The IG used in this study was purchased from a local pharmaceutical company (Hyosung Pharmaceutical Co., Daegu, Korea) and screened for use through microscopic inspection. Table 1 presents the composition of a sachet of IG. The ingredients contained in a selected sachet (68 g) were dissolved in 2000 ml distilled water and heat-extracted, followed by suction filtration of the extracted liquid and the depressurization and concentration of the filtered extract using a rotary vacuum evaporator (N-N type; LAB Camp, Daejeon, Korea). The extract with viscous consistency thus acquired underwent freeze drying in a programmable freeze dryer (PVTFD10A; Ilshin Lab., Seoul, Korea), which yielded a total of 17.00 g (yield rate: 25.01%) dark brown fluid extract. BK-1202 was then produced by adding 4 g Ostrea gigas to the IG extract, which was them pulverized and stored in a freezer (−20ºC) to be used for the experiments. The prepared BK-1202 powder dissolved well in distilled water up to the concentration of 40 mg/ml. Drugs for the experimental and reference groups, i.e. indomethacin, omeprazole, and DA-9601 were purchased from Sigma (MO, USA) and Dong-A Pharmaceutical Co. (Yongin, Korea).

A total of 56 male Sprague-Dawley rats (6-week old upon receipt, SLC, Japan) were used for the experiments after a 7-day acclimation. Throughout the duration of acclimation and experiments, they were kept in polycarbonate boxes designed for rats, 5 rats per box, in constant temperature (20–25°C) and humidity (30–35%) conditions in a 12-h light:dark cycle, and were given ad libitum access to feed (Samyang, Korea) and water. Of 56 rats, 48 were used for developing indomethacin-induced gastric mucosal damage, and the remaining 8 were used as an intact control group. All animals were administered their respective drugs (except for the two control groups, which were administered sterilized distilled water) after 24-h feed deprivation (with free access to water), and gastric mucosal damage was induced by administering indomethacin 30 min after the initial treatment (except for the intact control group, which was administered sterilized distilled water). Individuals were identified by picric acid. All animals were treated in compliance with the “Guide for the Care and Use of Laboratory Animals”7).

Lab animals were divided in 7 groups, 8 individuals per group, as listed in Table 2: 1) Intact control group treated with sterilized distilled water after the initial administration of sterilized distilled water used as medium, 2) Indomethacin control group with indomethacin-induced gastric mucosal damage after the administration of the medium sterilized distilled water, 3–7) Experimental groups with indomethacin-induced gastric mucosal damage after being administered omeprazole, DA-9601, and three different doses (200, 100, and 50 mg/kg) BK-1202, respectively (Table 2). BK-1202, omeprazole, and DA-9601 were dissolved in sterilized distilled water and orally administered to the animals using a 3-ml syringe with metallic probe at the ratio of 5 ml per 1 kg animal weight. The omeprazole and DA-9601 doses were set at 10 mg/kg8) and 100 mg/kg9), respectively, doses known to be have sure efficacy in reducing indomethacin-induced gastric mucosal damage.

After 24-h feed deprivation of all animals, the two reference groups were administered with omeprazole and DA-9601, respectively, and the three experimental groups were administered with three different doses of BK-1202. After 30 min, a single dosage of 25 mg/kg indomethacin dissolved in sterilized distilled water was orally administered in accordance with the Guidobono method10) to induce gastric mucosal damage. The two control groups were administered sterilized distilled water in the initial treatment. In the second treatment, the intact and indomethacin control groups were administered sterilized distilled water and 25 mg/kg indomethacin, respectively.

In accordance with the method proposed by Süleyman et al.11), all animals were sacrificed with cervical dislocation 6 h after being treated with indomethacin. From each animal, the stomach was harvested and, with the greater curvature cut open, fixed in 10% neutral buffered formalin for 24 h. The total surface area of the gastric mucosal lesions with hemorrhagic foci was estimated by the unit of mm² by overlapping a grid (area: 1 mm²) on the area of gastric mucosal damage.

In accordance with the method proposed by Morai et al12), the gastric mucus layer was separated and homogenized in an ice-cold solution of 10 mL KCl (100 g/L; pH 7.4; Sigma, MO, USA), and the MPO content within the stomach tissue homogenate was measured using the method proposed by Bradley et al13) with a light modification as follows. The homogenate underwent 3 freeze-thaw cycles and centrifuged at 1500×g for 10 min at 4°C. Of its supernatant, 100 ml was harvested and added with 1 ml of 1.5 mM/L o-dianisidinehydrochloride (Sigma, MO, USA) containing 1.9 ml of 10 mM/l phosphate buffer (pH 6.0) and 0.0005% (w/v) hydrogen peroxide (Merk, CA, USA). Finally, PMO activity was estimated at the unit of μM/minute/mg tissue by measuring absorbance at 450 nm with a UV-vis spectrophotometer (UV-3600, Shimadzu Scientific Instruments, CO, USA).

Lipid peroxidation was measured by the method of Ohkawa et al14), by measuring the content of malondialdehyde (MDA) in the prepared stomach tissue homogenate using thiobarbituric acid (Sigma, MO, USA). First, 0.5 ml homogenate was mixed with 0.2 ml solution containing 80 g/l sodium lauryl sulfate (Sigma, MO, USA), 1.5 ml 200 g/l acetic acid (Merck, CA, USA), 1.5 ml 8 g/l 2-thiobarbiturate, and 0.3 ml sterilized distilled water. The mixed solution was heated to 98°C for 1 h and added with 5 ml n-butanol:pyridine (15:1) (Merck, CA, USA), which was mixed by shaking for 1 min. The solution was then centrifuged at 4000 rpm for 30 min, and the absorbance of the supernatant was measured at 532 nm. Finally, MDA was estimated at the unit of nM/g tissue by comparing the measured absorbance values with the standard curve predetermined using 1,1,3,3-tetramethoxypropane (Sigma, MO, USA).

Tissue GSH content was measured by the method of Sedlak and Lindsay15). The harvested stomach was homogenized in 2 ml of 50 mM Tris–HCl buffer (pH 7.5) containing 20 mM EDTA (Sigma, MO, USA) and 0.2 mM sucrose (Merck, CA, USA) and immersed in 0.1 ml 25% trichloroacetic acid (Merck, CA, USA). Precipitates were then removed by centrifuging at 4200 rpm for 40 min at 4°C. After taking its supernatant, absorbance was measured at 412 nm using 5,5′-dithiobis(2-nitrobenzoic acid) (Sigma, MO, USA). Finally, GSH content was measured at the unit of nM/mg tissue.

Catalase activity was measured by measuring absorbance of hydrogen peroxide (H₂O₂) decomposition at 240 nm in the presence of catalase16). Catalase activity was defined as the amount of enzyme necessary for decomposing 1 nM H₂O₂ in one min at pH 7.8 and 25°C, and catalase was estimated the unit of mM/min/mg tissue.

SOD activity was estimated at the unit of mM/min/mg tissue using the method proposed by Sun et al17). The amount of superoxide radicals produced was estimated by measuring absorbance at 560 nm with respect to the formazan dye formation from xanthine and xanthine oxidase reacted with nitrotetrazolium blue (Sigma, MO, USA).

Gastric fundus tissue was harvested, cut into cell layers, fixed in 10% neutral-buffered formalin over 18 h, dehydrated, and embedded in paraffin wax. The paraffin block was then sliced into 4 μm sections. These tissue sections were stained with hematoxylin & eosin (H&E). The tissue samples thus prepared were analyzed under an optical microscope. In the histomorphometry, tissue damage such as gastric mucosal damage, edema, and hyperemia was estimated by means of semiquantitative scoring on a 4-grade scale (0 = normal, 1 = slight: mucosal surface injury, 2 = moderate: moderate-to-severe mucosal injury and edema, 3= severe: total mucosal damage). Additionally, using a conventional method used by Ku et al18), the invasion rate of gastric mucosal damage [(thickness of the damaged gastric mucosa / thickness of the total gastric mucosa) ×100, %] and the average mucosa thickness around the lesion were measured using CCD image analyzer (DMI-300, DMI, Korea).

All numerical values were expressed as mean ± standard deviation. Statistical analysis was performed using multiple comparison testing, and the equality of variance was tested with Levene’s test. In case of equal variance, one way ANOVA test was used, followed by post-hoc testing with the least-significant differences (LSD) test, to estimate the minimum difference required for the significance of between groups. When variances were unequal, Kruskal-Wallis non-parametric H test was used, and if significance was established, Mann-Whitney U (MW) test was performed to test the significance between groups. All statistical analysis was performed using SPSS for Windows (Release 14.0K, SPSS Inc., USA), and significance was considered to be present at the 5% level (p≤0.05). Moreover, the degree of indomethacin-induced gastric mucosal damage was estimated by a percent change between the intact control group and indomethacin control group using Eq. [2]. For a more concrete demonstration of the anti-ulcerative drugs, a percent change was performed between treatment group and indomethacin control group using Eq. [3].

In all of the 7 indomethacin-induced gastric mucosal damage groups, hemorrhagic gastric ulcers were observed throughout the gastric mucus layer in broad distribution. While indomethacin control group showed significantly greater (p<0.01) increase in visually inspected lesions than the intact control group, the two reference groups (omeprazole- and DA-9601-treated groups) and three experimental groups (BK-1202-treated groups of three different doses) demonstrated significant decrease (p<0.01) of visually inspected lesions compared with the indomethacin control group (Figs. 1, 2).

The indomethacin control group exhibited changes in the area affected by gastric mucosal damage as high as 833.71% with respect to the intact control group. However, the groups treated with omeprazole, DA-9601, and 200, 100, and 50 mg/kg BK-1202 showed changes of −21.16, −35.88, −51.57, −36.40, and −22.09% with respect to the indomethacin control group.

The indomethacin control group showed significantly greater (p<0.01) increase in gastric MPO activity compared to the intact control group, with the amount of change of 302.09% with respect to the intact control group. In contrast, the two reference groups (omeprazole- and DA-9601-treated groups) and three experimental groups (200, 100, and 50 mg/kg BK-1202-treated groups) demonstrated significant decrease (p<0.01) in gastric MPO activity, −22.55, −31.06, −50.85, −33.02, and −33.91%, respectively, when compared with the indomethacin control group (Fig. 3).

The indomethacin control group showed significantly greater (p<0.01) increase in gastric MDA content, i.e. increase in lipid peroxidation, compared to the intact control group, with the magnitude of change amounting to 570.04% with respect to the intact control group, whereas the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups demonstrated significant decrease (p<0.01) in gastric MDA content with changes of −21.51, −37.43, −48.98, −38.66, and −23.11%, respectively, when compared with the indomethacin control group (Table 3).

The indomethacin control group showed significantly greater (p<0.01) decrease in gastric GSH content compared to the intact control group, exhibiting a change of −59.88% with respect to the intact control group. In contrast, the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups demonstrated significant increase (p<0.01 or p<0.05) in gastric GSH content, marking changes of 20.86, 33.98, 51.40, 34.70, and 22.15%, respectively, when compared with the indomethacin control group (Table 3).

The indomethacin control group showed significantly greater (p<0.01) increase in gastric catalase activity compared to the intact control group, exhibiting 122.76% increase with respect to the intact control group. In contrast, the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups demonstrated significant decrease (p<0.01) in gastric catalase activity, exhibiting changes of −20.65, −32.49, −45.88, −34.04, and −22.27%, respectively, with respect to the indomethacin control group (Table 3).

The indomethacin control group showed significantly greater (p<0.01) decrease in gastric SOD activity compared to the intact control group, changing −42.73% with respect to the intact control group. In contrast, the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups demonstrated significant increase (p<0.01 or p<0.05) in gastric SOD activity, marking changes of 23.19, 34.23, 47.63, 36.44, and 25.24%, respectively, when compared with the indomethacin control group (Table 3).

While the indomethacin control group exhibited typical manifestations of gastric ulcer, such as superficial desquamation, necrosis, and inflammatory cell infiltration, such symptoms were considerably suppressed in omeprazole-, DA-9601- and all BK-1202-treated groups (Fig 4). Moreover, while significant increases (p<0.01) were confirmed in the indomethacin control group in terms of the invasion rate of gastric mucosal damage, mean thickness of the gastric mucus layer around the lesion, and semiquantative score, significant decrease (p<0.01) in these pathological conditions were confirmed in the omeprazole-, DA-9601-, and all BK-1202-treated groups compared to the indomethacin control group (Table 4).

The ratio of gastric mucosal damage in the indomethacin control group to that in the intact control group was as high as 5452.01%, whereas the same for the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups were −17.52, −53.92, −76.82, −55.81, and −18.85% with respect to the indomethacin control group.

The change in the gastric mucosa thickness in the indomethacin control group was 61.45% with respect the intact control group, whereas the same for the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups were 33.33, 43.94, 67.68, 48.99, and 37.37% with respect to the indomethacin control group.

The change in the gastric mucosa thickness in the indomethacin control group was 61.45% with respect the intact control group, whereas the same for the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups were 33.33, 43.94, 67.68, 48.99, and 37.37% with respect to the indomethacin control group.

The change in the semiquantative score in the indomethacin control group with respect the intact control group was as high as 1000.00%. In contrast, the omeprazole- and DA-9601-treated reference groups and the 200, 100, and 50 mg/kg BK-1202-treated experimental groups showed semiquantative score changes of −27.27, −59.09, −68.18, −63.64, and −36.36% with respect to the indomethacin control group.