Giardia duodenalis Infection Reduces Granulocyte Infiltration in an In Vivo Model of Bacterial Toxin-Induced Colitis and Attenuates Inflammation in Human Intestinal Tissue

Giardia duodenalis (syn. G. intestinalis, G. lamblia) is a predominant cause of waterborne diarrheal disease that may lead to post-infectious functional gastrointestinal disorders. Although Giardia-infected individuals could carry as much as 106 trophozoites per centimetre of gut, their intestinal mucosa is devoid of overt signs of inflammation. Recent studies have shown that in endemic countries where bacterial infectious diseases are common, Giardia infections can protect against the development of diarrheal disease and fever. Conversely, separate observations have indicated Giardia infections may enhance the severity of diarrheal disease from a co-infecting pathogen. Polymorphonuclear leukocytes or neutrophils (PMNs) are granulocytic, innate immune cells characteristic of acute intestinal inflammatory responses against bacterial pathogens that contribute to the development of diarrheal disease following recruitment into intestinal tissues. Giardia cathepsin B cysteine proteases have been shown to attenuate PMN chemotaxis towards IL-8/CXCL8, suggesting Giardia targets PMN accumulation. However, the ability of Giardia infections to attenuate PMN accumulation in vivo and how in turn this effect may alter the host inflammatory response in the intestine has yet to be demonstrated. Herein, we report that Giardia infection attenuates granulocyte tissue infiltration induced by intra-rectal instillation of Clostridium difficile toxin A and B in an isolate-dependent manner. This attenuation of granulocyte infiltration into colonic tissues paralled decreased expression of several cytokines associated with the recruitment of PMNs. Giardia trophozoite isolates that attenuated granulocyte infiltration in vivo also decreased protein expression of cytokines released from inflamed mucosal biopsy tissues collected from patients with active Crohn’s disease, including several cytokines associated with PMN recruitment. These results demonstrate for the first time that certain Giardia infections may attenuate PMN accumulation by decreasing the expression of the mediators responsible for their recruitment.


Introduction
Giardia duodenalis (syn. G. intestinalis, G. lamblia) is a noninvasive, small intestinal protozoan parasite infecting a variety of animal species, including humans. Giardia duodenalis is currently subdivided into eight distinct genetic assemblages, with only assemblages A and B being infective to humans [1,2]. Some reports suggest that these assemblages may be unique Giardia species, but much controversy remains on this topic [3,4]. This parasite induces a response within its host that may cause malabsorptive diarrheal disease ( [5]). Recent evidence also indicates that giardiasis can lead to the development of postinfectious disorders, in the intestine as well as extra-intestinally, via mechanisms that remain obscure ( [6,7]). During the acute stage of the infection, parasite loads can exceed 10 6 trophozoites/cm of gut but the intestinal mucosa of most Giardia-infected patients is devoid of overt signs of inflammation [8]. These observations are counter-intuitive, considering that Giardia infections increase small intestinal permeability via several mechanisms and, therefore, may indeed facilitate the translocation of luminal antigens into underlying host tissues [9][10][11]. However, microscopic duodenal inflammation has been observed in some patients with chronic assemblage B Giardia infections following metronidazole treatment [12]. Giardia assemblage B infections in vivo have been shown to induce small intestinal inflammation [13]. Therefore, it remains to be determined whether Giardia infections modulate pro-inflammatory responses within the intestinal mucosa and if these events are assemblage or isolate specific.
Transmission of Giardia infections occurs via ingestion of infectious cysts in contaminated food or water, or directly via the fecal-oral route (reviewed in [14]). This route of infection is adopted by many gastrointestinal pathogens, and as a result, Giardia co-infections may occur. Indeed, Giardia infections have been reported to occur concurrently with Ascaris sp. [15], Cryptosporidium sp. [16], Helicobacter pylori [16,17], Vibrio cholerae [18], Salmonella sp. [19] and rotavirus [18,20,21]. Many of these pathogens are known to promote inflammatory responses and simultaneously induce diarrheal disease within their hosts. A recent report suggested that Tanzanian children infected with Giardia have reduced incidence rates of diarrheal disease and fever, while also displaying lower serum C-reactive protein (CRP) levels [22]. Giardia infections have also been shown to attenuate symptoms of diarrheal disease during rotavirus infection [20]. The mechanisms remain unclear, it has been suggested that Giardia infections may create an advantageous environment for other coinfecting gastrointestinal pathogens [18]. Contradictorily, a separate study suggests that Giardia infections may instead enhance symptoms of diarrheal disease during rotavirus infection [16].
Polymorphonuclear leukocytes (PMNs) are heavily involved in inflammatory responses, and essential to host defence against many bacterial and fungal pathogens. Indeed, genetic mutations or drug therapy that result in defective PMN function make individuals highly susceptible to life-threatening infections [23][24][25]. These cells have been implicated in a variety of inflammatory disorders, in the gut and beyond [26,27]. PMN recruitment into intestinal tissues is a multistep process involving egression from bone marrow and circulation, migration through host tissues, and transepithelial migration [28][29][30]. Moreover, multiple mediators are known to promote tissue accumulation of PMNs. During acute inflammatory responses, granulocyte colony stimulating factor (G-CSF) promotes PMN bone marrow egression and increases the number of circulating PMNs [31,32]. Production of interleukin-17A (IL-17A) by tissue-resident cells increases circulating G-CSF levels and promotes subsequent PMN bone marrow egression [32][33][34]. PMN chemokines containing a consecutive glutamateleucine-arginine sequence (ELR+ chemokines) in their N-terminal region, such as CXCL1, CXCL2, and IL-8/CXCL8, are produced by a variety of cells within the intestinal mucosa. ELR+ chemokines are involved in multiple processes of PMN tissue recruitment, including bone marrow egression [31], exit from the blood stream [30,35,36], and migration through host tissues [37,38].
Previous research has suggested that Giardia excretory/ secretory products induce IL-8/CXCL8 expression in intestinal epithelial cells [39]. However, we and others have demonstrated that Giardia trophozoites attenuate IL-8/CXCL8 secretion from in vitro epithelial monolayers [40,41]. Our research further demonstrated that Giardia cathepsin B family proteases contributed to the degradation of IL-8/CXCL8 and attenuated PMN chemotaxis [41]. As PMN accumulation contributes to the development of diarrheal disease via several different mechanisms [28,[42][43][44], we hypothesized that during a Giardia infection, the attenuation of diarrheal disease or the creation of a favourable environment for other co-infecting gastrointestinal pathogens may stem, at least partially, from Giardia's ability to attenuate PMN recruitment. These observations support the notion that Giardia parasites may reduce host inflammatory responses. Such effects and their mechanisms have yet to be established in vivo. The present report demonstrates that Giardia infections attenuate granulocyte infiltration during acute experimental colitis induced by Clostridrium difficile toxins TcdA and TcdB, and markedly modulate the expression of several pro-inflammatory mediators, in an isolate-dependent manner. This study also shows that Giardia trophozoites decrease expression of a variety of inflammatory mediators when parasites are co-incubated with inflamed biopsy tissues from patients with Crohn's disease. Collectively, these results indicate for the first time that certain Giardia isolates are capable of attenuating intestinal inflammatory responses in inflammatory settings in vivo and ex vivo.

Ethics statement
All studies involving human colonic mucosal biopsy tissues were approved by the Conjoint Health Research Ethics Board (CHREB) at the University of Calgary and the Calgary Health Region. In accordance with CHREB guidelines, adult subjects used in this study provided informed, written consent and a parent or guardian of any child participant provided informed, written consent on their behalf. Animal experiments were approved by the Life and Environmental Sciences Animal Care Committee of the University of Calgary, conducted in compliance with that approval, and followed guidelines established by the Canadian Council of Animal Care. All isolates used in this study have been used for over 15 years in the laboratory. Therefore, their isolation did not require their approval from an ethical review board at their time of collection.

Human biopsy tissues
Adapted from previous protocols [41,45], intestinal biopsy tissues were collected from the descending colon of patients with active Crohn's disease (CD). Upon collection, samples were washed once in Dulbecco's PBS (Sigma-Aldrich) containing 0.016% 1,4-Diothioerythritol (Sigma-Aldrich) to remove loosely adherent mucous and bacteria followed by three washes with phosphate-buffered saline (PBS). Following this, biopsy tissues were placed in 96-well plates and incubated in 300 mL of OptiMEM (Life Technologies) at 37uC, 5% CO 2 , and 96% humidity. All studies were performed in duplicate, whereby multiple biopsy tissues were collected from each patient.

Parasites
Giardia NF trophozoites were originally obtained from a water sample during an outbreak of giardiasis in Newfoundland, Canada [46], while Giardia GS/M clone H7 trophozoites were obtained from ATCC (50581) as previously described [47]. Trophozoites were grown axenically in 15 mL polystyrene tubes (Becton-Dickinson Falcon) in Keister's modified TYI-S-33 medium [48,49] supplemented with piperacillin (Sigma-Aldrich) and used at peak density culture.

Giardia trophozoite isolation
Confluent tubes of Giardia NF or GS/M trophozoites were harvested by cold shock on ice for 30 minutes, pooled into 50 mL polypropylene tubes (Falcon), and centrifuged at 5006g for 10 minutes. The resulting pellets were re-suspended in a total of 10 mL of ice cold PBS (Sigma-Aldrich) and centrifuged for 10 minutes at 5006g. Pellets were then re-suspended in 3 mL of fresh PBS, trophozoites were enumerated with a hemocytometer, and adjusted to the required concentration. For ex vivo human biopsy experiments, Giardia trophozoites were adjusted to a concentration of 5.0610 6 trophozoites/well, while trophozoites were adjusted to a concentration of 1.0610 8 trophozoites/mL for in vivo experiments. For ex vivo experiments, cells were coincubated with Giardia trophozoites at 37uC, 5% CO 2 , and 96% humidity for 6 hours.

Giardia in vivo infection
Using a previously described Giardia infection model [50,51], male C57BL/6 mice aged 4 to 6 weeks (Charles River Laboratories, Sherbrooke, QC) were acclimatized for 1 week prior to infection with Giardia NF or GS/M trophozoites. Fortyeight hours prior to oral gavage, all mice were administered broadspectrum antibiotics to their drinking water ad libitum (1.4 mg/ mL neomycin (Alfa Aesar), 1.0 mg/mL ampicillin (Alfa Aesar) and 1.0 mg/mL vancomycin (Alfa Aesar)). This regimen was maintained for the entire duration of the study. After 48 hours, mice were administered 10 7 Giardia NF or GS/M trophozoites in 0.1 mL of TYI-S-33 medium or 0.1 mL TYI-S-33 medium by oral gavage. After 7 days, small intestinal parasite loads were quantified. Mice were euthanized and the first 3 cm of the small intestine distal to the ligament of Treitz were opened longitudinally, placed in 1.5 mL Eppendorf tubes containing 1 mL of PBS, and kept on ice for 15 minutes. After 15-minute incubation, tubes were vortexed, and trophozoite numbers were visually enumerated under light microscope, and total concentration was extrapolated using a hemocytometer.

Clostridium difficile TcdAB-induced colitis
Following a previously described model of C. difficile toxininduced colitis [52], mice infected with Giardia NF or GS/M trophozoites for 7 days were intra-rectally (i.r.) administered a 100 mg solution of C. difficile A/B toxin (TcdA/TcdB) for 3 hours. Briefly, a 5F infant feeding tube catheter containing side ports (Mallinckrodt Inc.) was lubricated with a water-soluble personal lubricant and inserted 2.5 cm into the colon. With pressure applied to the anal area to prevent leakage, 100 mL of a 1.0 mg/mL TcdA/TcdB solution diluted in PBS was slowly administered over a period of 30 seconds. Following this, the tube was slowly removed and pressure was maintained for another 30 seconds. Vehicle control-treated control groups were administered PBS.

Tissue collection
Mice were euthanized via cervical dislocation 3 hours post intracolonic instillation of TcdAB or PBS. The colon was removed and samples were collected for bead-based cytokine analysis, myeloperoxidase (MPO) assays, or fixed in a fresh 4% paraformaldehyde solution for immunofluorescence, as described below. Colonic tissue samples collected for histology and immunohistochemistry were fixed in 4% paraformaldehyde solution overnight at 4uC, and, subsequently, washed and transferred to PBS at 4uC.

Cytokine analysis
Supernatants from ex vivo human biopsy tissue experiments were collected and centrifuged at 10,000 g for 15 minutes at 4uC. Resulting supernatants were decanted and stored at 270uC. Colonic tissue samples and biopsy tissues were weighed and collected in 2 mL Fast-Prep tubes (MP Biomedicals) containing a mixture of 0.9-2.0 mm stainless steel beads (NextAdvance). Tissue samples were suspended in lysis buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% Tween-20 and a Complete Minitab protease inhibitor cocktail (Roche)] at a ratio of 50 mg tissue per 1 mL lysis buffer and homogenized using a Fast-Prep24 device (MP Biochemicals) at speed 6.0 for 40 seconds. The resulting homogenate solution was collected into pyrogen-free 1.5 mL Eppendorf tubes and centrifuged at 10,000 g for 15 minutes at 4uC. Supernatants and tissue homogenate cytokine levels were assessed using a Luminex XMap assay according to manufacturer's instructions (Luminex Corp.).

Tissue myeloperoxidase assay
Myeloperoxidase (MPO) activity was used as a marker of granulocyte infiltration [53]. Colonic tissue samples were weighed and collected in 2 mL Fast-Prep tubes (MP Biomedicals) containing a mixture of 0.9-2.0 mm stainless steel beads (NextAdvance). Tissue samples were suspended in 50 mM potassium phosphate buffer containing 5 mg/mL hexadecyltrimethylammonium bromide (Sigma-Aldrich) at a ratio of 50 mg tissue per 1 mL lysis buffer. Samples were then homogenized using a Fast-Prep24 device (MP Biochemicals) at speed 6.0 for 40 seconds. The resulting homogenate solution was collected into pyrogen-free 1.5 mL Eppendorf tubes and centrifuged at 10,0006g for 15 minutes at 4uC. Seven mL of the resulting supernatant was added to a standard 96-well plate along with 200 mL of the reaction mixture (comprised of 0.005 g Odianisidine (Sigma-Aldrich), 30 mL of distilled H 2 O, 3.33 mL of potassium phosphate buffer, and 17 mL of 1% H 2 O 2 ). Using a microplate scanner (SpectraMax M2 e , Molecular Devices, Sunnyvale, CA), three absorbance readings at 450 nm were recorded every 30 seconds. MPO activity was measured as units of activity per milligram of tissue, with 1 unit of MPO being defined as the amount required to degrade 1 mmol of H 2 O 2 per minute at room temperature.

Immunohistochemistry
Samples were dehydrated and embedded in paraffin wax, cut on a cryostat into 8 mm sections, and mounted on poly-D-lysine coated slides. Slides were deparaffinised and incubated in antigen   retrieval solution (Tris/EDTA, pH 9.0 solution for 30 minutes at 95 degrees C. Slides were then blocked in tissue blocking buffer (1% BSA, 0.1% Triton X-100 in PBS) for 1 hour (3620 minute washes) and incubated with an anti-MPO antibody (abcam ab9535) (1 in 200 dilution) overnight at 4uC. At room temperature, slides were washed for 1 hour (3620 minutes), incubated for 1 hour with secondary antibody (1 in 1000), and then washed for 1 hour (3620 minutes). After this, slides were mounted with Fluoroshield TM containing DAPI (Sigma-Aldrich) and visualized using a Leica DMR fluorescence microscope at 406 magnification.

Statistics
Parametric data are expressed as means 6 standard error of measurement, while non-parametric data is expressed as a dotplots with median and range. All statistical analyses were performed using GraphPad Prism 6 software, and normality of the data was assessed prior to analysis. Parametric comparisons were made using one-way ANOVA with Tukey's post hoc analysis. Nonparametric comparisons were made using a Mann Whitney test. Statistical significance was established at p,0.05 (*).

Giardia infections attenuate granulocyte infiltration during TcdAB-induced colitis in an isolate-dependent manner
Experiments were performed to determine whether in vivo assemblage A Giardia NF or assemblage B Giardia GS/M infections were capable of attenuating acute intestinal inflammatory responses to intracolonic administration of C. difficile toxin A (TcdA) and toxin B (TcdB). This model of intestinal inflammation was selected because it induces a rapid, acute intestinal inflammatory response resulting from the administration of bacterial toxins, and Giardia infections have been reported to occur concurrently with C. difficile [16]. Small intestinal trophozoites numbers in Giardia GS/M-infected animals were higher than numbers from NF-infected animals ( Figure 1A). These results are consistent with previous observations that assemblage B Giardia isolates, such as GS/M, have higher parasite burdens compared to assemblage A isolates, such as NF trophozoites [50,54]. Administration of C. difficile TcdAB for 3 hours did not change trophozoite numbers ( Figure 1A).
As previously demonstrated [52], intracolonic administration of 100 mg of C. difficile TcdAB significantly increased colonic tissue MPO activity ( Figure 1B and C). As determined via immunofluorescence, numbers of MPO-positive cells increased in colonic tissues of TdcAB-trated animals compared to control animals Giardia NF infections attenuate neutrophil-associated cytokines during TcdAB-induced colitis A bead-based cytokine assay on colonic tissue samples was performed to determine whether Giardia NF infections attenuated protein levels of pro-inflammatory cytokines and chemokines associated with granulocyte tissue recruitment. Compared to control animals, intracolonic instillation of 100 mg of TcdAB significantly increased tissue levels of several PMN-associated mediators, including CXCL1, CXCL2, IL-17 and G-CSF (Figure 3). Colonic tissues collected from Giardia NF infectedanimals administered TcdAB. demonstrated significantly lower protein levels of CXCL1, CXCL2, and IL-17 ( Figure 3A to C), while G-CSF protein levels were similar from uninfected TcdABtreated animals ( Figure 3D). Notably, colonic protein levels of CXCL1, CXCL2, IL-17, and G-CSF in Giardia GS/M-infected animals were similar from uninfected TcdAB-treated animals ( Figure 3E to H). These results support above observations that granulocyte infiltration is not attenuated in Giardia GS/M infected animals instilled with TcdAB, and suggest attenuation of granulocyte infiltration in Giardia NF-infected animals given TcdAB may result from decreased expression of PMN-associated mediators. These results also indicate that Giardia-mediated attenuation of PMN-associated mediators induced via TcdAB occurs in an isolate-dependent manner.
Bead-based cytokine analysis of colonic tissues indicated Giardia GS/M infections upregulated colonic expression levels of a subset of pro-inflammatory mediators, following TcdAB. In uninfected animals, colonic administration of TcdAB significantly increased colonic expression levels of IL-1b and CXCL10 ( Figure 5), and protein levels of these mediators were significantly increased in Giardia GS/M-infected animals compared to uninfected TcdAB-treated animals ( Figure 5). Giardia GS/M-infected animals given TcdAB also demonstrated increased colonic protein levels of several inflammatory mediators not initially increased via colonic instillation of TcdAB ( Figure S3). This did not appear to be a global increase in expression of inflammatory mediators, as CCL3 and CCL11 concentrations were similar between Giardia GS/Minfected animals given TcdAB and uninfected TcdAB controls ( Figure S4). Several inflammatory mediators also remained unchanged between all experimental groups ( Figure S5). These results suggest Giardia GS/M infections enhance expression of pro-inflammatory cytokines following colonic instillation of TcdAB. These data indicate that Giardia-infections can enhance or decrease the expression of inflammatory mediators following administration of a bacterial toxin, events which are isolatespecific.

Giardia NF trophozoites attenuate various inflammatory mediators from inflamed human mucosal biopsy tissues ex vivo
Our above results demonstrated in vivo that Giardia NF attenuated PMN accumulation and expression of pro-inflammatory cytokines following colonic administration of TcdA/TcdB. Therefore, we next assessed whether this same isolate was capable of attenuating similar inflammatory mediators from inflamed human colonic tissue. Using a previously published model [41], Giardia NF trophozoites were co-incubated ex vivo with colonic mucosal biopsy tissues collected the descending colon of patients with active CD, where supernatants and biopsy tissue homogenates were analyzed via a bead-based cytokine assay. Supernatants collected from co-incubation of Giardia NF trophozoites with inflamed biopsy tissues displayed significantly reduced supernatant levels of cytokines associated with granulocyte recruitment, including IL-8/CXCL8, growth-related oncogene (GRO) family proteins (CXCL1-3), IL-17A, G-CSF, and GM-CSF ( Figure 6). Co-incubation of Giardia NF trophozoites with biopsy tissues also resulted in decreased IL-8/CXCL8 levels detected within biopsy tissue homogenates ( Figure 7A), but not G-CSF and GM-CSF.
As in vivo Giardia NF attenuated colonic levels of CCL2, IL-6, LIF, and, potentially, IL-12p70 (Figure 4), we investigated whether live trophozoites were capable of attenuating protein expression of other chemokines. Supernatant levels of CCL2-5, CCL7, CCL11, CCL22, CXCL10, and CX 3 CL1 were significantly reduced in biopsy tissues co-incubated with Giardia NF trophozoites compared to control biopsy tissues (Figure 8). Tissue homogenate levels of these mediators were similar when biopsy tissues were incubated in the presence or absence of Giardia NF trophozoites ( Figure 9). Therefore, attenuation of these mediators by G. duodenalis NF trophozoites appears to occur following their release into supernatants. These results also suggest that Giardia NF trophozoites are capable of attenuating CCL2 in different experimental models of inflammation. Compared to control biopsy supernatant, co-incubation of Giardia NF trophozoites with inflamed biopsy tissues ex vivo resulted in significant attenuation of IL-6, IL-7, IL-10, IL-12p70, IFN-a, soluble CD40 ligand (sCD40L) and vascular endothelial growth factor (VEGF) (Figure 10). Tissue homogenate levels of IL-1a, IL-6, and TNF-a were significantly reduced in biopsy tissue homogenates coincubated with Giardia NF trophozoites, compared to biopsy tissue homogenates incubated in the absence of Giardia NF trophozoites ( Figure 11). These results suggest that Giardia NF trophozoites attenuate secreted IL-7, IL-10, IL-12p70, IFN-a, sCD40L, and VEGF from inflamed mucosal biopsy tissues ex vivo, while also attenuating tissue levels of IL-1a, IL-6, and TNF-a. These results also suggest that Giardia infections are capable of attenuating IL-6 and IL-12p70 in during an in vivo infection and following incubation with human intestinal biopsy tissues ex vivo. Our data also indicated that Giardia NF trophozoites did not attenuate the expression of pro-inflammatory cytokines (Figure S6 and S7) or growth factors ( Figure S8 and S9) within supernatants or biopsy tissue homogenates.

Discussion
Results from the present study demonstrate novel mechanisms through which certain Giardia infections were able to attenuate pro-inflammatory responses elicited by an inflammatory bacterial toxin in vivo, or from inflamed human intestinal tissues ex vivo. Attenuation of granulocyte infiltration in Giardia NF infected animals given TcdAB occurred concomitantly with reduced colonic expression of several inflammatory mediators, including those associated with granulocyte tissue recruitment. Several of these factors were also decreased following co-incubation of Giardia NF trophozoites and ex vivo inflamed human colonic mucosal biopsy tissues, and, therefore, support our in vivo observations that Giardia NF trophozoites are capable of attenuating factors associated with PMN recruitment. These results also reinforce previous observations that Giardia trophozoites are capable of attenuating IL-8/CXCL8 production from intestinal tissues in vitro and ex vivo [41]. Furthermore, our data indicate that Giardia trophozoites can attenuate the production of additional inflammatory mediators from intestinal tissues of differing origin, strengthening the notion that infection with specific Giardia isolates may modulate host immune responses in the gut. Interestingly Giardia GS/M-infected animals did not display reduced granulocyte infiltration nor decreased expression of inflammatory mediators. On the contrary, Giardia GS/M infections enhanced the expression of several factors initially upregulated following TcdAB administration and increased the expression of several factors not initially induced by TcdAB exposure. Therefore, our data suggest that Giardia infections attenuate intestinal pro-inflammatory responses in an isolatedependent manner, and attenuation of such responses is associated with reduced tissue granulocyte infiltration. The results show that Giardia NF attenuate a variety of pro-inflammatory mediators when co-incubated with inflamed colonic mucosal biopsy tissues ex vivo, but the inability to decrease expression of all mediators examined suggests this process may be targeted towards certain inflammatory mediators. Moreover, observations that some inflammatory mediators are preferentially degraded within supernatants while others within tissues suggests that Giardia NF trophozoites possess multiple mechanisms capable of attenuating expression of inflammatory mediators in inflamed intestinal mucosal biopsy tissues.
Excessive PMN infiltration and activation contribute to several gastrointestinal inflammatory disorders, including the inflammatory bowel diseases (IBD) and C. difficile colitis [55][56][57]. The present findings indicate that Giardia NF infections in vivo attenuate PMN tissue recruitment induced by a pro-inflammatory bacterial toxin by reducing tissue expression of mediators that can contribute to the tissue recruitment of PMNs. Release of ELR+ chemokines, such as CXCL1, CXCL2, and IL-8/CXCL8, from a variety of intestinal tissue-resident cells promotes PMN accumulation and chemotaxis through intestinal tissues. In mice, these chemokines include CXCL1 and CXCL2, while in humans this also includes IL-8/CXCL8 [58][59][60][61]. Separately, enhanced expression of G-CSF promotes the release of PMNs into the bloodstream [31,62], in a process that can be initiated via IL-17 production [32,33]. Our results suggest that Giardia NF infections attenuate granulocyte infiltration into colonic tissues following colonic. TcdA/B by reducing colonic expression levels of CXCL1, CXCL2, and IL-17; these results are supported by observations that Giardia GS/M infections failed to attenuate colonic granulocyte infiltration and expression of PMN-associated mediators. In addition, Giardia NF parasites attenuated supernatant levels of IL-8/CXCL8, GRO-family proteins (CXCL1-3), IL-17, G-CSF, and GM-CSF when co-incubated with ex vivo descending colon mucosal biopsy tissues. Together, these data show for the first time that Giardia infections attenuate the release of a variety of mediators associated with PMN recruitment in different experimental models of inflammation, in an isolate-dependent manner.
Further research is needed to demonstrate whether the results from this study explain how Giardia infections can, in a straindependent manner, modulate the incidence of diarrheal disease [20,22], and create an environment that is favourable for colonization by other gastrointestinal pathogens [18]. Indeed, the data suggest that Giardia may attenuate pro-inflammatory responses elicited by other co-infecting pathogens that can cause diarrheal disease, while other strains may enhance its development. Following their recruitment to tissues, PMNs are known to induce multiple pathophysiological events within intestinal tissues capable of causing diarrhea [28,[42][43][44]. Therefore, Giardia NF infections may protect against the development of diarrheal disease by attenuating PMN recruitment. Giardia GS/M infections did not enhance granulocyte infiltration following colonic of TcdAB; however, these infections resulted in enhanced expression of several pro-inflammatory mediators that could, potentially, enhance the development of diarrheal disease via other pathways. For example, IL-1b and IFNc have been shown to modulate intestinal barrier dysfunction and ion secretion, and, resultantly, cause diarrhea [63][64][65][66][67][68]. As PMNs are essential to host defence against a variety of pathogens, more research is needed to determine whether attenuation of PMN recruitment by Giardia may also facilitate colonization by other pathogens. To date, research in vitro and in vivo has demonstrated differences in the pathogenesis of Giardia isolates, but has failed to clearly establish whether these differences are assemblage-specific. [10,50,54,69,70].
Data from this study also demonstrate that Giardia infections reduce the expression of pro-inflammatory mediators not associated with PMN recruitment, in an isolate-dependent manner. Giardia NF infections attenuated colonic expression of IL-6 and the related cytokine LIF induced following i.r. TcdAB, whereas Giardia GS/M did not. Moreover, co-incubation of Giardia NF trophozoites with inflamed human intestinal mucosal biopsy tissues ex vivo also resulted in attenuation of supernatant and tissue IL-6. The ability of Giardia sp. to modulate IL-6 during in vivo infection remains incompletely understood. These observations are consistent with previous studies that have shown that animals infected with Giardia muris display decreased IL-6 levels in jejunal tissues 5 days post-infection [71], while mice infected with Giardia GS/M display elevated IL-6 intestinal tissue levels [72]. The release of IL-6 and the related cytokine LIF can induce the expression of acute phase response proteins, including CRP [73,74]. As Tanzanian children infected with Giardia were found to have lower serum CRP than their non-infected counterparts [22], future studies should examine whether Giardia infections can modulate the acute phase response in a strain-dependent manner, and establish whether the parasite-mediated attenuation of IL-6 expression is involved in this process.
The present findings illustrate that Giardia NF infections attenuate colonic expression of CCL2 and IL-12p70 in vivo, as well as in inflamed human intestinal mucosal biopsy tissues ex vivo. CCL2 is an important chemokine for monocytes and macrophages (reviewed in [75]). Conversely, co-incubation of Giardia trophozoites with in vitro Caco-2 monolayers results in increased mRNA expression of CCL2 [76]. Additional research is required in order to understand how Giardia infections may modulate monocyte/ macrophage recruitment. IL-12p70 is composed of IL-12p35 and IL-12p40, and is the bioactive form responsible for inducing helper T1 (T H 1) adaptive immune responses ( [77]). Recent research has demonstrated that Giardia trophozoites products are capable of attenuating IL-12p70 expression from dendritic cells (DCs) stimulated with bacterial lipopolysaccharide [78]. In contrast other studies found that Giardia products may increase DC IL-12p70 expression [79,80]. More research is required in order to determine how Giardia infections may modulate CCL2 and IL-12p70 expression in the context of intestinal inflammatory responses. We have recently shown that Giardia cathepsin B (catB) cysteine proteases degrade intestinal epithelial IL-8/CXCL8 and can attenuate PMN chemotaxis [41]. It remains to be determined whether Giardia catB proteases are involved in attenuating PMN accumulation in vivo and during human giardiasis. Genetic manipulation of the Giardia genome with Cre/loxP system has recently been demonstrated [81]. This would be extremely beneficial in determining the anti-inflammatory potential of Giardia catB proteases in vivo. Whether Giardia may exert immune-modulatory effects using other secretory-excretory products apart from Giardia catB proteases remains largely unknown. Giardia arginine deiminase has been shown to attenuate nitric oxide production from intestinal epithelial cells [82][83][84], to inhibit intestinal epithelial proliferation [85], to induce expression of IL-12p70 from DCs [80], and to inhibit T-cell proliferation [84], however these effects have yet to be characterized in vivo.
In conclusion, results from this study demonstrate that Giardia infections can attenuate granulocyte infiltration and expression of PMN chemokines in an isolate-dependent manner in an in vivo model of infectious colitis. Giardia trophozoites were also able to reduce the expression of a variety of inflammatory mediators released from inflamed colonic mucosal biopsy tissues from CD patients. These findings establish the strain-dependent antiinflammatory properties of Giardia in the intestine. Figure S1 Giardia NF infections in vivo do not attenuate colonic expression levels of IL-1b, IL-5, CXCL10, or CCL11 during C. difficile TcdA/TcdB-induced colitis. Male C57BL/6 mice, aged 4 to 6 weeks, were infected with Giardia NF trophozoites. 7 days post-infection, animals were given 100 mg of TcdA/TcdB or PBS i.r. and, after 3 hours, animals were euthanized. Levels of (A) IL-1b, (B) IL-5, (C) CXCL10, and (D) CCL11 were compared between Giardia NF-infected animals and uninfected animals given i.r. 100 mg of TcdA/TcdB or PBS.

Supporting Information
Values were determined via bead-based cytokine assay. All data are representative of two independent experiments (n = 5-11/ group) and represented as mean 6 SEM. n.s. = not significant * p,0.05. (TIF) Figure S2 Giardia NF infections in vivo do not attenuate colonic expression levels of inflammatory mediators not increased during C. difficile TcdA/TcdB-induced colitis. Male C57BL/6 mice, aged 4 to 6 weeks, were infected with Giardia NF trophozoites. 7 days post-infection, animals were given 100 mg of TcdA/TcdB or PBS i.r. and, after 3 hours, animals were euthanized. Levels of IL-1a  Figure S4 Giardia GS/M infections in vivo do not modulate colonic expression levels of CCL3 and CCL11 during C. difficile TcdA/TcdB-induced colitis. Male C57BL/6 mice, aged 4 to 6 weeks, were infected with Giardia GS/M trophozoites. 7 days post-infection, animals were given 100 mg of TcdA/TcdB or PBS i.r. and, after 3 hours, animals were euthanized. Levels of (A) CCL3 and (B) CCL11 were compared between Giardia GS/Minfected animals and uninfected animals given i.r. 100 mg of TcdA/TcdB or PBS. Values were determined via bead-based cytokine assay. All data are representative of two independent experiments (n = 5-11/group) and represented as mean 6 SEM. * p,0.05. (TIF) Figure S5 Giardia GS/M infections in vivo do not attenuate colonic expression levels of inflammatory mediators not increased during C. difficile TcdA/TcdB-induced colitis. Male C57BL/6 mice, aged 4 to 6 weeks, were infected with Giardia GS/M trophozoites. 7 days post-infection, animals were given 100 mg of TcdA/TcdB or PBS i.r. and, after 3 hours, animals were euthanized. Levels of (A) IL-1a, (B) IL-7, (C) IL-9, (D) IL-10, (E) IL-12p40, and (F) TNFa were compared between Giardia GS/Minfected animals and uninfected animals given i.r. 100 mg of TcdA/TcdB or PBS. Values were determined via bead-based cytokine assay. All data are representative of two independent experiments (n = 5-11/group) and represented as mean 6 SEM. (TIF) Figure S6 Giardia NF trophozoites do attenuate supernatant levels of several inflammatory mediators released from descending colon mucosal biopsy tissues. Human descending colon mucosal biopsy tissues obtained in areas of active inflammation from patients with active Crohn's disease (CD) were incubated with 5.0610 6 Giardia NF trophozoites for 6 hours. Supernatant levels of (A) IL-1ra, (B) IL-1b, (C) IL-2, (D) IL-12p40, (E) IL-13, (F) IL-15, (G) IFNc, and (H) TNFb were determined via a bead-based cytokine assay. All data are representative of three independent experiments (n = 4-6/group) and represented as median with the range. n.s. = not significant. (TIF) Figure S7 Giardia NF trophozoites do attenuate levels of several inflammatory mediators released from descending colon mucosal biopsy homogenates. Human descending colon mucosal biopsy tissues obtained in areas of active inflammation from patients with active Crohn's disease (CD) were incubated with 5.0610 6 Giardia NF trophozoites for 6 hours. Tissue homogenate levels of (A) IL-1ra, (B) IL-1b, (C) IL-2, (D) IL-12p40, (E) IL-13, (F) IL-15, (G) IFNc, and (H) TNFb were determined via a bead-based cytokine assay. All data are representative of three independent experiments (n = 4-6/group) and represented as median with the range. n.s. = not significant. (TIF) Figure S8 Giardia NF trophozoites do attenuate supernatant levels of several growth factors released from descending colon mucosal biopsy tissues. Human descending colon mucosal biopsy tissues obtained in areas of active inflammation from patients with active Crohn's disease (CD) were incubated with 5.