SEARCH

SEARCH BY CITATION

Keywords:

  • Flow cytometry;
  • Surface localization;
  • Antigen;
  • Helicobacter pylori

Abstract

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Previous studies on the localization of several different Helicobacter pylori antigens have been contradictory. We have therefore examined by using both one- and two-color flow cytometry (FCM), immunofluorescence (IF), and immunoelectron microscopy (IEM), the possible surface localization of some H. pylori antigens that may be important virulence factors. All four methods detected the lipopolysaccharide and the N-acetyl-neuroaminyllactose-binding hemagglutinin protein (HpaA) as surface-exposed, while the urease enzyme was not detected at all and the neutrophil activating protein only in low concentration on the surface of the H. pylori bacteria during culture of H. pylori in liquid broth for 11 days. The FCM analysis was found to be quite sensitive and specific and also extremely fast compared with IF and IEM, and therefore the preferred method for detection of surface-localized antigens of H. pylori.


1Introduction

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Approximately half of the world's population is colonized by Helicobacter pylori, an important causative agent of gastrointestinal diseases [1]. The bacteria possess numerous virulence factors among which the urease enzyme is by far the most studied [2,3]. H. pylori urease enzyme cleaves urea to HCO3 and NH4+, resulting in increased pH, thereby enabling the bacteria to survive in the acid gastric milieu of the host. The previous studies using cryo-immunoelectron microscopy have suggested that urease is secreted by the bacteria and is then reassociated to the bacterial surface [4]. However, a recent study suggests that the enzyme is mainly cytoplasmic, since surface-exposed urease is irreversibly inactivated at pH below 4.5, while the activation of cytoplasmic urease is increased 10–20-fold as the external pH falls down to 2.5 [3]. Like urease, the N-acetyl-neuraminyllactose-binding hemagglutinin, HpaA (H. pylori adhesin A), is a conserved protein in H. pylori. HpaA is a 30-kDa protein which has been reported to mediate binding to sialic acid in vitro [5] and to be either a flagellar sheath [6,7], a cytoplasmic [8], or an outer membrane protein of H. pylori[9]. Another H. pylori protein also believed to play a role in pathogenesis is the neutrophil activating protein (NAP). NAP has been reported to be located on the cell surface and to have the same function and similar structure as a bacterioferritin [10,11]. It has also been shown that NAP can bind to sulfated carbohydrates in mucins [12] and to glycosphingolipids of neutrophils [13].

In this study we have evaluated the surface localization of these different proteins by applying a new approach, i.e. flow cytometry (FCM). The lipopolysaccharide (LPS) was used for comparative purposes since it is known to be located in the outer membrane of the bacteria. The location of LPS, urease, HpaA, and NAP was analyzed by monoclonal antibodies (MAbs) raised against the different antigens. FCM analyses of bacteria are still not much in use, despite being proposed as an application method for bacteria as early as 1947 [14]. However, for eukaryotic cells this technique has been used extensively. FCM is multiparametric and processes samples extremely fast compared to other methods such as immunofluorescence (IF) and immunoelectron microscopy (IEM). Furthermore, FCM has sufficient sensitivity to allow detection of microscopic particles and has the ability to analyze large numbers of particles. In this study we compared the distribution of the different proteins on H. pylori cells in a one-color, i.e. phycoerythrin (PE), analysis and a two-color combination analysis of fluorescein isothiocyanate (FITC) and PE, as well as with IF and IEM. We also studied whether the proteins were localized differently during different growth phases of H. pylori.

2Materials and methods

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

2.1Growth conditions

H. pylori, strain SS1 [15], was used in all experiments. Also, a reference strain, CCUG 17875, with high urease production in vitro (Svennerholm, A.-M.; unpublished data) was included for FCM one-color analyses of urease. Cultures were grown under microaerobic conditions at 37°C, initially on Columbia agar plates, supplemented with Isovital, for 2 days and then in shaken (200 rpm) Brucella broth containing dimethyl-β-cyclodextrin (kindly provided by Teijin Ltd., Tokyo, Japan) and antibiotics (20 U ml−1 polymyxin, 10 μg ml−1 vancomycin, and 5 μg ml−1 trimetoprim). The optical density (OD600) in the Brucella broth was 0.06, corresponding to 108 cells ml−1 at the start of culture, i.e. on day 0. Samples were then taken after 1, 2, 3, 4, 7, and 11 days to cover different stages of a growth curve and analyzed for surface antigens and viable counts, as well as for morphology by phase contrast microscopy at 100× magnification. Viable counts were done by making 10-fold dilutions of the bacteria in phosphate-buffered saline (PBS) which were plated on horse blood agar plates and colonies were counted after growth under microaerobic conditions for 3–4 days. At each time point of sample collection, the culture was gassed with 6% O2, 10% CO2, and 84% N2 and contamination tested by growing an aliquot of the bacterial culture on horse blood agar plates in a microaerobic and in an aerobic incubator, respectively.

2.2Antibodies

Specific MAbs raised against H. pylori NAP, HpaA, UreB and UreA (the 66-kDa and 30-kDa subunits of urease, respectively) and LPS purified from H. pylori, strains SS1 and 17875, respectively, were produced and characterized for specificity against different H. pylori antigens in enzyme-linked immunosorbent assays (ELISAs) and immunoblots as previously described [16]. For control purposes, a MAb was raised against the Escherichia coli fimbriae protein, CS17 [17]. The anti-SS1 LPS MAb 3:6 was shown by us in immunoblotting to recognize the O-side chain of LPS and in IEM to stain the whole surface of H. pylori, strain SS1 bacteria (not shown), and therefore used as a positive control in the FCM analyses. The anti-17875 LPS MAb [18] similarly stained all 17875 bacteria. The properties of the different MAbs are shown in Table 1. For the two-color FCM analysis, the anti-LPS MAbs were labelled with FITC (Sigma-Aldrich, Stockholm, Sweden) as described by The et al. [19]. The MAbs specific for the H. pylori and the E. coli proteins were labelled with biotin (Sigma), according to Bayer et al. [20]. The MAb U-8:1 specific for the UreA subunit of urease was only tested in one-color analyses.

Table 1.  MAbs
  1. aNegative control; raised against a fimbrial protein of enterotoxigenic E. coli.

  2. bThe HP30-1:1:6 is specific for HpaA as shown by immunoblotting experiments with purified HpaA (Lundström, A.M. et al., to be published).

  3. cOwn production; Svennerholm, A.-M. et al.

  4. dOwn production; Evans, D.G. and Svennerholm, A.-M.

SpecificityMAbIsotypeIg concentration (μg ml−1)ELISA titerReference
HpaAHP30-1:1:6IgG1351/30 000[19]b
UreAU-8:1IgG1421/10 500(unpublished)c
UreBU-9:16:11IgG1291/13 000(unpublished)c
NAPNAP-6:8IgG1311/3000(unpublished)d
LPS17875LPS17-3:4IgM1151/3500[21]
SS1 LPSSS1 LPS-3:6IgG11201/8000(unpublished)b
CS17aCS17-8:1IgG11031/3000[20]

2.3FCM

Liquid cultures of SS1 and CCUG 17875 bacteria, respectively, were harvested by centrifugation at 11 000×g for 5 min and resuspended in 0.1% bovine serum albumin (BSA)–PBS to an OD600 of 0.1 corresponding to 2.5×108 cells ml−1, before FCM analysis. A total of 2.5×106 cells were then subjected to either one- or two-color analysis. Strain CCUG 17875 was only analyzed in one-color analyses for UreA and UreB. In the one-color analysis the cells were mixed with 30 μg ml−1 of either of the following primary MAbs; SS1 LPS-3:6 (positive control), HP30-1:1:6, NAP-6:8, U-9:16:11, U-8:1, and CS17-8:1 (negative control) and then with 6.25 μg ml−1 rat anti-mouse IgG1, labelled with PE (Becton Dickinson (BD), Temse, Belgium). The respective antibodies were added, diluted 1:10 in 0.1% BSA–PBS to give a final concentration of 3 μg ml−1 and 0.625 μg ml−1, respectively, and incubations were performed for 20 min at 4°C. In between and after the incubations the cells were washed with PBS once and twice, respectively. In the two-color analysis the bacterial cells were resuspended in 0.1% BSA–PBS and mixed with the H. pylori specific LPS MAb labelled with FITC and either of the biotinylated MAbs. Both the FITC- and biotin-labelled MAbs were added 1:10 to give a final concentration of 5 μg ml−1 in the mixture. This mixture was incubated at 4°C for 20 min and then washed once in PBS; thereafter 10 μl of PE-conjugated streptavidin (RPE) (STAR4B; Serotec Ltd., 22 Bankside, Station Approach, Kidlington, Oxford, UK) in 0.1% BSA–PBS was added 1:6, and the mixture was incubated for 20 min at 4°C and then washed twice with PBS before analysis.

The pelleted bacteria labelled with either one or two colors were resuspended in 400 μl PBS and analyzed by using a FACSCalibur (BD). In order to analyze only the bacterial cells, the cytometer data were viewed as scattergrams showing the relation between particle size and fluorescence. The threshold was set in the sidescatter channel to exclude noise and artefacts in the fluorescence channel. For each sample 50 000 events were collected. In all experiments a positive control, MAb SS1 LPS–3:6 when studying strain SS1 and MAb LPS17-3:4 when testing strain 17875, and a negative control, MAb CS17-8:1, were used. In two-color analysis, a FITC-labelled mouse IgG (BD) of irrelevant specificity was also used as a negative control to confirm that the FITC-labelled positive control, i.e. SS1 LPS-3:6, did not bind non-specifically to the bacterial cells.

2.4IF microscopy

In IF, two-color analyses were done as described for FCM but the pelleted H. pylori bacteria were resuspended in 100 μl PBS instead of 400 μl PBS. The bacteria were then applied to slides by cytospin (Shandon Southern Inc., Pittsburg, PA, USA) run for 5 min at 300×g and covered with one drop of mounting gel and cover slip. The samples were stored in a dark box at 4°C until analyzed within 6 h using a Leica microscope at 100× magnification with a G/R filter (no. 513803), which enabled the analysis of both the red and green color simultaneously.

2.5IEM

Bacterial samples of H. pylori, strain SS1, were collected after 2 days of growth in Brucella broth with shaking. The bacteria were subjected to IEM to study the localization of the H. pylori antigens; LPS, HpaA, urease, and NAP. The bacterial specimens with OD600 of 1.5 were incubated together with one drop of the respective H. pylori specific MAbs (diluted 1:2 of the original concentration, see Table 1) on Formvar-coated copper grids and detected with gold-labelled goat anti-mouse IgG (British BioCell International, Cardiff, UK) diluted 1:30. The grids were negatively stained with 1% ammonium molybdate and viewed at 25 000–100 000× magnification.

3Results

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

3.1FCM

The distribution of four different H. pylori antigens during different growth phases of H. pylori strain SS1 in liquid culture was analyzed in one- and two-color analyses, respectively (Fig. 1). The different growth phases (log, stationary, and declination phases) corresponded to the optical density values and colony forming units (CFUs), which in turn reflected the observed morphology of the cells, i.e. being more coccoid forms as the CFUs decreased (Table 2).

image

Figure 1. Histograms showing the intensity of fluorescence emission of H. pylori, strain SS1, after 1 day of liquid culture, stained with (A) one color (PE) and (B) two colors (FITC and PE). Negative controls, CS17 (PE) and mouse IgG (FITC), were used to enable the gating of only positively stained H. pylori bacteria (denoted ‘M’).

Download figure to PowerPoint

Table 2.  Viable counts, OD values, and % coccoids of H. pylori, strain SS1, in specimens collected at various times after culture with shaking in Brucella broth
  1. aThe % coccoids was subjectively estimated by phase contrast microscopy.

Days of cultureCFUs (log10)Optical density (OD600 nm)Coccoids (%)a
080.060
19.10.161
29.80.31
38.50.475
48.80.5650
77.50.470
1150.4390

As expected, the SS1 LPS MAb stained nearly 100% of the bacterial cells in both the one– and two-color analyses during the entire growth curve (Fig. 2). Moreover, in IEM the SS1 LPS MAb was shown to stain the cell surface of all H. pylori strain SS1 bacteria studied (not shown). Therefore, the SS1 LPS MAb was used as the positive control in our analyses. The negative control MAb (CS17-8:1) did not stain any of the cells at any time point. Similarly, the bacteria were not stained for H. pylori urease, using the MAb 9:16:11 raised against the UreB subunit, at any time point studied between 1 and 11 days of growth (Figs. 1 and 2) independent of the morphological stage (bacillary or coccoid forms) of the H. pylori bacteria. Analogous results were obtained with the MAb U-8:1 raised against the UreA subunit of urease (not shown). Similarly, no surface staining of urease was seen in strain 17875 when testing bacteria cultured for 4 or 11 days using the MAb against UreA and UreB, respectively. HpaA, on the other hand, was detected on the surface of SS1 bacteria, both when using one-color and two-color analyses, and the percentage of positive cells varied between 30% and 70% during the period studied (Fig. 2). Using one-color analysis, no cells were found to expose the NAP protein on the surface of SS1 bacteria, while in two-color analysis the distribution of NAP increased steadily from no staining on day 4 up to 5% on day 7 (Fig. 2).

image

Figure 2. Surface localization of different H. pylori antigens at different times during culture of H. pylori, strain SS1 in Brucella broth. The reactivity with the specific MAbs (SS1 LPS–3:6 (▪), HP30-1:1:6 (▴), NAP-6:8 (•), and urease 9:16:11 (*)) and the negative control (CS17-8:1 (♦)) were plotted as % labelled bacteria in (A) one-color analysis and in (B) two-color analysis.

Download figure to PowerPoint

3.2IF and IEM

Using IF, neither urease nor NAP could be detected on the surface of the SS1 cells at any time point (not shown). In contrast homologous LPS and HpaA were observed on the surface of several cells and using double staining it could be shown that the same cells expressed both antigens (Fig. 3).

image

Figure 3. IF pictures (A, B and C) and FCM contour plots (D and E) of H. pylori, strain SS1, stained for expression of LPS (FITC, green) and/or HpaA (PE, red). The cells were stained with MAbs against HpaA only (A), HpaA and LPS (B and D) or LPS only (C and E).

Download figure to PowerPoint

Using the same MAbs in IEM, urease was not detected on the surface of SS1 cells, whereas gold labelling was observed on the bacterial surface when using the MAbs against LPS (Fig. 4) and HpaA (Lundström et al., to be published), respectively; only sparse labelling was observed with the MAb against NAP (not shown).

image

Figure 4. IEM picture of H. pylori strain SS1 stained with MAbs LPS-3:6. Original magnification 20 000×. The black bar corresponds to 1 μm.

Download figure to PowerPoint

4Discussion

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Little is still known about the antigen composition of the H. pylori outer membrane. Only a few antigens have been characterized as being localized on the surface of the bacterial cell. The LPS, including the O-antigen specific sugar chain, is among the antigens that are located on the surface, since it constitutes a substantial part of the cell wall. However, the location of other known H. pylori antigens such as HpaA, NAP, and urease has been quite controversial. We therefore evaluated the surface exposure of these H. pylori antigens using FCM. FCM has previously been used to analyze mainly eukaryotic cells, while the use of FCM has not been widely applied for studying bacteria, especially not for the quantification of antigens exposed on the bacterial surface. There is one study describing the quantification of the surface-exposed heat shock protein (HSP) 60 [21] using FCM. However, in that study the authors were using a MAb and polyclonal serum raised against HSP60 of Yersinia enterocolitica in one-color analysis to detect the antigen on H. pylori. Using cross-reactive MAb and polyclonal antiserum may not be optimal for the identification of a heterologous antigen. Thus, if the recognition of the antigen is not perfect, it will result in lower intensity which lowers the detection level. Also, the use of one-color analysis will probably lead to lower sensitivity and specificity since the antigens of interest will not be analyzed within a defined and labelled cell population. The two-color FCM analysis that we have developed was made quite sensitive and specific, since artefacts influencing the results were efficiently excluded by staining the whole cell population with one color, i.e. FITC, and then among those cells stained with FITC detect the antigen of interest with another color, i.e. PE.

By using FCM, we could show that HpaA, unlike both urease subunits (UreA and UreB) and NAP, is localized on the surface of H. pylori in vitro, during all stages of growth, both on bacteria in the bacillary and coccoid phases; these findings were confirmed by IF and IEM. However, there was an initial decrease in the surface expression of HpaA from 1 to 3–4 days of growth. This may be a response to the changed culture media, i.e. from solid media to broth, but once the cells have adapted to their new milieu, they might increase the transport of the HpaA protein out to the bacterial surface. Other studies have also suggested that the H. pylori bacteria show a shift in antigenic surface expression when the bacteria are transferred from plate culture to broth [22]. Our results concerning the distribution of the HpaA protein are in agreement with those of Evans et al. and Bölin et al. [9,16], who characterized the proteins located onto the surface using polyclonal rabbit serum and MAb, respectively, raised against HpaA. However, our results are in contrast with those of O'Toole et al. [8] who reported the protein to be predominantly in the cytoplasmic fraction of sarcosyl-solubilized cells. Jones et al. [7] and Luke et al. [6] showed by immunogold labelling, using a MAb against HpaA, that the protein was restricted to the flagellar sheath, whereas we could detect the HpaA protein both on the flagellar sheaths and on the bacterial cell surface by using IEM (Lundström, A.M. et al., to be published).

NAP was first reported to be located in the outer membrane and like the HpaA protein, NAP has been suggested to bind to carbohydrates in mucus [12]. Therefore, NAP was considered to be located on the surface. In a recent study, however, NAP was shown to be similar to a dps (DNA-protecting protein) homologue and a bacterioferritin [11], indicating a cytoplasmic location. Our results confirm that NAP is mainly intracellularly located and may suggest that a small percentage of the antigen may be transported out to the bacterial surface. Using IF, NAP could not be detected on the surface whereas in IEM, NAP could be seen sparsely distributed on some of the bacterial cells. However, the sparsely positive signals observed in IEM may be due to non-specific staining. The reasons for the discrepancy between the different methods may be due to the inherent low intensity of the red stain in IF, which is difficult to detect in the fluorescence microscope if the distribution of a protein is sparse on the cell surface. The surface detection of NAP on some cells by two-color FCM may be explained by the fact that FCM is sensitive and fast, analyzing thousands of cells in a short time, i.e. 30 s. The difference seen in the surface exposure of NAP on the H. pylori bacteria, when using one-color compared with two-color analysis, may also be explained by the higher sensitivity of the two-color analysis method, which was performed in such a way that only the H. pylori cell population was analyzed, excluding any eventual artefacts.

Studies by Bode et al. [5] indicate that the urease is secreted by the bacteria and then reassociated to the bacterial surface. However, using FCM analyses we have shown that neither UreA nor UreB is exposed on the cell surface during any stage of growth, not even when the majority of the cells were coccoids. These results were confirmed by our IF and IEM analyses. The absence of the urease enzyme on the surface was observed both for strain SS1 and strain CCUG 17875; the latter is an H. pylori strain shown by us to have a very high urease activity. These results might be due to the possibility that the cells have not yet lysed so that the urease enzyme could have leaked out and reassociated with intact neighboring cells, or alternatively that under the growth conditions used (Brucella broth with 5%β-cyclodextrin with shaking) the urease enzyme was not exported into the media. However, in all other bacterial species the urease enzyme is in the cytoplasm [23] while in H. pylori it has been suggested to be located not only in the cytoplasm but also on the surface [4] and also to be secreted extracellularly [23]. It has also been assumed that only the surface-located urease is active [24,25]. However, a recent study has shown that it is the cytoplasmic urease that acts to reduce the acidity and that only a small percentage of the urease leaks out and is associated with the surface [3]. The study of Weeks et al. [3] affirms our results that the urease is mainly located intracellularly and not on the surface of the bacterial cell.

Compared with IF and IEM that are very labor intensive, FCM is simple and extremely fast. In future studies we plan to study the surface distribution of different H. pylori antigens expressed on H. pylori cells in vivo, which would add valuable information about the pathogenicity of H. pylori and on the identification of potential protective antigens in these bacteria.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

This study was financially supported by the Swedish Medical Research Council Grant 16X–09084.

References

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References
  • [1]
    Hopkins, R.J. Morris, J.G. Jr. (1994) Helicobacter pylori: the missing link in perspective. Am. J. Med. 97, 265277.
  • [2]
    Hazell, S.L. (1992) The role of Helicobacter pylori urease: a contentious issue. Eur. J. Gastroenterol. Hepatol. 4 (Suppl. 1), 5559.
  • [3]
    Weeks, D.L., Eskandari, S., Scott, D.R., Sachs, G. (2000) A H+-gated urea channel: the link between Helicobacter pylori urease and gastric colonization. Science 287, 482485.
  • [4]
    Bode, G., Malfertheiner, P., Lehnhardt, G., Nilius, M., Ditschuneit, H. (1993) Ultrastructural localization of urease of Helicobacter pylori. Med. Microbiol. Immunol. (Berl.) 182, 233242.
  • [5]
    Evans, D.J. Jr. Evans, D.G., Smith, K.E., Graham, D.Y. (1989) Serum antibody responses to the N-acetylneuraminyllactose-binding hemagglutinin of Campylobacter pylori. Infect. Immun. 57, 664667.
  • [6]
    Luke, C.J., Penn, C.W. (1995) Identification of a 29 kDa flagellar sheath protein in Helicobacter pylori using a murine monoclonal antibody. Microbiology 141, 597604.
  • [7]
    Jones, A.C., Logan, R.P., Foynes, S., Cockayne, A., Wren, B.W., Penn, C.W. (1997) A flagellar sheath protein of Helicobacter pylori is identical to HpaA, a putative N-acetylneuraminyllactose-binding hemagglutinin, but is not an adhesin for AGS cells. J. Bacteriol. 179, 56435647.
  • [8]
    O'Toole, P.W., Janzon, L., Doig, P., Huang, J., Kostrzynska, M., Trust, T.J. (1995) The putative neuraminyllactose-binding hemagglutinin HpaA of Helicobacter pylori CCUG 17874 is a lipoprotein. J. Bacteriol. 177, 60496057.
  • [9]
    Evans, D.G., Karjalainen, T.K. Evans, D.J. Jr., Graham, D.Y., Lee, C.H. (1993) Cloning, nucleotide sequence, and expression of a gene encoding an adhesin subunit protein of Helicobacter pylori. J. Bacteriol. 175, 674683.
  • [10]
    Evans, D.J. Jr. Evans, D.G., Takemura, T., Nakano, H., Lampert, H.C., Graham, D.Y., Granger, D.N., Kvietys, P.R. (1995) Characterization of a Helicobacter pylori neutrophil-activating protein. Infect. Immun. 63, 22132220.
  • [11]
    Tonello, F. et al. (1999) The Helicobacter pylori neutrophil-activating protein is an iron-binding protein with dodecameric structure. Mol. Microbiol. 34, 238246.
  • [12]
    Namavar, F., Sparrius, M., Veerman, E.C., Appelmelk, B.J., Vandenbroucke-Grauls, C.M. (1998) Neutrophil-activating protein mediates adhesion of Helicobacter pylori to sulfated carbohydrates on high-molecular-weight salivary mucin. Infect. Immun. 66, 444447.
  • [13]
    Teneberg, S., Miller-Podraza, H., Lampert, H.C. Evans, D.J. Jr., Evans, D.G., Danielsson, D., Karlsson, K.A. (1997) Carbohydrate binding specificity of the neutrophil-activating protein of Helicobacter pylori. J. Biol. Chem. 272, 1906719071.
  • [14]
    Davey, H.M., Kell, D.B. (1996) Flow cytometry and cell sorting of heterogeneous microbial populations: the importance of single-cell analyses. Microbiol. Rev. 60, 641696.
  • [15]
    Lee, A., O'Rourke, J., De Ungria, M.C., Robertson, B., Daskalopoulos, G., Dixon, M.F. (1997) A standardized mouse model of Helicobacter pylori infection: introducing the Sydney strain. Gastroenterology 112, 13861397.
  • [16]
    Bölin, I., Lönroth, H., Svennerholm, A.M. (1995) Identification of Helicobacter pylori by immunological dot blot method based on reaction of a species-specific monoclonal antibody with a surface-exposed protein. J. Clin. Microbiol. 33, 381384.
  • [17]
    Viboud, G.I., Binsztein, N., Svennerholm, A.M. (1993) Characterization of monoclonal antibodies against putative colonization factors of enterotoxigenic Escherichia coli and their use in an epidemiological study. J. Clin. Microbiol. 31, 558564.
  • [18]
    Tinnert, A., Mattsson, A., Bölin, I., Dalenbäck, J., Hamlet, A., Svennerholm, A.M. (1997) Local and systemic immune responses in humans against Helicobacter pylori antigens from homologous and heterologous strains. Microb. Pathog. 23, 285296.
  • [19]
    The, T.H., Feltkamp, T.E. (1970) Conjugation of fluorescein isothiocyanate to antibodies. II. A reproducible method. Immunology 18, 875881.
  • [20]
    Bayer, E.A., Wilchek, M. (1980) The use of the avidin–biotin complex as a tool in molecular biology. Methods Biochem. Anal. 26, 145.
  • [21]
    Yamaguchi, H., Osaki, T., Taguchi, H., Hanawa, T., Yamamoto, T., Kamiya, S. (1996) Flow cytometric analysis of the heat shock protein 60 expressed on the cell surface of Helicobacter pylori. J. Med. Microbiol. 45, 270277.
  • [22]
    Walsh, E.J., Moran, A.P. (1997) Influence of medium composition on the growth and antigen expression of Helicobacter pylori. J. Appl. Microbiol. 83, 6775.
  • [23]
    Mobley, H.L., Island, M.D., Hausinger, R.P. (1995) Molecular biology of microbial ureases. Microbiol. Rev. 59, 451480.
  • [24]
    Dunn, B.E., Campbell, G.P., Perez-Perez, G.I., Blaser, M.J. (1990) Purification and characterization of urease from Helicobacter pylori. J. Biol. Chem. 265, 94649469.
  • [25]
    Evans, D.J. Jr. Evans, D.G., Kirkpatrick, S.S., Graham, D.Y. (1991) Characterization of the Helicobacter pylori urease and purification of its subunits. Microb. Pathog. 10, 1526.