The stringent response (SR) is a bacterial defense mechanism that is activated in response to various growth-inhibiting stresses, by accumulation of the small nucleotide regulator (p)ppGpp, and induces global changes in bacterial transcription and translation [1, 2]. In Vibrio cholerae (the causative agent of pandemic disease cholera), the relA and relV genes are involved in the production of (p)ppGpp, whereas the spoT gene encodes an enzyme that hydrolyzes it [3].

Recent work from our group showed that the SR is activated by the alternative electron acceptor trimethylamine oxide (TMAO) in V. cholerae, stimulating anaerobic and steady-state growth and modulating cholera toxin (CT) production [4, 5]. A mutant strain defective in (p)ppGpp production (i.e., ΔrelAΔrelVΔspoT; (p)ppGpp⁰) lacked the ability to produce CT and exhibited excessive growth under anaerobic TMAO respiration. In contrast, the ΔrelAΔspoT mutant overproducing (p)ppGpp generated high levels of CT and exhibited growth retardation [5]. Thus, accumulation of intracellular (p)ppGpp results in inverse regulation of bacterial cell growth and CT production.

In this study, we carried out a genome-wide screen for genes related to SR-mediated regulation of virulence and growth. Transposon (Tn) random mutagenesis was used as a global genetic screening system to identify genes controlled by (p)ppGpp [6]. We constructed a random Tn mutant library based on the ΔrelAΔspoT mutant using the plasmid pBTK30, which contains a gentamicin resistance-marked transposable element and mariner C9 transposase. Escherichia coli strain SM10/λ pir harboring pBTK30 and the V. cholerae recipient strain (ΔrelAΔspoT mutant) were mixed and spread onto Luria-Bertani (LB) agar plates followed by incubation for 6 h at 37°C. The cells were then resuspended in LB medium, and dilutions were spread on LB plates containing 50 μg/ml gentamicin and 200 μg/ml streptomycin to select Tn insertion mutants. A total of 1,500 bacterial colonies were screened for mutants exhibiting growth retardation. Bacterial colonies were inoculated in a 96-well plate containing LBT (LB medium containing 50 mM TMAO) broth and were grown under anaerobic conditions for 16 h. Bacterial growth was assessed spectrophotometrically by measuring the optical density at 600 nm. A total of 42 candidate mutants that exhibited higher growth rates than the ΔrelAΔspoT mutant were obtained in the initial screen.

The 42 candidate mutants were tested for their capacity for CT production, which was measured in culture supernatants by GM1 enzyme-linked immunosorbent assay as previously described [7]. We ultimately selected six candidate mutants (ΔrelAΔspoT::Tn-1–6) that showed decreased CT production and recovery of (p)ppGpp-induced growth retardation (Figs. 1A and 1B). Compared with the parental strain, the OD₆₀₀ value of the six candidate mutants was higher than that of wild-type strain N16961 after 16 h of culture (Fig. 1A). On the other hand, the mutants produced low levels of CT (Fig. 1B). Consistent with our previous observation that CT production and growth in response to (p)ppGpp accumulation are inversely regulated, we found an inverse relationship between CT production and growth in mutants with the Tn insertion. The site of Tn insertion in the mutants was determined by arbitrary PCR as described elsewhere [5] using the specific primer BTK30-Tnp1 (CACCGCTGCGTTCGGTCAAG) and the two random primers RP-1 (CTTACCAGGCCACGC GTCGACTAGTACNNNNNNNNNNGATAT) and RP-2 (CTTACCAGGCCACGCGTCGACTAGTACNNNNNN NNNNACGCC). This was followed by a second round of PCR using the product from the first round as a template and the primers BTK30-Tnp2 (CGAACCGAACAGGCC TTATGTTCAATTC) and RP-3 (CTTACCAGGCCACGC GTCGACTAGTAC). The product was sequenced using the primer BTK30-Tnp3 (TGGTGCTGACCCCGGATGAAG). The Tn insertion site was identified by comparison with the genome sequence of V. cholerae 7th pandemic strain N16961 available in public databases. Interestingly, the genes carried by ΔrelAΔspoT::Tn-1 and 6 and ΔrelAΔspoT::Tn-2, 4, and 5 that were disrupted by Tn insertion were identified as VC2187(flaC) and its promoter region, respectively (Fig. 1C).

Figure 1. Selection of transposon (Tn) insertional mutant strains derived from the ΔrelAΔspoT mutant strain that reverse (p)ppGppinduced growth and virulence phenotypes. (A) Changes in the growth of wild-type strain N16961 and (p)ppGpp⁰, ΔrelAΔspoT, and six selected mutant strains are shown. Bacterial cells were inoculated in LB broth containing trimethylamine oxide (LBT) and cultured under anaerobic conditions for 16 h. OD₆₀₀ values were determined as a measure of relative growth. Values represent the mean ± SD of three independent experiments. *P < 0.03 vs. ΔrelAΔspoT mutant strain. (B) Cholera toxin (CT) production in bacterial cells grown in LBT under anaerobic conditions. The culture supernatant was harvested and CT levels were determined by enzyme-linked immunosorbent assay. Values represent the mean ± SD of three independent experiments. *P < 0.001 vs. ΔrelAΔspoT mutant strain. (C) Schematic depiction of the V. cholerae VC2178 locus. Arrowheads indicate the position of Tn insertions. (D) Intracellular ppGpp was detected by TLC analysis. Bacterial cells were anaerobically grown in LBT with [³²P]-orthophosphate for overnight. Cellular extracts were prepared and analyzed by TLC.

To confirm whether the accumulation of (p)ppGpp was reduced by flaC mutation, we next measured intracellular (p)ppGpp levels of bacterial cells under growth by anaerobic TMAO respiration. Intracellular (p)ppGpp concentration was measured as previously described [5]. The ΔrelAΔspoT, flaC::Tn mutant accumulated high levels of intracellular (p)ppGpp, which was not significantly different from the ΔrelAΔspoT mutant (Fig. 1D). These results indicate that the mutation of flaC was not affected by the accumulation of (p)ppGpp.

V. cholerae has five flagellin genes arranged at two loci, flaA, C and flaE, D, B. The flaA, a major flagellin coding gene, is associated with flagellum formation and motility; however, the precise function of the other flagellar filament-coding genes, including flaC, is not clearly defined [8]. Previous reports have shown that the gene expression of relA, which encodes an enzyme involved in the synthesis of (p)ppGpp, is correlated with that of flaC under starvation environment conditions in V. cholerae [9]. Furthermore, some reports indicate that flagellin-coding genes play an essential role in bacterial growth and virulence [10–13]. Here, we found that the Tn insertion in flaC in the ΔrelAΔspoT mutant reduced CT production, which reversed SR-induced growth retardation. Together, our results suggest that FlaC has potential role as a regulator of (p)ppGpp-mediated toxin production and cell growth. However, there is no report that FlaC can act as a transcriptional factor, and we do not yet know the mechanistic relationship between flagellin and the SR, so further studies are needed.

Interestingly, the ΔrelAΔspoT, flaC::Tn mutant strain showed a severe decrease in cell motility compared with the parent strain (Fig. 2A). Swimming and swarming assays were performed to measure bacterial cell motility as described previously [14]. Bacterial cells grown anaerobically in LBT were spot-inoculated on 0.3% (w/v) agar LB plate for 12 h at 37°C, with the diameter of the circular halo measured as an estimate of swimming motility; 0.5% (w/v) agar LB plates were used for the swarming assay. It is worth noting that bacterial cell motility is (p)ppGpp dependent. We observed that motility was decreased in the (p)ppGpp⁰ mutant, whereas the ΔrelAΔspoT mutant was hypermotile as compared with wild-type strain N16961 (Fig. 2). These results confirm that (p)ppGpp is involved in the motility of V. cholerae. Furthermore, the ΔrelAΔspoT, flaC::Tn mutant showed a similar degree of motility as the (p)ppGpp⁰ mutant, indicating that FlaC is associated with the SR-induced hypermotile phenotype.

Figure 2. FlaC is involved in regulation of stringent response-mediated cell motility in Vibrio cholerae. Cells were cultured in LB broth containing trimethylamine oxide under anaerobic conditions for 16 h and used for motility assays. (A) Bacterial strains were spot-inoculated on a 0.3% (w/v) agar LB plate to evaluate swimming motility. (B, C) Swarming motility was assessed on 0.5% (w/v) agar LB plates.

We next compared the morphology of the flagellum of wild-type strain N16961 and mutant strains. Cells were anaerobically cultured overnight in LBT and examined by transmission electron microscopy (JEM 1010; JEOL, Japan) as previously described [15], after negative staining with a 2% aqueous solution of phosphotungstic acid (pH 7.4). Interestingly, we found that the ΔrelAΔspoT, flaC::Tn mutant strain had thinner flagella than the parent mutant strain ΔrelAΔspoT. Furthermore, the diameter of the flagellum was correlated with the ability of (p)ppGpp production (Fig. 3). These results indicate that (p)ppGpp alters bacterial cell motility by modulating flagellar thickness and that FlaC plays an important role in this process.

Figure 3. Transmission electron microscopy analysis of V. cholerae strains. Representative transmission electron micrographs of flagella from wild-type V. cholerae strain N16961 and (p)ppGpp⁰, ΔrelAΔspoT, and ΔrelAΔspoT, flaC::Tn mutant strains. Scale bar, 200 nm. The flagellar diameter was measured using iTEM acquisition and analysis software (Olympus Soft Imaging Solutions GmbH, Germany).

In conclusion, the mutation in flaC in the ΔrelAΔspoT mutant strain reduced CT production and abolished (p)ppGpp-induced growth retardation in V. cholerae. In addition, the ΔrelAΔspoT, flaC::Tn mutant showed decreased flagellar diameter as compared with the parent strain, which was associated with reduced motility. This report describes a potential function for FlaC that involves the regulation of SR-induced toxin production, growth, and hypermotility.

This work was supported by a grant from the National Research Foundation (NRF) of Korea (Grant 2014R1A1A2056139) funded by the Korean government. This work was also supported by a Nakdonggang National Institute of Biological Resources (NNIBR) grant funded by the Ministry of Environment, Republic of Korea.

The authors have no financial conflicts of interest to declare.