p53-Dependent p21-mediated growth arrest pre-empts and protects HCT116 cells from PUMA-mediated apoptosis induced by EGCG

Abstract

The tumor suppressor protein p53 plays a key role in regulation of negative cellular growth in response to EGCG. To further explore the role of p53 signaling and elucidate the molecular mechanism, we employed colon cancer HCT116 cell line and its derivatives in which a specific transcriptional target of p53 is knocked down by homologous recombination. Cells expressing p53 and p21 accumulate in G1 upon treatment with EGCG. In contrast, same cells lacking p21 traverse through the cell cycle and eventually undergo apoptosis as revealed by TUNEL staining. Treatment with EGCG leads to induction of p53, p21 and PUMA in p21 wild-type, and p53 and PUMA in p21(-/-) cells. Ablation of p53 by RNAi protects p21(-/-) cells, thus indicating a p53-dependent apoptosis by EGCG. Furthermore, analysis of cells lacking PUMA or Bax with or without p21 but with p53 reveals that all the cells expressing p53 and p21 survived after EGCG treatment. More interestingly, cells lacking both PUMA and p21 survived ECGC treatment whereas those lacking p21 and Bax did not. Taken together, our results present a novel concept wherein p21-dependent growth arrest pre-empts and protects cells from otherwise, in its absence, apoptosis which is mediated by activation of pro-apoptotic protein PUMA. Furthermore, we find that p53-dependent activation of PUMA in response to EGCG directly leads to apoptosis with out requiring Bax as is the case in response to agents that induce DNA damage. p21, thus can be used as a molecular switch for therapeutic intervention of colon cancer.

Figure 1. EGCG selectively induces apoptosis of HCT116 cells lacking p21

(A) WT and p21^−/− cells were treated with 100 µM EGCG for 96 h and stained with methylene blue. The intensity of methylene blue take-up by live cells was measured calorimetrically after eluting the dye in 0.1N HCl and compared with untreated cells. (B) Cells were treated with 100 µM EGCG for 96 h, harvested, treated with 1% para-formaldehyde for 1h, fixed in 70% ethanol and stained for TUNEL positive cells. (C) Cell treated with 100 µM EGCG for 24, 48 and 72 h were harvested, stained with PI and analyzed by flow cytometry to measure Sub-G1 and G1 population. (D) Cells were treated with 100 µM EGCG for 48 and 96 h. Total cell lysates were subjected to Western blotting for p53, p21, and PUMA. All the data are representative of three independent experiments with highly reproducible results.

Figure 2. p53 dependent expression of p21 or PUMA determine the cellular response to EGCG

(A) Total cell lysate from Wild type and p21−/− cells transduced with shGFP or shp53 RNA vector were used for Western blot analysis for p53. (B and C) Total cell lysates from WT or p21−/− cells transduced with shp53 or GFP and treated with 100 µM EGCG for 48 and 96 h were subjected to Western blotting for p21 or PUMA respectively. (D and E) WT or p21−/− HCT116 cells with shp53 or shGFP were harvested after 96 h of 100 µM EGCG treatment and subjected to cell cycle analysis by flow cytometry, respectively. Reproducibility of the data were confirmed by at least three independent experiments.

Figure 3. p21-mediated cell cycle arrest predominates and prevent cells from PUMA-dependent, Bax-independent apoptosis

(A) Cells were treated with 100 µM EGCG for 96 h and stained with methylene blue. (B) Cells were treated with 100 µM EGCG for 96 h, harvested, treated with 1% paraformaldehyde for 1 h, fixed in 70% ethanol and stained for TUNEL positive cells.

Figure 4. EGCG does not cause DNA damage or ROS generation

(A) ROS generation in cells treated with 100 µM EGCG after 24 h. Camptothecin and nutlin were used as positive and negative controls, respectively. Treated (unshaded) are superimposed on untreated (shaded) samples. (B) DNA damage was measured after 24 h of 100 µM of EGCG treatment, using Oxy DNA assay kit (Calbiochem, San Diego, CA) as described in the methods. Camptothecin and nutlin were used as positive and negative controls, respectively, for DNA damage. Reproducibility of the data were confirmed by multiple independent experiments.

Figure 5. Proposed working model

(A) EGCG treatment leads to induction of p53 followed by induction of p21 and PUMA. Induction of p21 causes cell cycle arrest thereby protecting cells from EGCG induced apoptosis, however in the absence of p21 cells continue in cell cycle and undergo PUMA mediated apoptosis. (B) DNA damaging agents require both PUMA and Bax for apoptosis. Unlike apoptosis caused by DNA damaging agents, apoptosis caused by EGCG does not require Bax, and PUMA is sufficient to cause apoptosis.