Correspondence: Yanguang Cong, Department of Microbiology, Third Military Medical University, Chongqing 400038, China. Tel.: +86 23 68752241; fax: +86 23 68752240; e-mail: email@example.com
TonB-dependent transporters (TBDTs) are bacterial outer membrane proteins that are usually involved in the uptake of certain key nutrients, for example iron. In the genome of Salmonella enterica ssp. enterica serovar Typhi, the yncD gene encodes a putative TBDT and was identified recently as an in vivo-induced antigen. In the present study, a yncD-deleted mutant was constructed to evaluate the role of the yncD gene in virulence. Our results showed that the mutant is attenuated in a mouse model by intraperitoneal injection and its virulence is restored by the transformation of a complement plasmid. The competition experiments showed that the survival ability of the yncD-deleted mutant decreases significantly in vivo. To evaluate its vaccine potential, the yncD-deleted mutant was inoculated intranasally in the mouse model. The findings demonstrated a significant immunoprotection against the lethal wild-type challenge. The regulation analysis showed that yncD gene promoter is upregulated under acidic condition. The present study demonstrates that the yncD gene plays an important role in bacterial survival inside the host and is suitable for the construction of attenuated vaccine strains as a candidate target gene.
TonB-dependent transporters (TBDTs) are transporter proteins located in the outer membrane of Gram-negative bacteria. They are dependent for their function on contact with the TonB complex, which transduces the proton motive force of the cytoplasmic membrane to energize substrate transport through specific TBDTs across the outer membrane (Schauer et al., 2008). The TonB system, including the TonB complex and TBDTs for key nutrients such as iron and nickel, is of great medical relevance because the survival of pathogenic bacteria in their hosts depends on their capability to take up these nutrients (Perkins-Balding et al., 2004; Miethke & Marahiel, 2007; Schauer et al., 2007, 2008).
In the genome of Salmonella enterica ssp. enterica serovar Typhi Ty2 (S. Typhi Ty2), six genes encode TBDTs – fepA, iron, cirA, foxA, btuB and t1497. The homologous gene of t1497 is designated as yncD in Escherichia coli, hence the gene name is used as such throughout this paper. The functions of these genes have been determined experimentally except for yncD. The products of fepA and iron are receptors of ferric enterobactin and colicins B and D. CirA is a receptor protein for siderophores (colicin IA, IB and V) and microcins (E492, H47 and M). FoxA is a ferrioxamine B receptor. BtuB is a vitamin B12 (cobalamin) transporter. These five characterized TBDTs are required for the virulence of Salmonella, with the exception of BtuB (Sampson & Gotschlich, 1992; Kingsley et al., 1999; Rabsch et al., 2003).
To date, no direct functional study has been conducted on yncD; however, it was mentioned in several studies. In a previous study, YncD protein was identified as an in vivo-induced antigen in S. Typhi Ty2 (Hu et al., 2009). In an assay to screen pH-regulating genes in E. coli, yncD gene expression was showed to be regulated by pH stresses and its highest expression was induced at pH 5.0 (Maurer et al., 2005). In a DNA microarray analysis of the heat- and cold-shock stimulons in Yersinia pestis, the transcription of the yncD gene was identified to be enhanced 12.5-fold after heat-shock (Han et al., 2005). Marchal et al. (2004) reported a putative PmrA binding sequence upstream of the yncD gene in S. enterica ssp. enterica serovar Typhimurium (S. Typhimurium). The binding sequence also exists upstream of the S. Typhi yncD gene, which indicates that the expression of the yncD gene may be regulated by the PmrAB system. The PmrAB regulatory system responds to acid and ferric iron, and is required for resistance to cationic antibiotic polymyxin B (Roland et al., 1993) and Fe3+-mediated killing (Wösten et al., 2000). These indirect studies suggest that the yncD gene may be a stress gene subject to regulation by certain conditions, such as acid or heat, and as a putative TBDT, YncD may play a role in bacterial survival in vivo. The present study attempts to verify this hypothesis.
Materials and methods
Strains and growth conditions
The bacterial strains, plasmids and primers used in this study are listed in Table 1. Unless noted, the bacterial strains were maintained in lysogeny broth (LB) media. A defined α-minimum essential medium (α-MEM; Invitrogen) was used as the basic medium for gene regulation analysis. As required, antibiotics were used at the following concentrations: ampicillin 100 μg mL−1 and kanamycin 50 μg mL−1.
Table 1. Bacterial strains, plasmids and primers used in this study
A suicide vector for allelic exchange was constructed to facilitate the generation of knockout mutants. Two complementary oligonucleotide chains (M1 and M2, Table 1), containing multiple clone sites, including EcoRI, XbaI, ApaI, SfiI, SacI, NotI, SpeI, NdeI, SacII and BglII, were synthesized. The two oligonucleotide chains were boiled for 15 min and then annealed. After digestion with BglII and EcoRI, the DNA fragment was ligated to a 3599-bp BglII-EcoRI-digested fragment from the pUT-mini-Tn5 plasmid containing ampicillin resistance, a π protein-dependent origin, and mob RP4, resulting in pYG1. A kanamycin resistance cassette from pACYC177 was amplified using primers kana1 and kana2 (Table 1) and then cloned into the ApaI-XbaI site of the pYG1 to generate pYG2. The sacB gene of pYLTAC7 was removed by EcoRI-restriction, generating a 1.7-kb fragment. Then, the sacB-containing fragment was cloned into the EcoRI site of the pYG2 resulting in pYG3. Finally, the vector pYG3 was digested by ApalI to remove the ampicillin resistance and was self-ligated to create the final plasmid pYG4.
Construction of the yncD knockout strain
As described by Link et al. (1997), the 2067-bp in-frame deletion of the yncD gene was constructed by cross-over PCR with primers k1, k2, k3 and k4 (Table 1). The product was ligated directly to the pMD18-T vector (Takara Co., Dalian, China) and confirmed by sequencing. The recombinant plasmid was digested by NdeI and the fragment containing the deletion copy of the yncD gene was ligated to pYG4. The resulting vector was introduced into E. coli S17-1/λpir by electroporation.
The hybrid plasmid was transferred into YGC101 (wild type) by electroporation to perform mutagenesis. Integrons were selected from the LB plates containing kanamycin and were confirmed through PCR analysis. Overnight cultures of the identified integron grown in the absence of antibiotics were streaked onto LB agar containing 5% sucrose. Selected colonies with normal colony phenotypes were patched onto LB agar with and without kanamycin. The colonies that were sensitive to kanamycin were analyzed for the deletion by PCR with the primers O1 and O2, as well as I1 and I2 (Table 1). The strain carrying the desired deletion was selected and designated as YGC102.
Complementation of yncD gene
The gene yncD was PCR-amplified from the wild-type strain using the primers C1 and C2 (Table 1), which were designed based on sequences external to the yncD coding region. After amplification, the DNA fragment was digested by EcoRI and HindIII and ligated to the pBR322 to obtain PYN plasmid. The resulting vector was introduced into the mutant strain YGC102 by electroporation to produce the strain YGC103.
Virulence in the mouse mucin model
To determine the involvement of yncD in virulence, the median lethal dose (LD50) of YGC101, YGC102 and YGC103 was determined as described by Wang et al. (2001) with minor modifications. Female BALB/c mice aged 6–8 weeks (three mice per group, three groups per strain) were injected intraperitoneally with various dilutions of the different strains mixed with 7% (w/v) mucin from porcine stomach (Sigma) at a final volume of 0.5 mL in phosphate-buffered saline (PBS). The number of deaths that occurred within 72 h after inoculation was counted. The LD50 was calculated as described by Reed & Muench (1938).
Bacterial competition test in vivo
To evaluate the effect of yncD gene deletion on the survival capability in vivo, we performed bacterial competition experiments in the mouse model. Overnight cultures of YGC101 and YGC102 were harvested, washed twice with PBS, and mixed at an equal density of 2 × 103 CFU mL−1 with 7% (w/v) porcine gastric mucin. Three female BALB/c mice were injected intraperitoneally with the bacterial suspension at a volume of 0.5 mL. Twenty-four hours later, the mice were sacrificed, injected intraperitoneally with 1 mL of sterile PBS, kept for 1 min with gentle massage over the abdomen and then extracted. After serial dilution, the samples were spread on the LB plates and incubated at 37 °C overnight. Of the colonies recovered from the same mice, 20 were randomly picked and identified by PCR with primers O1 and O2. To calculate the competitive indices, the ratio of yncD-deleted mutant to wild type recovered from the abdominal cavity was determined and then normalized by dividing by the ratio of yncD-deleted mutant to wild type in the initial inoculum.
Female BALB/c mice aged 6–8 weeks (five groups with three mice per group) were immunized once intranasally with 109 CFU of YGC102 or PBS (as control). Thirty days later, the mice of the control group were challenged with 103 CFU of wild type, whereas the mice of the other four groups were challenged respectively with 104, 105, 106 and 107 CFU of the strain using the porcine gastric mucin model as described above. The survival of the mice was monitored for 7 days.
Regulation analysis of yncD gene promoter
A promoterless egfp gene from pEGFP-N2 was isolated by digestion with EcoRI and HindIII and was subcloned into the corresponding sites of the pBR322 plasmid, resulting in the pBGPL plasmid. The yncD promoter region was amplified by PCR using the primers EPR1 and EPR2 (Table 1). The promoter fragment was ligated directly with PMD18-T vector and subcloned as NcoI fragments into the corresponding sites of pBGPL resulting in the pBGP plasmid. The generated plasmid was electroporated into the YGC101 strain to generate YGC104 strain.
The YGC104 strain cells were inoculated into the indicated media (for the heat-shock experiment, cells were incubated at 45 °C for 10 min) and grown at 37 °C for 5 h to allow expression of enhanced green fluorescent protein (EGFP). Then, the bacteria were diluted with PBS and analyzed in a flow cytometer (BD FACSCanto II) with the gates set to forward and side scatters characteristic of the bacteria. The optical detector FL1-H was used for this measurement. For each condition assessed, 10 000 bacterial cells were analyzed and the mean fluorescent intensity of the bacteria was obtained. Each experiment was performed in triplicate. Comparisons of expression values among the groups were performed by t-test.
Quantitative real-time RT-PCR analysis
Total RNA was isolated from bacterial cells of Ty2 wild type incubated under each condition using the SV Total RNA Isolation System (Promega). Additional treatments with RNase-Free DNase I (Takara) were performed to eliminate any genomic DNA. The quantity and quality of the total RNA was determined with an ND-1000 spectrophotometer (NanoDrop). The cDNAs were synthesized using the PrimeScript RT reagent kit (Takara). Real-time PCR reactions were performed in a Rotor Gene 3000 real-time PCR instrument (Corbett Research) using the SYBR Premix Ex Taq kit (Takara). The primers are listed in Table 1. Real-time cycling conditions were as follows: 95 °C for 30 s; 40 cycles of 95 °C for 5 s, 55 °C for 30 s and 72 °C for 30 s. Quantitative real-time PCR experiments were performed in triplicate. The transcriptional levels of yncD gene were normalized to the transcripts of a housekeeping gene, rpoD, which served as an internal control.
The yncD-deleted mutant is attenuated in the mouse mucin model
The YncD protein of S. Typhi Ty2 is annotated as a TBDT in NCBI, which was confirmed with our bioinformatics analysis (Supporting Information, Appendix S1). To verify whether YncD plays a role in pathogenesis, a yncD deletion mutant was constructed by homologous recombination using a suicide vector pYG4 (Fig. S1). The LD50 of S. Typhi Ty2 and its yncD deletion mutant were measured using the mouse mucin model. As shown in Table 2, the ΔyncD mutant is 1000 times less virulent than the wild-type strain. When the pBR322 plasmid carrying the intact yncD gene with its native promoter was transformed into the mutant, the virulence was almost completely restored. These data show that the deletion of the yncD gene results in attenuation.
Table 2. LD50 measurement of S. Typhi Ty2 and its derivative strains
2.25 × 102
2.25 × 105
7.11 × 102
To understand why yncD knockout leads to reduced virulence, we determined the growth characteristics of the LB media-cultured YGC101, YGC102 and YGC103. Fig. S2 shows that the yncD-deleted mutant grows in the LB media as well as the wild-type and the complemented strain. The bacterial growth curves showed no significant deviation among the three strains. However, the competitive indices of the yncD-deleted mutant in the bacterial competition tests is 0.149 ± 0.093, which indicates a decreased survival capability of the mutant in vivo compared with that of the wild type.
Immunoprotective efficacy of the yncD deletion mutant
As the yncD deletion mutant was attenuated in the mouse mucin model, we examined its vaccine potential. Among the mice immunized with the yncD deletion mutant, a protection rate of 100% was produced in the groups challenged with 104 and 105 CFU of the wild-type strain, and a protection rate of 33% was produced in the group challenged with 106 CFU. As all control mice died 2 days after they were challenged with 103 CFU of the wild-type strain, the yncD deletion mutant showed a significant immunoprotective effect (Table 3).
Table 3. Efficacy of YGC102 (ΔyncD) in protecting mice against intraperitoneal challenge with wild-type serovar Typhi suspended in porcine gastric mucin, 4 weeks after a single intranasal immunization
Strain or control
Dosage of challenge (CFU)
Induction of the yncD gene under acid conditions
The yncD gene was supposedly a target of the PmrAB system by an early in silico analysis (Marchal et al., 2004). The PmrAB regulatory system is required for resistance to the cationic antibiotic polymyxin B and Fe3+-mediated killing. Therefore, the responses of the yncD mutant and the wild-type strain to polymyxin B and Fe3+-mediated killing were assessed. The results showed that no difference exists between the two strains (data not shown).
To investigate the regulation pattern of the yncD gene, the yncD promoter region was cloned and inserted into a site before a promoterless egfp gene, which was carried into the pBR322 plasmid. The expression of the EGFP reporter fusion was determined by fluorescence-activated cell sorter analysis of the wild-type strain under different conditions. We also conducted quantitative real-time RT-PCR to analyze the transcriptional level of the yncD gene in the wild-type cells under different conditions. As shown in Fig. 1a and b, the yncD gene expression showed an acid induction feature. However, other conditions such as supplementation with 10 mM FeCl3, an inducing factor for PmrAB two-component regulatory system in S. Typhimurium (Marchal et al., 2004), or heat shock, shown to induce yncD gene expression in Y. pestis (Han et al., 2005), have no significant effect on yncD gene expression in our experiments. The disparity is believed to be due to the presence of magnesium in the α-MEM. Millimolar magnesium represses the two-component regulatory system PhoPQ, which indirectly represses the PmrAB by reducing the expression of PmrD, which regulates PmrA activity at a post-transcriptional level (Garcia-Véscovi et al., 1996; Kox et al., 2000; Kato & Groisman, 2004). However, as a common activation signal of both the PmrAB and PhoPQ systems, acidic pH had been shown to activate PmrAB in spite of the presence of magnesium (Perez & Groisman, 2007).
Blanvillain et al. (2007) performed a survey of TBDTs in 226 completely sequenced eubacterial genomes revealing a broad variation in TBDT number in the surveyed bacteria. Interestingly, except for Pseudomonas aeruginosa, no important human pathogen was found among the bacteria with TBDT-overrepresentation. However, many human pathogens, e.g. Borrelia, Chlamydia, Coxiella, Francisella and Legionella, were found among bacteria without TBDT. Most of them were human or animal obligate parasites. Thus, the number of TBDTs in a bacterial strain seems to depend on the ecological niche diversity of the strain and is inversely related to a close relationship with human or animal, as in parasitism. As proteins located on the surface of bacterial cells, TBDTs are undoubtedly antigenic candidates. If a pathogen enters a host body, these antigens can induce specific antibodies that may inhibit the growth, propagation and pathogenesis of the pathogen. A large number of TBDTs are seemingly not optimal choices for pathogens if other selections are available. However, in some human pathogens such as S. Typhi, notwithstanding the long process of evolution, six TBDTs are still reserved, indicating their essential role in habitat survival, e.g. in the human body.
In the present study, we found that deleting the yncD gene of S. Typhi leads to significant attenuation in the porcine gastric mucin model. The model has been used to evaluate the degree of attenuation of some S. Typhi vaccine strains, CVD 906, CVD 908, CVD 908-htrA and CVD 915 (Hone et al., 1991; Wang et al., 2001). Although the model does not closely mimic the pathogenesis of human typhoid infection, it reflects the survival capability of pathogen in vivo. When cultured in the LB media, the yncD-deleted mutant showed normal growth, which indicated that the function of YncD is related to in vivo survival rather than growth in vitro.
As an IVI antigen identified in a previous study using IVIAT method, the regulation of YncD expression usually can be induced in certain conditions encountered in vivo. In the genome of S. Typhi, the yncD gene is adjacent to the yncE gene but it has the opposite transcriptional orientation. The yncE gene is induced under iron restriction through the action of the global iron regulator Fur in E. coli; however, the regulator and the iron restriction did not affect the transcription of the yncD gene (McHugh et al., 2003). Upstream of the yncD gene, a possible PmrAB-box sequence, cattttcttaacttaat, was found, which indicated that the expression of the yncD gene may be regulated by the PmrAB system (Marchal et al., 2004). In agreement with this anticipation, acidic pH, a main activation signal of the PmrAB system, was proved to induce the expression of yncD gene in the present study.
The acid condition is an ecological niche that pathogens usually encounter in vivo. Enteric pathogens share an oral route of infection (Gorden & Small, 1993; Maurer et al., 2005). During the initial infection, enteric bacteria encounter low pH stresses in the human digestive tract (Drasar et al., 1969). Successful colonization requires survival through the stomach at pH 1–2 or the intestinal tract at pH 2–7 (Dressman et al., 1990). The bacteria respond to low pH stresses by regulating gene expression, which maintains internal pH homeostasis (Gorden & Small, 1993). Moreover, low pH is an important inducing factor of virulence genes as well. Low pH enhances the expression of numerous virulence factors, such as the ToxR-ToxT virulence regulon in Vibrio cholerae (Behari et al., 2001) and the phoP-phoQ regulon of Salmonella enterica (Bearson et al., 1998). It also enhances expression of genes for flagellar motility and catabolism (Maurer et al., 2005).
Due to lack of information, the exact function of YncD remains unclear. However, our study showed that YncD plays a role in the in vivo survival of S. Typhi. As the yncD gene knockout significantly reduces bacterial virulence and the attenuated strain shows an effective immunoprotection, the yncD gene is undoubtedly a good candidate gene for the construction of attenuated vaccine strains.
This study was supported by the National Natural Science Foundation of China (Grant No. 30500435). We gratefully acknowledge Victor de Lorenzo of the Centro Nacional de Biotecnologia CSIC, Spain, for providing the Mini-Tn5 plasmid.