Seaweeds are rich in minerals and dietary fiber and are thus typically used as a health food. One such seaweed, Eisenia bicyclis (Kijellman), is used in various dishes including appetizers, casseroles, muffins, pilafs, and soups (Maegawa, 1990). Eisenia bicyclis was previously reported to be a potent source of bioactive phenolic compounds known as phlorotannins. Phlorotannins with antioxidant activity isolated from E. bicyclis include eckol (a trimer), phlorofucofuroeckol A (a pentamer), dieckol (a hexamer), and 8,8′-bieckol (a hexamer) (Okada et al., 2004). The biological activities of this brown algae have also shown activities against tumors (Noda et al., 1989), Alzheimer’s disease (Jung et al., 2010), inflammatory diseases (Shibata et al., 2003), and allergic diseases (Shibata et al., 2002). Despite its key biological activities, E. bicyclis has largely been used raw, frozen, or salted in processed products and in products that confer none of the advantages of highly processed seaweeds (Heo and Jeon, 2005). The development of processed foods containing seaweeds is necessary to attract consumers and enhance seaweed consumption (Bae and Kim, 2010; Eom et al., 2010; Song et al., 2011b).

Fermentation by living beneficial microorganisms may be useful in achieving this goal. Generally speaking, fermentation is a biochemical reaction that metabolizes high-molecular weight organic compounds into simpler molecules. Fermentation not only enhances the nutrient content of foods through the biosynthesis of vitamins, essential amino acids, and proteins, it also improves protein quality and fiber digestibility. Furthermore, fermentation improves micronutrient bioavailability and aids in the degradation of anti-nutritional factors (Achinewhu et al., 1998; Adewusi et al., 1999; Bae and Kim, 2010). Here, we report a microbial strain that effectively improved the functional qualities of E. bicyclis water extract and describe the physiochemical and nutritional changes that occurred in the extract as a result of microbial fermentation.

E. bicyclis was collected on Ulleung Island, Korea, in September 2010. As described previously, the sample was rinsed with fresh water to eliminate foreign materials such as sand and shells (Song et al., 2011b). Next, the sample was desalinated by soaking two to three times in two volumes of water for 1 h each. The desalinized sample was cut into an efficient size for extraction using a grinding mill. To prepare a water extract of E. bicyclis, 20 volumes of water were added, and the mixture was autoclaved and then centrifuged to remove all E. bicyclis residue. After cooling at room temperature, the extract was stored at -70°C until use.

The microbial strains used in this study were isolated from traditional Korean fermented foods and maintained at the Food Microbiology Laboratory, Department of Food Science and Technology, Pukyong National University. Selection of a strain for fermentation was done by assaying the total phenolic compound (TP) content and antioxidant activity during fermentation of the E. bicyclis extract. The following strains were tested: three Lactobacillus strains (LB-1, -2, and -3) isolated from kimchi by enrichment culture using MRS broth (Difco, Detroit, MI, USA), two Bacillus strains (BC-1 and -2) isolated from meju using Mannitol Egg Yolk Polymyxine Agar (Oxoid Ltd., Hampshire, UK), and three yeast strains (YM-1, -2, and -3) isolated from meju using YM broth (Difco). Fermentation was conducted in a flask containing 500 mL of the extract that had been inoculated with 10% (v/v) fresh culture. The samples were incubated aerobically with agitation (120 rpm) at 30°C. Sampling was done at regular intervals (Bae and Kim, 2010).

The TP content was determined according to a modified version of the Folin-Ciocalteu method using phloroglucinol as the standard (Kim et al., 2006a; Bae and Kim, 2010). The samples were diluted to match the measurable range of the spectrophotometer. A 0.1 mL aliquot of the diluted sample was mixed in an Eppendorf tube with 0.5 mL of 1 N Folin-Ciocalteu reagent. The mixture was allowed to stand for 3 min following the addition of 0.4 mL of 20% Na₂CO₃. The samples were incubated in the dark at room temperature for 45 min and then centrifuged at 1,600 g for 8 min. The optical density of the supernatant was measured spectrophotometrically with a GENios^® microplate reader (Tecan Austria GmbH, Grodig, Austria) at 765 nm. The TP content (%) was calculated using a standard graph. The TP content in the non-fermented or fermented extract was calculated using the following linear equation based on the calibration curve:

Y=0.5337X+0.3615, r2 = 0.9874

where, Y is the absorbance and X is the total phenolic contents in milligram phloroglucinol equivalents per gram extract.

According to the manufacturer’s instructions, portions of each yeast strain were picked up with a sterile loop from the YM plates and then suspended in API basal medium ampoules (Gunde？ et al., 2001). The suspensions were used to fill the cupules of the test strip (API 20C AUX system; bioMerieux, Marcy l’Etoile, France). The API trays were inoculated and incubated for 72 h at 30°C. The growth in each well was recorded after 24, 48, and 72 h of incubation. A profile number based on the reactions was assigned and identifications were made by reference to the API Analytical Profile Index (bioMerieux). Cupules showing turbidity that was significantly greater than that in the negative control cupules were considered positive (Fenn et al., 1994).

Filtered extracts were analyzed for pH, salinity, and proximate composition. The moisture content was measured by oven-drying at 105°C to a constant weight. The contents of crude ash, crude protein, and crude lipid were determined by standard methods (Association of Official Analytical Chemicals, 1995). Reducing sugars were measured by the Somogy method with a minor modification (Somogyi, 1952). All experiments were carried out in triplicate. The data are shown as mean values.

The pH was measured with a tip probe electrode. Salinity was measured by converting the chlorine content into sodium chloride using a YSI 63 pH meter (YSI Inc., Yellow Springs, OH, USA).

The determination of FAAs was done by ion-exchange chromatography using a Hitachi model L-8900 amino acid analyzer (Hitachi Co. Ltd., Tokyo, Japan) with methods adapted from the literature. A total of 10 μL of supernatant filtered through a 0.45-μm syringe filter (Advantec, Tokyo, Japan) was applied to an ion exchange column (#2622SC; Hitachi Co. Ltd.) using lithium citrate buffer.

DPPH radical scavenging activity was measured as described by Qureshi et al. (2010). A total of 1 mL of non-fermented or fermented extract (diluted 100-fold) was added to 1 mL of 0.2 mM DPPH in ethanol. After mixing vigorously for 30 s, the solution was incubated at 37°C for 30 min. The DPPH radical scavenging activity of the solution was measured using a GENios^® microplate reader (Tecan Austria GmbH) at 517 nm. The DPPH radical scavenging activity was calculated using the following equation, in which H and H₀ are the relative peak height of radical signals with and without sample, respectively:

Seaweeds possess metabolites with useful biological activi-

[Table1.] Change of total phenolic contents in Eisenia bicyclis water extract by microbial fermentation

Change of total phenolic contents in Eisenia bicyclis water extract by microbial fermentation

ties, including antifungal, antioxidant, antiviral, and antitumor activities (Noda et al., 1989; Nagayama et al., 2002; Kim et al., 2006b). It was previously reported that a seaweed extract containing commercial enzymes showed enhanced antioxidant (Heo et al., 2005) and anticoagulant activities (Athukorala et al., 2006). However, such enzymes are expensive. Therefore, our laboratory has focused on the microbial fermentation of seaweeds as an alternative method (Bae and Kim, 2010; Eom et al., 2010; Song et al., 2011b). The biological activities of plant materials have been reported to relate to their content of phenolic compounds (Lee et al., 2009), and these activities are based on the physiological functionalities of polymers of polyphenol (McDougall et al., 2005). In this study, we used an increase in TP content to judge the effectiveness of microbial fermentation. The change in antioxidant activity during fermentation of the extract was monitored as an indicator of the biological activity of the sample.

As shown in Table 1, the TP content of the E. bicyclis extract was enhanced by fermentation using all of the microbes tested compared to the control (unfermented extract). The highest TP content (47.5 mg phloroglucinol equivalents [PGE]/g) was observed after one day of fermentation with YM-1, followed by the Lactobacillus sp. BS-1 extract (44.7 mg PGE/g) on 5 days of microbial fermentation. These results are consistent with previous data showing that Lactobacillus fermentation of a Hizikia fusiforme extract and fungal fermentation of a Laminaria japonica extract enhanced the TP content (Bae and Kim, 2010; Song et al., 2011a, 2011b). We suspect that the rise in TP content resulted from the fermentation-enhanced breakdown of plant cell walls by various enzymes.

Polyphenolic compounds are known to possess antioxidant activity (Jimenez-Escrig et al., 2001). We measured the antioxidant activity of the extract before and after microbial fermentation. As shown in Fig. 1, the DPPH radical scavenging activity rose as the TP content of the E. bicyclis extract increased due to microbial fermentation. Peak scavenging activity (about 72%) was observed in the extract after one day of fermentation by yeast strain YM-1. However, the DPPH radical-scavenging activity gradually decreased as fermentation progressed. In contrast, a reduction in antioxidant activity was

[Fig. 1.] Effect of microbial fermentation on DPPH radical scavenging activity of Eisenia bicyclis water extract. Experiments were repeated three times. LB-1 LB-2 and LB-3 stand for Lactobacillus sp. LB-1 LB-2 and LB-3 respectively. BS-1 and BS-2 are Bacillus sp. BS-1 and BS-2 respectively. YM-1 2 and 3 are yeast strains isolated in this study.

[Table 2.] Physiological characteristics of yeast YM-1 strain

Physiological characteristics of yeast YM-1 strain

observed in the other samples, reflecting the different physiological functions of the enzymes from the other microorganisms (De Mot and Verachtert, 1985; Bar-Shimon et al., 2004). Considering the enhanced TP content and DPPH radical scavenging activity obtained through microbial fermentation, we selected yeast strain YM-1 for our subsequent experiments.

The selected strain (YM-1) was identified based on an analysis of its physiological characteristics and the API Analytical Profile Index as described in the Materials and Methods. The physiological characteristics of the strain are given in Table 2. YM-1 exhibited 94.9% identity with Candida utilis, and thus we identified the strain as C. utilis. C. utilis is a high protein-yielding species used in food and feed supplements for livestock (Sakata and Kurokawa, 1992). It is also an industrially important microorganism approved by the US Food and Drug Administration as a safe substance (Misawa and Shiamada, 1997).

Microbial fermentation causes changes in food constituents (Jeng et al., 2007). Accordingly, it was expected that a change in the properties of E. bicyclis extract would occur during the fermentation of YM-1. As shown in Table 3, no significant difference was observed in the proximate composition (e.g., moisture, crude lipid, crude protein, and crude ash). However, the reducing sugar content decreased gradually as fermentation progressed, suggesting that carbohydrates are used as a substrate for growth of the yeast (Bae and Kim, 2010). In practice, the yeast cells grew gradually until 3 days of fermentation and showed reduced growth thereafter (data not shown). These results are in accordance with our previous data (Eom et al., 2010; Song et al., 2011b). We also found that there was no significant change in pH or salinity during fermentation (Table 3).

We additionally investigated the effect of fermentation by YM-1 on the FAA content of E. bicyclis extract. In total, 29 FAAs were detected in the control (unfermented) extract. The E. bicyclis extract exhibited large amounts of cysteine, phosphorylated serine, ornithine, and sarcosine compared to the FAA contents of the other seaweed extracts (Table 4). Song et al. (2011b) reported that glutamic acid, alanine, valine, and leucine were the predominant FAAs in an H. fusiforme water extract. Furthermore, it was reported that glutamic acid, aspartic acid, and alanine were the predominant FAAs in an L. japonica water extract (Eom et al., 2010).

The FAA content was also affected by C. utilis (YM-1) fermentation. A comparison of the FAA content revealed that the total FAA content increased slightly after one day of fermentation and then decreased gradually as fermentation

[Table 3.] Change of proximate components pH and salinity by Candida utilis YM-1 fermentation in Eisenia bicyclis water extract

Change of proximate components pH and salinity by Candida utilis YM-1 fermentation in Eisenia bicyclis water extract

[Table 4.] Effect of Candida utilis YM-1 fermentation on free amino acid content of Eisenia bicyclis water extract

Effect of Candida utilis YM-1 fermentation on free amino acid content of Eisenia bicyclis water extract

progressed (Table 4). Similar to the reducing sugar content, the FAAs appeared to be used as a substrate for yeast growth during fermentation. For example, the phosphorylated serine content, which has proposed anti-obesity effects (Cremades et al., 2004), increased until 5 days of fermentation.

Considering the results obtained in the current study, we conclude that C. utilis YM-1 is a good candidate strain for the fermentation of E. bicyclis extracts. Additionally, the TP content, DPPH radical scavenging activity, and FAA content measured in this study strongly indicate that the optimum level of fermentation by strain YM-1 was achieved after one day of fermentation. In this study, we examined the possibility of using fermentation to develop value-added food ingredients from E. bicyclis. Microbial fermentation by C. utilis YM-1 resulted in enhanced biological activity, including an increased TP content and antioxidant activity, suggesting that C. utilis YM-1 fermentation is an attractive strategy for developing value-added food ingredients.