Vanadium accelerates horizontal transfer of tet(M) gene from marine Photobacterium to Escherichia coli


Correspondence: Satoru Suzuki, Center for Marine Environmental Studies, Ehime University, Matsuyama, Ehime, 790-8577 Japan. Tel./fax: +81 89 9278552; e-mail:


Vanadium is a contaminant from steel additive and ship fuel in coastal and port areas, and its effect on marine microbes remains largely unknown. We showed that vanadium accelerates transfer of the tetracycline resistance gene tet(M) from Photobacterium to Escherichia coli, and found a positive correlation between the concentration of vanadium in natural marine sediment and the rate of oxytetracycline resistance. These results suggest the possibility that vanadium may play a role in the preservation and horizontal transfer of antibiotic resistance genes in the marine environment.


Vanadium (V) is used as a steel additive (Moskalyk & Alfantazi, 2003) and is contained in jet and ship fuels, which may be released into the air and oceans (Viana et al., 2008; Pondolfi et al., 2011). Oil combustion alone accounts for 91% of total worldwide atmospheric V emissions. Atmospheric concentrations of V range from 10–50 mg g−1 in fly ash particulates of residual (heavy) fuel oils, which release V into the air when combusted (Mamane & Pirrone, 1998; International Agency for Research on Cancer (IARC) (2006). Vanadium pollution thus raises serious marine environmental concerns. High levels of V have been found in coastal sediment (Beg et al., 2001).

Vanadium (especially as VOSO4) is toxic to the mammalian respiratory system (Wörle-Knirsch et al., 2007) and also exerts adverse physiological effects on various microbes (Fukuda & Yamase, 1997; Aendekerk et al., 2002; Denayer et al., 2006). Bacterial resistance to V can be caused by mutations in efflux pump (Aendekerk et al., 2002) and tricarboxylic acid (TCA) cycle enzymes (Denayer et al., 2006). In contrast, some V-containing metabolic enzymes have been identified in both eukaryotes (Rehder, 1992) and soil and enteric bacteria (van Marwijk et al., 2009; Lee et al., 2010), and it appears that V serves as an essential trace element in these organisms. However, the effect of V pollution on the marine microbial ecosystem is unknown.

In some cases, antibiotic resistance can be correlated with metal exposure (Baker-Austin et al., 2006; Stepanauskas et al., 2006). Exposure to toxic metals such as cadmium (Cd) and nickel (Ni) represents a selective pressure that may lead to the development of antibiotic resistance (Stepanauskas et al., 2006), and major resistance mechanisms are based on common efflux of metals and antibiotics and a reduction in permeability (Baker-Austin et al., 2006). Some metals, such as Ca2+ and Mg2+, are capable of inducing competence at millimolar concentrations (Takeo, 1972; Page & von Tigerstrom, 1979), resulting in accelerated DNA intake by bacteria and horizontal gene transfer (HGT) between bacteria. We therefore hypothesized that V contamination in the ocean may facilitate development of antibiotic resistance through HGT. To determine the how exposure to V and other metals influences the acquisition of antibiotic resistance, we cultured oxytetracycline (OTC)-sensitive Escherichia coli in the presence of OTC-resistant Photobacterium. Then the occurrence rate of OTC resistant-E. coli was enumerated. Transfer of the tetracycline resistance gene, tet(M), was also confirmed in transconjugants. Furthermore, the concentration of V and the rate of OTC resistance in natural marine sediment were quantified.

Materials and methods

Strains and gene transfer

The marine bacterium Photobacterium damselae subspecies damselae strain 04Ya311, first reported as a Vibrio sp. (Neela et al., 2007), was used as the donor of the tet(M) gene. It has already been confirmed that transfer of tet(M) from P. damselae 04Ya311 to E. coli occurs through mating (Neela et al., 2009) via the conjugative plasmid pAQU1 (Nonaka et al., 2012), and this transfer is reversible. Mating gene-transfer experiments were performed as described previously (Neela et al., 2009) with E. coli JM109, which does not possess tet(M), serving as the recipient strain. The ratio of donor to recipient cells in the mating experiment was 10−3 : 1 because of the high conjugation rate (6.49 ± 1.97 × 10−3, = 3) in the absence of V. Conjugation rate was calculated as OTC-resistant recipient cell number/total recipient cell number. Mating experiments were performed at concentrations of each metal that were below the minimum inhibitory concentration (MIC). The definition of ‘transconjugants’ is recipient strains acquired OTC resistance by the mating. The tet(M) gene was detected in transconjugants using PCR (Rahman et al., 2008).

Vanadium quantitation

Chemical determination of V in Pacific marine sediment was performed. Details of sampling sites and condition are reported in elsewhere (Rahman et al., 2008). Analysis was performed by using inductively coupled plasma-mass spectrometry (ICP-MS) according to the method of Ha et al. (2009). Briefly, the sediment samples were treated with a mixture of HF–HNO3 (1 : 5) and digested by a closed vessel microwave system (Ethos D, Milestone S.r.l., Sorisole, BG, Italy). The digested solution was heated until acid was removed. The residue was dissolved by HNO3 and diluted with Milli-Q water. Concentrations of 28 trace elements were measured with an ICP-MS (Hewlett-Packard, HP-4500, Avondale, PA). Units of concentrations of trace elements were represented as μg g−1 dry weight.

Oxytetracycline resistance assay

The OTC resistance rate in the sediment has been reported previously (Rahman et al., 2008). The correlation between the V concentration and OTC resistance rate was analyzed in this study. Susceptibility of all strains used in this study to OTC and various metals was determined as the MIC according to the method described by the National Committee for Clinical Laboratory Standards (NCCLS) (2003). The MICs of OTC and various metals for each strain are shown in Table 1.

Table 1. Susceptibility to OTC and detection of tet(M) from strains
 MIC for OTC (μg mL−1)tet(M) positive/tested
Vibrio sp. 04Ya311128+
E. coli JM1092
Without metal25610/10
Ca 500 μM2569/10
Ca 1000 μM25610/10
Ca 5000 μM25610/10
V 100 μM25610/10
V 500 μM25610/10
V 1000 μM25610/10
Zn 150 μM25610/10
Zn 300 μM25610/10
Cu 150 μM25610/10
Cu 300 μM25610/10
Cu 500 μM2565/5
Cd 30 μM25610/10
Hg 2 μM25610/10
Hg 4 μM25610/10
Hg 8 μM25610/10

Results and discussion

Exposure to 500 and 1000 μM V (as VCl3) resulted in a significant increase in the conjugation rate (< 0.05) (Fig. 1a), and exposure to Ca (CaCl2) also increased the conjugation rate in a dose-dependent manner (Fig. 1b). Exposure of E. coli JM109 to zinc (Zn as ZnSO4), copper (Cu as CuSO4), and cadmium (Cd as CdCl2) resulted in a decrease in the conjugation rate (Fig. 1c–e). The conjugation rate also increased in an apparent dose-dependent manner upon exposure to mercury (Hg as HgCl2) (Fig. 1f); however, the increase was not significant. It is known that Ca2+ can increase the competency of bacterial cells and induce DNA compaction due to compensation of the DNA electrostatic charge and hydrophobic interactions of the complex sites (Kabanov & Kabanov, 1995), and it is believed that this mechanism contributes to DNA incorporation. Although the mechanism(s) leading to increased rates of OTC resistance following V exposure are not clear, the present study is the first to demonstrate that V can promote conjugation leading to OTC resistance.

Figure 1.

Conjugation rate of Escherichia coli JM109 to OTC-resistance in the presence of various concentrations of V (a), Ca (b), Zn (c), Cu (d), Cd (e) and Hg (f). Each symbol shows the average value and standard deviation of triplicate experiments. Conjugation rate was calculated as OTC-resistant recipient cell number/total recipient cell number.

The MIC of OTC for the recipient E. coli strain was 2 μg mL−1, whereas that of all the transconjugants was significantly higher (256 μg mL−1) (Table 1). The results of PCR analyses indicated that the tet(M) gene was transferred to the recipient E. coli cells, suggesting that the acquisition of OTC resistance occurred through HGT. The MICs of both OTC and V for the transconjugant were twice as high as those for the donor Vibrio, possibly due to HGT of genes related to OTC- and V-resistance other than tet(M) that are also encoded on pAQU1 (Nonaka et al., 2012). It has been reported that V(IV) binds to the surface of certain proteins (Nishida et al., 2009); however, it is not known whether this property is shared by the V(III) used in this study. Since exposure to Zn, Cu and Cd resulted in a decrease in the conjugation rate, the increased conjugation rate observed following V exposure might have been the result of specific physiological effects similar to those associated with Ca (Takeo, 1972). Chemical interactions between biomolecules and V should be studied to determine the mechanism by which V facilitates the acquisition of OTC resistance through HGT.

To determine whether the observed increased rate of OTC resistance also occurs in the natural environment, we determined the V concentration and rate of OTC resistance in samples of marine sediment. As shown in Fig. 2, the proportion of OTC-resistant bacteria increased with an increase in the concentration of V. Although regression analysis revealed a significant positive correlation between the proportion of OTC-resistant bacteria and V concentration on medium containing 120 μg mL−1 of OTC (= 0.023), this correlation was not significant on medium containing 60 μg mL−1 of OTC (> 0.1). Similarly, no positive correlation was observed between the sediment concentrations of Zn, Cu or Cd and OTC resistance, even though exposure to these metals suppressed acquisition of OTC resistance in E. coli JM109 (data not shown). The positive correlation between V concentration and OTC resistance suggests that more copies of OTC resistance genes may be present in sediments containing higher V concentrations. The rate of HGT increased at V concentrations of 500–1000 μM (1000 μM is equivalent to 157 μg mL−1). The maximum concentration of V in marine sediment was 140 μg g−1 of dry sediment (Fig. 2), which is within the range of HGT elevating concentrations. Despite the fact that our sediment sample was collected in the open ocean, where ship traffic level is not high, the concentration of V was at a level sufficient to stimulate HGT, thus confirming that the V does appear to accumulate in open ocean sediment. Tamminen et al. (2011) reported that tet genes are highly persistent and do not disappear from aquaculture sites, even after several years without antibiotic use. The presence of residual V in coastal marine sediments is thus of concern as this may lead to the preservation and/or spread of antibiotic resistance genes in the marine environment.

Figure 2.

Correlation between V concentration and OTC resistance rate in Pacific marine sediment. Percentage values were calculated from the following equation: number of colony forming units (CFU) on OTC-supplemented marine agar 2216E/number of CFU on marine agar 2216E without OTC × 100. Open symbols denote marine agar medium supplemented with 60 μg mL−1 OTC, and closed symbols denote medium supplemented with 120 μg mL−1 OTC.

The susceptibility of bacteria to V-containing compounds varies (Fukuda & Yamase, 1997; Aendekerk et al., 2002; Denayer et al., 2006). For example, vanadate inhibits Streptococcus pneumonia at 4–32 μg mL−1 (Fukuda & Yamase, 1997), whereas Pseudomonas aeruginosa, which is resistant to VOSO4, can grow in the presence of mM levels (Aendekerk et al., 2002). The HGT might be accelerated in the presence of V in the environment.


This work was partly supported by G-COE Program at Ehime University, from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), and Grant-in-Aid for Scientific Research (22241014) from Japan Society for the Promotion of Science (JSPS). We thank Dr T. Yokokawa for his support in data processing.