Inhibition of Notch and PEA3 inhibits anchorage-dependent growth in vitro and tumor growth in vivo. (A) and (B) MDA-MB-231 cells were transfected with either scrambled or PEA3 siRNAa and treated with vehicle (DMSO) or γ-secretase inhibitor (MRK-003 GSI) independently or in combination for 48 hours. A colony formation assay was performed using methylcellulose, and the colonies were incubated for 14 days. (A) Colonies were photographed using a standard light microscope (×40 original magnification). The photomicrographs are representative of three independent experiments. (B) Twelve fields per well were counted, and the means plus or minus standard deviations of three independent experiments were plotted. Statistical significance was determined by performing analysis of variance for multiple comparisons. (C) Cells were injected into each of two mammary fat pads of female athymic nude mice which were number-tagged prior to surgery to monitor each tumor. The mice were randomized to vehicle or MRK-003 GSI, which was fed orally by gavage on a schedule of three days on, four days off. Tumor areas (length × width) were measured twice per week using vernier calipers for up to 3.5 weeks. The y-axis is cross-sectional in centimeters squared, which is calculated using the formula [(area × π)/4]. The x-axis is weeks of treatment. The graph shows the mean cross-section of the tumors plus or minus the standard deviations of 20 tumors/10 mice/group. Linear regression analysis was performed on individual tumors to calculate the slope of a line with a correlation coefficient >0.85. Statistical analysis was performed using a two-tailed, unpaired Student's t-test on the slopes of each line. *Denotes statistical differences between vehicle + PEA3i and vehicle + Controli. **Denotes statistical differences between MRK-003 GSI + Controli and vehicle + Controli. (D) MDA-MB-231 cells were transfected with control or PEA3 siRNAa for 24 hours. Real-time PCR was performed on MDA-MB-231 cells prior to injection to detect PEA3 transcripts.