Introduction

The genus Shewanella comprises several Gram-negative species which are widely distributed in marine and freshwater environments. Shewanella algae (formerly classified as S. putrefaciens ) has been frequently isolated from marine water samples and spoiling fish [13]. They are capable of reducing trimethylamine N-oxide (TMAO) to trimethylamine and producing hydrogen sulfide, both of which are main components of the fishy odor present during low temperature storage. They were also isolated from human feces, skin and other clinical samples [4, 5]. The collected strains were heterogeneous with G + C values ranging from 43 % to 55 %. However, there were differences between environmental and clinical isolates. Most of the strains isolated from human clinical specimens and identified as S. putrefaciens showed beta-hemolysis on sheep blood agar whereas environmental strains were nonhemolytic [6, 7]. During a screening study of heterotrophic bacteria from the sediment of Arcachon Bay [8], a large set of isolates was obtained from different sampling sites and years (2). Among the 24 isolates, 15 strains belong to Shewanella genus and were able to reduce Mn(III/IV) and/or nitrate. The genus seems to play an important role in the turnover of organic matter coupled to anaerobic respiration electron acceptors, such as Fe(III), Mn(III/IV) and NO2/3. Here we report on further taxonomic and physiological studies on strain Shewanella algae strain C6G3 and present its main genomic features.

Organism information

Classification and features

The genus Shewanella currently contains 62 species [9] including Shewanella algae ATCC 51192T, the first described [2]. Ribosomal gene of strain C6G3 exhibits 99 % similarity with available ribosomal gene of Shewanella algae (strains ATCC 51192T, ACDC [3], BrY [10] and FeRed [11]) and was affiliated to this specie (Fig. 1). Cells of strain C6G3 are straights rods (Fig. 2), Gram-negative, motile, free-living and non-sporulating. Different growth temperatures, pH and % NaCl were tested (Table 1). Optimal growth occurs at 30 °C, pH 8 and 10 % NaCl (w/v). For strain C6G3, ATCC 51192T and S. oneidensis MR-1T [12], the use of 95 carbon sources was tested with Biolog GN2 microplate™ (Microlog) (Additional file 1: Table S1). Strain C6G3 presents 32 positive results: 5/30 carbohydrates, 14/29 organic acids, 8/19 amino-acids. Similar results were obtained for strain ATCC 511392T (38/95). Among the two strains of S. algae , slightly different patterns of carbon source were noticed; however, profil of S oneidensis MR-1T was different (16/95). Some electron acceptors were also tested for strain C6G3 according to genome annotation and Shewanella algae literature [1] (Additional file 1: Table S1).

Fig. 1
figure 1

Phylogenetic position of Shewanella algae C6G3 relative to the genus Shewanella and other strains of Shewanella algae. This Neighbor-joining tree is based on 1243 aligned characters of the 16S rRNA gene. The bootstrap percentages higher than 50 % are indicated at the node after 1000 resampled data sets. Branch length corresponds to sequence differences as indicated on the scale bar (substitutions per position). The proposed Shewanella species have been chosen from the List of Prokaryotic names with Standing in Nomenclature (type strain and sequence accession number) for their ability to use nitrate, nitrite and / or metal oxides. Species whose genome has been sequenced are marked with star (*). Escherichia coli [44] was used as out-group

Fig. 2
figure 2

Phase contrast micrograph of Shewanella algae C6G3. Bar scale: 2 μm

Table 1 Classification and general features of S. algae C6G3 [17]

Chemotaxonomic data

The fatty acid analysis was performed on two strains of S. algae (C6G3, ATCC 51192T) and on S. oneidensis MR-1T (Additional file 1: Table S2). At the end of aerobic culture, fatty acids were extracted from cell pellet by alkaline hydrolysis and analyzed using chromatography-electron ionization mass spectrometry (GC-EIMS) following the protocol described in Zabeti et al., [13]. The overall fatty acid pattern of S. algae C6G3 is rather common for the genus Shewanella . The major ones were C16:1ω7 (35.2 %), C16:0 (34.6 %) (generally reported between 16–55 % and 5–31 %, respectively [14]) and 3OH-C12:0 (7.7 %). Interestingly, the C15:0 br is much lower in strain C6G3 (2.3 %) than in S. algae ATCC 51192T and S. oneidensis MR-1T (27.4 % and 20.4 %, respectively). S. algae C6G3 presents also a relatively higher percentage of short-chain fatty acids (shorter than C15, 16.0 %) than both S. oneidensis MR-1T and S. algae ATCC 51192T (8.5 %).

Genome sequencing information

Genome project history

S. algae C6G3 was isolated from intertidal marine sediment on the basis of its ability to use large range of electron acceptors particularly nitrate, nitrite and metal-oxides [8]. The genome of S. algae C6G3 is the second to be reported from that species, the other one being S. algae ACDC [15]. The genome project of S. algae C6G3 is deposited in the Genome On Line Database [16]. A summary of the project and information on compliance with MIGS version 2.0 [17] are shown (Table 2).

Table 2 Genome project information

Growth conditions and DNA isolation

S. algae C6G3 was grown aerobically at 30 °C under stirring condition on artificial sea water [18] amended with lactate (3 g/L), yeast extract (1 g/L) and tryptone (5 g/L). DNA was extracted from cells collected in exponential growth phase using the protocol of Marteinsson et al. [19]. DNA concentration and purity were checking on biophotometer® (Eppendorf) before sequencing.

Genome sequencing and assembly

The genome sequencing of S. algae C6G3 was generated at the Molecular Research LP MR DNA Laboratory (USA). De novo whole-genome shotgun sequencing was performed using the Ion Torrent PGM (Life Technologies [20]) sequencing platform. This produced 1,444,981 reads with an average length of 200 bp for a total number of sequenced bases of 288,996,200 representing a sequencing depth of 50-fold. The assembly of S. algae C6G3 genome was generated at MR DNA Laboratory using the SeqMan NGen® software assembler (DNASTAR). The final assembly identified 43 contigs generating a genome size of 4,9 Mb.

Genome annotation

Genome annotation was performed on two platforms: on RAST [21] and on IMG/ER [22] (DOE Joint Genome Institute [23]). The tRNAscan-SE tool [24] (RAST and IMG/ER) was used to find tRNA genes, whereas ribosomal RNAs were detected using RMAmmer [25] (IMG/ER) and tool “search_for_rnas” (developed by Niels Larsen (available by the author), RAST). Open Reading Frames (ORFs) were predicted using GLIMMER2 [26] in RAST and using GeneMark (v.2.6.r) [27] and GenePRIMP [28] as a part of the DOE-JGI genome annotation pipeline. Gene prediction analyses and functional annotations were performed in RAST with a series of BLAST against FIG hands-curated subsystems [29]. They were also analyzed with FIGfams collection databases, and through comparative approaches with Integrated Microbial Genome – Expert Review platform (RPS-BLAST, BLAST, BLASTp, Hmmsearch (HMMER)) against non-redundant databases including COGs, Pfam [30], TIGR-fam [31], KEEG [32], IMG. Additional functional annotations were performed within the SEED framework (RAST) [33] and the IMG/ER (DOE-JGI) platform.

Genome properties

The assembly of non-contiguous finished draft genome consists of 43 contigs representing overall 4,879,425 pb. The DNA G + C content was 53.08 %. Using RAST and IMG/ER, 5770 and 5795 genes were respectively predicted. Among them 4149 and 5660 protein-coding genes and 108 and 135 RNAs were identified by RAST and IMG/ER, respectively. The properties and the statistics of the genome (IMG/ER data) are summarized in Table 3 and Fig. 3. Putative COG functions were assigned for 39.28 % of the protein-coding genes. The distribution of genes into COGs functional categories is presented in Table 4.

Table 3 Genome statistics of S. algae C6G3 (IMG/ER DOE-JGI)
Fig. 3
figure 3

Graphical map of 6 contigs containing ORF involved in nitrate, nitrite and metal oxides utilization. Nitrate reduction (contig #203 and #65), dissimilative reduction of nitrite into ammonium, (contig #82 and #35) and metal reduction (contig #105 and #166). From bottom to the top: genes on forward strand (color by COG), genes on reverse strand (color by COG), operon/gene cited (pointed red), RNA genes (tRNAs green, rRNAs red, other RNAs black), GC content, GC skew

Table 4 Number of genes associated with general COG functional categories

Insights into the genome sequence

Genome of S. algae C6G3 encodes genes for complete glycolysis and tricarboxylic acid (TCA) cycle. A focus has been made on the enzymes involved in the reductive respiratory reactions of the N-cycle (NAP, NRFA) and in extracellular electron transfer through the outer-membrane (Metal Transfer Reducing such as Fe(III) and Mn(III/IV)) on the basis of protein system described in S. oneidensis MR-1T.

Nitrate reduction

Nitrate respiration involves two distinct enzyme systems: the NapAB localized in the periplasm and the membrane-bound nitrate reductase NarGHI enzyme localized on the cytoplasmic face of the cytoplasmic membrane. The sole nitrate reductase of S. algae C6G3 is NapAB (Fig. 4). As most of the Shewanella species, S. algae C6G3 genome encodes the two NAP isoforms, each comprising three catalytic subunits: NapA where nitrate reduction takes place, a di-haem cytochrome NapB and a maturation chaperone NapD. The two isoforms present different membrane-intrinsic subunits [34] named NAP-α (NapEDABC) and NAP-β (NapDAGHB). NAP-β (NapDAGBH) possesses NapGH, an iron–sulfur cluster ferredoxins instead of NapC. The functional differences between these systems may be explained by differential regulation in the composition of the available quinol pool. S. oneidensis MR-1T encodes only Nap-β isoform in which NapC is lacking (Fig. 4). This membrane-anchored tetrahem c-Cyt mediates electron transport from the quinol pool to NapB. The function of NapC in NAP-β in S. oneidensis MR-1T may be met by CymA, an homologue of periplasmic tetrahem c-Cyt of the NapC/NirT family, which is also found in S. algae C6G3 genome.

Fig. 4
figure 4

NAP complex functioning. The proposed electron-transfer pathway of periplasmic nitrate reductase and membrane bound electron donors of S. algae C6G3 (NAP-α-β) (a) and S. oneidensis MR-1T (NAP-β) (b). Colored proteins are annotated from KEGG and putatively functional

Dissimilative reduction of nitrite into ammonium

Nitrite can be reduced to ammonium (NH4) by a periplasmic nitrite reduction system (NRF) [35]. As NAP systems, there are two types of NRF: NrfABCD and NrfAH types. NrfA is the terminal reductase while NrfBCD/NrfH are responsible for electron transfer from menaquinol pool to NrfA. Shewanella strains are known for encoding NrfABCD system only. Genome annotations of S. algae C6G3 and S. oneidensis MR-1T identify nrfA in both cases but nrfBCD were found in S. algae C6G3 only (Fig. 5). As previously described for NAP system, S. oneidensis MR-1T genome lacks genes encoding for the specific compounds that deliver electrons to the terminal reductase (nrfBCD). Indeed, nrfB is lacking and nrfCD are present but proposed to be pseudogene because of truncation [36]. Gao et al., [37] suggested that the tetraheme c-Cyt CymA, a cytoplasmic membrane electron transport protein, is likely to be the functional replacement of both NapC and NrfBCD/NrfH allowing to NrfAH-like system to be efficient in S. oneidensis MR-1T.

Fig. 5
figure 5

NRF complex functioning. The proposed electron-transfer pathway of dissimilatory nitrite reduction to ammonium of S. algae C6G3 (a) and S. oneidensis MR-1T (b). Colored proteins are annotated from KEGG and putatively functional. Streaked genes are annotated but probably not functional

Metal oxide reduction

S. algae C6G3 can utilize extracellular mineral metal oxides of Fe(III) and Mn(III/IV) as respiratory electron acceptors (unpublished data). Inspection of its genome confirmed the presence of genes involved in pathway (i.e. metal-reducing or MTR pathway) for transferring electrons from the inner membrane through the periplasm and across the outer membrane where metal oxides are reduced [38]. In S. algae C6G3 genome, genes that encode MtrCBA and OmcA are located in the same region, which also includes mtrD (an mtrA homologue), mtrE (an mtrB homologue) and mtrF (an mtrC homologue) (Fig. 6).

Fig. 6
figure 6

MTR pathway functioning. The proposed MTR extracellular electron-transfer pathway of S. algae C6G3 (a) and S. oneidensis MR-1T (b). Colored proteins are annotated from KEGG and putatively functional

Table 5 shows the BLASTP for amino acid sequences of MTR pathway in S. algae C6G3 versus those in metal-reducing S. oneidensis MR-1T .

Table 5 BlastP of MTR gene of S. algae C6G3 against MTR gene of S. oneidensis MR-1

The numbers of genes found in the MTR clusters of the analyzed Shewanella strains varies from four, such as omcA1-mtrC-mtrA-mtrB in Shewanella frigidimarina [11], to nine, such as mtrD-mtrE-mtrF-omcA1-undB-omcA1-mtrC-mtrA-mtrB in Shewanella halifaxensis [39, 40]. CymA identified as the entry point for electrons into the MTR pathway [41] is not located in the MTR gene cluster in S. algae C6G3 as described in S. oneidensis MR-1T. Furthermore, S. algae C6G3 has an additional protein encoding for decahem c-Cyt of the OmcA/MtrC family. The role of this cytochrome is not defined. On RAST platform, this ORF has been annotated as mtrH in S. algae C6G3 and S. halifaxensis HAW-EB4T.

Conclusion

The Shewanella genus comprises a diverse group of facultative anaerobes. Their ability to couple the oxidation of various carbon sources to the reduction of a broad range of terminal electron acceptors imparts a respiratory flexibility that allows colonization of varied and changeable marine and freshwater environments [39, 42, 43]. The occurrence of the two different NAP operons, NRF, and that of CymA in S. algae C6G3 accords with the renowned anaerobic respiratory flexibility of Shewanella . S. algae C6G3 is also capable of using solid Fe(III) and Mn(III/IV) as terminal electron acceptors. Reduction of these particulates occurs at the cell surface and is catalyzed by multihaem cytochromes whose properties are beginning to emerge.