Over the past several years, streptococcal infections have dramatically increased due to the growth of aquaculture (Baeck et al., 2006; Park et al., 2009). Streptococcosis is a common disease in fish caused by six different Gram-positive streptococcal species (Bercovier et al., 1997). One species, Streptococcus parauberis (formerly known as S. uberis type II), is an alpha hemolytic Gram-positive bacterium that causes mastitis in cows (Leigh, 2002). To date, this species has emerged as a causative agent of streptococcosis in several fish species, including turbot and olive flounder Paralichthys olivaceus (Domenech et al., 1996). In addition, this bacterium was reported to be a major cause of streptococcosis in Korea and the causative agent in an outbreak affecting olive flounder on Jeju Island, Korea in 2005 (Baeck et al., 2006). Streptococcosis caused by S. parauberis is becoming an endemic disease that results in external hemorrhaging, a pale friable liver with hepatomegaly in olive flounder (Kim et al., 2006). In trout, streptococcosis results in eye discharge, spleen and kidney congestion, and mucohaemorrhagic enteritis (Domenech et al., 1996).

Broad-spectrum antibiotics are commonly used to control streptococcal infections in aquaculture. Park et al. (2009) reported the isolation of antibiotic-resistant S. parauberis strains from olive flounder. Their study suggested the need for new approaches to control and overcome an increase in antibioticresistant S. parauberis. One alternative approaches is to use compounds that interact with antibiotics to improve their antibacterial efficacy. However, synthetic compounds can be toxic and cause several negative side effects when utilized in combination with antibiotics. Alternatively, phytochemicals from natural resources, including plants, have been reported to alter the efficacy of antibiotics against several pathogenic bacteria (Adwan and Mhanna, 2008; Gupta et al., 2009; Eom et al., 2014; Lee et al., 2014).

Seaweeds contain various metabolites that are capable of inhibiting bacterial growth (Rhimo et al., 2010; Cox et al., 2010; Eom et al., 2013). Among these seaweed species, edible brown alga Ecklonia cava, an organism native to the cost of Korea and Japan, has been utilized as an herbal remedy. Extracts from this alga, including fucoidan, fucan sulphate and phlorotannins, exhibit antitumor, anticoagulant, antioxidants and anti-inflammatory activity (Kim et al., 2006; Kang et al., 2012). In addition, this seaweed exhibits antibacterial activity against several human pathogenic bacteria (Hornsey and Hide, 2007; Kandhasamy and Arunachalam, 2008; Eom et al., 2013). Although various studies have shown the effectiveness of seaweed extracts in combating human pathogenic bacteria, there is less information on the antibacterial efficacy of this extract for fish pathogens. Thus, the objective of this study was to evaluate: (i) the antibacterial activity of E. cava extract toward the fish pathogenic bacterium S. parauberis and (ii) the synergistic effect of the extract and antibiotics in combination as an antibacterial agent against S. parauberis.

In this study, we used S. parauberis KSP strains (KSP2, 3, 5, 7, 8, 9, 26, 44, 46, and 47) provided by the Department of Aquatic Life and Medicine, Pukyong National University (Busan, Korea). The Korean Collection for Type Cultures (KCTC; Daejeon, Korea) provided the reference strain of S. parauberis (KCTC 3651). All strains were cultured in Brain Heart Infusion broth (BHI) (Difco, Sparks, MD, USA) and incubated at 25°C for 18-24 h.

In total, we used thirteen seaweed species. Dried seaweed powder (1.0 kg) was extracted and fractionated using organic solvents as described by Lee et al. (2014). The methanolic (MeOH) extract of E. cava (101.3 g) was fractionated with n-hexane (Hexane; 1.0 L × 3), dichloromethane (1.0 L × 3), ethyl acetate (EtOAc) (1.0 L × 3), and n-butanol (BuOH) (1.0 L × 4). The water fraction was obtained by filtering the remaining materials from the BuOH fraction. Finally, each extract was evaporated using a rotary evaporator (Eyela, Tokyo, Japan) under a vacuum at 45°C. Another twelve methanolic extracts from various marine algae, previously prepared and stored under appropriate conditions (at −70°C), were used for the same purpose.

We evaluated the antibacterial activity of seaweed extracts and antibiotics against S. parauberis strains using a disk diffusion assay as described by the Clinical Laboratory Standard Institute (CLSI, 2012). In brief, S. parauberis strains in 3 mL of BHI were grown at 25°C for 24 h. An aliquot of 100 μL of each strain was spread on Mueller Hinton agar (Difco) supplemented with 5% sheep blood. On the inoculated plates, we placed disks containing 5 mg of seaweed extract and disks with standard antibiotics. After incubation for 18-24 h at 25°C, antibacterial activity was measured by the diameter (mm) of the zone of growth inhibition.

The sensitivity of the S. parauberis strains to the E. cava extract and commercial antibiotics was determined by the two-fold serial dilution method using Cation Adjusted MullerHinton Broth (CAMHB) (MBCell, Seoul, Korea) supplemented with 5% lysed horse blood (MBCell) (CLSI, 2014). The minimum inhibitory concentration (MIC) was determined as the lowest concentration of E. cava extracts and antibiotics that resulted in the complete inhibition of visible growth in CAMHB after incubation at 25°C with shaking for 12 h.

Interactions between the E. cava EtOAc fraction and antibiotics were tested by the checkerboard method using an FIC assay (Hseish et al., 1993; Meletiadis et al., 2010). The sum of the FICs (ΣFIC) was calculated for each well using the following equation: ΣFIC = FIC_A + FIC_B = (C_A/MIC_A) + (C_B/ MIC_B), where MIC_A and MIC_B are the MICs of drugs A and B, respectively, and C_A and C_B are the concentrations of the drugs in combination, respectively, in all wells corresponding to an MIC (isoeffective combinations). Among the ΣFICs calculated for all isoeffective combinations, we reported the minimum ΣFIC (ΣFIC_min) and the maximum ΣFIC (ΣFIC_max) to capture synergistic and antagonistic interactions, respectively. Synergistic effects were evaluated on an FIC index as described by Lee et al. (2014). Interactions were defined as synergistic if the FIC index was < 1.0, additive if the FIC index equaled 1.0, subadditive if the FIC index was between 1.0 and 2.0, indifferent if the FIC index equaled 2, and antagonistic if the FIC index was > 2.0.

To evaluate the antibacterial activity of seaweed extracts toward S. parauberis strains, MeOH extracts from thirteen marine algae were utilized in a disk diffusion assay. The E. cava MeOH extract exhibited broad-spectrum antibacterial activity against all tested S. parauberis strains, expect for S. parauberis KCTC 3651. The diameters of the zones of inhibition were within the range of 8.0-12.0 mm at 5 mg/disk (Table 1). The greatest sensitivity was observed for S. parauberis KSP 8 and KSP 47, while strains KSP 3 and KSP 7 were the least affected. Kanagasabhapathy et al. (2006) classified the zones of inhibition as follows: weak (< 2 mm), moderate (2-5 mm) and strong (> 5 mm) antibacterial activity. Thus, the E. cava extract had a strong inhibitory effect on the S. parauberis strains. Furthermore, these results indicate that the E. cava extract had strong antibacterial activity against S. parauberis compared to the other seaweed extracts. These results are consistent with our previous study of antibacterial activities of an E. cava MeOH extract against Listeria monocytogenes (Nshimiyumukiza et al., 2015), Staphylococcus aureus (Eom et al., 2013) and Enterococcus faecalis (Kim et al., 2015).

To further investigate the antibacterial effects of E. cava, we fractionated the MeOH extract using several solvents. Of those, the Hexane fraction exhibited the greatest antibacterial activity against all S. parauberis strains (zone range: 6.0-11.0 mm at 5 mg/disk) compared to the other fractions (Table 2). Although the Hexane fraction had an inhibitory effect on the growth of S. parauberis KCTC 3651, the MeOH extract did not have any significant inhibitory effect.

From these results, we hypothesized that an antibacterial substance against S. parauberis could be abundant in the Hexane fraction of the E. cava extract. Interestingly, these results differed from our previous studies in which the E. cava EtOAC fraction exhibited the strongest antibacterial activity against human pathogenic bacteria (Eom et al., 2013; Kim et al., 2015; Nshimiyumukiza et al., 2015). Also, Rangaiaha et al. (2010) reported that seaweed extracts in different solvents exhibited different antimicrobial activities. The antimicrobial activity of organic solvent extracts against microorganisms could be related to the presence of different bioactive metabolites (Kolajinathan et al., 2009; Manivannan et al., 2011).

Thus, we performed a quantitative analysis using an MIC assay to precisely evaluate the antibacterial activity of the E. cava MeOH extract and its soluble fractions against S. parauberis. As shown in Table 3, the solvent-soluble fractions had broad-spectrum antibacterial activity against S. parauberis strains, with MIC values ranging from 256 to > 1,024 μg/mL. Among the solvent-soluble fractions, the Hexane fraction resulted in the lowest MIC values, by completely inhibiting the growth of S. parauberis strains (except KSP26) in the range of 256 to 1,024 μg/mL (Table 3). Kanjana et al. (2011) reported the following MIC scale for plant extracts: < 100 μg/mL, strong inhibitors; 100-500 μg/mL, moderate inhibitors; 500-1,000 μg/mL, weak inhibitors; and > 1,000 μg/mL, not inhibitors. Thus, we observed moderate inhibition against S. parauberis strains KSP8 and KSP47 using the Hexane fraction. Meanwhile, weak inhibition was observed against S. parauberis strains KSP2, KSP9, and KSP44 using the Hexane fraction.

To determine the antibiotic resistance of S. parauberis, ten commercial antibiotics were tested against several strains As shown in Table 4, all S. parauberis strains tested in this study exhibited susceptibility to amoxicillin/clavulanic acid (AMC), ampicillin (AMP), amoxicillin (AMX), gentamycin (GEN), enrofloxacin (ENR), florfenicol (FFC), and trimethoprim/sulfamethoxazole (SXT), with zones of inhibition ranging in diameter from 16 to 40 mm. Alternatively, most of the S. parauberis strains (KSP2, 3, 7, 8, 9, 44, 46, and 47) were resistant to doxycycline (DOX), erythromycin (ERY), and oxytetracycline (OTC). However, S. parauberis strains KCTC 3651 (a nonpathogenic strain not isolated from fish), KSP5, and KSP26 did not exhibit resistance to these antibiotics. In addition, antibiotic susceptibility testing by disk diffusion assays revealed that the antibiotic-resistant S. parauberis strains (KSP2, 3, 7, 8, 9, 44, 46, and 47) exhibited the highest resistance to ERY and OTC compared to DOX. Thus, the S. parauberis strains were highly resistant to ERY (zone diameter range: 6.0-9.0 mm) and OTC (zone diameter range: 6.0-7.0 mm), which far exceeded the breaking point of resistance for these antibiotics (≤ 23 mm for ERY and ≤ 17 mm for OTC) (Gray and Shryock, 2005).

The results obtained in this study are consistent with those of previous reports indicating that ERY and OTC have limited efficacy toward S. parauberis isolated from olive flounder (Jeong et al., 2006; Park et al., 2009). Meng et al. (2009) reported that S. parauberis strains with high-level resistance to OTC are often associated with resistance to ERY. This is likely because most erm genes encoding ERY resistance proteins are carried on transposons, and through transposition become linked to tet genes encoding tetracycline or OTC resistance proteins.

ERY and OTC were used in an MIC assay with the antibiotic-resistant S. parauberis strains to verify the antibiotic concentrations needed to inhibit bacterial growth. As shown in Table 5, the MIC values for both antibiotics against the antibiotic-resistant strains ranged between 256 and 1,024 μg/mL. The MIC values for ERY and OTC against the S. parauberis strains were significantly higher than the MIC breakpoint values of 1 μg/mL for ERY and 16 μg/mL for OTC (Gray and Shryock, 2005). These results suggest that ERY and OTC are no longer effective for treating infections caused by antibioticresistant S. parauberis.

One effective strategy for overcoming antibiotic resistance is to restore the activity of the antibiotic. This has been demonstrated by the use of antibiotics in combination with antibacterial agents derived from natural resources (Eom et al., 2014; Lee et al., 2014; Nshimiyumukiza et al., 2015). Here, we evaluated the interaction between the Hexane fraction of an E. cava MeOH extract and commercial antibiotics (ERY and OTC) against antibiotic-resistant S. parauberis strains using an FIC assay.

As shown in Table 5, the MIC values for ERY and OTC in combination with the Hexane fraction were markedly reduced, with 64-fold inhibition against the antibiotic-resistant S. parauberis strains. Thus, the MIC for OTC against S. parauberis KSP9 was reduced from 1,026 to 16 μg/mL in combination with 126 μg/mL of the Hexane fraction. These results suggest that the antibacterial activity of the conventional antibiotics ERY and OCT was restored when used in combination with the Hexane fraction. The synergistic antibacterial activity between the antibiotics and the Hexane fraction of E. cava was assessed by FIC analysis.

FIC indices of ERY and OCT in combination with the Hexane fraction are presented in Table 5. For all strains, the Hexane fraction and ERY combination resulted in a ΣFIC_min range of 0.118 to 0.500 and a ΣFIC_max range of 1.016 to 1.125. The Hexane fraction and OTC combination resulted in a ΣFIC_min range of 0.094 to 0.313 and a ΣFIC_max range of 1.004 to 1.016 for all strains. These results suggest that the Hexane fraction had greater synergistic effects in combination with OTC than with ERY. In addition, the median ΣFIC of the OTC and Hexane fraction combination ranged from 0.502 to 0.516. As reported by Lee et al. (2014), the synergistic ranges of median ΣFIC < 1 were observed for all combinations of OTC and the Hexane fraction against the antibiotic-resistant S. parauberis strains.

Despite the multitude of studies on the synergistic effects of natural compounds in combination with antibiotics to treat human pathogenic bacteria, few studies have focused on using this therapy to treat fish pathogens. Although the use of antibiotics is an effective way to control fish pathogenic bacteria in aquaculture farms, it may result in increased antibiotic resistance (Eom et al., 2013). Therefore, restoring the antibacterial activity of conventional antibiotics by combining them with natural compounds is an attractive approach. Natural products can increase the antibacterial efficacy of antibiotics without increasing antibiotic resistance among the bacterial pathogens in aquatic animals. In this study, we demonstrated that an E. cava extract could inhibit S. parauberis and that the synergistic effect created by the E. cava extract and conventional antibiotics significantly increased the level of antibacterial activity against S. parauberis. Thus, the results obtained in this study will contribute to the development of an alternative phytotherapeutic agent for use against infections with the fish pathogenic bacterium S. parauberis.