Partial characterization of an exopolysaccharide secreted by a marine bacterium, Vibrio neocaledonicus sp. nov., from New Caledonia

Authors


Correspondence

Eleftherios Chalkiadakis, IFREMER, Centre de Brest, BRM/BMM, BP 70, 29 280 Plouzané, France. E-mail: echalkiadakis@pasteur.nc

Abstract

Aims

Exopolysaccharides (EPS) are industrially valuable molecules with numerous useful properties. This study describes the techniques used for the identification of a novel Vibrio bacterium and preliminary characterization of its EPS.

Methods and Results

Bioprospection in marine intertidal areas of New Caledonia followed by screening for EPS producing brought to selection of the isolate NC470. Phylogenetic analysis (biochemical tests, gene sequencing and DNA–DNA relatedness) permitted to identify NC470 as a new member of the Vibrio genus. The EPS was produced in batch fermentation, purified using the ultrafiltration process and analysed by colorimetry, Fourier Transform Infrared spectroscopy, gas chromatography, Nuclear Magnetic Resonance and HPLC-size exclusion chromatography. This EPS exhibits a high N-acetyl-hexosamines and uronic acid content with a low amount of neutral sugar. The molecular mass was 672 × 103 Da. These data are relevant for possible technological exploitation.

Conclusions

We propose the name Vibrio neocaledonicus sp. nov for this isolate NC470, producing an EPS with an unusual sugar composition. Comparison with other known polymers permitted to select applications for this polymer.

Significance and Impact of the Study

This study contributes to evaluate the marine biodiversity of New Caledonia. It also highlights the biotechnological potential of New Caledonia marine bacteria.

Introduction

Marine bacteria, whether attached or free-living, are found in almost all parts of the marine world, from the Arctic Ocean to deep-sea hydrothermal vents (Nichols et al. 2005). They have developed various strategies to survive in extreme conditions, such as metabolic pathways adaptation including the production of protective structures like biofilms. Biofilms are highly hydrated matrix formed by extracellular polymeric substances such as exopolysaccharides (EPS), proteins, lipids and extracellular DNA (Flemming et al. 2007).

Polysaccharides, commonly found in plants, are also produced by some micro-organisms where they act as nutrient stores components in the cell walls (Work 1961); some are secreted in form of EPS. Polysaccharides have numerous interesting properties relevant to industrial applications; they can be used as stabilizers, emulsifiers, thickeners, gelling agents, coagulating agents and for their water retention capacity (Rasmussen et al. 2007). They are currently used in diverse applications in various sectors including food and beverage industries (Suresh Kumar et al. 2007), health industry (Okutani 1984, 1992) and industrial waste treatment and mining industries (Iyer et al. 2004, 2005). Polysaccharides can be chemically modified to impart or improve particular functional properties (Guezennec et al. 1998a) further increasing their exploitation potential. Bacteria present the advantage to grow rapidly and can be used for rapid fermentative production of polysaccharides. The polymers produced by batch fermentation are highly reproducible from a batch production to another; they are not sensitive to marine pollution, crop failure or climatic events. Consequently, these systems are valuable biotechnological resources. Bacterial polysaccharides have a wide variety of chemical arrangement and properties that are not found elsewhere, they often outperform polysaccharides from other sources, for example algae (alginates, carrageenans) or crustacean (chitin) (Rasmussen et al. 2007).

Bacteria excrete polysaccharides that then forms a protective coat on their cell surfaces against environmental stresses including variations in abiotic factors such as temperature, pH, salinity or UV (Boyle and Reade 1983). EPS also play a role in protection against antibiotics (Høiby et al. 2009) and heavy metals (Gutnick and Bach 2000). Indeed, hydrated biofilms offer a buffered microenvironment in which bacteria can live, adapt, resist to stressful conditions and thereby ensure their survival and proliferation (Decho and Herndl 1995).

Aiming at the discovery of novel biopolymers and biomolecules of biotechnological significance, it is now widely accepted that micro-organisms from unusual environments not only provide valuable resources for exploiting novel biotechnological processes but also serve as models for investigating how biomolecules are stabilized when subjected to changing conditions.

Bioprospection in microbial mats presents on some French Polynesian atolls in the Pacific Ocean led to the discovery of new bacterial species able to produce biopolymers with unusual chemical structures (Moppert et al. 2009). In that Pacific Ocean seems to be a promising source of new bacterial molecules and other marine biotechnological developments, such as novel drugs and healthcare products (Guezennec et al. 2011). A similar programme started in New Caledonia in April 2010 with the aim to isolate marine bacteria and evaluate their biotechnological potential.

As part of this programme, this work describes a new strain, designated NC470 and able to produce under laboratory conditions an EPS. A preliminary chemical characterization of this biopolymer is also reported.

Materials and methods

Bioprospection and sampling

New Caledonia is a Melanesian archipelago in the South West Pacific. It possesses one of the largest lagoons in the world and has exceptionally diverse marine areas, including reef flats, mangrove swamps and sandy beaches. During bioprospection, samples of waters, sediments, intertidal rocks, invertebrates, plants, fish and biofilms found on inorganic substrates (Guezennec et al. 1998b) were collected, mainly in the western part of the main island of New Caledonia (Fig. 1). Strain NC470 was isolated from a biofilm found on an invertebrate animal (Holothuroidea) in St-Vincent Bay (S 21°55′ 631″/E 166°04′ 840″).

Figure 1.

Map of New Caledonia showing the sites sampled during the 2010 bioprospection programme.

Isolation and culture of the bacterial strain

The samples collected were used to inoculate liquid Zobell (ZoBell 1941) medium at pH 7·6 and incubated at 28°C for 24 h. Successful cultures were subcultured on a solid medium under the same conditions; this procedure was repeated several times until pure cultures were obtained, and the isolates were stored at −80°C in 20% glycerol (v/v).

Physiological and biochemical characterization

Tests to identify optimal growth conditions were performed in 20-ml tubes with 8 ml of Zobell liquid medium. The tubes were inoculated (10%, v/v) with a primary bacterial culture in the exponential phase of growth and incubated with rotary shaking (Unitron, Infors, Massy, France) at 200 rev min−1. The environmental conditions at the sampling site include a temperature range of 28–30°C; pH 7–8; and, 37–38‰ salinity. Thus, the growth rates were determined by studying the following range of culture conditions:

  1. Temperature from 20 to 45°C, with 5°C steps.
  2. pH from 6 to 9, with 0·5 steps, in 50 mmol l−1 MOPS (3-(N-morpholino) propanesulfonic acid, pH 6–6·5) and Tris (pH 7–9) buffers.
  3. Salinity from 20 to 70 g l−1 with 10 g l−1 step using NaCl.

Bacterial growth was measured every hour by spectrometry (Uvikon XS, Secoman, Ales, France) at λ = 600 nm, until the end of the exponential phase of growth.

API 50CH, API 20NE, API 20E, ATB G- and ATB PSE tests kits were used to determine metabolic properties and antibiotic susceptibilities.

Phylogenetic analyses

Genomic DNA was isolated using the QIAmp® kit (QIAGEN S.A., Courtaboeuf, France) and the extracted DNA was adjusted to 20 ng μl−1. The 16S ribosomal RNA (16S rRNA) gene sequence was amplified by PCR using two primers: SADIR (5′-AGAGTTTGATCATGGCTCAGA-3′) and S17REV (5′-GTTACCTTGTTACGACTT-3′) (Cambon-Bonavita et al. 2002). The PCR involved of a first denaturation at 95°C, then a 35 cycles of 94°C for 2 min, 55°C for 30 s and 72°C for 90 s, followed by a final extension of 72°C for 7 min and ended with cooling to 4°C (Verity; Applied Biosystems, Pasteur Institute of New Caledonia, Noumea, New Caledonia).

The primers described by Sawabe (Sawabe et al. 2007) were used for multilocus sequence typing (MLST) study using the housekeeping genes gapA, gyrB, pyrH and topA. The PCR programme used was the same as for 16S rRNA gene sequence but with annealing temperature of 58, 53, 53 and 58°C for gapA, gyrB, pyrH and topA, respectively.

Sequence reads on the ABI 3130 xl were made at the regional genomic core research facilities for life science in New Caledonia ‘Plate-Forme du Vivant de Nouvelle-Calédonie (PFV-NC)’.

The 16S rRNA gene sequence obtained was compared with the GeneBank sequence database (http://blast.ncbi.nlm.nih.gov) using the BLAST algorithm to determine the bacterial genus of NC470. Ninety-six type strains from this genus (Balcázar et al. 2010; Beaz-Hidalgo et al. 2010; Rameshkumar et al. 2010; Sheu et al. 2010; Chimetto et al. 2011a; Wang et al. 2011; Gomez-Gil et al. 2012) were used for phylogenetic analyses, type strain DNA sequences of were aligned with that of NC470 using the BioEdit software package (Hall 1999).

The phylogenetic analyses were performed using a combination of three methods and distances corrected according to the Kimura two-parameter model (Kimura 1985). The neighbour-joining (Saitou and Nei 1987), maximum parsimony (Kluge and Farris 1969) and maximum-likelihood (Felsenstein 1992) methods were implanted in the Phylo_win program (Galtier et al. 1996) with bootstrap values determined after 500 replications. The tree was constructed with the njplot program (Perrière and Gouy 1996).

The strains and sequence Accession Numbers of 16S rRNA, gapA, gyrB, pyrH and topA genes of the species most closely related to NC470 are listed in Table 1.

Table 1. Gene sequence accession numbers used for phylogenetic study
SpeciesStrain16S rRNA gapA gyrB pyrH topA
  1. a

    Sequences submitted to GeneBank data base.

Vibrio alginolyticus ATCC 17749X56576DQ907274AB298202FM202578DQ907472
Vibrio azureus LC2-005AB428897 
Vibrio campbellii ATCC 25920X56575EF596565EU130500FM202551DQ907475
Vibrio cholerae CECT 514TX76337DQ907273FM202624FM202582DQ907478
Vibrio communis R-40496GU078672AB609125GU078680GU078691GU078704
Vibrio diabolicus HE800X99762JQ934821aJQ972871aJF739414JQ934822a
Vibrio harveyi NCIMB1280TAY750575EF596145AB298221FM202541GQ428295
Vibrio jasicida TCFB 0772TAB562592AB562597AB562598AB562600AB562602
Vibrio mytili CECT 632X99761DQ907293AB298231GU266287DQ907499
Vibrio natriegens ATCC 14048TX74714DQ907294AB298232FM202573DQ907500
Vibrio parahaemolyticus ATCC 17802AF388386DQ449618AB298239EU118240JQ934827a
Vibrio rotiferianus LMG 21460AJ316187DQ449619AB298244FM202568DQ907515
Vibrio sagamiensis LC2-047AB428909 
Vibrio xiamenensis G21TGQ397859 
Vibrio neocaledonicus sp. nov NC470JQ934828aJQ934823aJQ934823aJQ934825aJQ934826a

The DNA base composition was determined by a fluorimetric method involving quantitative PCR (Light-Cycler 2 from Roche diagnostic, Auckland, New Zeland): the thermal denaturation of DNA was determined, in duplicate, as described elsewhere (Xu et al. 2000; Gonzalez and Saiz-Jimenez 2002). The Tm values of 12 Vibrio type species with a known percentage G+C were used to construct a standard curve equation that allows the percentage G+C of NC470 to be determined. Using these results, DNA–DNA hybridization study was performed between NC470 and the three most closely related type strains as assessed from the 16S rRNA gene sequences and the concatenated MLST gene study. The method used was based on the study of Vibrio genus performed by Moreia (Moreira et al. 2010) and the work of De Ley (De Ley 1970) for optimal DNA denaturation (Tor). Aliquots of 10 μl of each DNA preparation (adjusted to 50 ng μl−1) were mixed and heated to 99°C for 10 min. The samples were then incubated for 8 h at 70°C and the temperature decreased 10°C per hour to 25°C. The samples were then cooled to 4°C and tested by quantitative PCR apparatus as explained previously to obtain the denaturation curve. The DNA–DNA relatedness was determined from the ∆Tm value for the hybrid formed between the DNAs from NC470 and the type strain.

Production of exopolysaccharides

EPS was produced by culturing NC470 for 72 h in a 3·5 l fermenter (Minifors, Infors, Massy, France) containing 3 l of Zobell medium, supplemented with 30 g l−1 glucose.

Each batch of culture medium was inoculated at 10% (v/v) with a suspension of cells in exponential growth. The temperature was maintained at 30°C, and the pH was adjusted to 7·5 by automatic addition of NaOH (0·2 mol l−1) or HCl (0·2 mol l−1). Foaming was avoided by adding Struktol (40 μl l−1). The airflow was set to 30 l h−1, and the agitation rate was controlled to maintain dissolved O2 at c. 40% saturation.

Isolation, purification and characterization of the EPS

Cells were removed from the medium after 72 h of culture by centrifugation at 2000 g (Sorvall Evolution RC, Saint Herblain, France). The EPS was then purified from the culture supernatant by ultrafiltration using a Sartocon Slice system (Sartorius Stedim Biotech, Aubagne, France) equipped with a 100-kDa filter and lyophilized.

Chemical analyses

The total neutral carbohydrate and hexuronic acid content were determined by the phenol-sulfuric (DuBois et al. 1956) and the meta-hydroxydiphenyl (Blumenkrantz and Asboe-Hansen 1973) methods, respectively. Hexosamines were assayed by staining with the Ehrlich reagent and spectrophotometry (Belcher et al. 1954) using N-acetyl-glucosamine and N-acetyl-galactosamine as standards. Protein content was determined by the BCA method (Lowry et al. 1951) with bovine serum albumin as the standard.

The molar ratio of monosaccharides was estimated after acidic methanolysis of the EPS using 3 mol l−1 of MeOH/HCl, 4 h at 100°C, followed by gas chromatography (GC) analyses of the trimethylsilyl derivatives. Methyl glycosides were converted to trimethylsilyl derivatives as described elsewhere (Montreuil et al. 1986) and analysed by GC using myo-inositol (1 mg ml−1) as the internal reference (Rougeaux et al. 1996, 2001). GC analysis was performed on an Agilent Technologies GC 7890A Series fitted with a HP1-fused silica column (60 m × 0·32 mm) with helium as carrier gas. The temperature programme was 50°C for 1 min, then increasing 20°C per min up to 120°C, 2°C per min up to 250°C, 50°C per min up to 280°C and 5 min at 280°C.

The attenuated total reflectance Fourier Transform Infrared spectroscopy (ATR-FTIR) spectrum was recorded on a Nicolet iS10 ThermoScientific (Saint Herblain, France) FTIR spectrometer. The exopolysaccharide sample was deposited on a germanium disc. Scans were performed in the wave frequency range of 600–4000 (per cm).

Nuclear Magnetic Resonance (NMR) analysis

The EPS was dissolved in D2O for 1D-NMR analysis in a Bruker DRX 500 spectrometer equipped with a 5 mm TBI 1H/(BB)13C probehead at 50°C. The chemical shifts are expressed in parts per million with reference to the external standard TSP (tetrasilyl propionate deutero acid).

High-performance size exclusion chromatography

Size exclusion chromatography (SEC) experiments were carried out at room temperature in water (0·1 mol l−1 NaNO3 + 50 mg l−1 NaN3 as bacteriostatic agents) using a Thermo Sep pump operating at a flow rate of 1 ml min−1 in combination with an automatic Gilson 234 injector (injection volume = 0·3 ml with polymer concentration c. 0·2 g l−1). A dual flow refractive index detector (RI171 from Shodex, Pays de la Loire, France) and a UV detector (UV2000 from Spectra Physics) operating at 280 nm were used for detection. A column set constituted of one PW6000 TSK column (17 μm, 7·5 × 600 mm) and of a guard DUPONT zorbax Bio series GF 450 column (9·4 × 250 mm) filled with hydroxylated polyether to prevent polymer adsorption was used for separation. The columns were calibrated with pectin. Molecular weights were read from the refractometer trace and are expressed as pectin equivalents.

Results

Bioprospection and sampling

Sampling areas were selected to be physically and chemically diverse; a collection of 493 marine bacterial isolates was obtained. Those isolates were then screened for their ability to produce EPS: 58% of the isolates produced EPS, and 10% showed a very mucoid aspect on marine agar medium supplemented with 30 g l−1 glucose.

Morphology

One of the strains isolated, strain NC470, is a mobile, facultative anaerobe, and a nonluminescent, nonpigmented and Gram-negative rod. Three-day-old colonies on Zobell-glucose were opaque, smooth and gummy colonies of about 1·2 cm in diameter. In the absence of glucose, hemi-translucent white swarming colonies measured about 0·9 cm in diameter formed after 3 days of growth.

Cultural characteristics

Strain NC470 was selected for its ability to show a mucoid phenotype after 2 days on Zobell medium supplemented with glucose 30 g l−1 at 28°C.

The optimal growth conditions were identified as being a temperature of 35°C, a pH of 7 and a salinity of 50 g l−1.Under these conditions, the doubling time was 30 min.

Metabolic properties

The API 50CH, API 20NE, API 20E kits showed positive responses for: catalase, oxydase, reduction of nitrate to nitrite, indole production, gelatinase, cytochrome oxidase, lysine decarboxylase, acetoïne production and ornithin decarboxylase. The biochemical and nutritional tests showed that the NC470 isolate can use a wide range of carbohydrate substrates as the sole carbon source: N-acetyl-glucosamine, gluconate, malate, inositol, potassium glucuronate, d-glucose, d-mannitol, sucrose and amygdalin.

NC470 was resistant to ampicillin, ticarcillin, ticarcillin–clavulanic acid, piperacillin, cefalotin, cefuroxime, tobramycin and colistin as determined by ATB G- and ATB PSE.

Phylogenetic analyses

BLAST analysis of the 16S rRNA gene sequence indicated that NC470 belonged to the gamma subdivision of the Proteobacteria phylum and was included a member of the Vibrionaceae family, order Vibrionales. The 16S rRNA gene sequence was aligned with and compared with 96 type Vibrio strains. The three phylogenetic methods used (neighbour-joining, maximum parsimony and maximum likelihood) placed NC470 in a group that includes Vibrio natriegens as the most closely related species (Fig. 2a) with a percentage of similarity of 95·7%. MLST with four housekeeping genes of NC470 showed a similarity with other Vibrio species between 71·4 and 90·9%, 61·3 and 95·4%, 61·9 and 81·8% and 65·8 and 95·3%, for the gapA, gyrB, pyrH and topA genes. A phylogenetic tree was constructed with the concatenated sequence of 16S rRNA, gapA, gyrB, pyrH and topA genes (Fig. 2b) in which NC470 is closest to Vibrio diabolicus (92·2% similarity) and Vibrio alginolyticus (90·4% similarity).

Figure 2.

Neighbour-joining phylogenetic tree using a subset of the Vibrio genus and showing the position of NC470 on basis of the 16S rRNA gene sequence (a) and a second tree based on the analysis of a concatenated 16S rRNA, gapA, gyrB, pyrH and topA gene sequence analysis (b). Evolutionary distances were evaluated using the Kimura-2 parameter. Bootstrap values are indicated at nodes, and the level is based on 500 repetitions. Vibrio cholerae was used as an out-group to root the tree.

The 16S rRNA gene (1420 bp) gapA (727 bp), gyrB (851 bp), pyrH (535 bp) and topA (711 bp) sequences of NC470 and the V. diabolicus gapA (717 bp) and topA (657 bp), and Vibrio parahaemolyticus topA (662 bp) used for this phylogenetic analysis were deposited in NCBI Genebank. The Accession Numbers are provided in Table 1.

The percentage G+C was estimated by the Tm value of Vibrio type strains that led to a standard curve conforming to the equation percentage G+C = 0·3677 Tm + 15·682 such that the percentage G+C of NC470 was estimated to be 45·9% ± 0·21 (mean Tm = 82·4 ± 0·56) (Fig. S1). The DNA–DNA relatedness between NC470 and the three closest species was determined by hybridization: the ∆Tm values were 5·7°C for V. alginolyticus, 6°C for V. diabolicus and 7·6°C for V. natriegens (Fig. S2).

Production of the EPS

During batch fermentation of NC470, EPS production began at the end of the exponential phase of growth and continued during the entire stationary phase. The concentration of EPS was reaching 2 g l−1 (dry weight) at the end of the experiment (72 h).

Characterization of the EPS

The gross chemical composition of the EPS produced by NC470 is listed in Table 2 as well as some other EPS from unusual environments (Loaëc et al. 1997; Cambon-Bonavita et al. 2002; Ortega-Morales et al. 2007; Guezennec et al. 2011). Hexosamines predominated with concentrations up to 40·5%, while uronic acids and neutral sugars accounted for 10 and 5% of the total sugars, respectively. The low amount of protein (2·5%) indicated a good efficiency of the purification protocol applied to this polymer.

Table 2. Gross chemical composition of NC470 exopolysaccharide and comparison with other micro-organisms (% w/w)
EPSProteinHexosaminesHexuronic acidsNeutral sugarsReferences
  1. EPS, exopolysaccharides.

NC4702·540·5105Present study
Vibrio diabolicus HE8002·533322·5Loaëc et al. (1997)
Vibrio sp (Mo 245)2302711Guezennec et al. (2011)
Microbacterium sp (MC6B-22)8·921·1514·75·5Ortega-Morales et al. (2007)
Alteromonas infernus 5·53751Cambon-Bonavita et al. (2002)

FTIR on the EPS produced by strain NC470 showed a broad band beyond 3000 cm−1 resulting from O-H and C-H bond stretching absorption bands at 3291 and 2930 cm−1, respectively, and an intense absorption band at 1652 cm−1 with a shoulder at 1732 cm−1 due to the presence of carboxylic groups on the uronic acids. Moreover, the bands at 1558 and 1645 cm−1 were assigned to amino I and amino II groups of osamines, as confirmed by the low amount of protein (2·5%). The absence of a doublet at 1250–1230 cm−1 indicated that no sulfate groups were present in this EPS (Fig. 3a).

Figure 3.

The chemical analysis performed on the exopolysaccharides produced by NC470 exposing four panel data with a) Fourier Transform Infrared spectroscopy spectra, b) gas chromatography spectra, c) 1D RMN spectra and d) size exclusion chromatography plots.

GC analysis of the monosaccharides as per-O-trimethylsilyl methylglycosides showed N-acetyl glucosamine, N-acetyl-galactosamine, glucose, galacturonic acid and glucuronic acid as the main constituents in a molar ratio of 1·5 : 0·7 : 1·6 : 1·1 : 0·3 (Fig. 3b). The 1D NMR spectra of the NC 470 EPS in native state showed a complex anomeric region with at least four signals and confirmed the presence of a methyl at 2·08 ppm of acetyl groups, which can be assigned to N-acetylation of the hexosamines (Fig. 3c).

HPSEC showed that the polysaccharide molecular weight (Mw) was of c. 672 × 103 Da, with a polydispersity (Ip) of 1·7, thus indicating an homogeneous biopolymer (Fig. 3d). One should note that the lack of low-molecular-weight polymer might be explained by the ultrafiltration method used that selected polymers higher than 100 kDa.

Discussion

Bioprospection in the New Caledonian lagoons aimed to constitute a bank of marine bacteria isolated from these atypical environments. During this programme, more than 490 bacterial isolates were collected from various sources along the west coast of New Caledonia.

The bacterial isolate NC470 was obtained from the surface of an invertebrate integument. On the basis of morphological and biochemical data, this aero-anaerobic, mesophilic and heterotrophic bacterium clearly belongs to the Vibrio genus. Phylogenetic analyses of 16S rRNA gene sequences demonstrate that strain NC470 was different from any other known Vibrio species with <97% of sequence identity the closest species. According to Stackebrandt (Stackebrandt and Goebel 1994), this degree of identity is clearly lower than the cut-off defining interspecific relationships, and therefore, strain NC470 appears to be a novel Vibrio species.

MLST with four housekeeping genes – gapA, gyrB, pyrH and topA (Thompson et al. 2005; Sawabe et al. 2007) – confirmed that NC470 is a new Vibrio species. According to Thompson et al. (2009), a Vibrio species is defined as a group of strains that share >95% of DNA identity in MLST. The concatenated 16S rRNA, gapA, gyrB, pyrH and topA gene sequences clearly defined a monophyletic cluster including NC470, V. alginolyticus and V. diabolicus with which it displays similarity of 90·4 and 92·2%, respectively. The study of concatenated gene for phylogenetic analysis has been previously described in recent article on Vibrio genus (Chimetto et al. 2011a,b; Gomez-Gil et al. 2012). The percentage G+C of NC470 was estimated as being 45·9% calculated using type strain denaturation curve. DNA–DNA relatedness was further investigated by measuring the ΔTm value for hybrids between NC470 and type strains denaturation curves. All the hybrids obtained showed a ∆Tm above 5°C. According to the recommendation of the ad hoc committees of 1987 and 2002 (Wayne et al. 1987; Stackebrandt et al. 2002) the cut-off for interspecies discrimination for this type of study is 5°C. Therefore, we suggest that NC470 should be considered to be a new member of the Vibrio genus for which the name Vibrio neocaledonicus is proposed. The type strain V. neocaledonicus NC470 has been deposited in the Collection de l'Institut Pasteur (Institut Pasteur, Paris: reference number CIP 110538T).

This study was part of a larger project to assess the potential of New Caledonian marine bacteria for biotechnological applications. Under nonoptimized culture conditions, the yield of EPS reached 2 g l−1 dry weight. EPS synthesized under laboratory conditions by strain NC470 was mainly characterized by high content of hexosamines (as N-acetyl-glucosamine and N-acetyl galactosamine) accounting for up to 40% of the EPS dry weight. To date, only a few marine bacteria have been described to produce EPS with hexosamines as the major constituents with the exception of members of the Vibrio genus. Vibrio diabolicus strain HE800 (Rougeaux et al. 1996) isolated from a Alvinella pompejana worm tube collected from a deep-sea hydrothermal field of the East Pacific Rise (Raguenes et al. 1997) produced an EPS characterized by equal amounts of uronic acid and hexosamines (N-acetyl glucosamine and N-acetyl galactosamine) in a molar ratio of c. 1 : 0·5 : 0·5, respectively. EPS from V. alginolyticus, a marine fouling bacterium revealed the presence of glucose aminoarabinose, aminoribose and xylose in the molar ratio of 2 : 1 : 9 : 1 (Muralidharan and Jayachandran 2003). More recently, a new EPS bacterial producer was isolated from a microbial mat located on atolls in French Polynesia. This biopolymer was characterized by the presence of N-acetyl glucosamine, N-acetyl galactosamine, glucuronic acid, glucose and galactose in a molar ratio of 2 : 1 : 1 : 0·1·0·2 (J. Guezennec, personal data).

Vibrio parahaemolyticus was shown to produce an EPS consisting of four major sugars, that is, glucose, galactose, fucose and N-acetylglucosamine (Enos-Berlage and McCarter 2000, Kavita et al. 2011). Vibrio cholerae is also an EPS-producing bacterium with as major components N-acetyl-d-glucosamine, d-mannose, 6-deoxy-d-galactose and d-galactose in a molar ratio of 7·4 : 10·2 : 2·4 : 3·0 (Wai et al. 1998). EPS from other Vibrio sp such as Vibrio harveyi (Bramhachari and Dubey 2006) and Vibrio vulnificus (Reddy et al. 1992) also were also characterized by the presence of amino sugars.

The presence of high content of hexosamines and uronic acids raise interesting questions about the biotechnological potential of (Gomez-Gil et al. 2012) such EPS. The high-molecular-weight linear bacterial exopolysaccharide produced by V. diabolicus strain HE 800 appears to be a strong bone-healing material. Introducing this new glycosylaminoglycan-like (GAG) polysaccharide in critical size defects in bone in rats induces a nearly complete healing within no longer than 2 weeks with a total restore of the anatomy of the defect with trabecular and cortical structure. This hyaluronic acid-like EPS constitutes a material that potentiates bone repair and its particular activity has to be related its original physicochemical characteristics (Zanchetta et al. 2003a,b). Exopolysaccharide produced by strain NC 470 also has a composition similar to both other well-known biologically active glycosaminoglycans (GAGs) and to the extracellular matrix. Both of these bioactive molecule polysaccharides and the extracellular matrix are known to play a major role in the first healing steps and during bone or injury healing (Jerdan et al. 1991; Esford et al. 1998; Lipscombe et al. 1998).

Mining represents the major economic activity in New Caledonia (Dalvi et al. 2004) and generates residues laden with heavy metals that are released into the environment (http://www.goronickel-icpe.nc/global/pages/02-impacts/section). Therefore, alternative solutions to heavy metal discharge would be of great interest from an industrial and ecological point of view. In that bacterial EPS can serve as an alternative source of low-cost biosorbents. EPS contain ionizable functional groups such as carboxyl, amine, sulfate and to a lesser extent hydroxyl groups that enable these biopolymers to bind heavy metals (Loaëc et al. 1997).

Preliminary experiments conducted on EPS produced NC470 strain showed a high metal binding capacity with values up to 370 mg Cu(II) g−1 EPS and up to 70 mg Ni(II) g−1 EPS. Similar high uptake capacities were found for hexosamine-rich exopolysaccharide produced by a marine bacterium isolated from a microbial mat in a Polynesian atoll with 400 mg Cu(II) g−1 EPS and 65 mg Ni(II) g−1 EPS, respectively (Guezennec et al. 2011). Further structural elucidation will allow a better understanding of the mechanism of this affinity and also determine the potential commercial value of this EPS as a biosorbent for a variety of heavy metals.

Acknowledgements

This study was partly supported by a grant from the Ministry of Overseas France (Secretariat d'Etat à l'Outre-Mer), no 09/1217976/BF. The authors would like to thank the following people for participating in the bioprospection programme (April 2010): Laurane Palandre, Marlène Vic, Etienne Pita, Yannick Labreuche and Dominique Ansquer [IFREMER Lagons, Ecosystèmes et Aquaculture Durable (LEAD/NC)]. We thank Cyrille Goarant (Laboratoire de Recherche en Bactériologie, Institut Pasteur de Nouvelle-Calédonie) for advice, helpful discussions technical support in regard to the phylogenetic work and for his comments on the manuscript. We also would like to thank Laurent Millet in charge of the sequencer ‘Plate-Forme du Vivant de Nouvelle-Calédonie (PFV-NC)’. Finally, we thank Dominique Girault for technical support and Carmen Taxier for preparing media.

Ancillary