Agar is a complex polysaccharide composed of agarose and agaropectin that originates from marine red algae such as Gelidium latifolium, G. amansii and Gracilaria verrucosa (Duckworth and Yaphe, 1970). Agar is an abundant natural resource and has been used as a food for several hundred years in oriental nations. At present, agar has a wide variety of uses due to its stabilizing and gelling characteristics. It has been used in microbiological media and for gel electrophoresis because agar is difficult for microorganisms to metabolize and because it forms clear, stable, and firm gels. Agar is also categorized as a “generally recognized as safe” (GRAS) food additive and is used in icings, glazes, processed cheese, jelly sweets, and marshmallows (Fu and Kim, 2010).

Agarolytic enzymes can be divided into two groups depending on their mode of action, α- and β-agarases, which produce agaro-oligosaccharides and neoagaro-oligosaccharides, respectively (Suzuki et al., 2002). Neoagaro-oligosaccharides possess various biological activities, including antitumor activity, macrophage-stimulating activity to increase immune function, anti-oxidizing activity, probiotic activity, and a skin moisturizing and whitening effect (Rhee et al., 2010). The agaro-oligosaccharide produced by α-agarase possesses anti-oxidative and anti-cancer activities (Ohta et al., 2005).

Some researchers have investigated the possibility of obtaining agarolytic enzymes from bacteria (Ohta et al., 2005; Lee et al., 2007). Although many genera of agarolytic bacteria have been identified, mainly from marine environments, since the first isolation from seawater by Gran in 1902, novel and more active agarolytic enzymes are required for industrial applications (Rhee et al., 2010). Therefore, to investigate the possibility of valuable industrial applications, we isolated a novel agar-degrading bacterium from seawater and characterized its culture properties and enzymatic activity.

Seawater, seaweeds, and mud sampled from the seashore in Busan, Korea, were used for bacterial isolation. Media for isolation were marine broth 2216 and nutrient broth, which was prepared with 0.4 μm-filtered seawater and 1.5% agar. One hundred microliters of each sample was spread on an agar plate and incubated at 37℃ for five days. Lugol’s iodine solution (2 g KI and 1 g I2 in 200 mL water) was then sprayed onto the agar medium. Bacterial colonies that formed a clear zone were selected as agarolytic bacteria. Selected colonies were stored at -80℃ with glycerol until required. Media used in this study were obtained from Difco Co. (Detroit, MI, USA) and chemicals were purchased from Sigma Co. (St. Louis, MO, USA).

The isolate was identified by 16S rRNA sequencing and scanning electron microscopy. Chemicals and reagents for DNA manipulation were purchased from Takara Co. (Tokyo, Japan) and SolGent Co. (Daejeon, Korea). Chromosomal DNA purification for 16S rRNA sequencing was performed using the method of Berns and Thomas (1965). The DNA oligonucleotide primers were 27F (5'-AGAGTTTGATCCTGGCTCAG-3') and 1492R (5'-GGCTACCTTGTTACGACTT-3') and were synthesized by SolGent Co. (Dunbar et al., 2000). The reaction mixture (final volume, 50 μL) contained Taq polymerase 0.5 μL (2.5 U), Taq polymerase buffer 5 μL (Χ10), 2.5 mM dNTP 4 μL, ddH₂O 34.5 μL, 20 pmol each primer 2 μL, and template DNA 2 μL. PCR conditions consisted of an initial denaturation at 94℃ for 5 min, followed by 30 cycles at 94℃ for 30 s, 55℃ for 30 s, and 72℃ for 90 s, with a final 7 min extension at 72℃ . The amplification products were examined by 1% agarose gel electrophoresis, purified with a QIAquick PCR cleanup kit (Qiagen Inc., Valencia, CA, USA), and directly sequenced with the same primers that were used for PCR (SolGent Co.). The resulting sequence data were analyzed using the Basic Local Alignment Search Tool (BLAST) program with the ribosomal RNA gene sequence from GenBank (www.ncbi.nlm.nih.gov/BLAST). Morphological observations were performed using a scanning electron microscope (Hitachi S-2400; Hitachi, Tokyo, Japan).

Nutrient broth (Difco Co.) with an additional 3% NaCl was used to culture the isolate. To determine the optimal temperature and pH for growth, the strain was cultured at various temperatures between 20 and 80℃ at pH 7, and at pH 5.0-10.0 at 37℃ . Cultures were incubated at 150 rpm for 24 h and the absorbance at 600 nm was measured.

The isolate was cultured in Nutrient broth (pH 6.0) containing 0-10% (w/v) NaCl at 30℃ for 24 h and the absorbance at 600 nm was measured.

The effect of agar concentration on growth of the isolate was determined. The isolate was cultured at 30℃ for 30 h at 150 rpm in Nutrient broth containing 0-0.4% (w/v) agar with 3% NaCl at pH 6.0 and the absorbance at 600 nm was measured every 3 h.

Thin-layer chromatography (TLC) of agar hydrolysate was performed on a Silica Gel 600 glass plate (F254; Merck Co., Darmstadt, Germany). To confirm the agarolytic activity of Microbulbifer sp. SD-1, culture supernatant containing agar hydrolysate from 24 h culture broth with 0.3% (w/w) agar was applied. Five microlitres of culture supernatant were spotted and developed with n-butanol-acetic acid-ddH₂O (2:1:1, v/v) for 1 h. Oligosaccharides were detected by spraying with ethanolic sulfuric acid (375 mL ethanol+100 mL sulfuric acid) followed by drying at 110℃ for 10 min (Duckworth and Yaphe, 1970). Galactose as a monosaccharide and lactose as a disaccharide were used as standards.

A bacterial strain that produced an extracellular agarolytic enzyme was isolated from seawater near Busan, Korea. Bacterial cells that can hydrolyze agar create a shallow depression or liquefaction around colonies on a 1.5% agar plate. Additionally, Lugol’s iodine solution was used to confirm the presence of haloes surrounding the colonies. Unstained haloes around colonies on agar plates treated with iodine show that the properties of the gel have been changed by the action of agarase diffusing out from the bacteria. The binding of iodine to agar gels to give a colored complex depends on the structure of the gel matrix. As shown in Fig. 1, the isolate produced extracellular agarase, resulting in a noticeable halo around the colony after staining with Lugol’s iodine solution.

For taxonomic identification, the 16S rRNA sequence (1,457 bp) was determined and compared with sequences available in the GenBank database using the BLAST search program. Microbulbifer maritimus strain TF-17 showed the greatest homology (97% identity), followed by M. salipaludis and M. elongatus (95 and 94% identity, respectively) (Table 1). Scanning electron microscopy showed that the isolate is rod-shaped, non-flagellum, and 0.5-0.7Χ2.0-2.5 μm in size, orgwhich are typical properties of Microbulbifer sp. (Fig. 2) (Jonnadula et al., 2009). Therefore, the isolate was renamed Microbulbifer sp. SD-1.

The genus Microbulbifer was first proposed by Gonzalez

[Fig. 1.] Staining of agar using Lugol’s iodine solution on Marine agar medim plate presenting the diffusion of extracellular agarase.

[Table 1.] 16s rRNA sequence (1457 bp) and identification of the isolated strain SD-1 by homology search based on 16S rRNA

which are typical properties of Microbulbifer sp. (Fig. 2) (Jonnadula et al., 2009). Therefore, the isolate was renamed Microbulbifer sp. SD-1.

The genus Microbulbifer was first proposed by Gonzalez

[Fig. 2.] Scanning electron microscopy of strain showing typical shape of Microbulbifer sp. SD-1.

et al. (1997) for Microbulbifer hydrolyticus, a Gram-negative, rod-shaped, strictly aerobic, and biopolymer-decomposing marine gammaproteobacterium. Since then, ten species of the genus have been isolated from various marine sources, including seawater, seaweeds, mud, sandstone, sediments, and solar saltern (Yoon et al., 2004; Jonnadula et al., 2009). One of the prominent features of these bacteria is their ability to degrade more than one insoluble complex polysaccharide, such as agar, carrageenan, cellulose, xylan, alginate, and chitin. Microbulbifer degradans 2-40, the only strain reported to degrade 10 polysaccharides, has been reclassified as Saccharophagus degradans, and various industrial applications are under development (Ensor et al., 1999).

The growth rate of Microbulbifer sp. SD-1 at various temperatures was measured (Fig. 3). Microbulbifer sp. SD-1 showed maximum growth at 30℃ , and 86, 81, 80, and 68% of this maximum growth rate at 37, 40, 50 and 60℃ , respectively. No or little growth was observed at 20 and 70℃ . The optimal growth temperature of Microbulbifer sp. SD-1 was

[Fig. 3.] Effect of temperature on the cell growth by Microbulbifer sp. SD-1.

[Fig. 4.] Effect of pH on the cell growth by Microbulbifer sp. SD-1.

similar to that of M. elongatus, which was 25-30℃ . The maximal temperature for growth of Microbulbifer sp. SD-1 was 60℃ (68% of maximum growth rate), which is much higher than that in homologues. The maximal temperatures for growth of M. maritimus and M. salipaludis, which showed the highest identity based on 16S rRNA analysis, were 48 and 45℃ , respectively. These data suggest that Microbulbifer sp. SD-1 has thermophilic properties and raises the possibility of thermophilic agarase production by this isolate (Jonnadula et al., 2009).

The optimal pH for growth of Microbulbifer sp. SD-1 was determined using culture media pH-adjusted to 5.0-10.0 (Fig. 4). The data indicated that pH 6.0 was optimal, and that 92-74% of the maximum growth rate occurred at pH 7-9. No or little growth was observed below pH 5.0 or over pH 10.0. These results are similar to those for M. maritimus (optimal pH 6.5-7.5), meaning that Microbulbifer sp. SD-1 is a neutrophilic bacterium (Yoon et al., 2004; Jonnadula et al., 2009). Accordingly, the growth medium was adjusted to pH 6.0.

The effect of NaCl concentration on the growth of Microbulbifer sp. SD-1 was determined (Fig. 5). Growth rate increased with NaCl concentration from 0-5% and then rapidly decreased at NaCl concentrations of 6% or greater; little growth occurred at 9-10% NaCl. The optimum NaCl concentration for growth of Microbulbifer sp. SD-1 was 5%, which is similar to those (3-5% NaCl) of many halophilic marine bacteria. Around 45% of the maximum growth rate occurred in the absence of NaCl. The optimal NaCl concentration for growth of M. maritimus was 2-4% and no growth occurred in the absence of NaCl. However, growth was evident at 10% NaCl, illustrating the various NaCl tolerances of Microbulbifer species.

[Fig. 5.] Effect of NaCl on the cell growth by Microbulbifer sp. SD-1.

[Fig. 6.] Growth kinetics of Microbulbifer sp. SD-1 at various agar concentration in batch culture codition (● without agar; ○ 0.2% agar; ■ 0.4% agar).

[Fig. 7.] Thin-layer chromatography analysis of agarolytic products hydrolyzed from agar by Microbulbifer sp. SD-1. Gal; galactose Lac; lactose 24 h; 24h culture supernatant 12h; 12 h culture supernatant.

Agar, a type of polysaccharide, can be a source of carbohydrate for strains that can hydrolyze agar into smaller saccharides. The effect of agar on growth rate was determined (Fig. 6). No effect on growth rate was evident at more than 30 h culture or at the time to reach the stationary phase at 12 h, with or without agar in the medium. These results suggest that Microbulbifer sp. SD-1 may use agar as a secondary carbon source, after other nutrients were exhausted. Most Microbulbifer strains, including M. salipaludis and M. elongatus but not M. maritimus, have been reported to exhibit agarase activity (Yoon et al., 2004; Jonnadula et al., 2009).

To confirm the agarolytic activity of Microbulbifer sp. SD-1, supernatant from a 24 h culture with 0.3% (w/w) agar was applied to a silica gel TLC plate along with galactose and lactose as standards (Fig. 7). When the plate was developed and visualized, two spots were detected. Microbulbifer sp. SD-1 produced a monosaccharide and an oligosaccharide longer than lactose after 12 h culture. After 24 h culture, it produced mainly a monosaccharide similar to galactose. These results are similar to those for Cellvibrio mixtus SC-22, in that production of monosaccharide increased with culture time, and different from those of Agarivorans sp. JA-1 which produced only a hexamer (Lee et al., 2006; Rhee et al., 2010). Whether Microbulbifer sp. SD-1 produces α - and/or β-agarase remains unclear; however, this novel strain may have industrial applications due to its rapid, thermophilic monosaccharide production. Research on the characteristics, over-expression, activity on other marine polysaccharides, and optimal monosaccharide production conditions of this strain and agarase will continue.