Characterization of a novel zinc transporter ZnuA acquired by Vibrio parahaemolyticus through horizontal gene transfer

Abstract

Vibrio parahaemolyticus is a clinically important foodborne pathogen that causes acute gastroenteritis worldwide. It has been shown that horizontal gene transfer (HGT) contributes significantly to virulence development of V. parahaemolyticus. In this study, we identified a novel znuA homolog (vpa1307) that belongs to a novel subfamily of ZnuA, a bacterial zinc transporter. The vpa1307 gene is located upstream of the V. parahaemolyticus pathogenicity island (Vp-PAIs) in both tdh-positive and trh-positive V. parahaemolyticus strains. Phylogenetic analysis revealed the exogenous origin of vpa1307 with 40% of V. parahaemolyticus clinical isolates possessing this gene. The expression of vpa1307 gene in V. parahaemolyticus clinical strain VP3218 is induced under zinc limitation condition. Gene deletion and complementation assays confirmed that vpa1307 contributes to the growth of VP3218 under zinc depletion condition and that conserved histidine residues of Vpa1307 contribute to its activity. Importantly, vpa1307 contributes to the cytotoxicity of VP3218 in HeLa cells and a certain degree of virulence in murine model. These results suggest that the horizontally acquired znuA subfamily gene, vpa1307, contributes to the fitness and virulence of Vibrio species.

Keywords: Vibrio parahaemolyticus; fitness; horizontal gene transfer; virulence; znuA.

Figure 1

Schematic of vpa1307 gene location in V. parahaemolyticus RIMD2210633 and TH3996. Vp-PAI (open box) is flanked by direct repeats (DRs, 5′-AACTC-3′). The white arrows (vpa1309 and vpa1397) indicate the first and last genes outside the VP-PAI. The vpa1307 gene and T3SS2 gene cluster are also indicated.

Figure 2

Multiple sequence alignment of Vpa1307 and ZnuA proteins. The amino acid sequences (GenBank accession No.CAB72627, P73085, AAC74927, Q8Z5W7, AAC45725, and Vpa1307 from C. jejuni, Synechocystis sp., E. coli, S. enterica, Treponema pallidum, and V. parahaemolyticus, respectively) were aligned using the CLUSTAL W2. Three conserved histidine residues were indicated by black arrows.

Figure 3

Structural alignments between VPA1307 and ZnuA from Synechocystis sp. (A) VPA1307 modeled structure. Structure of VPA1307 was modeled using SWISS-MODEL program and three conserved histidine residues were labeled. (B) Structural comparison of VPA1307 (green) and the crystal structure of ZnuA from Synechocystis sp. (PDB accession number 1PQ4) template (gray). A TM score of 0.97 was obtained over 252 aligned residues.

Figure 4

Neighbor-joining tree of VPA1307 and related genes. The protein sequences were obtained from NCBI except that L8XH86, M2S449, M7R955, and K5UB29 were obtained from EBI. ZnuA homologs were represented in red (except that of Aquifex aeolicus) and manganese transporters were represented in black. Bootstrap values (>50%) are shown at branch nodes. ZnuA homolog protein sequence from Aquifex aeolicus was used as outgroup. Bar, 0.2 difference at the amino acid level.

Figure 5

Growth rates of V. parahaemolyticus strains. V. parahaemolyticus strains were cultured in LB or LB supplemented with 35 μ M TPEN. V. parahaemolyticus growth (OD600) was monitored. The data represents three independent experiments ± the SD.

Figure 6

Detection of vpa1307 transcript by RT-PCR in V. parahaemolyticus. V. parahaemolyticus was grown in LB and RNA was extracted after incubation with/without 35 μ M TPEN. 16S RNA was used as a loading control. Lane 1, 16S RNA (with TPEN); lane 2, 16S RNA (without TPEN); lane 3, vpa1307 (without TPEN); lane 4, vpa1307 (with TPEN); lane M, 100-bp maker (Thermo Scientific).

Figure 7

Relative growth rate of different V. parahaemolyticus strains under normal and zinc depletion conditions. Different V. parahaemolyticus strains as indicated were cultured in LB or LB supplemented with 35 μ M TPEN. Their growth (OD600) was monitored at 6 h and relative growth rates were calculated as culture grown with TPEN to that of grown without TPEN. The data represents three independent experiments ± the SD.

Figure 8

Cytotoxic effect of vpa1037 on HeLa cells. HeLa cells were infected with V. parahaemolyticus at MOI of ~50 cfu per cell. Supernatants were collected at specific time points and the amounts of LDH released were determined using CytoTox 96 Non-Radioactive Cytotoxicity kit (Promega) following the manufacturer's instructions. Percentage of cytotoxicity was calculated using formula: (test LDH release—spontaneous release)/maximal release. Test LDH release represents the LDH release after infection with different V. parahaemolyticus strains; spontaneous release represents the baseline cell LDH release without infecting with any bacteria, whereas maximal release represents the release of LDH when cells were lysed using lysis solution from the kit. The data represents three independent experiments ± the sem.

Figure 9

Survival rates of murine model infected with different V. parahaemolyticus strains. C57BL/6 mice (n = 10) were infected intraperitoneally with WT or Δ vpa1307 strains (10⁸ CFU) and mice were monitored for the duration of 96 h. The mortality rate was measured at different time points (0, 3, 6, 9, 12, 24, 48, 72, and 96 h) for three independent experiments. Each data point in the figure represents the average of the data from three experiments. Kaplan–Meier and log rank tests were used to analyze the data.