|Home | About | Journals | Submit | Contact Us | Français|
In Helicobacter pylori the stringent response is mediated solely by spoT. The spoT gene is known to encode (p)ppGpp synthetase activity and is required for H. pylori survival in the stationary phase. However, neither the hydrolase activity of the H. pylori SpoT protein nor the role of SpoT in the regulation of growth during serum starvation and intracellular survival of H. pylori in macrophages has been determined. In this study, we examined the effects of SpoT on these factors. Our results showed that the H. pylori spoT gene encodes a bifunctional enzyme with both a hydrolase activity and the previously described (p)ppGpp synthetase activity, as determined by introducing the gene into Escherichia coli relA and spoT defective strains. Also, we found that SpoT mediates a serum starvation response, which not only restricts the growth but also maintains the helical morphology of H. pylori. Strikingly, a spoT null mutant was able to grow to a higher density in serum-free medium than the wild-type strain, mimicking the “relaxed” growth phenotype of an E. coli relA mutant during amino acid starvation. Finally, SpoT was found to be important for intracellular survival in macrophages during phagocytosis. The unique role of (p)ppGpp in cell growth during serum starvation, in the stress response, and in the persistence of H. pylori is discussed.
Helicobacter pylori is a helical or spiral-shaped gram-negative bacterium. This gastric pathogen infects more than one-half of the world's population. H. pylori infection has been linked to human gastritis, ulcers, and gastric cancer (10, 19, 39). Gastric cancer is the fourth most common cancer and the second leading cause of death from cancer worldwide (38). Thus, understanding H. pylori pathogenesis and factors that affect establishment of infection has public health significance.
The stringent response is a bacterial adaptation which affects global gene expression during nutrient limitation and under other stress conditions. In a well-studied Escherichia coli system, two homologous genes, relA and spoT, are important for the stringent response (13). RelA is a synthetase for guanosine tetra- and pentaphosphate [(p)ppGpp], while in E. coli SpoT is a bifunctional enzyme with both synthetase and hydrolase activities. These two small molecular effectors bind to RNA polymerase (7, 14, 50) and affect global gene expression under nutrient limitation conditions, such as amino acid starvation (13, 17, 51). E. coli relA mutants defective in (p)ppGpp synthesis have a “relaxed” phenotype and growth advantage during amino acid starvation compared to wild-type strains (13, 17, 46, 51), while E. coli spoT mutants defective in the (p)ppGpp hydrolase have a slow-growth phenotype (43). The relA and spoT genes are conserved in eubacteria, implying that the functions of (p)ppGpp are important for the organisms (31). The roles of (p)ppGpp in the survival of several bacterial pathogens during infection and transmission have been described previously (11, 20, 22, 26, 35). Many bacteria have both the relA and spoT genes, but members of the alpha- and epsilonproteobacteria, including H. pylori, have only one member of the RelA/SpoT family of proteins, SpoT (5, 20, 49, 53).
We speculated that SpoT is a potential global transcriptional regulator for H. pylori growth and persistence. H. pylori lives deep in the gastric mucus layer in the human stomach, where it encounters low pH and a constantly changing environment (44). H. pylori infection in humans persists for a lifetime unless it is eliminated by antibiotic treatment. In order to establish an infection and to persist in the stomach, H. pylori must overcome the host innate immune response, including macrophages (1, 9, 32, 41). For optimal in vitro growth, microaerophilic conditions (low levels of O2) and capnophilic conditions (high levels of CO2) are used and media are supplemented with serum (6, 12, 33). However, in vitro the viability of H. pylori is reduced to the noncultivatable level within 7 to 10 days. The cessation of growth is accompanied by a morphological change from a helical shape to a nonhelical shape, including a coccoid form (6). H. pylori lacks sigma S, sigma H, and sigma E, which are typically associated with various stress responses in many gram-negative bacteria; however, spoT is conserved in H. pylori, implying that SpoT and (p)ppGpp are key regulators for various stress responses in this bacterium, including survival in macrophages.
While this study was in progress, it was reported that the H. pylori spoT gene encodes (p)ppGpp synthetase activity and that spoT is required for H. pylori survival in the stationary phase, during exposure to low pH, and during aerobic shock (34, 52). However, the hydrolase activity of the H. pylori SpoT protein has not been determined. Further, no “relaxed” phenotype associated with a spoT mutant during nutrient starvation has been reported. Moreover, the role of spoT in survival during phagocytosis by macrophages is unknown. In this study, we examined SpoT activity in vivo and also the effects of spoT on cell growth under serum starvation conditions and on H. pylori survival in macrophages. Our data provide additional evidence that spoT is important in sensing serum limitation, in infection persistence, and in the pathogenesis of H. pylori.
All E. coli strains used are K-12 derivatives. The relA spoT double null mutant [(p)ppGpp0] and the spoT203 mutant (obtained from Mike Cashel, NIH) have been described previously (43, 55). The general bacterial techniques and media used have also been described previously (30). H. pylori strains J99 (obtained from the American Type Culture Collection [ATCC]), 26695 (obtained from ATCC), G27 (obtained from D. Scott Merrell, Uniformed Services University of the Health Sciences), strain HP1061 (obtained from Paul S. Hoffman, Dalhousie University, Halifax, Nova Scotia, Canada), and SS1 (obtained from A. Lee and J. O'Rourke, University of New South Wales, Sydney, Australia) have been described previously (5, 23, 27, 49, 54). The G27 spoT* strain (obtained from Karen Guillemin, University of Oregon) has also been described previously (34).
Unless mentioned otherwise, H. pylori cells were grown in bisulfiteless brucella broth (BLBB) (24) supplemented with (i) Glaxo selective supplement A (20 μg/ml bacitracin, 1.07 μg/ml nalidixic acid, 0.33 μg/ml polymyxin B, 10 μg/ml vancomycin) (36) and (ii) 10% (vol/vol) fetal bovine serum (FBS; triple sterile filtered, 0.1-μm filter) (SH30088.03HI; HyClone) (referred to below as BLBB with 10% FBS). To make solid medium plates, Difco agar (1.7%, wt/vol) was added to BLBB with 10% FBS described above. Kanamycin (15 μg/ml) or chloramphenicol (4 to 8 μg/ml) was added when necessary. H. pylori cells were cultivated at 37°C in a sealed jar or a humidified incubator with a microaerophilic atmosphere (5% O2, 10% CO2, 85% N2), which was generated with an Anoxomat Mark II microprocessor (www.spirabiotech.com) or was provided from a gas mixture.
H. pylori strains were maintained as frozen stocks at −80°C. A frozen stock was prepared by mixing equal volumes of a fresh 1-day culture (optical density at 600 nm [OD600], ~1) and BLBB supplemented with 10% (vol/vol) FBS and 50% (vol/vol) glycerol. To ensure physiological reproducibility, experiments were started using frozen stocks; thus, the initial cultures were prepared 3 days before experiments were performed. On day 1, about 50 to 100 μl of a frozen stock was placed on a solid medium plate (60 by15 mm), which was then incubated for ~24 h. The next day, the bacteria were transferred evenly onto the entire surface of a new solid medium plate using a moistened cotton swab, and the plate was then incubated for ~24 h. On day 3, the fresh confluent bacterial lawn was collected with a swab and suspended in 1 ml of BLBB without FBS in a 1.5-ml tube, and this preparation was used as the inoculum for the starting cultures after appropriate dilution. In general, a starting culture (25 to 30 ml in a 225-ml cell culture flask) with an OD600 of ~0.05, which was equivalent to ~3 × 107 CFU ml−1, was used for experiments. In each experiment, the wild-type strain and isogenic derivatives of this strain were analyzed in parallel in the same environment (jar) to minimize variables. Cultures were shaken gently at 60 rpm.
Bacterial growth was monitored by using three methods: (i) measurement of the OD600 of cultures using a spectrophotometer, with the growth medium serving as the blank; (ii) determination of the number of CFU by plating serial dilutions of cultures in duplicate on plates, which were then incubated for 3 to 6 days; and (iii) measurement of intracellular ATP levels using a bioluminescent ATP assay kit as described previously (47) (see below).
ATP levels in duplicate cultures were determined using the luciferase-based BacTiter-Glo microbial cell viability assay kit (catalog number G8230; Promega) according to the manufacturer's instructions; the growth medium was used as a blank in these experiments. Cultures were diluted appropriately to ensure that the range of measurements was linear. Light production was measured using a microplate reader (Vector3; PerkinElmer), and the data were expressed in relative light units. Because the wild-type and mutant strains were always compared in the same experiment, determination of absolute ATP concentrations was unnecessary.
Briefly, H. pylori cultures were prepared as described above. On day 3, ~50 μl of an inoculum (~108 cells) was spotted onto a fresh solid medium plate, and the plate was incubated for 3 h. Subsequently, ~50 μl (≥50 ng) of DNA was placed on top of the bacterial spot, and the plate was incubated for another 3 to 5 h. Then the bacteria and DNA mixture were transferred onto the entire surface of a fresh solid medium plate supplemented with the appropriate antibiotic using a moistened cotton swab. The plates were incubated for 3 to 6 days to select for recombinants or transformants, which were purified at least once on selective solid medium plates before frozen stocks were prepared.
Standard molecular biology techniques for DNA purification, cloning, and PCR were performed as described previously (42). Genomic DNA was prepared using a Wizard Genomic DNA purification kit (Promega) according to the manufacturer's instructions. Plasmid DNA was purified from H. pylori as described previously (16). DNA sequencing was performed by the National Cancer Institute Intramural DNA Sequencing MiniCore facility.
DNA fragments containing the spoT gene from different H. pylori strains (genomic DNAs) were generated by PCR and cloned into either the BamHI/PstI sites of vector pQE80L (Qiagen) or the SphI/KpnI sites of the shuttle vector pHel2 (obtained from Rainer Haas ) using standard DNA techniques. The primers used for generating the DNA fragments containing the spoT gene flanked by BamHI and PstI sites were SpoT/BamHI F (5′-GCGGATCCATGAACGAAATTGATAAATC-3′; BamHI site underlined) and SpoT/PstI R (5′-AAAACTGCAGTTATGATTCATAAGCGTCAT-3′; PstI site underlined). The primers used for generating the DNA fragments containing the spoT gene flanked by SphI and KpnI sites were SpoT-SphI-F (5′-ACATGCATGCGATTCGCTGAGAATGTAGG-3′; SphI site underlined) and SpoT-KpnI-R (5′-CGGGGTACCTTATGATTCATAAGCGTCAT-3′; KpnI site underlined). The sequences of the primers were based on the previously published genome sequence of H. pylori strain 26695 (49). The cloned spoT genes were confirmed by DNA sequencing.
A nonpolar spoT null mutation, in which about one-half of the spoT gene encoding amino acid residues 101 to 511 of SpoT was deleted and replaced with an insertion of the nonpolar aphA-3 gene encoding a kanamycin resistance cassette (29), was constructed in several steps. First, a ~4.1-kb DNA fragment containing the spoT null mutation was constructed directly by performing three PCRs without cloning. For the first two PCRs, H. pylori strain J99 genomic DNA was used as the template. In the first PCR, using primer 1 (5′-AATCCCTCACTACATCCTTAAAGAAG-3′) and primer 2 (5′-AGCCATTTATTCCTCCTAGTTAGTCACCTCACAAGGCGTGTCTTCTACCAC-3′), a 1.5-kb DNA fragment (fragment A) was produced. The 5′ end of this DNA fragment covered the JHP715 gene sequence located upstream of spoT (JHP712), and its 3′ end had an internal spoT sequence, followed by translational stop codons, a ribosome binding site, and the beginning of aphA-3 in pUC18K-2 (29). In the second PCR, primer 3 (5′-GAATTGTTTTAGTACCTGGAGGGAATAATGACCGACACTAAGAGCATGATCAATATC-3′) and primer 4 (5′-CACGATTTCACTAGCGAGATCCTC-3′) were used, and a 1.7-kb DNA fragment (fragment B) was produced. The 5′ end sequence of fragment B overlapped the end of the aphA-3 sequence, followed by an extra ribosome binding site which was in frame with the remaining spoT sequence, and the 3′ end of fragment B overlapped the JHP711 gene sequence located downstream of spoT. In the third PCR, the first two PCR products, DNA fragments A and B, as well as pUC18K-2 DNA, were used as templates along with primers 1 and 4, which resulted in a 4.1-kb DNA fragment that contained the spoT null mutation. Finally, H. pylori strains were transformed with the 4.1-kb DNA fragment, and Kanr recombinants were selected. The spoT null mutation in the Kanr recombinants was verified by PCR and sequencing of the genomic locus. The resulting spoT mutant was not able to produce (p)ppGpp when cells were shifted from a rich growth medium to a minimal medium (data not shown), as reported previously for another spoT mutant (34).
Bacterial cells were fixed on glass slides by heat, followed by Gram staining (56). The slides were examined using a Zeiss Axiophot II microscope, and images were captured with a charge-coupled device camera. For each condition and/or time point, more than 100 cells were scored. Due to the limited resolution, only helical and nonhelical forms, including coccoid forms, were categorized.
Murine macrophage cell line RAW 264.7 was obtained from the ATCC and grown in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) supplemented either with only 10% FBS or with Glaxo selective supplement A plus 10% FBS. This cell line was cultured in a humidified incubator with 5% CO2 and 95% air at 37°C.
An invasion assay and an intracellular viability bacterial assay were performed as described previously (18), with some modifications. One day before invasion assays were performed, RAW 264.7 cells were seeded in 96-well tissue culture plates to obtain a concentration of ~2 × 104 macrophages per well. The bacterial cells were prepared similarly to the inoculum used for the starting cultures described above, except that DMEM containing 10% FBS was used for the final dilution. In each independent experiment, 3 to 12 wells were used for each bacterial strain tested. The monolayers of macrophages in the wells were infected with 0.1 ml of a chilled bacterial suspension (OD600, ~0.025; CFU ~1.5 × 106 CFU). Immediately after infection, the plates were centrifuged at 4°C for 4 min at 600 × g to synchronize bacterial contact with the monolayers (4), and this was followed by three washes with cold DMEM. After centrifugation and washing, 0.1 ml of DMEM containing 10% FBS at 37°C was added (designated time zero), and the infected macrophage monolayers were incubated at 37°C in a humidified incubator with 5% CO2 and 95% air. After 1 h, the infected monolayers were washed once with DMEM and then were incubated in 0.1 ml of DMEM containing 10% FBS and 100 μg of gentamicin/ml for 1 h to kill the extracellular bacteria but not macrophages. The infected monolayers were then washed once with phosphate-buffered saline (PBS) or DMEM and then either lysed (2-h time point) or incubated in 0.1 ml of DMEM containing 10% FBS for a total of 24 h (24-h time point). To lyse the macrophage monolayers and release H. pylori, 0.1 ml of H2O was added to each well after it was washed with PBS or DMEM either three times (time zero) or once (other time points). Finally, the numbers of viable bacteria (CFU) in the macrophage lysates were determined by plating serial dilutions on solid plates as described above.
Live H. pylori cells within macrophages were stained using a BacLight kit (Invitrogen) according to the manufacturer's instructions. Bacterial invasion and intracellular survival assays were performed essentially as described above, with some modifications. Twenty-four-well tissue culture plates with clean glass coverslips (12 mm) placed in the bottom of wells were used. About 1 × 105 RAW 264.7 macrophages (seeding) and 0.5 ml H. pylori cells (OD600, ~0.025; ~7.5 × 106 CFU) in DMEM containing 10% FBS per well were used for infection. At the 2-h (at the end of gentamicin treatment) and 24-h time points, the medium was removed from each well by aspiration, and the wells were rinsed once with PBS. The macrophage cells were permeabilized with 0.5 ml of saponin (1 mg/ml; freshly made in PBS) at room temperature for 10 to 15 min. After the saponin treatment, the solution was removed very carefully (macrophages were fragile following this treatment), and 0.25 ml of diluted BactLight solution was added to each well to stain the viable bacteria. After incubation at room temperature for 15 min, we removed the coverslips from the wells and placed each coverslip on a glass slide for visualization. The slides were examined using a Zeiss Axiophot II microscope equipped with epifluorescence filters. Images were captured with a charge-coupled device camera. The images were processed with Adobe Photoshop.
H. pylori has the spoT gene but no relA gene. This suggests that (p)ppGpp synthesis and degradation are mediated solely by SpoT in H. pylori. We wished to determine whether the H. pylori spoT product has (p)ppGpp hydrolase activity in addition to its known (p)ppGpp synthetase activity. We cloned the spoT gene from several H. pylori strains (G27, J99, 1061, SS1, and 26695) into either the expression vector pQE80L or the shuttle vector pHel2 (25). The resulting spoT clones were tested to determine their abilities to complement known E. coli relA and spoT mutations, which has been used previously as a way to define the (p)ppGpp synthetase and hydrolase activities of the products of spoT genes from different bacteria in vivo (15, 21, 28). All spoT clones complemented the defective functions in the E. coli mutants (Fig. (Fig.11 and data not shown). Sequence analysis of these clones revealed only a few minor variations in them (data not shown); therefore, the results obtained for a representative clone, pHPspoT (pQE80L-spoT clone from strain 1061), are presented below.
The E. coli relA spoT double null mutant was used as a host to test the (p)ppGpp synthetase activity of H. pylori SpoT in vivo. This E. coli double null mutant does not grow on minimal medium due to the absence of (p)ppGpp in the cell (55). The pHPspoT clone complemented the double null mutation, allowing the host to grow on minimal medium. As a control, we showed that the cloning vector pVector (pQE80L) did not complement the mutation (Fig. (Fig.1A).1A). Our results are consistent with the results of a previous report that showed that the H. pylori SpoT protein has (p)ppGpp synthetase activity (34). However, in the previous study the authors observed only partial complementation of the E. coli double null mutant by the H. pylori spoT construct used in their study. The differences may be explained as follows: the spoT clones used in this study encode the full length of SpoT (~775 amino acid residues), whereas the spoT construct used in the previous report lacked the last 25 amino acid residues in the C terminus of SpoT (34).
To determine if the H. pylori SpoT protein has (p)ppGpp hydrolase activity, we took advantage of the mutant E. coli spoT203 mutant gene, which encodes a SpoT mutant protein defective in (p)ppGpp hydrolase activity without abolishing the (p)ppGpp synthetase activity. The spoT203 mutation results in a ~10-fold-higher level of (p)ppGpp in the cell and, because this high level is toxic, a slow-growth phenotype (43). The growth defect of the E. coli spoT203 mutant was abolished by clones expressing the H. pylori spoT gene but not by the vector alone (Fig. (Fig.1B).1B). These complementation results demonstrated that H. pylori SpoT does have both (p)ppGpp hydrolase and synthetase activities in vivo.
The domains for the two enzyme activities associated with SpoT are highly conserved in bacteria (31). H. pylori SpoT is more similar to SpoT of Campylobacter jejuni (~39% identity), a closely related bacterium, than to E. coli SpoT (~27% identity). Results of this study (see below) and other studies (34) indicate the importance of spoT in regulation of H. pylori growth during serum starvation and in stress responses, which are consequences of stationary growth or phagocytosis. It is expected that both the synthetase and hydrolase activities of SpoT are sensitive to environmental cues. Recently, it was reported that the SpoT-dependent stress response is linked to fatty acid metabolism in E. coli (8). It is plausible that SpoT behaves similarly in H. pylori. The challenge is to identify the signals which modulate the SpoT activities and the mechanism by which (p)ppGpp regulates gene expression in H. pylori.
E. coli mutants defective in (p)ppGpp synthesis have a “relaxed” phenotype during amino acid starvation; therefore, we hypothesized that a spoT mutant of H. pylori might have a growth advantage over the wild-type strain under certain nutrient-limiting conditions. Brucella broth supplemented with FBS has often been used for culturing H. pylori (33). We used BLBB supplemented with FBS to obtain optimal growth of H. pylori (24). The basis of the requirement for serum for optimal or enhanced H. pylori growth in vitro is unknown. Components other than bovine serum albumin present in the serum have been suggested to be the factor that enhances the growth of H. pylori (48). The serum requirement for H. pylori growth prompted us to test whether a spoT null mutation confers a growth advantage during serum starvation.
For this study we chose the H. pylori G27 strain (54), because this strain has been studied by several laboratories and could serve as a prototype H. pylori strain for detailed genetic and physiology studies. We constructed a G27 nonpolar spoT null mutant (G27ΔspoT) using a method that eliminated an intermediate cloning step as described in Materials and Methods. In addition, a “complemented” or spoT* strain (G27ΔspoT rdxA::Pure::HP0775) was included for comparison because Pure-driven spoT (HP0775) expression has been reported to be higher than wild-type spoT expression in the G27 background (34), although the extent of the increase in SpoT expression in this strain has not been determined. To determine whether the spoT null mutation provides a growth advantage in serum-free BLBB, we compared the growth of the spoT mutant with the growth of the wild-type and spoT* strains in either BLBB with no FBS or BLBB with 10% FBS using three different measurements, OD600, intracellular ATP levels, and CFU.
After initial brief growth in serum-free BLBB, the wild-type and spoT* strains stopped growing when the OD600 reached ~0.2 at 48 h. However, the spoT mutant reproducibly continued to grow until the OD600 reached ~0.8 at 48 h (Fig. (Fig.2A).2A). This finding is in contrast to the growth curves for cells grown in BLBB with 10% FBS, which indicated that there were no differences among the spoT mutant, wild-type, and spoT* strains. Consistent with the OD600 values, the peak ATP levels were reproducibly approximately threefold higher in the spoT mutant than in the wild-type and spoT* strains at 48 h in the serum-free medium (Fig. (Fig.2B).2B). The ATP level in the spoT mutant, however, rapidly decreased after the peak was reached and was close to the basal level at 72 h, a pattern similar to the pattern observed for the same mutant grown in serum-supplemented medium. In the wild-type and spoT* strains grown in the serum-free medium, however, the overall ATP levels were low and remained low with no apparent peak at 48 h. In contrast, the ATP levels were high and peaked at 48 h for both the wild-type and spoT* strains in the serum-supplemented medium. The ATP levels in the wild-type and spoT* strains were ~15-fold lower in the serum-free medium than in the serum-supplemented medium at 48 h.
Determination of the viable bacterial counts confirmed that the number of CFU ml−1 of the spoT mutant was consistently approximately threefold higher than the numbers of CFU ml−1 of the wild-type and spoT* strains at 48 h in the serum-free medium (Fig. (Fig.2C),2C), in close agreement with the data for the OD600 and ATP levels. At 72 h and later time points, however, the number of CFU ml−1 was lowest for the spoT mutant, and this was accompanied by a rapid decline in the ATP levels (Fig. (Fig.2B).2B). These results are consistent with the report that the spoT mutation reduced the survival rate in the stationary phase (34), apparently independent of the presence of serum in the medium. Thus, our results showed that the spoT mutant exhibits a “relaxed” growth phenotype in a serum-free medium. To the best of our knowledge, this is the first reported “relaxed” phenotype associated with a spoT mutant in H. pylori.
Overall, our data suggested that one of the functions of SpoT or (p)ppGpp in H. pylori is to limit cellular metabolic activity and restrict cell growth during serum starvation. This function of SpoT appears to be a general function because the growth advantage of the spoT mutant not only is observed in serum-free BLBB, which is a complex medium, but also is apparent in a serum-free chemically defined Ham's F-12 nutrient mixture, which has been reported to be able to support limited growth of H. pylori (48). Our preliminary results showed that the number of CFU ml−1 of the spoT mutant was also approximately threefold higher than the number of CFU ml−1of the wild-type strain at 48 h in a serum-free F-12 medium (data not shown). How SpoT senses serum starvation is not known at present. It has been reported that expression of spoT in Borrelia burgdorferi is induced during serum starvation (15). We speculate that a spoT mutation in B. burgdorferi might also stimulate cell grow in a serum-free medium. The mechanism by which SpoT and (p)ppGpp regulate H. pylori growth in response to serum starvation needs to be studied further.
We also monitored the morphology of H. pylori cells grown in the serum-free medium by using microscopy. Images of more than 100 cells at different time points during bacterial growth were analyzed, and cell shapes were categorized as either helical or nonhelical, including coccoid forms. Interestingly, the morphological conversion of wild-type cells from a helical form to a nonhelical form was significantly delayed during serum starvation compared to the cells grown in BLBB with 10% FBS (Fig. (Fig.3A).3A). Almost all (≥95%) of the exponentially growing cells (up to 24 h) in either a serum-free medium or serum-supplemented BLBB were helical (data not shown). After 24 h, the conversion to nonhelical cells was reproducibly faster for cells grown in serum-supplemented medium than for cells grown in serum-free medium. For example, at 48 h the values were ~31% and ~4% for the wild-type strain grown in the serum-supplemented and serum-free media, respectively. At 72 h, however, the values were ~97% and ~21%, respectively. Consistent with these results, the numbers of CFU ml−1 between 48 and 72 h were similar for the cells grown in the serum-free medium; however, the number of CFU ml−1 at 72 h was about fivefold lower than the number of CFU ml−1 at 48 h when cells were grown in BLBB with 10% FBS (Fig. (Fig.2C).2C). This trend was also observed for both the spoT mutant (Fig. (Fig.3B)3B) and the spoT* strain (Fig. (Fig.3C).3C). Compared to the wild-type strain, however, the spoT mutant initiated the morphological transformation prematurely, as reported previously (34). Most impressively, however, the vast majority (~88%) of the spoT* cells, which had higher SpoT activity than the wild-type cells, were still helical even at 96 h in the serum-free medium (Fig. (Fig.3C).3C). It is likely that the higher SpoT activity in spoT* cells amplifies the function of SpoT during serum starvation.
Our results suggest that serum starvation restricts H. pylori growth due to the SpoT function and consequently prolongs the time that the bacterial cells are helical. Because helical H. pylori cells are consistently found in chronically infected hosts, it is tempting to speculate that H. pylori growth in vivo is minimal or restricted due to nonoptimal growth conditions in the human stomach. The majority of H. pylori cells are located deep in the gastric mucus, close to the surface of the epithelium (44). It has been suggested that tight junctions between epithelium cells in healthy gastric tissue are unlikely to leak significant amounts of serum, thus making the natural habitat of H. pylori a serum-free microenvironment (47). Thus, SpoT may play an important role in restricting cellular metabolic activity and limiting H. pylori growth during serum starvation in order to maintain cell vitality under these poor growth conditions.
H. pylori has developed multiple mechanisms to evade elimination by host innate immune responses, including macrophages, which may account for the persistence of this bacterium in the host (1, 40). Once inside macrophages, H. pylori interferes with phagosome maturation, which leads to the formation of large phagosomes called megasomes. H. pylori resides in these megasomes during phagocytosis (2-4). The megasomes apparently do not fuse with the lysosomes in macrophages, thus protecting the bacteria from elimination. It is conceivable that ingested H. pylori cells following phagocytosis by macrophages likely invoke a bacterial cell stress response as a survival mechanism. H. pylori virulence factors, VacA, urease, and type IV secretion components have been reported to be important for bacterial survival in macrophages (37, 40, 45, 57).
We wished to determine whether spoT plays a role in the survival of H. pylori in macrophages. We investigated the effect of the spoT mutation on intracellular survival by determining the number of CFU ml−1 24 h after phagocytosis, as previously described (4, 18). The number of surviving spoT mutant cells (CFU ml−1) in macrophages after 24 h was significantly lower than the numbers of surviving cells of the wild-type and spoT* strains (Fig. (Fig.44).
To determine whether the spoT mutant was invasion defective, which could contribute to the apparent reduced survival, we determined the numbers of CFU ml−1 of the infected bacteria at several time points after phagocytosis by macrophages. While the level of survival of the spoT mutant was significantly lower than the levels of survival of the wild-type and spoT* strains at 24 h after infection, the number of spoT mutant cells that could be recovered from infected macrophages was essentially the same as the numbers of wild-type and spoT* colonies recovered 2 h after invasion (Fig. (Fig.4).4). Note that the 2-h samples were samples that were obtained after gentamicin treatment to kill extracellular bacteria; therefore, the CFU represented the intracellular bacteria that had invaded. Thus, our data demonstrated that the ability of the spoT mutant to invade macrophages appears to be normal, but its ability to survive in the macrophages is defective. Consistent with these results, microscopic imaging of live H. pylori cells showed that while the numbers of intracellular bacteria per macrophage were comparable for the spoT mutant, wild-type, and spoT* strains at 2 h after infection, the numbers of spoT mutant cells were significantly reduced after 24 h compared to the numbers of wild-type and spoT* strain cells (data not shown).
Our results demonstrated that SpoT or (p)ppGpp plays an important role in bacterial evasion of elimination by macrophages. At present, the mechanism(s) by which SpoT or (p)ppGpp mediates the survival of the bacterium in macrophages is unknown. It may be that SpoT or (p)ppGpp affects a pathway(s) different from the pathways mediated by either VacA or urease, because the activities of the VacA and urease proteins in the spoT mutant were not significantly different from the activities of these proteins in the isogenic wild-type strain (34; unpublished results). Currently, we are studying the effects of SpoT on the interactions between H. pylori and macrophages.
In summary, in this study we investigated the function of the H. pylori SpoT protein in vivo and its roles in cell growth under serum starvation conditions and in bacterial survival in macrophages. Our results showed that the H. pylori spoT gene encodes a bifunctional enzyme for the synthesis and degradation of (p)ppGpp. Also, while wild-type H. pylori maintains helical morphology and cell growth is restricted during serum starvation, the spoT mutant has a growth advantage over the wild-type strain in a serum-free medium and exhibits a “relaxed” growth phenotype. Moreover, SpoT is critical for intracellular survival in macrophages. Thus, our results demonstrate that H. pylori SpoT or (p)ppGpp regulates cell growth in response to serum starvation and plays a critical role in the persistence of this bacterium.
We thank the leadership of the Gene Regulation and Chromosome Biology Laboratory and the National Cancer Institute Center for Cancer Research for the support in the initiation of the H. pylori project in D.J.J.'s laboratory, and we thank colleagues in the H. pylori research community and in our laboratories for helpful discussions. We also thank Scotty Merrell, Karen Guillemin, Kevin Bourzac, Rainer Haas, Mike Cashel, and Lee-Anne Allen for providing strains, plasmids and/or protocols. Y.N.Z. thanks Susan Gottesman for her support. We are grateful for helpful comments on the manuscript from Don Court, Susan Gottesman, Mikhail Kashlev, Peter McPhie, Scotty Merrell, Herbert Tabor, and Julie Torruellas Garcia.
This research was supported by the Intramural Research Program of the NIH National Cancer Institute Center for Cancer Research, by the Intramural Research Program of the National Institute of Digestive Diseases, Diabetes and Kidney Diseases, and by an intra-agency agreement with the NIH National Center on Minority Health and Health Disparities.
Published ahead of print on 3 October 2008.