinal cells were infected with C. parvum or C. hominis. The effects of the carbohydrate lectin galactose-N-acetylgalactosamine (Gal/GalNAc) and inhibitors of cytoskeletal function and signal transduction mechanisms on entry of the parasites into host cells were tested. HCT-8 cells (human ileocecal adenocarcinoma cells) were used for the purpose of comparison. Pretreatment of parasites with Gal/GalNAc inhibited entry of C. parvum into HCT-8 cells and primary bovine cells but had no effect on entry of either C. parvum or C. hominis into primary human cells or on entry of C. hominis into HCT-8 cells. Both Cryptosporidium species entered primary cells by a protein kinase C (PKC)- and actin-dependent mechanism. Staurosporine, in particular, attenuated infection, likely through a combination of PKC inhibition and induction of apoptosis. Diversity in the mechanisms used by Cryptosporidium species to infect cells of different origins has important implications for understanding the relevance of in vitro studies of Cryptosporidium pathogenesis.
INTRODUCTION
Cryptosporidium is an obligate enteric parasite of the phylum Apicomplexa and an important cause of diarrheal disease worldwide. The infection is transmitted by fecal, food, and waterborne routes and is initiated when sporozoites released from oocysts in the intestinal tract attach to and invade mucosal epithelial cells. In susceptible immunocompetent humans (mainly children), infection leads to a self-limiting diarrhea. In immunocompromised patients, especially AIDS patients and those with congenital immune deficiencies, chronic diarrhea frequently develops and leads to considerable fluid and weight loss and even death. In addition to the upper intestine, the parasite can also infect the stomach, pancreas, liver, and bile ducts of immunocompromised hosts. Although promising chemotherapeutic treatments against cryptosporidiosis are beginning to appear (22), elimination of the parasite is extremely difficult and recurrence is common.
The genus Cryptosporidium consists of various different species and genotypes that infect a wide range of hosts. Originally, almost all isolates from human patients were assigned to Cryptosporidium parvum. Within C. parvum, two genotypes were distinguished: the "human genotype," type I, and the "cattle genotype," type II, both of which are capable of initiating human infection. The two genotypes differ significantly in their host range and genetics (24, 31) and maintain separate reproductive cycles (27). They are now widely recognized as two separate species, Cryptosporidium hominis (formerly known as type I) and C. parvum (formerly known as type II). So far, natural infections of C. hominis have only been reported from humans, although experimental infections have been carried out in neonatal pigs (21). In contrast, C. parvum infects most, if not all, mammals, including humans, and is a major pathogen of calves. Our knowledge of how these species differ in relation to host range, infectivity, or pathogenicity is only beginning to evolve.
Generally, humans are infected with C. hominis in an anthroponotic cycle and with C. parvum in a zoonotic cycle. Most work to date on the biology and pathogenicity of Cryptosporidium in relation to human infection has been done using C. parvum. Therefore, compared to C. parvum little is known of the biology of invasion of the human-restricted C. hominis. In a recent study we compared the pathogenesis of C. hominis and C. parvum using both HCT-8 cells and primary culture of both bovine and human intestinal cells. C. hominis infection of HCT-8 cells differed from C. parvum infection (17). Entry of C. hominis into HCT-8 cells was less efficient than C. parvum and showed a distinct focal pattern of infection in the monolayer, whereas C. parvum infection of HCT-8 cells was evenly distributed throughout the monolayer. In contrast to HCT-8 cells, there was no difference between the two species of Cryptosporidium during infection of primary human intestinal cells. However, only C. parvum was capable of infecting primary bovine cells. These data suggested that the species restriction of C. hominis is due to host tissue tropism of the infecting isolate and that different adhesin-receptor interactions may promote entry of these organisms in cells of different origins.
Identification of the parasite and host molecules that mediate the initial host-parasite interactions during host-cell invasion is crucial for designing preventive and interventional strategies to combat cryptosporidiosis. A number of sporozoite glycoproteins are known to act as mediators of C. parvum attachment in vitro including a galactose-N-acetylgalactosamine (Gal/GalNAc)-specific sporozoite epitope (7, 18, 19, 26). Studies have shown that sporozoite motility and invasion are dependent on parasite and host cytoskeletal elements (7, 14, 15). In the present study we have investigated the role of a Gal/GalNAc-specific lectin carbohydrate interaction in mediating colonization of primary human cells by C. hominis and C. parvum and colonization of primary bovine cells by C. parvum. In addition, we have studied the cytoskeletal and signal transduction events that occur following invasion of primary human and bovine intestinal cells by C. hominis and C. parvum.
MATERIALS AND METHODS
C. parvum and C. hominis. C. parvum oocysts (Iowa strain) were obtained from a commercial source (Pleasant Hill Farms, Troy, ID). This strain was originally isolated from a calf by Harley Moon. It has been passaged through calves and purified from the fecal material by ether extraction, followed by a one-step sucrose gradient.
C. hominis TU502 oocysts were a gift from Donna Akiyoshi, Tufts University School of Veterinary Medicine. Boston, Mass. C. hominis strain TU502 is a well-characterized isolate that has been passaged through gnotobiotic piglets (2), and its genome has been sequenced (32). Before host cells were infected, oocysts were decontaminated. Briefly, they were washed twice with distilled water and then incubated with freshly prepared 10% (vol/vol) Clorox bleach (Sigma-Aldrich) for 10 min on ice. Following two further washes with ice-cold distilled water and one with RPMI 1640 medium (Bio-Whittaker), oocysts were resuspended at a concentration of 2 x 105 oocysts/ml and then used to infect cell monolayers.
Cell culture. Human ileocecal adenocarcinoma cells (HCT-8) (ATCC CCL 244) were obtained from the American Type Culture Collection. Cells were maintained in 75-cm2 tissue culture flasks in RPMI 1640 medium and 10% (vol/vol) fetal bovine serum (Sigma). The cells were grown as adherent monolayers at 37°C in a 5% CO2-95% air humidified incubator. Twenty-four hours prior to infection, the monolayers were trypsinized with trypsin-EDTA (Bio-Whittaker) for 15 min at 37°C. The cells were grown on 25-mm Thermonox coverslips in six-well Costar tissue culture plates (Gibco) for 24 h, at which time they were 80 to 85% confluent.
Isolation and culture of primary human and bovine intestinal epithelial cells. Human and bovine intestinal epithelial cells were isolated as described previously (17). Following approval by the ethics committee at the hospital and parental consent, human small-bowel biopsy tissue was obtained from children undergoing endoscopy at Our Lady's Hospital for Sick Children. Only children undergoing endoscopy for clinical reasons were recruited (e.g., investigation of abdominal pain, gastroesophageal reflux, and failure to thrive). Only grossly normal tissue was sampled. Bovine small-duodenal tissue was removed from cattle under 30 months of age immediately after slaughter at a local abattoir. Cells from both tissues were then isolated using similar procedures. Briefly, the tissue was washed with Hank's balanced salt solution (Bio-Whittaker) with Ca2+ or Mg2+ containing 0.1 mM EDTA (Sigma) and 0.1 mM dithiothreitol (Sigma) at 37°C for 10 min with vigorous shaking. The crypts and cells were isolated using 0.05% (wt/vol) collagenase (Sigma). The isolated cells and crypts were kept on ice until ready for use. A total of 500 μl was plated in 24-well Costar culture plates on 13-mm plastic coverslips using Dulbecco's modified Eagle's medium-Ham's F-12 with 10% fetal bovine serum, 8 μg/ml insulin, 10 μg/ml gentamicin, 50 μg/ml hydrocortisone, 100 μg/ml streptomycin, 100 U/ml penicillin, and 2.5 μg/ml amphotericin B. Cells were grown for 24 to 48 h at 37°C in a 5% CO2-95% air humidified incubator, at which time isolated cells and crypts attached to the coverslips. Cells grew as isolated epithelial colonies and propagated by growing out from the crypts. The epithelial origin of the cultured cells was demonstrated by immunofluorescence staining with a monoclonal anti-pan cytokeratin (Sigma) as previously described (17).
Infection of HCT-8 cells and primary human and bovine intestinal cells. Maintenance medium was removed from the cells, and 2 x 105 Cryptosporidium oocysts in 2 ml of growth medium were added to the cells. The plates were placed at 37°C in a 5% CO2-95% air humidified incubator. After incubation for 3 h, the infected cells were washed with phosphate-buffered saline (PBS) to remove unexcysted oocysts, empty oocyst walls, and toxic materials that may have been liberated from the oocysts. Then, 2 ml of growth medium containing 100 U of penicillin/ml and 100 μg of streptomycin/ml was added. Primary cells were infected for 24 h, and HCT-8 cells were infected for up to 72 h. During this time period the cells remained viable and did not appear to be damaged by infection.
Lectin VVL staining. Parasites were detected in the monolayers using lectin VVL (a plant lectin from Vicia villosa) staining, as previously described (16). Monolayers were fixed with 4% formaldehyde in PBS for 30 min, permeabilized with 1% Triton X-100 in PBS for 10 min, and blocked with 0.5% (wt/vol) bovine serum albumin for a further 10 min. They were then incubated with 1 μg/ml of conjugated lectin VVL-biotin (B1235; Vector Laboratories) in 0.5% (wt/vol) bovine serum albumin followed by 1 μg/ml of conjugated fluorophore streptavidin-CY3 (Sigma). The coverslips were mounted using fluorescence mounting medium (Dako) and examined by fluorescence microscopy using a 50x water immersion lens. The numbers of infected cells in 10 high-power fields were counted and the percent rate of infection ([number of infected cells/total number of cells] x 100) was calculated. All experiments were performed in triplicate and expressed as the mean percentage of cells infected ± the standard deviation of the mean. Uninfected monolayers were used as a control.
Colocalization of host cell actin and Cryptosporidium. In order to stain cells for both actin and parasite, cells were stained with lectin VVL as described above with a slight modification. After permeabilization with Triton X-100, cells were incubated with 5 μg/ml phalloidin conjugated to fluorescein isothiocyanate (FITC) (Sigma) together with 1 μg/ml lectin VVL for 30 min. Slides were examined using a laser scanning confocal microscope (Bio-Rad MRC 1024).
Effect of Gal/GalNAc on C. hominis and C. parvum entry of HCT-8 cells and primary human and bovine intestinal cells. The effect of Gal/GalNAc and bovine submaxillary mucin (BSM; BSM contains Gal/GalNAc) on entry of sporozoites of C. hominis and C. parvum to HCT-8 cells and primary human and bovine intestinal cells was tested. Sporozoites were prepared by excystation of oocysts using a previously described method (28). Oocysts were suspended in 1 ml of PBS and incubated for 60 min at 37°C using a water bath. The suspension was pelleted by centrifugation at 5,000 x g for 5 min at 4°C using a microcentrifuge (Sigma). The supernatant was discarded, and the pellet was resuspended in 1 ml of isotonic Percoll solution and centrifuged at 5,000 x g for 5 min at 4°C. The pellet contained some sporozoites and intact oocysts. Immediately above the pellet was a buffy coat consisting of highly enriched sporozoites, while oocyst walls lay on top of the Percoll. By using a micropipette tip, the sporozoites were aspirated into a 1.5-ml Eppendorf tube. They were resuspended in four times their volume with PBS and were spun at 5,000 x g for 5 min at 4°C. Sporozoites were then incubated with 10 μM and 5 μM Gal/GalNAc and 0.1 mg/ml and 0.01 mg/ml BSM for 1 hour at 37°C. The sporozoites were then washed twice with PBS and used to infect HCT-8 cells and primary human and bovine intestinal cells. Infected HCT-8 cells and primary human and bovine intestinal cells were stained with lectin VVL as described above, and the number of infected cells in a total of 250 cells was counted and expressed as the infection rate. Untreated sporozoites were used as controls.
Effect of inhibitors of cytoskeletal components on C. parvum and C. hominis entry of HCT-8 cells and primary human and bovine intestinal cells. Stock solutions of 1 mg/ml of cytochalasin D and 5 mg/ml of cytochalasin B (Sigma) (actin inhibitors) were prepared and diluted to their final concentrations using culture medium. Three different experiments were performed. In the first set of experiments, cells were incubated for 1 h in an assay medium containing either 1 μg/ml cytochalasin D or 10 μg/ml cytochalasin B (7) for 1 h at 37°C (30). The treated cells were washed twice with PBS and infected with 2 x 105 C. parvum oocysts for either 48 h or 22 h. In a second set of experiments, cells were infected as previously described, but the cytoskeletal inhibitors were present throughout the infection period. Finally, in order to assess the effect of cytoskeletal inhibition on the parasite, oocysts were incubated with 1 μg/ml cytochalasin D or 10 μg/ml cytochalasin B for 15 min at 37°C. Oocysts were then washed twice with PBS before being used to infect monolayers. Untreated cells or oocysts were used as controls.
Effect of signal transduction inhibitors on C. hominis and C. parvum invasion of HCT-8 cells and primary human and bovine intestinal cells. A number of signal transduction inhibitors were used to assess the importance of various host signal transduction pathways during invasion of C. hominis and C. parvum into host cells. Stock solutions of different inhibitors (Table 1) were prepared according to the manufacturers' recommendations and diluted to their final concentrations using culture medium. The final concentrations tested were the concentrations recommended by the manufacturers for inhibition of cellular function. Cells were incubated with the different inhibitors for 1 h. The cells were then washed twice with PBS and were infected with oocysts for up to 48 h. Untreated monolayers were used as controls.
Statistical analysis. The above results were expressed as the mean percentage of cells infected ± standard deviation of triplicate experiments. Means were compared using a nonparametric Mann-Whitney U test. A P value of <0.05 was considered statistically significant.
Annexin V and PI staining. Annexin V and propidium iodide (PI) staining was used to detect apoptotic cells. Cells were treated with different concentrations of staurosporine for 1 h, trypsinized with trypsin-EDTA from the bottom of 24-well plates, washed twice in ice-cold PBS, and resuspended in binding buffer (10 mM HEPES-NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) at a concentration of 106 cells/ml. One hundred microliters of cells was transferred to a FACScan tube to which were added 10 μl of FITC-conjugated annexin V solution (10 μg/ml) (IQ Corp., Groningen, The Netherlands) and 10 μl of PI (50 μg/ml). Annexin V is a phosphatidylserine binding protein that detects phosphatidylserine on the surface of cells undergoing apoptosis, whereas PI associates with nuclear DNA, indicating necrotic cell death. Cells were vortexed gently and then incubated at room temperature for 15 min in the dark. Then, 400 μl of binding buffer was added to the cells, and they were analyzed by flow cytometry within 1 h of staining. The data from 10,000 cells were collected and analyzed using LYSIS II software (Becton Dickinson). The signals for green (FL1; annexin V) and orange fluorescence (FL2; PI) were measured by logarithmic amplification.
RESULTS
Role of Gal/GalNAc-specific lectin carbohydrate interaction in mediating attachment of C. hominis and C. parvum to HCT-8 cells and primary human and bovine intestinal cells. In a recent study (17) we have shown that restriction of C. hominis to human hosts reflects species-specific tissue tropism of this species. To investigate the basis for this tropism, the role of a Gal/GalNAc-specific lectin-carbohydrate interaction in mediating attachment to host cells was examined. Pretreatment of sporozoites with both BSM and Gal/GalNAc itself reduced invasion of primary bovine cells and HCT-8 cells by C. parvum but did not affect entry into primary human cells. In addition, pretreatment of C. hominis sporozoites with both BSM and Gal/GalNAc did not affect entry into either HCT-8 cells or into primary human cells (Fig. 1). These data suggest that C. parvum enters HCT-8 cells and bovine cells using a Gal/GalNAc-specific lectin-carbohydrate interaction. However, entry of C. parvum and C. hominis into primary human cells is Gal/GalNAc independent. In addition, C. hominis enters HCT-8 cells via a Gal/GalNAc-independent mechanism.
Role of cytoskeletal components during infection of primary intestinal cells with C. parvum and C. hominis. Previous studies using cultured cell lines have demonstrated that host actin polymerization occurs at the host-parasite interface during C. parvum (12) and C. hominis (17) invasion and that inhibition of polymerization inhibits parasite invasion. To investigate whether host or parasite actin are involved during entry of C. hominis and C. parvum into primary human and bovine intestinal epithelial cells, cells or parasites were treated with actin inhibitors prior to infection. Pretreatment of host cells with cytochalasin D and cytochalasin B significantly reduced invasion of C. parvum into primary human and bovine cells. The inhibitory effect of cytochalasin D and cytochalasin B on C. parvum invasion into primary human cells and bovine cells was more pronounced when both inhibitors were left throughout the infection period (Fig. 2). Similarly, pretreatment of host cells with cytochalasin D and cytochalasin B significantly reduced invasion of C. hominis into primary human cells (Fig. 2). Treatment of cells with cytochalasins did not affect cell viability or induce any gross morphological changes in the cells. However, pretreatment of oocysts of both species with cytochalasin D or cytochalasin B for 15 min prior to infection did not affect invasion (results not shown). Double staining of C. parvum- or C. hominis-infected primary human cells with phalloidin conjugated to FITC and lectin VVL demonstrated host actin accumulation in association with internalized sporozoites (Fig. 3).
Signal transduction pathways involved during infection of HCT-8 cells by C. parvum and C. hominis. To assess the signal transduction pathways involved during infection of HCT-8 cells by C. parvum and C. hominis, cells were treated with a variety of protein kinase inhibitors for 1 h prior to infection. Exposure of cells to staurosporine (a general protein kinase inhibitor that inhibits tyrosine kinase, protein kinase C [PKC] and serine/threonine kinases) completely abolished entry of C. parvum and C. hominis into HCT-8 cells (Fig. 4). Two PKC inhibitors, calphostin C and G6976, had an inhibitory effect on invasion of C. hominis. In addition the PKC inhibitors chlerythrine chloride and G6976 and the serine/threonine kinase inhibitor KN-93 had inhibitory effects on C. parvum invasion of cells (Fig. 5). However, it was noteworthy that none of these inhibitors reduced C. parvum or C. hominis invasion of HCT-8 cells to the levels seen following staurosporine treatment. Moreover, combining PKC inhibitors with genistein or with serine/threonine kinase inhibitors could not reproduce the complete inhibition seen with staurosporine (Tables 2 and 3).These data suggested that staurosporine was affecting invasion by an additional or alternative mechanism. As C. parvum has been shown not to invade apoptotic cells (8, 29) and staurosporine is a known apoptosis-inducing agent (10), we examined if it could induce apoptosis in HCT-8 cells. The number of apoptotic cells increased from 19.5% for control (untreated cells) to a maximum of 67.4% as the concentration of staurosporine was increased to 2.5 μM (Fig. 6).
Signal transduction pathways involved during infection of primary intestinal cells by C. parvum and C. hominis. Primary cells were more resistant to the effect of staurosporine than HCT-8 cells. A concentration of 5 μM was required to completely abolish invasion of all primary cells by either C. parvum or C. hominis. Staurosporine at a concentration of 2.5 μM reduced C. parvum infection of primary human cells from 97.3% ± 0.2% to 60.4% ± 0.4% and of bovine cells from 93.6% ± 0.2% to 51.3% ± 0.1%, while C. hominis entry into primary human cells was reduced from 95.2% ± 0.2% to 60.0% ± 0.2%. This inhibitory effect was mimicked by the PKC inhibitor G6976 that reduced C. parvum infection of human cells from 97.4% ± 0.2% to 49.6% ± 1.6% and of bovine cells from 92.7% ± 0.7% to 48.3% ± 1.5%, while C. hominis infection of human cells was reduced from 95.2% ± 0.2% for control cells to 60.0% ± 0.2% (Fig. 7). None of the other tyrosine phosphorylation inhibitors, PKC inhibitors, serine/threonine kinase inhibitors, wortmannin, or suramin sodium salt had any effect on entry of C. parvum into primary human or bovine cells or on entry of C. hominis into primary human cells (data not shown).
DISCUSSION
In this study a primary cell model was used to study the interactions of C. parvum and C. hominis with host primary intestinal epithelial cells. Primary intestinal cells have a number of advantages over immortalized tissue culture cell lines for the study of host-parasite interactions in vitro. They may more accurately reflect in vivo conditions than immortalized cells. In particular, primary cells allow direct and meaningful examination ex vivo of species tropism and the importance of specific-receptor ligand interaction, as they are likely to occur in vivo.
In order to investigate the basis for species tropism, the mechanisms by which C. parvum and C. hominis enter human and bovine cells were examined. It has been shown previously that Gal/GalNAc-specific lectin-carbohydrate interactions play a role in mediating attachment of C. parvum to intestinal and biliary epithelial cells (7). Therefore, we wanted to investigate whether Gal/GalNAc and BSM could be used to block the ligand on the parasite surface prior to infection of primary human and bovine intestinal cells. The results obtained showed that C. parvum entry into primary bovine and HCT-8 cells is dependent on the Gal/GalNAc-specific lectin-carbohydrate interaction. However, neither C. parvum nor C. hominis entry into human cells involved such an interaction. In addition, C. hominis entry into HCT-8 cells was Gal/GalNAc independent. These data clearly underline the differences between the invasion mechanisms used by the two species of parasite and the differences between the different cell types. These results also provide an explanation for the findings of our earlier study (17), where we showed that C. hominis did not invade primary bovine cells and that there was a different pattern of infection when HCT-8 cells were used (infection localized to certain cells) compared to when primary human cells were used (infection dispersed evenly throughout the monolayer). These results suggest that C. hominis may be using a pathway distinct from that of C. parvum for infection. Alternatively, both C. parvum and C. hominis may have evolved a specialized mechanism of host-parasite interaction specific for infecting human intestinal cells. Comparison of the genomes of C. parvum and C. hominis showed that the two genomes are very similar, exhibiting only 3 to 5% sequence divergence, with no large insertions, deletions, or rearrangements evident (32). Extensive arrays of potentially variant surface proteins were not observed in either the C. hominis or the C. parvum genome (1, 32). The identification of the surface molecules which interact with human cells and/or the receptor(s) on human cells for C. parvum and C. hominis could potentially be a significant development in the search for therapeutics to specifically treat infections.
Studies have shown that C. parvum induces host cell actin rearrangement upon invasion of different cell lines and that this process is necessary for invasion to occur (6, 9, 12, 13).We have previously shown that this also occurs when C. hominis enters HCT-8 cells (17). Our demonstration of the colocalization of C. parvum and C. hominis with host actin using primary human intestinal cells corroborates the importance of this cellular event previously observed in immortalized cells during infection with Cryptosporidium. Inhibition of entry of both species to primary cells by the cytoskeletal inhibitors cytochalasin B and cytochalasin D underlines the importance of actin rearrangement for successful invasion of the host cell.
A previous study has shown that treatment of primary bovine fallopian tube epithelial cells with genistein, staurosporine, suramin sodium salt, or wortmannin inhibited entry of C. parvum sporozoites (14). C. parvum sporozoites were shown to induce tyrosine phosphorylation of cortactin in a human bile duct epithelial cell line, and inhibition of c-Src, a host protein tyrosine kinase inhibitor, inhibited invasion (5). In contrast, Elliot and Clark infected HCT-8 cells with C. parvum oocysts and were unable to demonstrate tyrosine phosphorylation at the site of developing trophozoites and merozoites (12). We found that only staurosporine had an effect on entry of C. parvum or C. hominis oocysts into HCT-8 cells and into primary bovine or human intestinal epithelial cells. The effect of staurosporine was shown to be mediated in primary cells through inhibition of a PKC signaling pathway. However, the effect of staurosporine on HCT-8 cells could only be partly mimicked by a PKC inhibitor. A plausible mechanism to explain the potency of staurosporine compared with other PKC inhibitors is the induction of apoptosis (10). However, whether apoptosis or some other mechanism underlies these different effects remains to be proven. These results highlight the importance of the type of host cell used to study the invasion mechanisms of Cryptosporidium species and suggest that some of the conflicting data in the literature about the role of different signaling pathways in the invasion process may be due to the use of a variety of host cells for infection studies. The use of sporozoites in some assays and oocysts in others to initiate infections could possibly also explain some conflicting results. Some inhibitors may be more potent when sporozoites are used as they can adhere to and invade host cells faster than oocysts, which need to excyst prior to infection. In the time that it takes the oocysts to excyst, it is possible that the host cells may recover from the effect of the inhibitor.
Our results clearly point toward a role for the classical Ca2+-dependent PKC isoenzyme PKC and/or PKC in the invasion process by C. parvum of primary human and primary bovine intestinal cells and C. hominis entry into primary human intestinal cells. PKC is a ubiquitous phospholipid-dependent serine/threonine kinase involved in major signaling events that regulate a wide variety of biological responses to stimuli (11). It has been shown that the discharge of C. parvum sporozoite apical organelle contents is dependent on both intracellular Ca2+ and the cytoskeleton and is required for host cell invasion (9). We have also found that chelation of intracellular Ca2+ inhibits entry of C. parvum oocysts into HCT-8 cells (unpublished observation). Interestingly, PKC has been shown to play an important role in the invasion of human brain microvascular endothelial cells by Escherichia coli. Like Cryptosporidium, E. coli induces host actin condensation at the site of infection when it invades host cells. Immunocytochemical studies indicated that activated PKC is associated with actin condensation beneath the bacterial entry site (25). Further studies are required to investigate whether a similar situation occurs upon invasion of host cells by Cryptosporidium species. The PKC inhibitor used in this study, G6976, has been shown to promote the formation of tight junctions in urinary bladder carcinoma cells. Specifically, G6976 was shown to induce the formation of adherens and desmosomal cell-cell junctions (20). Disruption of tight junctions is known to contribute to intestinal disorders caused by enteric pathogens (4). Infection of bovine and human epithelial cell lines with Cryptosporidium andersoni disrupted tight junctional zonula occludens 1 (3). PKC expression has been correlated with tight junctional leakiness in renal epithelial cells (23). It is plausible that the induction of tight junctions in host cells and inhibition of disruption of those junctions is a mechanism that could prevent entry of C. parvum and C. hominis.
This study demonstrates the importance of the model system used to study Cryptosporidium invasion of host cells and suggests that the use of primary cells, although technically more difficult to work with, may provide a more biologically relevant model system than conventional cell lines. There is a need now to identify the potential receptor-ligand interactions that promote C. hominis invasion of human intestinal epithelial cells. The identification of such receptors and further characterization of the signaling pathways used by C. parvum and C. hominis to enter primary human cells could lead to the development of new therapeutic agents for the treatment of infection.
ACKNOWLEDGMENTS
This work was supported by grants from the Health Research Board, Ireland, and The Children's Medical and Research Foundation, Crumlin, Dublin 12, Ireland.
REFERENCES
1. Abrahamsen, M. S., T. J. Templeton, S. Enomoto, J. E. Abrahante, G. Zhu, C. A. Lancto, M. Deng, C. Liu, G. Widmer, S. Tzipori, G. A. Buck, P. Xu, A. T. Bankier, P. H. Dear, B. A. Konfortov, H. F. Spriggs, L. Iyer, V. Anantharaman, L. Aravind, and V. Kapur. 2004. Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 304:441-445.
2. Akiyoshi, D. E., X. Feng, M. A. Buckholt, G. Widmer, and S. Tzipori. 2002. Genetic analysis of a Cryptosporidium parvum human genotype 1 isolate passaged through different host species. Infect. Immun. 70:5670-5675.
3. Buret, A. G., A. C. Chin, and K. G. Scott. 2003. Infection of human and bovine epithelial cells with Cryptosporidium andersoni induces apoptosis and disrupts tight junctional ZO-1: effects of epidermal growth factor. Int. J. Parasitol. 33:1363-1371.
4. Chen, M. L., C. Pothoulakis, and J. T. LaMont. 2002. Protein kinase C signaling regulates ZO-1 translocation and increased paracellular flux of T84 colonocytes exposed to Clostridium difficile toxin A. J. Biol. Chem. 277:4247-4254.
5. Chen, X. M., B. Q. Huang, P. L. Splinter, H. Cao, G. Zhu, M. A. McNiven, and N. F. LaRusso. 2003. Cryptosporidium parvum invasion of biliary epithelia requires host cell tyrosine phosphorylation of cortactin via c-Src. Gastroenterology 125:216-228.
6. Chen, X. M., B. Q. Huang, P. L. Splinter, J. D. Orth, D. D. Billadeau, M. A. McNiven, and N. F. LaRusso. 2004. Cdc42 and the actin-related protein/neural Wiskott-Aldrich syndrome protein network mediate cellular invasion by Cryptosporidium parvum. Infect. Immun. 72:3011-3021.
7. Chen, X. M., and N. F. LaRusso. 2000. Mechanisms of attachment and internalization of Cryptosporidium parvum to biliary and intestinal epithelial cells. Gastroenterology 118:368-379.
8. Chen, X. M., S. A. Levine, P. L. Splinter, P. S. Tietz, A. L. Ganong, C. Jobin, G. J. Gores, C. V. Paya, and N. F. LaRusso. 2001. Cryptosporidium parvum activates nuclear factor kappaB in biliary epithelia preventing epithelial cell apoptosis. Gastroenterology 120:1774-1783.
9. Chen, X. M., S. P. O'Hara, B. Q. Huang, J. B. Nelson, J. J. Lin, G. Zhu, H. D. Ward, and N. F. LaRusso. 2004. Apical organelle discharge by Cryptosporidium parvum is temperature, cytoskeleton, and intracellular calcium dependent and required for host cell invasion. Infect. Immun. 72:6806-6816.
10. Couldwell, W. T., D. R. Hinton, S. He, T. C. Chen, I. Sebat, M. H. Weiss, and R. E. Law. 1994. Protein kinase C inhibitors induce apoptosis in human malignant glioma cell lines. FEBS Lett. 345:43-46.
11. Dempsey, E. C., A. C. Newton, D. Mochly-Rosen, A. P. Fields, M. E. Reyland, P. A. Insel, and R. O. Messing. 2000. Protein kinase C isozymes and the regulation of diverse cell responses. Am. J. Physiol. Lung Cell Mol. Physiol. 279:L429-L438.
12. Elliott, D. A., and D. P. Clark. 2000. Cryptosporidium parvum induces host cell actin accumulation at the host-parasite interface. Infect. Immun. 68:2315-2322.
13. Elliott, D. A., D. J. Coleman, M. A. Lane, R. C. May, L. M. Machesky, and D. P. Clark. 2001. Cryptosporidium parvum infection requires host cell actin polymerization. Infect. Immun. 69:5940-5942.
14. Forney, J. R., D. B. DeWald, S. Yang, C. A. Speer, and M. C. Healey. 1999. A role for host phosphoinositide 3-kinase and cytoskeletal remodeling during Cryptosporidium parvum infection. Infect. Immun. 67:844-852.
15. Forney, J. R., D. K. Vaughan, S. Yang, and M. C. Healey. 1998. Actin-dependent motility in Cryptosporidium parvum sporozoites. J. Parasitol. 84:908-913.
16. Gut, J., and R. G. Nelson. 1999. Cryptosporidium parvum: synchronized encystation in vitro and evaluation of sporozoite infectivity with a new lectin-based assay. J. Eukaryot. Microbiol. 46:56S-57S.
17. Hashim, A., M. Clyne, G. Mulcahy, D. Akiyoshi, R. Chalmers, and B. Bourke. 2004. Host cell tropism underlies species restriction of human and bovine Cryptosporidium parvum genotypes. Infect. Immun. 72:6125-6131.
18. Joe, A., D. H. Hamer, M. A. Kelley, M. E. Pereira, G. T. Keusch, S. Tzipori, and H. D. Ward. 1994. Role of a Gal/GalNAc-specific sporozoite surface lectin in Cryptosporidium parvum-host cell interaction. J. Eukaryot. Microbiol. 41:44S.
19. Joe, A., R. Verdon, S. Tzipori, G. T. Keusch, and H. D. Ward. 1998. Attachment of Cryptosporidium parvum sporozoites to human intestinal epithelial cells. Infect. Immun. 66:3429-3432.
20. Koivunen, J., V. Aaltonen, S. Koskela, P. Lehenkari, M. Laato, and J. Peltonen. 2004. Protein kinase C alpha/beta inhibitor Go6976 promotes formation of cell junctions and inhibits invasion of urinary bladder carcinoma cells. Cancer Res. 64:5693-5701.
21. Morgan-Ryan, U. M., A. Fall, L. A. Ward, N. Hijjawi, I. Sulaiman, R. Fayer, R. C. Thompson, M. Olson, A. Lal, and L. Xiao. 2002. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporidiidae) from Homo sapiens. J. Eukaryot. Microbiol. 49:433-440.
22. Rossignol, J. F., A. Ayoub, and M. S. Ayers. 2001. Treatment of diarrhea caused by Cryptosporidium parvum: a prospective randomized, double-blind, placebo-controlled study of Nitazoxanide. J. Infect. Dis. 184:103-106.
23. Rosson, D., T. G. O'Brien, J. A. Kampherstein, Z. Szallasi, K. Bogi, P. M. Blumberg, and J. M. Mullin. 1997. Protein kinase C-alpha activity modulates transepithelial permeability and cell junctions in the LLC-PK1 epithelial cell line. J. Biol. Chem. 272:14950-14953.
24. Spano, F., L. Putignani, J. McLauchlin, D. P. Casemore, and A. Crisanti. 1997. PCR-RFLP analysis of the Cryptosporidium oocyst wall protein (COWP) gene discriminates between C. wrairi and C. parvum, and between C. parvum isolates of human and animal origin. FEMS Microbiol. Lett. 150:209-217.
25. Sukumaran, S. K., and N. V. Prasadarao. 2002. Regulation of protein kinase C in Escherichia coli K1 invasion of human brain microvascular endothelial cells. J. Biol. Chem. 277:12253-12262.
26. Thea, D. M., M. E. Pereira, D. Kotler, C. R. Sterling, and G. T. Keusch. 1992. Identification and partial purification of a lectin on the surface of the sporozoite of Cryptosporidium parvum. J. Parasitol. 78:886-893.
27. Tzipori, S., and H. Ward. 2002. Cryptosporidiosis: biology, pathogenesis and disease. Microbes Infect. 4:1047-1058.
28. Upton, S. 1997. In vitro cultivation, p. 181-207. In R. Fayer (ed.), Cryptosporidium and cryptosporidiosis. CRC Press, Boca Raton, Fla.
29. Widmer, G., E. A. Corey, B. Stein, J. K. Griffiths, and S. Tzipori. 2000. Host cell apoptosis impairs Cryptosporidium parvum development in vitro. J. Parasitol. 86:922-928.
30. Wiest, P. M., J. H. Johnson, and T. P. Flanigan. 1993. Microtubule inhibitors block Cryptosporidium parvum infection of a human enterocyte cell line. Infect. Immun. 61:4888-4890.
31. Xiao, L., U. M. Morgan, J. Limor, A. Escalante, M. Arrowood, W. Shulaw, R. C. Thompson, R. Fayer, and A. A. Lal. 1999. Genetic diversity within Cryptosporidium parvum and related Cryptosporidium species. Appl. Environ. Microbiol. 65:3386-3391.
32. Xu, P., G. Widmer, Y. Wang, L. S. Ozaki, J. M. Alves, M. G. Serrano, D. Puiu, P. Manque, D. Akiyoshi, A. J. Mackey, W. R. Pearson, P. H. Dear, A. T. Bankier, D. L. Peterson, M. S. Abrahamsen, V. Kapur, S. Tzipori, and G. A. Buck. 2004. The genome of Cryptosporidium hominis. Nature 431:1107-1112.(Amna Hashim, Grace Mulcah)