In Vitro Synergistic Inhibition of HT-29 Proliferation and 2H-11 and HUVEC Tubulogenesis by Bacopaside I and II Is Associated with Ca2+ Flux and Loss of Plasma Membrane Integrity

Abstract

We previously showed how triterpene saponin bacopaside (bac) II, purified from the medicinal herb Bacopa monnieri, induced cell death in colorectal cancer cell lines and reduced endothelial cell migration and tube formation, and further demonstrated a synergistic effect of a combination of bac I and bac II on the inhibition of breast cancer cell line growth. Here, we assessed the effects of bac I and II on the colorectal cancer HT-29 cell line, and mouse (2H-11) and human umbilical vein endothelial cell (HUVEC) lines, measuring outcomes including cell viability, proliferation, migration, tube formation, apoptosis, cytosolic Ca²⁺ levels and plasma membrane integrity. Combined bac I and II, each applied at concentrations below IC₅₀ values, caused a synergistic reduction of the viability and proliferation of HT-29 and endothelial cells, and impaired the migration of HT-29 and tube formation of endothelial cells. A significant enhancement of apoptosis was induced only in HUVEC, although an increase in cytosolic Ca²⁺ was detected in all three cell lines. Plasma membrane integrity was compromised in 2H-11 and HUVEC, as determined by an increase in propidium iodide staining, which was preceded by Ca²⁺ flux. These in vitro findings support further research into the mechanisms of action of the combined compounds for potential clinical use.

Keywords: Ca2+ flux; apoptosis; bacopaside I; bacopaside II; colorectal cancer cells; endothelial cells; migration; plasma membrane; proliferation; tube formation.

Figure 1

Comparison of the half maximal inhibitory concentration (IC₅₀) values of bac I and II, alone and combined, and their combination index (CIx) in HT-29, 2H-11 and HUVEC. Cell growth was measured by a crystal violet assay and IC₅₀ was determined by nonlinear regression analysis of 6 replicates. Dose–response curves show IC₅₀ for bac I (A–C), bac II (D–F) and combinations (G–I). Error bars represent the 95% confidence interval (CI). On the isobolograms (J–L), theoretical IC₅₀ lines (red solid lines) acquired from the IC₅₀ for each compound used individually, and the measured IC₅₀ (green square) for bac I and II combined, are plotted for each cell line, with 95% CI being represented by the red dotted lines for theoretical IC₅₀ lines and by the black lines for measured IC₅₀. Response surface methodology (RSM) analysis was performed on HUVEC viability, and the contour plot (M) and surface plot (N) are drawn, with the red line indicating the 50% probability isobole.

Figure 2

Bac I and II combined impaired proliferation of HT-29, 2H-11 and HUVEC. HT-29 (A), 2H-11 (B), and HUVEC (C) were treated with bac I and II, either alone or in combination, and cellular proliferation was measured over 3 days by a crystal violet assay. Data represent the mean absorbance of 5 to 6 replicates. Error bars represent the standard deviation (SD). Significant difference as compared to vehicle after 3-day treatment is indicated by asterisks (* p < 0.05; ** p < 0.01; **** p < 0.0001).

Figure 3

Bac I and II combined impaired migration of HT-29. (A) HT-29 was treated with bac I and bac II, either alone or in combination, and wound closure was measured after 48 h. Data represent the mean wound closure as a percentage of 5 to 6 replicates. Error bars represent the SD. Significant difference as compared to vehicle is indicated by asterisks (* p < 0.05; ** p < 0.01). (B) Representative images of scratch wounds at times 0 and 48 h are shown for treatment with vehicle and bac I/II 10/5 μM. Magnification ×40, scale bar = 0.5 mm.

Figure 4

Bac I and II combined impaired tube formation of 2H-11 and HUVEC. 2H-11 (A) and HUVEC (B) were treated with bac I and bac II, either alone or in combination, and the number of loops formed was counted after 2 h. Data represent the mean percentage in the number of loops formed, with respect to the vehicle for triplicates. Error bars represent the SD. Significant difference as compared to vehicle is indicated by asterisks (**** p < 0.0001). Representative images of 2H-11 (C) and HUVEC (D) treated with the vehicle, bac II 5 µM and bac I/II 10/5 µM are shown. Magnification 100×, scale bar = 0.2 mm.

Figure 5

Bac I and II combined induced caspase 3/7 activation in HUVEC, but not in HT-29. HT-29 (A), 2H-11 (B) and HUVEC (C) were treated with bac I and II, either alone or in combination, and caspase 3/7 activation was monitored using CellEvent Caspase-3/7 Green Detection Reagent. Data represent the mean number of caspase 3/7-positive cells per image for 6 replicates. Error bars represent the SD. A significant difference at 36 h as compared to the vehicle is indicated by asterisks (** p < 0.01; *** p < 0.001 **** p < 0.0001).

Figure 6

Bac I and II combined induced an increase in cytosolic Ca²⁺ in HT-29, 2H-11 and HUVEC. HT-29 (A), 2H-11 (B) and HUVEC (C) were treated with bac I and II, either alone or in combination, and cytosolic Ca²⁺ was monitored over 5 h by a Fluo-8 assay. Data represent the mean fractional increase in fluorescence with respect to time 0 of triplicates. Error bars represent the SD. Significant difference as compared with the vehicle after 5-h of treatment is indicated by asterisks * p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 7

Disruption of the plasma membrane by bac I and II combined was cell line dependent. HT-29 (A), 2H-11 (B) and HUVEC (C) were treated with bac I and II, either alone or in combination, and the disruption of the plasma membrane was monitored over 24 h using propidium iodide (PI). Data represent the mean number of PI-stained cells per image for 6 replicates. Error bar represents SD. Significant difference as compared with vehicle after 24-h treatment is indicated by asterisks * p < 0.05; **** p < 0.0001).