Antiadhesive property of microalgal polysaccharide extract on the binding of Helicobacter pylori to gastric mucin

Authors


  • Editor: Johannes Kusters

Correspondence: Bow Ho, Department of Microbiology, Yong Loo Lin School of Medicine, National University of Singapore, MD4, 5, Science Drive 2, Singapore 117597. Tel.: +65 6516 3672; fax: +65 6776 6872; e-mail: michob@nus.edu.sg

Abstract

The emergence of antibiotic-resistant Helicobacter pylori is of concern in the treatment of H. pylori-associated gastroduodenal diseases. As the organism was reported to bind gastric mucin, we used porcine gastric mucin as substrate to assess the antiadhesive property of polysaccharides derived from Spirulina (PS), a commercially available microalga, against the binding of H. pylori to gastric mucin. Results show that polysaccharides prevented H. pylori from binding to gastric mucin optimally at pH 2.0, without affecting the viability of either bacteria or gastric epithelial cells, thus favouring its antiadhesive action in a gastric environment. Using ligand overlay analysis, polysaccharide was demonstrated to bind H. pylori alkyl hydroperoxide reductase (AhpC) and urease, which have shown here to possess mucin-binding activity. An in vivo study demonstrated that bacteria load was reduced by >90% in BALB/c mice treated with either Spirulina or polysaccharides. It is thus suggested that polysaccharides may function as a potential antiadhesive agent against H. pylori colonization of gastric mucin.

Introduction

Helicobacter pylori is reported to colonize the stomach of half of the global human population, causing peptic ulcer diseases, gastric MALT lymphoma, and distal gastric cancer (Perez-Perez et al., 2004). Triple therapy (comprising a proton-pump inhibitor combined with clarithromycin and amoxicillin or metronidazole) is the recommended first-line eradication therapy for H. pylori, but the eradication failure rate is as high as 25% and has been shown to be on the rise (Malfertheiner et al., 2002). Eradication failure may result from lack of patient compliance, acquired resistance, acidic gastric pH, high bacterial load, impaired mucosal immunity, and early reinfection. Furthermore, eradication therapy is not employed for the majority of asymptomatic patients infected with H. pylori in order to avoid increasing medical expenditure and hastening the emergence of resistant strains (Malfertheiner et al., 2002). Resistance to clarithromycin is 8–30% (McLoughlin et al., 2004), while that to metronidazole is 15–66% (Lui et al., 2003). There is therefore a need for a nonantibiotic-based approach that is economical, effective and safe in preventing the expression of gastric pathologies among asymptomatic H. pylori-infected patients as well as in preventing the colonization of H. pylori in healthy individuals.

The human gastrointestinal tract is covered with a highly impermeable polymer matrix of a mucus layer that consists of gastric mucin, which comprises high-molecular-mass glycoproteins. The continuous layer of mucus gel acts as a protective barrier for the underlying mucosa. During infection, H. pylori is found mainly in the gastric mucus layer, with some attaching to the epithelial cell surface or in the lamina propria (Shimizu et al., 1996). The secreted gel-forming mucin MUC5AC serves as the primary receptor for H. pylori in the human stomach (Van de Bovenkamp et al., 2003). In order for H. pylori to colonize the underlying gastric epithelium, the bacteria must be able to adhere to and penetrate the mucus layer. Because adhesion to gastric mucin is essentially the initial stage in H. pylori colonization and establishment of pathogenesis, abolishing the adhesion of H. pylori to gastric mucin can serve as a novel preventive strategy. As bacterial adhesion is primarily mediated by the interaction of lectin-like molecules with highly specific carbohydrate structures present on gastric mucin and cell surfaces, carbohydrate-based compounds are ideal candidates for antiadhesives. Compared with bactericidal agents, antiadhesives are less likely to generate resistance, because it is not as compelling a selective pressure to prevent adhesion since adhesion is a nonlethal function in the survival of the bacteria (Bavington & Page, 2005).

Current difficulties and the cost of synthesizing rationally designed high-affinity carbohydrate-based antiadhesives make it cost-effective to exploit naturally occurring polysaccharide-based compounds as inhibitors of bacterial binding. Chlorella and Spirulina are two species of microalgae produced by artificial open-air cultures that have received much attention not only as promising and nutritious food sources but also as functional foods (Otles & Pire, 2001). This study aimed to evaluate the polysaccharides derived from these commercially available dietary microalgae for their antiadhesive properties. In addition, H. pylori proteins with mucin-binding activity were analysed for their role in adhesion to porcine gastric mucin.

Materials and methods

Helicobacter pylori strain and culture

Mouse-adapted H. pylori Sydney strain 1 (SS1) was cultured on chocolate blood agar (blood agar base 2) (Oxoid, Hampshire, UK) supplemented with 5% lysed horse blood (Quad Five, Ryegate, MT) or in brain heart infusion (BHI) broth (Oxoid) supplemented with 0.4% yeast extract (Oxoid) and 10% horse serum (JRH Biosciences, Lenexa, KA). Cultures were incubated for 3 days under microaerophilic conditions (10% CO2 and 95% humidity) at 37°C.

Gastric mucin

Type II crude porcine stomach mucin was purchased from Sigma Chemical Co. (St Louis, MO).

Extraction of Chlorella and Spirulina polysaccharides

Polysaccharides were extracted and partially purified from Chlorella tablets (PC) and Spirulina (PS) according to an organic extraction method as described by Guzman-Murillo and Ascencio (2000).

Competitive and blocking inhibition assay

A modified method as described by Shibata et al. (2003) was used to assess the competitive and blocking capability of the polysaccharides against the binding activity of H. pylori to mucin. For competitive inhibition assay, polysaccharide extracts were incubated together with H. pylori suspended in PBS (107 CFU per well) for 90 min at 37°C on a mucin-coated 96-well plate. For blocking inhibition assay, the immobilized mucin was preincubated with polysaccharide extracts for 90 min, after which excess polysaccharides were rinsed off and the bacterial suspension was added to the wells.

Controls were wells containing PBS in place of polysaccharide extracts and wells containing no bacteria. After removing unbound bacteria and excess polysaccharides, bacteria adhering to mucin were quantified using a urease-based method (Wadström et al., 1997).

Polysaccharide agglutination assay

The assay was carried out according to the modified method as described by Khin et al. (2000). Polysaccharide extract or mucin dissolved in sterile PBS was mixed with an equal volume of H. pylori (1012 CFU mL−1) in PBS. The mixture was constantly agitated for 10 min, and observed for clumping visually and microscopically.

Animal study

Twenty-one BALB/c mice (Animal Resources Centre, Murdoch, Western Australia) were divided into seven groups of three mice each. Two groups of mice were treated intragastrically with 350 μg of PS suspended in PBS at 1- and 2-hour intervals before inoculating with 106 CFU of H. pylori SS1. Similarly, two other groups were treated with 35 mg of Spirulina powder suspended in PBS. The positive control group was fed with H. pylori only. Two remaining groups that were fed with either PS or Spirulina served as the negative controls. The treatment was repeated for each of the seven groups of mice for 2 consecutive days. All mice were sacrificed 2 weeks after the second treatment, and the stomach of each animal was removed for further analysis. Helicobacter pylori that were present in the animal stomachs were isolated, identified and enumerated according to the method described in Liu & Lee (2003). To study the effect of a longer period of treatment with PS or Spirulina before infection with H. pylori, another two groups of three mice each were fed with either PS or Spirulina three times per week for 4 weeks before infection with H. pylori. Mice (n=3 each) fed with either PS or Spirulina served as a negative control, while another group of three mice fed with H. pylori alone served as a positive control. All mice were sacrificed 2 weeks after the 4-week treatment with PS or Spirulina. The stomach of each animal was removed for analysis according to the method described in Liu & Lee (2003). These studies were approved by the NUS Institutional Animal Care and Use Committee (IACUC) and carried out as according to the International Guiding Principles for Animal Research.

Localization of H. pylori mucin-binding proteins

Acid-glycine-extracted proteins (cell-associated proteins), membrane proteins, periplasmic proteins, and whole-cell lysates of H. pylori SS1 were prepared as described in Du & Ho (2003). Helicobacter pylori subcellular fraction proteins and whole-cell lysates were labelled with d-biotin-N-hydroxysuccinimide (Roche Diagnostics, Mannheim, Germany) and incubated with mucin immobilized onto 96-well plates. Instead of bacterial protein, PBS was added to the wells coated with mucin that were designed as blank. The unbound H. pylori proteins were removed by washing with PBS containing 0.05% Tween-20; the biotin-labelled proteins that were bound to the mucin were then detected by horseradish peroxidase-conjugated streptavidin (Roche Diagnostics) and O-phenylenediamine-dichloride (OPD) (Sigma). Absorbance was determined at 495 nm.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)

Protein concentration was estimated by modified Bradford protein assay (Bio-Rad, Hercules, CA). Protein samples were electrophoresed under denaturing conditions according to the method of Laemmli (1970). Samples separated under partially denaturing conditions in loading buffer consisting of Tris-HCl (pH 6.8), glycerol and bromophenol blue without boiling. Protein bands of interest were excised from Coomassie blue-stained gels and subjected to trypsin digestion (Shevchenko et al., 1996), and then desalted and concentrated using Zip-Tip C18 pipette tips (Millipore, Bedford, MA, USA) prior to analysis by Matrix Assisted Lase Desorption Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS).

Ligand overlay analysis

Adjacent hydroxyl groups in sugars of mucin and polysaccharide extracts were oxidized to aldehyde groups and labelled with digoxigenin-3-O-succinyl-ɛ-aminocaproic acid hydrazide (DIG) (Roche Diagnostics). Electrophoresed H. pylori acid glycine extract (AGE) or recombinant proteins were transblotted onto Immobilon membranes (Millipore, Bedford, MA). BSA was used as the negative control. Ligand overlay analysis was carried out as described by Wampler et al. (2004). DIG-labelled mucin or polysaccharide-bound proteins were detected using a Glycan labelling and detection kit (Roche Diagnostics).

Cloning and expression of ahpC

The ahpC gene (HP1563) was cloned into the pET16b vector, overexpressed in Escherichia coli BL21. The recombinant AhpC (rAhpC) protein was purified by affinity chromatography using a nickel-chelating column (GE Healthcare, Bucks, UK).

Statistics

Experimental data are expressed as mean±SD. Statistical analyses were conducted using Student's t-test. P<0.05 is considered as significant.

Results

Antiadhesive property of microalgal polysaccharides

The competitive inhibition assay was carried out to screen compounds for any antiadhesive property of the polysaccharides. Both PS and PC were found to have an inhibitory effect on the adhesion of H. pylori to gastric mucin (Fig. 1). It was found that 35 μg of PS reduced the H. pylori adhering to mucin by about 90% (Fig. 1a). However, more than twice the amount or 80 μg, of PC was required to achieve this effect (Fig. 1b).

Figure 1.

 Competitive and blocking inhibitory effect of microalgal polysaccharide extracts on Helicobacter pylori adhesion to mucin. For the competitive inhibition assay, H. pylori SS1 (106 CFU mL−1) was incubated in the presence of (a) PS and (b) PC (dissolved in PBS, pH 7.0) on 96-well tissue culture plates coated with mucin, after which unbound bacteria and excess polysaccharide were rinsed off. For the blocking inhibition assay, mucin coated onto 96-well tissue culture plates was blocked with (c) PS and (d) PC before interaction with H. pylori. (e) Mucin was blocked with PS dissolved in buffer (pH 2.0 to 7.0). Helicobacter pylori SS1 (106 CFU mL−1) was incubated and allowed to adhere after washing to remove excess polysaccharide. Subsequently, unbound bacteria were rinsed off. For both assays, adhering H. pylori was quantified by modified urease-based assay. The result represents the mean urease activity in triplicate in the absence of polysaccharide (control). Error bars are SD. *P<0.05 compared with the control.

The blocking inhibition assay was used to assess the feasibility of using these polysaccharide extracts in preventing H. pylori from colonizing mucin. Both polysaccharide extracts (PS and PC) inhibited bacterial adhesion to mucin in the blocking inhibition assay (Fig. 1c and d). PS at a concentration of 35 μg inhibited H. pylori adhesion to mucin by more than 99% (Fig. 1c), whereas a similar amount of PC reduced the bacteria binding to mucin only by about 80% (Fig. 1d). It was also noted that a higher concentration of PC did not increase its effectiveness in inhibition (data not shown). In order to determine the effectiveness of PS in an acidic environment, PS was allowed to react with mucin at different levels of pH and it was shown to bind to mucin optimally at pH 2.0 (Fig. 1e).

Polysaccharide agglutination assay

PS, but not PC nor mucin, was found to cause agglutination of H. pylori (Fig. 2). Allowing mucin to interact with PS before the addition of H. pylori did not prevent the agglutination of H. pylori by PS. Similarly, the presence of mucin did not affect PS in agglutinating H. pylori.

Figure 2.

 Interaction between Helicobacter pylori, polysaccharide extracts and mucin by agglutination reaction. Helicobacter pylori SS1 (1012 CFU mL−1) was incubated with polysaccharide extracts and mucin in PBS for 10 min at room temperature with constant agitation. Agglutination was observed macroscopically and microscopically. Results are representative of two independent experiments.

Animal study

Treatment of BALB/c mice with PS up to 2 h before inoculation with H. pylori was shown to significantly (P<0.05) reduce the mean bacterial load in the stomach of these animals by more than 50% (Fig. 3a). Helicobacter pylori load in the stomach of mice treated with Spirulina was also observed to be lower than that in animals in the positive control group, although the difference was not statistically significant (P≥0.05). When the mice were fed with PS and Spirulina for 4 weeks before infection with H. pylori, reduction in H. pylori load in the stomach was found to be significantly (P<0.05) reduced by 94% and 87%, respectively.

Figure 3.

 Reduction in Helicobacter pylori colonization of gastric mucosa in BALB/c mice. Mice were treated with PS and Spirulina (a) 1–2 h and (b) 4 weeks before infection with H. pylori SS1. Helicobacter pylori load was determined by the spread plate method. Each of the three mice in each group was examined (denoted by ♦), and the mean H. pylori count of the group is represented by ⊗ (P<0.05).

Helicobacter pylori mucin-binding proteins

Mucin-binding activity was found mainly in the AGE fraction containing cell-associated proteins of H. pylori. Ligand overlay analysis was performed using DIG-labelled ligand as probes to identify AGE proteins with mucin-binding activity (Fig. 4). Under denaturing conditions, two low-molecular-weight H. pylori proteins (Y and Z) in the AGE fraction were found to bind DIG-labelled mucin, as well as DIG-labelled polysaccharide extracts, PS and PC. Proteins Y and Z were identified using MALDI-TOF MS as urease A subunit (UreA; 26 kDa; score: 145) and alkyl hydroperoxide reductase C (AhpC; 22 kDa; score: 590), respectively. Mascot protein scores that were greater than 76 were considered to be significant (P<0.05) (Perkins et al., 1999). When subjected to partially denaturing conditions, protein X was found to bind DIG-labelled mucin too. Interestingly, protein X was also identified as AphC. In addition, when protein X was excised and run under denaturing conditions, it gave a protein band of similar size to that of protein Z. The mucin-binding activity of AhpC was further confirmed by the binding of DIG-labelled mucin to recombinant AhpC (Fig. 4d).

Figure 4.

 Identification of Helicobacter pylori mucin and polysaccharide binding proteins by DIG-labelled ligand overlay analysis and mass spectrometry. (a) Helicobacter pylori SS1 AGE proteins were separated by 12% SDS-PAGE and stained by Coomassie blue R250. Lane M, marker; lane 1, denaturing conditions; lane 2, partially denaturing conditions; lane 3, protein band X subjected to SDS-PAGE under denaturing conditions; lane 4, BSA (control). Proteins X and Z were identified by MALDI-TOF MS as H. pylori alkyl hydroperoxide reductase, and protein Y as H. pylori urease protein UreA. (b) Proteins were electrophoretically transferred to Immobilon membrane. The blot was probed with DIG-labelled mucins, anti-DIG-AP, and developed with NBT/X-phosphate substrate. Lane M, marker; lane 5, denaturing conditions; lane 6, partially denaturing conditions; lane 7, BSA. (c) The blot was probed with DIG-labelled PC (lane 8) and PS (lane 9), anti-DIG-AP, and developed with NBT/X-phosphate substrate. (d) Verification of recombinant AhpC mucin-binding activity by ligand overlay analysis. DIG-labelled mucin was used as probe for recombinant AhpC. Lane M, marker; lanes 10 and 12, recombinant AhpC; lanes 11 and 13, BSA. Results are representative of two independent experiments.

Discussion

With the increasing prevalence of antibiotic-resistant organisms, new strategies to combat infection are being sought. One such strategy is the use of antiadhesive molecules targeting the primary step of infection, i.e. adhesion of the organism to the host (Ofek et al., 2003). Inhibition of adhesion works on the principle that by blocking bacteria attaching to host tissue there is a lower risk that these bacteria will develop resistance compared with the use of bactericidal compounds (Bavington & Page, 2005). Investigations into algal materials as sources of antiadhesives have become common over the past few years, as many algal species contain polysaccharide substances with various biological activities at relatively high concentrations. Microalgal sulphated exopolysaccharides (Guzman-Murillo & Ascencio, 2000) and cyanobacterial exocellular polysaccharides (Ascencio et al., 2004) have been shown to inhibit the attachment of H. pylori to human gastric cell lines. Shibata et al. (2003) demonstrated that Cladosiphon fucoidan inhibited H. pylori attachment to porcine gastric mucin and reduced H. pylori-induced gastritis and the prevalence of H. pylori in infected Mongolian gerbils. Prevention of H. pylori adhesion to gastric mucin is important for antiadhesive-based preventive therapy because the impermeable nature of the mucus gel barrier may mean that high-molecular-weight polysaccharide compounds cannot cross the gel layer to block binding sites present on the surface of underlying cells.

In this in vitro study, polysaccharides extracted from two commercially available dietary microalgae were demonstrated to have an antiadhesive effect against H. pylori adhering to mucin, without killing H. pylori and AGS gastric epithelial cells at concentrations of twice the effective inhibitory dosage used in the inhibition study (data not shown). This ensured that the reduction in urease production demonstrated in the inhibitory assays was truly caused by the reduction of H. pylori attaching to mucin, and not by the bactericidal activity of the polysaccharides on the bacteria. The microalgae used in this study are generally regarded as safe to be used as functional food (Otles & Pire, 2001). Furthermore, these compounds did not affect the growth of the mammalian cells, suggesting that they should not have adverse effects on the host if used as an antiadhesive agent. Results from the inhibition and agglutination assays suggested that these polysaccharide extracts prevented H. pylori from binding to mucin by attaching to the bacteria (supplementary material, Fig. S1). More interestingly, the blocking inhibition assay showed that PS can bind to mucin, and that the binding of PS to mucin was optimum at pH 2.0, indicating its potential to be active in the acidic gastric environment in vivo. The agglutination assay also showed that the interaction of PS with mucin did not affect the binding of PS to H. pylori. In a follow-up study, we found that there was a decrease in H. pylori load in the stomach of mice fed with PS or Spirulina even for a short period of 2 h before infection. When treatment with PS or Spirulina was extended to 4 weeks, the reduction in H. pylori load was even more obvious. This may suggest that PS and Spirulina may have a protective effect against the colonization of H. pylori.

Most of the putative H. pylori adhesins identified were members of the large outer membrane protein (OMP) family. Among these, the BabA protein was shown to recognize the Leb antigen, which is present on red blood cells, gastrointestinal cells, and gastric mucins of secretors. Another adhesin, the SabA adhesin, binds to sialylated antigens, which are upregulated in inflamed gastric tissue (Odenbreit, 2005). Experiments have been performed to study the relationship between the presence of these adhesin genes and mucin binding activity. Results from these experiments suggested that H. pylori may bind to gastric mucin independent of these two adhesins (supplementary material, Fig. S2).

In this study, it is interesting to note that DIG-labelled mucin binds to AhpC and UreA, indicating the probable role of these two H. pylori proteins in mucin binding. These surface proteins may be important in H. pylori colonization. Surface-localized H. pylori urease, which is a nonmembrane protein, was reported by Icatlo et al. (1998) to have affinity for gastric mucin. However, other nonmembrane bound proteins that are expressed on the bacterial surfaces have not been adequately studied for their roles in adhesion. Although UreA and AhpC do not possess any characteristic membrane localization sequences, they were observed in the membrane fraction of H. pylori (Nilsson et al., 2000). Furthermore, AhpC can elicit immune responses in experimental animals exposed to whole H. pylori cells, supporting its localization on the bacterial surface (Yan et al., 2001). It was also found that both PC and PS are capable of binding to AhpC and UreA of H. pylori. Thus, our findings suggest that PS competes for mucin-binding sites present on H. pylori to prevent H. pylori from binding to mucin.

In conclusion, PS was demonstrated to be effective as a carbohydrate-based antiadhesive in preventing H. pylori adhering to gastric mucin without affecting the host.

Acknowledgements

This project was supported by an Academic Research Fund (Ministry of Education, Singapore) grant: no. R-182-000-070-112. L.M.F. is a research scholar of the National University of Singapore under grant no. R-182-000-071-112.

Ancillary