Figure 3

PDK4 inhibition promotes EMT. (A) HCC827 cells were transfected with pCMV6-AC-IRES-GFP vector or pCMV6-PDK4-IRES-GFP (WT, 254A, or 257AA), then cultured in the presence of 1.5 μg/ml puromycin for 3 weeks. After that, the GFP-positive cells were enriched through FACS, cultured in the presence or absence of 2 ng/ml TGFβ for 10 days, and lysed for immunoblotting. (B) A549 and HCC827 cells were transfected with siNTC pool no. 2 or siPDK4 pool at 1 day and 3 days post-seeding. Two days after the second transfection, the cells were lysed for immunoblotting and qRT-PCR. VDAC and GAPDH are protein loading controls for mitochondrial lysate and whole cell lysate, respectively. In A and B, the red asterisk denotes the upper band that is specific for PDK4. (C, D) HCC827 (C) and HCC4006 (D) cells were transfected with siNTC pool no. 2 or siPDK4 pool at 20 nM siRNA at 1 day and 3 days post-seeding. Twenty-four hours after the second transfection, the media was replaced with growth media containing 2 μM erlotinib and cells were continuously cultured in such media with fresh media replenished every 3 to 4 days for 2 to 3 weeks. For the control plates without erlotinib treatment, the day before the second transfection, a fraction of the cells was re-plated at low density and subjected to a second transfection the next day. At the end of the experiment, cells were stained with crystal violet. The quantification of colony numbers represents four independent experiments, and for each experiment, the colony number in the siNTC plate is normalized to that in the siPDK4 plate. Paired t-test was performed for C and D. Data are plotted as mean ± SEM. *p < 0.05; **p < 0.01.