Helicobacter pylori protein response to human bile stress

The reflux of bile into the stomach post-prandially is a normal occurrence, especially in patients with gastroduodenal disease (Worku et al., 2004). Although it has been established that Helicobacter pylori is bile-sensitive, the mechanism allowing it to colonize a bile-containing environment is unclear. It has been suggested previously that the acidification of bile through reflux into the low pH of the stomach precipitates inhibitory bile acids, thereby allowing colonization by H. pylori (Hynes et al., 2003). Many studies support the suggestion that the faecal–oral route is important for H. pylori entering the digestive tract (Cellini et al., 1999; Stone, 1999; Kivi & Tindberg, 2006). During its transmission, H. pylori must tolerate and resist bile stress when traversing the bile-rich intestinal tract. Furthermore, H. pylori has been detected in bile, gall-bladder tissue and liver samples by PCR and Southern blot hybridization (Nilsson et al., 2000; Fallone et al., 2003; Neri et al., 2005). All these points indicate that the ability of H. pylori to tolerate bile is likely to be important for its colonization and survival in the gastrointestinal tract of humans.

Sublethal bile concentrations can trigger a physiological adaptive response in bacteria, resulting in the employment of mechanisms similar to those used in other stress responses (Begley et al., 2005). For example, a study by Bernstein et al. (1999) in Escherichia coli showed that genes including micF, osmY and dinD, which can protect DNA against damage and oxidative stress, are activated by bile salts. A DNA microarray assay in Salmonella showed that the AcrAB pump is required for bile resistance (Prouty et al., 2004). However, proteome studies focus on the complete set of proteins encoded by the genome and thus complement transcriptional studies. Fox et al. (2007) used two-dimensional electrophoresis (2-DE) to compare protein expression profiles of Campylobacter jejuni grown with and without ox bile. They found that the bile could modulate proteins such as elongation factors, ferritin, chaperones, ATP synthase and others. A proteomic approach was also used to analyse the regulation of bifidobacterial proteins during growth in the presence of bile salts (Sánchez et al., 2005). However, global changes in protein expression of H. pylori during bile and acidified bile exposure have not yet been reported.

The genomic sequencing of many micro-organisms, including H. pylori, has been completed and proteomic technologies have proved to be particularly useful for studying the physiological responses of bacteria to various environmental stresses (Tomb et al., 1997). Using this information, we analysed protein expression changes of H. pylori under bile and acidified bile stress using comparative proteomics. In an attempt to reveal the possible molecular mechanisms underlying bile resistance in this important human pathogen and to simulate accurately the effects of human bile on H. pylori, fresh human bile of normal appearance was used in this study.

Treatment of bile. Treatment of bile samples was conducted as described previously (Graham & Osato, 2000). Fresh, normal-looking human bile was collected from patients (28–65 years, five male and seven female) undergoing bile drainage for various indications without complicated chronic cholecystitis. The protocol was directed by the ethics committee and with the informed consent of patients. Bile samples were mixed, centrifuged, sterilized through a 0.22 µm pore-size membrane and separated into equal portions. Some of the bile was acidified to pH 2.0 by the addition of 1 M HCl after sterilization. The samples were then centrifuged at 8000 r.p.m. in a microfuge for 20 min to remove the precipitated bile salt, the supernatant was decanted and the pH was adjusted to pH 7.0 by the addition of 1 M NaOH for the acidified bile samples. Acidified and unacidified bile solutions were diluted 1 : 3 in Brucella broth containing 10 % fetal bovine serum and then serially diluted twofold up to a dilution of 1 : 768.

Bacterial strain and culture conditions. H. pylori strain 26695 was kindly provided by Dr Zhang Jianzhong (Chinese Disease Control and Prevention Center). The bacteria were cultured on Skirrow agar with 5 % (v/v) sheep blood for 48 h under microaerobic conditions (5 % O₂/10 % CO₂/85 % N₂, by vol.) at 37 °C. To examine the inhibitory activity of bile and acidified bile on H. pylori, plate-grown bacteria were scraped and resuspended in Brucella broth. Each test solution, comprising the various dilutions of acidified and unacidified bile, was inoculated with H. pylori cells (approx. 1x10⁷ bacteria ml^–1). The control was Brucella broth containing 10 % fetal bovine serum without bile or acidified bile. These cultures were incubated at 37 °C under microaerobic conditions for 72 h. At the end of the incubation, 10 µl aliquots were removed from each test solution and used to inoculate Skirrow agar plates containing 5 % (v/v) sheep blood to determine whether growth was inhibited. The results were scored as growth or no growth. The maximum bile and acidified bile dilution with macroscopic growth of H. pylori was defined as the maximum growth dilution. H. pylori colonies were then counted at intervals of 12 h to determine the growth rate of H. pylori in the absence of human bile and at the maximum growth dilution of unacidified and acidified bile.

Preparation of H. pylori proteins. H. pylori, cultured in the absence of bile and in the maximum growth dilution of unacidified and acidified bile, was harvested by centrifugation at 5000 g for 10 min at 4 °C and washed three times with ice-cold PBS (pH 7.2). The pellet was then dissolved in lysis buffer containing 8 M urea, 2 M thiourea, 4 % CHAPS, 1 % DTT, 1 % Pharmalyte (pH range 3–10), 1 % protease inhibitor mix and 1 % nuclease mix (Amersham Biosciences). After sonication, the solution was centrifuged at 20 000 g for 60 min at 4 °C. The protein concentration was determined using the Bradford method and proteins were stored at –80 °C until 2-DE analysis.

2-DE and image analysis. Following a typical protocol, 100 µg total protein was adjusted to a total volume of 340 µl with rehydration buffer containing 8 M urea, 4 % CHAPS, 20 mM DTT, 0.5 % Pharmalyte and a trace of bromophenol blue. For isoelectric focusing, samples were loaded onto 18 cm immobilized pH gradient (IPG) strips (pH 3–10 NL) and placed on an IPGphor instrument (Amersham Biosciences). The strips were rehydrated for 12 h at 60 V and then run for 2 h at 100 V and 1 h at 500, 1000 and 5000 V; the final step was kept constant at 8000 V to a total of 80 kVh.

After isoelectric focusing, the IPG strips were equilibrated with gentle shaking for 15 min each in buffer [50 mM Tris/HCl (pH 8.8), 6 M urea, 30 % glycerol, 2 % SDS, trace of bromophenol blue] with 0.5 % (w/v) DTT and 2 % (w/v) iodoacetamide. SDS-PAGE was performed on 13 % acrylamide gels using a PROTEAN IIxi two-dimensional cell (Bio-Rad) at 15 mA per gel for 30 min followed by 30 mA per gel until the bromophenol blue front reached the bottom of the gel. The gels were then silver-stained and scanned using ImageScanner II (Amersham Biosciences) at the 256 greyscale and 300 d.p.i. degree levels. A gel of H. pylori cultured without bile was used as a reference. The test gels were matched against the reference gel using ImageMaster 2D Elite v5.0 (Amersham Biosciences) in order to find differentially expressed protein spots (greater than twofold difference, P<0.05).

In-gel digestion and MALDI-TOF/TOF MS. Differentially expressed protein spots were excised, tryptically digested and identified by tandem matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF/TOF MS) using a 4700 MALDI-TOF/TOF mass spectrometer (Applied Biosystems). Gel spots were excised manually and then digested and desalted with trypsin and a ZipPlate (Millipore). Digested peptides were mixed with α-cyano-4-hydroxycinnamic acid and spotted onto MALDI target plates. Peptide mass fingerprints were collected in the positive-ion mass spectrometry (MS) reflector mode. MS/MS was performed with air as the collision gas at 6x10^–7 torr (800x10^–7 Pa). The MS and MS/MS spectra were analysed with a 50 p.p.m. mass tolerance by GPS Explorer v2.0.1 and Mascot v1.9 based on NCBI SWISS-PROT and local H. pylori databases (updated April 2006). Background peaks from known trypsin autodigestion fragments and common contaminants were automatically excluded prior to database searches. Oxidation of methionine and carbamidomethylation of cysteines were allowed as variable modifications. Identifications with a GPS confidence interval of greater than 99 % were accepted.

One of the originally described characteristics of H. pylori was that its growth is inhibited by ox bile (Marshall et al., 1984). After reflux into the human stomach, bile is absorbed by the underlying mucosa and forms a concentration gradient that decreases from the luminal to the epithelial surface of the mucus layer. The study of Worku et al. (2004) showed that all bile acids are chemorepellents for H. pylori. Thus, the bile gradient may drive H. pylori into the epithelial surface. This suggests that H. pylori has to endure a certain amount of bile or acidified bile stress before colonizing the gastric mucosa epithelial cells.

Inhibitory influence of bile on H. pylori before and after acidification
To approximate the in vivo conditions of the human stomach, where H. pylori mainly exists, we examined the inhibitory activity on H. pylori of fresh, normal-looking human bile before and after acidification. The maximum dilution of unacidified bile in which H. pylori could grow was 1 : 192, whilst that of acidified bile was 1 : 48. Fig. 1 shows the growth rate of H. pylori cultured in medium without human bile and with 1 : 192-diluted bile and 1 : 48-diluted acidified bile. This demonstrated that the growth of H. pylori was remarkably inhibited by human bile. Furthermore, acidification of human bile reduced its inhibitory influence on H. pylori considerably, although a certain level of activity remained. However, H. pylori was still able to propagate at this level of sublethal bile and acidified bile stress.

(15K): Fig. 1. Growth of H. pylori cultured in medium without human bile () and with 1 : 192-diluted bile () and 1 : 48-diluted acidified bile ().

Bile acids constitute approximately 50 % of the organic components of bile. Before secretion, all bile acids are conjugated as N-acyl amidates with either glycine or taurine. The ratio of glycoconjugates to tauroconjugates in human bile is usually 3 : 1 (Begley et al., 2005). The pK_a of glycine conjugates ranges between 4.3 and 5.2 and that of taurine conjugates between 1.8 and 1.9. Thus, whereas both glycine and taurine conjugates are soluble in neutral solutions, only taurine conjugates are soluble in acidic conditions. This is likely to be because glycine-conjugated bile acids are precipitated under acidic conditions and are removed by centrifugation. The reflux of bile into the stomach is a normal occurrence. Under the acidic conditions of the stomach where H. pylori mainly colonizes, glycoconjugates are precipitated and the inhibitory factors of bile will be markedly reduced.

Proteomic profiles of H. pylori under bile and acidified bile stress
Global protein expression profiles of H. pylori strain 26695 grown in the absence of human bile or with 1 : 192-diluted bile and 1 : 48-diluted acidified bile were analysed using high resolution 2-DE. To confirm the data obtained, we repeated the 2-DE analysis with three sets of independently grown cultures. Fig. 2 shows the 2-DE maps of H. pylori under normal, bile and acidified bile stress conditions. In the present study, the expression of a protein was considered to have changed if the percentage volume of its spot in the gels between bacteria cultured with and without bile showed a twofold or greater difference (P<0.05). Protein spots that were regulated significantly were cut from the gel and further identified by MALDI-TOF/TOF MS.

(95K): Fig. 2. Representative 2-DE maps of H. pylori strain 26695 proteins. The results show proteins from H. pylori grown in medium without bile or acidified bile (a), in medium with 1 : 192-diluted bile (b) or in medium with 1 : 48-diluted acidified bile (c).

We obtained information on 28 proteins whose expression changed under bile or acidified bile stress (Table 1). Among these, 23 proteins were upregulated and the other five proteins were downregulated by bile and acidified stress. These proteins are involved in various physiological functions such as molecular chaperones, iron storage, chemotaxis, flagellar mobility and the energy metabolism possibly necessary for enhanced flagellar motion required for chemotaxis. These observations indicated that H. pylori responds to bile stress through multiple mechanisms involving many signalling pathways (Fig. 3). What was interesting was that the changes in proteomic profile under bile and acidified bile were similar when compared with that of normal H. pylori, indicating that H. pylori resists bile and acidified bile through similar mechanisms.

Table 1. Summary of protein spots showing altered expression under bile and acidified bile stress

(15K): Fig. 3. A simple model of the bile stress response in H. pylori. H. pylori can perceive bile and acidified bile stress signals by putative sensors and transmit them to the cellular machinery by signal transduction to regulate protein expression. Through regulation of DNA, RNA and protein processing, H. pylori thus escapes an unfavourable environment by mobility modulation or establishing a new cellular homeostasis by changing the abundance and activities of functional proteins involved in iron storage, morphological change, proteolysis and metabolism of energy, amino acids and lipids. These processes might work cooperatively under bile and acidified bile stress.

In this experiment, three proteins involved in iron storage and metabolism were induced by bile and acidified bile stress. These proteins were non-haem iron-containing ferritin, neutrophil-activating protein (bacterioferritin) and NifU-like protein. A previous study reported that bile can chelate calcium and iron in bacterial cells resulting in iron starvation (Begley et al., 2005). Iron is essential for maintaining the basic energy and redox metabolism in bacteria and also acts as a signal to regulate the expression of many genes via the ferric uptake regulator (Fur) protein. Bacteria have two classes of iron storage protein: ferritin and bacterioferritin. Like other pathogens, H. pylori has developed several iron-acquisition systems including its Fur protein (Dhaenens et al., 1997). H. pylori non-haem iron-containing ferritin is a major component of iron storage as well as iron distribution in the cells. It can protect the bacteria from metal toxicity (Bereswill et al., 2000). It has been reported that this protein is downregulated by Fur to secure the availability of free iron when iron is scarce or titrated out by other metals (Bereswill et al., 2000). Neutrophil-activating protein is important for the pathogenesis of H. pylori because it can promote neutrophil and monocyte infiltration into the gastric mucosa. Recent studies have shown that this protein is regulated by environmental iron levels and may play a role as bacterioferritin in the response of H. pylori to iron changes (Leakey et al., 2000; Cooksley et al., 2003; Lee et al., 2004). Non-haem iron-containing ferritin and neutrophil-activating protein (bacterioferritin) were both induced by bile and acidified bile stress in our experiment. Although the mechanism by which they are regulated is not clear, we believe it is related to the change in iron levels caused by bile. The expression of these two proteins was also shown to be upregulated in H. pylori under acid stress (Toledo et al., 2002; Merrell et al., 2003). NifU-like protein is a component participating in iron–sulfur cluster biosynthesis. This protein was also upregulated by bile and acidified bile. In short, these three proteins may play important roles in the response of H. pylori to bile stress.

Mobility has been shown to play an important role in the pathogenicity of H. pylori. The flagellum, its principal mobile apparatus, is a complex system. The presence of bile is an unfavourable condition and bile is a chemorepellent for H. pylori. Thus, the flagellar system of H. pylori should exert its function in response to bile stress. Under bile and acidified bile stress, one flagellar protein, the flagellar motor switch protein, was upregulated. This protein is a component of the switch complex on the rotor of the bacterial flagellum, which modulates the direction of rotation of the flagellar motor (Toker & Macnab, 1997). Additionally, the expression of one chemotaxis protein, CheW, was downregulated by bile. This is consistent with a previous study on C. jejuni (Fox et al., 2007). When the bacterial cell senses the presence of bile, its flagellar switch modulates its direction of mobility in response to the environmental information received through the chemotaxis signal transduction pathway (Toker & Macnab, 1997). Previous studies have reported that acid shock also influences the expression of a large number of genes that encode components of the flagellar apparatus and chemotaxis protein in H. pylori (Merrell et al., 2003; Wen et al., 2003). Moreover, the upregulation of ATP synthase, isocitrate dehydrogenase and glutamate dehydrogenase may contribute to the energy needs of the enhanced flagellar motion required for chemotaxis.

Many molecular chaperones are affected by bile stress, as well as other types of stress. In our study, the expression of chaperone and heat-shock protein 70 and chaperone and heat-shock protein GroEL increased, whilst that of co-chaperone GroES decreased under bile and acidified bile stress. In addition, the intensity of protein spot 2, identified as translation elongation factor EF-Tu, increased in gels of H. pylori cultured with bile and acidified bile. This factor was also induced by ox-bile stress in a previous report on C. jejuni (Worku et al., 2004). EF-Tu can act as a molecular chaperone participating in protein folding and protection from stress (Caldas et al., 2000).

In this experiment, we found that the expression of other proteins was altered under bile stress. Rod shape-determining protein has been shown to be required for maintaining the rod-like shape in many bacteria (Jones et al., 2001; Kruse et al., 2003; Figge et al., 2004). This protein was shown to be induced by bile and acidified bile stress in our study. A previous study has shown that H. pylori changed from its primary rod-like shape to a spherical shape with blebs on the cell surface when exposed to bile (Itoh et al., 1999). These results indicate that this protein plays an important role in the morphological changes of H. pylori. Expression of the gene encoding this protein was reproducibly altered by a shift to a low pH in H. pylori (Ang et al., 2001; Merrell et al., 2003). The modulator of drug activity encoded by mda66 also changed its expression in the presence of acidified bile. The exact function of this protein in H. pylori is unclear; however, it is possible that acidified bile acts on H. pylori as a drug because of its antimicrobial actions, which induced upregulation of the modulator of drug activity. Proteolysis is essential for quick adaptation to stress and regulates important developmental processes. Clp-mediated proteolysis is also required for disease progression and virulence of several bacterial pathogens, favouring survival in the host or modulating the activity of genuine virulence factors (Gaillot, 2004). In this study, one ATP-dependent Clp protease proteolytic component was upregulated by bile and acidified bile. In Bacillus subtilis, the deletion mutant of this protein could not grow under several stress conditions, most severely during starvation and at high temperatures (Tomoyasu et al., 2002). Finally, there were several proteins with unknown function that were influenced by bile and acidified bile stress. One of these was H. pylori predicted coding region Hp0721. This protein has been isolated as a sialic acid-specific lectin from H. pylori, which implies that it is involved in sialic acid-specific adhesion (Bennett & Roberts, 2005).

In conclusion, this work focused on the global proteomic changes and possible molecular mechanisms of H. pylori in response to bile and acidified bile stress, and indicated that at least 28 proteins were shown to be influenced by the presence of bile and acidified bile. These proteins are related to various kinds of biological function such as iron storage, chemotaxis, flagellar mobility and the energy metabolism possibly necessary for enhanced flagellar motion required for chemotaxis. These results indicate that H. pylori responds to bile stress through multiple mechanisms involving many signalling pathways. These findings may provide valuable information in understanding the interactions between H. pylori and human bile and acidified bile, as well as the survival mechanism of this bacterium in humans.

This work was supported by the National Natural Science Foundation of China (no. 30770118), the National Basic Research Program of China (973 Program2007CB512001) and the Science Foundation of Shandong Province, PR China (nos Y2004C03 and 2005GG3202087). We thank Jessica Link who corrected the English version of the manuscript.

Footnotes

^,†, Qunye Zhang² †These authors contributed equally to this work.