Interaction Between the Quorum Sensing and Stringent Response Regulation Systems in the Enterohemorrhagic Escherichia coli O157:H7 EDL933 Strain

Abstract

Quorum sensing and the stringent response are well-known regulation systems for the expression of virulence genes in enterohemorrhagic Escherichia coli (EHEC). However, how these two systems interact is not well known. E. coli strains with mutations in two regulation systems, ${\Delta}luxS$ (ECM101) and ${\Delta}luxS{\Delta}relA{\Delta}spoT$ (ECM201), and the ${\Delta}luxS$ complement strain to ECM201 (ECM202) were created from EHEC O157:H7 EDL933 to investigate how the regulatory systems interact. The phenotypic changes of the mutant strains were characterized and compared with the wild type. The mutant strains exhibited no obvious growth defects, although acid resistance and cellular cytotoxicity were decreased significantly in all the mutant strains. Phenotypic characterization revealed that mutations in the stringent response system (ECM201 and ECM202) influenced the metabolic (defective utilization of arabinose and L-sorbose) and enzymatic activities (decreased trypsin activity, and increased ${\alpha}$-glucosidase activity). In contrast, the quorum sensing system mutant (ECM101) did not display these phenotypes. The motility of the quorum sensing system mutant (ECM101) was unchanged, but mutation in the stringent response system influenced the motility. Our results suggest that quorum sensing interacts with the stringent response regulation system.

Introduction

Outbreaks of enterohemorrhagic Escherichia coli (EHEC) O157:H7 in humans were first reported in 1983 [15, 22, 31]. Since then, EHEC has been recognized as an important foodborne pathogen that causes hemorrhagic colitis and hemolytic uremic syndrome [4, 20]. EHEC strains express Shiga toxins (Stx), cause attaching and effecting lesions on epithelial cells, and possess an ca. 60 MDa plasmid [20].

Quorum sensing is a mechanism of cell-to-cell signaling involving the production of autoinducers (AIs), which are hormone-like compounds [25]. When they reach a certain concentration threshold, they interact with bacterial regulatory proteins to regulate gene expression [24]. Specifically, AI-2 and -3, which mediate quorum sensing signaling, have been extensively studied in EHEC [19, 29]. In addition, LuxS is responsible for the production of AI-2 [5], and is indirectly involved in the production of AI-3 [16, 25, 30].

Bacterial cells encounter various environmental stresses, and make appropriate adjustments to ensure survival [1, 3, 21]. The stringent response is a stress response that is important for bacterial survival under conditions of stress, including amino acid starvation, fatty acid limitations, lack of iron, heat shock, and low pH [3, 27]. Stringent response signaling is achieved by the accumulation of the effector nucleotides, guanosine tetra- and pentaphosphate (ppGpp and pppGpp) [21], of which ppGpp is the predominant and more stable of the two species. Two enzymes, RelA and SpoT, regulate the levels of ppGpp. RelA is the major synthetase of ppGpp under stressful conditions, whereas SpoT is a bifunctional enzyme with weak synthetase activity of (p)ppGpp, and hydrolase activity for degrading (p)ppGpp [11]. As such, ΔrelA spoT mutants are unable to produce (p)ppGpp and have multiple defects, including auxotrophies for several amino acids [32].

The quorum sensing and stringent response systems are known to regulate gene expression in various bacteria [20]. Previous studies showed that the luxS mutation can cause global changes in the expression of many genes that are involved in biochemical and metabolic cellular processes [12, 14, 23, 24, 25, 28, 29]. Other studies demonstrated that relA and spoT mutations in bacterial species, including Yersinia and Enterococcus, caused growth defects compared with wild type [1, 26]. However, how these two systems interact remains unclear. We therefore assessed the interaction between these systems in EHEC O157:H7 EDL933.

 

Materials and Methods

Strains, Media, and Growth Conditions

Bacterial strains and plasmids used in this study are listed in Table 1. All cultures were grown at 37℃ on Luria-Bertani agar plates (BD, Sparks, USA), and aerobically at 37℃ with shaking at 200 rpm in LB broth. To confirm the mutations in the stringent response system, all strains were grown at 37℃ in 1–50% LB medium, giving a minimum concentration of 5% LB. In this study, we therefore used both standard and 5% LB medium for studies of bacterial growth, acid resistance, cell cytotoxicity, and motility (Supplementary data, Tables S1–S5). Antibiotics were used at the following concentrations: ampicillin (Sigma, USA, 100 μg/ml), kanamycin (Sigma, 50 μg/ml), and chloramphenicol (Sigma, 24 μg/ml).

Table 1.Bacterial strains and plasmids used in this study.

Construction of Deletion Mutants and the Complementation Strain

An isogenic ΔluxS mutant lacking the entire structural luxS gene was constructed using a one-step inactivation method [9, 18]. A linear DNA fragment harboring the kanamycin-resistant gene kan and homologous luxS-flanking sequences was amplified using the gene-specific primers LuxS-kan-F and LuxS-kan-R, as shown in Table 2, and the pKD13 plasmid as the template. The fragment was introduced into EDL933 competent cells by electroporation (GenePulser Xcell; BIO-RAD Laboratories, Hercules, CA, USA), and transformants were selected on LB agar plates containing kanamycin. The kan gene was then removed by recombination using ELP recombinase. The deletion of luxS was confirmed by PCR, using luxS-specific primers (Fig. 1). For the stringent response mutant, the two donor strains S17-λ pir (pCVD442/ΔrelA::aphA-3) and S17-λ pir (pCVD442/ΔspoT::cat) were a generous gift from Jang W. Yoon at Seoul National University (J.W. Yoon and J.B. Kaper, unpublished). A triple mutant (ΔluxSΔrelAΔspoT) was then constructed using the quorum sensing mutant (ΔluxS) as a recipient by conjugation, as described previously [10]. All donor and recipient strains were grown to the late exponential phase, washed in LB, and mixed at a 1:10 ratio. Mating mixtures were resuspended in 50 μl of LB, and plated on to LB agar plates. After incubation at 37℃ for 12 h, they were resuspended by vortexing, and diluted in liquid medium. Transconjugants were then selected on LB plates supplemented with the appropriate antibiotics [8]. For the construction of the complementation strain, an isopropyl-β-D-thiogalactopyranoside-inducible expression plasmid carrying the intact luxS gene (pEXP5-CT/luxS) was created, and transformed into the mutant strain ECM201. Expression of luxS was then induced using 10 mM IPTG.

Table 2.Primers used in this study.

Fig. 1.Genetic mutation of the luxS gene in enterohemorrhagic E. coli O157:H7 EDL933. PCR products were separated on 1.5% agarose gel, and the deletion of luxS was confirmed by PCR using gene-specific primers adjacent to luxS. Lane M = 1.0 kb plus DNA ladder; lane 1 = EDL933 (wild type strain); lane 2 = ECM101 (EDL933 ΔluxS).

Bacterial Growth Test

The wild-type and mutant strains were grown aerobically at 37℃ with shaking at 200 rpm in LB broth. Bacterial growth was then monitored every hour by measuring the optical density at 600 nm (OD600). At least three independent experiments were performed, and the data are presented as the mean ± standard error (SE). Viable cell counts were calculated after serial dilution in 0.85% saline and plating on Tryptic Soy Agar (BD, Sparks, USA) plates [2].

Acid Resistance Assay

The acid resistance of bacteria was assessed by inoculating 30 ml of LB broth with a 1:1,000 dilution of cultures at pH 7.0, 2.5, and 1.5, and incubating for 8 h. After incubation for 2 h, the colony-forming units (CFU) of the cultures were counted [6].

Carbohydrate Utilization and Enzyme Activity Tests

The ability of the strains to grow with carbon sources were assessed using the API 50CH (API system; bioMerieux, Paris, France) standardized system, following the manufacturer’s instructions [17]. Enzyme activities were measured using the API ZYM (API system; bioMerieux) [13].

Cytotoxicity Assay

To assess cytotoxicity, the dehydrogenase activity in the culture medium was assessed using a cytotoxicity detection kit (EZ-Cytox; Daeillab Service Co. Ltd, Seoul, Korea). Vero cells cultured in DMEM (Dulbecco’s modified Eagle’s medium; Gibco, Grand Island, USA) were plated onto 96-well tissue culture plates at a density of 1 × 104 cells per well, and incubated at 37℃ for 24 h in the presence of 5% CO2. The cytotoxicity of the EHEC O157:H7 EDL933 and mutant strains was determined by treating cells with 10 μl of bacterial culture supernatant for 8 h. After incubation for 3 h at 37℃, the absorbance of each well at 450 nm was measured, with a reference wavelength at 650 nm (Molecular Devices, US/VERSAmax, Sunnyvale, USA).

Motility Assay

To investigate bacterial motility, EHEC O157:H7 EDL933 and mutant strains were stabbed into motility agar plates (1% tryptone, 0.25% NaCl, and 0.3% agar) and incubated at 37℃. The diameter of the motility halos was then measured after 10 h. The experiment was repeated using three independent cultures.

 

Results and Discussion

Construction of Deletion Mutants and Complemented Strain

To investigate the interaction of quorum sensing and stringent response regulation, single and triple mutant strains were constructed from EHEC O157:H7 EDL933 (Table 1). The resulting ΔluxS mutant strain was designated ECM101. The stringent response mutant to ECM101 (ΔluxSΔrelAΔspoT) was constructed by conjugation, and designated ECM201. Finally, the ΔluxS complemented strain was constructed by the insertion of pEXP-CT/luxS to ECM201 and designated ECM202.

Effects of Quorum Sensing and Stringent Response Systems on Metabolic Activities

To assess whether the mutations in quorum sensing and stringent response systems influenced the metabolic and enzyme activities in EHEC O157:H7 EDL933, the bacterial growth of wild-type and mutant strains was monitored by measuring ODs and counting viable cells. All the mutant strains showed similar growth rates to the wild-type strain in LB medium, suggesting that there were no obvious growth defects attributable to the mutations. However, the stringent response system mutation (in ECM201 and ECM202) affected both metabolic and enzymatic activities, whereas the quorum sensing system mutant strain (ECM101) had no effect on either metabolic or enzymatic activity. Specifically, the ECM101 strain showed the same metabolic activity pattern as wild type, whereas ECM201 and ECM202 showed defects in the ability to utilize D-arabinose and L-sorbose (Table 3). This is consistent with a previous report, where a ΔluxS mutant of a clinical EHEC O157:H7 strain isolated in Korea was unable to utilize L-sorbose as a single carbon source [18]. Other studies suggested that some enteric bacterial pathogens might use their ability to degrade L-sorbose as their pathomechanism [7]. A defective ability to utilize L-sorbose therefore suggests that the stringent response system may have more correlation with the pathomechanism than the quorum sensing system. When enzymatic activities were compared, ECM201 and ECM202 exhibited different activities to wild type and ECM101. As shown in Table 4, trypsin activity was decreased and α-glucosidase activity was increased in ECM201 and ECM202 compared with wild type and ECM101. Interestingly, the cystine arylamidase activity of ECM201 was recovered, whereas wild type, ECM101, and ECM202 all lacked cystine arylamidase activity (Table 4).

Table 3.a − = no utilization of the carbohydrate; + = good utilization; ++ = very good utilization.

Table 4.a − = no activity; + = low activity; ++ = medium activity; +++ = high activity.

Changes in Acid Resistance and Cell Cytotoxicity of the Mutant Strains

To assess the effect of the mutations on acid resistance, we measured the survival ratio of wild-type and mutant strains under low pH conditions. After incubation for 2 h, the survival ratio of wild type was 69% at pH 2.5 and 60% at pH 1.5. However, the mutant strains all had poor survival at both pH 2.5 and 1.5 (Fig. 2). To determine whether mutations in quorum sensing and stringent response systems influenced cellular cytotoxicity, we carried out an LDH test of wild-type and mutant strains using Vero cells. Although the cytotoxicity of all mutant strains was lower than wild type, the effects were more pronounced in ECM201 and ECM202 than in ECM101 (Fig. 3).

Fig. 2.The survival ratio of EDL933 and mutant strains ECM101 (ΔluxS mutant), ECM201 (ΔluxSΔrelAΔspoT mutant), and ECM202 (ECM201 containing pEXP5-CT/luxS). All strains were precultured in LB medium and then incubated in low pH LB.

Fig. 3.Cytotoxic activity of the culture supernatants from the wild-type, ECM101 (ΔluxS mutant), ECM201 (ΔluxSΔrelAΔspoT mutant), and ECM202 (ECM201 containing pEXP5-CT/luxS) strains.

Changes in the Motility of Bacterial Cells by Interaction Between the Two Regulation Systems

To compare the motility of the wild-type and mutant strains, bacterial swarm motility was assessed. The motility of the quorum sensing system mutant (ECM101) was unchanged, and was similar to wild type (Fig. 4). Interestingly, however, quorum sensing influenced the motility of the stringent response system. The motility of the stringent response mutant was decreased (data not shown), but the motility of the quorum sensing system and stringent response mutant (ECM201) was recovered to the levels of wild type. This result was confirmed by luxS complemented strain (ECM202) (Fig. 4).

Fig. 4.The swarm motility of wild-type, ECM101 (ΔluxS mutant), ECM201 (ΔluxSΔrelAΔspoT mutant), and ECM202 (ECM201 containing pEXP5-CT/luxS) strains in 0.3% tryptone agar after a 10 h incubation.

In summary, our data suggest that quorum sensing interacts with the stringent response regulation system. However, further studies are necessary to define the mechanisms and genes responsible for the phenotypic changes.