Agar is a kind of complex polysaccharide that exists widely in the marine environment and is hard to degrade. Flammeovirga pacifica WPAGA1, which is isolated from deep-sea sediment of west Pacific Ocean (157° 24′ 31″ E 19° 30′ 30″ N) at the depth of 5,378 m, was cultured as described previously (Xu et al., 2012). To evaluate the ability of the bacterial isolate to utilize agar, F. pacifica WPAGA1 was grown on modified 2216E medium that contains 0.2% yeast extract, sea water and 2% agar. Then, we observed agar degradation termly. After 36 h, the medium was processed using 2% (m/v) iodine staining.

For additional analysis, the agarose-utilizing ability of the crude extracellular enzymes of F. pacifica WPAGA1 were prepared. The WPAGA1 strain was cultured at 28°C for 48 h in 1,500 mL of 2216E medium, and the following preparation was performed on ice. The cultures were centrifuged at 12,000 × g for 15 min, and ammonium sulfate was added to the supernatant to reach 80% saturation. After 4 h incubation to precipitate proteins, the mixture was centrifuged at 15,000 × g for 30 min. The pellet was then dissolved in 10 mL 10 mM PBS buffer (pH 7.4), and the resultant suspension was dialyzed three times against the same buffer for 6 h. Then, the dialyzed enzyme solution was stored in 20% glycerol (v/v) at −20°C. The extracellular agarase system in the crude enzyme extracts was analyzed to identify the agarase components. Crude enzyme extracts (8 μL each) were each mixed with 4 μL 3 × loading buffer and separated by 12% non-denaturing polyacrylamide gel electrophoresis (PAGE). After electrophoresis, the gel was washed three times with 10 mM PBS buffer (pH 7.4) for total of 30 min. The gel was then overlaid onto a 2% agar sheet and incubated overnight at 37°C. The zymogram was analyzed by iodine staining, and a partial gel was developed using silver staining as contrast material.

To further investigate the abilities of other complex polysaccharides-degradation of strain WPAGA1, arabinogalactan, amylum, sodium alginate, colloidal chitin, carrageenan, carboxymethyl cellulose (CMC), fucoidan, heparin, pectate, porphyran, xylan were used as sole carbon source for the growth of strain WPAGA1 (2 g/L (NH₄)SO₂ and 10 g/L of each tested polysaccharide in sea water). The OD₆₀₀ was detected after 24 h of culturing at 28°C and 200 rpm to determine whether strain WPAGA1 could utilize the tested polysaccharide as the carbon source for its growth. Moreover, to directly detect the enzymatic activities of strain WPAGA1 toward different polysaccharides, the secreted extracellular proteins from the WPAGA1 culture were collected and examined for their polysaccharides-degrading activities. In brief, the WPAGA1 strain was cultured at 28°C for 48 h in modified 2216E medium (0.2% yeast extract, 1% peptone, and 0.5% each tested polysaccharide in sea water), and the crude extracellular proteins were prepared as mentioned above. The enzymatic activities of crude extracellular proteins toward various polysaccharides were assayed by detecting the release of the reducing sugar from the corresponding substrates using the 3,5-dinitrosalicylic acid (DNS) method (Miller, 1959) with a slight adaptation. The standard reaction contained 100 μL protein solution and 900 μL of PBS buffer (pH 7.4) containing 0.2% (w/v) of each tested polysaccharide. After incubation at 28°C for 10 min, the reaction was stopped by immersion in boiling water for 5 min. The heat-inactivated crude extracellular proteins were used as negative controls. Following this, 250 μL of the reaction solution was mixed with 750 μL of DNS reagent and heated for 10 min in a boiling water bath and then cooled. The absorbance of reducing sugar was measured at 540 nm. Enzyme activity (U) was defined as the amount of enzymes that liberated 1 μmol of reducing sugar per minute under the assay conditions.

The cells of strain WPAGA1 were collected from the 2 mL culture by centrifugation at 10,000 × g for 10 min. Then, the pellet was washed by sterilized TE buffer twice to remove the residual medium before DNA extraction. The genomic DNA of strain WPAGA1 was extracted using a bacterial DNA kit (OMEGA) and subjected to quality control by agarose gel electrophoresis and quantified by Qubit. The obtained DNA was verified to be of high quality (DNA amount: ≥20 μg and purity: 1.8 ≤ OD260 nm/280 nm ≤ 2.0).

Library construction and sequencing was performed at the Beijing Novogene Bioinformatics Technology Co., Ltd. The genome of strain WPAGA1 was sequenced with massively-parallel-sequencing Illumina technology. Two DNA libraries were constructed, namely, a paired-end library with an insert size of 500 bp and a mate-pair library with an insert size of 5 kb. The 500 bp and 5 kb libraries were sequenced using an Illumina HiSeq2500 apparatus by the PE125 strategy. Quality control of both paired-end and mate-pair reads were performed using an in-house program. After this step, the Illumina PCR adapter reads and low quality reads were filtered. The filtered reads were then assembled by SOAPdenovo (Li et al., 2008, 2010) (http://soap.genomics.org.cn/soapdenovo.html) to generate scaffolds. All reads were used for further gap closure.

Transfer RNA (tRNA) genes were predicted with a tRNAscan-SE (Lowe and Eddy, 1997), ribosomal RNA (rRNA) genes were predicted with rRNAmmer (Lagesen et al., 2007) and sRNAs were predicted by BLAST against Rfam (Gardner et al., 2009) database. Repetitive sequences were predicted using RepeatMasker (Saha et al., 2008) (http://www.repeatmasker.org/). Tandem repeats were analyzed using Tandem Repeat Finder (Benson, 1999). PHAST (Zhou et al., 2011) was used for prophage prediction (http://phast.wishartlab.com/) and CRISPRFinder (Grissa et al., 2007) was employed for CRISPR (clustered regularly interspaced short palindromic repeat sequences) identification. This Whole-Genome Shotgun project was deposited at the DDBJ/EMBL/GenBank database under the accession number JRYR00000000.

Gene prediction was performed for the F. pacifica WPAGA1 genome assembly by GeneMarkS (Besemer et al., 2001) (http://topaz.gatech.edu/) with an integrated model that combines GeneMarkS generated (native) and Heuristic model parameters. A whole-genome BLAST (Altschul et al., 1990) search (E-value less than 1e−5, minimal alignment length percentage larger than 40%) was performed against six databases. These databases were KEGG (Kanehisa, 1997; Kanehisa et al., 2004, 2006) (Kyoto Encyclopaedia of Genes and Genomes), COG (Tatusov et al., 1997, 2003) (Clusters of Orthologous Groups), NR (Non-Redundant Protein Database databases), Swiss-Prot (Magrane and UniProt Consortium, 2011), GO (Ashburner et al., 2000) (Gene Ontology) and TrEMBL (Magrane and UniProt Consortium, 2011). A whole-genome BLAST (Altschul et al., 1990) search (E-value less than 1e−5, minimal alignment length percentage larger than 40%) was performed against four databases for pathogenicity and drug resistance analysis. These databases included PHI (Vargas et al., 2012) (Pathogen Host Interactions), VFDB (Chen et al., 2012) (Virulence Factors of Pathogenic Bacteria), ARDB (Liu and Pop, 2009) (Antibiotic Resistance Genes Database), and CAZy (Cantarel et al., 2009) (Carbohydrate-Active enZYmes Database). Secretory proteins were then detected on the genome assembly by SignalP (Petersen et al., 2011). Types I–VII secretion system related proteins were extracted from all the annotation results. Type III secretion system effector proteins were also detected by EffectiveT3 (Jehl et al., 2011). Secondary metabolite gene clusters were predicted by antiSMASH (Medema et al., 2011; Blin et al., 2013). All these annotation files were further combined into one table. Genome overview was created by Circos (Krzywinski et al., 2009) to show annotation information. Searches for the polysaccharide-related metabolic enzymes, including glycoside hydrolases (GHs), glycosyl transferases (GTs), carbohydrate esterases (CEs), polysaccharide lyases (PLs), and auxiliary activities (AAs) were performed manually on the basis of the BLASTP and the CAZy database. These metabolic pathways of polysaccharides were deduced on the basis of such enzymes and the KEGG database.

As reported, all bacteria belonging to genus Flammeovirga can degrade complex polysaccharides, especially the agar that derived from the cell wall of red seaweeds. To investigate the polysaccharide-degrading ability of F. pacifica WPAGA1, the agar-degrading profile was first examined. After a 36 h culture in modified 2216E medium plate, strain WPAGA1 spread and formed clear, etched colonies. The plate was stained with iodine solution, and some clear agar degradation zones developed around individual colonies (Figure 1A). The solid medium where strain WPAGA1 was cultured became liquid after 4 days of incubation at 28°C (Figure 1B). Thus, these culture characteristics suggest that WPAGA1 hold potent agarolytic activities.

To further investigate the agar-degrading ability of the strain WPAGA1, extracellular enzymes were extracted from a 48 h culture at 28°C. F. pacifica WPAGA1 produced various extracellular proteins (Figure 1C). Moreover, the crude enzymes were run using native PAGE to estimate the agarase components in the extracellular proteins. The iodine staining of the agar gel showed the presence of at least six transparent bands on the developed sheet in Figure 1C. The molecular masses of these agarase bands mainly lay between 35 and 100 kDa. These results show that the strain WPAGA1 harbors a powerful extracellular agarase system for degrading agar.

Besides of the agar, the capabilities of strain WPAGA1 to degrade other complex polysaccharides were further examined. As showed in Table 1, most of the tested polysaccharides could be utilized for the growth of strain WPAGA1 as the sole carbon source. Strain WPAGA1 grew best on alginate (OD₆₀₀ = 1.868), followed by arabinogalactan, carrageenan, porphyran, fucoidan, cellulose, amylum, and xylan. However, strain WPAGA1 could not grow on heparin and pectate. To confirm the above results, the extracellular proteins secreted by strain WPAGA1 were prepared from the culture supernatant, and their enzymatic activities toward the various tested polysaccharides were assayed by the DNS method. As showed in Table 1, strain WPAGA1 exhibited enzymatic activities toward most of the tested polysaccharides except of heparin and pectate, which was consistent with the above results. Therefore, our results showed that the strain WPAGA1 harbors powerful enzymes for degrading various complex polysaccharides, which would enable it to adapt to various environments.

To further explore the genetic basis for the ability of complex-polysaccharide degradation, the genome of F. pacifica WPAGA1 was sequenced. By using an Illumina HiSeq2500 apparatus and PE125 genome sequencing strategy, we constructed the following two DNA libraries: a paired-end library with an insert size of 500 bp and a mate-pair library with an insert size of 5 kb. As a result, a total of 625 Mb (500 bp) and 1,317 Mb (5 kb) reads were obtained. The generated sequences were assembled into 65 contigs with an N₅₀ length of 445,839 bps, and these contigs were assembled into 21 scaffolds. Accordingly, the fine genome of F. pacifica WPAGA1 consists of 6,610,326 bases with 32.89% G+C content, 333 interspersed repeated sequences, 229 tandem repeats finder, 83 tRNA genes, 5 rRNAs (two 5S rRNA, two 16S rRNA, and one 23S rRNA), 8 genomics islands and especially, 5 questionable CRISPRs (Figure 2, Table 2). A total of 5,036 ORFs, which account for 87.37% of the total nucleotides, were obtained. Amongst these genes, 3,189 protein-coding sequences (63.32% of the total) were annotated and identified by NR search with GenBank sequences. In addition, 957, 1,660, 1,090, 1,487 and 2,983 proteins were functionally annotated from UniProtKB/Swiss-Prot, KEGG, COG, GO, and UniProtKB/TrEMBL databases, respectively. The detailed results are shown in Tables 2, 3.

The F. pacifica WPAGA1 genome contains a number of regions, such as genomic island, with atypical GC content. Eight genomic islands were found, and the average length was 16,651 bp. Genomic island (GIs) 1 (GIs1, 2,634,781 bp to 2,663,274 bp, Scaffold1) comprises a cluster of 29 genes with products predicted to be involved in tRNA-modifying proteins. Interestingly, three genes that encode proteins showed the highest similarity to protein HigA, phage antirepressor protein and plasmid maintenance system killer protein belonging to phage genes or plasmid gene sequences. GIs 4 (3,802,724–3,584,014 bp, Scaffold1) carries a gene cluster for Clp protease ClpC. GIs 7 (314,546–333,456 bp, Scaffold2) comprises a series of genes with products predicted to be involved in stress-resistant proteins, such as the acriflavin resistance protein, tellurium resistance protein, beta-lactamase and arsenic resistance protein ArsB. Some compound genes, such as the cobalamin synthase gene, spore coat protein gene, ABC transporter genes, transposase genes and rich phosphatase genes were found in these genomic islands.

The F. pacifica WPAGA1 genome harbors a considerable number of genetic elements, such as 333 interspersed nuclear elements and 88 non-coding RNA. Especially, five regions of nucleotide sequences with products predicted to be CRISPR-associated genes were found. These CRISPR-associated genes show that F. pacifica WPAGA1 was invaded by exogenous DNA or RNA, such as phage.

A distinguishing feature of the Flammeovirga genus is the capability to produce a series of enzymes for polysaccharide hydrolysis or lysis. To search for genes related to polysaccharide-degrading enzymes in the genome of F. pacifica WPAGA1, the CAZy database was used for annotating carbohydrate-related genes. A total of 1,022 carbohydrate-related genes were annotated by the CAZy database (Figure 3). Amongst these genes, 483 genes belong to glycoside hydrolases (GHs), followed by 249 genes identified with glycosyl transferases (GTs) and 87, 35, and 6 genes that belong to carbohydrate esterases (CEs), polysaccharide lyases (PLs) and auxiliary activities (AAs), respectively. Meanwhile, 218 proteins contain carbohydrate-binding modules (CBM) defined as contiguous amino-acid sequences within a carbohydrate-active enzyme with a discreet fold having carbohydrate-binding activity. Bacteria catalyze polysaccharide depolymerization by secreting an array of lyases or glycosidase that generate oligosaccharides or monosaccharides. These bacteria then undergo a complex series of random or regular cleavage reactions. The major components of polysaccharide-degrading enzymes are GHs and PLs. Simultaneously, 483 putative GH proteins and 35 putative PL proteins were found in the F. pacifica WPAGA1 genome (Figure 3). These masses of carbohydrate-active proteins comprise a complex system for carbohydrate catabolism in F. pacifica WPAGA1. We next investigated the detailed polysaccharide-degrading pathways in the WPAGA1 strain with respect to various polysaccharides on the basis of genomic analysis. Detailed gene information related to polysaccharide metabolism is shown in Table 4 and Table S1.

As described above, substantial amounts of polysaccharide-degrading enzymes are found in F. pacifica WPAGA1. Abundant sugar source is produced by the polysaccharide-degrading enzyme system. Our data show that the pentose phosphate pathway (PPP), pentose and glucoronate interconversion, fructose and mannose metabolism, galactose metabolism and starch and sucrose metabolism constitute the main pathways that enable F. pacifica WPAGA1 to utilize these sugars for growth. We deduced a schematic of glycometabolism in F. pacifica WPAGA1 (Figure 5).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.