The shade-dried aerial section of the Artemisia indica plant (3.0 kg) collected from a local region of Gansu, China (Specimen number: GNS-762) was subjected to chloroform extraction three times. The solvent was evaporated in vacuo to obtain a crude extract of 300 g. The obtained extract was subjected to column chromatography over a silica-gel to obtain compounds a (700 mg), b (2.1 g) and c (100 mg) using hexane-EtOAc (Guoyao Chemical Co., Ltd., Shanghai, China) as an eluent with increasing polarity of 20, 25 and 35%, respectively. Similarly, the root section (1.5 kg) was shade-dried and macerated with CHCl₃ (Guoyao Chemical Co., Ltd.) to yield 100 g crude extract, which upon fractionation, produced fraction I (7.4 g), II (13.2 g) and III (17.3 g) with 10, 20 and 50% EtOAc, respectively. Repeated column chromatography of fraction I yielded three more compounds: D (50 mg), e (65 mg) and f (2.5 g), fraction II and III yielded c (200 mg; also isolated from the shoot) and g (10.0 mg), respectively. A number of the compounds were purified by re-crystallization. The isolated compounds were characterized by spectral techniques, such as ¹nuclear magnetic resonance (NMR), ¹³Carbon (C)-NMR, distortionless enhancement by polarization transfer (DEPT)-NMR (all using a Bruker AV III NM Spectrometer; Bruker Scientific Technology Co., Ltd., Beijing, China) and liquid chromatography–mass spectrometry (LC-MS) using an Agilent 1200 system (Agilent Technologies, Waldbronn, Germany) coupled with a Bruker micro QTOF mass spectrometer (Bruker Daltonics, Ettlingen, Germany).

Growth medium RPMI-1640, minimum essential medium and fetal calf serum (FCS) were obtained from Gibco-BRL (Carlsbad, CA, USA) and trypsin, penicillin, MTT, streptomycin, dimethylsulfoxide (DMSO) and phosphate-buffered saline (PBS) were obtained from Tianjin Hanyang Biologicals Technology Co. Ltd. (Tianjin, China).

MCF-7 human breast cancer, BHY human oral squamous carcinoma, Miapaca-2 human pancreatic cancer, Colo-205 human colon cancer, A-549 human lung cancer cell lines and NIH-3T3 mouse embryonic fibroblasts were procured from the Institute of Cancer Research (Gansu, China). All cells were grown in a humidified incubator with 5% CO₂ atmosphere at 37°C and cultured in RPMI-1640 medium supplemented with 10% heat-inactivated FCS, 100 IU/ml penicillin and 100 μg/ml streptomycin.

Serial dilutions of the compounds (10–100 μM) were prepared by dissolving them in DMSO. The cell cytotoxicity was determined in MCF-7, BHY, Miapaca-2, Colo-205, A-549 and BHY cells using MTT assay, and the the half maximal inhibitory concentration values (IC₅₀) were calculated. The cells were plated in 96-well plates and treated with different concentrations of the seven compounds. Subsequent to incubation for 24, 48 and 72 h at 37°C in a humidified incubator, MTT (5 mg/ml) was added to each well followed by incubation for a further 4 h. The medium was carefully removed, then 0.2 ml DMSO was added to each well and the plates were agitated to allow for mixing. The absorbance was measured at 545 nm. The inhibitory effect of the compounds on cell growth was assessed as the percentage cell cytotoxicity, where vehicle-treated cells were considered to be 100% viable. The final concentration of DMSO was 0.5% in all treatment protocols.

ΛΨm was measured by Rhodamine-123 (Rh-123) dye (Tianjin Hanyang Biologicals Technology Co., Ltd.). Briefly, 5×10⁵/ml MCF-7 cells were treated with different concentrations (25 and 50 μM) of compound b and d and ΛΨm was measured by flow cytometry. Rh-123 (1 mM) was added 1 h prior to the termination of experiment. The cells were collected, washed in PBS and incubated with PI (5 μg/ml) for 20 min. The reduction in fluorescence intensity, as a result of ΛΨm loss was analyzed in the FL-1 channel.

Fluorescence microscopy was performed to evaluate the morphological alterations on the cells following drug treatment. Cells (1×10⁶ cells/ml) were seeded in 6-well plates and treated with the tested compounds at 25 and 50 μM. After 24 h, cells were spun at 1000 × g for 5 min. The pellet was resuspended in PBS. Cells were stained with DAPI (Guoyao Chemical Co., Ltd.). Cells were then observed, and images were captured, under a phase contrast microscope (Nikon TMS, Nikon Corporation, Tokyo Japan) for morphological analysis. Fluorescence-based dyes were employed for staining the cellular components.

IC₅₀ values were calculated by non-linear regression analysis with Graph Pad Prism, version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). Values are expressed as the means ± standard error of at least three independent experiments. A P<0.05 was considered to indicate a statistically significant difference.

The seven compounds (a–g) were obtained from shoot and root extract of A. indica. Their structures are presented in Fig. 1, and the molecules were identified as (a) 5-hydroxy-3,7,4′-trimethoxyflavone, (b) ludartin, (c) maackiain, (d) lupeol, (e) cis-matricaria ester, (f) trans-matricaria ester and (g) 6-methoxy-7,8-methylenedioxy coumarin, by comparison of spectroscopic [1D-NMR, 2D-NMR, infrared spectroscopy (IR), and high resolution (HR)-MS] and analytical data comparison with previous literature (18–28). The majority of the compounds were isolated from A. indica for the first time in the current study, to the best of our knowledge, with the exception of maackiain, which has been reported previously (16). A number of the isolated compounds have previously only been isolated from other species of the genus Artemisia (29–32).

The isolated compounds (a–g) were evaluated for antiproliferative activity by MTT assay against the human cancer cell lines, MCF-7, BHY, Miapaca-2, Colo-205 and A-549, and a normal mouse embryonic fibroblast cell line NIH-3T3 at 24, 48 and 72 h following treatment with the compounds. All of the compounds exhibited cytotoxic activity in a dose-dependent manner at different time intervals with a maximum effect at 72 h. The antiproliferative effect obtained as a result of 72-h treatment and the IC₅₀ data are summarized in Table I. Among the seven compounds, compounds b and d exhibited the greatest antiproliferative effect against all the cell lines. The IC₅₀ values were 25.18 and 28.09 (MCF-7); 28.05 and 32.31 (BHY); 31.21 and 36.45 (Miapaca-2); and 34.33 and 39.01 μM (Colo-205) for compounds b and d, respectively. The MCF-7 cell line was the most susceptible to these two compounds. Furthermore, the cytotoxic effects of the compounds (a–g) were evaluated against NIH-3T3 mouse embryonic fibroblasts. This was performed in order to provide further insight into whether these compounds are specific or non-specific to cancer cells. The normal cells were treated with the IC₅₀ equivalent concentrations of the compounds. Table II lists the cytotoxicity data of compounds a–g in the normal cell line. The data shows that these compounds also exhibit cytotoxic effects towards the normal cell lines, which in this case are NIH 3T3 mouse embryonic fibroblasts. The results demonstrate that compounds b and d exhibited lower toxicity towards the normal cell lines. This implies that these compounds show specific cytotoxic effect towards cancer cell lines.

The effect of compounds b and d on cell cycle phase distribution was analyzed by flow cytometry. The MCF-7 cells were stained with PI and then treated with different concentrations (25 and 50 μM) of the compounds for 72 h. Significant DNA damage was indicated by the fluorescence patterns obtained from the flow cytometer (Fig. 2). Apoptotic cells were designated as shrunken cells with degraded chromatin, high side scatter and low forward scatter properties. The rise in the sub-G₁ cell population (hypodiploid DNA content) may be due to DNA fragmentation, which results in apoptotic cell death. The inhibition of cell cycle progression may be one of the molecular events concomitant with the cytotoxic activity of these compounds.

The effect of compounds b and d were further studied by evaluating their effects on the ΛΨm. MCF-7 cells were treated with 25 and 50 μM compound b and d and ΛΨm was measured by flow cytometry. The untreated control cells had intact mitochondria, and the majority of the cells were bioenergetically active, as evidenced by high Rh-123 uptake as compared with the positive control, which exhibited damaged mitochondria. The two compounds induced a significant increase in mitochondrial membrane potential loss. Ludartin (compound d) produced a more potent effect on the mitochondrial membrane potential loss as compared with lupeol (compound b) (Fig. 3). Mitochondria have a key function in the induction of apoptosis, as they are involved at an early period in the apoptotic pathway. Mitochondrial membrane potential is crucial in the regulation of apoptosis, and any reduction leads to increased mitochondrial penetrability and the opening of the permeability transition pore, which is a key stage in apoptosis (33). The opening of the permeability transition pore permits the release of factors such as cytochrome c and apoptotic inducing factor, which trigger the finishing and degradative stage of apoptosis.

The MCF-7 cells were treated with compound b and d for 24 h, stained with DAPI and visualized to determine the resulting morphological alterations. The untreated cells displayed a normal nuclear morphology, but ludartin- and lupeol-treated cells presented apoptotic bodies. The number of apoptotic bodies was greater at a higher dose. Camptothecin, which is a known apoptotic inducer, was used as a positive control and also led to the appearance of typical apoptotic bodies (Fig. 4).