Anti-Lewis X Antibody Promotes Helicobacter pylori Adhesion to

Download PDF
0 downloads 0 Views 431KB Size Report
Comment
When AGS cells were pretreated with anti-Lex MAb, the adhesion of the babA2 mutant ..... fragment was hybridized by both babA (left) and the Cm resistance.
INFECTION AND IMMUNITY, June 2007, p. 2661–2667 0019-9567/07/$08.00⫹0 doi:10.1128/IAI.01689-06 Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Vol. 75, No. 6

Anti-Lewis X Antibody Promotes Helicobacter pylori Adhesion to Gastric Epithelial Cells䌤 Shew-Meei Sheu,1 Bor-Shyang Sheu,3 Hsiao-Bai Yang,5 Huan-Yao Lei,1 and Jiunn-Jong Wu1,2,4* Institutes of Basic Medical Sciences1 and Molecular Medicine2 and Departments of Medicine3 and Medical Laboratory Science and Biotechnology,4 College of Medicine, National Cheng-Kung University, Tainan, and Department of Pathology, Ton-Yen General Hospital, Hsinchu,5 Taiwan Received 23 October 2006/Returned for modification 6 December 2006/Accepted 28 February 2007

Lewis X (Lex) antigen is expressed on the human gastric mucosa and the O-specific chain of lipopolysaccharides of Helicobacter pylori. This antigen can induce autoantibodies, which may be involved in bacterial colonization and thus deserve further investigation. Flow cytometry was used to examine the effects of anti-Le monoclonal antibodies (MAbs) on H. pylori adhesion. A babA2 mutant was also constructed to evaluate the effect of an anti-Lex MAb on adhesion. The bacterial agglutination and in situ adhesion assays were used to confirm the anti-Lex MAb effect on H. pylori adhesion. This study revealed that an anti-Lex MAb, but not an anti-Leb MAb or an anti-Ley MAb, could enhance the adhesion of H. pylori strains that expressed high levels of Lex antigen to AGS cells. The enhancement was not found on an H. pylori strain with a low level of Lex antigen. Anti-Lex MAb could increase the adhesion of both the wild-type strain and its isogenic babA2 mutant to AGS cells. When AGS cells were pretreated with anti-Lex MAb, the adhesion of the babA2 mutant also increased. Only anti-Lex MAb could promote bacterial agglutination, and the in situ adhesion assay further confirmed that adding anti-Lex MAb resulted in denser bacterial adhesion on the gastric epithelia collected from clinical patients. These results suggest anti-Lex MAb could specifically enhance the adhesion abilities of H. pylori strains through a mechanism by which anti-Lex MAb promotes bacterial aggregation and mediates bivalent interaction (antigen-antibody-antigen) between bacteria and host cells. Helicobacter pylori infection has been found to increase the risk of development of peptic ulcer and gastric adenocarcinoma (9). The mechanisms of persistent infection by H. pylori are not clearly understood. Lewis (Le) blood group antigens are commonly present in the human gastric mucosa and are also expressed on the O-specific chain of the lipopolysaccharide (LPS) of H. pylori (4, 5, 18, 24). The molecular mimicry between bacteria and host could be involved in colonization, adaptation to the host, the induction of autoreactive antibodies (Abs), or even gastric damage (3, 19, 20, 21). On the O-specific chain of LPS, H. pylori strains predominantly express type 2 Le antigens, including Lex and Ley, but only a few strains express type 1 antigens, Lea and Leb (28, 31). Several studies imply that the Lex and Ley expression of H. pylori facilitates bacterial adhesion and stimulates the host gastric inflammation response (10, 12, 17, 29). It has been shown that Lex and Ley antigens on H. pylori can induce autoantibodies, including anti-Lex and anti-Ley Abs, in both mouse models and clinical patients (2, 13, 22). Interestingly, in a mouse model lacking B cells (and thus unable to exert humoral immunity), H. pylori colonization was limited (1). These data imply that the Abs or autoantibodies (such as anti-Lex and anti-Ley Abs) generated by antigenic mimicry play a role in bacterial colonization. We thus aimed to determine whether anti-Lex and anti-Ley Abs can mediate H. pylori adhesion un-

der different conditions of Le antigen expression. In addition, BabA, encoded by babA2-positive H. pylori, reacting with Leb facilitates bacterial colonization (14, 26). This study thus used a babA2 mutant to validate the exact impact of anti-Le Ab on H. pylori adhesion. We found that anti-Lex monoclonal Ab (MAb) could specifically increase the bacterial adhesion of those H. pylori isolates expressing greater amounts of Lex antigen. Moreover, the effect of anti-Lex MAb was found in either the wild-type strain or the babA2 mutant. Such a positive effect on colonization could be due to the increase in bacterial aggregation or to the anti-Lex MAb mediating a possible bivalent interaction, antigen-Ab-antigen, between bacteria and host cells. MATERIALS AND METHODS Bacterial strains, cells, and culture conditions. Forty-two H. pylori clinical isolates, including HP352 and HP266, were collected from National Cheng-Kung University Hospital, Tainan, Taiwan. All isolates were identified as H. pylori by positive tests for cytochrome oxidase, catalase, and rapid urea hydrolysis and by the API Campy kit (BioMe´rieux, Marcy-l’Etoile, France) and were stored at ⫺70°C in brain heart infusion with 30% glycerol until they were tested. H. pylori isolates were grown on brucella plates containing 10% fetal cafe serum (FCS) at 37°C under microaerophilic conditions. Escherichia coli HB101 (Food Industry Research and Development Institute, Hsinchu, Taiwan) was grown on LuriaBertani agar. The human gastric adenocarcinoma cell line AGS was obtained from the Food Industry Research and Development Institute in Taiwan and was maintained in Ham’s F-12 medium (GIBCO BRL, Grand Island, NY) containing 10% FCS. The other human gastric cancer cell line, MKN-45, was obtained from the Health Science Research Resources Bank in Japan and was maintained in RPMI 1640 medium (GIBCO BRL, Grand Island, NY) containing 10% FCS. The cells were subcultured every second day. Determination of Le antigens. Le antigen expression was measured by enzyme-linked immunosorbent assays (ELISA) as described by Taylor et al. (29) with minor modifications. H. pylori isolates grew for 44 to 48 h and were sus-

* Corresponding author. Mailing address: Department of Medical Laboratory Science and Biotechnology, National Cheng Kung University, College of Medicine, No. 1 University Road, Tainan, Taiwan 701. Phone: 886-6-2353535, ext. 5775. Fax: 886-6-2363956. E-mail: jjwu @mail.ncku.edu.tw. 䌤 Published ahead of print on 19 March 2007. 2661

2662

SHEU ET AL.

pended in 1⫻ phosphate-buffered saline (PBS) at a final concentration of ⬃1 ⫻ 108 bacteria/ml. Then, 100 ␮l of the bacterial suspensions was added to each well of microtiter plates and incubated at 4°C overnight. After three washes with washing buffer (0.05% bovine serum albumin [BSA] and Tween 20 in 1⫻ PBS), the blocking solution (2.5% BSA, 5% FCS, and 0.05% Tween 20 in 1⫻ PBS) was applied at room temperature (RT) for 2 h. Then, the microtiter plate was washed three times and developed with primary Abs, including anti-Leb, anti-Lex, and anti-Ley murine MAbs (Signet Laboratories, Inc., Dedham, MA) at RT for 2.5 h. After being washed, each well was treated with horseradish peroxidase-conjugated anti-mouse immunoglobulin M (IgM) and IgG Abs (Chemicon International Inc., Temecula, CA) at RT for 1.5 h. Subsequently, three washes were done, and 1 mmol/liter 2,2⬘-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (Sigma Chemical Co., St. Louis, MO) and 0.03% H2O2 in 0.01 mol/liter citrate buffer (pH 4.2) was added at RT for 15 min. The reaction was stopped with 10 ␮l of 10% sodium dodecyl sulfate, and the absorbance was recorded at 405 nm. Escherichia coli HB101 was used as the negative control. Labeling H. pylori with the PKH2 Green Fluorescent Cell Linker Kit. In order to assay H. pylori adhesion to cells by flow cytometry, the bacterium was labeled with a PKH2 Green Fluorescent Cell Linker Kit (Sigma). The H. pylori strain was cultured for 44 to 48 h and suspended in F-12 medium to an optical density at 600 nm (OD600) of 1.0. After being centrifuged at 3,000 rpm for 10 min, the bacterial pellet was resuspended in 0.5 ml diluent A. Then, 0.5 ml diluent A containing 1.5 ␮l PKH2 dye was added and mixed with the bacterial suspension. The bacterial suspension was mixed again at 2 min, and the labeling reaction was stopped with 1 ml FCS at 4 min. Two ml of F-12 medium was added to the bacterial suspension, and the mixture was centrifuged at 3,000 rpm for 10 min. The bacterial pellet was washed one time with F-12 medium supplemented with 2% FCS and twice with 0.1% BSA in 1⫻ PBS. Finally, the bacterial pellet was suspended in 0.5 ml of 0.1% BSA, and 100 ␮l was taken for each assay. Treatment with different anti-Le MAbs in the adhesion assay. The in vitro adhesion assay with AGS and MKN-45 cells was based on the method of Osaki et al. (23). Three MAbs, anti-Leb, anti-Lex, and anti-Ley (Signet Laboratories, Inc. Dedham, MA), were used to investigate their effects on H. pylori adhesion. These three MAbs belong to the IgG, IgM, and IgM isotypes, respectively. H. pylori 266 and 352, which had low and high Lex expression, respectively, were selected for the subsequent in vitro study. Both strains were labeled with the PKH2 kit. Before the bacteria infected AGS cells, 100-␮l bacterial suspensions were left untreated or treated with anti-Le MAb (1:100 dilution) at 4°C for 1 h. After being washed two times with 0.1% BSA to remove free anti-Le MAb, bacteria and AGS cells (5 ⫻ 105/tube) were suspended in 250 ␮l of F-12 medium without or with anti-Le MAb (1:50 or 1:100 dilution), followed by incubation at 37°C for 1 h with gentle shaking. Then, 300 ␮l of buffer (2% FCS and 0.1% NaN3 in 1⫻ PBS) was added to the cell solution to be analyzed by flow cytometry. The procedure performed with MKN-45 cells was the same as that using AGS cells, except the suspension volume of bacteria and cells was 500 ␮l of RPMI medium. Construction of the babA2 mutant. The babA2 fragment, obtained from PCRs using babA2-3 (5⬘-CGGGATCCACCTCTTCAACCACCATCTTC-3⬘) and babA2-4 (5⬘-CGGAATTCCGCTCCAAAACCATAAGTCC-3⬘) primers, was ligated into the pGEM-T Easy vector (Promega, Madison, WI) to construct plasmid pMW303. Plasmid pMW303 was digested with XbaI and then filled in with Klenow (Promega, Madison, WI). A chloramphenicol (Cm) resistance cassette was cut from a plasmid, vector 78, with HincII and inserted into the blunt-end site of pMW303. This plasmid was designated pMW304. Natural transformation was used to transform pMW304 into HP352 (16). Cm (20 ␮g/ml) was used as a selection marker, and a mutant of the babA2 gene was obtained. Southern blotting and Western blotting. Extraction of genomic DNA of HP352 and its isogenic babA2 mutant was described previously (27). Genomic DNA was digested by HindIII and separated by 0.8% agarose gel electrophoresis. Southern blot analysis was based on the method of Sambrook et al. (25). The babA (411-bp) probe was obtained from the PCR product (primers babA2-3 and babA2-4) digested by HindIII. The babB probe was obtained from the PCR product amplified with primers babB-1 (5⬘-GGGCCTATATCCACTGCAAA3⬘) and babB-2 (5⬘-CCACCATTTTGGAAGTTTGG-3⬘). The Cm resistance cassette was digested with HindIII, and the 528-bp fragment was used as a probe. Western blot analysis was performed as described previously (26). Applying the babA2 mutant in the adhesion assay and the competition assay. The babA2 mutant was labeled with PKH2 and used to perform the adhesion assay by coincubation with AGS cells without or with anti-Lex MAb (1:50 dilution). AGS cells pretreated with anti-Lex MAb (1:100 dilution) were also used to determine the adhesion of the babA2 mutant. After the AGS cells were incubated with anti-Lex MAb at 4°C for 1 h, free Ab was washed two times with 0.1% BSA, and the cell pellet was incubated with the suspension of the babA2 mutant. In the competition assay, 12.5 ␮g of synthetic trimeric Lex conjugated to human

INFECT. IMMUN. serum albumin (Isosept AB, Tullinge, Sweden) was added to the coincubated solution containing anti-Lex MAb to compete away the effect of the Ab. The adhesion of the babA2 mutant was assayed as described before and analyzed by flow cytometry. Bacterial aggregation test. H. pylori 352 grown for 44 to 48 h was suspended in 0.1% BSA at an OD600 of 1.0. The bacterial suspension (1.6 ml) was centrifuged at 3,000 rpm for 10 min. Then, the bacterial pellet was resuspended in 1 ml of 0.1% BSA, and each 250-␮l aliquot was left untreated or treated with Ab in 1:25, 1:50, and 1:100 dilutions. After incubation at 4°C for 1 h, the bacterial suspensions were analyzed by flow cytometry to investigate whether bacterial aggregation existed. The intensity of forward scatter (FSC) of light during flow cytometry indirectly indicates the aggregated size of the bacterial particle. Gastric tissue samples and the in situ adhesion assay. The antral biopsy specimens were collected from two clinical patients, one with and the other without H. pylori infection. The gastric Lex expressions of both patients were confirmed to be positive by immunohistochemistry staining (26). The in situ adhesion assay was modified according to the method of Falk et al. (11). Formalin-fixed paraffin-embedded tissue sections were deparaffinized with xylene and hydrated with ethanol. The slides were washed with distilled water and then placed in 1⫻ PBS for 5 min. Blocking solution (0.2% BSA-0.05% Tween 20 in 1⫻ PBS) was applied for 25 min, and then 75 ␮l of PKH2 kit-labeled bacterial suspensions without or with 1.5 ␮l anti-Lex MAb were added to the antral tissues. Then, after incubation at RT for 1 h in a humidified chamber, the slides were washed three times with 1⫻ PBS-0.05% Tween 20 containing 0.01% Evan blue. Finally, the slides were covered with fluorescent mounting gel for investigation. The interleukin 8 (IL-8) secretion effect of anti-Lex Ab-enhanced adhesion in AGS cells. AGS cells (2.5 ⫻ 104/well) suspended in F-12 medium containing 2% FCS were cultured in single wells of 96-well plates for 1 day. Each well was then washed two times with 100 ␮l of F-12 medium. H. pylori 352 was cultured for 44 to 48 h and suspended in 0.1% BSA at an OD600 of 1.0. The bacterial suspension (0.2 ml), combined with 0.1% BSA (0.8 ml), was centrifuged at 3,000 rpm for 10 min. After the bacterial pellet was suspended in 500 ␮l F-12 medium, 100-␮l bacterial suspensions with or without 2 ␮l anti-Lex MAb were applied to 96-well plates that contained AGS cells. Each well was washed twice with 100 ␮l of F-12 medium 30 min later, and then 200 ␮l of F-12 medium containing 2% FCS was added. The cell supernatants were collected at 24 h, and IL-8 levels were determined with an ELISA kit (BioSource International, Inc., Camarillo, CA). The method was performed according to the standard procedure for the kit. Statistics. Statistical analysis was performed by using a paired t test. Differences were considered significant at P values of ⬍0.05.

RESULTS Lewis antigen expression on H. pylori. Forty-two H. pylori isolates were cultured, and Le antigen expression was detected by ELISA. Leb expression was determined by the OD value and ranged between 0 and 3.171. Nineteen isolates had OD values at 405 nm of ⬍0.1, and seven isolates had OD values of ⬎1.0. OD values of Lex expression ranged from 0 to 2.875. There were 8 isolates that expressed OD values of ⬍0.1, but 24 strains had OD values of ⬎1.0. In this study, two isolates, one with a high Lex expression value (OD ⫽ 2.039; HP352) and the other with a low Lex expression value (OD ⫽ 0.142; HP266), were selected for further study. The OD value for Leb reactivity in both strains was ⬍0.1. The OD values for Ley expression for both strains were ⬎1.0. Effect of anti-Le MAb on H. pylori adhesion. Anti-Leb, antiLex, or anti-Ley MAb was added to determine whether the adhesion activities of H. pylori isolates would be affected. As shown in Fig. 1A, when HP352 was pretreated with anti-Lex MAb (mean fluorescence, 247.84 ⫾ 19.48), the adhesion of HP352 was not significantly changed compared to that of the strain without pretreatment (mean fluorescence, 231.32 ⫾ 33.44). Nevertheless, the peak of fluorescence had a right shift, which indicated increased adhesion of HP352, when the antiLex MAb (1:100 and 1:50 dilutions) was added to coincubate with bacteria and AGS cells. The mean fluorescences were

VOL. 75, 2007

ANTI-Lex ANTIBODY PROMOTES H. PYLORI ADHESION

2663

FIG. 2. Southern blot and Western blot analyses of HP352 and its babA2 mutant. (A) Southern blot analysis of HindIII-digested DNA isolated from HP352 (lane 1) and its babA2 mutant (lane 2). Hybridization was performed with three different probes, babA (left), babB (middle), and the Cm resistance cassette (right). Two markers, ␭ HindIII (lane M1) and 100 bp (lane M2), were used as molecular size standards. Hybridization to DNA of HP352 with the babA probe identified two fragments (left), and the babB probe confirmed that the 2.3-kb fragment was the babB gene (middle). In the babA2 mutant, the 0.87-kb fragment was hybridized by both babA (left) and the Cm resistance cassette (right). The faint band above the 0.87-kb fragment was incompletely digested, and the additional high-molecular-weight band for the babA2 mutant probed with babB represented undigested product (middle). (B) Western blot of whole bacterial extracts of HP352 (lane 1) and the babA2 mutant (lane 2) performed by using anti-BabA Ab (top) and anti-Hsp60 MAb (bottom). Hybridization to bacterial protein with anti-Hsp60 MAb was the internal control.

FIG. 1. The ability of H. pylori 352, with and without anti-Lex, anti-Leb, and anti-Ley MAb treatment, to adhere to AGS or MKN-45 cells was examined with flow cytometry. (A) Effect of anti-Lex MAb on the adhesion of HP352. PKH2-labeled HP352 was incubated with AGS cells (wide black peak). HP352 pretreated with anti-Lex MAb (1:100 dilution) and incubated with cells is shown as a blue peak. The green peak represents HP352 pretreated with anti-Lex MAb (1:100 dilution) with additional anti-Lex MAb (1:100 dilution) applied in the coincubation solution that included bacteria and cells. The yellow peak represents HP352 pretreated with anti-Lex MAb (1:100 dilution) and additional anti-Lex MAb (1:50 dilution) applied in the coincubation solution. The thin black peak is the cell control. (B) Confirmation of the anti-Lex MAb specificity in HP352 adhesion by using anti-Leb MAb (top) and anti-Ley MAb (bottom). The method was the same as for panel A, and the peaks represent the same experimental treatments.

327.64 ⫾ 47.12 and 389.70 ⫾ 71.78, respectively. This was significantly increased compared to that without pretreatment (HP352 alone; P ⬍ 0.05). Such a shift was not found when anti-Leb or anti-Ley MAbs were used to pretreat HP352 or added to the coincubation solution (Fig. 1B). The adhesion enhancement from treatment with anti-Lex MAb also occurred when MKN45 cells were used as targets, and pretreatment of HP352 with anti-Lex MAb had a clearer effect on increasing adhesion than using AGS cells (Fig. 1C). In contrast, both pretreatment and further coincubation with anti-Lex MAb could not enhance the adhesion of HP266 (expressing low levels of Lex antigen) to MKN45 cells (Fig. 1D). Effect of anti-Lex MAb on the adhesion of the babA2 mutant. The babA2 mutant was constructed and confirmed by Southern and Western blot analyses (Fig. 2). Because the babA probe had partial sequence similarity to the babB gene, two hybridizing fragments were detected in HP352 (Fig. 2A, left). The

The thin black peak also represents the cell control. (C) MKN45 cells were used for the assay, and the experimental procedure was the same as for panels A and B. (D) HP266 and MKN45 cells were used, and the procedure was the same as for the other panels.

2664

INFECT. IMMUN.

SHEU ET AL.

FIG. 3. Effects of anti-Lex MAb on H. pylori adhesion with and without BabA. PKH2-labeled HP352 or the babA2 mutant coincubated with AGS cells without or with anti-Lex MAb (1:50 dilution) was analyzed by flow cytometry. The relative fluorescence was determined with the adhesion fluorescence of HP352 as the reference value, and the bacterial adhesion value was divided by that value. The asterisks indicate a significant difference between adding MAb (p) or not (■) (paired t test; P ⬍ 0.05). In addition, tri-Lex antigen was added to compete the effect of anti-Lex MAb (z) and antigen control (䊐). AGS cells were pretreated with anti-Lex MAb and then incubated with the labeled babA2 mutant (o). ⫹, added; ⫺, not added.

specific babB probe was used further and found a fragment of about 2.3 kb in both HP352 and the babA2 mutant (Fig. 2A, middle). The 0.87-kb fragment from the babA2 mutant (Fig. 2A, left) was also detected by the probe of the Cm resistance cassette (Fig. 2A, right). No BabA expression was found in the babA2 mutant (Fig. 2B). To determine whether the effect of anti-Lex MAb was still active under non-BabA adhesion conditions, the babA2 mutant was used. As shown in Fig. 3, the adhesion ability of the babA2 mutant was decreased compared with the that of the wild-type strain of HP352 (P ⬍ 0.001). Anti-Lex MAb treatment could significantly enhance the adhesion of both HP 352 and the babA2 mutant (P ⬍ 0.05). The adhesion ability of the babA2 mutant could be significantly restored to levels similar to that of HP352 when incubated with the anti-Lex MAb (a 1.77-fold increase). When tri-Lex Ag was added to compete with the effect of anti-Lex MAb, the adhesion ability was similar to that of the babA2 mutant alone. In addition, when AGS cells were pretreated with anti-Lex MAb, the adhesion of the babA2 mutant also increased 1.37-fold. Identification of the mechanism of adhesion enhancement mediated by anti-Lex MAb. To determine what mechanism might be involved in increasing bacterial adhesion, different concentrations of anti-Lex MAb were applied to evaluate whether the MAb could cause bacterial aggregation. As shown in Fig. 4A, we found that the FSC of bacteria increased with the increase of the anti-Lex MAb concentration in a dosedependent manner. However, when the same concentration of anti-Ley MAb was added, the size of the bacterial particle did not obviously increase (Fig. 4B). Adding 5 ␮l of anti-Leb to HP352 (1:50 dilution) also resulted in an FSC similar to that of HP352 alone (199.45 versus 198.27). An in situ adhesion assay showed that when HP352 and anti-Lex MAb were coincubated with antral tissues (Fig.

FIG. 4. The effects of anti-Lex and anti-Ley MAbs on bacterial agglutination. (A) Bacterial suspensions of HP352 were treated without (bar 1) or with different concentrations (bar 2, 1:100 dilution; bar 3, 1:50 dilution; bar 4, 1:25 dilution) of anti-Lex MAb for 1 h to evaluate the size of the bacterial particle by flow cytometry. The FSC of light during flow cytometry indicates the size of the particle. The asterisks indicate that the bacterial particle size significantly increased compared with HP352 alone (paired t test; P ⬍ 0.05). (B) Bacterial suspensions of HP352 were performed without (bar 1) or with different concentrations (bar 2, 1:50 dilution; bar 3, 1:25 dilution) of anti-Ley MAb. The error bars indicate standard deviations.

5B and D), there were large bacterial particles adhering to the gastric epithelial cells compared to those after incubation with HP352 alone (Fig. 5A and C). In addition, the epithelial cells adhered to by HP352 were also increased (Fig. 5B and D). As shown in Fig. 5E, the babA2 mutant of HP352 had much less bacterial adherence to the gastric epithelium, and increased adhesion was not found when the babA2 mutant and anti-Lex MAb were coincubated with the antral tissues (Fig. 5F). The IL-8 level is stimulated by HP352 and anti-Lex MAb. Based on anti-Lex MAb increasing HP352 adhesion, we determined whether IL-8 secretion by AGS cells was also affected. As shown in Fig. 6, the IL-8 production of AGS cells infected with HP352 and anti-Lex MAb increased compared with HP352 alone (P ⫽ 0.055). DISCUSSION Akhiani et al. (1) demonstrated that Abs have a role in maintaining H. pylori colonization. They found that most bacteria were cleared after 8 weeks of H. pylori infection in a B-cell-deficient mouse model, but bacterial colonization was maintained in wild-type mice. Additionally, experiments using IgA-deficient mice have shown that IgA Abs promote, rather than prevent, bacterial colonization (7). However, it is still unclear why Abs promote H. pylori colonization and whether there are IgM isotype Abs with the antigen specificity to mediate bacterial colonization. In the present study, we have shown that anti-Lex MAb could promote the adhesion ability of an H. pylori strain with high Lex expression and that this effect could be produced through a mechanism by which anti-

VOL. 75, 2007

ANTI-Lex ANTIBODY PROMOTES H. PYLORI ADHESION

2665

FIG. 5. Comparison of H. pylori alone versus supplemented with anti-Lex MAb in binding to antral tissues by the in situ adhesion assay. Fluorescence microscopy was used to investigate adherence of PKH2-labeled HP352 or the babA2 mutant to the antral tissues from patient 1 (A and B) and patient 2 (C, D, E, and F). H. pylori 352 unsupplemented (A and C) or with (B and D) anti-Lex MAb and the babA2 mutant unsupplemented (E) or with (F) anti-Lex MAb were used to investigate their abilities for in situ adhesion. The Lex distribution of gastric tissue (G) detected by immunohistochemistry stain is shown as a brick red color. The tissue sections in panels C, D, E, and F were from the same patients as in panel G. Lewis X antigen was expressed on most of the epithelial cells.

2666

SHEU ET AL.

FIG. 6. IL-8 secretion responses of AGS cells induced by HP352 alone or combined with anti-Lex MAb. The cell supernatants of AGS cells infected with HP352 alone or supplemented with anti-Lex MAb were collected at 24 h. The mean level of IL-8 stimulated by HP352 and anti-Lex MAb was close to a significant difference compared to that of HP352 alone (paired t test; P ⫽ 0.055). The error bars indicate standard deviations.

Lex MAb promotes bacterial aggregation and mediates the Lex-Lex interaction. Previously, Wyatt et al. (32) demonstrated that H. pylori coated with host IgM Ab colonized the gastric mucosa, which was found in patients with active gastritis. Chmiela et al. (8) also suggested that anti-Lex IgM Ab may be produced naturally in humans. These results indicate that anti-Lex IgM Ab may be present in the local gastric lumen of H. pylori-infected patients and that it has the ability to bind H. pylori in vivo. Our study showed that anti-Lex MAb enhanced the adhesion ability of HP352 (Fig. 1A) and that this adhesion enhancement was specifically mediated by anti-Lex MAb (Fig. 1A and B and 3). The same finding could be reproduced in the other two H. pylori strains with high Lex expression (data not shown). We also confirmed this by using the Lex mutant (strain 98-1014) of strain 26695 provided by M. J. Blaser (30). When strain 26695 and anti-Lex MAb (1:50 dilution) were coincubated with AGS cells, the peak of adhesion had a right shift compared to incubation without Abs. However, the adhesion fluorescence of the Lex mutant (strain 98-1014) did not increase even when incubated with anti-Lex MAb. Based on these data, we proposed that anti-Lex MAb could be one of the IgM Abs that binds to H. pylori in vivo and may play a role in maintaining H. pylori density. Although Osaki et al. (23) used high concentrations of anti-H. pylori LPS IgM MAb (125 to 500 ␮g/ml) and demonstrated that the MAb had inhibitory effects on H. pylori adhesion, we believed the discrepancy between these data might be due to the two Abs having different epitope recognition. BabA protein of H. pylori binding to Leb antigen on gastric epithelial cells has been shown to serve as a major adhesion receptor interaction (14, 26). In this study, the in vitro adhesion assay showed that a babA2 mutant decreased in adhesion ability by 45% compared to that of a wild-type strain, whereas the increasing adhesion level of the babA2 mutant was even higher than that of the wild-type strain when both were coincubated with anti-Lex MAb (Fig. 3). These data implied that the anti-Lex MAb can serve as an alternative pathway to enhance the adhesion of H. pylori when it lacks BabA to interact with the Leb antigen of the host. This implication is supported by the previous finding of Kurtenkov et al. (15), who showed that most H. pylori-infected Leb-negative individuals were stronger responders to the Lex determinant and produced anti-

INFECT. IMMUN.

Lex Ab. Since patients without Leb expression should have limited interaction with H. pylori BabA, the stronger responders may significantly produce anti-Lex IgM Ab, which thus serves as an alternative pathway to help H. pylori adhesion in Leb-negative individuals. However, we also found that there was low babA2 mutant adhesion to antral tissue in the in situ adhesion assay even with anti-Lex MAb (Fig. 5). The washing process in the in situ adhesion assay could account for the discrepancy between it and the in vitro adhesion assay (Fig. 3). Our data indicate that the binding affinity of the babA2 mutant could be weak, and it was washed away easily. Although the interaction of the babA2 mutant and host cells was weak, anti-Lex MAb could obviously increase its adhesion (Fig. 3). Anti-Lex MAb belongs to a polymeric IgM isotype. Therefore, anti-Lex MAb may promote the aggregation of bacteria with high Lex antigen expression. AGS cells also have Lex expression on their surfaces (data not shown). Therefore, it should be possible for the MAb to serve as a bridge to connect the Lex antigen of bacteria and the Lex antigen of AGS cells and thus enhance bacterial adhesion. Although anti-Ley MAb is also a polymeric IgM isotype and both HP352 and AGS cells have Ley expression, the adhesion abilities of HP352 with and without anti-Ley MAb treatment were similar (Fig. 1B, bottom). Therefore, the specific recognition epitope of the Ab is a crucial factor in mediating adhesion enhancement. However, we cannot exclude the concentration and affinity of the Ab as other likely possibilities. In Fig. 4A, anti-Lex MAb is shown to directly promote the aggregation of HP352, but anti-Ley MAb did not have this effect (Fig. 4B). In the in situ adhesion assay, large bacterial clusters adhered to the gastric epithelium and the number of epithelial cells adhered to by HP352 increased when they were incubated with HP352 supplemented with anti-Lex MAb (Fig. 5). Our results indicated that the bacterial aggregation mediated by anti-Lex MAb could be one means to enhance H. pylori adhesion. The aggregated bacteria could interact with epithelial cells through the bridge formed by anti-Lex MAb or with receptors expressed on the surfaces of epithelial cells. Blanchard et al. (6) postulated that anti-UreB Abs may inhibit colonization by agglutinating Helicobacter felis and promoting its removal by gastric mucus flow. However, the authors reported that their other H. felis-specific MAbs also agglutinate the bacteria in vitro yet fail to inhibit colonization. Therefore, it seems that some Abs inhibit or increase bacterial adhesion by agglutinating bacteria, depending on some property of the Abs. The other possible mechanism for increasing adhesion is for anti-Lex MAb to become the interaction bridge between bacteria and AGS cells. As shown in Fig. 3, when AGS cells were pretreated with anti-Lex MAb, the adhesion of the babA2 mutant was increased 1.37-fold. However, the increased level was lower than that of the babA2 mutant coincubated with anti-Lex MAb and AGS cells (a 1.77-fold increase). This suggested that anti-Lex MAb has an ability to form a bridge between H. pylori and epithelial cells and has a partial effect on enhanced adhesion. Taylor et al. (29) illustrated that the interaction of Lex between host and bacteria has a role in H. pylori colonization. We thus postulate that anti-Lex Ab may also mediate the Lex-Lex interaction between bacteria and host cells to enhance H. pylori adhesion.

ANTI-Lex ANTIBODY PROMOTES H. PYLORI ADHESION

VOL. 75, 2007

As shown in Fig. 6, when HP352 supplemented with anti-Lex MAb adhered to AGS cells, it also induced more IL-8 secretion. These data indicate that adhesion enhancement by antiLex MAb has the additional effect of stimulating the IL-8 production of gastric epithelial cells and supports the hypothesis that agglutinating bacteria would not promote their removal when mediated by anti-Lex Ab. In summary, anti-Lex IgM MAb specifically increased the adhesion abilities of H. pylori strains with high Lex expression. It could cause bacterial aggregation and mediate Lex-Lex interaction, enhancing H. pylori adhesion. Therefore, when antiLex IgM Ab exists in the gastric lumen, H. pylori strains with high Lex expression may use one or both mechanisms to enhance adhesion. In the future, the association between the titer of anti-Lex IgM Ab in gastric juice and H. pylori density needs to be analyzed to investigate the role of antiLex IgM Ab in vivo. ACKNOWLEDGMENTS We thank Martin J. Blaser for giving us the Lex mutant and Shigeru Kamiya for providing anti-Hsp60 MAb (H20 MAb). The study was supported by grants NSC92-2314-B006-146, NSC932314-B006-022, NSC94-3112-B-006-016, and NSC94-2320-B-006-076 from the National Science Council, Taiwan. REFERENCES 1. Akhiani, A. A., K. Schon, L. E. Franzen, J. Pappo, and N. Lycke. 2004. Helicobacter pylori-specific antibodies impair the development of gastritis, facilitate bacterial colonization, and counteract resistance against infection. J. Immunol. 172:5024–5033. 2. Appelmelk, B. J., I. Simoons-Smit, R. Negrini, A. P. Moran, G. O. Aspinall, J. G. Forte, T. De Vries, H. Quan, T. Verboom, J. J. Maaskant, P. Ghiara, E. J. Kuipers, E. Bloemena, T. M. Tadema, R. R. Townsend, K. Tyagarajan, J. M. Crothers, Jr., M. A. Monteiro, A. Savio, and J. De Graaff. 1996. Potential role of molecular mimicry between Helicobacter pylori lipopolysaccharide and host Lewis blood group antigens in autoimmunity. Infect. Immun. 64:2031–2040. 3. Appelmelk, B. J., M. A. Monteiro, S. L. Martin, A. P. Moran, and C. M. Vandenbroucke-Grauls. 2000. Why Helicobacter pylori has Lewis antigens. Trends Microbiol. 12:565–570. 4. Aspinall, G. O., and M. A. Monteiro. 1996. Lipopolysaccharides of Helicobacter pylori strains P466 and MO19: structures of the O antigen and core oligosaccharide regions. Biochemistry 35:2498–2504. 5. Aspinall, G. O., M. A. Monteiro, H. Pang, E. Walsh, and A. P. Moran. 1996. Lipopolysaccharides of Helicobacter pylori type strain NCTC 11637 (ATCC 43504): Structures of the O-antigen and core oligosaccharide regions. Biochemistry. 35:2489–2497. 6. Blanchard, T. G., S. J. Czinn, R. Maurer, W. D. Thomas, G. Soman, and J. G. Nedrud. 1995. Urease-specific monoclonal antibodies prevent Helicobacter felis infection in mice. Infect. Immun. 63:1394–1399. 7. Blanchard, T. G., S. J. Czinn, R. W. Redline, N. Sigmund, G. Harriman, and J. G. Nedrud. 1999. Antibody-independent protective mucosal immunity to gastric helicobacter infection in mice. Cell Immunol. 191:74–80. 8. Chmiela, M., M. Jurkiewicz, M. Wisniewska, E. Czkwianianc, I. PlanetaMalecka, T. Rechcinski, and W. Rudnicka. 1999. Anti-Lewis X IgM and IgG in H. pylori infections in children and adults. Acta Microbiol. Pol. 48:277– 281. 9. Dunn, B. E., H. Cohen, and M. J. Blaser. 1997. Helicobacter pylori. Clin. Microbiol. Rev. 10:720–741. 10. Edwards, N. J., M. A. Monteiro, G. Faller, E. J. Walsh, A. P. Moran, I. S. Roberts, and N. J. High. 2000. Lewis X structures in the O antigen side-chain promote adhesion of Helicobacter pylori to the gastric epithelium. Mol. Microbiol. 35:1530–1539. 11. Falk, P., K. A. Roth, T. Boren, T. U. Westblom, J. I. Gordon, and S. Normark. 1993. An in vitro adherence assay reveals that Helicobacter pylori exhibits cell lineage-specific tropism in the human gastric epithelium. Proc. Natl. Acad. Sci. USA 90:2035–2039.

Editor: J. N. Weiser

2667

12. Heneghan, M. A., C. F. McCarthy, and A. P. Moran. 2000. Relationship of blood group determinants on Helicobacter pylori lipopolysaccharide with host Lewis phenotype and inflammatory response. Infect. Immun. 68:937–941. 13. Heneghan, M. A., C. F. McCarthy, D. Janulaityte, and A. P. Moran. 2001. Relationship of anti-Lewis X and anti-Lewis Y antibodies in serum samples from gastric cancer and chronic gastritis patients to Helicobacter pylorimediated autoimmunity. Infect. Immun. 69:4774–4781. 14. Ilver, D., A. Arnqvist, J. Ogren, I. M. Frick, D. Kersulyte, E. T. Incecik, D. E. Berg, A. Covacci, L. Engstrand, and T. Boren. 1998. Helicobacter pylori adhesin binding fucosylated histo-blood group antigens revealed by retagging. Science 279:373–377. 15. Kurtenkov, O., K. Klaamas, L. Miljukhina, L. Shljapnikova, M. Ellamaa, N. Bovin, and T. Wadstrom. 1999. IgG antibodies to Lewis type 2 antigens in serum of H. pylori-infected and noninfected blood donors of different Lewis (a, b) blood-group phenotype. FEMS Immunol. Med. Microbiol. 24:227–232. 16. Lee, W. K., Y. S. An, K. H. Kin, S. H. Kin, J. Y. Song, B. D. Ryu, Y. J. Choi, Y. H. Yoon, S. C. Baik, K. H. Rhee, and M. J. Cho. 1997. Construction of a Helicobacter pylori-Escherichia coli shuttle vector for gene transfer in Helicobacter pylori. Appl. Environ. Microbiol. 63:4866–4871. 17. Martin, S. L., A. A. McColm, and B. J. Appelmelk. 2000. H. pylori adhesion and Lewis X. Gastroenterology 119:1414–1416. 18. Montiero, M. A., K. H. N. Chan, D. A. Rasko, D. E. Taylor, P. Y. Zheng, B. J. Appelmelk, H. P. Wirth, M. Yang, M. J. Blaser, S. O. Hynes, A. P. Moran, and M. B. Perry. 1998. Simultaneous expression of type 1 and type 2 Lewis blood group antigens by Helicobacter pylori lipopolysaccharides. Molecular mimicry between H. pylori lipopolysaccharides and human gastric epithelial cell surface glycoforms. J. Biol. Chem. 273:11533–11543. 19. Moran, A. P., B. J. Appelmelk, and G. O. Aspinall. 1996. Molecular mimicry of host structures by lipopolysaccharides of Campylobacter and Helicobacter spp.: implications in pathogenesis. J. Endotoxin Res. 3:521–531. 20. Moran, A. P. 1999. Helicobacter pylori lipopolysaccharide-mediated gastric and extragastric pathology. J. Physiol. Pharmacol. 50:787–805. 21. Moran, A. P., E. Sturegård, H. Sjunneson, T. Wadstro ¨m, and S. O. Hynes. 2000. The relationship between O-chain expression and colonisation ability of Helicobacter pylori in a mouse model. FEMS Immunol. Med. Microbiol. 29:263–270. 22. Negrini, R., L. Lisato, I. Zanella, L. Cavazzini, S. Gullini, V. Villanacci, C. Poiesi, A. Albertini, and S. Ghielmi. 1991. Helicobacter pylori infection induces antibodies cross-reacting with human gastric mucosa. Gastroenterology 101:437–445. 23. Osaki, T., H. Yamaguchi, H. Taguchi, M. Fukuda, H. Kawakami, H. Hirano, S. Watanabe, A. Takagi, and S. Kamiya. 1998. Establishment and characterisation of a monoclonal antibody to inhibit adhesion of Helicobacter pylori to gastric epithelial cells. J. Med. Microbiol. 47:505–512. 24. Sakamoto, J., T. Wantanabe, T. Tokumara, H. Takagi, H. Nakazato, and K. O. Lloyd. 1989. Expression of Lewisa, Lewisb, Lewisx, Lewisy, sialyl-Lewisa and sialyl-Lewisx blood group antigens in human gastric carcinoma and in normal gastric tissue. Cancer Res. 49:745–752. 25. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 26. Sheu, B. S., S. M. Sheu, H. B. Yang, A. H. Huang, and J. J. Wu. 2003. Host gastric Lewis expression determines the bacterial density of Helicobacter pylori in babA2 genopositive infection. Gut 52:927–932. 27. Sheu, S. M., B. S. Sheu, H. B. Yang, C. Li, T. C. Chu, and J. J. Wu. 2002. Presence of iceA1 but not cagA, cagC, cagE, cagF, cagN, cagT, or orf13 genes of Helicobacter pylori is associated with more severe gastric inflammation in Taiwanese. J. Formos. Med. Assoc. 101:18–23. 28. Simoons-Smit, I. M., B. J. Appelmelk, T. Verboom, R. Negrini, J. L. Penner, G. O. Aspinall, A. P. Moran, S. F. Fei, B. S. Shi, W. Rudnica, A. Savio, and J. deGraaff. 1996. Typing of Helicobacter pylori with monoclonal antibodies against Lewis antigens in lipopolysaccharide. J. Clin. Microbiol. 34:2196– 2200. 29. Taylor, D. E., D. A. Rasko, R. Sherburne, C. Ho, and L. D. Jewell. 1998. Lack of correlation between Lewis antigen expression by Helicobacter pylori and gastric epithelial cells in infected patients. Gastroenterology 115:1113–1122. 30. Takata, T., E. El-Omar, M. Camorlinga, S. A. Thompson, Y. Minohara, P. B. Ernst, and M. J. Blaser. 2002. Helicobacter pylori does not require Lewis X or Lewis Y expression to colonize C3H/HeJ mice. Infect. Immun. 70:3073– 3079. 31. Wirth, H. P., M. Yang, M. Karita, and M. J. Blaser. 1996. Expression of the human cell surface glycoconjugates Lewis X and Lewis Y by Helicobacter pylori isolates is related to cagA status. Infect. Immun. 64:4598–4605. 32. Wyatt, J. I., B. J. Rathbone, and R. V. Heatley. 1986. Local immune response to gastric Campylobacter in non-ulcer dyspepsia. J. Clin. Pathol. 39:863–870.