Role of the pJM1 plasmid-encoded transport proteins FatB, C and D in ferric anguibactin uptake in the fish pathogen Vibrio anguillarum

Abstract

Vibrio anguillarum serotype O1 is part of the natural flora in the aquatic habitat, but under certain circumstances it can cause terminal haemorrhagic septicemia in marine and fresh water fish due to the action of the anguibactin iron uptake system encoded by the virulence plasmid pJM1. This plasmid harbours the genes for the biosynthesis of the siderophore anguibactin and the ferric anguibactin transport proteins FatD, C, B and A encoded in the iron transport operon. The FatA protein is the outer membrane receptor for the ferric siderophore complex and the FatB lipoprotein provides the periplasmic domain for its internalization, whereas the FatC and D proteins are located in the cytoplasmic membrane and might play a role as part of the ABC transporter for internalization of the ferric siderophore. In this work we demonstrate the essential role of these two inner membrane proteins in ferric anguibactin transport and that the lipo-protein nature of FatB is not necessary for ferric anguibactin transport.

Fig. 1

The iron transport and biosynthesis operon (ITBO) in the pJM1 plasmid of V. anguillarum 775(pJM1). FatC and FatD are polytopic integral cytoplasmic membrane proteins. FatB is a periplasmic binding lipoprotein. FatA is a ferric-anguibactin outer membrane receptor. AngR is involved in anugibactin biosynthesis and regulates the expression of the ITBO. AngT is involved in anguibactin biosynthesis.

Fig. 2

Effect of mutations in either fatC or fatD on growth under iron-limiting conditions. Vibrio anguillarum strains were grown first in TSBS and then in CM9 minimal medium. A 40 μl aliquot of an overnight culture in CM9 (adjusted OD₆₀₀ to 1) was inoculated into CM9 supplemented with 10 μg ml⁻1 Cm and 0.5 mM IPTG. EDDA (1μM) was added to achieve iron-limiting conditions. OD600 was measured after 24 h incubation at 25°C. Experiments were carried out three times, and the error bar shows the standard deviation.

Fig. 3. Bioassay for anguibactin uptake

A. Bioassay to assess whether fatC or fatD is necessary for ferric anguibactin transport. B. Bioassay anguibactin uptake of strains with the wild-type FatB and C23A FatB. Indicator strains were grown in TSBS, and then CM9 minimal medium. A 50 μl aliquot of an overnight culture was mixed with 20 μl of melted CM9 1.5% agar, 20 μM EDDA, 500 μM IPTG and 10 μg μl⁻1 Cm. After the agar became solid 5 μl of V. anguillarum 775(pJM1) culture (labelled as ferric anguibactin) and 1 μl of 1 mg ml⁻1 ferric ammonium citrate [labelled as free iron (FAC)] were spotted on each plate as sources of anguibactin and free iron respectively. Plates were incubated at 25°C for 24 h. The experiments were repeated three times, with consistent results. Strain CC9-16(pMMB208) was used as a positive control; strain CC9-16ΔtonB2(pMMB208), constructed as described (Lopez et al., 2009), was used as a negative control.

Fig. 4

Kinetics of ⁵⁵Fe-anguibactin uptake by V. anguillarum. Characterization of FatC (A), FatD (B) and FatB (C). Vibrio anguillarum strains grown to exponential phase (OD₆₀₀ = 0.3–0.6) in CM9 were washed twice and resuspended in casamino acid-free CM9 containing the chelator sodium nitrilotriacetate at a concentration of 100 μM. The anguibactin siderophore was loaded with ⁵⁵Fe by incubation with ⁵⁵FeCl₃(1 μCi ml⁻1) for 6 h and then mixing it with an equal volume of V. anguillarum in CM9 salts. At each time point 1 ml of mixture was withdrawn, filtered through a 0.45 μm filter (Millipore Corporation) and immediately washed twice with10 ml of 100 mM Sodium citrate. The filters were air-dried and the radioactivity was measured in a liquid scintillation counter. The values were normalized to an OD (OD₆₀₀ = 1), and results were fitted using the GraphPad Prism4 program.

Fig. 5

The FatB protein attaches to the cytoplasmic membrane. A. Western blots of proteins separated by sucrose density gradient centrifugation. Sucrose density gradient centerifugation was conducted as described by Nikaido (1994) with modification. One litre of V. anguillarum 775 overnight culture in CM9 broth was harvested and resuspended in 20 ml Hepes buffer (pH 7.4). After French pressure treatment and centrifugation, 1 ml of total protein was layered on top of a sucrose gradient (0.25 ml of saturated sucrose, 1.5 ml of 2.02 M sucrose, 5 ml of 1.44 M sucrose and 3 ml of 0.77 M sucrose) in the 14 × 89 mm polyallomer centrifuge tubes (Beckman). After 20 h ultracentrifugation at 4°C in the SW28 rotor (Beckman) at 100 000 g each 1 ml of fractions was collected from the top of the tube, and 3 μl (for FatA detection) and 15 μl (for FatB detection) of samples were used for Western bots. B. NADH oxidase activity of fractions obtained from sucrose gradient sedimentation. The NADH oxidase activity was measured as described by Osborn and colleagues (1972) with modification. Samples (30 μl) was mixed with 240 μl of assay buffer (50 mM Tris-HCl (pH 7.5), 0.2 mM dithiothreitol). After 5 min incubation at 25°C, 30 μl of 1.2 mM NADH was added, and the decrease in OD₃₄₀ was measured at 25°C.

Fig. 6

Western blot detection of FatB in cell fractions of V. anguillarum carrying either the wild-type FatB or C23A FatB. Periplasmic proteins were extracted as described by Wunderlich and colleagues (1993) with some minor modifications. Briefly, an exponential phase bacterial culture in CM9 minimal medium with 10 μg ml⁻1 Cm and 0.5 mM IPTG was centrifuged. The pellets were washed with CM9 minimal medium and resuspended in 2 ml BBS/EDDA (200 mM boric acid/NaOH, pH 8.0, 160 mM NaCl, 1 mM EDTA) per gram cell. The suspension was incubated for 45 min at 4°C with gentle agitation, and centrifuged (27 000 g, 1 h, 4°C) to pellet the spheroplasts. The supernatant containing periplasmic proteins was carefully transferred into new tubes. To obtain total protein the bacterial cells prepared as described above were resuspended into 10 mM Tris buffer (pH 7.6), sonicated 5 × 5 s and centrifuged at 15 000 r.p.m. at 4°C for 5 min. Supernatants were transferred into new tubes and used as total proteins. Membrane proteins were obtained by centrifuging the total proteins for 1 h at 30 000 g, and after resuspension this step was repeated. The presence of FatB or C23A FatB in the periplasm fractions and in total proteins was determined by Western blotting using anti-FatB serum.