The organic solvents, including methanol (MeOH), n-hexane, dichloromethane (CH₂Cl₂), ethyl acetate (EtOAc), and n-butanol (n-BuOH), were obtained from Duksan Chemical (Anseong, Korea). Purification of the compounds was performed by column chromatography by using silica gel (70 – 230 mesh, Merck #7734, Darmstadt, Germany) and octadecyl silica (ODS, 150 µm, YMC #AA12SA5, Kyoto, Japan). Pre-coated thin-layer chromatography (TLC) silica gel 60 F₂₅₄ (Merck #1.05554.0001) and TLC silica gel 60 RP-18 F_254S (Merck #1.05559.0001) were used to monitor the eluted compounds. Nuclear magnetic resonance (NMR) was performed on a Bruker Avance Digital 500 NMR spectrometer (Karlsruhe, Germany). The NMR solvents (CDCl₃ and CD₃OD) were purchased from Sigma-Aldrich (St. Louis, MO, USA). The absolute configuration was determined by a Jasco P-2000 polarimeter (Tokyo, Japan).

The dried stems of BP grown in Korea were purchased in 2015 at the local market in Daegu, Republic of Korea, and identified by Prof. Kyung-Sik Song. A voucher specimen (#KNUNPM BP-1) was deposited at the Laboratory of Natural Products Medicine, College of Pharmacy, Kyungpook National University.

BP (10 kg) was refluxed three times in MeOH for 6 h. The extracted solution was filtered and concentrated to obtain the MeOH extract (321.4 g). The MeOH extract was suspended in distilled water and successively partitioned with n-hexane, CH₂Cl₂, EtOAc, and n-BuOH. After evaporation, the fractions soluble in n-hexane (9.9 g), CH₂Cl₂ (49.6 g), EtOAc (42.8 g), and n-BuOH (35.6 g) were obtained. The CH₂Cl₂-soluble fraction was applied to a silica gel column (9 × 82 cm; CH₂Cl₂-MeOH, 500:1 → 1:1) to yield 25 fractions. Compounds 1 (51.4 mg) and 2 (76.0 mg) were obtained from fractions 2 and 5, respectively. Fraction 7 was rechromatographed on an ODS column (489.6 mg; 3 × 30 cm; water-MeOH, 7:3 → 1:9; flow rate 15 mL/min) to isolate compounds 3 (118.0 mg) and 4 (27.5 mg). Compounds 5 (20.0 mg), 6 (7.0 mg), and 7 (17.0 mg) were purified from fraction 11 (288.3 mg; 3 × 30 cm; water-MeOH, 4:1 → 1:9; flow rate 15 mL/min) by ODS medium pressure liquid chromatography (MPLC).

The human alveolar adenocarcinoma cell line A549 and a human normal epithelial cell line BEAS-2B were purchased from Korean Cell Line Bank (Seoul, Korea) and American Type Culture Collection (ATCC, Manassas, VA, USA), respectively. The cell culture medium used was Roswell Park Memorial Institute (RPMI) 1640 (Welgene, Gyongsan, Korea) supplemented with 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA) (v/v), 100 U/mL penicillin, and 100 µg/mL streptomycin (Welgene). The cells were cultured at 37 ℃ in an atmosphere of 5% CO₂. Subculturing was performed every 48 h.

To examine the cytotoxicity of the test samples on A549 and BEAS-2B cells, the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Biosesang, Seongnam, Korea) assay was performed. The cells were seeded at a density of 2 × 10⁴ cells/mL on a 48-well plate, incubated for 24 h, and the test sample was applied to the cells for 48 h. Subsequently, the media was removed, and 0.2 mg/mL MTT in phenol red-free and serum-free RPMI1640 medium was added. The insoluble formazan crystals produced by the live cells were dissolved in dimethyl sulfoxide after the removal of MTT solution. The optical density (OD) was measured at 545 nm by using a microplate reader. The cell viability was expressed as a percentage relative to the OD of the vehicle-treated control. For comparison, the anti-cancer reagent 5 µM doxorubicin (DOX; Sigma-Aldrich) was used as a positive control.

The cytotoxicity of the test sample on A549 cells was additionally evaluated through the use of a lactate dehydrogenase (LDH) release assay kit (CytoTox96 Non-Radioactive Cytotoxicity Assay, Promega, Madison, WI, USA) in accordance with the manufacturer's instructions. The supernatant and the cell lysate were prepared after the cell treatment described in section of MTT cell viability assay. The OD value was measured at 490 nm. The LDH leakage was calculated as follows:

LDH leakage (fold) = A/(A + B)

A, LDH in supernatant; B, LDH in cell lysate

A549 cells were seeded at a density of 2 × 10⁵ cells/mL in a 6-well plate. After 24 h, the cells were treated with test samples for 24 h. The cells were detached from the plate using 0.25% trypsin/ethylenediaminetetraacetic acid (EDTA; Welgene), washed once with phenol red-free and serum-free RPMI1640 medium, and collected. The collected cells were incubated with 5 µM of 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate and acetyl ester (CM-H₂DCFDA; Invitrogen, Eugene, OR, USA) for 15 min in the dark. The cells were washed once and resuspended in phenol red-free and serum-free RPMI1640 medium before flow cytometric analysis of 10⁴ cells was conducted on a FACS ARIA III (BD, San Diego, CA, USA).

The data are expressed as the mean ± standard deviation (SD), and differences were considered to be statistically significant at p < 0.05 determined by one-way analysis of variance (ANOVA; GraphPad Prism 7, La Jolla, CA, USA).

The A549 cell viability was significantly reduced to 45.83 ± 3.65% after treatment with 200 µg/mL MeOH extract of BP compared to control (CNT, 100.00 ± 23.12%; Fig. 1). Next, the four fractions obtained from BP were used to treat A549 cells: 100 µg/mL of the n-hexane and CH₂Cl₂ fractions decreased cell viability to 56.10 ± 11.99% and 52.76 ± 3.14%, respectively, whereas the EtOAc and n-BuOH fractions were not effective in reducing of cell viability. As the CH₂Cl₂ fraction was the most effective (Fig. 2), it was selected for further experiments.

Seven compounds were isolated from the CH₂Cl₂ fraction and identified as lupeol (1),2122 betulinic acid (2),23 (−)-rhododendrol (3),2425 platyphyllenone (4),26 platyphyllone (platyphyllonol, 5; obtained as a racemic mixture),27 (−)-centrolobol (6),2728 and oleanolic acid (7)2930 through the comparison of their NMR data with those of previous studies (Fig. 3) in which compounds 1 – 7 were identified as the constituents of BP.14313233

As shown in Fig. 4, the treatment with 2 and 4 – 7 at 50 µM resulted in a less than 50% survival rate of A549 cells compared to the control (100.00 ± 23.12%). In contrast, 1 and 3 did not reduce A549 cell viability to less than 50%, even at 50 µM, as described in previous reports.34353637 Compounds 2 and 7 decreased the cell viability to below 50% at 5 µM (30.49 ± 1.95% and 43.84 ± 2.62%, respectively). Compounds 238 and 734 have been previously examined for cytotoxicity against A549 cells. In these studies, it was shown that the concentration that resulted in 50% cytotoxicity (CC₅₀) was 43.4 µM after treatment with 2 for 72 h38 and 52.97 ± 1.22 µg/mL (approximately 116.16 µM) after treatment with 7 for 48 h.27 This difference might be a result of the cell concentrations used. At 50 µM, 4 reduced A549 cell viability to 18.93 ± 0.82%. In addition, 5 (46.54 ± 2.02%) and 6 (42.15 ± 1.13%) showed moderate cytotoxicity to A549 cells at 50 µM. The cytotoxic effects of 4 on the human NSCLC cell line NCI-H460 (CC₅₀, 8.2 ± 0.1 µM) has been reported.39 In addition, 5 and 6 effectively induced cytotoxicity in the human NCI-H187 SCLC cell line (CC₅₀ = 59.24 and 9.55 µM, respectively).40 However, to the best of our knowledge, this is the first report about the cytotoxic effects of 4 – 6 on A549 cells. Among the three compounds 4 – 6, 4 was the most effective and was applied to further experiments.

After treatment with 4, the cytosolic lactate dehydrogenase (LDH) leakage in A549 was induced in a concentration-dependent manner. In particular, the LDH leakage in the group treated with 50 µM 4 was 7.50 ± 0.50-fold greater compared with that of the control (CNT, 1.00 ± 0.04-fold; Fig. 5). Moreover, the production of intracellular reactive oxygen species (ROS) was significantly increased by 3.86 ± 1.05-fold by the treatment of 50 µM 4 compared with CNT (1.00 ± 0.12-fold; Fig. 6).

To determine the selectivities of compounds, the effects of compounds on the viability of human normal epithelial cell line BEAS-2B were examined. As shown in Fig. 7, compounds 1 and 3 did not show any cytotoxicities at the concentrations of 1–50 µM, although they revealed the reduction of A549 cell viability. In contrast, the viabilities of BEAS-2B cells were decreased to below 50% by 2 (20.53 ± 1.27%), 4 (13.55 ± 0.37%), and 5 (45.40 ± 4.07%) at 50 µM relative to the control (100.00 ± 3.78%). In particular, the reduction pattern of BEAS-2B cell viability induced by the most A549 cytotoxic compound 4, was similar to that of A549. Therefore, the selectivity of 4 for lung cancer cells has to be considered to develop the hit compounds. To the best of our knowledge, the effects of 1 – 6 on the BEAS-2B cell viability were firstly examined in this study, except for 7.41

The diarylheptanoid platyphyllenone (4) is a constituent of not only BP,16 but also Alnus japonica,4243 A. viridis,44 A. hirsute,45 and A. nepalensis.46 In a previous report, 4 showed anti-influenza activity against KBNP-0028 (H9N2) avian influenza virus, with a 50% effective concentration (EC₅₀) of 29.9 ± 2.5 µM.43 Additionally, 4 was reported to inhibit the proliferation of the human pancreatic ductal carcinoma cell line PANC-1 (CC₅₀= 6.06 µM) through the modulation of the shh-Gli-FoxM1 pathway.42 The anti-microbial activity47 of 4 has been also reported, however, other biological activities have been rarely investigated. Especially, there is only one report of the anti-lung cancer effect of 4 against NCI-H460 cells.39

The evaluation of the structure–activity relationships (SAR) by Novakoviæ et al. (2014)48 and Diniæ et al. (2016)39 revealed the structural aspects responsible for the cytotoxic effects of diarylheptanoids against NCI-H460 cells: i) a carbonyl group (C=O) at the C-3 position; ii) a double bond between C-4 and C-5, instead of a C-5 substitution with -OCH₃, -Oxyl, -OGlc, and -OH; iii) the absence of the-OH groups at C-3 and C-3.3948 A549 is a NSCLC cell line, for which considerable information about SAR is already known.39,48 The most effective diarylheptanoid 4 also satisfied the above three conditions, which suggested the importance of the positions of the carbonyl and hydroxyl groups and the double bond. This study is the first to report about the cytotoxic effects of 4 – 6 on the human NSCLC cell line, A549. Further studies are required to determine the value of 4 as a potential drug lead for NSCLC.