Since methicillin was introduced in 1959 to resolve infec-tions caused by penicillin-resistant Staphylococcus aureus, the incidence of methicillin resistance in staphylococci has increased rapidly (Kaplan, 2005). As a result of widespread methicillin use, methicillin-resistant S. aureus (MRSA) has become a major problem globally (Lee et al., 2008). MRSA infections are difficult to treat because of the multidrug-re-sistance properties of MRSA, which is resistant to β-lactams as well as several other classes of antibiotics (Bramley et al., 1989; Hiramatsu et al., 1997). Vancomycin is commonly used for the treatment of MRSA-related infection. Vancomy-cin resistance was discovered first in enterococci and later in staphylococci (Isnansetyo and Kamei, 2009). The emergence of vancomycin-resistant S. aureus has recently been recog-nized and leaves physicians with few options available to treat MRSA infections. Therefore, much attention has been given to the search for new antimicrobial agents (Lee et al., 2008).

In an effort to decrease usage of vancomycin and discover an alternative therapeutic agent for treating MRSA infection, we have screened marine algae for anti-MRSA compounds. Eisenia bicyclis is a common perennial phaeophyceae (brown alga) that generally inhabits the coast of Ulleung Island in the East Sea of Korea. This seaweed is used in various dishes, including appetizers, casseroles, muffins, pilafs, and soups (Eom et al., 2011; Kim et al., 2011). The antioxidant activity of E. bicyclis phlorotannins such as eckol (a trimer), phlorofu-cofuroeckol A (a pentamer), dieckol, and 8,8′-bieckol (hexam-ers) has been reported (Okada et al., 2004). This brown algae also exhibits activity against tumors (Noda et al., 1989), Al-zheimer’s disease (Jung et al., 2010), atherosclerosis (Kang et al., 2003), inflammatory diseases (Shibata et al., 2003), aller-gic disease, and cancer (Shibata et al., 2002). However, to our knowledge, no study has reported on the antimicrobial activ-ity of E. bicyclis. Here, we examined the antibacterial activity of E. bicyclis against several pathogenic bacteria, including MRSA.

In late September 2010, the brown seaweed E. bicyclis was gathered from Ulleung Island, Korea. Dried E. bicyclis was ground and then finely powdered using a food mixer (HMF-1000A; Hanil Electronics, Seoul, Korea). The dried powder was stored at -20℃ until required. Powdered E. bicyclis (1 kg) was extracted with methanol (3 times × 10 L) for 3 h. The combined filtrate was concentrated by rotary evaporation at 40℃. After suspending in water (1 L), the methanol extract was partitioned with n-hexane (hexane), dichloromethane (DCM), ethyl acetate (EtOAc), and n-butanol (BuOH), in se-quence.

Standard bacterial strains were obtained from the Ko-rean Collection of Type Cultures (KCTC; Daejeon, Korea) and the Korean Culture Center of Microorganisms (KCCM; Seoul, Korea). The bacterial strains used were methicillin-susceptible S. aureus (MSSA; KCTC 1927), two MRSA strains (MRSA; KCCM 40510 and KCCM 40511), Bacillus cereus (KCTC 3624), B. subtilis (KCTC 1028), Enterococcus faecalis (KCTC 3206), Escherichia coli (KCTC 1682), Lis-teria monocytogenes (KCTC 3710), Salmonella typhimurium (KCTC 1925), and Vibrio parahaemolyticus (KCTC 2729). All strains were grown aerobically at 37℃ in Mueller-Hinton broth (MHB; Difco, Detroit, MI, USA) or tryptic soy broth (TSB; Difco) for determination of the minimum inhibitory concentration (MIC) and in Mueller-Hinton agar (MHA; Dif-co) for disk diffusion and minimum bactericidal concentration (MBC) assays.

Antibacterial activity was evaluated by a disk diffusion assay, as described by the National Committee for Clinical Labo-ratory Standards (National Committee for Clinical Laboratory Standards, 2004). In brief, bacterial strains were cul-tured in TSB at 37℃ until an OD₆₀₀ of 0.5. Bacterial culture (1 mL), containing approximately 10⁴ CFU, was spread on a MHA plate, and a paper disk (6 mm diameter) containing 1 mg extract was then placed on the agar surface. After incuba-tion for 24 h at 37℃, the diameter of the inhibition zone was measured. The determination was performed three times and the mean values presented.

The MIC is the lowest concentration of antimicrobials that will inhibit the visible growth of microorganisms after over-night incubation (Grierson and Afolayan, 1999). MICs of the extracts and vancomycin were determined by the twofold seri-al dilution method in MHB (NCCLS, 2003). MIC was defined as the lowest concentration of crude extract that inhibited vi-sual growth after incubation at 37℃ for 20-24 h, and was per-formed in triplicate (Grierson and Afolayan, 1999).

The MBC value was defined as the lowest concentration of E. bicyclis extracts required for a 99.9% reduction in the viable cell population (Shen et al., 2002; NCCLS, 2003). For determining MBC values, an aliquot (0.1 mL) of MIC mix-tures that showed no growth was inoculated onto MHA plates and incubated at 35℃ for 48 h (Syu et al., 2004).

To identify the active compound(s) in E. bicyclis extracts, high-performance liquid chromatography (HPLC) was per-formed using a Hitachi 2000 series HPLC system (Hitachi Tech, Tokyo, Japan) equipped with Shiseido C₁₈ reverse-phase column (250 mm × 4.6 mm, I.D. 5 ㎛; Shiseido Co., Tokyo, Japan). For detection of the bioactive substance, a linear gradi-ent elution of 90% water with 10 to 100% (v/v) methanol was used at a flow rate of 1.0 mL per min for 45 min. Eluates were monitored at 230 nm.

Standard phlorotannins (dieckol, eckol, and eckstolonol) were used to allow quantification. Standard compound con-centrations were calculated in the range of 0.01 to 0.1 mg by linear regression from the respective calibration curves. Eckol (Y = 12.973X - 0.4261; -r² = 0.9991), dieckol (Y = 17.065X + 1.3499; -r² = 0.9993), and eckstolonol (Y = 10.192X + 0.7048; -r² = 0.9963) standard curves were obtained (data not shown).

E. bicyclis methanolic extract exhibited anti-MRSA activi-ty, suggesting the presence of an antibacterial substance (Table 1). Also, the extract exhibited similar activity against MSSA (Table 1). To identify the antimicrobial substance, the extract was further fractionated using organic solvents. Lyophilized E. bicyclis extract (1.0 kg) was percolated in methanol (3 times × 1,000 mL), followed by fractionation with organic sol-vents to yield hexane- (0.05 g), DCM- (0.05 g), EtOAc- (0.11 g), BuOH ion- (0.39 g), and water-soluble (0.57 g) fractions. The anti-MRSA activity of the hexane, DCM, EtOAc, BuOH, and water-soluble fractions was evaluated by measuring the inhibition zones. Of these, the EtOAc-soluble fraction showed the strongest anti-MRSA activity, followed by DCM, BuOH, and hexane, in that order (Table 1). No anti-MRSA activity was detected in the water-soluble fraction. These results were consistent with the reports of Lee et al. (2008) and Choi et al. (2010) stating that the EtOAc-soluble fraction of Ecklonia stolonifera and Ecklonia cava exhibited the strongest anti-MRSA activity. The antibacterial substance was subsequently identified.

[Table 1.] Growth inhibitory of Eisenia bicyclis extracts against methicillin-resistant Staphylococcus aureus (MRSA) and other strains

[Table 2.] Minimum inhibitory concentrations (MICs) and minimum bactericidal concentrations (MBCs) of Eisenia bicyclis extracts against methicillin-resistant Staphylococcus aureus (MRSA) and other strains

The current study focused on the antibacterial activity of E. bicyclis extracts against MRSA. To quantitatively evaluate this antibacterial activity, we investigated the MIC and MBC values of the extract (Table 2). The highest anti-MRSA and MSSA activities were present in the EtOAc-soluble fraction at 32-64 μg/mL. These results were consistent with those of the disk diffusion assay. The MICs and MBCs of the methanolic extract against MSSA and MRSA ranged from 64 to 256 μg/mL; those of the other fractions (hexane, DCM and BuOH) were between 32 and 64 μg/mL for MICs and 64 and 256 μg/mL for MBCs. However, no antibacterial activity was detected in the water-soluble fraction (Table 2). These results strongly suggest the presence of an anti-MRSA compound in the EtO-Ac-soluble fraction of the E. bicyclis methanolic extract.

Kim et al. (2002) reported that the EtOAc extract of this seaweed showed the highest antibacterial activity against the cavity-causing Gram-positive bacterium Streptococcus mu-tans. Therefore, we investigated the antibacterial activity of E.bicyclis extract against several pathogenic and spoilage bac-teria (Table 2). The EtOAc fraction showed the highest anti-bacterial activity against these bacterial strains. The MICs and MBCs of the EtOAc fraction were 32-256 μg/mL.

Vancomycin, a tricyclic glycopeptide antibiotic, is used to treat MRSA infections (Lee et al., 2008). Vancomycin in-terferes with bacterial cell wall synthesis, as does penicillin, eventually leading to cell lysis (Barna and Williams, 1984). Most Gram-negative bacteria are less sensitive to vancomycin than Gram-positives (Totsuka et al., 1999; Lee et al., 2008). As expected, the MICs of vancomycin against Gram-neg-ative bacteria were over 512 μg/mL compared to 0.5-4 μg/mL against Gram-positives (Table 2). Unlike vancomycin, the EtOAc fraction exhibited strong antibacterial activity (MICs 32-128 μg/mL; MBCs 32-256 μg/mL) against Gram-negative bacteria. These data suggest the existence of different mecha-nisms of inhibiting cell growth and bacterial cell wall synthe-sis.

The biological activities of plant materials are related to their phenolic compound content (Kim et al., 2006; Lin et al., 2008). Such activity is based on the physiological functions of polyphenol polymers (McDougall et al., 2005). Seaweed polyphenol (phlorotannin) is the predominant EtOAc-soluble compound in brown algae (Lee et al., 2008; Choi et al., 2010). Phloroglucinols, such as eckol, phlorofucofuroeckol-A, diek-

col, and 8,8′-bieckol, also exhibit antibacterial activity (Isnan-setyo et al., 2001; Nagayama et al., 2002).

We detected anti-MRSA activity in E. bicyclis methanolic extract. Additionally, the ethyl acetate-soluble fraction exhib-ited the highest antibacterial activity against other pathogenic bacteria, suggesting that the antibacterial compound is abun-dant in the EtOAc fraction (Table 2). To identify the active compound, we conducted HPLC analysis (Figs. 1 and 2). Pu-rified phloroglucinol compounds (dieckol, eckol, and ecksto-lonol) were used as controls. Identification of unknown phlo-roglucinol compounds was achieved by comparing retention time with those of control compounds. The retention times of dieckol, eckol, and eckstolonol were 12.98 ± 0.30, 10.87 ± 0.29, and 21.52 ± 0.45 min, respectively (Fig. 1). HPLC analy-sis showed that only the methanolic extract and the EtOAc-

[Fig. 1.] High-performance liquid chromatography (HPLC) profile of standard phlorotannins. HPLC analysis was performed as described in Materials and Methods. EK 100 μg per mL of eckol; DK 100 μg per mL of dieckol; ES 100 μg per mL of eckstolonol.

[Fig. 2.] HPLC profiles of Eisenia bicyclis extracts after 5 days of fermentation. A B C D E and F are the HPLC profiles for methanol extract n-hexane-soluble extract dichloromethane-soluble extract ethyl acetate-soluble extract n-butanol-soluble extract water-soluble extract. The injection amount was 1000 ppm.

soluble fraction contained sizeable quantities of dieckol (76.9 mg/g in the EtOAc-soluble fraction) (Table 2).

Dieckol is a known antibacterial substance with activity against MRSA (Lee et al., 2008). Of the soluble fractions, di-eckol was present only in the EtOAc fraction, which exhibited the highest anti-MRSA activity. Our previous report showed that of the pholorotannins tested, dieckol showed the highest anti-MRSA activity (Lee et al., 2008). Considering these data, we suggest that the anti-MRSA activity of E. bicyclis is likely mediated by phlorotannins such as dieckol.