Fetal bovine serum (FBS), 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT), and dimethylsulfoxide (DMSO) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). RPMI1640 medium, penicillin and streptomycin (PEST), and trypsin-ethylenediaminetetraacetic acid (T-EDTA) were obtained from GibcoBRL (Gaithersburg, MD, USA). All solvents used were analytical grade.

The subfractions from I. obliquus extracts were collected by the method of Ham et al. [11]. The I. obliquus specimen used in this study was imported from Russia, dried, powdered, and stored at -20℃. The powdered specimen (500 g) was extracted twice with 3 L of methanol (99.8%) at 45-50℃ for 3 h. The methanol fraction (45 g) was then rotary evaporated, emulsified, dissolved in water, and extracted three times with ethyl acetate (H₂O:EtOAc, 5:7, v/v). The ethyl-acetate fraction (12 g) was obtained after rotary evaporation, and the dried extract was reconstituted in methanol. This stock solution was subjected to vacuum chromatography (N-2N; Eyela, Tokyo, Japan) with silica gel (60 g, 230-400 mesh) mixed with dichloromethane. The mobile phases were dichloromethane (A) and methanol (B), with a gradient of increasing methanol (0-100%) in dichloromethane and re-equilibration of the column with 100% A for 3 min prior to the next run (Fig. 1). The success of the fractionation was confirmed by thin-layer chromatography (TLC). Based on the Ames test and the DPPH method, amongst the three ethyl-acetate-extract fractions, VCHG-fraction 1 was the most bioactive fraction in the antimutagenic and antioxidant effects [5,6]. Thus, the VCHG-fraction 1 was selected and subjected to column chromatography using a reversed-phase column (ODS-C₁₈) containing LiChroprep RP-18 (25-40 µm) mixed with H₂O: MeOH (1:30, v/v) in a gradient program ranging from 30% to 100% methanol (mobile phase A: methanol; mobile phase B: ultra-pure water). Four fractions were isolated, of which fraction 2 exhibited the most bioactivity. Fraction 4 also had strong bioactivity. Fraction 2 and fraction 4 were further fractionated by chromatography with a normal-phase column (2.4 cm×15 cm) containing silica gel (200 g, 70-230 mesh; Merck Co., Darmstadt, Germany) mixed with dichloromethane and eluted with a mobile-phase gradient ranging from 100% dichloromethane to 50% methanol. The flow rate was 1.0 ml/min (Fig. 1). This resulted into two subfractions from fraction 2, designated as subfractions 1 and 2, both of which showed bioactivity. In our previous study, subfractions 1 and 2 were identified as 3β-hydroxy-lanosta-8,24-dien-21-al and inotodiol, respectively [11]. Subfraction 3 was obtained from fraction 4 (Fig. 1). The compound in subfraction 3 was crystallized and characterized by determining its mass spectrometry (MS) (JEOL), ¹H nuclear magnetic resonance (¹H NMR), and ¹³C nuclear magnetic resonance (¹³C NMR) (Bruker DRX 500 MHz) spectra.

The human cancer cells used in this experiment were lung carcinoma A-549 cells, breast adenocarcinoma MCF-7 cells, stomach adenocarcinoma AGS cells, Sarcoma-180 cells, and cervical adenocarcinoma HeLa cells. These, and the normal transformed primary human embryonic kidney (HEK) 293 cell line, were obtained from the Korean Cell Line Bank (Seoul, Korea). All cells were cultured in RPMI 1640 medium supplemented with 10% FBS and 1% PN/ST. The cells were incubated at 37℃ in a humid atmosphere containing 5% CO₂.

The cytotoxic potential of the subfractions was evaluated by the MTT assay [12] against four human tumor cell lines: A549, MCF-7, AGS, and HeLa. The inhibitory activity against the proliferation of cancer cell lines was compared with that against the normal transformed primary human 293 cell line. The cells were seeded in 96 wells at a density of 5×10⁴ cells/ml and cultured for 24 h in RPMI 1640 medium with 10% FBS and 1% PEST at 37℃, 5% CO₂. The three subfractions separated from I. obliquus were dissolved in DMSO, and the stock solutions were diluted with RPMI 1640 medium (DMSO concentration < 0.01%). The DMSO was added to all test solutions at a final concentration of 0.01%. The control was the well containing 0.01% DMSO only. Various concentrations of the subfractions (0, 62.5, 125.0, 187.5, and 250.0 µg/ml) were added to each culture well and incubated for 48 h. The supernatants were then removed by aspiration, and MTT (5 mg/ml) was added to each culture well (one-tenth of the original culture volumes). After incubation for 4 h, the medium was removed, and the cells were allowed to dry for 30 min. The precipitates were then solubilized in DMSO with shaking for 30 min. The extent of the reduction of MTT was quantified by measuring the absorbance at 570 nm.

Anticancer activity in vivo against sarcoma growth in a mouse model was examined using the method of Ham et al. [13] and Maeda et al. [14] with minor modification. Briefly, Balbc/c mice (six-to seven-weeks-old) were obtained from Samtaco (Gyeonggi-do, Korea), randomly divided into the control and six experimental groups, and provided with water and normal chow. Animals were housed in an air-conditioned room at 21-26℃ with 45-55% moisture and a 12-hour dark/light cycle. The mice were treated in accordance with Kangwon National University guidelines for the care and use of laboratory animals. The Sarcoma-180 cell line was cultured for 7-10 days in the abdominal cavity of a Balbc/c mouse, and the cultured cells were harvested with the peritoneal fluid and centrifuged at 1,200 rpm for 10 min in phosphate-buffered saline (PBS). The separated sarcoma cells were floated in PBS and centrifuged, and their concentration was adjusted to 1.0×10⁶ cells/ml. The cells (0.2 ml) were implanted subcutaneously in the left groins of mice. After 24 h, the mice were fed normal chow supplemented with water (control group), 0.1 mg subfraction 1 per mouse per day (S-180+subfraction 1A), 0.2 mg subfraction 1 per mouse per day (S-180+subfraction 1B), 0.1 mg subfraction 2 per mouse per day (S-180+subfraction 2A), 0.2 mg subfraction 2 per mouse per day (S-180+subfraction 2B), 0.1 mg subfraction 3 per mouse per day (S-180+subfraction 3A) or 0.2 mg subfraction 3 per mouse per day (S-180+subfraction 3B) for 20 days. The mice were fed normal chow for 7 days after the last oral administration of subfractions and then euthanized with an ip injection of Avertin (Sigma-Aldrich Co., St. Louis, MO, USA) [15] on the 27^th day after cancer-cell implantation. Solid tumors were removed and weighed, and the tumor-growth-inhibition ratio (I.R.) was calculated using the following formula:

I.R. (%) = (Cw-Tw)/Cw×100

Cw: Average tumor weight of the control.

Tw: Average tumor weight of the experimental group.

The data were expressed as means ± SD. Differences between the groups were determined by one-way Analysis of Variance using the SPSS package (Version 10.0; SPSS, Chicago, IL, USA). The level of statistical significance was set at P<0.05. The comparisons of cancer cell growth inhibition in vitro and solid tumor growth inhibition in vivo among the groups were performed using Duncan's multiple range test [15].

As shown in Fig. 1, the ethyl-acetate fraction from I. obliquus was separated into three fractions by vacuum chromatography. VCHG-fraction 1 was further separated into four fractions by reversed-phase (ODS-C₁₈) column chromatography, of which the second fraction showed the greatest in vitro anticancer activity. Fraction 4 also had inhibitory activity against the proliferation of human cancer cells (data not shown). Fractions 2 and 4 from VCHG-fraction 1 were, in turn, separated by normal-phase silica-gel column chromatography into subfraction 1, subfraction 2 and subfraction 3, respectively. Each of the three subfractions yielded a single spot by TLC, confirming single-compound purity (data not shown). Purity was also confirmed by MS, ¹H NMR, and ¹³C NMR analyses. Each of compounds 1, 2, and 3 (subfractions 1, 2, and 3) had a purity of above 99.5%. The pure compounds were assayed against human lung carcinoma A549, breast adenocarcinoma MCF-7, stomach adenocarcinoma AGS, and cervical adenocarcinoma HeLa cells and normal HEK 293 cell line.

Exposure of normal HEK 293 cell culture to 0-250 µg/ml of subfractions for 48 h resulted in low cytotoxicity (Table 1), but high cytotoxicity was observed at a concentration of 300 µg/ml subfractions (data not shown). Thus, to avoid high cytotoxicity in normal cells from subfractions, 0-250 µg/ml subfractions were chosen for in vitro study.

As shown in Table 1, subfraction 1 was more active than subfraction 2 or subfraction 3 against the selected tumor cell lines A549, AGS, MCF-7, and HeLa. Subfractions 1, 2, and 3 (250 µg/ml) inhibited the growth of A549 cells by 66.7%, 51.7%, and 48.4%, respectively, and inhibited that of AGS cells by 72.2%, 51.0%, and 56.9%, respectively (Table 1). Subfractions 1, 2, and 3 (250 µg/ml) inhibited the growth of MCF-7 cells by 67.4%, 55.1%, and 37.6%, respectively, and inhibited that of HeLa cells by 69.5%, 58.3%, and 71.2%, respectively (Table 1). In contrast, the subfractions inhibited the growth of normal HEK 293 cells by less than 20% (Table 1). The results indicate that these subfractions inhibit cancer cell growth and have low cytotoxicity against normal cells.

In our previous study, after oral administration with 0, 0.2, 0.4, 0.8 or 1.6 mg ethyl acetate extract from I. obliquus per mouse per day, no genotoxic activity was observed, and the extract inhibited the genotoxic activity by N-methyl-N'-nitro-N-nitrosoguanidine in a dose-dependent manner [6]. Thus, to avoid toxicity from the subfractions, oral administration of 0.1 mg per mouse per day and that of 0.2 mg per mouse per day were chosen for in vivo experiment.

The average tumor weight of the Sarcoma-180-cell-bearing mice after water feeding (control) was 4.02 ± 0.19 g, whereas that of the mice after subfraction 1 administration at low level (0.1 mg/mouse per day, S-180+subfraction 1A) was 3.06 ± 0.22 g and that of the mice after subfraction 1 feeding at high level (0.2 mg/mouse per day, S-180+subfraction 1B) was 2.67 ± 0.34 g (Table 2). Thus, subfraction 1 at high level inhibited tumor weight by 33.71% of the control tumor weight, and treatment following tumor transplant resulted in a dose-dependent reduction of tumor weight in comparison with the untreated control (Fig. 2). Subfraction 2 at high level (0.2 mg/mouse per day) and subfraction 3 at both low and high levels (0.1 mg/mouse per day and 0.2 mg/mouse per day, respectively) also inhibited solid tumor growth significantly (Table 2). Subfraction 1 showed much greater inhibition of tumor growth than subfractions 2 and 3 in the mice bearing Sarcoma-180 cells, in agreement with our in vitro results.

As shown in Fig. 1, the subfractions containing pure compounds were obtained after reversed-phase (ODS-C₁₈) column chromatography of the ethyl-acetate extract of I. obliquus, followed by normal-phase silica-gel column chromatography. In our previous study, subfractions 1 and 2 were identified by MS, ¹H NMR, and ¹³C NMR analyses as 3β-hydroxy-lanosta-8,24-dien-21-al and inotodiol, respectively [11].

Inotodiol and lanosterol have similar chemical structures. Inotodiol has an OH group at position 22 (C-22) of lanosterol [9]. The chemical structure of the compound in subfraction 3 was determined by comparing the MS, ¹H-NMR, and ¹³C-NMR spectra (data not shown) to published library spectra and spectra in the literature [7,9,15,16]. These analyses revealed that subfraction 3 is lanosterol. Thus, 3β-hydroxy-lanosta-8,24-dien-21-al, inotodiol, and lanosterol are bioactive compounds of I. obliquus with anticancer properties.