Adlay Seed (Coix lacryma-jobi L. var. Ma-yuen Stapf.) Ethanolic Extract Fractions and Subfractions Induce Cell Cycle Arrest and Apoptosis in Human Breast and Cervical Cancer Cell Lines

Different dosages of adlay seeds and extracts from four parts of the plant (adlay hull (AHE), testa (ATE), bran (ABE) and polished adlay (PAE)) were tested. Treatment with AHE-Ea, ATE-Ea, ATE-Bu and ABE-Bu that started from 12.5 μg/mL for 72 h significantly inhibited MCF-7 cell viability in a dose-dependent manner. Moreover, fractionation that was based on polarity resulted in the subfractions having different effects. All subfractions showed antiproliferative effects on MCF-7 cell except ATE-Wa, ABE-Wa, PAE-He, PAE-Ea and PAE-Bu. (Figure 1A).

As for effects on HeLa cells, all fractions showed significant growth inhibitory effects at 100 μg/mL concentrations. Notably, the ATE-Ea, ABE-Ea and ABE-Wa subfractions showed promising antiproliferative effects in a dose-dependent manner, unlike ATE-He, ABE-Bu, PAE-He and PAE-Ea (Figure 1B).

Since AHE and ATE showed the most antiproliferative effects on cancer cell lines, we further explored the effects of their subfractions. The AHE and its fourteen subfractions were obtained according to the similarity of each collection by thin layer chromatography (TLC). In MCF-7 cells (Figure 2A), the AHE-Ea-(A–N) subfractions significantly decreased MCF-7 cell viability with IC50 ranging from 13.7 to 146.3 μg/mL (Table 1). In HeLa cells (Figure 2B), aside from the AHE-Ea-(A–C) subfractions, all of the subfractions had significant growth reduction effects. As for the ATE and its eight subfractions, they significantly decreased MCF-7 cell viability with IC50 ranging from 13.7 to 146.3 μg/mL (Table 2). Similar effects were demonstrated in HeLa cells with IC50 ranging from 13.7 to 146.3 μg/mL (Table 2) and the subfraction ATE-Ea-F showing the best growth inhibition effects (Figure 2B).

Following the MTT results, the AHE-Ea-K, AHE-Ea-L, ATE-Ea-E and ATE-Ea-F subfractions showed the best growth inhibitory effects on cancer cell lines. Therefore, we further explored their effects on cancer cell cycle distribution using flow cytometry in both MCF-7 and HeLa cells. Different concentrations of AHE-Ea-K and AHE-Ea-L (0, 25, 50 and 75 μg/mL) were used for 48 h. The treatments resulted in the significant cell cycle arrest of MCF-7 cells in the Go/G1 phase (Table 3) and HeLa cells (Table 3) in the S phase. The ATE-Ea-E and ATE-Ea-F treatments of 0, 25, 50 and 75 μg/mL for 48 h caused the significant cell cycle arrest of MCF-7 cells and HeLa cells in the G0/G1 phase (Table 3).

Since activation and cleavage of caspase-3 is considered to be the key modulator in cell apoptosis [6], we used flow cytometry to explore the changes in caspase-3 activity in cancer cell lines following treatment with 75 or 100 μg/mL of AHE-Ea-K, AHE-Ea-K L, ATE-Ea-E or ATE-Ea-F for 48 h. The treatments significantly increased the caspase-3 activity, as shown by the strengthening of the relative fluorescence signal (Figure 3), as well as quantitation (Table 4).

To explore the anticancer effects of AHE and ATE on Hela and MCF-7 cells, the cells were treated with 50 or 75 μg/mL of ATE or AHE for 48 h. The ATE and AHE treatment could successfully suspend the cyclin D1 and CDK4 protein expression (Figure 4A). To explore the potential components in the adlay extracts, the results of the LC-MS analysis of the combinations of active subfractions from AHE-Ea and ATE-Ea are shown in the Supplementary Material, Figure S1. According to their inhibitory activities on HeLa and MCF-7 (Table 1 and Table 2), AHE-EA-E–L, ATE-EA-E and ATE-Ea-F were combined to be regarded as active subfractions. Among the compounds that were isolated from the Coix seeds, naringenin, quercetin and vanillin were identified in the sum intensity of the total ion chromatogram (TIC) of the AHE-Ea active combined subfraction. In comparison to AHE-Ea, ATE-Ea-E and ATE-Ea-F contained more phenolic compounds, including quercetin, luteolin, apigenin, homoeriodictyol, eriodictyol, naringenin, isoliquiritigenin, chrysoeriol and vanillic acid. Based on the LC-MS analyses, quercetin and naringenin both existed in the active combined subfractions from AHE and ATE, which means that these two compounds could play important roles in anticancer properties. As shown in Figure 4B, quercetin increased the PARP and decreased CDK4 and Cyclin D1 protein expression, which led the apoptosis progression and caused the cell cycle arrest in the G0/G1 phase when treated for 48 h. Otherwise, naringenin did not show any significant activity under the same dosages as quercetin (data not shown).

The Coix seeds were processed by a dehulling machine to obtain the outer shell (35.8% of the weight of the Coix seeds) and then the skin and Coix seeds were air-sieved to obtain the testa (6.8% of the weight of the Coix seeds). The Coix seeds were further divided into Coix bran (4.3% of the weight of the Coix seeds) and refined white Coix seed (53.1% of the weight of the Coix seeds) [8]. Dehulled adlay, consisting of bran and endosperm, is regarded as a healthy cereal material, while the adlay hull and testa are treated as waste. The extracts were first extracted using 90% ethanol and then they were dried, resuspended in 10% methanol (aq.) and partitioned with 1:1 volume n-hexane until they were colorless. Then, they were dried under a vacuum to obtain the n-hexane fraction (He). The aqueous layer was further partitioned using ethyl acetate (EA) and n-butanol (Bu) to obtain the next two fractions (Figure 5) [8].

The EA fractions of the different parts of the Coix seeds (bran and testa) were eluted using silica gel open column chromatography (OCC) with gradually increasing proportions of He or EA, from 0% to 100%, and were then eluted using methanol. The eluents were expanded with thin-layer chromatography (TLC). Subfractions with similar chemical compositions from the TLC were combined to obtain AHE-EA-A–N and ATE-EA-A–H.

After the fractions were air-dried, DMSO was used as a solvent to make the 100 mg/mL of stock solution for further experiments.

HeLa (human cervix adenocarcinoma cell line) and MCF-7 (human breast adenocarcinoma) cells were obtained from the Bioresource Collection and Research Center (BCRC, Hsinchu, Taiwan). The cells were maintained in 10% fetal bovine serum (FBS, Biological Inc., Kibbutz Beit-Haemek, Israel) and supplemented with 1% penicillin–streptomycin–amphotericin B solution in Dulbecco’s modified Eagle’s medium (DMEM, BRL Gibco Inc., Grand Island, NY, USA) or in minimum essential medium (MEM) in a humidified incubator (Asheville NC, North Carolina, USA (37 °C; 5% CO₂)).

The cells were planted in 96-well plates (2 × 10³ cells/well) and treated with different concentrations (12.5, 25, 50, 100, 200 μg/mL) of the adlay seed extracts. The cell viability was analyzed using an MTT assay (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; Abcam, Cambridge, MA, USA). At the end of the treatment, a serum-free medium with 1 mg/mL of MTT was substituted by the conditional medium and incubated for an additional 4 h. Subsequently, the media were removed. The crystal formazan was dissolved by in 100 μL/well of dimethyl sulfoxide (DMSO; ECHO Chemical Co. Ltd., Taipei, Taiwan). The optical density was measured using a VERSA Max microplate reader (Molecular Devices, San Jose, CA, USA) at 570 nm and 630 nm as the reference wavelength.

To analyze the cell cycle distribution, the cells (3 × 10⁵) were planted into 6-well plates. After attaching overnight, the cells were treated with the AHE-Ea-K, AHE-Ea-L, ATE-Ea-E and ATE-Ea-F extracts in a culture medium containing 10% FBS for 48 h. At the end of the incubation, the cells were detached using trypsin. The cells were fixed in 70% alcohol at −20 °C and stained with a propidium iodide (PI) solution (2 mg of DNAse-free RNAse A and 0.4 mL of 500 μg/mL PI added to 10 mL of 0.1% Triton X-100 in PBS; Sigma-Aldrich, St. Louis, MO, USA) at room temperature for 30 min. Then, using a BD FACSCanto flow cytometer (BD Biosciences, San Jose, CA, USA), a minimum of 10,000 cells per sample was collected and further analyzed using CellQuest software (BD Biosciences).

The cells were planted in 6-well plates (3 × 10⁵ cell/well) and treated for 48 h. The cells were then trypsinized to detach the cell and then, we added the caspase-3-specific fluorescence reagent PhiPhiLuxTM–G2D2 (Millipore, Billerica, MA, USA) for 1 h at 37 °C, following the manufacturer’s instructions. We used flow cytometry to capture the fluorescence signals. The signals were quantified and compared to the untreated control and unstained groups.

The cells were lysed in an RIPA lysis buffer containing protease and phosphatase inhibitors (Roche, Mannheim, Baden-Württemberg, Germany). The protein was quantitated using a bicinchoninic acid (BCA) assay and sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with 95 voltages and a 2 h transfer to a polyvinylidene fluoride (PVDF) membrane. We used a 5% bovine serum albumin (BSA) solution to block for 1 h. Subsequently, the membranes were incubated with the primary antibodies CDK4 (1:1000; Abcam, Cambridge, UK), cyclin D1 (1:1000; Cell Signaling, Boston, MA, USA), PARP (1:1000; Cell Signaling) and GAPDH (1: 10,000; Proteintech, Rosemont, IL, USA) at 4 °C overnight and then with a horseradish peroxidase (HRP)-conjugated secondary antibody (1: 10,000) for 1 to 2 h. The signals were captured by an eBlot Touch Imager (eBlot Photoelectric Technology, Shanghai, China). The band densities were determined using the Image J software program, version 1.52 (NIH, Bethesda, MD, USA). The expression levels of these target proteins were analyzed in three individual experiments.

All data are presented as the mean ± SD and the MTT results were analyzed using Student’s t-tests. The statistical significance was presented by p-values of less than 0.05 and p-values of less than 0.001. The flow cytometry results were analyzed using one-way analysis of variance and Duncan’s new multiple range tests to examine the differences between the treatments using the statistical package for social science (SPSS) software.