The MDA-MB-231 cellular model was established from the pleural effusion of a patient with metastatic breast cancer. These cells lack the expression of receptors for estrogen, progesterone, and human epidermal growth factor and represent a model for clinical TNBC.14 The MDA-MB-231 cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA), and was maintained in RPMI medium with L-glutamine and supplemented with 5% fetal bovine serum (Life Technologies, Grand Island, NY, USA). The cultures were routinely expanded at a 1:4 ratio when 70% confluent.

The Chinese nutritional herb VY is a common constituent in herbal formulations used in TCM. This herb was provided by Dr. George YC Wong (coauthor). Traditionally, herbal formulations in TCM are prepared as an aqueous decoction in boiling water for patient consumption. To simulate this, a nonfractionated aqueous extract from the leaves of VY was prepared following an optimized, published protocol.5-9 The stock solution (aqueous supernatant) was obtained after multiple boiling and centrifugation cycles at a VY concentration of 20mg/20mL. This stock solution was reconstituted in the RPMI culture medium at a concentration of 1mg/mL. For the experiments on cell cultures, the RPMI stock was serially diluted using the RPMI culture medium to obtain the final concentration range in μg/mL.

To determine the effective growth-inhibitory concentration range for VY, cell viability was monitored using the CellTiter-Glo assay (Promega Corporation). The cells were treated with VY at a concentration range of 20, 40, 80, 100, and 400 μg/mL for 6 days, and the cultures were maintained at 37°C in a CO2 incubator. Cells maintained without any treatment served as the control. Cell viability was measured using a Fluoroskan plate reader (Thermo Fisher Scientific Inc). The data were presented as relative luminescent units (RLUs) and as the percentage of inhibition relative to the untreated control.

To perform the anchorage-independent (AI) growth assay, a suspension of MDA-MB-231 cells was prepared in 0.33% agar at a cellular density of 5.0x10^5/mL. The cell suspension was treated with VY at concentrations of 100, 500, and 1000 μg/mL cells without any treatment served as the untreated control. The cell suspensions were overlaid on a basement layer of 0.6% agar. The cultures were maintained at 37°C in a CO2 incubator for 21 days, and AI colonies formed in 0.33% agar were counted at x10 magnification.8 The data were presented as the AI colony number and as the percentage of inhibition relative to the untreated control.

For the cell cycle progression assay, cells were treated with VY at concentrations of 20, 40, and 80 μg/mL. Cells maintained in the culture medium without any treatment served as the control. Cell cycle analysis was performed according to an optimized and published protocol. The DNA content of cells was determined using a Becton Dickinson FACScan Flow Cytometer (BD Biosciences, Research Triangle Park, NC, USA) and analyzed using FACS Express software (version 3.06; De Novo Software, Glendale, CA, USA). The distribution of the individual cell population in the G1 (quiescent), S, and G2 (proliferative) phases of the cell cycle was determined.

The Western blot assay was performed according to an optimized and published protocol. The cells were treated with VY at concentrations of 40 and 80 μg/mL. Cells without any treatment served as the control. An equal quantity of cellular proteins was separated on 10% sodium dodecyl sulfatepolyacrylamide gels (SDS-PAGE; Mini-PROTEAN TGX, Bio-Rad Laboratories). The gels were directly incubated with relevant primary and secondary antibodies (see Supplementary Table).

The chemiluminescent signal was developed with ECL-plus reagent (Bio-Rad Laboratories) and detected by autoradiography. The signal intensity of β-actin (loading control) was used to normalize the signal intensity of cellular proteins. The signal intensity of proteins was quantified using a Molecular Imager GS800 and Quantity One software (Bio-Rad Laboratories) and was presented as arbitrary scanning units. Additionally, these data are presented as the total protein: β-actin ratio.

The presence of cells in the sub-G0/G1 (apoptotic) phase of the cell cycle was monitored by flow cytometry. Cells treated with VY at concentrations of 20, 40, and 80 μg/mL and the untreated control were sorted, and the apoptotic cell population was determined. The data are presented as the percentage of sub-G0/G1 cells.

An optimized, published protocol for caspase 3/7 activity used the Caspase-Glo assay kit (Promega Corporation, Madison, WI, USA). A cellular homogenate prepared from the cells treated with VY at 20, 40, and 80 μg/mL and a homogenate from untreated cells were used to measure caspase 3/7 activity. The luminescence was measured using a luminometer (Thermo Fisher Scientific Inc, Waltham, MA, USA). The data were expressed as RLUs.

To examine the functional specificity of caspase activity, the synthetic peptide pan-caspase inhibitor Z-VAD-FMK (Cell Signaling Technology; catalog No. 60332) was used. The untreated control, cells treated with 80 μg/mL of VY, cells treated with 10 μM of Z-VAD-FMK, and cells treated with VY + Z-VAD-FMK were processed for the measurement of caspase 3/7 activity. The data were presented as RLUs.

The experiments for dose response, caspase activity, and pan-caspase inhibitor were conducted in triplicate, and data are presented as mean (SD). The experiments for cell cycle progression and RB signaling were conducted in duplicate, and the data are presented as arithmetic means. Dunnett multiple comparison test was used as a post hoc test of 1-way analysis of variance (ANOVA). Moreover, an independent sample Student t test was used for comparison between multiple treatment groups and a common control group. Data analysis was carried out using Microsoft Excel 2013 XLSTAT-Base software. The generated P values are reported. P< 0.05 was considered to indicate a statistically significant difference.

This experiment was conducted to determine the range of growth inhibitory concentrations in response to treatment with VY. The growth of MDA-MB-231 cells was inhibited in a dose dependent manner as evidenced by reductions in cell viability. VY concentrations of 20, 40, 80, 100 and 400 μg/mL exhibited inhibition rates of 53.8%, 61.3%, 76.9%, 84.6% and 92.3% respectively, relative to the untreated control (Table 1). This dose response identified 18.6 μg/mL of VY as the 50% inhibitory concentration.

This experiment was conducted to examine the effect of VY on the formation of AI colonies. AI colony formation represents an in vitro surrogate endpoint for tumorigenic cells. The data presented demonstrates that in response to treatment with VY at the concentrations of 100, 500 and 1,000 μg/mL the AI colony number was found to decrease the inhibition rates by 42.9%, 63.7% and 91.9% respectively, relative to untreated control (Table 2).

These data on AI colony formation identified 196.5 μg/mL of VY as the 50% inhibitory concentration.

The data generated from this experiment demonstrated that VY treatment resulted in arrest of cells at the S phase of cell cycle in a dose-dependent manner. The G2 phase of the cell cycle was found to be progressively decreased and was undetectable in response to the maximum cytostatic concentration of 80 μg/mL VY. Control versus VY 80 μg/mL. % S P=0.01. G2 phase: Not detectable.

The VY treatment groups were compared with a common control group. The P values for VY at 20 μg/mL were as follows: % S, P=0.5. % G2, P=0.5. The P values for VY at 40 μg/mL were as follows: % S, P=0.04. % G2, P=0.04 (Figure 1A).

Representative DNA histograms documented the effects of VY on cell cycle progression. Similar to the data in Figure 1A, DNA histograms demonstrated a progressive decrease in the G2 phase of the cell cycle and abrogation of the G2 phase in response to treatment with the higher dose of VY (Figure 1B-E).

The RB signaling pathway operates during the G1-S phase and/or S-G2 phase transition of cells. For the S-G2 phase transition, the expression status of cyclin E (encoded by CCNE1), CDK2 (CDK2), pRB (RB1), and E2F1 (E2F1) is critical. The data demonstrated that VY at a concentration of 80 μg/mL inhibited the expression of cyclin E, CDK2, pRB, and E2F1, relative to the untreated control (Figure 2A).

The data presented as the total protein: β-actin (ACTB) ratio demonstrated that in response to treatment with 80 μg/mL of VY, cyclin E expression decreased by approximately 78% P=0.02, CDK2 expression decreased by approximately 47% P=0.04, pRB expression decreased by approximately 31% P=0.04, and E2F1 expression decreased by approximately 92.3% P=0.01 relative to the control (Figure 2B).

Treatment with VY at 40 μg/mL did not significantly affect the expression of cyclin E P=0.5, CDK2 P=0.5, pRB P=0.5, or E2F1 P=0.5 relative to the control.

The experiment designed to examine the effect of VY on cellular apoptosis monitored the population of cells in the sub-G0 phase of the cell cycle. The data demonstrated that VY treatment resulted in a dose-dependent increase in the cell population in the subG0 phase of the cell cycle. Treatment with VY at a concentration of 80 μg/mL induced an approximately 9.8-fold increase P=0.01 relative to the untreated control (Figure 3A).

At the mechanistic level, VY treatment resulted in a dose-dependent increase in caspase 3/7 activity. VY at a concentration of 80 μg/mL induced an 18.8-fold increase P=0.01 in caspase 3/7 activity relative to the untreated control (Figure 3B).

The experiment designed to confirm the functional specificity of caspase 3/7 induction used the pancaspase inhibitor Z-VAD-FMK. The data presented in Figure 3C demonstrated that VY at a concentration of 80 μg/mL increased caspase 3/7 activity by approximately 8.8-fold P=0.01 relative to the untreated control, and the VY-induced caspase activity was inhibited by approximately 57.1% P=0.04 in response to treatment with Z-VAD-FMK relative to treatment with VY alone (Figure 3C).