Bioactive Compounds from the Roots of Asiasarum heterotropoides

A new tetrahydrofuran lignan, (7S,8R,7'S,8'S)-3-methoxy-3',4'-methylenedioxy-7,9'-epoxylignane-4,7',9-triol (1), and 21 known compounds 2–22 were isolated from the roots of Asiasarum heterotropoides by chromatographic separation methods. The structures of all compounds 1–22 were elucidated by spectroscopic analysis including 1D- and 2D-NMR. Fourteen of these compounds (1–3, 7, 10, 12–17, 19, 21, and 22) were isolated from this species in this study for the first time. All of the isolates were evaluated for their anticancer activities using in vitro assays. Among the 22 tested compounds, two (compounds 5 and 7) induced the downregulation of NO production, FOXP3 expression, and HIF-1α transcriptional activity.

Some phytochemical and pharmacological studies of A. heterotropoides have reported several types of secondary metabolites, including essential oils, monoterpenes, lignans, alkaloids, and phenyl propanoids, that display antimicrobial, anti-tumor, anti-inflammatory, and larvicidal activities [10][11][12][13][14]. Despite their medicinal importance and availability, knowledge of the chemical constituents and biological activities of A. heterotropoides remains insufficient to evaluate their pharmacological effects.
In the course of our research to discover anticancer agents from medicinal herbs, the chromatographic separation of an 80% EtOH extract from the roots of A. heterotropoides resulted in the isolation and identification of a new tetrahydrofuran lignan (1) and 21 known compounds 2-22. The structures of all compounds 1-22 were elucidated by spectroscopic analysis, including NMR, and through comparison of the data with published values. All of the compounds 1-22 were evaluated for cytotoxic activity against human cancer cell lines and for inhibitory effects on nitric oxide production, FOXP3 promoter activation, and HIF-1α transcriptional activity. Here, we report the isolation, structural elucidation, and biological activities of these compounds from the roots of A. heterotropoides.
Our previous study showed that ethanol extracts of Asiasari radix have anticancer effects on human non-small cell lung carcinoma (NSCLC) cells [9]; therefore, we examined the biological activities of the 22 compounds related with anticancer effects. First, the cytotoxic effects of the isolated compounds were investigated using A549 human NSCLC cells and IMR90 human normal lung fibroblast cells. The cells were exposed to the test compounds (20 μM) or vehicle (DMSO) for 48 h and the cell viabilities were determined based on the mitochondrial dehydrogenase activities (Ez-Cytox). Remarkably, compounds 5 and 7 inhibited the growth of A549 lung cancer cells at 20 μM, as shown in Figure 4. The viability of A549 cells was reduced to 50% and 30% by compounds 5 and 7, respectively, when compared with the vehicle treatment. Notably, compounds 5 and 7 only marginally affected the growth of IMR90 normal lung cells, and their viabilities were maintained at over 85% using the same compound concentration (20 μM). Therefore, compounds 5 and 7 selectively killed A549 lung cancer cells without affecting normal cells. Nitric oxide (NO) synthesized from L-arginine by a group of nitric oxide synthases has been shown to promote tumor growth by regulating the expression of genes involved in cell mobility, invasion, and angiogenesis [39][40][41]. Chronic and substantial exposure to NO was also reported to contribute to cancer development [42,43]. In addition, the application of a highly selective NOS inhibitor, N G -monomethyl-L-arginine, monoacetate and the consequent reduction of the NO concentration was shown to sensitize radio-resistant NSCLC cells to radiotherapy [44]. Therefore, the inhibition of up-regulated NO production should be considered an anticancer strategy. In the present study, we used LPS-mediated NO production from Raw264.7 murine macrophage cells as an in vitro screening system to investigate the potential of the isolated compounds to inhibit NO production. As shown in Figure 5, four compounds (4, 5, 7, and 20) reduced the production of NO by more than 50% by Raw264.7 cells treated with 1 μg/mL LPS when compared with vehicle treatment ( Figure 5A), and their inhibitory effects were observed to be in dose-dependent ( Figure 5B). (B) Raw264.7 cells were treated with combination of 1 μg/mL LPS and the indicated concentrations of selected compounds. Extracellular NO was quantified and relative NO production was determined by comparing with LPS (+)/vehicle (V) treatment group. The data are presented as the means ± S.D. of three independent experiments. Differences between each treatment groups and LPS control group (LPS (+)/vehicle (V)) were compared and statistical significances are denoted as * p < 0.05, ** p < 0.01 or *** p < 0.001.
As a member of the forkhead box (FOX) family protein, FOXP3 is a master regulator of regulatory T cells (T reg ) and functions in the maintenance of immune tolerance [45]. Abnormally high expression of FOXP3 was observed in various types of tumors [46][47][48][49] and was known to affect the sensitivity of cancer cells to anticancer drugs [50] and radiotherapy [51]. Therefore, the down-regulation of FOXP3 is one antitumor strategy. In this study, we investigated the effects of isolated compounds on FOXP3 expression using the FOXP3 promoter-luciferase reporter system, FOXP3-Luc#3 stable cells. The expression of luciferase in FOXP3-Luc#3 cells was derived by phorbol 12-myristate 13-acetate (PMA) treatment. As shown in Figure 6A, compounds 4, 5, and 7 significantly inhibited the PMA-mediated FOXP3 promoter activation by 82%, 55%, and 63%, respectively, compared with vehicle treatment. In addition compounds 4, 5, and 7 could inhibit FOXP3 promoter activity in a dose-dependent manner ( Figure 6B). The vehicle, as the control (V, 0.1% DMSO), did not affect PMA-induced FOXP3 promoter activity. Figure 6. Effects of compounds on FOXP3 promoter activity. (A) FOXP3-Luc#3 cells were exposed to combinations of PMA (5 ng/mL) and each compound (20 μM) or vehicle as a control (V, DMSO); and (B) FOXP3-Luc#3 cells were exposed to increasing concentrations of each selected compound (compounds 4 (C4), 5 (C5), or 7 (C7)). After 24 h incubation, luciferase activity in the whole cell lysate (WCL) was quantified, and the relative activity was determined by comparing with the PMA (+)/vehicle (V) treatment group. The data are presented as the means ± S.D. of three independent experiments. Differences between each treatment groups and the PMA control group (PMA (+)/vehicle (V)) were compared and statistical significances are denoted as * p <0.05, ** p < 0.01 or *** p < 0.001.
As a key regulator during tumor development, hypoxia inducible factor-1 (HIF-1), in particular HIF-1α is a promising target for anticancer drug development. To investigate whether the isolated compounds could modulate the transcriptional activity of HIF-1α under hypoxic stress, we performed a luciferase reporter assay using a NIH3T3/HIF-luc stable cell line carrying hypoxia response element (HRE) sequences. To introduce hypoxic stress, the cells were transferred into a hypoxic chamber containing 1% O 2 and maintained for 24 h. Cells were pretreated with compound or vehicle 2 h before exposure to hypoxic gas. At the end of hypoxic stress, WCL was prepared and applied to the luciferase activity assay. As shown in Figure 7A, hypoxic stress increased luciferase activity up to 20-folds, compared with the normoxic condition (20% O 2 ). Among the twenty-two tested compounds, two (compounds 5 and 7) successfully inhibited HIF-1α transcriptional activity. Compounds 5 and 7 significantly inhibited hypoxic gas-induced luciferase activity by 92.5% and 94.6%, respectively, compared with the hypoxia control (1% O 2 (+), vehicle treatment). As expected, both compounds could dramatically inhibit HIF-1α transcriptional activity in a dose-dependent manner ( Figure 7B). Their effective doses at 50% inhibition (ED 50 ) were 7.6 and 8.1 μM, respectively. The data are presented as the means ± S.D. of three independent experiments. The differences between each treatment group and the hypoxia control group (1% O 2 (+)/vehicle (V)) were compared and statistical significance is denoted as * p < 0.05, ** p < 0.01 or *** p < 0.001 and (C) Compounds 5 and 7 inhibited hypoxic gas-induced HIF-1α expression in A549 cells in a dose-dependent manner. β-actin was applied as a loading control.
To elucidate the mechanism of the decrease in HIF-1α transcriptional activity induced by compounds 5 and 7, we first investigated changes in the intracellular levels of HIF-1α in A549 human lung cancer cells treated with a combination of 1% O 2 exposure and compound pretreatment. Increasing concentrations of compounds 5 and 7 were applied to A549 cells 2 h before exposure to hypoxic gas. After 20 h, the total protein was extracted, and expression of intracellular HIF-1α was visualized by western blotting. As shown in Figure 7C, hypoxic gas-induced HIF-1α expression was reduced by compounds 5 (C5) and 7 (C7) in a dose-dependent manner. However, HIF-1β, which is constitutively expressed, was not affected by C5 and C7. Furthermore, we investigated the PI3K/Akt/mTOR signaling, a HIF-1α synthetic pathway, by observing phosphorylation of the S6 ribosomal protein, which is the substrate of the PI3K/Akt/mTOR downstream kinase, p70 S6 kinase. As shown in Figure 7C, neither C5 nor C7 affected the level of phosphorylated S6 protein. Taken together, compounds 5 and 7 can inhibit HIF-1α transcriptional activity ( Figure 7B) by decreasing intracellular levels of HIF-1α by regulating its stability.
Previous pharmacological studies have reported that compounds 5 and 7 showed cytotoxic activity against only a human proximal tubular epithelial cell line (HK-2) [52,53]. To the best of our knowledge, this is the first biological report of the inhibitory effects of compounds 4, 5, and 7 on NO, FOXP3, and HIF-1α.
Resveratrol can protect RAW264.7 cells from LPS-mediated inflammatory insults by inhibition of proinflammatory mediators including NO [54]. Al Dhaheri et al. demonstrated that ethanolic extract of Origanum majorana exerts its antitumor and antimetastatic potential in MDA-MB-231 breast cancer cells through down-regulation of NO production [40]. Aizman et al. demonstrated that farnesylthiosalycilic acid can down-regulate FoxP3 protein by Ras inhibition leading to induction of anti-tumor cytotoxic T-cell reactivity in glioma cells [55]. A bunch of previous reports showed that single compounds which can down-regulate HIF-1 signaling exert anticancer activities in diverse in vitro and in vivo systems [56][57][58][59][60][61].

Plant Material
The

FOXP3-Lluciferase Stable Cell Line
The FOXP3-luciferase reporter vector (pGL4.17-FOXP3-luc) was a generous gift from Professor Im, S.H. (School of Life Sciences and Immune Synapse Research Center, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea). The conserved noncoding sequences 1 and 3 (CNS1 and CNS3, respectively), which were shown to be critical regulatory elements controlling FOXP3 gene expression [62] were subcloned into the upstream region of luciferase in the pGL4.17 empty vector (Promega, Madison, WI, USA). The pGL4.17-FOXP3-luc vector was transfected into E6-1 cells by electroporation. Drug resistant cells were selected under 800 μg/mL G418 (Invitrogen). Expression of luciferase was induced by stimulation of cells with 5 ng/mL PMA (Sigma, St Louis, MO, USA). The stable and PMA-responsive single clone (FOXP3-Luc#3) was obtained by limited dilution of the parental resistant cells in 96-well culture plates. The FOXP3-Luc#3 clone was maintained at 400 μg/mL G418.

Cell Viability
Cell viability was quantified using the Ez-Cytox cell viability assay kit (Daeil Lab Service, Seoul, Republic of Korea) as previously described [63].

NO Assay
Raw264.7 cells were inoculated at 4.5 × 10 5 cells/well in 48-well tissue culture plates containing 500 μL of fresh medium and grown overnight. The cells were treated with a combination of lipopolysaccharide (LPS, Sigma, 1 μg/mL) and test compound (20 μM). The NO released from LPS-treated cells into the culture medium was quantified using a commercially available NO detection kit (iNtRON Biotechnology, Seoul, Republic of Korea) as described in the manufacturer's guide.

Luciferase Reporter Assay
After treatment as designated, the cells were transferred to 1.5 mL tubes and pelleted by centrifugation at 500 g for 5 min. Cells were washed twice with ice-cold phosphate-buffered saline (PBS), and WCL was prepared by repeated freezing and thawing in 1X passive lysis buffer (Promega). The insoluble cellular debris was removed by centrifugation at 14,000 rpm for 10 min. Luciferase activities in the WCL were determined using the luciferase assay system (Promega) according to the manufacturer's instructions. Light intensities were monitored using a GloMax luminometer (Promega).

Western Blot Analysis
Changes in intracellular protein levels were examined by Western blotting as previously described [9]. In short, A549 cells were cultured under normoxic (20% O 2 ) or hypoxic (1% O 2 ) environments, and WCL was prepared using ice-cold RIPA buffer (Thermo Scientific, Rockford, IL, USA). Equal amounts of proteins (20 μg) were separated on SDS-PAGE gels and electro-blotted onto nitrocellulose membranes. The membranes were blocked with a 5% (w/v) skim milk solution in 0.1% (v/v) Tween-20 PBS and probed with primary antibodies at 4 °C overnight. The primary antibodies were captured by horseradish peroxidase-labeled secondary antibodies. Immuno reactive bands were visualized with the SuperSignal West Femto Kit (Thermo Scientific). All primary antibodies except HIF-1α (Bethyl Laboratories, Montgomery, TX, USA) and β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were obtained from Cell Signaling Technology (Danvers, MA, USA).