Molecular Insights Into Withaferin-A-Induced Senescence: Bioinformatics and Experimental Evidence to the Role of NFκB and CARF

Molecular Insights Into Withaferin-A-Induced Senescence: Bioinformatics and Experimental Evidence... Abstract Withaferin-A (Wi-A) has been shown to possess anticancer activity. Molecular mechanism(s) of its action has not been fully resolved. We recruited low dose of Wi-A that caused slow growth arrest in cancer cells and was relatively safe for normal cells. Consistently, we detected nuclear translocation of nuclear factor kappa B (NFκB) and activation of p38MAPK selectively in cancer cells. Bioinformatics analyses revealed that Wi-A did not disrupt IKKα/IKKβ–Nemo complex that regulates NFκB activity. However, it caused moderate change in the conformation of IKKβ–Nemo interacting domain. Experimental data revealed increased level of phosphorylated IκBα in Wi-A-treated cells, suggesting an activation of IKK complex that was supported by nuclear translocation of NFκB. Molecular docking analysis showed that Wi-A did not disrupt; however, decreased the stability of the NFκB–DNA complex. It was supported by downregulation of DNA-binding and transcriptional activities of NFκB. Further analysis revealed that Wi-A caused upregulation of CARF (collaborator of ARF) demonstrating an activation of DNA damage oxidative stress response in both cancer and normal cells. In line with this, upregulation of p21WAF1, p16INK4A, and hypophosphorylated pRB and induction of senescence were observed demonstrating that Wi-A-induced senescence is mediated by multiple pathways in which CARF-mediated DNA damage and oxidative stress play a major role. Withaferin-A, Cancer, Cellular senescence, p53–p21 pathway, NFκB, CARF Nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) family of proteins is a class of transcription factors that possess a Rel homology domain in their N-terminus and control expression of genes involved in regulation of growth, apoptosis, immunoregulatory, and inflammatory processes (1,2). It regulates cellular response to oxidative stress, cytokines, bacterial, and viral antigens and plays a key role in regulating the immune response to infection (3). NFκB1 and NFκB2 are transcription factors synthesized as large precursors, p105, and p100 that are cleaved to mature NFκB p65 and p50 proteins by selective degradation of their C-terminal region containing ankyrin repeats (4). These are present in cells in an inactive (sequestered in cytoplasm) state by binding to their inhibitors called IκBs (inhibitor of NFκB). Intracellular/extracellular stresses or cytokine signals activate its upstream regulator, IKK (Inhibitor of IkB Kinase), which phosphorylate IkBα at Ser32 and Ser36 residues. The phosphorylated IκBα undergoes proteasomal degradation resulting in active NFκB that translocates to the nucleus and performs transcriptional activation function (5). The NFκB proteins lack intrinsic transcriptional activation ability and function by binding as homodimers (6). Its crucial role in processes such as proliferation, apoptosis, and invasiveness requires controlled activation (7). Dysregulated activation has been shown to be associated with various conditions such as arthritis, asthma, and inflammatory disorders (1,8). Aberrant or constitutive activation of NFκB has been observed in various types of cancer cells marking it as a potential therapeutic target (1,2,9). In line with this, identification and functional characterization of natural and synthetic inhibitors of NFκB have been initiated. Inhibition of NFκB by small molecules, β-mangostin (βM) isolated from Cratoxylum arborescens (10) and Wi-A (withaferin-A) isolated from medicinal herb Withania somnifera (Ashwagandha) (8,9,11,12), has been reported. It was shown that Wi-A inhibits DNA binding of NFκB resulting in reactive oxygen species generation, mitochondrial dysfunction, oxidative stress, and apoptosis in cancer cells (11–14). DNA damage response (DDR) is differentially regulated in cancer and normal cells. Whereas normal cells respond by executing growth arrest, cancer cells are refractory and keep proliferating (15). Several studies have established that DDR signaling is essential for execution of senescence, an established powerful tumor suppressor mechanism (16,17). The p38MAPK is a kinase that is activated through environmental stress and DNA damage stress, including ionizing radiation, ultraviolet, chemotherapeutic drugs, and lead to the induction of a G2/M cell cycle checkpoint through p53-dependent and -independent mechanisms (18,19). p38MAPK has been shown to transcriptionally activate NFκB leading to senescence (20–22), mediated by upregulation of p53-p21WAF1 pathway. CARF (collaborator of ARF), an ARF (Alternative Reading Frame, p14ARF)-interacting protein, has been shown to regulate activities of p53-tumor suppressor protein in an ARF-dependent or -independent manner (23–25). It has been shown to regulate DDR in a dose-dependent manner and regulates cell proliferative fates in normal and cancer cells (25,26). Wi-A, a withanolide extracted from medicinal herb W. somnifera, has been shown to cause inhibition of IKK activity in some earlier studies (8,12,27). Through bioinformatics analysis, we had previously shown that Wi-A disrupts important hydrophobic interaction between IKKβ and Nemo chain residues, L93:F734, T735:F92, F734:M94, W739:F97, W741:A100, W741:R101, thereby inhibiting their complex (28). However, an experimental study failed to reveal the inhibition of IKKβ–Nemo interaction by Wi-A; instead, it inhibited IκB kinase activity by interacting with C179 residue located in catalytic domain of IKKβ (8). It was also demonstrated that Wi-A hyperphosphorylates IKKβ at S181 causing inhibition of TNF-induced IKK activity and, thereby, causing inhibition of IκB degradation and p65 translocation (12). These reports described three different binding sites of Wi-A on IKK complex suggesting its interaction at multiple sites. Inhibition of IKK activity by Wi-A and/or NFκB translocation to the nucleus has been reported in some studies (8,12,27). However, the molecular insights and overall impact of interactions of Wi-A with IKKβ or NFκB structure and activity remain undefined. We hypothesized that the Wi-A might affect the conformation of the binding complex and hence set out to determine such impact on IKKβ–Nemo interaction domain and NFκB activity by molecular docking and dynamic simulations. We found that Wi-A caused significant change in conformation of IKKβ–Nemo and NFκB–DNA interaction domains. We provide experimental evidence of activation of kinase activity of IKK complex resulting in phosphorylation of IκBα, its degradation, and nuclear translocation of NFκB. Wi-A was further seen as not disrupting, but destabilizing NFκB–DNA interactions and was supported by the decreased level of expression of cyclin D1, cyclin E, and CDK2/4. We provide evidence that Wi-A triggers DDR, as supported by upregulation of γH2AX and CARF that yielded senescence in both cancer and normal cells. Material and Methods Cell Culture and Antibodies Human lung carcinoma (H1299 and A549) and normal fibroblasts (MRC5) were cultured in RPMI-1640 and Dulbecco’s modified Eagle’s medium, respectively, supplemented with 10% (vol/vol) fetal bovine serum in 5% CO2 and 95% air humidified incubator. The antibodies were purchased from Santa Cruz Biotech Inc., CA (NFκB-p65, p-IκBα p53, p38MAPK, p16INK4A, CDK2, cyclin E, cyclin D1, and CDK4), Cell Signaling, Beverly, MA (HP1γ, p-p38MAPK, pRB [S780], and γH2AX, p21WAF1), and Abcam, Cambridge, UK (β-actin). Anti-CARF antibodies were generated in our laboratory. Cell Viability Assays Short-term cell viability (2–3 days) and long-term colony forming assays (5–12 days) were performed in 96- and 6-well tissue culture plates, respectively, as described earlier (25,26). Western Blot Analysis The cells were treated with Wi-A (0.2–2.0 µg/mL, as indicated) for 24 hours following which cell lysates were prepared in RIPA buffer (Thermo Scientific, Rockford, IL) containing protease inhibitor cocktail (Roche Applied Science, Mannheim, Germany). Control cells were treated with dimethyl sulfoxide. Cell lysates (containing 20-µg protein) were subjected to Western blot analysis with antibodies as indicated. The blots were then developed using chemiluminescence (GE Healthcare, UK) and visualized using Lumino Image Analyzer equipped with CCD camera (LAS3000-mini; Fuji Film, Tokyo, Japan) as described earlier (25,26). Nuclear and cytoplasmic fractions were prepared using the Qproteome cell compartment kit (Qiagen, Hilden, Germany). Immunostaining Cells (1 × 104), cultured on a glass coverslips placed in a 12-well plate, were treated with Wi-A and fixed with prechilled methanol:acetone (1:1) for 5 minutes at 4°C. Fixed cells were incubated with primary antibodies (as indicated) followed by extensively washing (0.2% Triton X-100 in phosphate-buffered saline) and incubation with the fluorochrome-conjugated secondary antibodies (Alexa-488-conjugated goat anti-rabbit or anti-mouse and Alexa-594-conjugated goat anti-rabbit or anti-mouse [Molecular Probes, OR]) for 45 minutes as described earlier (25,26). Stained cells were examined under LSM700 laser scanning confocal microscope from Carl Zeiss. The images were taken using AxioVision 4.6 software (Carl Zeiss Microimaging Inc., Jena, Germany) and IMARIS software (Bitplane, Zurich, Switzerland). Immunoprecipitation Cell lysates of control and Wi-A-treated cells (1-mg protein) in Nonidet (NP)-P40 lysis buffer were incubated with p65 rabbit polyclonal antibody and control IgG (Cell Signaling, Beverly, MA) for 3 hours in slow rotation at 4°C. Immunocomplex was precipitated by incubation with Protein-A/G plus Agarose (20 mL; Santa Cruz Biotech. Inc., CA, sc-2003) for 45 minutes followed by centrifugation at 2,500 rpm for 5 minutes at 4°C. Pellets were washed five times with NP-40 buffer with repeated centrifugation at 2,500 rpm for 5 minutes at 4°C. Immunocomplex was resolved on SDS/10%PAGE and electroblotted onto a polyvinylidene fluoride membrane. The proteins in immunocomplex were detected by Western blotting with the indicated monoclonal antibody and captured using ECL (GE Healthcare, UK). Cell Cycle Analysis Cells treated with indicated doses of Wi-A for 24 hours were collected in 1.5-mL tube, washed with cold phosphate-buffered saline, and fixed with 70% ethanol at 4°C for 12 hours. The fixed cells, centrifuged (2,000 rpm for 10 minutes), washed with cold phosphate-buffered saline, and resuspended in 0.25-mL phosphate-buffered saline, were stained with Guava Cell Cycle Reagent (Millipore, Tokyo, Japan) for 30 minutes in dark. To avoid false DNA-PI staining, RNA was removed by RNase A treatment (5 µL of 1 mg/mL at 37°C for 1 hour). Cell cycle analysis was performed using Guava PCA-96 System (Millipore, Tokyo, Japan) and CytoSoft TM Software, version 2.5.6 (Millipore, Tokyo, Japan). Apoptosis and Senescence Assays Cells (2 × 105 cells/well in six-well plates) were treated with Wi-A. Apoptosis was detected and analyzed by Guava Nexin Reagent (EMD Millipore Corporation, Tokyo, Japan) and Flow Jo Software, respectively. Senescent cells were detected using senescence-associated β-galactosidase kit (Cell Signaling Technology, Danvers, MA) by methods as recommended by the manufacturers. NFκB–DNA Binding Assay NFκB–DNA binding assay was performed using the NFκB–DNA binding Assay Kit (AB#133112, Abcam, Cambridge, MA) following the manufacturer’s instructions. Nuclear cell extracts of control and Wi-A-treated samples were prepared using Qproteome Cell Compartment Kit (#37502, Qiagen Inc., Manchester, UK) following the manufacturer’s protocol. Statistical Analysis All the experiments were performed in triplicate, and variables were expressed as mean ± SEM of triplicate experiments. Unpaired t test (GraphPad Prism, GraphPad Software, San Diego, CA) has been performed to determine the degree of significance between the control and experimental samples. Statistical significance was defined as p value and represented by *p < .05, **p < .01, ***p < .001, whereas no mark denotes insignificant correlation. Computational Methods Docking of Wi-A With IKKβ–Nemo Association Domain The structure of IKKβ–Nemo association domain, complex of two IKKβ kinase subunits and two regulatory subunits of NEMO (NFκB essential modulator), was obtained from Protein Data Bank (PDB ID: 3BRV). Flexible docking of Wi-A with IKKβ–Nemo complex was carried out by assigning flexibility to side chains of their key interacting residues (28). The structure of the protein and ligand molecule, Wi-A (PubChem ID: 265237), deposited in the databases was preprocessed, followed by generation of grid around the flexible residues and docking of Wi-A within the designed grid using AUTODOCK 4.2 software (29). Binding efficiency of another Nemo chain with IKKβ:Nemo complex was also predicted in the presence and absence of Wi-A using HADDOCK (High Ambiguity Driven protein–protein DOCKing) (30). Docking of Wi-A With NFκB Dimer DNA Binding Domain The structure of NFκB dimer that is p50–p65 subunit bound to DNA (PDB ID: 3GUT) was retrieved from Protein Data Bank. The DNA chain was removed from the complex, and protein structure was prepared using PrepWizard of maestro (31). The structure of Wi-A was preprocessed using LigPrep version 3.5 of Schrodinger suite. Glide extra precision algorithm was used to dock Wi-A around residues forming hydrogen bonds with the DNA molecule (31,32). Binding affinity of DNA with NFκB dimer both in the presence and absence of Wi-A was predicted by carrying out protein–DNA docking using HADDOCK. Molecular Dynamics Simulations Amber Molecular Dynamics Suite was used to perform all the simulations on DELL T3610 workstation with 16-GB DDR RAM and NVIDIA GeForce GTX TITAN Black Graphics Card (33). Amber protein force field, ff12SB, was used to perform simulations of all complexes solvated with TIP3P water octahedral box using a spacing distance of 10 Å around the molecule. Solvated molecule was then neutralized using appropriate number of counterions followed by minimization, heating up to 300 K temperature and equilibration of molecule and molecular dynamic simulations. Three sets of molecular dynamic simulations were performed to analyze the effect of Wi-A on complete IKKβ–Nemo association domain. First set of complexes, IKKβ:Nemo:Wi-A docked complex and IKKβ:Nemo complex, were simulated for time duration of 251 and 200 ns, respectively. The second set of simulation was performed for IKKβ:Nemo2:Wi-A complex and IKKβ:Nemo2 complex for 50 and 48 ns, respectively. The final set of molecular dynamic simulations was performed on complete IKKβ/Nemo association domain complexes in the presence and absence of Wi-A, namely, IKKβ2:Nemo2:Wi-A complex and IKKβ2:Nemo2 complex, which were simulated for the time period of 60 and 50 ns, respectively. The NFκB dimer:DNA complex and NFκB dimer:Wi-A docked complex were simulated for 40 and 50 ns, respectively, to check the effect of Wi-A on DNA binding activity of NFκB dimer. Coordinates of DNA were incorporated in NFκB dimer:Wi-A, and resulting NFκB dimer:DNA:Wi-A complex was simulated for a period of 50 ns to study the stability of Wi-A on DNA-bound complex of NFκB dimer. Analysis of Molecular Docking and Molecular Dynamics Simulations Root mean square deviation (RMSD) computation and conformational analysis over the range of simulation trajectories was performed using VMD version 1.9.2 (34). Superimposition of protein structures and generation of images were performed using the PyMol molecular graphics system (35). Protein–ligand interactions were studied using LigPlot + v.1.4.5 (36). Results Wi-A, at Low Dose, Triggers Growth Arrest in Human Lung Carcinoma Human lung cancer (H1299) cells were treated with serially increasing doses of Wi-A for 48 hours. As shown in Supplementary Figure 1A, cells showed serial increase in NFκB at doses 0.08–1.0 µg/mL followed by decrease at 1.6 and 2.0 µg/mL. We selected two doses of Wi-A (low, 0.2 µg/mL and high, 2.0 µg/mL) and performed cell proliferation assays. As shown in Supplementary Figure 1B, whereas high dose caused 50% reduction in viability within 48 hours, low dose caused only ~10% reduction followed by slow growth arrest. Annexin IV cytometric analysis revealed that the high, not low, dose instigated apoptosis in about 48 hours (Supplementary Figure 1C). In contrast to 5.12% in control, 8.71% and 34.59% cells were detected in apoptosis in low-dose- and high-dose-treated cultures, respectively (Figure 1C). Low-dose-induced slow growth arrest leads to reduction in colonigenicity in long-term colony forming assays (Supplementary Figure 1D), and most interestingly, we found that whereas low dose (24- to 48-hour treatment) induced G0/G1 cell cycle arrest in H1299 cells, the normal (MRC5) cells remained unaffected (Supplementary Figure 1E). Figure 1. View largeDownload slide Low dose of withaferin-A (Wi-A) caused nuclear translocation of nuclear factor kappa B (NFκB) in lung carcinoma. Immunostaining of NFκB, in control and Wi-A-treated cells, showed its nuclear translocation in cancer (H1299 and A549) cells (A and B). Normal cells (MRC5) did not show nuclear translocation of NFκB. Confocal images showing nuclear NFκB staining in cancer cells (B). Western blotting exhibited increase in NFκB expression in Wi-A-treated cells (C). Cell fractionation and Western blotting analyses confirmed the nuclear translocation of NFκB in Wi-A-treated cells (D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 1. View largeDownload slide Low dose of withaferin-A (Wi-A) caused nuclear translocation of nuclear factor kappa B (NFκB) in lung carcinoma. Immunostaining of NFκB, in control and Wi-A-treated cells, showed its nuclear translocation in cancer (H1299 and A549) cells (A and B). Normal cells (MRC5) did not show nuclear translocation of NFκB. Confocal images showing nuclear NFκB staining in cancer cells (B). Western blotting exhibited increase in NFκB expression in Wi-A-treated cells (C). Cell fractionation and Western blotting analyses confirmed the nuclear translocation of NFκB in Wi-A-treated cells (D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Low Dose of Wi-A Caused Nuclear Translocation of NFκB in Cancer Cells We next treated H1299 (cancer) and MRC5 (normal) cells with low dose of Wi-A (0.2–1.0 µg/mL) for 24–48 hours and examined the localization of NFκB in control and treated cells. As shown in Figure 1A, NFκB showed nuclear translocation only in cancer cells. The results were confirmed by confocal laser scanning microscopy (Figure 1B) and biochemical analyses including cell fraction and Western blotting with anti-NFκB, anti α-tubulin (cytoplasmic), and lamin A/C (nuclear) antibodies (Figure 1C and D). These data suggested the activation of NFκB in cancer cells and were in contrast to the earlier reports that have predicted inactivation of NFκB by Wi-A (8,12). Molecular Docking and Computational Inputs on Wi-A Targeting to IKKβ–Nemo Complex: Change in Conformation, Not Disruption, of the Complex We targeted IKKβ–Nemo association domain to unveil the effect of Wi-A on its conformation. Flexible docking of Wi-A was performed by generating grid around the key interacting residues in IKKβ and Nemo chain to check whether Wi-A can disrupt these interactions (37). IKKβ:Nemo:Wi-A docked complexes with minimum binding energy, −7.95 Kcal/mol, were selected. Wi-A formed two hydrogen bonds with R101 of Nemo chain as shown in Figure 2A. This lead to disruption of intermolecular interaction between Nemo and IKKβ chain residues R101 and D738, respectively, as previously reported in an X-ray crystallographic study of IKKβ–Nemo association domain (37). The complex was simulated for 251 ns to check the stability of Wi-A at the binding site. The analysis of snapshots of IKKβ:Nemo:Wi-A complex over the range of simulation showed that Wi-A was stably interacting with both chains with slight change in its position. Effect of such shift on ligand interaction at the binding site showed that Wi-A formed hydrophobic interactions with both IKKβ and Nemo chain residues, of which F97 and F734 were known to be originally involved in IKKβ–Nemo hydrophobic interactions (Figure 2B). The interaction site of Wi-A constituted amino acid residues ranging from 732 to 743 of IKKβ chain and residues 92–109 of Nemo chain. Furthermore, analysis of conformations of the complex over the simulation trajectories also showed the bending of Nemo chain at Wi-A interaction site. To confirm that the bending was induced by Wi-A, a 200-ns simulation of IKKβ:Nemo chain complex was performed. Comparison of two structures obtained after simulations confirmed that Wi-A does not inhibit interaction between IKKβ and Nemo chains, instead, caused bending of the Nemo chain (Figure 2C). To further confirm the inability of Wi-A to disrupt IKKβ–Nemo interaction, RMSD of IKKβ:Nemo complex and IKKβ:Nemo:Wi-A complex over the range of 200- and 251-ns simulation, respectively, was computed and compared. It was observed that the presence of Wi-A does not affect the RMSD trajectory of IKKβ:Nemo complex that endorsed the inability of Wi-A to disrupt IKKβ–Nemo interaction (Supplementary Figure 2A). Figure 2. View largeDownload slide Binding of withaferin-A (Wi-A) with IKKβ–Nemo complex and its effect on structural stability. Representation of hydrogen bond interactions of Wi-A with IKKβ–Nemo complex after molecular docking (A). Interactions formed by Wi-A with IKKβ–Nemo complex after 251-ns simulation. Wi-A is hydrophobically interacting with Phe97, Gly108, and Leu109 of Nemo chain and Gln732 and Phe734 of IKKβ chain (B). Superimposition of IKKβ–Nemo–Wi-A complex (purple) with IKKβ–Nemo complex taken as control (red) reveals the effect of Wi-A on confirmation of IKKβ–Nemo complex (C). Comparison of confirmation of IKKβ–Nemo2 complex generated after docking of Nemo Chain (pink) with IKKβ–Nemo complex (orange) in (D) the presence of Wi-A and (E) the absence of Wi-A. The comparison of the structures shows bending of both the Nemo chains toward Wi-A, whereas no such confirmation change was observed in the absence of Wi-A. Comparison of confirmation of complete IKKβ–Nemo association domain generated after inclusion of another IKKβ chain (pink) with IKKβ–Nemo2 complex (orange) in (F) the presence of Wi-A and (G) the absence of Wi-A. Figure 2. View largeDownload slide Binding of withaferin-A (Wi-A) with IKKβ–Nemo complex and its effect on structural stability. Representation of hydrogen bond interactions of Wi-A with IKKβ–Nemo complex after molecular docking (A). Interactions formed by Wi-A with IKKβ–Nemo complex after 251-ns simulation. Wi-A is hydrophobically interacting with Phe97, Gly108, and Leu109 of Nemo chain and Gln732 and Phe734 of IKKβ chain (B). Superimposition of IKKβ–Nemo–Wi-A complex (purple) with IKKβ–Nemo complex taken as control (red) reveals the effect of Wi-A on confirmation of IKKβ–Nemo complex (C). Comparison of confirmation of IKKβ–Nemo2 complex generated after docking of Nemo Chain (pink) with IKKβ–Nemo complex (orange) in (D) the presence of Wi-A and (E) the absence of Wi-A. The comparison of the structures shows bending of both the Nemo chains toward Wi-A, whereas no such confirmation change was observed in the absence of Wi-A. Comparison of confirmation of complete IKKβ–Nemo association domain generated after inclusion of another IKKβ chain (pink) with IKKβ–Nemo2 complex (orange) in (F) the presence of Wi-A and (G) the absence of Wi-A. To investigate the effect of Wi-A on IKKβ:Nemo dimer formation, binding affinity of Nemo chain to IKKβ:Nemo complex in the presence/absence of Wi-A was computed by docking another Nemo chain to both complexes around residues, 47–56, that are essential for Nemo–Nemo interaction (38). We observed minor reduction in binding affinity of Nemo chain to IKKβ:Nemo complex from −66.1 to −53.3 in the presence of Wi-A that were deemed not significant enough to inhibit the formation of IKKβ:Nemo dimer. Furthermore, both docked complexes, IKKβ:Nemo2:Wi-A and IKKβ:Nemo2, were also simulated for 50 and 48 ns, respectively. Comparison of IKKβ:Nemo2 structures in the presence/absence of Wi-A revealed that Wi-A caused bending of both Nemo chains in IKKβ:Nemo2:Wi-A complex, whereas no bending was observed in IKKβ:Nemo2 complex (Figure 2D and E). Wi-A Induced Change in Conformation of IKKβ–Nemo Interaction Domain The effect of Wi-A on conformation of complete IKKβ–Nemo interaction domain was examined by addition of coordinates of another IKKβ chain to IKKβ:Nemo2:Wi-A and IKKβ:Nemo2 complex. The resulting complexes, IKKβ2:Nemo2:Wi-A and IKKβ2:Nemo2, simulated for 60 and 50 ns, respectively. Both simulated structures were compared to unveil the effect of Wi-A on IKKβ–Nemo interaction domain, and change in conformation of IKKβ:Nemo association domain at Wi-A interaction site was observed in the presence of Wi-A (Figure 2F and G). Furthermore, Wi-A-induced instability of IKKβ:Nemo association domain structure at Wi-A interaction site was also supported by RMSD trajectories of Wi-A interaction site computed both in the presence/absence of Wi-A (Supplementary Figure 2B). We found that Wi-A interaction with IKKβ:Nemo association domain triggered instability of the structure at the interaction site, which was quite stable in the absence of Wi-A. To evaluate whether this instability is limited to the interaction site of Wi-A, RMSD of whole IKKβ:Nemo association domain excluding the Wi-A interaction site was also computed and compared (Supplementary Figure 2C). This comparison revealed that the presence of Wi-A has no effect on the stability of rest of the complex, but causes drastic changes in its interaction site that is evident in the form of bending of the Nemo chains. Wi-A Caused Nuclear Translocation of NFκB and p38MAPK To confirm the above computational findings, we performed experiments using human lung cancer (H1299 and A549) and normal (MRC5) cells. As shown in Figure 3A, immunostaining of control and Wi-A-treated H1299 cells with phosphorylation-specific anti-IκBα antibody revealed its increased level in cancer cells, but not in normal cells. These data endorsed that Wi-A did not inhibit the activity of IKKβ:Nemo complex and was in line with above computational analysis. Furthermore, Wi-A induced increase in phosphorylated IκBα, supporting the nuclear translocation of NFκB in cancer cells (Figure 1). These data were further supported by Western blotting and IκBα-p65 coimmunoprecipitation that revealed decrease (due to degradation of the phosphorylated form) in IκBα and its complex formation with p65 (Figure 3B). Several studies have shown that NFκB is regulated by p38MAPK and activated in response to various environmental stresses including DNA damage signaling, ultraviolet, ionization radiation, oxidative stress, and cytokines (39). Activation causes its phosphorylation, essential for its nuclear translocation (18), which in turn stimulates NFκB signaling (40). In view of these reports, we investigated whether Wi-A-induced activation of NFκB was mediated by p38MAPK. As shown in Figure 3C and D, we found nuclear p38MAPK in cancer cells. It was confirmed to be in activated phosphorylated form by Western blotting with phosphorylation-specific antibodies (Figure 3E). Normal cells did not show increase in p38MAPK (Figure 3C). These data suggested that Wi-A preferentially instigated DDR and activation of NFκB in cancer cells only. Figure 3. View largeDownload slide Upregulation of IκBα and p38MAPK in withaferin-A (Wi-A)-treated cells. Immunostaining of phosphorylated IκBα (A) in control and Wi-A-treated lung carcinoma and normal cells revealed their upregulation in cancer cells only. Western blotting for total IκBα in Wi-A-treated cells showing its decrease (B) coimmunoprecipitation (IP) of nuclear factor kappa B (NFκB) (p65) and IκBα showing decrease IκBα in IκBα-p65 complex (B). Increase in phosphorylated p38MAPK in Wi-A-treated cancer cells (C). Nuclear translocation of p38MAPK in Wi-A-treated cells as confirmed by confocal laser scanning microscopy (D) and Western blotting (E) with phosphorylation-specific antibody. Figure 3. View largeDownload slide Upregulation of IκBα and p38MAPK in withaferin-A (Wi-A)-treated cells. Immunostaining of phosphorylated IκBα (A) in control and Wi-A-treated lung carcinoma and normal cells revealed their upregulation in cancer cells only. Western blotting for total IκBα in Wi-A-treated cells showing its decrease (B) coimmunoprecipitation (IP) of nuclear factor kappa B (NFκB) (p65) and IκBα showing decrease IκBα in IκBα-p65 complex (B). Increase in phosphorylated p38MAPK in Wi-A-treated cancer cells (C). Nuclear translocation of p38MAPK in Wi-A-treated cells as confirmed by confocal laser scanning microscopy (D) and Western blotting (E) with phosphorylation-specific antibody. Effect of Wi-A on DNA Binding Activity of NFκB Dimer In view of the nuclear translocation and activation of NFκB in Wi-A-treated cells, we next examined the effect of Wi-A on DNA-binding ability of NFκB. DNA molecule was removed from X-ray determined structure of NFκB:DNA complex, and Wi-A was docked around amino acid residues R33, R35, Y36, E39, and Y57 of p65 subunit and R354, R356, Y357, E360, H364, and K541 of p50 subunit, which were involved in hydrogen bond interactions with DNA molecule. The docking analysis revealed that Wi-A interacted with NFκB dimer near DNA-binding site with binding energy of −4.129 kcal/mol. NFκB:Wi-A docked complex was simulated for 50 ns to check its stability and to observe changes in structure of the protein in the presence of the ligand. Molecular dynamic simulation revealed slight shift in the position of Wi-A. It, however, formed hydrogen bond with R354 and hydrophobic interactions with Y357, C359, and E360 among which R354, Y357, and E360 were originally involved in hydrogen bond interaction with DNA molecule (Figure 4A). To study the effect of Wi-A on NFκB dimer–DNA interactions, NFκB dimer:DNA:Wi-A complex was obtained by superimposition of docked complex over NFκB dimer:DNA complex. This complex was simulated for 50 ns, and the effect of DNA on NFκB–Wi-A interactions was examined. It was observed that Wi-A no longer formed any hydrogen bond interaction rather formed hydrophobic interactions with p50 subunit’s amino acid residues R354, Y357, C359, E360, S540, and K541, in addition to deoxycytidine of DNA molecule (Figure 4B). These data strongly suggested that Wi-A did not hinder the binding of DNA molecule with NFκB dimer. It, instead, was found to interact with the DNA and protein at their interaction site. Comparison of DNA binding in the presence and absence of Wi-A also revealed the differences in binding mode of DNA with protein dimer (Figure 4C). Figure 4. View largeDownload slide Binding of withaferin-A (Wi-A) with nuclear factor kappa B (NFκB) dimer (p65–p50) complex and its effect on structural stability. Interactions of Wi-A with NFκB dimer complex after molecular docking and simulation (A). Interactions of Wi-A with NFκB dimer–DNA complex (B). Superimposition of NFκB dimer–DNA complex in the presence (pink) and absence (blue) of Wi-A revealed the alterations caused by Wi-A in the confirmation of protein–DNA complex and the clash between DNA and Wi-A interaction site (C). Root mean square deviation (RMSD) fluctuations of NFκB dimer–DNA complex during 40- and 50-ns simulation of NFκB dimer–DNA complex (red) and NFκB dimer–DNA–Wi-A complex (purple), respectively (D). NFκB–DNA binding activity (E) and immunostaining and Western blotting for cyclin D1 and cyclin E in control and Wi-A-treated cancer and normal cells showed downregulation of cyclin D1 (F) and cyclin E cancer cells only (G and H). Normal cells showed downregulation of cyclin E, but not of cyclin D1. p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 4. View largeDownload slide Binding of withaferin-A (Wi-A) with nuclear factor kappa B (NFκB) dimer (p65–p50) complex and its effect on structural stability. Interactions of Wi-A with NFκB dimer complex after molecular docking and simulation (A). Interactions of Wi-A with NFκB dimer–DNA complex (B). Superimposition of NFκB dimer–DNA complex in the presence (pink) and absence (blue) of Wi-A revealed the alterations caused by Wi-A in the confirmation of protein–DNA complex and the clash between DNA and Wi-A interaction site (C). Root mean square deviation (RMSD) fluctuations of NFκB dimer–DNA complex during 40- and 50-ns simulation of NFκB dimer–DNA complex (red) and NFκB dimer–DNA–Wi-A complex (purple), respectively (D). NFκB–DNA binding activity (E) and immunostaining and Western blotting for cyclin D1 and cyclin E in control and Wi-A-treated cancer and normal cells showed downregulation of cyclin D1 (F) and cyclin E cancer cells only (G and H). Normal cells showed downregulation of cyclin E, but not of cyclin D1. p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. We next determined the binding affinity of NFκB dimer to DNA both in the presence and in the absence of Wi-A using HADDOCK. Residues of protein and DNA molecules involved in hydrogen bond interactions were treated as anchoring point for protein–DNA docking and specified as active residues, whereas passive residues were automatically selected by HADDOCK around active residues. Amino acid residues R33, R35, Y36, E39, R41, S42, K123, and R246 of p65 subunit and R354, R356, Y357, E360, K374, K444, K445, K572, K575, R605, and Q606 of p50 subunit were assigned as active residues for NFκB dimer:DNA complex. Although amino acid residues R50, T52, T55, T57, and R246 of p65 and C359, K444, K503, L507, S508, D539, K541, A545, R605, and R633 of p50 subunit were treated as active residues for NFκB dimer:DNA:Wi-A complex. Protein–DNA docking results revealed that the DNA binding affinity of NFκB dimer was −154.9 ± 2.9 that was reduced to −112.1 ± 4.7 in the presence of Wi-A. This reduction in binding affinity of DNA with NFκB dimer:Wi-A complex showed that Wi-A may result in decreased affinity of NFκB dimer. RMSD of NFκB dimer:DNA complex in the presence and absence of Wi-A was computed to decipher the effect of Wi-A on stability of NFκB dimer:DNA complex (Figure 4D). Comparison of RMSDs in the presence and absence of Wi-A revealed that after 10 ns, the presence of Wi-A added instability to NFκB dimer:DNA complex and increases fluctuations ranging from 2–4 to 6–8 Å. These data suggested that in spite of the above-described NFκB activation in Wi-A-treated cells, it may lead to downregulation of its transcriptional activation function. To examine such effect, we performed enzyme-linked immunosorbent assay (ELISA) for DNA–NFκB (p65) binding. As shown in Figure 4E, we found that Wi-A resulted in 20% decrease in the DNA binding activity of NFκB (p65). Consistently, we found decrease in NFκB effectors, cyclin D1 and cyclin E, in Wi-A-treated cancer cells (Figure 4F and G). Interestingly, cyclin E, but not cyclin D1, showed decrease in normal cells also (Figure 4F–H). Wi-A Caused Upregulation of DDR in Cancer and Normal Cells To further clarify, we examined the level of CARF expression, an established important regulator of DDR and proliferation fate of cells (23,25). As shown in Figure 5A and B, we found an upregulation of CARF in Wi-A-treated cancer cells. In contrast to the activation of NFκB and p38MAPK, occurred selectively in cancer cells (Figures 1 and 3), the level of CARF was upregulated in normal cells in response to Wi-A treatment. In line with these data, both cancer and normal cells also showed increase in γH2AX foci and reactive oxygen species (Figure 5A) consequent to Wi-A treatment, suggesting that it may cause growth arrest/senescence both in cancer and normal cells. Analyses of p21WAF1 and p16INK4A proteins that regulate growth arrest and senescence in cells revealed increase in Wi-A-treated cancer and normal cells (Figure 5C and D). As expected, the downstream effector of these proteins CDK4 and CDK2 showed decrease that in turn decrease the phosphorylated pRB leading to cell cycle arrest (Figure 6A–D). Although normal cells did not show decrease in expression of cyclin D1 (Figure 4F) or CDK4 (Figure 6A and B) expression levels, phosphorylated pRB showed distinct decrease in Wi-A-treated cells (Figure 6A and B) suggesting decrease in kinase activity of cyclinD1–CDK4 complex. Indeed, induction of senescence was confirmed by β-galactosidase staining and HP1γ foci in Wi-A-treated both cancer and normal cells (Figure 6C and D). Of note, because H1299 cells lacked p53 protein (Supplementary Figure 2D), an increase in p21WAF1 was therefore attributed to p53-independent mechanism, such as increase in CARF, resulting in decreased levels of HDM2 and stabilization of p21WAF1 (23,26,41,42). Figure 5. View largeDownload slide Upregulation of DNA damage response in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CARF, γH2AX, and reactive oxygen species (ROS) revealed their upregulation in Wi-A treated both cancer and normal cells as detected by immunostaining (A) and Western blotting (B). p38 MAPK showed selective increase in cancer cells (B). Upregulation of p21WAF1 and p16INK4A was also observed both in cancer and normal cells (C and D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 5. View largeDownload slide Upregulation of DNA damage response in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CARF, γH2AX, and reactive oxygen species (ROS) revealed their upregulation in Wi-A treated both cancer and normal cells as detected by immunostaining (A) and Western blotting (B). p38 MAPK showed selective increase in cancer cells (B). Upregulation of p21WAF1 and p16INK4A was also observed both in cancer and normal cells (C and D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 6. View largeDownload slide Downregulation of CDK4, CDK2, and pRB phosphorylation in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CDK4, CDK2, and phosphorylated pRB showed their downregulation in Wi-A-treated cancer cells (A). Normal cells showed decrease in CDK2 and phosphorylated pRB (A). Western blotting showing decrease in CDK2 and phosphorylated pRB both in cancer and normal cells; CDK4 showed decrease in cancer cells only (B). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Induction of senescence in Wi-A-treated cancer and normal cells. Senescence-associated β-galactosidase staining (C) and HP1γ (D) foci appeared both in H1299 (p53−/−) and MRC5 (p53+/+) cells subsequent to Wi-A treatment. Schematic diagram showing the effect of Wi-A on DNA damage signaling in cancer and normal cells. Whereas p38 and nuclear factor kappa B (NFκB) axis (E) were selectively activated in cancer cells (shown by gray arrows), CARF (collaborator of ARF) and its downstream activators p53, p21WAF1, and p16INK4A were activated both in cancer and normal cells (shown by black arrow) resulting in induction of senescence. Figure 6. View largeDownload slide Downregulation of CDK4, CDK2, and pRB phosphorylation in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CDK4, CDK2, and phosphorylated pRB showed their downregulation in Wi-A-treated cancer cells (A). Normal cells showed decrease in CDK2 and phosphorylated pRB (A). Western blotting showing decrease in CDK2 and phosphorylated pRB both in cancer and normal cells; CDK4 showed decrease in cancer cells only (B). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Induction of senescence in Wi-A-treated cancer and normal cells. Senescence-associated β-galactosidase staining (C) and HP1γ (D) foci appeared both in H1299 (p53−/−) and MRC5 (p53+/+) cells subsequent to Wi-A treatment. Schematic diagram showing the effect of Wi-A on DNA damage signaling in cancer and normal cells. Whereas p38 and nuclear factor kappa B (NFκB) axis (E) were selectively activated in cancer cells (shown by gray arrows), CARF (collaborator of ARF) and its downstream activators p53, p21WAF1, and p16INK4A were activated both in cancer and normal cells (shown by black arrow) resulting in induction of senescence. Discussion NFκB is a transcription factor that regulates several proteins involved in growth arrest and apoptosis (1). Constitutional activation of NFκB in cancer cells has been related to their anti-apoptosis and continued proliferation characteristics. It has been shown to regulate cyclin D1 and cyclin E that along with CDK2 and CDK4/6 cause phosphorylation of tumor suppressor pRB protein, essential for cell cycle progression. Some studies have reported that Wi-A might inhibit IκB phosphorylation and NFκB translocation into the nucleus (8,11,12,27). NF-kB has been shown to play a major role in cellular senescence (43), an innate tumor suppressor mechanism, triggered by either activation of oncogenes and/or DNA damage signaling (44), and inflammatory response involving changes in extracellular matrix proteins (45) also related to senescence-associated secretory phenotype (21). Considering that Wi-A is capable of binding to multiple sites, we examined its effect on the conformation of IKKβ–Nemo complex. Contrary to the reports on its effect on local domains (28), we found that Wi-A neither disrupts interaction between IKKβ and Nemo chains nor inhibits the formation of IKK complex. Furthermore, change in conformation of IKKβ–Nemo in the presence of Wi-A was associated with increased level of IκB phosphorylation and NFκB translocation to the nucleus suggesting that Wi-A caused activation of IκB–NFκB signaling. Indeed, it was supported by the experimental data (Figures 1 and 3). Bioinformatics analysis on the effect of Wi-A on binding of NFκB dimer to DNA showed that it stably interacted with both DNA and protein at the DNA-binding site. However, it caused some instability in the NFκB–DNA interactions and was endorsed by experiments on the DNA-binding-based ELISA for transcriptional activity (Figure 4E). Analyses of NFκB downstream effectors such as cyclin D1, cyclin E, CDK2, and CDK4 in control and Wi-A-treated cells indeed revealed their lower level of expression in Wi-A treated cells when compared with control cancer cells. Interestingly, decrease in cyclin D1 and CDK4, but not cyclin E and CDK2, was selective in cancer cells suggesting that Wi-A therapy may be particularly beneficial for cancers with constitutively active cyclin D1 and CDK4. Decrease in CDK2 and CDK4 expression and activities (as determined by decrease in phosphorylated pRB) was assigned to increase in its upstream inhibitors (p21WAF1 and p16INK4A) that showed increase in Wi-A-treated cells. Such increase was assigned to activation of DNA damage and oxidative stress signaling as determined by increased level of expression of CARF (23–26,41,42) and γH2AX, suggesting induction of senescence. Induction of DNA damage and reactive oxygen species was observed and was supported by earlier studies (14,46–49). Senescence-associated β-galactosidase and HP1γ assays, indeed, endorsed Wi-A-induced senescence both in cancer and normal cells. The study demonstrated that Wi-A caused activation of DDR by pleiotropic mechanisms (Figure 6E). Although one axis of activated DNA damage signaling, p38MAPK–NFκB, was activated selectively in cancer cells, the other one, CARF-p21WAF1/p16INK4A–cyclin/CDK-pRB, was activated both in cancer and normal cells suggesting that CARF plays an important role in Wi-A-induced senescence. Funding This study was supported by grants from DBT (Government of India) and AIST (Japan). Conflict of Interest None reported. Acknowledgments Computations were performed at the Bioinformatics Centre at IIT Delhi. In silico analyses were performed by V.M. Experiments were performed by P.B., Y.L., and J.R. S.C.K., D.S., and R.W. conceived and designed the experiments and contributed reagents/materials/analysis tools for this study. P.B., V.M., S.C.K., D.S., and R.W. wrote the article. P.B. and V.M. contributed equally as co-first authors. D.S. and R.W. contributed equally as co-senior (corresponding) authors. References 1. Cartwright T , Perkins ND , L Wilson C . NFKB1: a suppressor of inflammation, ageing and cancer . FEBS J . 2016 ; 283 : 1812 – 1822 . doi: 10.1111/febs.13627 Google Scholar Crossref Search ADS PubMed 2. Dolcet X , Llobet D , Pallares J , Matias-Guiu X . NF-kB in development and progression of human cancer . Virchows Arch . 2005 ; 446 : 475 – 482 . doi: 10.1007/s00428-005-1264-9 Google Scholar Crossref Search ADS PubMed 3. Yu JH , Kim H . Oxidative stress and cytokines in the pathogenesis of pancreatic cancer . J Cancer Prev . 2014 ; 19 : 97 – 102 . doi: 10.15430/JCP.2014.19.2.97 Google Scholar Crossref Search ADS PubMed 4. Hoesel B , Schmid JA . The complexity of NF-κB signaling in inflammation and cancer . Mol Cancer . 2013 ; 12 : 86 . doi: 10.1186/1476-4598-12-86 Google Scholar Crossref Search ADS PubMed 5. Viatour P , Merville MP , Bours V , Chariot A . Phosphorylation of NF-kappaB and IkappaB proteins: implications in cancer and inflammation . Trends Biochem Sci . 2005 ; 30 : 43 – 52 . doi: 10.1016/j.tibs.2004.11.009 Google Scholar Crossref Search ADS PubMed 6. Wan F , Lenardo MJ . Specification of DNA binding activity of NF-kappaB proteins . Cold Spring Harb Perspect Biol . 2009 ; 1 : a000067 . doi: 10.1101/cshperspect.a000067 Google Scholar Crossref Search ADS PubMed 7. Park MH , Hong JT . Roles of NF-kappaB in cancer and inflammatory diseases and their therapeutic approaches . Cells . 2016 ; 5 : 15 – 27 . doi: 10.3390/cells5020015 Google Scholar Crossref Search ADS 8. Heyninck K , Lahtela-Kakkonen M , Van der Veken P , Haegeman G , Vanden Berghe W . Withaferin A inhibits NF-kappaB activation by targeting cysteine 179 in IKKβ . Biochem Pharmacol . 2014 ; 91 : 501 – 509 . doi: 10.1016/j.bcp.2014.08.004 Google Scholar Crossref Search ADS PubMed 9. Jackson SS , Oberley C , Hooper CP et al. Withaferin A disrupts ubiquitin-based NEMO reorganization induced by canonical NF-κB signaling . Exp Cell Res . 2015 ; 331 : 58 – 72 . doi: 10.1016/j.yexcr.2014.09.034 Google Scholar Crossref Search ADS PubMed 10. Syam S , Bustamam A , Abdullah R et al. β Mangostin suppress LPS-induced inflammatory response in RAW 264.7 macrophages in vitro and carrageenan-induced peritonitis in vivo . J Ethnopharmacol . 2014 ; 153 : 435 – 445 . doi: 10.1016/j.jep.2014.02.051 Google Scholar Crossref Search ADS PubMed 11. Maitra R , Porter MA , Huang S , Gilmour BP . Inhibition of NFkappaB by the natural product withaferin A in cellular models of cystic fibrosis inflammation . J Inflamm (Lond) . 2009 ; 6 : 15 . doi: 10.1186/1476-9255-6-15 Google Scholar Crossref Search ADS PubMed 12. Kaileh M , Vanden Berghe W , Heyerick A et al. Withaferin a strongly elicits IkappaB kinase beta hyperphosphorylation concomitant with potent inhibition of its kinase activity . J Biol Chem . 2007 ; 282 : 4253 – 4264 . doi: 10.1074/jbc.M606728200 Google Scholar Crossref Search ADS PubMed 13. Ghosh K , De S , Das S , Mukherjee S , Sengupta Bandyopadhyay S . Withaferin a induces ROS-mediated paraptosis in human breast cancer cell-lines MCF-7 and MDA-MB-231 . PLoS One . 2016 ; 11 : e0168488 . doi: 10.1371/journal.pone.0168488 Google Scholar Crossref Search ADS PubMed 14. Malik F , Kumar A , Bhushan S et al. Reactive oxygen species generation and mitochondrial dysfunction in the apoptotic cell death of human myeloid leukemia HL-60 cells by a dietary compound withaferin A with concomitant protection by N-acetyl cysteine . Apoptosis . 2007 ; 12 : 2115 – 2133 . doi: 10.1007/s10495-007-0129-x Google Scholar Crossref Search ADS PubMed 15. O’Connor MJ . Targeting the DNA damage response in cancer . Mol Cell . 2015 ; 60 : 547 – 560 . doi: 10.1016/j.molcel.2015.10.040 Google Scholar Crossref Search ADS PubMed 16. Olivieri F , Albertini MC , Orciani M , Ceka A , Cricca M , Procopio AD et al. DNA damage response (DDR) and senescence: shuttled inflamma-miRNAs on the stage of inflamm-aging . Oncotarget . 2015 ; 6 : 35509 – 35521 . doi: 10.18632/oncotarget.5899 Google Scholar PubMed 17. von Zglinicki T , Saretzki G , Ladhoff J , d’Adda di Fagagna F , Jackson SP . Human cell senescence as a DNA damage response . Mech Ageing Dev . 2005 ; 126 : 111 – 117 . doi: 10.1016/j.mad.2004.09.034 Google Scholar Crossref Search ADS PubMed 18. Wood CD , Thornton TM , Sabio G , Davis RA , Rincon M . Nuclear localization of p38 MAPK in response to DNA damage . Int J Biol Sci . 2009 ; 5 : 428 – 437 . doi: 10.7150/ijbs.5.428 Google Scholar Crossref Search ADS PubMed 19. Borodkina A , Shatrova A , Abushik P , Nikolsky N , Burova E . Interaction between ROS dependent DNA damage, mitochondria and p38 MAPK underlies senescence of human adult stem cells . Aging (Albany, NY) . 2014 ; 6 : 481 – 495 . doi: 10.18632/aging.100673 Google Scholar Crossref Search ADS 20. Salminen A , Kauppinen A , Kaarniranta K . Emerging role of NF-κB signaling in the induction of senescence-associated secretory phenotype (SASP) . Cell Signal . 2012 ; 24 : 835 – 845 . doi: 10.1016/j.cellsig.2011.12.006 Google Scholar Crossref Search ADS PubMed 21. Freund A , Patil CK , Campisi J . p38MAPK is a novel DNA damage response-independent regulator of the senescence-associated secretory phenotype . EMBO J . 2011 ; 30 : 1536 – 1548 . doi: 10.1038/emboj.2011.69 Google Scholar Crossref Search ADS PubMed 22. Vaughan S , Jat PS . Deciphering the role of nuclear factor-κB in cellular senescence . Aging (Albany, NY) . 2011 ; 3 : 913 – 919 . doi: 10.18632/aging.100390 Google Scholar Crossref Search ADS 23. Cheung CT , Hasan MK , Widodo N , Kaul SC , Wadhwa R . CARF: an emerging regulator of p53 tumor suppressor and senescence pathway . Mech Ageing Dev . 2009 ; 130 : 18 – 23 . doi: 10.1016/j.mad.2008.05.002 Google Scholar Crossref Search ADS PubMed 24. Hasan K , Cheung C , Kaul Z et al. CARF is a vital dual regulator of cellular senescence and apoptosis . J Biol Chem . 2009 ; 284 : 1664 – 1672 . doi: 10.1074/jbc.M805778200 Google Scholar Crossref Search ADS PubMed 25. Cheung CT , Singh R , Kalra RS , Kaul SC , Wadhwa R . Collaborator of ARF (CARF) regulates proliferative fate of human cells by dose-dependent regulation of DNA damage signaling . J Biol Chem . 2014 ; 289 : 18258 – 18269 . doi: 10.1074/jbc.M114.547208 Google Scholar Crossref Search ADS PubMed 26. Singh R , Kalra RS , Hasan K et al. Molecular characterization of collaborator of ARF (CARF) as a DNA damage response and cell cycle checkpoint regulatory protein . Exp Cell Res . 2014 ; 322 : 324 – 334 . doi: 10.1016/j.yexcr.2014.01.022 Google Scholar Crossref Search ADS PubMed 27. McKenna MK , Gachuki BW , Alhakeem SS et al. Anti-cancer activity of withaferin A in B-cell lymphoma . Cancer Biol Ther . 2015 ; 16 : 1088 – 1098 . doi: 10.1080/15384047.2015.1046651 Google Scholar Crossref Search ADS PubMed 28. Grover A , Shandilya A , Punetha A , Bisaria VS , Sundar D . Inhibition of the NEMO/IKKβ association complex formation, a novel mechanism associated with the NF-κB activation suppression by Withania somnifera’s key metabolite withaferin A . BMC Genomics . 2010 ; 11 ( suppl 4 ): S25 . doi: 10.1186/1471-2164-11-S4-S25 Google Scholar Crossref Search ADS PubMed 29. Morris GM , Huey R , Lindstrom W et al. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility . J Comput Chem . 2009 ; 30 : 2785 – 2791 . doi: 10.1002/jcc.21256 Google Scholar Crossref Search ADS PubMed 30. de Vries SJ , van Dijk M , Bonvin AM . The HADDOCK web server for data-driven biomolecular docking . Nat Protoc . 2010 ; 5 : 883 – 897 . doi: 10.1038/nprot.2010.32 Google Scholar Crossref Search ADS PubMed 31. Schrödinger Suite 2016-3: Protein Preparation Wizard; Epik version 3.3; Impact version 6.8; Prime version 4.1; LigPrep; Glide . New York, NY : Schrödinger, LLC ; 2016 . 32. Friesner RA , Murphy RB , Repasky MP et al. Extra precision glide: docking and scoring incorporating a model of hydrophobic enclosure for protein-ligand complexes . J Med Chem . 2006 ; 49 : 6177 – 6196 . doi: 10.1021/jm051256o Google Scholar Crossref Search ADS PubMed 33. Case D , Darden T , Cheatham T III et al. AMBER 12 . San Francisco, CA : University of California ; 2012 . 34. Humphrey W , Dalke A , Schulten K . VMD: visual molecular dynamics . J Mol Graph . 1996 ; 14 : 33 – 38 . doi: 10.1016/0263-7855(96)00018-5 Google Scholar Crossref Search ADS PubMed 35. DeLano WL. PyMOL: The PyMOL Molecular Graphics System, Version 1.3 . Schrödinger, LLC ; 2002 . 36. Laskowski RA , Swindells MB . LigPlot+: multiple ligand-protein interaction diagrams for drug discovery . J Chem Inf Model . 2011 ; 51 : 2778 – 2786 . doi: 10.1021/ci200227u Google Scholar Crossref Search ADS PubMed 37. Rushe M , Silvian L , Bixler S et al. Structure of a NEMO/IKK-associating domain reveals architecture of the interaction site . Structure . 2008 ; 16 : 798 – 808 . doi: 10.1016/j.str.2008.02.012 Google Scholar Crossref Search ADS PubMed 38. Marienfeld RB , Palkowitsch L , Ghosh S . Dimerization of the I kappa B kinase-binding domain of NEMO is required for tumor necrosis factor alpha-induced NF-kappa B activity . Mol Cell Biol . 2006 ; 26 : 9209 – 9219 . doi: 10.1128/MCB.00478-06 Google Scholar Crossref Search ADS PubMed 39. Herlaar E , Brown Z . p38 MAPK signalling cascades in inflammatory disease . Mol Med Today . 1999 ; 5 : 439 – 447 . doi: 10.1016/S1357-4310(99)01544-0 Google Scholar Crossref Search ADS PubMed 40. Olson CM , Hedrick MN , Izadi H , Bates TC , Olivera ER , Anguita J . p38 mitogen-activated protein kinase controls NF-kappaB transcriptional activation and tumor necrosis factor alpha production through RelA phosphorylation mediated by mitogen- and stress-activated protein kinase 1 in response to Borrelia burgdorferi antigens . Infect Immun . 2007 ; 75 : 270 – 277 . doi: 10.1128/IAI.01412-06 Google Scholar Crossref Search ADS PubMed 41. Hasan MK , Yaguchi T , Harada JI , Hirano T , Wadhwa R , Kaul SC . CARF (collaborator of ARF) interacts with HDM2: evidence for a novel regulatory feedback regulation of CARF-p53-HDM2-p21WAF1 pathway . Int J Oncol . 2008 ; 32 : 663 – 671 . doi: 10.3892/ijo.32.3.663 Google Scholar PubMed 42. Kalra RS , Cheung CT , Chaudhary A , Prakash J , Kaul SC , Wadhwa R . CARF (collaborator of ARF) overexpression in p53-deficient cells promotes carcinogenesis . Mol Oncol . 2015 ; 9 : 1877 – 1889 . doi: 10.1016/j.molonc.2015.07.003 Google Scholar Crossref Search ADS PubMed 43. Lee YK , Cha HJ , Hong M , Yoon Y , Lee H , An S . Role of NF-κB-p53 crosstalk in ultraviolet A-induced cell death and G1 arrest in human dermal fibroblasts . Arch Dermatol Res . 2012 ; 304 : 73 – 79 . doi: 10.1007/s00403-011-1176-2 Google Scholar Crossref Search ADS PubMed 44. Schmitt CA , Fridman JS , Yang M et al. A senescence program controlled by p53 and p16INK4a contributes to the outcome of cancer therapy . Cell . 2002 ; 109 : 335 – 346 . doi: 10.1016/S0092-8674(02)00734-1 Google Scholar Crossref Search ADS PubMed 45. Freund A , Orjalo AV , Desprez PY , Campisi J . Inflammatory networks during cellular senescence: causes and consequences . Trends Mol Med . 2010 ; 16 : 238 – 246 . doi: 10.1016/j.molmed.2010.03.003 Google Scholar Crossref Search ADS PubMed 46. Widodo N , Priyandoko D , Shah N , Wadhwa R , Kaul SC . Selective killing of cancer cells by Ashwagandha leaf extract and its component withanone involves ROS signaling . PLoS One . 2010 ; 5 : e13536 . doi: 10.1371/journal.pone.0013536 Google Scholar Crossref Search ADS PubMed 47. Kakar SS , Jala VR , Fong MY . Synergistic cytotoxic action of cisplatin and withaferin A on ovarian cancer cell lines . Biochem Biophys Res Commun . 2012 ; 423 : 819 – 825 . doi: 10.1016/j.bbrc.2012.06.047 Google Scholar Crossref Search ADS PubMed 48. Fong MY , Jin S , Rane M , Singh RK , Gupta R , Kakar SS . Withaferin A synergizes the therapeutic effect of doxorubicin through ROS-mediated autophagy in ovarian cancer . PLoS One . 2012 ; 7 : e42265 . doi: 10.1371/journal.pone.0042265 Google Scholar Crossref Search ADS PubMed 49. Yu Y , Katiyar SP , Sundar D et al. Withaferin-A kills cancer cells with and without telomerase: chemical, computational and experimental evidences . Cell Death Dis . 2017 ; 8 : e2755 . doi: 10.1038/cddis.2017.33 Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences Oxford University Press

Molecular Insights Into Withaferin-A-Induced Senescence: Bioinformatics and Experimental Evidence to the Role of NFκB and CARF

Loading next page...
 
/lp/ou_press/molecular-insights-to-withaferin-a-induced-senescence-bioinformatics-Y7szwjg0TT
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5006
eISSN
1758-535X
DOI
10.1093/gerona/gly107
Publisher site
See Article on Publisher Site

Abstract

Abstract Withaferin-A (Wi-A) has been shown to possess anticancer activity. Molecular mechanism(s) of its action has not been fully resolved. We recruited low dose of Wi-A that caused slow growth arrest in cancer cells and was relatively safe for normal cells. Consistently, we detected nuclear translocation of nuclear factor kappa B (NFκB) and activation of p38MAPK selectively in cancer cells. Bioinformatics analyses revealed that Wi-A did not disrupt IKKα/IKKβ–Nemo complex that regulates NFκB activity. However, it caused moderate change in the conformation of IKKβ–Nemo interacting domain. Experimental data revealed increased level of phosphorylated IκBα in Wi-A-treated cells, suggesting an activation of IKK complex that was supported by nuclear translocation of NFκB. Molecular docking analysis showed that Wi-A did not disrupt; however, decreased the stability of the NFκB–DNA complex. It was supported by downregulation of DNA-binding and transcriptional activities of NFκB. Further analysis revealed that Wi-A caused upregulation of CARF (collaborator of ARF) demonstrating an activation of DNA damage oxidative stress response in both cancer and normal cells. In line with this, upregulation of p21WAF1, p16INK4A, and hypophosphorylated pRB and induction of senescence were observed demonstrating that Wi-A-induced senescence is mediated by multiple pathways in which CARF-mediated DNA damage and oxidative stress play a major role. Withaferin-A, Cancer, Cellular senescence, p53–p21 pathway, NFκB, CARF Nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) family of proteins is a class of transcription factors that possess a Rel homology domain in their N-terminus and control expression of genes involved in regulation of growth, apoptosis, immunoregulatory, and inflammatory processes (1,2). It regulates cellular response to oxidative stress, cytokines, bacterial, and viral antigens and plays a key role in regulating the immune response to infection (3). NFκB1 and NFκB2 are transcription factors synthesized as large precursors, p105, and p100 that are cleaved to mature NFκB p65 and p50 proteins by selective degradation of their C-terminal region containing ankyrin repeats (4). These are present in cells in an inactive (sequestered in cytoplasm) state by binding to their inhibitors called IκBs (inhibitor of NFκB). Intracellular/extracellular stresses or cytokine signals activate its upstream regulator, IKK (Inhibitor of IkB Kinase), which phosphorylate IkBα at Ser32 and Ser36 residues. The phosphorylated IκBα undergoes proteasomal degradation resulting in active NFκB that translocates to the nucleus and performs transcriptional activation function (5). The NFκB proteins lack intrinsic transcriptional activation ability and function by binding as homodimers (6). Its crucial role in processes such as proliferation, apoptosis, and invasiveness requires controlled activation (7). Dysregulated activation has been shown to be associated with various conditions such as arthritis, asthma, and inflammatory disorders (1,8). Aberrant or constitutive activation of NFκB has been observed in various types of cancer cells marking it as a potential therapeutic target (1,2,9). In line with this, identification and functional characterization of natural and synthetic inhibitors of NFκB have been initiated. Inhibition of NFκB by small molecules, β-mangostin (βM) isolated from Cratoxylum arborescens (10) and Wi-A (withaferin-A) isolated from medicinal herb Withania somnifera (Ashwagandha) (8,9,11,12), has been reported. It was shown that Wi-A inhibits DNA binding of NFκB resulting in reactive oxygen species generation, mitochondrial dysfunction, oxidative stress, and apoptosis in cancer cells (11–14). DNA damage response (DDR) is differentially regulated in cancer and normal cells. Whereas normal cells respond by executing growth arrest, cancer cells are refractory and keep proliferating (15). Several studies have established that DDR signaling is essential for execution of senescence, an established powerful tumor suppressor mechanism (16,17). The p38MAPK is a kinase that is activated through environmental stress and DNA damage stress, including ionizing radiation, ultraviolet, chemotherapeutic drugs, and lead to the induction of a G2/M cell cycle checkpoint through p53-dependent and -independent mechanisms (18,19). p38MAPK has been shown to transcriptionally activate NFκB leading to senescence (20–22), mediated by upregulation of p53-p21WAF1 pathway. CARF (collaborator of ARF), an ARF (Alternative Reading Frame, p14ARF)-interacting protein, has been shown to regulate activities of p53-tumor suppressor protein in an ARF-dependent or -independent manner (23–25). It has been shown to regulate DDR in a dose-dependent manner and regulates cell proliferative fates in normal and cancer cells (25,26). Wi-A, a withanolide extracted from medicinal herb W. somnifera, has been shown to cause inhibition of IKK activity in some earlier studies (8,12,27). Through bioinformatics analysis, we had previously shown that Wi-A disrupts important hydrophobic interaction between IKKβ and Nemo chain residues, L93:F734, T735:F92, F734:M94, W739:F97, W741:A100, W741:R101, thereby inhibiting their complex (28). However, an experimental study failed to reveal the inhibition of IKKβ–Nemo interaction by Wi-A; instead, it inhibited IκB kinase activity by interacting with C179 residue located in catalytic domain of IKKβ (8). It was also demonstrated that Wi-A hyperphosphorylates IKKβ at S181 causing inhibition of TNF-induced IKK activity and, thereby, causing inhibition of IκB degradation and p65 translocation (12). These reports described three different binding sites of Wi-A on IKK complex suggesting its interaction at multiple sites. Inhibition of IKK activity by Wi-A and/or NFκB translocation to the nucleus has been reported in some studies (8,12,27). However, the molecular insights and overall impact of interactions of Wi-A with IKKβ or NFκB structure and activity remain undefined. We hypothesized that the Wi-A might affect the conformation of the binding complex and hence set out to determine such impact on IKKβ–Nemo interaction domain and NFκB activity by molecular docking and dynamic simulations. We found that Wi-A caused significant change in conformation of IKKβ–Nemo and NFκB–DNA interaction domains. We provide experimental evidence of activation of kinase activity of IKK complex resulting in phosphorylation of IκBα, its degradation, and nuclear translocation of NFκB. Wi-A was further seen as not disrupting, but destabilizing NFκB–DNA interactions and was supported by the decreased level of expression of cyclin D1, cyclin E, and CDK2/4. We provide evidence that Wi-A triggers DDR, as supported by upregulation of γH2AX and CARF that yielded senescence in both cancer and normal cells. Material and Methods Cell Culture and Antibodies Human lung carcinoma (H1299 and A549) and normal fibroblasts (MRC5) were cultured in RPMI-1640 and Dulbecco’s modified Eagle’s medium, respectively, supplemented with 10% (vol/vol) fetal bovine serum in 5% CO2 and 95% air humidified incubator. The antibodies were purchased from Santa Cruz Biotech Inc., CA (NFκB-p65, p-IκBα p53, p38MAPK, p16INK4A, CDK2, cyclin E, cyclin D1, and CDK4), Cell Signaling, Beverly, MA (HP1γ, p-p38MAPK, pRB [S780], and γH2AX, p21WAF1), and Abcam, Cambridge, UK (β-actin). Anti-CARF antibodies were generated in our laboratory. Cell Viability Assays Short-term cell viability (2–3 days) and long-term colony forming assays (5–12 days) were performed in 96- and 6-well tissue culture plates, respectively, as described earlier (25,26). Western Blot Analysis The cells were treated with Wi-A (0.2–2.0 µg/mL, as indicated) for 24 hours following which cell lysates were prepared in RIPA buffer (Thermo Scientific, Rockford, IL) containing protease inhibitor cocktail (Roche Applied Science, Mannheim, Germany). Control cells were treated with dimethyl sulfoxide. Cell lysates (containing 20-µg protein) were subjected to Western blot analysis with antibodies as indicated. The blots were then developed using chemiluminescence (GE Healthcare, UK) and visualized using Lumino Image Analyzer equipped with CCD camera (LAS3000-mini; Fuji Film, Tokyo, Japan) as described earlier (25,26). Nuclear and cytoplasmic fractions were prepared using the Qproteome cell compartment kit (Qiagen, Hilden, Germany). Immunostaining Cells (1 × 104), cultured on a glass coverslips placed in a 12-well plate, were treated with Wi-A and fixed with prechilled methanol:acetone (1:1) for 5 minutes at 4°C. Fixed cells were incubated with primary antibodies (as indicated) followed by extensively washing (0.2% Triton X-100 in phosphate-buffered saline) and incubation with the fluorochrome-conjugated secondary antibodies (Alexa-488-conjugated goat anti-rabbit or anti-mouse and Alexa-594-conjugated goat anti-rabbit or anti-mouse [Molecular Probes, OR]) for 45 minutes as described earlier (25,26). Stained cells were examined under LSM700 laser scanning confocal microscope from Carl Zeiss. The images were taken using AxioVision 4.6 software (Carl Zeiss Microimaging Inc., Jena, Germany) and IMARIS software (Bitplane, Zurich, Switzerland). Immunoprecipitation Cell lysates of control and Wi-A-treated cells (1-mg protein) in Nonidet (NP)-P40 lysis buffer were incubated with p65 rabbit polyclonal antibody and control IgG (Cell Signaling, Beverly, MA) for 3 hours in slow rotation at 4°C. Immunocomplex was precipitated by incubation with Protein-A/G plus Agarose (20 mL; Santa Cruz Biotech. Inc., CA, sc-2003) for 45 minutes followed by centrifugation at 2,500 rpm for 5 minutes at 4°C. Pellets were washed five times with NP-40 buffer with repeated centrifugation at 2,500 rpm for 5 minutes at 4°C. Immunocomplex was resolved on SDS/10%PAGE and electroblotted onto a polyvinylidene fluoride membrane. The proteins in immunocomplex were detected by Western blotting with the indicated monoclonal antibody and captured using ECL (GE Healthcare, UK). Cell Cycle Analysis Cells treated with indicated doses of Wi-A for 24 hours were collected in 1.5-mL tube, washed with cold phosphate-buffered saline, and fixed with 70% ethanol at 4°C for 12 hours. The fixed cells, centrifuged (2,000 rpm for 10 minutes), washed with cold phosphate-buffered saline, and resuspended in 0.25-mL phosphate-buffered saline, were stained with Guava Cell Cycle Reagent (Millipore, Tokyo, Japan) for 30 minutes in dark. To avoid false DNA-PI staining, RNA was removed by RNase A treatment (5 µL of 1 mg/mL at 37°C for 1 hour). Cell cycle analysis was performed using Guava PCA-96 System (Millipore, Tokyo, Japan) and CytoSoft TM Software, version 2.5.6 (Millipore, Tokyo, Japan). Apoptosis and Senescence Assays Cells (2 × 105 cells/well in six-well plates) were treated with Wi-A. Apoptosis was detected and analyzed by Guava Nexin Reagent (EMD Millipore Corporation, Tokyo, Japan) and Flow Jo Software, respectively. Senescent cells were detected using senescence-associated β-galactosidase kit (Cell Signaling Technology, Danvers, MA) by methods as recommended by the manufacturers. NFκB–DNA Binding Assay NFκB–DNA binding assay was performed using the NFκB–DNA binding Assay Kit (AB#133112, Abcam, Cambridge, MA) following the manufacturer’s instructions. Nuclear cell extracts of control and Wi-A-treated samples were prepared using Qproteome Cell Compartment Kit (#37502, Qiagen Inc., Manchester, UK) following the manufacturer’s protocol. Statistical Analysis All the experiments were performed in triplicate, and variables were expressed as mean ± SEM of triplicate experiments. Unpaired t test (GraphPad Prism, GraphPad Software, San Diego, CA) has been performed to determine the degree of significance between the control and experimental samples. Statistical significance was defined as p value and represented by *p < .05, **p < .01, ***p < .001, whereas no mark denotes insignificant correlation. Computational Methods Docking of Wi-A With IKKβ–Nemo Association Domain The structure of IKKβ–Nemo association domain, complex of two IKKβ kinase subunits and two regulatory subunits of NEMO (NFκB essential modulator), was obtained from Protein Data Bank (PDB ID: 3BRV). Flexible docking of Wi-A with IKKβ–Nemo complex was carried out by assigning flexibility to side chains of their key interacting residues (28). The structure of the protein and ligand molecule, Wi-A (PubChem ID: 265237), deposited in the databases was preprocessed, followed by generation of grid around the flexible residues and docking of Wi-A within the designed grid using AUTODOCK 4.2 software (29). Binding efficiency of another Nemo chain with IKKβ:Nemo complex was also predicted in the presence and absence of Wi-A using HADDOCK (High Ambiguity Driven protein–protein DOCKing) (30). Docking of Wi-A With NFκB Dimer DNA Binding Domain The structure of NFκB dimer that is p50–p65 subunit bound to DNA (PDB ID: 3GUT) was retrieved from Protein Data Bank. The DNA chain was removed from the complex, and protein structure was prepared using PrepWizard of maestro (31). The structure of Wi-A was preprocessed using LigPrep version 3.5 of Schrodinger suite. Glide extra precision algorithm was used to dock Wi-A around residues forming hydrogen bonds with the DNA molecule (31,32). Binding affinity of DNA with NFκB dimer both in the presence and absence of Wi-A was predicted by carrying out protein–DNA docking using HADDOCK. Molecular Dynamics Simulations Amber Molecular Dynamics Suite was used to perform all the simulations on DELL T3610 workstation with 16-GB DDR RAM and NVIDIA GeForce GTX TITAN Black Graphics Card (33). Amber protein force field, ff12SB, was used to perform simulations of all complexes solvated with TIP3P water octahedral box using a spacing distance of 10 Å around the molecule. Solvated molecule was then neutralized using appropriate number of counterions followed by minimization, heating up to 300 K temperature and equilibration of molecule and molecular dynamic simulations. Three sets of molecular dynamic simulations were performed to analyze the effect of Wi-A on complete IKKβ–Nemo association domain. First set of complexes, IKKβ:Nemo:Wi-A docked complex and IKKβ:Nemo complex, were simulated for time duration of 251 and 200 ns, respectively. The second set of simulation was performed for IKKβ:Nemo2:Wi-A complex and IKKβ:Nemo2 complex for 50 and 48 ns, respectively. The final set of molecular dynamic simulations was performed on complete IKKβ/Nemo association domain complexes in the presence and absence of Wi-A, namely, IKKβ2:Nemo2:Wi-A complex and IKKβ2:Nemo2 complex, which were simulated for the time period of 60 and 50 ns, respectively. The NFκB dimer:DNA complex and NFκB dimer:Wi-A docked complex were simulated for 40 and 50 ns, respectively, to check the effect of Wi-A on DNA binding activity of NFκB dimer. Coordinates of DNA were incorporated in NFκB dimer:Wi-A, and resulting NFκB dimer:DNA:Wi-A complex was simulated for a period of 50 ns to study the stability of Wi-A on DNA-bound complex of NFκB dimer. Analysis of Molecular Docking and Molecular Dynamics Simulations Root mean square deviation (RMSD) computation and conformational analysis over the range of simulation trajectories was performed using VMD version 1.9.2 (34). Superimposition of protein structures and generation of images were performed using the PyMol molecular graphics system (35). Protein–ligand interactions were studied using LigPlot + v.1.4.5 (36). Results Wi-A, at Low Dose, Triggers Growth Arrest in Human Lung Carcinoma Human lung cancer (H1299) cells were treated with serially increasing doses of Wi-A for 48 hours. As shown in Supplementary Figure 1A, cells showed serial increase in NFκB at doses 0.08–1.0 µg/mL followed by decrease at 1.6 and 2.0 µg/mL. We selected two doses of Wi-A (low, 0.2 µg/mL and high, 2.0 µg/mL) and performed cell proliferation assays. As shown in Supplementary Figure 1B, whereas high dose caused 50% reduction in viability within 48 hours, low dose caused only ~10% reduction followed by slow growth arrest. Annexin IV cytometric analysis revealed that the high, not low, dose instigated apoptosis in about 48 hours (Supplementary Figure 1C). In contrast to 5.12% in control, 8.71% and 34.59% cells were detected in apoptosis in low-dose- and high-dose-treated cultures, respectively (Figure 1C). Low-dose-induced slow growth arrest leads to reduction in colonigenicity in long-term colony forming assays (Supplementary Figure 1D), and most interestingly, we found that whereas low dose (24- to 48-hour treatment) induced G0/G1 cell cycle arrest in H1299 cells, the normal (MRC5) cells remained unaffected (Supplementary Figure 1E). Figure 1. View largeDownload slide Low dose of withaferin-A (Wi-A) caused nuclear translocation of nuclear factor kappa B (NFκB) in lung carcinoma. Immunostaining of NFκB, in control and Wi-A-treated cells, showed its nuclear translocation in cancer (H1299 and A549) cells (A and B). Normal cells (MRC5) did not show nuclear translocation of NFκB. Confocal images showing nuclear NFκB staining in cancer cells (B). Western blotting exhibited increase in NFκB expression in Wi-A-treated cells (C). Cell fractionation and Western blotting analyses confirmed the nuclear translocation of NFκB in Wi-A-treated cells (D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 1. View largeDownload slide Low dose of withaferin-A (Wi-A) caused nuclear translocation of nuclear factor kappa B (NFκB) in lung carcinoma. Immunostaining of NFκB, in control and Wi-A-treated cells, showed its nuclear translocation in cancer (H1299 and A549) cells (A and B). Normal cells (MRC5) did not show nuclear translocation of NFκB. Confocal images showing nuclear NFκB staining in cancer cells (B). Western blotting exhibited increase in NFκB expression in Wi-A-treated cells (C). Cell fractionation and Western blotting analyses confirmed the nuclear translocation of NFκB in Wi-A-treated cells (D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Low Dose of Wi-A Caused Nuclear Translocation of NFκB in Cancer Cells We next treated H1299 (cancer) and MRC5 (normal) cells with low dose of Wi-A (0.2–1.0 µg/mL) for 24–48 hours and examined the localization of NFκB in control and treated cells. As shown in Figure 1A, NFκB showed nuclear translocation only in cancer cells. The results were confirmed by confocal laser scanning microscopy (Figure 1B) and biochemical analyses including cell fraction and Western blotting with anti-NFκB, anti α-tubulin (cytoplasmic), and lamin A/C (nuclear) antibodies (Figure 1C and D). These data suggested the activation of NFκB in cancer cells and were in contrast to the earlier reports that have predicted inactivation of NFκB by Wi-A (8,12). Molecular Docking and Computational Inputs on Wi-A Targeting to IKKβ–Nemo Complex: Change in Conformation, Not Disruption, of the Complex We targeted IKKβ–Nemo association domain to unveil the effect of Wi-A on its conformation. Flexible docking of Wi-A was performed by generating grid around the key interacting residues in IKKβ and Nemo chain to check whether Wi-A can disrupt these interactions (37). IKKβ:Nemo:Wi-A docked complexes with minimum binding energy, −7.95 Kcal/mol, were selected. Wi-A formed two hydrogen bonds with R101 of Nemo chain as shown in Figure 2A. This lead to disruption of intermolecular interaction between Nemo and IKKβ chain residues R101 and D738, respectively, as previously reported in an X-ray crystallographic study of IKKβ–Nemo association domain (37). The complex was simulated for 251 ns to check the stability of Wi-A at the binding site. The analysis of snapshots of IKKβ:Nemo:Wi-A complex over the range of simulation showed that Wi-A was stably interacting with both chains with slight change in its position. Effect of such shift on ligand interaction at the binding site showed that Wi-A formed hydrophobic interactions with both IKKβ and Nemo chain residues, of which F97 and F734 were known to be originally involved in IKKβ–Nemo hydrophobic interactions (Figure 2B). The interaction site of Wi-A constituted amino acid residues ranging from 732 to 743 of IKKβ chain and residues 92–109 of Nemo chain. Furthermore, analysis of conformations of the complex over the simulation trajectories also showed the bending of Nemo chain at Wi-A interaction site. To confirm that the bending was induced by Wi-A, a 200-ns simulation of IKKβ:Nemo chain complex was performed. Comparison of two structures obtained after simulations confirmed that Wi-A does not inhibit interaction between IKKβ and Nemo chains, instead, caused bending of the Nemo chain (Figure 2C). To further confirm the inability of Wi-A to disrupt IKKβ–Nemo interaction, RMSD of IKKβ:Nemo complex and IKKβ:Nemo:Wi-A complex over the range of 200- and 251-ns simulation, respectively, was computed and compared. It was observed that the presence of Wi-A does not affect the RMSD trajectory of IKKβ:Nemo complex that endorsed the inability of Wi-A to disrupt IKKβ–Nemo interaction (Supplementary Figure 2A). Figure 2. View largeDownload slide Binding of withaferin-A (Wi-A) with IKKβ–Nemo complex and its effect on structural stability. Representation of hydrogen bond interactions of Wi-A with IKKβ–Nemo complex after molecular docking (A). Interactions formed by Wi-A with IKKβ–Nemo complex after 251-ns simulation. Wi-A is hydrophobically interacting with Phe97, Gly108, and Leu109 of Nemo chain and Gln732 and Phe734 of IKKβ chain (B). Superimposition of IKKβ–Nemo–Wi-A complex (purple) with IKKβ–Nemo complex taken as control (red) reveals the effect of Wi-A on confirmation of IKKβ–Nemo complex (C). Comparison of confirmation of IKKβ–Nemo2 complex generated after docking of Nemo Chain (pink) with IKKβ–Nemo complex (orange) in (D) the presence of Wi-A and (E) the absence of Wi-A. The comparison of the structures shows bending of both the Nemo chains toward Wi-A, whereas no such confirmation change was observed in the absence of Wi-A. Comparison of confirmation of complete IKKβ–Nemo association domain generated after inclusion of another IKKβ chain (pink) with IKKβ–Nemo2 complex (orange) in (F) the presence of Wi-A and (G) the absence of Wi-A. Figure 2. View largeDownload slide Binding of withaferin-A (Wi-A) with IKKβ–Nemo complex and its effect on structural stability. Representation of hydrogen bond interactions of Wi-A with IKKβ–Nemo complex after molecular docking (A). Interactions formed by Wi-A with IKKβ–Nemo complex after 251-ns simulation. Wi-A is hydrophobically interacting with Phe97, Gly108, and Leu109 of Nemo chain and Gln732 and Phe734 of IKKβ chain (B). Superimposition of IKKβ–Nemo–Wi-A complex (purple) with IKKβ–Nemo complex taken as control (red) reveals the effect of Wi-A on confirmation of IKKβ–Nemo complex (C). Comparison of confirmation of IKKβ–Nemo2 complex generated after docking of Nemo Chain (pink) with IKKβ–Nemo complex (orange) in (D) the presence of Wi-A and (E) the absence of Wi-A. The comparison of the structures shows bending of both the Nemo chains toward Wi-A, whereas no such confirmation change was observed in the absence of Wi-A. Comparison of confirmation of complete IKKβ–Nemo association domain generated after inclusion of another IKKβ chain (pink) with IKKβ–Nemo2 complex (orange) in (F) the presence of Wi-A and (G) the absence of Wi-A. To investigate the effect of Wi-A on IKKβ:Nemo dimer formation, binding affinity of Nemo chain to IKKβ:Nemo complex in the presence/absence of Wi-A was computed by docking another Nemo chain to both complexes around residues, 47–56, that are essential for Nemo–Nemo interaction (38). We observed minor reduction in binding affinity of Nemo chain to IKKβ:Nemo complex from −66.1 to −53.3 in the presence of Wi-A that were deemed not significant enough to inhibit the formation of IKKβ:Nemo dimer. Furthermore, both docked complexes, IKKβ:Nemo2:Wi-A and IKKβ:Nemo2, were also simulated for 50 and 48 ns, respectively. Comparison of IKKβ:Nemo2 structures in the presence/absence of Wi-A revealed that Wi-A caused bending of both Nemo chains in IKKβ:Nemo2:Wi-A complex, whereas no bending was observed in IKKβ:Nemo2 complex (Figure 2D and E). Wi-A Induced Change in Conformation of IKKβ–Nemo Interaction Domain The effect of Wi-A on conformation of complete IKKβ–Nemo interaction domain was examined by addition of coordinates of another IKKβ chain to IKKβ:Nemo2:Wi-A and IKKβ:Nemo2 complex. The resulting complexes, IKKβ2:Nemo2:Wi-A and IKKβ2:Nemo2, simulated for 60 and 50 ns, respectively. Both simulated structures were compared to unveil the effect of Wi-A on IKKβ–Nemo interaction domain, and change in conformation of IKKβ:Nemo association domain at Wi-A interaction site was observed in the presence of Wi-A (Figure 2F and G). Furthermore, Wi-A-induced instability of IKKβ:Nemo association domain structure at Wi-A interaction site was also supported by RMSD trajectories of Wi-A interaction site computed both in the presence/absence of Wi-A (Supplementary Figure 2B). We found that Wi-A interaction with IKKβ:Nemo association domain triggered instability of the structure at the interaction site, which was quite stable in the absence of Wi-A. To evaluate whether this instability is limited to the interaction site of Wi-A, RMSD of whole IKKβ:Nemo association domain excluding the Wi-A interaction site was also computed and compared (Supplementary Figure 2C). This comparison revealed that the presence of Wi-A has no effect on the stability of rest of the complex, but causes drastic changes in its interaction site that is evident in the form of bending of the Nemo chains. Wi-A Caused Nuclear Translocation of NFκB and p38MAPK To confirm the above computational findings, we performed experiments using human lung cancer (H1299 and A549) and normal (MRC5) cells. As shown in Figure 3A, immunostaining of control and Wi-A-treated H1299 cells with phosphorylation-specific anti-IκBα antibody revealed its increased level in cancer cells, but not in normal cells. These data endorsed that Wi-A did not inhibit the activity of IKKβ:Nemo complex and was in line with above computational analysis. Furthermore, Wi-A induced increase in phosphorylated IκBα, supporting the nuclear translocation of NFκB in cancer cells (Figure 1). These data were further supported by Western blotting and IκBα-p65 coimmunoprecipitation that revealed decrease (due to degradation of the phosphorylated form) in IκBα and its complex formation with p65 (Figure 3B). Several studies have shown that NFκB is regulated by p38MAPK and activated in response to various environmental stresses including DNA damage signaling, ultraviolet, ionization radiation, oxidative stress, and cytokines (39). Activation causes its phosphorylation, essential for its nuclear translocation (18), which in turn stimulates NFκB signaling (40). In view of these reports, we investigated whether Wi-A-induced activation of NFκB was mediated by p38MAPK. As shown in Figure 3C and D, we found nuclear p38MAPK in cancer cells. It was confirmed to be in activated phosphorylated form by Western blotting with phosphorylation-specific antibodies (Figure 3E). Normal cells did not show increase in p38MAPK (Figure 3C). These data suggested that Wi-A preferentially instigated DDR and activation of NFκB in cancer cells only. Figure 3. View largeDownload slide Upregulation of IκBα and p38MAPK in withaferin-A (Wi-A)-treated cells. Immunostaining of phosphorylated IκBα (A) in control and Wi-A-treated lung carcinoma and normal cells revealed their upregulation in cancer cells only. Western blotting for total IκBα in Wi-A-treated cells showing its decrease (B) coimmunoprecipitation (IP) of nuclear factor kappa B (NFκB) (p65) and IκBα showing decrease IκBα in IκBα-p65 complex (B). Increase in phosphorylated p38MAPK in Wi-A-treated cancer cells (C). Nuclear translocation of p38MAPK in Wi-A-treated cells as confirmed by confocal laser scanning microscopy (D) and Western blotting (E) with phosphorylation-specific antibody. Figure 3. View largeDownload slide Upregulation of IκBα and p38MAPK in withaferin-A (Wi-A)-treated cells. Immunostaining of phosphorylated IκBα (A) in control and Wi-A-treated lung carcinoma and normal cells revealed their upregulation in cancer cells only. Western blotting for total IκBα in Wi-A-treated cells showing its decrease (B) coimmunoprecipitation (IP) of nuclear factor kappa B (NFκB) (p65) and IκBα showing decrease IκBα in IκBα-p65 complex (B). Increase in phosphorylated p38MAPK in Wi-A-treated cancer cells (C). Nuclear translocation of p38MAPK in Wi-A-treated cells as confirmed by confocal laser scanning microscopy (D) and Western blotting (E) with phosphorylation-specific antibody. Effect of Wi-A on DNA Binding Activity of NFκB Dimer In view of the nuclear translocation and activation of NFκB in Wi-A-treated cells, we next examined the effect of Wi-A on DNA-binding ability of NFκB. DNA molecule was removed from X-ray determined structure of NFκB:DNA complex, and Wi-A was docked around amino acid residues R33, R35, Y36, E39, and Y57 of p65 subunit and R354, R356, Y357, E360, H364, and K541 of p50 subunit, which were involved in hydrogen bond interactions with DNA molecule. The docking analysis revealed that Wi-A interacted with NFκB dimer near DNA-binding site with binding energy of −4.129 kcal/mol. NFκB:Wi-A docked complex was simulated for 50 ns to check its stability and to observe changes in structure of the protein in the presence of the ligand. Molecular dynamic simulation revealed slight shift in the position of Wi-A. It, however, formed hydrogen bond with R354 and hydrophobic interactions with Y357, C359, and E360 among which R354, Y357, and E360 were originally involved in hydrogen bond interaction with DNA molecule (Figure 4A). To study the effect of Wi-A on NFκB dimer–DNA interactions, NFκB dimer:DNA:Wi-A complex was obtained by superimposition of docked complex over NFκB dimer:DNA complex. This complex was simulated for 50 ns, and the effect of DNA on NFκB–Wi-A interactions was examined. It was observed that Wi-A no longer formed any hydrogen bond interaction rather formed hydrophobic interactions with p50 subunit’s amino acid residues R354, Y357, C359, E360, S540, and K541, in addition to deoxycytidine of DNA molecule (Figure 4B). These data strongly suggested that Wi-A did not hinder the binding of DNA molecule with NFκB dimer. It, instead, was found to interact with the DNA and protein at their interaction site. Comparison of DNA binding in the presence and absence of Wi-A also revealed the differences in binding mode of DNA with protein dimer (Figure 4C). Figure 4. View largeDownload slide Binding of withaferin-A (Wi-A) with nuclear factor kappa B (NFκB) dimer (p65–p50) complex and its effect on structural stability. Interactions of Wi-A with NFκB dimer complex after molecular docking and simulation (A). Interactions of Wi-A with NFκB dimer–DNA complex (B). Superimposition of NFκB dimer–DNA complex in the presence (pink) and absence (blue) of Wi-A revealed the alterations caused by Wi-A in the confirmation of protein–DNA complex and the clash between DNA and Wi-A interaction site (C). Root mean square deviation (RMSD) fluctuations of NFκB dimer–DNA complex during 40- and 50-ns simulation of NFκB dimer–DNA complex (red) and NFκB dimer–DNA–Wi-A complex (purple), respectively (D). NFκB–DNA binding activity (E) and immunostaining and Western blotting for cyclin D1 and cyclin E in control and Wi-A-treated cancer and normal cells showed downregulation of cyclin D1 (F) and cyclin E cancer cells only (G and H). Normal cells showed downregulation of cyclin E, but not of cyclin D1. p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 4. View largeDownload slide Binding of withaferin-A (Wi-A) with nuclear factor kappa B (NFκB) dimer (p65–p50) complex and its effect on structural stability. Interactions of Wi-A with NFκB dimer complex after molecular docking and simulation (A). Interactions of Wi-A with NFκB dimer–DNA complex (B). Superimposition of NFκB dimer–DNA complex in the presence (pink) and absence (blue) of Wi-A revealed the alterations caused by Wi-A in the confirmation of protein–DNA complex and the clash between DNA and Wi-A interaction site (C). Root mean square deviation (RMSD) fluctuations of NFκB dimer–DNA complex during 40- and 50-ns simulation of NFκB dimer–DNA complex (red) and NFκB dimer–DNA–Wi-A complex (purple), respectively (D). NFκB–DNA binding activity (E) and immunostaining and Western blotting for cyclin D1 and cyclin E in control and Wi-A-treated cancer and normal cells showed downregulation of cyclin D1 (F) and cyclin E cancer cells only (G and H). Normal cells showed downregulation of cyclin E, but not of cyclin D1. p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. We next determined the binding affinity of NFκB dimer to DNA both in the presence and in the absence of Wi-A using HADDOCK. Residues of protein and DNA molecules involved in hydrogen bond interactions were treated as anchoring point for protein–DNA docking and specified as active residues, whereas passive residues were automatically selected by HADDOCK around active residues. Amino acid residues R33, R35, Y36, E39, R41, S42, K123, and R246 of p65 subunit and R354, R356, Y357, E360, K374, K444, K445, K572, K575, R605, and Q606 of p50 subunit were assigned as active residues for NFκB dimer:DNA complex. Although amino acid residues R50, T52, T55, T57, and R246 of p65 and C359, K444, K503, L507, S508, D539, K541, A545, R605, and R633 of p50 subunit were treated as active residues for NFκB dimer:DNA:Wi-A complex. Protein–DNA docking results revealed that the DNA binding affinity of NFκB dimer was −154.9 ± 2.9 that was reduced to −112.1 ± 4.7 in the presence of Wi-A. This reduction in binding affinity of DNA with NFκB dimer:Wi-A complex showed that Wi-A may result in decreased affinity of NFκB dimer. RMSD of NFκB dimer:DNA complex in the presence and absence of Wi-A was computed to decipher the effect of Wi-A on stability of NFκB dimer:DNA complex (Figure 4D). Comparison of RMSDs in the presence and absence of Wi-A revealed that after 10 ns, the presence of Wi-A added instability to NFκB dimer:DNA complex and increases fluctuations ranging from 2–4 to 6–8 Å. These data suggested that in spite of the above-described NFκB activation in Wi-A-treated cells, it may lead to downregulation of its transcriptional activation function. To examine such effect, we performed enzyme-linked immunosorbent assay (ELISA) for DNA–NFκB (p65) binding. As shown in Figure 4E, we found that Wi-A resulted in 20% decrease in the DNA binding activity of NFκB (p65). Consistently, we found decrease in NFκB effectors, cyclin D1 and cyclin E, in Wi-A-treated cancer cells (Figure 4F and G). Interestingly, cyclin E, but not cyclin D1, showed decrease in normal cells also (Figure 4F–H). Wi-A Caused Upregulation of DDR in Cancer and Normal Cells To further clarify, we examined the level of CARF expression, an established important regulator of DDR and proliferation fate of cells (23,25). As shown in Figure 5A and B, we found an upregulation of CARF in Wi-A-treated cancer cells. In contrast to the activation of NFκB and p38MAPK, occurred selectively in cancer cells (Figures 1 and 3), the level of CARF was upregulated in normal cells in response to Wi-A treatment. In line with these data, both cancer and normal cells also showed increase in γH2AX foci and reactive oxygen species (Figure 5A) consequent to Wi-A treatment, suggesting that it may cause growth arrest/senescence both in cancer and normal cells. Analyses of p21WAF1 and p16INK4A proteins that regulate growth arrest and senescence in cells revealed increase in Wi-A-treated cancer and normal cells (Figure 5C and D). As expected, the downstream effector of these proteins CDK4 and CDK2 showed decrease that in turn decrease the phosphorylated pRB leading to cell cycle arrest (Figure 6A–D). Although normal cells did not show decrease in expression of cyclin D1 (Figure 4F) or CDK4 (Figure 6A and B) expression levels, phosphorylated pRB showed distinct decrease in Wi-A-treated cells (Figure 6A and B) suggesting decrease in kinase activity of cyclinD1–CDK4 complex. Indeed, induction of senescence was confirmed by β-galactosidase staining and HP1γ foci in Wi-A-treated both cancer and normal cells (Figure 6C and D). Of note, because H1299 cells lacked p53 protein (Supplementary Figure 2D), an increase in p21WAF1 was therefore attributed to p53-independent mechanism, such as increase in CARF, resulting in decreased levels of HDM2 and stabilization of p21WAF1 (23,26,41,42). Figure 5. View largeDownload slide Upregulation of DNA damage response in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CARF, γH2AX, and reactive oxygen species (ROS) revealed their upregulation in Wi-A treated both cancer and normal cells as detected by immunostaining (A) and Western blotting (B). p38 MAPK showed selective increase in cancer cells (B). Upregulation of p21WAF1 and p16INK4A was also observed both in cancer and normal cells (C and D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 5. View largeDownload slide Upregulation of DNA damage response in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CARF, γH2AX, and reactive oxygen species (ROS) revealed their upregulation in Wi-A treated both cancer and normal cells as detected by immunostaining (A) and Western blotting (B). p38 MAPK showed selective increase in cancer cells (B). Upregulation of p21WAF1 and p16INK4A was also observed both in cancer and normal cells (C and D). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Figure 6. View largeDownload slide Downregulation of CDK4, CDK2, and pRB phosphorylation in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CDK4, CDK2, and phosphorylated pRB showed their downregulation in Wi-A-treated cancer cells (A). Normal cells showed decrease in CDK2 and phosphorylated pRB (A). Western blotting showing decrease in CDK2 and phosphorylated pRB both in cancer and normal cells; CDK4 showed decrease in cancer cells only (B). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Induction of senescence in Wi-A-treated cancer and normal cells. Senescence-associated β-galactosidase staining (C) and HP1γ (D) foci appeared both in H1299 (p53−/−) and MRC5 (p53+/+) cells subsequent to Wi-A treatment. Schematic diagram showing the effect of Wi-A on DNA damage signaling in cancer and normal cells. Whereas p38 and nuclear factor kappa B (NFκB) axis (E) were selectively activated in cancer cells (shown by gray arrows), CARF (collaborator of ARF) and its downstream activators p53, p21WAF1, and p16INK4A were activated both in cancer and normal cells (shown by black arrow) resulting in induction of senescence. Figure 6. View largeDownload slide Downregulation of CDK4, CDK2, and pRB phosphorylation in withaferin-A (Wi-A)-treated cancer and normal cells. Immunostaining of CDK4, CDK2, and phosphorylated pRB showed their downregulation in Wi-A-treated cancer cells (A). Normal cells showed decrease in CDK2 and phosphorylated pRB (A). Western blotting showing decrease in CDK2 and phosphorylated pRB both in cancer and normal cells; CDK4 showed decrease in cancer cells only (B). p Values are indicated as *p < .05, **p < .01, ***p < .001 by unpaired t test. Induction of senescence in Wi-A-treated cancer and normal cells. Senescence-associated β-galactosidase staining (C) and HP1γ (D) foci appeared both in H1299 (p53−/−) and MRC5 (p53+/+) cells subsequent to Wi-A treatment. Schematic diagram showing the effect of Wi-A on DNA damage signaling in cancer and normal cells. Whereas p38 and nuclear factor kappa B (NFκB) axis (E) were selectively activated in cancer cells (shown by gray arrows), CARF (collaborator of ARF) and its downstream activators p53, p21WAF1, and p16INK4A were activated both in cancer and normal cells (shown by black arrow) resulting in induction of senescence. Discussion NFκB is a transcription factor that regulates several proteins involved in growth arrest and apoptosis (1). Constitutional activation of NFκB in cancer cells has been related to their anti-apoptosis and continued proliferation characteristics. It has been shown to regulate cyclin D1 and cyclin E that along with CDK2 and CDK4/6 cause phosphorylation of tumor suppressor pRB protein, essential for cell cycle progression. Some studies have reported that Wi-A might inhibit IκB phosphorylation and NFκB translocation into the nucleus (8,11,12,27). NF-kB has been shown to play a major role in cellular senescence (43), an innate tumor suppressor mechanism, triggered by either activation of oncogenes and/or DNA damage signaling (44), and inflammatory response involving changes in extracellular matrix proteins (45) also related to senescence-associated secretory phenotype (21). Considering that Wi-A is capable of binding to multiple sites, we examined its effect on the conformation of IKKβ–Nemo complex. Contrary to the reports on its effect on local domains (28), we found that Wi-A neither disrupts interaction between IKKβ and Nemo chains nor inhibits the formation of IKK complex. Furthermore, change in conformation of IKKβ–Nemo in the presence of Wi-A was associated with increased level of IκB phosphorylation and NFκB translocation to the nucleus suggesting that Wi-A caused activation of IκB–NFκB signaling. Indeed, it was supported by the experimental data (Figures 1 and 3). Bioinformatics analysis on the effect of Wi-A on binding of NFκB dimer to DNA showed that it stably interacted with both DNA and protein at the DNA-binding site. However, it caused some instability in the NFκB–DNA interactions and was endorsed by experiments on the DNA-binding-based ELISA for transcriptional activity (Figure 4E). Analyses of NFκB downstream effectors such as cyclin D1, cyclin E, CDK2, and CDK4 in control and Wi-A-treated cells indeed revealed their lower level of expression in Wi-A treated cells when compared with control cancer cells. Interestingly, decrease in cyclin D1 and CDK4, but not cyclin E and CDK2, was selective in cancer cells suggesting that Wi-A therapy may be particularly beneficial for cancers with constitutively active cyclin D1 and CDK4. Decrease in CDK2 and CDK4 expression and activities (as determined by decrease in phosphorylated pRB) was assigned to increase in its upstream inhibitors (p21WAF1 and p16INK4A) that showed increase in Wi-A-treated cells. Such increase was assigned to activation of DNA damage and oxidative stress signaling as determined by increased level of expression of CARF (23–26,41,42) and γH2AX, suggesting induction of senescence. Induction of DNA damage and reactive oxygen species was observed and was supported by earlier studies (14,46–49). Senescence-associated β-galactosidase and HP1γ assays, indeed, endorsed Wi-A-induced senescence both in cancer and normal cells. The study demonstrated that Wi-A caused activation of DDR by pleiotropic mechanisms (Figure 6E). Although one axis of activated DNA damage signaling, p38MAPK–NFκB, was activated selectively in cancer cells, the other one, CARF-p21WAF1/p16INK4A–cyclin/CDK-pRB, was activated both in cancer and normal cells suggesting that CARF plays an important role in Wi-A-induced senescence. Funding This study was supported by grants from DBT (Government of India) and AIST (Japan). Conflict of Interest None reported. Acknowledgments Computations were performed at the Bioinformatics Centre at IIT Delhi. In silico analyses were performed by V.M. Experiments were performed by P.B., Y.L., and J.R. S.C.K., D.S., and R.W. conceived and designed the experiments and contributed reagents/materials/analysis tools for this study. P.B., V.M., S.C.K., D.S., and R.W. wrote the article. P.B. and V.M. contributed equally as co-first authors. D.S. and R.W. contributed equally as co-senior (corresponding) authors. References 1. Cartwright T , Perkins ND , L Wilson C . NFKB1: a suppressor of inflammation, ageing and cancer . FEBS J . 2016 ; 283 : 1812 – 1822 . doi: 10.1111/febs.13627 Google Scholar Crossref Search ADS PubMed 2. Dolcet X , Llobet D , Pallares J , Matias-Guiu X . NF-kB in development and progression of human cancer . Virchows Arch . 2005 ; 446 : 475 – 482 . doi: 10.1007/s00428-005-1264-9 Google Scholar Crossref Search ADS PubMed 3. Yu JH , Kim H . Oxidative stress and cytokines in the pathogenesis of pancreatic cancer . J Cancer Prev . 2014 ; 19 : 97 – 102 . doi: 10.15430/JCP.2014.19.2.97 Google Scholar Crossref Search ADS PubMed 4. Hoesel B , Schmid JA . The complexity of NF-κB signaling in inflammation and cancer . Mol Cancer . 2013 ; 12 : 86 . doi: 10.1186/1476-4598-12-86 Google Scholar Crossref Search ADS PubMed 5. Viatour P , Merville MP , Bours V , Chariot A . Phosphorylation of NF-kappaB and IkappaB proteins: implications in cancer and inflammation . Trends Biochem Sci . 2005 ; 30 : 43 – 52 . doi: 10.1016/j.tibs.2004.11.009 Google Scholar Crossref Search ADS PubMed 6. Wan F , Lenardo MJ . Specification of DNA binding activity of NF-kappaB proteins . Cold Spring Harb Perspect Biol . 2009 ; 1 : a000067 . doi: 10.1101/cshperspect.a000067 Google Scholar Crossref Search ADS PubMed 7. Park MH , Hong JT . Roles of NF-kappaB in cancer and inflammatory diseases and their therapeutic approaches . Cells . 2016 ; 5 : 15 – 27 . doi: 10.3390/cells5020015 Google Scholar Crossref Search ADS 8. Heyninck K , Lahtela-Kakkonen M , Van der Veken P , Haegeman G , Vanden Berghe W . Withaferin A inhibits NF-kappaB activation by targeting cysteine 179 in IKKβ . Biochem Pharmacol . 2014 ; 91 : 501 – 509 . doi: 10.1016/j.bcp.2014.08.004 Google Scholar Crossref Search ADS PubMed 9. Jackson SS , Oberley C , Hooper CP et al. Withaferin A disrupts ubiquitin-based NEMO reorganization induced by canonical NF-κB signaling . Exp Cell Res . 2015 ; 331 : 58 – 72 . doi: 10.1016/j.yexcr.2014.09.034 Google Scholar Crossref Search ADS PubMed 10. Syam S , Bustamam A , Abdullah R et al. β Mangostin suppress LPS-induced inflammatory response in RAW 264.7 macrophages in vitro and carrageenan-induced peritonitis in vivo . J Ethnopharmacol . 2014 ; 153 : 435 – 445 . doi: 10.1016/j.jep.2014.02.051 Google Scholar Crossref Search ADS PubMed 11. Maitra R , Porter MA , Huang S , Gilmour BP . Inhibition of NFkappaB by the natural product withaferin A in cellular models of cystic fibrosis inflammation . J Inflamm (Lond) . 2009 ; 6 : 15 . doi: 10.1186/1476-9255-6-15 Google Scholar Crossref Search ADS PubMed 12. Kaileh M , Vanden Berghe W , Heyerick A et al. Withaferin a strongly elicits IkappaB kinase beta hyperphosphorylation concomitant with potent inhibition of its kinase activity . J Biol Chem . 2007 ; 282 : 4253 – 4264 . doi: 10.1074/jbc.M606728200 Google Scholar Crossref Search ADS PubMed 13. Ghosh K , De S , Das S , Mukherjee S , Sengupta Bandyopadhyay S . Withaferin a induces ROS-mediated paraptosis in human breast cancer cell-lines MCF-7 and MDA-MB-231 . PLoS One . 2016 ; 11 : e0168488 . doi: 10.1371/journal.pone.0168488 Google Scholar Crossref Search ADS PubMed 14. Malik F , Kumar A , Bhushan S et al. Reactive oxygen species generation and mitochondrial dysfunction in the apoptotic cell death of human myeloid leukemia HL-60 cells by a dietary compound withaferin A with concomitant protection by N-acetyl cysteine . Apoptosis . 2007 ; 12 : 2115 – 2133 . doi: 10.1007/s10495-007-0129-x Google Scholar Crossref Search ADS PubMed 15. O’Connor MJ . Targeting the DNA damage response in cancer . Mol Cell . 2015 ; 60 : 547 – 560 . doi: 10.1016/j.molcel.2015.10.040 Google Scholar Crossref Search ADS PubMed 16. Olivieri F , Albertini MC , Orciani M , Ceka A , Cricca M , Procopio AD et al. DNA damage response (DDR) and senescence: shuttled inflamma-miRNAs on the stage of inflamm-aging . Oncotarget . 2015 ; 6 : 35509 – 35521 . doi: 10.18632/oncotarget.5899 Google Scholar PubMed 17. von Zglinicki T , Saretzki G , Ladhoff J , d’Adda di Fagagna F , Jackson SP . Human cell senescence as a DNA damage response . Mech Ageing Dev . 2005 ; 126 : 111 – 117 . doi: 10.1016/j.mad.2004.09.034 Google Scholar Crossref Search ADS PubMed 18. Wood CD , Thornton TM , Sabio G , Davis RA , Rincon M . Nuclear localization of p38 MAPK in response to DNA damage . Int J Biol Sci . 2009 ; 5 : 428 – 437 . doi: 10.7150/ijbs.5.428 Google Scholar Crossref Search ADS PubMed 19. Borodkina A , Shatrova A , Abushik P , Nikolsky N , Burova E . Interaction between ROS dependent DNA damage, mitochondria and p38 MAPK underlies senescence of human adult stem cells . Aging (Albany, NY) . 2014 ; 6 : 481 – 495 . doi: 10.18632/aging.100673 Google Scholar Crossref Search ADS 20. Salminen A , Kauppinen A , Kaarniranta K . Emerging role of NF-κB signaling in the induction of senescence-associated secretory phenotype (SASP) . Cell Signal . 2012 ; 24 : 835 – 845 . doi: 10.1016/j.cellsig.2011.12.006 Google Scholar Crossref Search ADS PubMed 21. Freund A , Patil CK , Campisi J . p38MAPK is a novel DNA damage response-independent regulator of the senescence-associated secretory phenotype . EMBO J . 2011 ; 30 : 1536 – 1548 . doi: 10.1038/emboj.2011.69 Google Scholar Crossref Search ADS PubMed 22. Vaughan S , Jat PS . Deciphering the role of nuclear factor-κB in cellular senescence . Aging (Albany, NY) . 2011 ; 3 : 913 – 919 . doi: 10.18632/aging.100390 Google Scholar Crossref Search ADS 23. Cheung CT , Hasan MK , Widodo N , Kaul SC , Wadhwa R . CARF: an emerging regulator of p53 tumor suppressor and senescence pathway . Mech Ageing Dev . 2009 ; 130 : 18 – 23 . doi: 10.1016/j.mad.2008.05.002 Google Scholar Crossref Search ADS PubMed 24. Hasan K , Cheung C , Kaul Z et al. CARF is a vital dual regulator of cellular senescence and apoptosis . J Biol Chem . 2009 ; 284 : 1664 – 1672 . doi: 10.1074/jbc.M805778200 Google Scholar Crossref Search ADS PubMed 25. Cheung CT , Singh R , Kalra RS , Kaul SC , Wadhwa R . Collaborator of ARF (CARF) regulates proliferative fate of human cells by dose-dependent regulation of DNA damage signaling . J Biol Chem . 2014 ; 289 : 18258 – 18269 . doi: 10.1074/jbc.M114.547208 Google Scholar Crossref Search ADS PubMed 26. Singh R , Kalra RS , Hasan K et al. Molecular characterization of collaborator of ARF (CARF) as a DNA damage response and cell cycle checkpoint regulatory protein . Exp Cell Res . 2014 ; 322 : 324 – 334 . doi: 10.1016/j.yexcr.2014.01.022 Google Scholar Crossref Search ADS PubMed 27. McKenna MK , Gachuki BW , Alhakeem SS et al. Anti-cancer activity of withaferin A in B-cell lymphoma . Cancer Biol Ther . 2015 ; 16 : 1088 – 1098 . doi: 10.1080/15384047.2015.1046651 Google Scholar Crossref Search ADS PubMed 28. Grover A , Shandilya A , Punetha A , Bisaria VS , Sundar D . Inhibition of the NEMO/IKKβ association complex formation, a novel mechanism associated with the NF-κB activation suppression by Withania somnifera’s key metabolite withaferin A . BMC Genomics . 2010 ; 11 ( suppl 4 ): S25 . doi: 10.1186/1471-2164-11-S4-S25 Google Scholar Crossref Search ADS PubMed 29. Morris GM , Huey R , Lindstrom W et al. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility . J Comput Chem . 2009 ; 30 : 2785 – 2791 . doi: 10.1002/jcc.21256 Google Scholar Crossref Search ADS PubMed 30. de Vries SJ , van Dijk M , Bonvin AM . The HADDOCK web server for data-driven biomolecular docking . Nat Protoc . 2010 ; 5 : 883 – 897 . doi: 10.1038/nprot.2010.32 Google Scholar Crossref Search ADS PubMed 31. Schrödinger Suite 2016-3: Protein Preparation Wizard; Epik version 3.3; Impact version 6.8; Prime version 4.1; LigPrep; Glide . New York, NY : Schrödinger, LLC ; 2016 . 32. Friesner RA , Murphy RB , Repasky MP et al. Extra precision glide: docking and scoring incorporating a model of hydrophobic enclosure for protein-ligand complexes . J Med Chem . 2006 ; 49 : 6177 – 6196 . doi: 10.1021/jm051256o Google Scholar Crossref Search ADS PubMed 33. Case D , Darden T , Cheatham T III et al. AMBER 12 . San Francisco, CA : University of California ; 2012 . 34. Humphrey W , Dalke A , Schulten K . VMD: visual molecular dynamics . J Mol Graph . 1996 ; 14 : 33 – 38 . doi: 10.1016/0263-7855(96)00018-5 Google Scholar Crossref Search ADS PubMed 35. DeLano WL. PyMOL: The PyMOL Molecular Graphics System, Version 1.3 . Schrödinger, LLC ; 2002 . 36. Laskowski RA , Swindells MB . LigPlot+: multiple ligand-protein interaction diagrams for drug discovery . J Chem Inf Model . 2011 ; 51 : 2778 – 2786 . doi: 10.1021/ci200227u Google Scholar Crossref Search ADS PubMed 37. Rushe M , Silvian L , Bixler S et al. Structure of a NEMO/IKK-associating domain reveals architecture of the interaction site . Structure . 2008 ; 16 : 798 – 808 . doi: 10.1016/j.str.2008.02.012 Google Scholar Crossref Search ADS PubMed 38. Marienfeld RB , Palkowitsch L , Ghosh S . Dimerization of the I kappa B kinase-binding domain of NEMO is required for tumor necrosis factor alpha-induced NF-kappa B activity . Mol Cell Biol . 2006 ; 26 : 9209 – 9219 . doi: 10.1128/MCB.00478-06 Google Scholar Crossref Search ADS PubMed 39. Herlaar E , Brown Z . p38 MAPK signalling cascades in inflammatory disease . Mol Med Today . 1999 ; 5 : 439 – 447 . doi: 10.1016/S1357-4310(99)01544-0 Google Scholar Crossref Search ADS PubMed 40. Olson CM , Hedrick MN , Izadi H , Bates TC , Olivera ER , Anguita J . p38 mitogen-activated protein kinase controls NF-kappaB transcriptional activation and tumor necrosis factor alpha production through RelA phosphorylation mediated by mitogen- and stress-activated protein kinase 1 in response to Borrelia burgdorferi antigens . Infect Immun . 2007 ; 75 : 270 – 277 . doi: 10.1128/IAI.01412-06 Google Scholar Crossref Search ADS PubMed 41. Hasan MK , Yaguchi T , Harada JI , Hirano T , Wadhwa R , Kaul SC . CARF (collaborator of ARF) interacts with HDM2: evidence for a novel regulatory feedback regulation of CARF-p53-HDM2-p21WAF1 pathway . Int J Oncol . 2008 ; 32 : 663 – 671 . doi: 10.3892/ijo.32.3.663 Google Scholar PubMed 42. Kalra RS , Cheung CT , Chaudhary A , Prakash J , Kaul SC , Wadhwa R . CARF (collaborator of ARF) overexpression in p53-deficient cells promotes carcinogenesis . Mol Oncol . 2015 ; 9 : 1877 – 1889 . doi: 10.1016/j.molonc.2015.07.003 Google Scholar Crossref Search ADS PubMed 43. Lee YK , Cha HJ , Hong M , Yoon Y , Lee H , An S . Role of NF-κB-p53 crosstalk in ultraviolet A-induced cell death and G1 arrest in human dermal fibroblasts . Arch Dermatol Res . 2012 ; 304 : 73 – 79 . doi: 10.1007/s00403-011-1176-2 Google Scholar Crossref Search ADS PubMed 44. Schmitt CA , Fridman JS , Yang M et al. A senescence program controlled by p53 and p16INK4a contributes to the outcome of cancer therapy . Cell . 2002 ; 109 : 335 – 346 . doi: 10.1016/S0092-8674(02)00734-1 Google Scholar Crossref Search ADS PubMed 45. Freund A , Orjalo AV , Desprez PY , Campisi J . Inflammatory networks during cellular senescence: causes and consequences . Trends Mol Med . 2010 ; 16 : 238 – 246 . doi: 10.1016/j.molmed.2010.03.003 Google Scholar Crossref Search ADS PubMed 46. Widodo N , Priyandoko D , Shah N , Wadhwa R , Kaul SC . Selective killing of cancer cells by Ashwagandha leaf extract and its component withanone involves ROS signaling . PLoS One . 2010 ; 5 : e13536 . doi: 10.1371/journal.pone.0013536 Google Scholar Crossref Search ADS PubMed 47. Kakar SS , Jala VR , Fong MY . Synergistic cytotoxic action of cisplatin and withaferin A on ovarian cancer cell lines . Biochem Biophys Res Commun . 2012 ; 423 : 819 – 825 . doi: 10.1016/j.bbrc.2012.06.047 Google Scholar Crossref Search ADS PubMed 48. Fong MY , Jin S , Rane M , Singh RK , Gupta R , Kakar SS . Withaferin A synergizes the therapeutic effect of doxorubicin through ROS-mediated autophagy in ovarian cancer . PLoS One . 2012 ; 7 : e42265 . doi: 10.1371/journal.pone.0042265 Google Scholar Crossref Search ADS PubMed 49. Yu Y , Katiyar SP , Sundar D et al. Withaferin-A kills cancer cells with and without telomerase: chemical, computational and experimental evidences . Cell Death Dis . 2017 ; 8 : e2755 . doi: 10.1038/cddis.2017.33 Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

The Journals of Gerontology Series A: Biomedical Sciences and Medical SciencesOxford University Press

Published: Jan 16, 2019

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off