|Home | About | Journals | Submit | Contact Us | Français|
Glycoconjugates expressed on gastric mucosa play a crucial role in host–pathogen interactions. The FUT2 enzyme catalyzes the addition of terminal α(1,2)fucose residues, producing the H type 1 structure expressed on the surface of epithelial cells and in mucosal secretions of secretor individuals. Inactivating mutations in the human FUT2 gene are associated with reduced susceptibility to Helicobacter pylori infection. H. pylori infects over half the world's population and causes diverse gastric lesions, from gastritis to gastric cancer. H. pylori adhesion constitutes a crucial step in the establishment of a successful infection. The BabA adhesin binds the Leb and H type 1 structures expressed on gastric mucins, while SabA binds to sialylated carbohydrates mediating the adherence to inflamed gastric mucosa. In this study, we have used an animal model of nonsecretors, Fut2-null mice, to characterize the glycosylation profile and evaluate the effect of the observed glycan expression modifications in the process of H. pylori adhesion. We have demonstrated expression of terminal difucosylated glycan structures in C57Bl/6 mice gastric mucosa and that Fut2-null mice showed marked alteration in gastric mucosa glycosylation, characterized by diminished expression of α(1,2)fucosylated structures as indicated by lectin and antibody staining and further confirmed by mass spectrometry analysis. This altered glycosylation profile was further confirmed by the absence of Fucα(1,2)-dependent binding of calicivirus virus-like particles. Finally, using a panel of H. pylori strains, with different adhesin expression profiles, we have demonstated an impairment of BabA-dependent adhesion of H. pylori to Fut2-null mice gastric mucosa, whereas SabA-mediated binding was not affected.
Carbohydrate structures expressed on cell surfaces by glycoproteins, glycolipids, and proteoglycans have been implicated in a wide variety of biological processes such as inflammation, host–pathogen interactions, and tumor metastasis (Varki 1993; Hollingsworth and Swanson 2004). The biosynthesis and modification of protein and lipid glycans are mediated by the concerted action of different glycosyltransferases, including sialyltransferases and fucosyltransferases. In humans, the α(1,2)fucosyltransferase activities are encoded by two closely linked genes corresponding to the H blood group locus (FUT1) and the Secretor locus (FUT2). The FUT2 enzyme catalyzes the addition of terminal α(1,2)fucose residues, producing the H type 1 carbohydrate present on the surface of epithelial cells and in mucosal secretions of secretor individuals (Mollicone et al. 1985). The secretor status depends on the FUT2 enzyme. Inactivating mutations in the FUT2 gene, responsible for the nonsecretor phenotype which affect 20% of human population, have been associated with reduced susceptibility to infections by Norwalk virus (Lindesmith et al. 2003), Campylobacter jejuni (Ruiz-Palacios et al. 2003), and Helicobacter pylori (Ikehara et al. 2001).
H. pylori is a Gram-negative bacterium, specialized in the colonization of human stomach, that infects more than half the world's population (Atherton 2006). The infection is associated with the increased risk to develop gastric lesions, from gastritis to gastric adenocarcinoma (Correa and Houghton 2007). The establishment of a successful infection is dependent on H. pylori tight adherence to the mucous epithelial cells and the mucus layer lining the gastric epithelium (Blaser 1993). The H. pylori glycan receptors include fucosylated ABO blood group antigens (Boren et al. 1993; Aspholm-Hurtig et al. 2004), glycans with charged groups such as sialic acid (Mahdavi et al. 2002) or sulfate (Guzman-Murillo et al. 2001) and neolacto core chains (Miller-Podraza et al. 2009). The blood group antigen-binding adhesin (BabA) was shown to bind the Lewis b (Fucα2Galβ3[Fucα4]GlcNAc-R) and H type 1 (Fucα2Galβ3GlcNAc-R) histo-blood group carbohydrate structures expressed in gastric epithelium and mucins (Ilver et al. 1998). Later, a sialic-acid binding adhesin (SabA) that recognizes sialylated carbohydrates was identified (Mahdavi et al. 2002). SabA mediates the adherence of H. pylori to inflamed gastric mucosa by binding carbohydrate structures such as Sialyl-Lewis x (NeuAcα3Galβ4[Fucα3]GlcNAc-R) and Sialyl-Lewis a (NeuAcα3Galβ3[Fucα4]GlcNAc-R) (Mahdavi et al. 2002; Aspholm et al. 2006).
Mucin-type O-gycans, including those described above, play a crucial role in the pathogenesis process of H. pylori infection (Kobayashi et al. 2009). Therefore, it is expected that the diversity of carbohydrate expression on gastric mucosa observed among individuals may confer distinct susceptibilities to H. pylori colonization and infection. In humans, this diversity can be atributed to several genetic polymorphisms in glycosyltransferase genes together with host immune response to pathogens. For this reason, genetically modified mice are an important tool to study the role of a specific glycosyltransferase in the biosynthesis of glycans involved in bacterial adhesion.
Domino et al. have developed a homozygous Fut2-null mouse model, by replacing Fut2 with the bacterial reporter gene lacZ (Domino, Zhang, Gillespie, et al. 2001). The mouse Fut2 gene is the ortholog of the human secretor gene and it has been shown that the FUT2 enzyme is responsible for α(1,2)fucosylation in the mouse gastrointestinal tract (Domino, Zhang, Lowe 2001; Hurd et al. 2005).
In this study, we have evaluated alterations in the glycosylation profile of Fut2-null mice gastric mucosa in comparison with the wild-type mice, using lectins, a panel of monoclonal antibodies for Lewis antigens, and mass spectrometry analysis. We further analyzed whether these modifications would have a role in H. pylori adhesion and observed that H. pylori strains expressing a functional BabA adhesin showed decreased adhesion to Fut2-null mice gastric mucosa. This observation was further supported by the fact that this decrease was not observed when testing strains that only express the sialic-acid binding adhesin SabA, showing that this impaired adhesion could be strictly attributed to reduced expression of BabA ligands in Fut2-null mice gastric mucosa.
No histological differences were observed when comparing the hematoxylin and eosin-stained gastric mucosa of wild-type mice with the Fut2-null mice. The PAS staining reveals the presence of neutral mucins in the superficial mucous cells of the foveolar epithelium and in most cells of pyloric glands both in wild-type and Fut2-null mice (Figure (Figure1).1). The superficial mucous cells of the foveolar epithelium and most cells from the antrum (distal) pyloric glands also stained positive for sialomucins, as indicated by the alcian blue staining (Figure (Figure1).1). Sialomucins were also detected in glands of the proximal stomach.
The major difference detected was a complete loss of binding of the α(1-2)fucose-specific lectin UEA-I to Fut2-null mice (Figure (Figure2,2, Table TableI).I). No major alterations in gastric mucosa sialylation were observed. The binding of the α2-6-specific SNA lectin was similar in both types of mice (Figure (Figure2,2, Table TableI).I). Staining with MAL-I lectin, which recognizes the products of α2,3 sialyltransferases, showed a minor increase in Fut2-null mice when comparing with the wild-type gastric mucosa (Figure (Figure2,2, Table TableII).
Using a panel of different monoclonal antibodies (mAbs) specific for Lewis antigens (Table (TableII),II), it was possible to observe different patterns of staining between wild-type and Fut2-null mice. Lea antibodies, Ca3F4 and 7LE, strongly stained the surface mucous cells of Fut2-null mice but only weakly stained a few cells in wild-type mice (Figure (Figure3A3A and B). In contrast, BG6 and 2.25LE, Leb recognizing antibodies, bound to superficial mucous cells of the foveolar epithelium of wild-type mice but not to the gastric mucosa of Fut2-null mice (Figure (Figure3C–F).3C–F). The pattern of staining of the 2.25LE mAb was restricted to some cells and less intense when compared with BG6 staining. Binding of the Lea (Ca3F4) and Leb (BG6) mAbs to gastric mucosa was specifically blocked by preincubation of the mAb with the corresponding soluble purified Lea or Leb oligosaccharide (supplementary Figure S1). No staining was observed with the CA19.9, anti sialyl-Lea, antibody (Figure (Figure3G3G and H).
Wild-type and Fut2-null mice gastric mucosa were stained in a similar manner with SH1 mAb. Lex expression was observed both on the surface mucous cells and in most of the neck cells and pyloric glands cells (Figure (Figure3I3I and J). The intensity and extent of staining of AH6 mAb, anti Ley, was higher in wild-type than in Fut2-null mice. In null mice gastric mucosa, the staining of AH6 was restricted to mucous cells of the foveolar epithelium (Figure (Figure3K3K and L). No major differences were observed between wild-type and Fut2-null mice regarding the staining of KM93, specific for sialyl-Lex, and FH6, that recognizes (dimeric)sialyl-Lex. Both antibodies showed reactivity with the surface mucous cells, as well as few cells from the antrum glands of both wild-type and Fut2-null mice (Figure (Figure33M–P).
Murine gastric mucosal scrapings were prepared for glycomic screening according to the methodology described in Material and methods. In summary, O-glycans were released from tryptic digests of mucosal homogenates and were permethylated prior to mass spectrometric analysis. O-Glycomic profiles of samples from both wild-type and Fut2-null mice were acquired by MALDI-TOF, and components of interest were sequenced by collision-induced tandem mass spectrometry using both MALDI and Electrospray ionization methods.
Representative MS and MS/MS data from these experiments are shown in Figure Figure4.4. Glycan structures were assigned based on their mass value, the fragment ion data, and knowledge of the biosynthetic pathways of mucin-type O-glycosylation.
The gastric mucosal scraping from wild-type mice shows O-glycans with compositions up to 12 monosaccharides (Figure (Figure4A).4A). The most abundant types of O-glycan are core 2 structures, many of which are branched on their core 1 arm, as noted in an earlier study of murine stomach glycans (Stone et al. 2009). The terminal epitopes are dominated by blood group H type 2-antigens (Fucα2Galß4GlcNAc), for instance m/z 2404. However, some of the larger glycans, for example m/z 2578, carry difucosylated antenna. The MS/MS data for this component gave evidence for Ley, by showing that one antenna is difucosylated and that there was a convincing signal for the elimination of fucose from the molecular ion, confirming that a portion of the fucose in the glycan is attached to position 3 of GlcNAc (data not shown).
As expected, the MALDI profiles of gastric mucosal scrapings from Fut2-null mice, when compared with wild type, showed a major loss of glycans carrying the H-epitope (Figure (Figure4B).4B). However, it is possible to observe minor fucosylated terminal galactose, as demonstrated by m/z 708 (Figure (Figure4B),4B), which on MS/MS analysis gave the same fragment ions (the major ions are m/z 298, 433 and 520, data not shown) as the H-core 1 sequence found in the wild type, indicating that a residual α(1,2)fucosyltransferase activity remains in the stomach of the null mice.
Both wild-type and Fut2-null samples gave a MALDI molecular ion at m/z 1157 whose composition is compatible with putative Lewis structures. In order to establish whether a Lewis structure contributes to this molecular ion, MS/MS experiments were carried out on the corresponding doubly charged ion at m/z 590.4 in the electrospray spectra. These experiments revealed that the dominant wild-type glycans carry the H-epitope on either the core 1 or the core 2 arm (Figure (Figure4C).4C). In the latter case, the fragment ions are consistent with the H-antigen being of the type 2 sequence in which the Gal is attached to the 4-position of GlcNAc. However, we cannot rule out minor amounts of the type 1 sequence. Also, because of the dominant H-containing structures it is not possible to determine whether any Lewis structure is contributing to m/z 1157 in the wild type. In contrast, the Fut2-null mice showed firm evidence for the Lex structure, despite the relevant glycan (m/z 1157 in the MALDI data) being only a small fraction of a percent of the total glycan mixture (Figure (Figure4B).4B). Because of the very low abundance of this component, contaminant signals are prominent below m/z 500 in the electrospray MS/MS spectrum (Figure (Figure4D).4D). Fortunately, however, the region of the MS/MS spectrum which contains structurally useful fragment ions (above m/z 600) is relatively free from contaminating ions. Especially relevant are the signals at m/z 951 and m/z 715 (Figure (Figure4D)4D) because they correspond to elimination of fucose and separate eliminations of fucose plus hexose, respectively. As shown on the cartoon annotations, these signals can only be derived from a Lex structure (Figure (Figure4D).4D). Nevertheless, some of the H-structure is also present because m/z 747 can only be derived from a glycan which carries the fucose attached to galactose because this ion is produced by elimination of fucose plus hexose via a single cleavage event (see cartoon annotations).
Since calicivirus virus-like particles (VLPs) have been shown to bind gastroduodenal epithelial cells from secretor individuals but not to cells from nonsecretors, deficient in the FUT2 gene, we further evaluated VLPs binding capacity to Fut2-null mice gastric mucosa (Marionneau et al. 2002; Lindesmith et al. 2003).
As shown in Figure Figure5,5, DVLP (Dijon strain VLP) and NVLP (Norwalk strain VLP) bound the surface mucous cells of wild-type gastric mucosa but no binding was observed to the gastric mucosa of Fut2-null mice. Negative controls were performed by replacing the incubation with VLPs by PBS and no nonspecific binding of the VLPs recognizing antibodies could be detected.
H. pylori strains 17875/Leb, 17875babA1A2 mutant, and J99 have been previously characterized regarding BabA and SabA expression (Ilver et al. 1998; Mahdavi et al. 2002; Aspholm et al. 2006). As shown in Figure Figure6A,6A, 17875/Leb and J99 strains express both BabA and SabA adhesins while the 17875babA1A2 mutant strain expresses SabA adhesin only. It has been demonstrated that besides expressing both BabA and SabA adhesins, the 17875/Leb strain is unable to bind to sialylated antigens (Mahdavi et al. 2002).
To better understand the importance of FUT2 enzymatic activity in the biosynthesis of potential H. pylori ligands, we evaluated the binding capacity of these strains to the gastric mucosa of Fut2-null mice in comparison with wild-type mice. As shown in Figure Figure6B6B and C, 17875/Leb strain bound the surface mucous cells of wild-type mice but no adhesion was observed to the gastric mucosa of Fut2-null mice. The SabA competent 17875babA1A2 strain adhered equally well to the surface mucous cells of wild-type and Fut2-null mice and, in addition, quantification revealed that an increased number of bacteria bound the glandular region of Fut2-null mice gastric mucosa when compared to the wild type. The J99 strain, that expresses both BabA and SabA adhesins (Figure (Figure6A),6A), showed decreased adhesion to superficial mucous cells of the foveolar region of Fut2-null mice when compared with the wild type.
Besides these largely characterized model H. pylori strains, we further evaluated the adhesion of a panel of strains from clinical isolates. These strains were characterized for BabA and SabA expression as shown in Figure Figure7A.7A. Quantification of adhered bacteria showed that adhesion of the BabA-positive strains, 094UK 127UK and BO265, was decreased in the Fut2-null mice gastric mucosa when compared to the wild-type mice (Figure (Figure7B).7B). H. pylori 101UK strain was an exception because although being considered BabA positive, binding was not affected by the Fut2 deficiency (Figure (Figure7B).7B). The exclusively SabA-positive strains, 111UK and 123UK, adhered equally to the gastric mucosa of both wild-type and Fut2-null mice. As expected, the levels of adhesion of strains 097UK and 131UK that did not express any of the adhesins were very low.
Glycoconjugates expressed on the gastric mucosa play a crucial role in the establishment of a close contact between H. pylori and mucous epithelial cells favoring bacterial adhesion and colonization of gastric mucosa (Ilver et al. 1998; Mahdavi et al. 2002; Kobayashi et al. 2009). In humans, the FUT2 enzyme is responsible for the biosynthesis of the lacto-series type 1 (A1, B1, H1, Leb) and type 2 (A2, B2, H2, Ley) in the gastric superficial epithelium (Mollicone et al. 1985). It has been demonstrated by our group and others that infection by H. pylori is influenced by the secretor phenotype, which is defined by the presence of an active FUT2 (secretor) enzyme (Ikehara et al. 2001; Lee et al. 2006; Azevedo et al. 2008). In this study, we have used an animal model of nonsecretors, Fut2-null mice, to characterize the glycosylation profile of the wild-type and null mice gastric mucosa and to evaluate the effect of the observed glycan expression modifications in the process of H. pylori adhesion to gastric epithelial cells.
First we characterized the alterations in the glycosylation profile of Fut2-null mice gastric mucosa compared to wild-type mice. The surface epithelium of wild-type gastric mucosa stained positive with the α(1-2)fucose-specific lectin UEA-I, whereas complete loss of staining in the Fut2-null mice was observed (Figure (Figure2).2). This result is in accordance with the absence of UEA-I staining in the uterine epithelia of Fut2-null mice, described by Domino, Zhang, Gillespie, et al. (2001).
Unexpectedly, we observed that both BG6 (T218) and 2.25LE, Leb recognizing antibodies, bound to the superficial mucous cells of wild-type gastric mucosa but not to Fut2-null mice (Figure (Figure3C–F).3C–F). This staining was not expected since mice lack an ortholog of the human FUT3 gene and no α1,4fucosyltranferase enzyme has ever been described in mice (Falk et al. 1995; Nairn et al. 2008). The mouse gene orthologous to the ancestral gene for human FUT3, FUT5, and FUT6 seems to be a pseudogene (Gersten et al. 1995; Smith et al. 1996; Costache et al. 1997).
Moreover Falk et al. (1995) have showed de novo expression of Leb, detected with a monoclonal antibody 96 FR2.10, by expressing a human α1,3/4fucosyltranferase enzyme in the gastric pit cell lineage of FVB/N transgenic mice (Falk et al. 1995). Recently, characterization of glycosphingolipid structures from these FVB/N transgenic mice identified a Leb-like glycosphingolipid (Fucα2Galβ3[Fucα4]GalNAcβ4Galβ4Glcβ1Cer) that was recognized by 96 FR2.10 but not by a BG6 (T218) antibody (Fagerberg et al. 2009). The same study indicated that the Leb epitope present in glycoproteins from the transgenic Leb mouse stomach can differ from the “traditional” Leb epitope because it is not recognized by the BG6 (T218) antibody (Fagerberg et al. 2009). One possible interpretation for the staining of wild-type mice gastric mucosa we observed with the BG6 (T218) and 2.25LE, Leb recognizing antibodies, is antibody cross-reactivity since it has been previously demonstrated that BG6 (T218) cross-reacted significantly with Lec, Ley, and blood group H1 (Monteiro et al. 1998; Manimala et al. 2007), whereas the 2.25LE antibody displayed weak cross-reactivity with Lea (Good et al. 1992; Amano et al. 1997; Manimala et al. 2007), although for the latter that would not explain the loss of binding to Fut2-null mice tissue sections. Noteworthy and also unexpected was the increased binding of the Lea antibodies Ca3F4 and 7LE to the surface mucous cells of Fut2-null mice, suggesting that lack of α(1,2)fucosyltransferase activity would lead to accumulation of Lea structures (Figure (Figure3A3A and B). Once again, possible antibody cross-reactivity may be considered, even though the 7LE antibody displayed excellent specificity within the context of the carbohydrate microarray performed by Manimala et al. (2007). Taken together, anti-Lea and anti-Leb antibody binding patterns suggest expression of Lewis type 1 structures in mice gastric mucosa corroborated by antibody inhibition using specific synthetic saccharides. However, mass spectrometry analysis of the O-glycans of mucosal scrapings could not detect these structures indicating that, if present, the levels are extremely low. Further structural studies will be performed using gastric mucosa samples enriched by using anti-Lea and anti-Leb binding proteins.
Regarding type 2 structures, the pattern of Lex expression detected with the monoclonal antibody SH1 was similar in both wild-type and null mice (Figure (Figure3I3I and J). However, Lex could only be detected in Fut2-null mice, for example m/z 1157 (Figure (Figure4D).4D). This could be explained by the fact that, contrary to the wild type, this sample is not dominated by H-epitopes. Therefore, it is possible to turn up the gain for detecting glycans bearing Lex which are originally low in abundance. Decreased Ley expression was detected in Fut2-null mice, both by immunohistochemistry using an AH6 monoclonal antibody (Figure (Figure3K3K and L) and mass spectrometry analysis (Figure (Figure4A),4A), suggesting that the FUT2 enzyme is participating in the fucosylation of both type 1 and type 2 structures.
We observed that both wild-type and Fut2-null mice stained positively with KM93, directed to sialyl-Lex, and FH6 that recognizes (dimeric)sialyl-Lex. Since the FH6 monoclonal antibody has been shown to recognize the N-acetylneuraminyl Lex oligosaccharide but not its N-glycolyl form (Mitoma et al. 2009), our results suggest that N-acetylneuraminyl Lex is being detected. The presence of sialyl-Lex was also described in the gastric mucosa of FVB/N Leb transgenic mice (Mahdavi et al. 2002).
The MALDI profile of wild-type mice gastric mucosal scrapings identified the Fuc-Gal-epitope (blood group H-type epitope) as the most abundant monofucosylated glycan. This observation is in accordance with the recently described analysis of O-linked oligosaccharides present in mice cervical mucins that also identified this structure as being highly expressed (Domino et al. 2009). As expected, Fut2-null mice showed major loss of glycans carrying the H-epitope. However, minor terminal fucosylation could be detected in Fut2-null mice gastric mucosa as demonstrated by the molecular ions m/z 708 and fragment ion m/z 747 (Figure (Figure4B4B and D, respectively) and the binding of the AH6 antibody. This α(1,2)fucosylation activity may be attributed to mouse Fut1 since fucosyltransferase activity assays using extracts from COS-7 cells expressing mouse Fut1 demonstrated that this murine enzyme can use both type 1 (Galβ1-3GlcNAc) and type 2 (Galβ1-4GlcNAc) soluble acceptors in vitro (Domino et al. 1997). Northern-blot analysis showed expression of Fut1 transcript in adult mouse stomach tissue, although an in situ hybridization technique did not reveal specific hybridization in stomach cells (Domino et al. 1997).
The different glycan profile of Fut2-null mice gastric mucosa could be further confirmed by the lack of binding of calicivirus VLPs of the Norwalk and Dijon strains. VLPs have been shown to bind H-type 1 structures present on gastroduodenal epithelial cells from secretor individuals but not on cells from nonsecretors (Marionneau et al. 2002; Lindesmith et al. 2003), indicating that Fut2-null mice surface mucous cells mimic nonsecretor gastric epithelial cells.
Binding of H. pylori to the fucosylated blood group antigens Leb and H-type 1 is mediated by the blood group antigen-binding adhesin BabA, an outer membrane protein expressed by most disease causing H. pylori strains (Ilver et al. 1998; Gerhard et al. 1999). After colonization, H. pylori is able to upregulate a set of inflammation-associated genes, including the β3GnT5 enzyme that participates in the synthesis of sLex (Marcos et al. 2008). SabA adhesin binds sLex and also other structurally related sialylated antigens contributing to the persistence of H. pylori infection (Mahdavi et al. 2002).
H. pylori reference strain J99 expresses both BabA and SabA active adhesins with binding affinity for Leb, H1, blood group A-Leb, blood group A-H1, sialyl-Lex, sialyl-di-Lex, sialyl-Lea, 3′-sialyl-lactose, and sialyl-lactosamine (Aspholm-Hurtig et al. 2004; Aspholm et al. 2006). We observed that FITC labeled J99 strain bound to surface mucous cells and glands of both wild-type and Fut2-null mice; however, levels of adhesion to the surface mucous cells were decreased in Fut2-null mice. In order to evaluate if this decrease in adhesion could be attributed to decreased expression of α(1,2)fucosylated glycans recognized by BabA adhesin, we evaluated the adhesion of the 17875/Leb strain, which besides expressing both BabA and SabA adhesins recognizes Leb, H1, blood group A-Leb, blood group A-H1, and blood group B-Leb but does not bind to sialylated glycans (Aspholm-Hurtig et al. 2004; Aspholm et al. 2006). Binding of the 17875/Leb strain was restricted to the surface mucous cells of foveolar epithelium of wild-type mice and completely absent in the null mice. In addition, we have tested a BabA mutant strain 17875babA1A2, with affinity for sialyl-Lex, sialyl-di-Lex, sialyl-Lea, 3′-sialyl-lactose, and sialyl-lactosamine (Aspholm-Hurtig et al. 2004; Aspholm et al. 2006) and observed that this strain bound similarly to wild-type and Fut2-null mice foveolar and surface mucous cells. Interestingly, this strain adhered more to the glandular region of Fut2-null mice in comparison with the wild type. This observation could be explained by the fact that these mice show decreased H-type 2 expression, as further demonstrated by mass spectrometry, so it is possible that they accumulate increased amounts of sialylated-type 2 structures. Similarly, it has been previously reported a reciprocal induction in fucosylation versus sialylation during infection by H. pylori of secretor positive macaques and that this reciprocity was much lower in weak secretor individuals (Lindén et al. 2008).
We further evaluated the adhesion of a larger panel of H. pylori strains and we observed that adhesion of BabA-positive strains, 127UK, BO265, and 094UK, was significantly decreased in the Fut2-null mice gastric mucosa when compared to the wild-type mice (Figure (Figure7B).7B). H. pylori 101UK strain did not seem to be affected by the FUT2 deficiency, suggesting that adhesion of this strain is mostly mediated by SabA, although by Western blot it was possible do detect a faint band corresponding to BabA adhesin (Figure (Figure7A7A and B). No significant differences were observed for SabA-positive strains, 111UK and 123UK. H. pylori strains 097UK and 131UK that did not express any of the adhesins did not adhere to either wild-type or mutant mice gastric mucosa.
In summary, we have demonstrated that Fut2-null mice show marked alteration in gastric mucosa glycosylation patterns, revealed by the absence of UEA-I lectin binding, increased Lea imunostaining, decreased Leb antibodies immunoreactivity, and diminished H-type 2 and Ley expression as detected by mass spectrometry. Furthermore, we have shown that this altered glycosylation profile of the Fut2-null mice gastric mucosa completely abolishes the binding of calicivirus VLPs. Finally, using a panel of H. pylori strains, we have demonstated that whereas SabA-mediated binding was not affected, BabA-dependent adhesion was impaired in the Fut2-null mice. Our results confirm the relevance of BabA adhesin and the alpha2-linked fucose residues in H. pylori binding to host gastric mucosa and set the ground for the use of this model in studies addressing H. pylori adhesion.
Wild-type C57BL/6 and FUT2-null mice (Domino, Zhang, Gillespie, et al. 2001) were obtained from the Consortium for Functional Glycomics (CFG). Mice were reproduced, maintained, and housed at IPATIMUP Animal House at the Medical Faculty of the University of Porto and handled accordingly to the Guidelines for the Care and Use of Laboratory Animals, directive 86/609/EEC. Mice genotyping was performed as previously described, using primers specific for the wild-type and the null alleles (supplementary Figure S2) (Domino, Zhang, Gillespie, et al. 2001).
Stomachs from 6- to 8-week-old wild-type and null mice were harvested and processed for paraffin sections. Every specimen was fixed in 10% formalin and embedded in paraffin wax. Serial sections were cut and used for histochemistry and immunohistochemistry. Tissue sections from all mice were stained with hematoxylin and eosin (H&E) for morphological analysis. Alcian blue (pH 2.5)/periodic acid-Schiff technique was used to identify neutral and sialomucins.
Paraffin sections, after deparaffination, rehydration, and blocking of endogenous peroxidase activity with 3%H2O2 in methanol, were incubated with 10% of BSA in PBS for 30 min at room temperature. Sections were then incubated with one of the biotynilated lectins UEA-I (Ulex europaeus, SIGMA), SNA (Sambucus nigra,Vector Laboratories), or MAL-I (Maackia amurensis lectin I, Vector Laboratories) for 1 h. The slides were subsequently washed in PBS and incubated for 30 min with avidin–biotin complex (Vectastain Elite ABC kit, Burlingame, CA) according to manufacturer's recommendations. Staining was performed with 3,3′-diaminobenzidine tetrahydrochloride (Sigma, St. Louis, MO) containing 0.02% hydrogen peroxide. Slides were counter-stained with Mayer's hematoxylin. Negative controls were performed by replacing lectin with PBS.
Monoclonal antibodies used in this study, their specificity, and their references are listed in Table TableII.II. Immunohistochemical staining was performed similarly to the lectin staining with the following alterations: sections were blocked with normal rabbit serum diluted 1:5 in PBS containing 10% BSA, and then incubation with the monoclonal antibody was performed overnight at 4°C; next, slides were washed in PBS and incubated for 30 min with a biotinylated rabbit anti-mouse antibody (E0354-DakoCytomation, Glostrup, Denmark) diluted 1:200 in PBS containing 5% of BSA prior to incubation with HRP-avidin-biotin complex. Negative controls were performed by replacing primary antibody with PBS.
Scrapings from gastric mucosa of 6 to 8-week-old mice were prepared for glycomic analysis according to methodology that have been modified and optimized from a previously reported methodology (Sutton-Smith et al. 2000; Jang-Lee et al. 2006). In brief, mice gastric mucosa scrapings were homogenized with ice-cold water. This is followed by extraction of proteins and glycoproteins which were then sequentially reduced carboxymethylated and digested with trypsin (Sigma-Aldrich, UK). O-Glycans were released from glycopeptides by reductive elimination using KBH4 in KOH. After purification on Dowex columns (Sigma-Aldrich), the O-glycan samples were permethylated and then further purified with Sep-Pak cartridges. O-Glycans were eluted in aqueous acetonitrile fractions and then lyophilized. Glycans are normally eluted in the 35% and 50% acetonitrile fractions; therefore, only these fractions were subjected to mass spectrometric analysis.
Permethylated samples were dissolved in 10 μL of methanol and 1 μL of the dissolved sample was pre-mixed with 1 μL of matrix (20 mg/mL 2,5-dihydroxybenzoic acid (DHB) in 70% (v/v) aqueous methanol) (Sigma-Aldrich) spotted onto a target plate. MALDI mass spectrometric (MS) and tandem mass spectrometric (MS/MS) data were acquired on a 4800 MALDI-TOF/TOF mass spectrometer (Applied Biosystems, UK) in the reflectron mode. The potential difference between the source acceleration voltage and the collision cell was set to 1 kV and argon was used as collision gas. The 4700 Calibration Standard kit (Applied Biosystems, UK) was used as the external calibrant for the MS mode and [Glu1]fibrinopeptide B human was used as an external calibrant for the MS/MS mode. Tandem mass spectrometric data were also acquired by using a QSTAR Pulsar-I mass spectrometer (Applied Biosystems, UK). Permethylated samples were dissolved in 10 μL of methanol and 1 μL of the dissolved sample was introduced into the electrospray source of the mass spectrometer via a nanoflow electrospray needle. Nitrogen was used as the collision gas with the pressure being maintained at 5.3 × 10−5 Torr and collision energies typically varied from 35 to 90 eV.
VLPs from the norovirus strains Norwalk (NVLPs) and Dijon 171/96 (DVLPs) of the GI.1 and GII.4 Grimsby type, respectively, as well as their respective antibodies were prepared as previously described (Rydell et al. 2009). Paraffin sections, after deparaffination, rehydration, and blocking of endogenous peroxidase activity as described above, were incubated with 3% BSA in PBS for 30 min. Then sequential sections were incubated with either NVLP (Norwalk VLP), 5 μg/mL in PBS for 90 min at room temperature, or DVLP (Dijon VLP) 3 μg/mL in PBS containing 1% BSA overnight at 4°C. After three washes with PBS, sections were incubated with a rabbit polyclonal antibody specific for each virus particle for 1 h at room temperature. Next, samples were incubated with goat anti rabbit-HRP (Santa Cruz Biotechnology, CA) and detection was achieved with 3,3′-diaminobenzidine tetrahydrochloride as described above. Slides were counter-stained with Mayer's hematoxylin. Negative controls were performed by replacing VLPs with PBS.
The H. pylori strains 17875/Leb, 17875babA1::kan babA2::cam, and J99 were obtained from the Department of Medical Biochemistry and Biophysics, Umeå University, Sweden (Mahdavi et al. 2002). The 17875/Leb strain is a spontaneous mutant that binds Leb but does not bind to sialylated antigens (Mahdavi et al. 2002). H. pylori 094UK, 097UK, 101UK, 111UK, 123UK, 127UK, 131UK, and BO265 were obtained from the Department of Molecular Biology, Max-Planck Institut für Infektionsbiology, Berlin, Germany.
Bacteria were grown in Pylori agar (BioMérieux, Marcy l’Étoile, France) at 37°C under microaerobic conditions. For strain 17875babA1::kan babA2::cam, culture media included also 20 mg/L Chloramphenicol (Sigma) and 25 mg/L Kanamycin (Sigma).
Bacteria were collected from plates using 1 mL PBS and centrifuged at 2400 g for 5 min; proteins were extracted with the lysis buffer RIPA containing 1 mM PMSF, 1 mM Na3VO4, and protease inhibitor cocktail (Roche, Germany). Protein concentration was determined using the BCA protein assay kit and 50 μg was loaded in acrylamide gel (stacking 5%/resolving 10%) for electrophoresis. Proteins were transferred to immunoblot polyvinylidene difluoride membranes (Hybond-P PVDF Membrane, Amersham Biosciences) at 50 V for 1 h. Membranes were blocked for 1 h in PBS-Tween containing 5% nonfat dried milk prior to incubation overnight at 4°C with BabA (AK277 1:10000) and SabA (AK278 1:5000) antibodies. Blots were incubated with a HRP-conjugated goat anti-rabbit antibody (Santa Cruz Biotechnology) and finally developed with Immun-StarTM HRP Chemiluminescent kit (Bio-Rad, CA).
Total DNA was extracted from H. pylori and PCR was performed, as previously described with forward primer babA_F1 (5′-ATCGGTGAAGCCGCTCAA-3′) and equimolar amounts of reverse primers babA_R1 (5′-TTACCCC- CGCATTGCG-3′) and babA_R2 (5′-TTACCGCCGC-ATTGCG-3′). The amplified product has 290 bp (Azevedo et al. 2008).
H. pylori was labeled with FITC and concentration adjusted as previously described (Falk et al. 1993). Adhesion assays were performed as previously described (Falk et al. 1993). Briefly, paraffin-embedded mice gastric sections were deparaffinized and rehydrated followed by incubation for 1 h with blocking buffer: 1% BSA in PBS containing 0.05% Tween 20. BSA was previously submitted to Periodate oxidation to destroy competitive carbohydrate receptors for H. pylori binding (Aspholm et al. 2006). The FITC-labeled bacterial suspension was diluted 5-fold in the blocking buffer and 100 μL were used to incubate each section for 2 h at room temperature. Slides were subsequently washed three times with PBS containing 0.05% Tween 20 and stained with DAPI. Evaluation of bacterial binding was estimated by the number of adhered bacteria under 400× magnification. Each value represents the mean of at least five different fields quantified using the ImageJ software, with significance evaluated with the Student's t-test. Whenever referred in the text H. pylori adhesion was evaluated separately in the foveolar and glandular regions of mice stomach. The foveolar region was considered as the superficial region of the gastric mucosa including the surface mucous cells, while the glandular region corresponds to the deeper portion of the gastric mucosa including the mucosal glands.
Fundação para a Ciência e a Tecnologia (FCT PDTC/ CTM65330/2006 and PTDC/SAU-MII64153/2006); financiados no âmbito Programa Operacional Ciência e Inovação 2010 do Quadro Comunitário de Apoio III e comparticipados pelo Fundo Europeu de Desenvolvimento Regional Fundação Luso-Americana para o Desenvolvimento the Biotechnology and Biological Sciences Research Council (B19088 and BBF0083091 to A.D. and S.M.H.); the Région des Pays de la Loire (CIMATH to J.L.P.); University Science Malaysia, Ministry of Higher Education (M.N.I.); FCT (SFRH/BD/36339/2007 to A.M. and SFRH/BD/40563/2007 to J.G.); and Consortium for Functional Glycomics funded by NIGMS-GM62116.
We thank Ulla Mandel from the School of Dentistry, University of Copenhagen, Denmark, and Jacques Bara from the U-673 INSERM, CNRS, Paris, France, for monoclonal antibodies. We thank Mark Achtman from the Max-Planck Institut für Infektionsbiology, Berlin, Germany, for providing the 094UK, 097UK, 101UK, 111UK, 123UK, 127UK, 131UK, and BO265 H. pylori strains.