ABSTRACT
Infection with Giardia duodenalis is one of the most common causes of diarrheal disease in the world. While numerous studies have identified important contributions of adaptive immune responses to parasite control, much less work has examined innate immunity and its connections to the adaptive response during this infection. We explored the role of complement in immunity to Giardia using mice deficient in mannose-binding lectin (Mbl2) or complement factor 3a receptor (C3aR). Both strains exhibited delayed clearance of parasites and a reduced ability to recruit mast cells in the intestinal submucosa. C3aR-deficient mice had normal production of antiparasite IgA, but ex vivo T cell recall responses were impaired. These data suggest that complement is a key factor in the innate recognition of Giardia and that recruitment of mast cells and activation of T cell immunity through C3a are important for parasite control.
INTRODUCTION
Giardia duodenalis is one of the most common protozoan infections of humans as well as other mammals throughout the world and is a leading cause of diarrheal disease in these species (1–3). Symptomatic infections occur in ∼20 to 80% of humans with positive stool samples and are characterized by nausea, vomiting, epigastric pains, and diarrhea (1, 2, 4, 5). These symptoms are associated with nutrient malabsorption and can lead to weight loss and malnutrition in children, exposing this vulnerable group to failure to thrive and developmental problems (6, 7). Disease resolves spontaneously in >85% of cases. In certain cases, in spite of a healthy and fully developed immune system, the acute phase of the disease develops into chronic disease. In these cases, symptoms of the disease will reappear for short and recurrent periods (3, 4, 8). The mechanisms explaining interactions between the host and the parasite leading to parasite clearance and disease pathogenesis are poorly understood.
Studies of immune responses against Giardia have demonstrated important roles for both innate and adaptive immunity (7, 9). Antibody production following infection is robust, and IgA is very effective at eliminating parasites (9). Antibody-independent roles of T cells can also eliminate infections (10). Early studies indicated that Giardia trophozoites are susceptible to killing by factors in nonimmune human serum, milk, and intestinal fluid (11–13). Recently, killing by normal serum was demonstrated to involve the lectin pathway (14), consistent with the expression of N-acetylglucosamine (GlcNAc) on the surface of trophozoites (15, 16). Studies of cells of the innate immune system indicate that macrophages, dendritic cells, mast cells, and intestinal epithelial cells are all involved. In vitro macrophages have been shown to be capable of ingesting Giardia, and we recently showed that macrophages accumulate in the lamina propria following infection (17, 18). Bone marrow-derived dendritic cells mature in response to Giardia extracts, but cytokine production in response to Toll-like receptor (TLR) agonists is modulated toward interleukin-10 (IL-10) and away from IL-12 (19). Mast cells are also recruited following infection and are required for the efficient control of infection (20, 21). Finally, intestinal epithelial cells produce several cytokines after exposure to Giardia in vitro and may produce nitric oxide to help control infections (22, 23). However, how this parasite is recognized by the innate immune system and the importance of these innate responses are less clear.
We recently reported a microarray-based transcriptomic analysis of intestinal responses to Giardia (24). Among the genes significantly induced following infection was mannose-binding lectin (MBL) (Mbl2). Because MBL was also shown to contribute to parasite lysis in vitro (14), we decided to investigate the role of complement activation by MBL in immune responses to Giardia using the adult mouse model of Giardia duodenalis infection (25). We show that recombinant MBL binds to Giardia trophozoites, infections resolve more slowly in the absence of MBL, and recruitment of mast cells to the intestinal mucosa is delayed. In addition, we show that mice deficient in the receptor for complement factor 3a (C3aR) have a phenotype identical to that of MBL-deficient mice and that, while IgA production is normal in these mice, mast cell and T cell responses are both diminished.
MATERIALS AND METHODS
Mice, parasites, and infections.C57BL/6J and BALB/c mice were obtained from Jackson Laboratories (Bar Harbor, ME). In addition, we also purchased breeding pairs of B6.129S4-Mbl1tm1Kata Mbl2tm1Kata/J and C.129S4-C3ar1tm1Cge mice from Jackson Laboratories (Bar Harbor, ME) for breeding at the Georgetown University Division of Comparative Medicine. All animals were housed under specific-pathogen-free conditions. Mice were provided neomycin sulfate (NEO) (1.4 mg/ml; Phoenix Pharmaceuticals, St. Louis, MO), metronidazole (MTZ) (1 mg/ml; Sidmak Labs, Hanover, NJ), ampicillin (AMP) (1 mg/ml; Sigma-Aldrich, St. Louis, MO), and vancomycin (VYN) (1 mg/ml; Abbott Labs, Worcester, MA) in drinking water in order to facilitate infection, as described previously (26). A combination of NEO-MTZ-VYN-AMP was provided for 5 days prior to infection, followed by a combination of NEO-VYN-AMP for the remaining course of infection, since MTZ kills Giardia. All experiments were performed with the approval of the Georgetown University Animal Care and Use Committee. G. duodenalis strain GS(M)H7 was cultured to confluence and used for infections, as previously described (27). Briefly, 5- to 6-week-old female mice (n = 4/group) were infected by gavage with 1 × 106 parasites each in 0.1 ml phosphate-buffered saline (PBS). Parasites were counted by collecting intestinal segments from the duodenum just below the ligament of Treitz. This is where maximal parasite growth occurs due to the presence of bile from the common bile duct. Collected fragments were minced in 4 ml of cold PBS and then kept on ice for 30 min for parasite release before counting on a hemocytometer.
Mast cell responses.Mast cells in the small intestine were identified by using chloroacetate esterase staining as described previously (21). The distal jejunum was fixed overnight in 4% formaldehyde in PBS, embedded in paraffin, and sectioned (5 μm) on glass slides. Sections were then stained and viewed with a Zeiss Axiophot microscope. Images were collected with a CoolSnap fx camera (Roper Scientific, Trenton, NJ) by using Volocity software (Improvision, Cambridge, MA). Imaging and scoring of mast cell numbers were performed by observers blind to the origin of the slides.
MBL-binding assay.Trophozoites were allowed to attach to glass slides at 37°C for 15 min. They were fixed in cold acetone for 5 min and then incubated with MBL-deficient serum supplemented with recombinant MBL (rMBL) (5 μg/ml), together with purified GlcNAc (200 mM) or methyl-galactopyranoside (200 mM) for 30 min. After blocking with Superblock blocking buffer (Pierce, Rockford, IL, USA) for 2 h at room temperature, the sections were incubated with a rat monoclonal antibody (MAb) specific for mouse Mbl2 (MAb 16A8) (diluted with Superblock blocking buffer). After overnight incubation at 4°C, sections were washed with PBS and then incubated with secondary goat anti-rat antibodies labeled with fluorescein isothiocyanate (FITC) (Southern Biotech Inc., Birmingham, AL). Two to five percent goat serum in Superblock was used for blocking of nonspecific binding sites and for dilution of primary and secondary antibodies. Sections incubated with secondary antibodies alone were used as controls. Slides were mounted with Vectashield plus 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, CA) and viewed with a Zeiss Axiophot microscope. Images were collected with a CoolSnap fx camera (Roper Scientific, Trenton, NJ) by using Volocity software (Improvision, Cambridge, MA). Images were processed with Adobe Photoshop (Adobe Systems, San Jose, CA).
Measurement of antiparasite IgA response.Giardia cells were harvested by centrifugation at 800 × g for 5 min at 4°C. The cell pellet was resuspended in cold buffer (0.01 M Tris-HCl [pH 8.6], 0.25 M sucrose, 1 mM MgSO4, 0.001% phenylmethylsulfonyl fluoride [PMSF] enzyme inhibitor, and 0.05% trypsin inhibitor) and homogenized for 5 min with a probe-type sonicator on ice. Lysates were then centrifuged for 10 min at 800 × g at 4°C and again for 20 min at 10,000 × g at 4°C to remove organelles and cytoskeletal fragments. Supernatants were centrifuged again for 1 h at 100,000 × g at 4°C, resulting in a pellet containing mainly Giardia cell plasma membrane. The pellet was dissolved in buffer, and the protein concentration was measured with a Bio-Rad protein assay.
Ninety-six-well plates (Nunc; Thermo Scientific) were coated overnight at 4°C with 100 μl of 2 μg/ml Giardia membrane protein in PBS. After blocking of the plate with 1% bovine serum albumin (BSA), 100 μl gut wash was added, and the plate was sealed for incubation for 2 h at room temperature. The addition of 100 μl peroxidase-conjugated goat anti-mouse immunoglobulin A was followed by the addition of the 2,2′-azinobis(3-ethylbenzthiazolinesulfonic acid) (ABTS) substrate. IgA levels were determined by using a microplate reader at 405 nm.
Spleen cell preparation and in vitro stimulation.Spleen cells were suspended in PBS after removal of red cells by using red blood cell lysis buffer (Sigma-Aldrich, St. Louis, MO). Spleen cells were counted by using a microscopy counting chamber (hemocytometer), and total cell numbers were calculated for each spleen organ. Spleen cells were adjusted to 5 × 107 cells/ml in complete culture medium (RPMI 1640 with 10% fetal calf serum, 1 mM glutamine, 50 μm 2-mercaptoethanol, and 100 μg/ml antibiotics) and plated in 24-well cell culture clusters at 5 × 106 cells/well in 1 ml. The cells were incubated with Giardia extract for 48 h at 37°C, and supernatants were harvested for analysis of cytokine activity as described previously (28). Mouse gamma interferon (IFN-γ), tumor necrosis factor (TNF) (BD OpEIA), IL-17 (R&D Systems), and IL-4 (Southern Biotech) levels in the supernatant were measured by an enzyme-linked immunosorbent assay (ELISA) using commercial kits.
Statistics.Comparisons of parasite numbers between mouse groups were done by using a one-tailed Mann-Whitney test, and comparisons of mast cell recruitment, cytokine production, and IgA levels were performed with t tests using GraphPad Prism software (GraphPad Software, San Diego, CA). A statistical probability (P) value of <0.05 was considered significant.
RESULTS
Mbl2 is abundantly expressed in epithelial cells of the small intestine, to which Giardia attaches and with which it interacts (24, 29). MBL has been shown to bind to surface carbohydrates on pathogens, including Giardia, leading to complement activation (14). In many infections, this recognition has a significant influence on disease outcome, e.g., in cryptosporidiosis, leishmaniasis, malaria, and HIV infection (30–33). Interestingly, several studies have established a link between human genotypes associated with low serum levels of MBL and susceptibility to severe infectious diseases such as malaria, HIV, and cryptococcosis (34–36).
Giardia modifies cytosolic and surface proteins by the addition of N-acetylglucosamine (GlcNAc) (15, 16). Since GlcNAc is a ligand for MBL, we decided to see if MBL would bind directly to Giardia trophozoites. Parasites were attached to glass sides, incubated with recombinant MBL, and then stained with anti-MBL and a FITC-conjugated antibody (Fig. 1). While parasites incubated in serum from Mbl-deficient mice showed no staining, the addition of MBL produced an evenly distributed pattern on the parasite. Binding could be inhibited by the addition of soluble GlcNAc but not by the addition of galactopyranoside, consistent with the specificity of MBL.
Binding of MBL to Giardia trophozoites. Parasites were allowed to attach to slides, fixed, and incubated with MBL-deficient serum alone (A) or supplemented with 5 μg/ml rMBL (B to D), as described in Materials and Methods. A total of 200 mM GlcNAc (C) or α-methyl-galactopyranoside (D) was included as a competitor. After washing, samples were treated with rat anti-mouse MBL followed by a FITC-conjugated goat anti-rat secondary antibody. Original magnification, ×400.
MBL binding to Giardia was recently shown to initiate complement-mediated lysis of trophozoites in vitro (14). We therefore sought to determine if MBL binding to the parasite in vivo had a role in immune recognition of Giardia. We infected wild-type and Mbl-deficient mice with parasites and examined the control of infection. As shown in Fig. 2A, mice deficient in Mbl had a delay in eliminating Giardia from the small intestine. Whereas wild-type mice had eliminated almost all detectable Giardia parasites by 10 days postinfection, elimination of parasites in MBL-deficient mice required 14 days. A similar result was seen after infection of mice without antibiotic treatments, although the numbers of parasites observed in all groups were much lower than those shown in Fig. 2. At 7 days postinfection, we observed almost four times as many parasites in Mbl-deficient mice as in wild-type controls (98,125 ± 33,235 versus 26,875 ± 11,290 parasites; P = 0.09).
Infections in mannose-binding-lectin-deficient mice. Wild-type (WT) C57BL/6 mice and mice deficient in both MBL1 and MBL2 (MBL knockout [MBLKO]) were infected with G. duodenalis as described in Materials and Methods. (A) Parasite numbers in the small intestine were determined at 5, 10, and 14 days postinfection (n = 4/group). Each point represents an individual mouse, and bars represent the means of the distribution for each group. The dashed line represents the limit of detection for this assay. (B) Mast cell recruitment was determined in 10 VCUs of chloroacetate esterase-stained sections of jejunum. The solid bar represents the mean of the distribution for each group. n.d., none detected; ***, P < 0.0001; **, P < 0.05. Data are representative of results from two independent experiments.
Mbl is known to activate the complement pathway, which can result in target cell lysis, opsonization of targets, and recruitment of inflammatory cells through the release of C3a and C5a. Since mast cell responses are known to have a role in controlling Giardia infection, we examined mast cell recruitment in MBL-deficient mice. Mast cell recruitment to the small intestine was significantly delayed in MBL-deficient mice (Fig. 2B). By day 5 postinfection, only a few mast cells were seen in any mouse. By day 10 postinfection, wild-type mice showed an average of 60 mast cells/10 villus-crypt units (VCU), compared to an average of 13 mast cells/VCU (P < 0.001) in MBL-deficient mice. There was a similar difference at day 14: wild-type mice showed on average 77 mast cells/VCU, compared to 34 mast cells/VCU (P < 0.004) in MBL-deficient mice.
To determine if recruitment of mast cells was occurring directly through C3a receptors (C3aRs), we next infected C3aR-deficient mice with Giardia. Similar to what was observed for MBL-deficient mice, the clearance of parasites from the small intestine and the recruitment of mast cells were delayed in mice lacking C3aR (Fig. 3A and B). The similarity in parasite clearance between MBL- and C3aR-deficient mice suggests that the major functions of complement during giardiasis are mediated through C3aR and that direct lysis of trophozoites by complement may not be important for parasite clearance in vivo. These data are also consistent with the previously reported essential role for mast cells in the elimination of Giardia in this model (21).
Infections in C3aR-deficient mice. Wild-type BALB/c and C3aR knockout mice were infected on day 0, and parasite loads (A) and mast cell accumulation (B) were determined at 5, 7, 14, and 18 days postinfection. Each point represents an individual mouse, and bars represent the means of the distribution for each group. The dashed line represents the limit of detection for this assay. ***, P < 0.0001; **, P < 0.05. Data are representative of results from three independent experiments.
We next asked whether other aspects of protective immunity were affected in C3aR-deficient mice. CD4+ T cell responses are required for parasite control (10, 37), and these cells are a major source of cytokines following infections. We therefore measured ex vivo cytokine production in response to parasite extracts. As shown in Fig. 4, C3aR-deficient mice had reduced IFN-γ, IL-4, IL-17, and TNF responses compared with those of wild-type BALB/c control mice at different time points following infection. In contrast, the production of antiparasite IgA appeared unaffected in mice lacking C3aR (Fig. 5). This is consistent with previously reported data for this model showing that IgA can help eliminate infection, although more time is required than for other CD4+ T cell-dependent mechanisms (27).
Cytokine response profiles of spleen cells from C3aR knockout and wild-type BALB/c mice. Spleen cells were cultured with 100 μg/ml G. duodenalis extract for 48 h, and supernatants were analyzed in duplicate for cytokines by an ELISA. The bars represent the means and standard errors based on results for four individual mice/group. *, P < 0.05; **, P < 0.01; ***, P < 0.008; ****, P < 0.0001; ns, not significant. Results are representative of data from three independent experiments.
IgA responses to parasite infection in wild-type and C3aR knockout BALB/c mice. Contents of the jejunum were assayed for antiparasite IgA as described in Materials and Methods. The points represent the means and standard errors based on results for four individual mice/group. *, P < 0.05; ****, P < 0.0001 (compared to uninfected mice). Results are representative of data from three independent experiments.
DISCUSSION
Our analysis of Giardia infections in mice lacking MBL or C3aR demonstrate that complement plays a significant role in initiating immune responses against this protozoan parasite. Mice lacking either MBL or C3aR take longer to eliminate intestinal trophozoites than do wild-type controls. These mice also have delayed recruitment of mast cells to the small intestine, consistent with previous reports that mast cells are a key component of the immune response against Giardia (20, 21, 38). Mice lacking C3aR also exhibited significantly reduced T cell responses against parasite antigens following infection, but antiparasite IgA levels in the small intestine were unaffected.
Several laboratories have investigated mechanisms by which the innate immune system can recognize and respond to Giardia infection. Our laboratory previously showed that dendritic cells responded to parasite extracts by upregulating the expression of surface molecules, including CD40, CD80, and CD86, but that very little cytokine production was induced. Indeed, parasite extracts actually blocked IL-12 production in response to several TLR agonists, although IL-10 production was enhanced (19). A recent report showed that the endoplasmic reticulum (ER) chaperone protein BiP from Giardia could signal through TLRs, but like our own studies with dendritic cells, extremely high levels of protein were required to detect a response (39).
Recognition of the parasite by MBL represents a mechanism by which the innate immune system can detect the parasite and may in fact be a significant pathway for activation of the innate response. Activation of complement can trigger direct pathogen lysis, opsonization of pathogens for uptake by macrophages, signaling to other immune cells via receptors for C3a and C5a, and enhancement of antibody responses through complement receptor 1 (CR1) signaling on B cells. While MBL and complement can lyse parasites in vitro, our data indicate that signaling through C3aR may be more important for inducing other immune responses in vivo to actually eliminate the infection.
The recruitment of mast cells by MBL and C3aR is consistent with data from previous studies demonstrating important roles for mast cells in parasite elimination (4, 7). Mice with mutations in c-kit lack functional mast cells and have profound defects in parasite elimination (20, 21). Treatment with a monoclonal antibody against c-kit also blocked mast cell responses and parasite elimination (21). Giardia infection was also shown to cause significant increases in serum levels of histamine and mast cell protease, indicating that mast cells are not only recruited but also activated to degranulate (21). Mast cells were further implicated in generating smooth muscle hypercontractility following Giardia infection through pharmacological inhibition of degranulation ex vivo (40). Finally, intestinal hypermotility was also shown to be an important mechanism for parasite elimination, although a direct role for mast cells in this motility response in vivo has not been shown (41, 42).
Several recent studies have focused on the role of IL-17A in the control of Giardia infections. The IL-17A mRNA level is elevated in the guts of cattle and mice infected with Giardia, and mice lacking IL-17A exhibit a severe defect in controlling infections with either Giardia muris or G. duodenalis (43–45). Moreover, IL-17A memory T cell responses correlated with more rapid elimination of infections in a cohort of travelers with giardiasis (46). Our data show that IL-17A responses are significantly reduced in the absence of signaling via C3aR. This is consistent with data from previous studies showing that C3aR signaling on dendritic cells can enhance IL-23 production and induce elevated IL-17 responses in a model of allergic airway hypersensitivity (47). Similarly, in a model of colitis-associated cancer induced by dextran sodium sulfate (DSS) along with azoxymethane, C3-deficient mice had reduced IL-17A production in myeloid cell populations and were therefore less susceptible to tumor development (48). In contrast, MBL-deficient mice had enhanced intestinal IL-17A and IL-23 mRNA levels in a DSS colitis model (49). The differential effects of complement in these two models may reflect differences in the time points analyzed (day 60 versus day 14), differences in the specific mouse mutants analyzed, or other differences between mouse colonies.
Variation in human levels of MBL is well documented (50). Polymorphisms in the promoter and in the coding region of Mbl2 lead to different levels of gene expression and to defects in the ability of proteins to form multimers, a prerequisite for the activation of complement. Furthermore, MBL levels have been correlated with differential infection outcomes in human cryptosporidiosis and malaria (34, 36, 51, 52). Mice with a deficiency in MBL were also shown to be more susceptible to West Nile virus than wild-type controls, although mice lacking C3 were even more susceptible to the virus, indicating that multiple pathways of complement activation could be involved (53–55). Trypanosoma cruzi, the causative agent of Chagas' disease, was shown to activate complement via the lectin pathway, leading to parasite lysis and reduced invasion of host cells in vitro (56). Furthermore, mice lacking MBL had higher parasite burdens in tissues and enhanced pathology following T. cruzi infection (57). Data from humans are less clear, however. Single-nucleotide polymorphisms (SNPs) in the gene for MBL-associated serine protease 2 (MASP-2) that are associated with low MASP-2 expression levels were found more frequently in Chagas' disease patients with cardiomyopathy than in patients in the indeterminate phase (58). Similarly, MBL genotypes with low expression were also correlated with increased pathology (59). However, a recent study found that low-MBL genotypes were associated with reduced pathology in Chagas' disease patients (60). Thus, variation in MBL levels can impact both early and late events during both acute and chronic Chagas' disease.
It is possible that humans with low levels of MBL activity might therefore activate complement less effectively, stimulate weaker T cell cytokine responses, and recruit fewer mast cells to the intestinal tract during giardiasis. Thus, MBL deficiency might lead to prolonged infections and/or reduced symptoms such as cramps and diarrhea. MBL recruitment of mast cells might also provide a mechanism for the immune response against this infection in patients with reduced numbers of CD4+ T cells, such as in AIDS patients. This may explain the occurrence of diarrhea in HIV-infected individuals, even at low CD4 counts (52). Given the natural variation in MBL levels in humans and the potential for mast cells to cause intestinal pathology, it will be important to determine the contribution of MBL and mast cell responses to the variation in the presentation of giardiasis in humans.
Several studies have examined the relationship between MBL and inflammatory bowel disease. It was first reported that variants of MBL that decrease the production of functional MBL were associated with ulcerative colitis (UC) but not with Crohn's disease (CD) (61). Additional studies found an association between low levels of MBL and severe forms of CD but not CD overall or UC (62). MBL deficiency was also associated with pediatric CD (63, 64), although the association with UC in these populations was inconsistent. Understanding how the immune system maintains homeostasis within the intestinal tract, as well as the rest of the body, is a major point of investigation for immunologists. Differences in the ways in which the system first recognizes and responds to microbes shape the eventual response to that organism and, along with the development of regulatory immunity, contribute to maintaining intestinal health. The outcomes of Giardia infection cover a broad spectrum ranging from parasite elimination to chronic infection and from asymptomatic to severe cramps, diarrhea, and nutrient malabsorption, and understanding the mechanisms involved in the early recognition of this parasite by the immune system provides another model for intestinal immunity. The data in this study demonstrate that recognition of the parasite by the lectin pathway of complement is a key contributor to early detection by the immune system and shapes downstream events related to both parasite control and immunopathology.
ACKNOWLEDGMENTS
We thank Nancy Noben-Trouth, Maria Donoghue, Joel Kamda, Diane Taylor, Francoise Selliers-Moisiewitsch, and Audrey Thevenon for helpful advice and discussion. We also acknowledge the Lombardi Comprehensive Cancer Center Microarray Core Facility, Bioinformatics and Biostatistics Resource, Histology Facility, and Department of Comparative Medicine for assistance with experiments.
This work was supported by Public Health Service grants AI-081033, AI-094492, and AI-109591 from the National Institute of Allergy and Infectious Diseases and grant 5D43 TW001264 from the Fogarty International Center. The Georgetown Lombardi Shared Resources are partially supported by grants CA-051008 and NCATS 8 UL1 TR000101 from the National Cancer Institute, and the Georgetown University Barrier Animal Facility was partially supported by grant RR-025828 from the National Center for Research Resources.
FOOTNOTES
- Received 25 January 2016.
- Accepted 26 January 2016.
- Accepted manuscript posted online 1 February 2016.
- Copyright © 2016, American Society for Microbiology. All Rights Reserved.
