Fig 1.
Schematic representation of a network of PI3K/AKT/mTOR signaling.
Inhibitors targeting the pathways are used in this study. Blue arrows and black lines represent activating and inhibitory connections, respectively. GDC-0941, pan-PI3K inhibitor; BEZ235 and GSK2126458, dual PI3K/mTOR kinase inhibitors; AZD8055, AZD82014 and KU-0063794, dual mTORC1/mTORC2 kinase inhibitors; and everolimus, allosteric mTORC1 inhibitor.
Fig 2.
Relationship between drug sensitivity, mutation status, receptor status and pathway utilization.
IC50 values for everolimus are represented on the y-axis and individual cell lines on the x-axis. Orange shading in the matrix below indicates PTEN, PIK3CA, RAS and BRAF mutations (Roche Cancer Genome Database 2.0 [45]), and identifies cell lines that are ER positive, HER2 positive and triple-negative (TN). Relative levels of phosphorylation of p70S6K, AKT and ERK from the respective breast cancer cell lines (untreated) are shown as bar graphs. Bands are normalized to tubulin control and bars represent changes in fold compared with BT20 and expressed as the mean of two experiments.
Fig 3.
Relationship between drug sensitivity and pathway utilization in MCF-7 parental and sub-lines.
IC50 values for everolimus are represented on the y-axis and individual cell lines on the x-axis. Orange shading in the matrix indicates estrogen receptor positive (ER), and triple-negative (TN) cell lines. Relative levels of phosphorylation of p70S6K, AKT and ERK of breast cancer cell lines are shown as bar graphs. Bands are normalized to tubulin or actin control and bars represent changes in fold compared with MCF-7 and expressed as the mean of two experiments.
Fig 4.
Signaling pathway usage and growth inhibitory response of MDA-MB-231, MDA-MB-436, BT20 and HCC1143 exposed to different drugs.
(A) IC50 values (50% inhibition of thymidine incorporation) are shown for BEZ235 (BEZ), GSK2126458 (GSK) and AZD8055 (AZD). Cells were treated with drugs for 3 days and cell proliferation was measured by the thymidine incorporation assay. Results are shown as the mean ± standard error from triplicate experiments. (B) Signaling pathway usage as measured by basal protein phosphorylation of AKT, p70S6K, rpS6 and 4E-BP1 in the four cell lines. Immunoblots with antibodies specific for phosphorylated proteins and their respective total protein are indicated. Actin is the loading control.
Fig 5.
The growth inhibitory effects of drug combinations on MDA-MB-231, MDA-MB-436, BT20 and HCC1143 breast cancer cell lines.
Growth inhibitory effects of combinations of everolimus with BEZ235 (BEZ) (left hand panel), GSK2126458 (GSK) (middle panel) and AZD8055 (AZD) (right hand panel) using the Bliss additivism method. Dashed line, Bliss additivity curve, representing the theoretical expectation if the combined effects of everolimus with kinase inhibitors were exactly additive. Averages of three independent experiments are shown.
Fig 6.
The cellular response to drug combinations of breast cancer cell lines.
(A) MDA-MB-231, (B) MDA-MB-436, (C) BT20 and (D) HCC1143 breast cancer cells were treated with drugs for 24 h and signaling pathway usage was measured by protein phosphorylation of AKT, p70S6K, rpS6, 4E-BP1 and ERK in the four cell lines. Immunoblots with antibodies specific for phosphorylated proteins are indicated. Actin is the loading control.
Table 1.
Bliss values for combinations of everolimus and PI3K/mTOR inhibitors in four cell lines.
Synergy, as noted by a positive Bliss value, was observed in all the cell lines in the combinations of everolimus and PI3K/mTOR or mTOR inhibitors.
Fig 7.
Comparison of viability assay using Alamar Blue and proliferation assay using thymidine incorporation, and the cell cycle changes in drug treatment.
The inhibitory effects of drug combinations on MDA-MB-231, MDA-MB-436, BT20 and HCC1143 breast cancer cell lines were compared using (A) H3-thymidine incorporation assay and (B) Alamar Blue viability assay. BEZ, BEZ235 (10nM); GSK, GSK2126458 (2.5nM); AZD, AZD8055 (10nM); EVL, everolimus (100nM). Averages of three independent experiments are shown. (C) Scatter plot showing the relative thymidine incorporation (x-axis) and Alamar Blue fluorescence (y-axis). Significant correlation (Pearson Correlation, R = 0.96; Linear Regression, P = 1.54 x 10−18) was found between the thymidine uptake assay and viability assay in MDA-MB-231 (blue), MDA-MB-436 (red), BT20 (yellow); HCC1143 (green). Averages of two independent experiments are shown. (D) Cell cycle analysis of treated MDA-MB-231 cells. Cells were treated with DMSO control, 100 nM everolimus (EVL), 100nM BEZ235 (BEZ), everolimus and BEZ235 (EVL BEZ) combination. Cell number in each phase of the cell cycle was determined and calculated as a percentage of the total cell population.
Fig 8.
The growth inhibitory effects of drug combinations on MDA-MB-231, MDA-MB-436, BT20 and HCC1143 breast cancer cell lines.
Growth inhibitory effects of combinations of everolimus with GDC-0941 (GDC) (left hand panel), KU-0063794 (KU) (middle panel) and AZD2014 (AZD) (right hand panel) using the Bliss additivism method. Dashed line, Bliss additivity curve, representing the theoretical expectation if the combined effects of everolimus with kinase inhibitors were exactly additive. Averages of three independent experiments are shown.
Fig 9.
Growth sensitivity to the combination of everolimus and kinase inhibitors.
Synergy, positive Bliss value, was observed for all drugs tested. Bliss = 0 indicates the combination is additive; Bliss > 0 indicates the percentage increase in maximal inhibition above additivity (synergy); and Bliss < 0 indicates the percentage increase in maximal inhibition below additivity (antagonism).