Inhibition of the p38 kinase suppresses the proliferation of human ER-negative breast cancer cells

Abstract

p38 kinases are members of the mitogen-activated protein kinase family that transduce signals from various environmental stresses, growth factors, and steroid hormones. p38 is highly expressed in aggressive and invasive breast cancers. Increased levels of activated p38 are markers of poor prognosis. In this study, we tested the hypothesis that blockade of p38 signaling would inhibit breast cancer cell proliferation. We studied breast cancer cell proliferation and cell cycle regulation upon p38 blockade by using three independent approaches: dominant-negative (DN) constructs, small interfering RNA (siRNA), and small molecule inhibitors. p38alpha and p38delta are the most abundant isoforms expressed by all examined human breast tumors and breast cancer cell lines. Expression of a DN p38 inhibited both anchorage-dependent and -independent proliferation of MDA-MB-468 cells. Silencing of p38alpha, but not p38delta, using siRNA suppressed MDA-MB-468 cell proliferation. Pharmacologic inhibitors of p38 significantly inhibited the proliferation of p53 mutant and ER-negative breast cancer cells. Whereas p38 has previously been considered as a mediator of stress-induced apoptosis, we propose that p38 may have dual activities regulating survival and proliferation depending on the expression of p53. Our data suggest that p38 mediates the proliferation signal in breast cancer cells expressing mutant but not wild-type p53. Because most ER-negative breast tumors express mutant p53, our results provide the foundation for future development of p38 inhibitors to target p38 for the treatment of p53 mutant and ER-negative breast cancers.

Figure 1. Expression of p38 in human breast tumors and cancer cell lines

(A) mRNA levels of p38α, β, γ, and δ in 37 human breast tumors were measured by Q-RT-PCR. Genomic equivalent copies for each p38 isoform were normalized to cyclophilin. Statistical analysis was performed using Friedman chi-square test and pair wise Wilcoxon signed-rank test and found that p38α is the most abundant isoform (P < 0.0001). (B) mRNA levels of p38α, β, γ, and δ in a panel of human breast cancer cells measured by Q-RT-PCR. Genomic equivalent copies for each p38 isoform were normalized to cyclophilin (n = 3). (C) Total p38 protein levels in the cell lines were measured by Western blots. The level of total p38 protein was normalized to that of β-actin (n = 3).

Figure 2. Expression of dominant-negative p38 inhibited MDA-MB-468 cell growth both in anchorage-dependent and independent manner

(A) Stable pools were cultured in serum free IMEM for 24 h and treated with 50ng/ml anisomycin for 15 min. The levels of total p38, total MAPKAPK-2, Flag and β-actin were measured by Western blot. (B) To measure growth, the stable pools of cells in A were plated in 6-well plates at 2 × 10⁴ cells per well. Cell proliferation was measured by counting cells (n = 3) every 24 h. Analysis was performed using a mixed linear model and P values were as follows for MDA-MB-468 (P < 0.0001), MDA-MB-231 (P = 0.0007), and MCF-7 cells (P =0.1809). (C) Clones of MDA-MB-468/pcDNA3 (#1, 2, and 4) and MDA-MB-468/pcDNA3-DNp38 (#9, 25, and 29) were plated maintained in serum free IMEM for 48 h. Medium was then replaced with full growth medium IMEM containing 10% heat inactivated FBS. Cells were harvested at either 0 (−) or 10 min (+) after serum stimulation. The levels of total p38, MAPKAPK-2, and β-actin were measured by Western blot. The arrow indicates the larger Flag-tagged p38 protein. (D) The above pcDNA3 clones were plated in 6-well plates at 2 × 10⁴ cells per well to measure anchorage-dependent growth. Cell proliferation was measured daily by counting cells (n = 3). Analysis was performed using a mixed linear model (P = 0.0005). The clones were also plated in soft agar in 60mm tissue culture dishes at 5 × 10³cells per dish. Anchorage independent growth was measured by counting colonies formed in soft agar 4 weeks after plating (n = 3). Analysis was performed using a mixed linear model (P <0.0001).

Figure 3. Silencing p38α gene with siRNA inhibited MDA-MB-468 cell growth

(A) Q-RT-PCR was used to measure p38α and p38δ mRNA levels 2 days after transfection. Relative p38α and p38δ mRNA abundance was normalized to that of cyclophilin (n = 3). (B) 4 days after transfection, cells were treated with DMSO (-) or 50ng/ml anisomycin for 15 min. Levels of total p38, total MAPKAPK-2, and β-actin were measured by Western blot. (C) 2 × 10³ transfected cells were plated in 96-well plates. Proliferation was measured by MTS assay every 24 h (n = 4). p38α significantly inhibited the growth of MDA-MB-468 cells (P = 0.0463) while p38δ had no significant effect (P = 0.3318).

Figure 4. siRNA against p53 and Wip1 causes addiction to p38 signaling

(A) Q-RT-PCR was used to measure P53 mRNA levels 2 days after transfection of MCF-7 cells and normalized to cyclophilin (n=3). MCF-7 cells transfected with P53 siRNA and treated with increasing doses of AZ10164773 show sensitivity to the p38 inhibitor compared to non-specific siRNA knockdown when analyzed using a mixed linear model (P < 0.0001). (B) Q-RT-PCR was also used to measure Wip1 mRNA as described above (n=3) and similar results were found when analyzed using a mixed linear model (P < 0.0001).

Figure 5. Proposed model for p38 signaling in p53 wild-type or mutant cells

p38 is activated by various extracellular stimuli. In cells with p53^WT (A), p38 phosphorylates and activates p53, leading to p53-dependent transcription and apoptosis. p53^WT is required for the induction of Wip1, which can dephosphorylate and inactivate p38. Thus, there exists a negative feedback regulation among p38, p53^WT, and Wip1. p38 can also regulate ATF-2, MAPKAPK-2 and other targets that are involved in regulating cell cycle and cell proliferation. However, in the context of p53^WT, the predominant effect of p38 signaling may be regulating p53-dependent apoptosis rather than cell proliferation. On the other hand, when cells have p53^MUT (B), they cannot undergo p53-dependent apoptosis and they lose the negative feedback regulation involving Wip1. Signaling through p38 continues to activate proliferative pathways and stimulates cell cycle progression and cell proliferation through mediators such as ATF-2 and MAPKAPK-2. In such p53^MUT cells, blockade of p38 signaling suppresses cancer cell growth. (PIGs: p53-induced genes, ds genes: downstream genes).