• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Biochem Parasitol. Author manuscript; available in PMC May 20, 2013.
Published in final edited form as:
PMCID: PMC3658456

Molecular & Biochemical Parasitology Release of metabolic enzymes by Giardia in response to interaction with intestinal epithelial cells


Giardia lamblia, an important cause of diarrheal disease, resides in the small intestinal lumen in close apposition to epithelial cells. Since the disease mechanisms underlying giardiasis are poorly understood, elucidating the specific interactions of the parasite with the host epithelium is likely to provide clues to understanding the pathogenesis. Here we tested the hypothesis that contact of Giardia lamblia with intestinal epithelial cells might lead to release of specific proteins. Using established co-culture models, intestinal ligated loops and a proteomics approach, we identified three G. lamblia proteins (arginine deiminase, ornithine carbamoyl transferase and enolase), previously recognized as immunodominant antigens during acute giardiasis. Release was stimulated by cell–cell interactions, since only small amounts of argi-nine deiminase and enolase were detected in the medium after culturing of G. lamblia alone. The secreted G. lamblia proteins were localized to the cytoplasm and the inside of the plasma membrane of trophozoites. Furthermore, in vitro studies with recombinant arginine deiminase showed that the secreted Giardia proteins can disable host innate immune factors such as nitric oxide production. These results indicate that contact of Giardia with epithelial cells triggers metabolic enzyme release, which might facilitate effective colonization of the human small intestine.

Keywords: Parasite, Cell–cell interaction, Innate immunity, Secretory product, Arginine deiminase, Enolase

1. Introduction

The intestinal protozoan pathogen Giardia lamblia annually infects around 300 million people worldwide [1]. Infection usually starts by ingestion of cysts, which differentiate into trophozoites that colonize the upper small intestine [2]. Clinical manifestations of giardiasis vary from asymptomatic carriage to chronic diarrhea and severe malabsorption [3]. The mechanisms by which G. lamblia causes disease are poorly understood. The parasite is not invasive and little or no mucosal inflammation is seen during acute infection [4]. The severity of disease may depend on multiple parasite and host factors [3,4]. One of the proposed mechanisms is secretion of proteins with toxin-like activities [5,6] but no giardial toxin has been identified to date. In vitro interaction models for G. lamblia and intestinal cells are well established [711], as are rodent models [1214] and this has increased the knowledge of the infection [1517].

The overall goal of this study was to identify proteins released by the parasite during its interaction with host cells. Earlier studies had indicated that parasite proteins are released into the growth medium during incubation of axenic cultures of Giardia trophozoites in serum-free medium but no specific proteins have been identified so far. However, the highly immuno-reactive variable surface proteins (VSPs) [18,19] and an unidentified 58 kDa protein causing intestinal fluid accumulation in mice [5,20] have been reported to be excretory–secretory factors.

Recently several cysteine-type proteases of G. lamblia origin were detected after co-incubations with rat small intestine epithelial cells [15], and other host cell–Giardia interaction experiments and giardiasis patient data have shown that Giardia reduces the epithelial barrier function [16,17] and induces apoptosis [14,2123], but the effector protein(s) have never been identified.

To improve our understanding of host–parasite interactions during infection, we asked whether exposure of G. lamblia to human intestinal epithelial cells (IEC) might lead to release of trophozoite proteins into the medium. We recently showed that a secreted parasite factor induces chemokine expression in human Caco-2 IECs during Giardia interaction [8]. Here we identified three major Giardia proteins released into the medium after only brief interaction between G. lamblia and IEC. These three enzymes (arginine deiminase (ADI), ornithine carbamoyl transferase (OCT) and enolase) function in giardial metabolism and are immunoreactive during human and murine infections [24,25]. This is the first study identifying proteins in the secretome of Giardia, and the first study in a human homologue system (Giardia of human origin-human IECs) presenting evidence for interaction related excretory–secretory products of Giardia lamblia.

2. Materials and methods

2.1. Reagents and cell culture

Unless otherwise indicated, reagents were obtained from Sigma Chemical Co, USA. Giardia lamblia strain WB (ATCC30957), clone C6, and GS, clone H7 (ATCC50581) trophozoites were grown as described [8]. Intestinal epithelial cell lines HT-29, Caco-2, cervical epithelial HeLa cells and human fibroblasts were all grown in high glucose DMEM supplemented with 10% FBS, 4mM L-glutamine, 1 × MEM non-essential amino acids, 160 μg/ml streptomycin and 160 U/ml penicillin G at 37 °C and 5% CO2. IEC-6 cells were grown according to ATCC specifications. Prior to interaction experiments, the Caco-2 cells were differentiated into small intestine-like enterocytes by post-confluence cultivation for 14–17 days, changing the medium twice weekly. The enterocyte phenotype was confirmed by immunolocalization of ZO-1 and presence of tight junctions and by elevated mRNA expression of intestinal alkaline phosphatase and aminopeptidase N compared to non-differentiated cells [8].

2.2. In vitro host–parasite interaction and protein precipitation

HT-29 or differentiated Caco-2 cells were washed 3 times in 37 °C PBS before initiating interaction with PBS washed trophozoites, with a cell ratio (parasite:IEC) of 3:1. The interacting cells, and controls of IECs and Giardia separately, were incubated in culture flasks filled with serum-free M199, supplemented with 6 mM ascorbic acid and 6 mM cysteine, pH adjusted to 7.2, at 37 °C for 2.5 h. The condition of the cells was monitored by phase-contrast light microscopy during the interaction and viable Giardia trophozoite numbers were counted before and after interaction using trypan blue staining according to instructions from the manufacturer (Sigma Chemical Co, USA).

After interaction, the culture medium was pre-cleared of cells by centrifugation (2500 rpm, 15 min at 4 °C) and then filtered through a 0.22 μm pore filter (Pall Corporation, USA). Proteins were precipitated over-night at 4 °C with 10% trichloroacetic acid, collected at 2500 rpm, 30 min at 4 °C followed by drying at room temperature. Pellets from interactions and control experiments were dissolved in PBS and equal volumes were analyzed on 2D gels. Samples for 1D Western blot analysis were precipitated similarly, changing the TCA concentration to 20% and adding 0.02% DOC 30 min before the over night precipitation, followed by a wash in cold acetone before drying of the pellet.

2.3. Protein identification by 2D gel electrophoresis and mass spectrometry

Precipitated proteins were separated by 2D gel electrophoresis according to the PS-1 protocol [24], stained and analyzed by mass spectrometry as described [24]. 70- or 180-mm ReadyStrip IPG Strips (BioRad Laboratories Inc., USA), in pI ranges nonlinear 3–10 and linear 5–8, were used for isoelectric focusing. For isolation of proteins for mass spectrometry analysis, gels were stained with Coomassie brilliant blue and all major spots were excised. Protein spots were subjected to in-gel digestion with trypsin (modified sequence grade, porcine; Promega, USA). Peptide spectra were internally calibrated against autolytic peptides from trypsin. Matrix-assisted laser desorption ionization (MALDI) analysis was performed on a Bruker Reflex III mass spectrometer (Bruker-Franzen Analytik, Germany) using alpha-cyano-4-hydroxycinnamic acid as matrix.

Peptide maps were searched against simulated tryptic digests in NCBInr database using the Mascot search engine at Matrix science (http://www.matrixscience.com). No miss-cleavages were allowed and the peptide mass tolerance was set to <0.2 Da. Significance of identification was evaluated using the Probability Based Mowse Score of the Mascot Program. For confirmation of spot 4 (G. lamblia enolase) tandem mass spectrometry was performed on a hybrid quadruple-orthogonal acceleration time-of-flight (TOF) instrument (Micromass, UK) and the partial amino acid sequence obtained was compared to protein sequences from the G. lamblia genome project (www.mbl.edu/giardia).

2.4. Production of recombinant proteins

Enolase, ADI, and OCT were expressed and purified as described [24].

2.5. Polyclonal antibody production

BALB/c mice were immunized intraperitoneally as described [26] on days 0, 15, and 30 with 50 μg of purified recombinant ADI, OCT or enolase re-suspended in 100 μl PBS and emulsified with an equal volume of RIBI adjuvant (Corixa, USA). Prior to initial immunization and after each boost, blood was collected from the tail vein and the serum fraction was assayed for specific antibody titers. The specificity of each polyclonal antibody was determined by Western blot analysis using total Giardia extract and the antibodies were shown to be specific.

2.6. Localization of ADI, OCT and enolase in trophozoites

Trophozoites were harvested by chilling on ice for 30 min, washed twice in ice-cold PBS, and fixed with 3% paraformaldehyde for 30 min at room temperature, followed by 5 min in 100 mM glycine in PBS. Fixed cells were permeabilized with 0.1% Triton X-100 in PBS for 30 min and blocked >1 h in 2% bovine serum albumin in PBS. The mouse polyclonal antibodies against ADI, OCT and enolase were diluted 1:200 in PBS containing 0.1% Triton X-100, incubated with fixed cells for 1 h, washed 3 times with PBS and incubated for 1 h with FITC-conjugated sheep anti-mouse antibody (dilution 1:2000, DAKO, Denmark). Fluorescence microscopy was performed on a Leica DM-IRBE microscope using a 100× HCX PL Fluotar lens (Leica Microsystems GmbH, Germany) and digital images were recorded using a cooled CCD camera (Diagnostic Instruments Inc., USA) and processed with the Metaview software package (Visitron Systems GmbH, Germany).

2.7. Localization of proteins during host–parasite interaction

Co-culture of human differentiated Caco-2 cells and G. lamblia trophozoites was performed at a 3:1 parasite:host cell ratio for 18 h in 8-well slide chambers. Complete DMEM with serum and antibiotics were used with 5% CO2 as described [8]. Wells were washed 3 times with warm PBS then fixed and immuno-stained as above. Cell interactions and protein distribution patterns were observed using a Zeiss Axioplan II Imaging fluorescence microscope equipped with appropriate filter sets, an Axiocam CCD camera and AXIOVISION software (Carl Zeiss Light Microscopy, Germany).

2.8. Detection of secreted proteins using Western blot analysis

Secreted proteins from 15 min to 2.5 h co-incubations of confluent HT-29 cells or differentiated Caco-2 cells, HeLa cells, human fibroblasts, IEC-6 rat small intestinal cells and WB-C6 Giardia trophozoites, was analyzed by Western blots with antibodies against ADI, enolase and OCT. The secreted proteins were precipitated with 0.02% DOC and 20% TCA followed by a wash in cold acetone for enhanced yield and the resulting pellets were dissolved in 1X SDS Loading Buffer to give a 200-fold final concentration of the released co-culture proteins. For acidic samples, the pH was adjusted to ≥pH 6.8, by addition of NaOH. 10 μl of each precipitation was separated by 10% SDS-PAGE and transferred to a PVDF membrane (Millipore Corporation, USA).

The PVDF membrane was blocked in 5% non-fat dried milk, 1X TBS, 0.1% Tween-20 at 4 °C over-night, incubated with primary antibodies against ADI, enolase or OCT, diluted 1:2000 in blocking solution. Specifically bound primary antibodies were detected by horseradish peroxidase-labeled goat anti-mouse Ig antibody (DAKO, Denmark), diluted 1:30,000 in blocking solution. The membrane was incubated, washed in TBS-Tween and developed using the ECL Plus Western blotting detection system (GE Healthcare, Sweden).

2.9. Identification of proteins secreted in vivo

To examine the secretion of Giardia proteins in vivo, ligated jejunal loops (2 cm) were prepared in deeply anesthesized adult C57BL/6 mice and infected with 107 trophozoites G. lamblia GS/M trophozoites in 200 μl PBS. Control loops were injected with 200 μl PBS alone. Two mice were used per condition. At 4 h the loops were removed, opened up and placed into 1 ml ice-cold intestinal wash buffer (PBS, 1 mM DTT, 2× Complete Protease Inhibitors; Roche). The intestinal washes were first centrifuged (2500 rpm, 15 min at 4 °C) to remove cells, followed by a second centrifugation (13,000 rpm, 15 min at 4 °C) to clear the supernatants from cellular debris. As positive controls the same number of trophozoites were incubated in 1 ml PBS for 4 h at 37 °C, sonicated and centrifuged at 13,000 rpm for 15 min at 4 °C. 200 μl of the supernatant was added to 1 ml intestinal washing buffer from uninfected control mice to mimic the conditions from the ligated loop experiments. The protein content in each sample was measured by BioRad Bradford analysis, and 5 μg protein was mixed with 6XSDS loading buffer and separated by 10% SDS-PAGE. The ligated loop samples were screened for presence of ADI, OCT and enolase by Western blots as described above. All animal experiments were approved by the UCSD Institutional Animal Care and Use Committee.

2.10. ADI activity and nitric oxide production

Enzymatic activity of purified recombinant ADI was determined after buffer change as described [27]. 100 ng reactive protein or boiled (not shown) in complete DMEM was added to differentiated Caco-2 epithelial cells in 12 well-plates (Corning, USA) containing 1 ml cell culture medium per well. The cells were stimulated for nitric oxide (NO) production by addition of IL-1 (20 ng/well), IFNα (50 ng/well) and TNF (20 ng/well) simultaneously with ADI or buffer control. At 18 h post cytokine addition, growth medium was collected and analyzed for total nitrite production using the Total Nitric Oxide kit (Promega, USA). Quantification of total nitrate of un-stimulated Caco-2 cells and stimulated Caco-2 with only protein buffer added was used as controls.

3. Results

3.1. Identification of secreted proteins by 2D gel electrophoresis and mass spectrometry

Giardia trophozoites were cultured in vitro with or without confluent intestinal epithelial cells (differentiated Caco-2 or HT-29) in serum-free medium for up to 2.5 h and monitored by light microscopy. The trophozoites were motile and maintained normal morphology throughout the experimental period. Live trophozoite numbers, determined by trypan blue staining, did not differ significantly before and after interaction, suggesting absence of major cell lysis (data not shown). After 3 h of co-incubation in a low oxygen environment (closed caps, filled 250 ml culture flasks) the intestinal epithelial cells began to swell and detach from the surface of the flask. This was not seen if epithelial cells were cultivated in co-cultures with serum and an oxygen-rich environment (in open 6-well plates).

Samples of culture media from 2.5 h interactions between Giardia and human intestinal cells were analyzed by 2D gel electrophoresis and silver staining of the gels (Fig. 1A–C). The same dominant proteins were detected on 2D gels with interaction media of both HT-29 (Fig. 1A) or differentiated Caco-2 cells (Fig. 1C) and Giardia cells. The dominant protein spots were much less abundant when parasites were incubated without intestinal cells (Fig. 1B). All visible spots, equivalent to the most dominant spots of silver stained gels (Fig. 1A–C), were excised from a Coomassie stained 2D gel, digested in-gel with trypsin and analyzed by mass spectrometry.

Fig. 1
Precipitated culture supernatant proteins from 2.5 h incubations of Giardia lamblia trophozoites with (A) and without human HT-29 cells (B) and with differentiated small intestine-like Caco-2 cells (C). Identified Giardia and human proteins are indicated ...

Peptide mass fingerprints were compared to theoretical digests of proteins from all species at NCBI. G. lamblia OCT and ADI were identified in the co-culture supernatant. Tandem mass spectrometry analysis was performed for final identification of spot number 4, which revealed Giardia enolase (Table 1). In addition, human alpha-enolase (syn. enolase-1) and heat shock 70 kDa protein 8 (HSPA8) were identified. HSPA8 is too highly conserved to distinguish between human and bovine origin, but it was only detected in the medium of co-incubated cells indicating specific release by the human cells during the interaction. A constant background of bovine serum proteins was identified in the medium of co-cultures and in the controls, and several spots of lower intensity, possibly corresponding to interaction related proteins, failed to be identified (Fig. 1A–C). The 2D pattern of soluble protein fractions of Giardia trophozoites is much more complex than that of the co-culture supernatants (data not shown, [24]) supporting the idea that the results are not due to large scale lysis of trophozoites. For example, the abundant cytoplasmic, metabolic enzymes fructose-1,6-biphosphate aldolase and UPL-1 [24] were not detected after co-incubation (Fig. 1).

Table 1
Identified proteins from 2D gels of interaction supernatant of G. lamblia and HT-29 co-cultures

3.2. Localization of ADI, OCT and enolase in Giardia trophozoites

The released Giardia proteins OCT, ADI and enolase identified here have important metabolic roles in the parasite [28], but have not previously been localized. We produced recombinant versions of the three proteins and raised polyclonal antibodies in mice. Each serum reacted with one band of the expected molecular weight in Western blots (data not shown). Immunofluorescence microscopy showed that ADI and Enolase exhibited a punctuate pattern within the cytoplasm (Fig. 2A and C), while OCT localized in the cytoplasm with highest accumulation prominently near the plasma membrane (Fig. 2B). During co-incubation with intestinal cells the cytoplasm was stained with all antibodies but there were also differences in the staining pattern from separately grown parasites; ADI localized to the most anterior part of the parasite (Fig. 2G) and OCT and enolase appeared to be more concentrated in the central anterior area (Fig. 2H and I).

Fig. 2
Localization of identified proteins in G. lamblia trophozoites by polyclonal mouse antibodies (green) raised against purified recombinant ADI (A), OCT (B) and enolase (C). Corresponding light microscopy pictures are shown in panels (D–F). Figs. ...

3.3. Western blot analysis of ADI, OCT and enolase release in vitro and in vivo

To further analyze protein release during host–parasite interactions, we used the specific polyclonal antibodies against OCT, ADI and enolase in Western blots of supernatants precipitated from interactions between trophozoites and human IECs. Fig. 3A shows that ADI and enolase, and to a lesser degree OCT, were secreted into the growth medium within 15 min of interaction. Co-cultivation of Giardia and differentiated Caco-2 cells for 15 min gave a 7-fold higher secretion of ADI and a 3-fold higher enolase secretion compared to Giardia grown without epithelial cells (Fig. 3A). Incubation of Giardia trophozoites with HT-29, human cervical epithelial cells (HeLa), human fibroblasts and rat small intestinal epithelial cells (IEC-6) also induced release of the enzymes (Fig. 3B). Note however that the level of released ADI relative to enolase is different compared to the differentiated Caco-2 cell experiment (Fig. 3A and B). The difference in relative amounts of released ADI and enolase between HT-29 and differentiated Caco-2 cells is also obvious in the 2D analysis (Fig. 1A and C, spots 3 and 4). These experiments show that there is no specificity for a certain cell-type but the relative levels of secreted ADI was higher after incubation with the differentiated human intestinal epithelial cells (Caco-2) compared to the other cell lines. It should also be noted that Giardia trophozoites only interact with intestinal epithelial cells during a natural infection, which can explain the relaxed cell type specificity for protein release.

Fig. 3
Detection of secreted Giardia proteins during human cell–trophozoite interactions in vitro. The presence of ADI, enolase and OCT was detected in concentrated culture supernatants from co-cultures of Giardia lamblia WB-C6 and differentiated Caco-2 ...

The in vivo release of Giardia enzymes was studied in C57BL/6 mice infected with G. lamblia GS trophozoites, the only human Giardia strain colonizing mouse small intestine. Fig. 4 shows that enolase and a fragment of ADI (indicated by an asterisk) are detected in the intestinal luminal fluid from Giardia-infected ligated loops whilst OCT could not be detected in these samples. These data are consistent with the observation that giardial ADI and enolase are released during host–Giardia interaction.

Fig. 4
Detection of secreted Giardia proteins during host cell–trophozoite interaction in vivo. ADI (A) and enolase (B) were detected by Western blot analysis of intestinal washes after 4 h Giardia GS-H7 infection of C57BL/6 mice ligated intestinal loops. ...

3.4. Characterization of the function of released Giardia proteins

Enolase and ADI are released both in vitro and in vivo during host–Giardia interactions. Enolase has been shown to be a multi-functional enzyme often secreted by pathogens at mucosal surfaces [29]. However, the function of secreted enolase is poorly understood and we were not able to show any specific extra-cellular activity of enolase.

The released Giardia protein with the relatively greatest increase upon interaction of Giardia and differentiated human intestinal epithelial cells was ADI (Fig. 3A and B). This enzyme plays an important role in the intracellular arginine metabolism of Giardia [28]. We showed earlier that Giardia deplete arginine required by intestinal epithelial cells for nitric oxide (NO) production [19]. Secreting ADI, may enable the parasite to disarm the potentially detrimental effects of NO-dependent host defenses. To test this, we evaluated the effect of recombinant Giardia ADI on epithelial cell NO-production. The differentiated, small intestine-like epithelial cell line Caco-2 was stimulated for nitric oxide production by the addition of cytokines (IL-1α, IFN-γ and TNF-α), together with enzymatically active recombinant G. lamblia ADI. As seen in Fig. 5, active, but not boiled (data not shown) Giardia ADI suppressed the production of NO, as evidenced by decreased levels of the stable NO end product, nitrite, in the growth medium. Thus, ADI may directly interfere with the mucosal immune system and thus facilitate Giardia infection in the small intestine.

Fig. 5
Secreted nitric oxide levels measured by colorimetric substrate conversion of total nitrate in the growth medium of differentiated Caco-2 epithelial cells after stimulation with cytokines (+) in the absence or presence of recombinant G. lamblia ADI. The ...

4. Discussion

The pathogenesis of intestinal infections with Giardia lamblia remains elusive [4]. This protozoan parasite is not invasive and no conventional toxin has been identified [1]. Any mechanism that supports its ability to remain in the host small intestine or to resist host defenses may be viewed as a virulence factor. Here we present data showing release of three Giardia enzymes, previously only characterized as being involved in the intracellular energy metabolism of Giardia [28].

Several previous studies have searched for secreted proteins correlated to disease and colonization by G. lamblia. The VSPs that cover the entire trophozoite surface [30], were originally identified as spontaneous excretory–secretory proteins [31,32]. A 58 kDa non-VSP protein, reported to localize to the surface of G. lamblia P-1 strain [5], was found in the culture supernatant of trophozoites incubated in serum-free medium [5,20]. However, the published partial amino-acid sequence from the 58 kDa protein (ADFVPQVST) does not resemble any sequence in the Giardia WB genome (data not shown), and it only partially fits with the contaminating protein bovine serum albumin (BSA). It is noteworthy that we have seen that full-length BSA and several different sized proteolytic fragments of BSA bind to the trophozoite surface strongly and are slowly released during incubation in serum-free medium (data not shown). It is possible that these fragments interfered with the identification of the 58 kDa protein. Moreover, several unknown proteins with masses ranging from 15 to 225 kDa as well as un-identified proteins with cysteine-type protease activity have been reported in supernatants of Giardia-host cell co-cultures [15,33,34].

This is the first time giardial ADI, OCT and enolase have been shown to be present in Giardia culture supernatants. These typically cytoplasmic proteins have been found to be surface-exposed or secreted in other organisms. We were not able to detect any signal sequences in the proteins but now there are many examples of secreted proteins in eukaryotes without signal sequences [35]. Enolase has been implicated in the pathogenesis of other microbes colonizing mucosal surfaces [29,36]; prokaryotic Streptococcus agalactiae, S. sobrinus, S. pyogenes and eukaryotic Candida albicans α-enolases are secreted [36,37], while other enolases are surface-associated [29,38,39]. Some of these enolases are immunodominant [29,36], as is Giardia enolase [24,25]. However, we were not able to show any specific activity of secreted Giardia enolase and its direct role, if any, during host–parasite interactions remains to be determined.

In addition to enolase we have demonstrated host-induced secretion of OCT and more extensively ADI, two Giardia enzymes of the arginine metabolic pathway. This is an unusual, bacterial-like pathway, not present in human host cells. Trophozoites use ADI and OCT to actively metabolize arginine for energy, thus depleting arginine from the growth medium. Arginine depletion is known to induce apoptosis in human cell lines [40] and human giardiasis patients show an increased rate of apoptosis of intestinal epithelial cells [23]. This has been suggested to be a major disease mechanism [22]. Further experiments will show if ADI is involved in this induction of apoptosis. Prior to inducing apoptosis, this depletion reduces the ability of IECs to produce NO, an anti-microbial innate defense molecule. Our previous studies implicate NO toxicity to Giardia because in vitro NO donors inhibit giardial growth but not viability [19]. NO also inhibits both encystation and excystation of Giardia and could thus interfere with parasite transmission [19]. Many intracellular pathogens are killed by NO, but the role of NO in controlling infections of extracellular pathogens is not well established [19]. Interestingly, the NO levels in intestinal epithelial cells have also been shown to be important in the regulation of adsorption/secretion of water [41], suggesting that it could be associated with symptoms of giardiasis. The secreted OCT and ADI of Giardia might reduce the levels of intestinal arginine further and lower the NO production by IECs. In support of this hypothesis, recombinant mycoplasmal ADI reduced NO production in human cells [18,42]. When we studied the effect of recombinant giardial ADI on IEC in vitro we also observed a decrease of NO production from the intestinal epithelial cells. Secretion of arginine metabolizing enzymes could thus be a general mechanism used by pathogenic microbes at mucosal surfaces. Therefore, arginine consumption and NO reduction define a novel cross talk between Giardia as a non-invasive pathogen and the host intestinal epithelium.

Our present and earlier results confirm that Giardia lamblia is involved in an extensive cross talk with the host intestinal epithelial cells. Immunomodulating proteins produced by microbes are important virulence factors since they are crucial for the survival of the microorganisms in the host. Our results show host-cell stimulated secretion of giardial proteins, which have the potential to be involved in the immunological events during initiation of infection and subsequent survival of Giardia in the host. We have also recently shown that the host response towards Giardia of differentiated Caco-2 cells is partly mediated by soluble factor(s) [8]. The secreted proteins we identify here are immunodominant in human infections [29,36] and experimental mouse infections [13]. The targeting of secreted giardial proteins for immunoneutralization may thus represent a useful therapeutic strategy, aimed at treating or preventing Giardia infections. Further experiments will show what roles these proteins may play during Giardia–host cell interactions.


SS, ER, DP and MW were supported by grants from the Swedish Natural Science Research Council, the Swedish Medical Research Council and Karolinska Institutet. BD, DR and FG were supported by NIH grants DK35108, RR17030, AI42488, GM61896-04 and AI051687, and LE by DK35108 and RR17030. ABH was supported by grants 3100A0-100270 and 3100A0-112327 of the Swiss National Science Foundation.


intestinal epithelial cell
arginine deiminase
ornithine carbamoyl transferase
nitric oxide


1. Morrison HG, McArthur AG, Gillin FD, et al. Genomic minimalism in the early diverging intestinal parasite Giardia lamblia. Science. 2007;317:1921–6. [PubMed]
2. Adam RD. Biology of Giardia lamblia. Clin Microbiol Rev. 2001;14:447–75. [PMC free article] [PubMed]
3. Farthing MJ. The molecular pathogenesis of giardiasis. J Pediatr Gastroenterol Nutr. 1997;24:79–88. [PubMed]
4. Roxstrom-Lindquist K, Palm D, Reiner D, et al. Giardia immunity—an update. Trends Parasitol. 2006;22:26–31. [PubMed]
5. Kaur H, Ghosh S, Samra H, et al. Identification and characterization of an excretory–secretory product from Giardia lamblia. Parasitology. 2001;123:347–56. [PubMed]
6. Meyer EA, Radulescu S. Giardia and giardiasis. Adv Parasitol. 1979;17:1–47. [PubMed]
7. Katelaris PH, Naeem A, Farthing MJ. Activity of metronidazole, azithromycin and three benzimidazoles on Giardia lamblia growth and attachment to a human intestinal cell line. Aliment Pharmacol Ther. 1994;8:187–92. [PubMed]
8. Roxström-Lindquist K, Ringqvist E, Palm D, Svärd S. Giardia lamblia-induced changes in gene expression in differentiated Caco-2 human intestinal epithelial cells. Infect Immun. 2005;73:8204–8. [PMC free article] [PubMed]
9. Müller J, Ruhle G, Müller N, et al. In vitro effects of thiazolides on Giardia lamblia WB clone C6 cultured axenically and in coculture with Caco2 cells. Antimicrob Agents Chemother. 2006;50:162–70. [PMC free article] [PubMed]
10. Jung HC, Eckmann L, Yang SK, et al. A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest. 1995;95:55–65. [PMC free article] [PubMed]
11. Perez PF, Minnard J, Rouvet M, et al. Inhibition of Giardia intestinalis by extra-cellular factors from Lactobacilli: an in vitro study. Appl Environ Microbiol. 2001;67:5037–42. [PMC free article] [PubMed]
12. Li E, Zhao A, Shea-Donohue T, Singer SM. Mast cell-mediated changes in smooth muscle contractility during mouse giardiasis. Infect Immun. 2007;75:4514–8. [PMC free article] [PubMed]
13. Davids BJ, Palm JED, Housley MP, et al. Polymeric immunoglobulin receptor in intestinal immune defense against the lumen-dwelling protozoan parasite Giardia. J Immunol. 2006;177:6281–90. [PubMed]
14. Panaro MA, Cianciulli A, Mitolo V, et al. Caspase-dependent apoptosis of the HCT-8 epithelial cell line induced by the parasite Giardia intestinalis. FEMS Immunol Med Microbiol. 2007;51:302–9. [PubMed]
15. Rodriguez-Fuentes GB, Cedillo-Rivera R, Fonseca-Linan R, et al. Giardia duodenalis: analysis of secreted proteases upon trophozoite–epithelial cell interaction in vitro. Mem Inst Oswaldo Cruz. 2006;101:693–6. [PubMed]
16. Teoh DA, Kamieniecki D, Pang G, Buret AG. Giardia lamblia rearranges F-actin and alpha-actinin in human colonic and duodenal monolayers and reduces transepithelial electrical resistance. J Parasitol. 2000;86:800–6. [PubMed]
17. Scott KG, Meddings JB, Kirk DR, et al. Intestinal infection with Giardia spp. reduces epithelial barrier function in a myosin light chain kinase-dependent fashion. Gastroenterology. 2002;123:1179–90. [PubMed]
18. Dillon BJ, Holtsberg FW, Ensor CM, et al. Biochemical characterization of the arginine degrading enzymes arginase and arginine deiminase and their effect on nitric oxide production. Med Sci Monit. 2002;8:BR248–53. [PubMed]
19. Eckmann L, Laurent F, Langford TD, et al. Nitric oxide production by human intestinal epithelial cells and competition for arginine as potential determinants of host defense against the lumen-dwelling pathogen Giardia lamblia. J Immunol. 2000;164:1478–87. [PubMed]
20. Shant J, Bhattacharyya S, Ghosh S, et al. A potentially important excretory–secretory product of Giardia lamblia. Exp Parasitol. 2002;102:178–86. [PubMed]
21. Chin AC, Teoh DA, Scott KG, et al. Strain-dependent induction of enterocyte apoptosis by Giardia lamblia disrupts epithelial barrier function in a caspase-3-dependent manner. Infect Immun. 2002;70:3673–80. [PMC free article] [PubMed]
22. Buret AG. Mechanisms of epithelial dysfunction in giardiasis. Gut. 2007;56:316–7. [PMC free article] [PubMed]
23. Troeger H, Epple HJ, Schneider T, et al. Effect of chronic Giardia lamblia infection on epithelial transport and barrier function in human duodenum. Gut. 2007;56:328–35. [PMC free article] [PubMed]
24. Palm JE, Weiland ME, Griffiths WJ, et al. Identification of immunoreactive proteins during acute human giardiasis. J Infect Dis. 2003;187:1849–59. [PubMed]
25. Tellez A, Palm D, Weiland M, et al. Secretory antibodies against Giardia intestinalis in lactating Nicaraguan women. Parasite Immunol. 2005;27:163–9. [PubMed]
26. Marti M, Regos A, Li Y, et al. An ancestral secretory apparatus in the protozoan parasite Giardia intestinalis. J Biol Chem. 2003;278:24837–48. [PubMed]
27. Knodler LA, Sekyere EO, Stewart TS, et al. Cloning and expression of a prokaryotic enzyme, arginine deiminase, from a primitive eukaryote Giardia intestinalis. J Biol Chem. 1998;273:4470–7. [PubMed]
28. Brown DM, Upcroft JA, Edwards MR, Upcroft P. Anaerobic bacterial metabolism in the ancient eukaryote Giardia duodenalis. Int J Parasitol. 1998;28:149–64. [PubMed]
29. Pancholi V. Multifunctional alpha-enolase: its role in diseases. Cell Mol Life Sci. 2001;58:902–20. [PubMed]
30. Nash TE. Surface antigenic variation in Giardia lamblia. Mol Microbiol. 2002;45:585–90. [PubMed]
31. Nash TE, Gillin FD, Smith PD. Excretory–secretory products of Giardia lamblia. J Immunol. 1983;131:2004–10. [PubMed]
32. Nash TE, Keister DB. Differences in excretory–secretory products and surface antigens among 19 isolates of Giardia. J Infect Dis. 1985;152:1166–71. [PubMed]
33. Jimenez JC, Uzcanga G, Zambrano A, et al. Identification and partial characterization of excretory/secretory products with proteolytic activity in Giardia intestinalis. J Parasitol. 2000;86:859–62. [PubMed]
34. Jimenez JC, Fontaine J, Grzych JM, et al. Systemic and mucosal responses to oral administration of excretory and secretory antigens from Giardia intestinalis. Clin Diagn Lab Immunol. 2004;11:152–60. [PMC free article] [PubMed]
35. Nickel W. Unconventional secretory routes: direct protein export across the plasma membrane of mammalian cells. Traffic. 2005;6:607–14. [PubMed]
36. Sundstrom P, Aliaga GR. A subset of proteins found in culture supernatants of Candida albicans includes the abundant, immunodominant, glycolytic enzyme enolase. J Infect Dis. 1994;169:452–6. [PubMed]
37. Fluegge K, Schweier O, Schiltz E, et al. Identification and immunoreactivity of proteins released from Streptococcus agalactiae. Eur J Clin Microbiol Infect Dis. 2004;23:818–24. [PubMed]
38. Pancholi V, Fischetti VA. Alpha-enolase, a novel strong plasmin(ogen) binding protein on the surface of pathogenic streptococci. J Biol Chem. 1998;273:14503–15. [PubMed]
39. Ge J, Catt DM, Gregory RL. Streptococcus mutans surface alpha-enolase binds salivary mucin MG2 and human plasminogen. Infect Immun. 2004;72:6748–52. [PMC free article] [PubMed]
40. Philip R, Campbell E, Wheatley DN. Arginine deprivation, growth inhibition and tumour cell death. Part 2. Enzymatic degradation of arginine in normal and malignant cell cultures. Br J Cancer. 2003;88:613–23. [PMC free article] [PubMed]
41. Kukuruzovic R, Robins-Browne RM, Anstey NM, Brewster DR. Enteric pathogens, intestinal permeability and nitric oxide production in acute gastroenteritis. Pediatr Infect Dis J. 2002;21:730–9. [PubMed]
42. Noh EJ, Kang SW, Shin YJ, et al. Characterization of mycoplasma arginine deiminase expressed in E. coli and its inhibitory regulation of nitric oxide synthesis. Mol Cells. 2002;13:137–43. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...