Helicobacter pylori is a Gram-negative bacterium, a member of the Epsilon proteobacteria that colonizes the human gastric mucosa and is responsible for causing peptic ulcers and gastric carcinoma (Marshall and Warren, 1984; Parsonnet et al., 1991; Uemura et al., 2001). H. pylori survives in the hostile environment found in the stomach, which is partially attributed to the expression of virulence factors, such as secretion systems, cytotoxins, flagella, and adhesins. Unlike other Gram-negative bacteria such as Escherichia coli or Salmonella enterica, the H. pylori genome encodes only few known transcriptional regulators, which control expression of genes involved in bacterial metabolism and pathogenicity. This limited repertoire is likely due to its life style highly adapted to one particular niche, the human gastric mucosa. H. pylori strain 26695 has a small genome of 1.67 Mb (Tomb et al., 1997), and possesses three genes that code for sigma factors: rpoD (σ⁸⁰), rpoN (σ⁵⁴), and fliA (σ²⁸). σ⁸⁰ is a homolog of Gram-negative vegetative sigma factors responsible for the transcription of housekeeping genes (Tomb et al., 1997; Beier et al., 1998), whereas σ⁵⁴ and σ²⁸ are two alternative sigma factors dedicated mostly to control expression of flagella components (Fujinaga et al., 2001; Josenhans et al., 2002; Niehus et al., 2004). The response regulator FlgR is also involved in regulation of flagella synthesis (Spohn and Scarlato, 1999b), whereas bacterial chemotaxis is controlled by the CheY protein (Foynes et al., 2000; Terry et al., 2005). Master regulators of response to metals such as Fur, NikR, and CrdR activate or repress genes in the presence of iron, nickel or copper, respectively (Contreras et al., 2003; Waidner et al., 2005; Pich and Merrell, 2013). Environmental cues such as acid pH and temperature influence expression of HrcA, HspR, and ArsR regulatory proteins (Spohn and Scarlato, 1999a; Spohn et al., 2004; Pflock et al., 2005). Although many microarrays analysis have been published (Ang et al., 2001; Merrell et al., 2003a,b; Thompson et al., 2003; Wen et al., 2003; Bury-Mone et al., 2004; Kim et al., 2004), little is known about the effects of environmental cues on the expression of H. pylori regulatory genes, including some poorly investigated transcriptional regulators.

In this work, we determined the expression profile of the transcriptional repertoire of H. pylori strain 26695 under several environmental conditions relevant for adaptation to its particular ecological niche of the human stomach, such as acidic pH, urea, nickel, and iron. In addition, we analyzed the expression of regulatory genes in biofilm formation and in the presence of AGS gastric epithelial cells. Finally, we studied the effect of the antibiotics kanamycin, chloramphenicol, and tetracycline on the transcription of regulatory genes. Our study describes the transcriptional expression of H. pylori regulatory genes in response to different environmental conditions.

H. pylori is a highly specialized bacterium that is exclusively found in the human gastric mucosa. In this work, we describe the expression of H. pylori transcriptional regulatory genes under different environmental conditions. Most transcription factors were highly expressed when H. pylori reached the early stationary phase. At this growth phase, H. pylori is exposed to specific stress signals such as pH changes, starvation, reactive oxygen species that would activate its transcriptional repertoire, suggesting that the stationary phase may mimic the conditions found by the bacteria in the host. Acid pH, the presence of urea, nickel, or iron are environmental cues required for optimal adaptation of H. pylori to its natural niche. Whereas all of the above conditions boosted fliA expression, rpoD showed only a mild increase in transcription after bacterial exposure to acid pH, urea, and nickel, and a decrease in response to iron. The expression of rpoN significantly increased upon treatment of bacteria with urea or nickel; RpoN was initially described as regulator of flagellar genes, but recent studies show that it also controls several bacterial regulatory processes involved in energy metabolism, biosynthesis, protein fate, oxidative stress, and virulence (Sun et al., 2013). fliA was strongly expressed even under environmental conditions that are not common inducers of flagella synthesis, but are known to affect other bacterial components or pathways. In fact, FliA regulates expression of outer membrane proteins, lipopolysaccharide synthesis, DNA restriction, and CagA (Josenhans et al., 2002; Baidya et al., 2015), a protein involved in virulence and associated with the development of gastric carcinoma (Ohnishi et al., 2008). Our data support the notion that FliA regulates different bacterial functions other than the flagellum.

H. pylori is exposed to changes in pH while passing from the stomach lumen through the mucus layer to interact with epithelial cells, and this pH gradient is used by the bacteria for spatial orientation (Schreiber et al., 2004). Accordingly, changes in pH strongly affect expression of transcriptional regulators that control genes involved in colonization and persistence in the human host. Our data show that an acidic pH repressed arsR, hp0564, and flgR, while stimulating hup, cheY, and crdR. The hup gene codes for the HU nucleoid protein, which has a regulatory role in the response to acid stress in H. pylori. Thus, hup mutants are less viable than wild type bacteria at pH 5.5 and during stomach colonization due to down regulation of both urease (ureA) and arginine decarboxylase (speA) in the absence of HU nucleoid protein (Wang et al., 2012; Almarza et al., 2015). CheY and CrdR response regulators are also crucial for a successful colonization of the animal stomach (Foynes et al., 2000; Panthel et al., 2003; McGee et al., 2005; Terry et al., 2005). CheY expression is essential for the chemotactic motility required to reach and colonize the gastric epithelia, and is likely to be triggered in the acidic milieu of the stomach lumen. In contrast, CrdR has not been shown to be involved in the regulation of gene expression in response to acidic pH (Pflock et al., 2007b), although copper can be present in the acidic environment of the human stomach, and regulation of its uptake is important for keeping the balance between supplying copper as respiration co-factor, and avoiding copper-induced toxicity (Haley and Gaddy, 2015). Interestingly, the master regulator of the acid response arsR was repressed in acidic pH, which confirms reports that ArsR may also act as transcriptional auto-repressor under an acidic pH (Dietz et al., 2002).

Urea, nickel, and iron are crucial for H. pylori pathogenesis and they control regulatory networks responding to their presence. We found that expression of most of the transcription regulators tested increased when bacteria were exposed to urea, nickel, or iron. Nickel serves as essential co-factor for the urease enzyme, which enables H. pylori survival at acidic pH (Khan et al., 2009). The up regulation of nikR expression that we observed contrasted with the auto-negative regulation reported for the nikR promoter (Delany et al., 2002; Contreras et al., 2003). The conditions of growth (stationary phase) and nickel concentrations (250 μM) that we tested resulted in increased nikR expression. However, nikR expression showed slight variations in response to low (1 μM) and high (100 μM) concentrations of nickel (Davis et al., 2006), suggesting that nickel may modulate nikR transcription in a concentration-dependent manner. Interestingly, rpoD and fur were down regulated in the presence of iron. While iron-mediated fur repression can be explained by the negative auto-regulation of this transcription factor upon iron-binding (Delany et al., 2002), the decrease in rpoD levels is hard to explain. Whilst the evaluation of each environmental condition provides relevant information about H. pylori physiology, the combination of these stimuli could better mimic the in vivo response of the bacteria in the infection context.

One of the strategies employed by H. pylori to persist and colonize the stomach is biofilm formation. Analysis of bacteria grown in biofilm revealed an interesting expression pattern of the three sigma factors: whereas rpoN was not affected, expression of rpoD and fliA increased during biofilm formation. FliA has been found to control the lpxC gene, which is involved in the early steps of lipid A synthesis in H. pylori (Josenhans et al., 2002). The marked increase of fliA expression that we found in sessile, aggregated bacteria is in agreement with reports about the effect of lipid A architecture on biofilm formation (Gaddy et al., 2015). In addition, with the exception of hrcA, expression of all transcription factor genes studied increased during biofilm formation, and the relative increase of several of them was the highest increase observed across all the conditions tested. This remarkably activated state of the regulatory transcriptome highlights the importance of forming sessile microbial communities in H. pylori ecology.

Similar to the response during biofilm formation, the presence of gastric epithelial cells significantly increased expression levels of several transcription factors, except for fur, nikR, and crdR. Expression of these three master regulators of virulence remained unaffected in our AGS cell model, which correlates with the previously reported lack of activation or repression of these regulators after the interaction with gastric epithelial cells (Kim et al., 2004).

It has been hypothesized that the reduced number of transcriptional regulators in the H. pylori genome has been compensated by gain of functions in the remaining transcription factors, as compared to their functions by homologs found in other species (Ernst et al., 2005a). For instance, the H. pylori Fur protein was not only found to be involved in iron homeostasis, but it also participated in several other additional pathways including those of oxidative stress resistance (Harris et al., 2002) and acid regulation (Bijlsma et al., 2002; van Vliet et al., 2004), and has been found essential for colonization of the gastric mucosa (Bury-Mone et al., 2004). Moreover, unlike Fur homologs in other species, H. pylori Fur has been found to mediate gene regulation even in its iron-free (apo) form (Bereswill et al., 2000; Delany et al., 2001; Carpenter et al., 2013). Interestingly, whereas most conditions tested here showed only moderate effects on Fur expression, biofilm formation resulted in a marked up regulation of the gene, suggesting functions beyond regulation of iron metabolism.

The presence of antibiotics can alter the expression of genes related to the bacterial stress and virulence on a transcriptional level. Interestingly, most regulatory genes were repressed in response to antibiotic treatment. rpoN, fliA, flgR, and crdR genes presented a negative regulation profile in the presence of kanamycin, chloramphenicol, and tetracycline. In contrast, expression levels of rpoD and hup were highly stimulated under chloramphenicol or tetracycline treatment. These last antibiotics inhibit bacterial translation, differentially affecting the 50S and 30S ribosomal subunits, respectively. The molecular mechanisms responsible for the regulation in expression of transcriptional regulators in the presence of antibiotics have been poorly studied. About this, 16S rRNA expression was affected in the presence of kanamycin and chloramphenicol, showing that this gene was not completely stable and that antibiotic treatment may have affected the expression of this reference gene under these conditions. For a better analysis in presence of these antibiotics, it is necessary the selection and validation of other reference genes for qRT-PCR normalization as was recently reported (Martins et al., 2017).

During analysis of Helicobacter sequences we found that the transcriptional regulators were highly identical among H. pylori species. Interestingly, H. acynonichis and H. cetorum grouped together with H. pylori, corroborating the close phylogenetic relation between these species. The transcriptional regulators HP0222 and HP0564 appear to be conserved in H. pylori and its closely related species, while they were absent in most of the remaining Helicobacter species. Since H. pylori, H. acynonichis, and H. cetorum are all found within mammalian stomachs, these two regulators may confer an adaptive advantage in this particular ecological niche. In line with these findings, expression levels of both, hp0564 and hp0222 increased in contact with AGS gastric epithelial cells, corroborating a report by Kim et al. (2004) on hp0222. However, we did not observe enhanced hp0222 expression under acidic pH, contrasting with the report by Ang et al. (2001).

Recently, we reported the transcriptional profiling of type II toxin-antitoxin genes under different environmental conditions (Cardenas-Mondragon et al., 2016). The type II antitoxins function as transcriptional repressors of their own expressions (Yamaguchi and Inouye, 2011) and also regulate the expression of other genes related with cellular functions such as biofilm formation, persistence, and the general stress response (Wang and Wood, 2011; Hu et al., 2012). Our findings here expand the transcriptional repertoire of H. pylori to respond to the different stresses found in the stomach niche.

In summary, our data show that the repertoire of transcriptional regulators of H. pylori possesses a functional plasticity needed to respond to different environmental cues and to integrate them for the survival and persistence of this bacterium in the stomach niche.

Conceived and designed the experiments: MDC. Performed the experiments: MDC, KvB, MA, LP, JM-C, HV-S, and CJ-G. Analyzed the data: MDC, KvB, and MA. Wrote the paper: MDC, KvB, and JT.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.