Parthenolide induces apoptosis via TNFRSF10B and PMAIP1 pathways in human lung cancer cells

Parthenolide and PMAIP1 antibody were purchased from Calbiochem (Darmstadt, Germany). Briefly, parthenolide was dissolved in dimethyl sulfoxide (DMSO) at a concentration of 10 mmol/L, and the aliquots were stored at -20°C. Stock solutions were diluted to the desired concentrations with growth medium before use. The antibodies of TNFRSF10B and ACTB were purchased from Sigma-Aldrich (St. Louis, MO, USA). CDH1 and CFLAR antibodies were obtained from BD Biosciences (San Jose, CA, USA) and Alexis (San Diego, CA) respectively. Anti-CASP8, CASP9, HSPA5, MCL1, p-EIF2A, and PARP1 antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). CASP3 anti-body was obtained from Imgenex (San Diego, CA, USA). Antibodies of ATF4, DDIT3 were obtained from Santa Cruz (Santa Cruz, CA).

Human lung cancer cell lines were obtained from the American Type Culture Collection (Manassas, VA). Cells were gown in monolayer culture with RPMI 1640 medium containing 5% new born calf serum at 37°C in a humidified atmosphere consisting of 5% CO2 and 95% air. The A549/Ctrl, A549/CFLAR, H157/Ctrl, H157/CFLAR, A549/shCtrl and A549/shCDH1 stable cell lines are established earlier by infection with lentiviral production [31, 32].

Cells were seeded in 96-well plates and treated on the second day with the given concentration of PTL for another 48 hours and then subjected to SRB or MTT assay. For SRB assay, live cell number was estimated as described earlier [33]. After treatment, the medium was discarded firstly. In order to fix the adherent cells, 100 μ1 of cold trichloroacetic acid (10% (w/v)) were adding to each well and incubating at 4°C for at least 1 hour. The plates were then washed five times with deionized water and dried in the air. Each well were then added with 50 μ1 of SRB solution (0.4% w/v in 1% acetic acid) and incubated for 5 min at room temperature. The plates were washed five times with 1% acetic acid to remove unbound SRB and then air dried. The residual bound SRB was solubilized with 100 μ1 of 10 mM Tris base buffer (pH 10.5), and then read using a microtiter plate reader at 495 nm. The MTT assay was executed following the manufacturer’s protocol of Cell Proliferation Kit I (Roche Applied Science, Brandford, CT, USA). 20 μl MTT (5 mg/ml) were added to each sample and incubate at 37° for 4 h, then 100 μl solubilization solution were added. Cell viability was determined at 595 nm.

Cell cycle was evaluated by DNA flow cytometry analysis. Cells were treated with different concentrations of PTL (0, 5, 10, 20 μM) for 24hours. After treatment, the cells were harvested and washed twice with ice PBS, then fixed in 70% ethanol at -20°C overnight. Before analysis, cells were washed again with ice PBS, incubated with PI (100 μg/ml) and RNase (50 μg/ml) in the dark for 30 min. Then samples were analyzed by FACScan flow cytometer (Becton Dickinson, San Jose, CA) [34].

Whole cell protein lysates were prepared and analyzed by Western blot according to the protocol described previously [35]. Cells were harvested and rinsed with pro-cold PBS. Then cell extracts were lysed and centrifuged at 4°C for 15 minutes. Whole cell protein lysates (40 μg) were electrophoresed through 12% denaturing polyacrylamide slab gels and then transferred to a Hybond enhanced chemiluminescence (ECL) membrane by electroblotting. The proteins were probed with the appropriate primary antibodies and subsequently with secondary antibodies. The antibody binding was detected by the ECL system (Millipore, Billerica, MA, USA), according to the manufacturer’s protocol.

siRNAs targeting sequences of TNFRSF10B, ATF4 and DDIT3 have been described previously and synthesized by GenePharma (Shanghai, China) [36]. The target sequence of PMAIP1 is 5′-GGAAGUCGAGUGUGCUACU-3′. The transfection of siRNA was following the manufacturer’s protocol of X-tremeGENE Transfection Reagent (Roche Molecular Biochemicals, Mannheim, Germany). Cells were seeded in 6-well plates and transfected with control or target siRNA on the second day. Cells were treated with indicated concentration of PTL for another 24 hours and harvested for Western blot analysis or Annexin V assay.

Apoptosis was evaluated using Annexin V-FITC/PI apoptosis detection kit purchased from BIO-BOX Biotech (Nanjing, China) following the manufacturer’s instructions. Briefly, 2×10⁶cells were harvested and washed twice with pre-cold PBS and then resuspended in 500 μl binding buffer. 5 μl of annexin V-FITC and 5 μl of Propidium Iodide (PI) were added to each sample and then incubated at room temperature in dark for 10 minutes. Analysis was performed by FACScan flow cytometer (Becton Dickinson, San Jose, CA).

It has been reported that parthenolide has antitumor effects on various cancer cells. Hence, we examined the inhibition effect of PTL on human NSCLC cells by treating the cells with various concentrations for 48 h and then conducting SRB and MTT assay. As is shown, PTL had a dose-dependent growth inhibition effect on NSCLC cells Calu-1, H1792, A549, H1299, H157, and H460 (Figure 1A, B). To characterize the mechanism by which PTL induces growth inhibition in human NSCLC cells, we first determined the effect of PTL on induction of apoptosis by western blot analysis. The data showed that PTL could induce cleavage of apoptotic proteins such as CASP8, CASP9, CASP3 and PARP1 both in concentration- and time-dependent manner in tested lung cancer cells, indicating that apoptosis was trigged after PTL exposure (Figure 1C, D). In addition to induction of apoptosis, PTL also induced G₀/ G₁ cell cycle arrest in a concentration- dependent manner in A549 cells and G₂/M cell cycle arrest in H1792 cells (Additional file 1: Figure S1). The difference in cell cycle arrest induced in these two cell lines may be due to the p53 status [37, 38]. Collectively, these results show that PTL inhibits the growth of human lung cancer cells through induction of apoptosis and/or cell-cycle arrest.

In order to understand the molecular mechanism of PTL-induced apoptosis in NSCLC cell lines, several apoptosis-related proteins were examined. Data showed that TNFRSF10B was up-regulated after exposure to PTL (Figure 2A, B). After TNFRSF10B expression was knocked down using siRNA method, the cleavage of CASP8, CASP9, CASP3 and PARP1 induced by PTL treatment were receded compared with control siRNA knockdown (Figure 2C). The analysis of Annexin V staining showed that apoptosis was inhibited when TNFRSF10B was knocked down (Figure 2D, E). It can be concluded that PTL up-regulates TNFRSF10B and contributes to apoptosis induction in lung cancer cells.

Since CFLAR is an important modulator of extrinsic apoptotic pathway, we also detected the levels of CFLAR and found that both CFLAR_L (Long form) and CFLAR_S (Short form) were down-regulated in a concentration- and time-dependent manner after PTL treatment (Figure 3A, B). Compared with control cells, cleavage of pro-caspases and PARP1 were weaker in A549/CFLAR_L cells which over-expressing CFLAR_L (Figure 3C). Annexin V staining showed PTL induced less apoptosis in A549/CFLAR_L cells than that in control cells (Figure 3D). We got same results in H157/CFLAR_L cells (Figure 3C, E). This implicated that CFLAR_L could prevent human lung cancer cells from apoptosis induced by PTL treatment. Therefore, we can summarize that TNFRSF10B and CFLAR_L are involved in PTL-induced extrinsic apoptosis.

We wonder if PTL could also activate intrinsic apoptotic pathway in lung cancer cells. Since PMAIP1 and MCL1 are both important proteins in intrinsic signaling pathway, we detected their expression after PTL treatment. Western blot analysis revealed that MCL1 was decreased in both concentration- and time-dependent manners after PTL exposure, while PMAIP1 was up-regulated (Figure 4A, B). Gene silencing experiment presented that when PMAIP1 was knocked down, the expression of MCL1 was partially increased and the cleavage of pro-caspases and PARP1 induced by PTL were reduced (Figure 4C). Annexin V staining analysis showed that apoptosis induced by PTL was weakened after knocking down of PMAIP1 (Figure 4D, E). It could be concluded that the intrinsic apoptosis process induced by PTL is through PMAIP1 and MCL1 axis.

DDIT3, which is a target protein of ATF4, is reported to regulate the expression of TNFRSF10B and PMAIP1 by binding to their promoter sites [27]. Therefore, we wonder if PTL induces TNFRSF10B and PMAIP1 through ATF4-DDIT3 axis. We examined expression of ATF4 and DDIT3 after PTL treatment. Western blot revealed that PTL could up-regulate ATF4 and DDIT3 in both concentration- and time-dependent manner (Figure 5A, B). When ATF4 was knocked down, DDIT3 was decreased, and activation of pro-caspases was weakened at the same time compared with control knockdown cells (Figure 5C). In addition, apoptosis was suppressed when DDIT3 was knocked down, while the expression of TNFRSF10B and PMAIP1 were decreased simultaneously (Figure 5D). Since ATF4 and DDIT3 are important hallmarks involved in ER stress pathway, we examined the expression of other molecules in ER stress signaling such as ERN1, HSPA5 and p-EIF2A as well [39]. We found that they were both increased after PTL treatment (Figure 6A, B). All these data indicated that PTL induces apoptosis through activation of ER stress response.

Weinberg et al. has demonstrated that knocking down of CDH1/E-cadherin with shRNA could make the cells have stem-like properties [40]. We had demonstrated that A549/shCDH1 cells in which CDH1/E-cadherin expression is inhibited had stronger capacity of proliferation, migration and invasiveness [32]. Furthermore, we found that the expression of SOX2 and POU5F1 which were considered to be the makers of stem cells were up-regulated in A549/shCDH1 cells (Additional file 1: Figure S2) [41, 42]. So in order to determine why PTL could selectively eradicate cancer stem-like cells, A549/shCDH1 cell line was used to mimic cancer stem cells and the A549/shCtrl cell line served as control. SRB assay showed PTL was more effective in inhibiting the growth of A549/shCDH1 cells than that of A549/shCtrl cells (Figure 7A). Western blot data showed that PTL could induce stronger cleavage of pro-caspases and PARP1 in A549/shCDH1 cell line (Figure 7B), which means that PTL could trigger stronger apoptosis in A549/shCDH1 cells compared with control cells. Furthermore, apoptosis-related proteins were detected in A549/shCtrl and A549/shCDH1 cells side by side. Both long form and short form of CFLAR levels were down-regulated even more clearly in A549/shCDH1 cells than that in control cells after PTL treatment. We also found that MCL1 was reduced more dramatically in A549/shCDH1 cells, while PMAIP1 was up-regulated on contrary after PTL treatment compared with the control cells (Figure 7C). Taken together, we conclude that both extrinsic apoptosis and intrinsic apoptosis induced by PTL are enhanced in A549/shCDH1 cells. The levels of p-EIF2A, ATF4 and DDIT3 were also examined. Data showed that their expression was further up-regulated in A549/shCDH1 cells after PTL treatment compared with A549/shCtrl cells (Figure 7C). DDIT3 was knocked down in the two cell lines simultaneously, and PMAIP1 was down-regulated and apoptosis was receded (Figure 7D). Therefore, we propose that the reason why PTL has a selective effect towards A549/shCDH1 cells is because PTL somehow triggers much more intensive ER stress response in cancer stem-like cells and further enhances the expression of ATF4 and DDIT3, leading to up-regulation of PMAIP1 and eventually, the induction of apoptosis.