| ObjectiveCancer is one of the major diseases in China and chemotherapy, as an important component of cancer treatment, has been wildly used in cancer treatment. Due to in recent years the success of application the small molecular tyrosine kinase inhibitors on cancer therapeutics, the discovery and development of anti-cancer drugs have been recognized as an hot field in drug discovery. Extensive studies have indicated that ROS are involved in anticancer drug-caused apoptosis. And increasing evidence has suggested the overexpression of antioxidants-associated enzyme in cancer cells mediate the drug resistance. Reactive oxygen species (ROS) are involved in a variety of physiological and pathological processes. Low levels of ROS regulate cellular signal transduction and play important roles in normal cell proliferation. High levels of ROS lead to apoptosis and necrosis. Evidence has generally suggested that ROS act as second messengers that are required for the downstream signaling effects. Similar to the regulation of protein function by phosphorylation, oxidation of cysteine residues by ROS results in conformational, structural, and direct catalytic consequences on targeted signaling proteins. However, the observation of these modifications in response to pharmacologically relevant signaling stimuli has been proven less. The cancer chemopreventive and therapeutic retinoid4-HPR (N-(4-hydroxyphenyl) retinamide,4-HPR), as a synthetic retinoid, has widely been studied in cancer prevention. Although several phase I and II clinical trails have suggested that4-HPR exhibit the potential anti-tumor activity against different types of cancer, the result from phase III clinical trail is still very controversial. Since generation of reactive oxygen species (ROS) have been implicated in4-HPR-induced apoptosis, the mechanism study of4-HPR on anti-tumor activity, especially focusing on the protein posttranslational modification, will greatly expand our knowledge to identify the potential biomarker for4-HPR therapeutic and the novel target for cancer treatment. The aim of this study was to investigate the involvement of protein posttranslational modification in4-HPR-mediated cell death. The current study was divided into two sections:1) ROS-driven Akt dephosphorylation at Ser-473is involved in4-HPR-mediated apoptosis in NB4cells.2) The oxidation states of DJ-1dictate the cell fate in response to oxidative stress triggered by4-HPR.Methods1) Cell apoptosis were detected by flow cytometry after staining with Annexin V/PI.The change of mitochondrial membrane potential were detected by flow cytometry after staining with JC-1.Cell lysates were subjected to western-blot analysis for the detection of PARP, caspase3, caspase9, Bax, Bcl-2, Akt, p-Akt(Ser473and Thr308), GSK3β, p-GSK30,HSP90and PP2Ac protein level. Immunofluorescence was carried out to detect the colocalisation of Akt and HSP90or PP2Ac.Immunoprecipitation was employed to detect the combination of Akt with HSP90or PP2Ac.The protein disulfides were assessed using mercurochrome.The production of intracellular ROS was measured in the NB4cell line using the oxidation-sensitive fluorescent dye carboxy-DCFDA with flow cytometry.The redox states of Akt proteins were assessed by modifying free thiol with N-ethylmaleimide (NEM) with western-blot.The different redox states of Akt proteins with HSP90were detected by ELISA.The AKT protein disulfides were assessed using improved modifying free thiol with N-ethylmaleimide (NEM) with western-blot. The thiol antioxidant N-acetylcysteine (NAC), glutathione (GSH),dithiothreitol (DTT), and catalase-polyethylene glycol were used to unravel the relationship between4-HPR and Akt and detect the role of4-HPR in induced apoptosis.Mice blood glucose were measured to detect the inhibition effect of4-HPR on Akt in vivo.2) Cell apoptosis were detected by flow cytometry after staining with Annexin V-fluorescein isothiocyanate/PI.The change of mitochondrial membrane potential were detected by flow cytometry after staining with JC-1.Cell lysates were subjected to-western-blot analysisi for the detection of PARP, LC-3, DJ-1, ASK1, JNK, p-JNK, p38, p-p38, IRE1, eIF2a,p-eIF2α,ATF4and CHOP protein level.Immunofluorescence was carried out to detect the colocalisation of DJ-1and ASK1.Immunoprecipitation was employed to detect the combination of ASK1with DJ-1or Trx.The production of intracellular ROS was measured using the oxidation-sensitive fluorescent dye carboxy-DCFDA with flow cytometry.The production of autophagic vacuole were measured by AO staining,MDC staining and GFP-LC3B transfection with fluorescence microscope.The production of autophagic vacuole and apoptosis were measured by using fluorescence microscope.Tumors were established by injection into athymic mice and measure the anti-tumor effect of different dose of4-HPR.TUNEL staining was performed to detect apoptosis.SDS-PAGE and protein mass spectrometry were performed to analyse which proteins were combined with DJ-1.The redox states of DJ-1proteins were assessed by two-dimensional gel electrophoresis.the effect of4-HPR on DJ-1dimerization was measured by using cross-linking reagent DSS.The thiol antioxidant N-acetylcysteine (NAC) and glutathione (GSH) were used to detect the role of4-HPR in induced apoptosis and its effect on DJ-1. the role of4-HPR in apoptosis were measured after silencing DJ-1by performing RNAi technology.The redox states of DJ-1proteins were assessed by using Probe-1with fluorescence microscope.Detection of free sulfhydryl of proteins and GSH were based on Ellman assay.The mutant DJ-1plasmid was constructed by using site-specific mutagenesis and assessed its effect on4-HPR induced apoptosis. ResultsPart1ROS-driven Akt dephosphorylation at Ser-473is involved in anticancer drugs-mediated apoptosis in NB4cells1) Akt signaling pathway was involved in4-HPR-induced apoptosis.We found that4-HPR induced apoptosis in NB4cells in an time-and concentration-dependent manner.4-HPR caused an obvious decrease of ΔΨm m in NB4cells in a time-dependent manner. Exposure to4-HPR resulted in decrease amount of Bcl-2and no changes of Bax. These results suggested that the activation of caspase-3, caspase-9, and the cleavage of PARP were ultimately responsible for the apoptotic process imposed by4-HPR. In addition, mitochondrial pathway might be involved in4-HPR-caused apoptosis.In NB4cells, although the protein expression of Akt and its upstream regulator PI3K (p110) were not affected, the decreased amount of two phosphorylation of Akt were observed. Treatment with4-HPR resulted in an acute increase in blood glucose that retuned to baseline level. These results indicated that the Akt signaling pathway was involved in4-HPR-induced apoptosis. Furthermore, our data also implied that the attenuated expression of Akt’s phosphorylation at Ser-473caused by4-HPR was PI3K-independent.2) ROS caused by4-HPR were involved in apoptosis and changes of p-Akt.A significant generation of ROS was observed after0.5-6h exposure period compared with untreated cells. All the used antioxidants effectively attenuated the apoptosis in NB4cells treated with4-HPR. In addition, the thiol antioxidants NAC and GSH and the protein thiol protector DTT were showed to be sufficient to maintain the production of4-HPR-mediated ROS at the basic level. Moreover all the used antioxidants blocked4-HPR-induced downregulation of p-Akt (Ser-473) and p-GSK3β in a dose-dependent manner.3)4-HPR-induced ROS contributed to the formation of the intramolecular disulfide bond of AktBoth4-HPR and H2O2, significantly increased the formation of disulfide bond in NB4cells. Notably, the addition of NAC remarkably reversed4-HPR-induced formation of disulfide bond. In addition,4-HPR increased the amount of oxidized Akt in a time-dependent manner. The fluorescent microscopy and Western blotting data under reducing and nonreducing conditions implied that4-HPR-induced ROS mediated the formation of disulfide bond in Akt.4) The formation of intracellular disulfide bond in Akt inhibited the binding ability of Akt-Hsp90complex and increased Akt dephosphorylation catalyzed by PP2ANeither the expression change of PP2Ac nor Hsp90was observed in NB4cells treated with4-HPR. Moreover, both the immunoprecipitation analysis and immunofluorescence studies revealed that interaction between Akt and Hsp90was decreased; however, interaction between Akt and PP2Ac was enhanced in NB4cells after exposure to4-HPR. In addition, both binding abilities were abrogated by NAC. Hsp90prefers binding with reduced Akt rather than that of oxidized Akt. Together with previous data, these results suggested that the oxidized Akt may fail in formatting the Akt-Hsp90complex and increasing the dephosphorylation of Akt catalyzed by PP2A.Part2The oxidation states of DJ-1dictate the cell fate in response to oxidative stress triggered by anticancer drugs1)4-HPR-driven ROS induce the autophagy-apoptosis transition in a concentration threshold dependent manner both in vitro and in vivoIn the present study, the autophagy-apoptosis transition was also observed in several cancer cells. AO staining, MDC staining and GFP-LC3transfection assay were further performed to visualize the autophagosome formation in cells. Electron microscopic observations also revealed that4-HPR treatment accumulated autophagosomes in HeLa cells as numerous autophagic vacuole, empty vacuoles and secondary lysosomes was observed. All the results suggested that apoptosis can only be triggered by high concentrations of4-HPR, while autophagy can be induced by very low concentrations of4-HPR.To determine whether this transition can be triggered by4-HPR in vivo, we established a HeLa xenograft nude mouse model treated with two doses of4-HPR. The data indicate that4-HPR also induce a dose-threshold dependent autophagy-apoptosis transition in vivo. A significant generation of ROS was observed after exposure of4-HPR (5and10μM). Comparing to untreated cells, the production of ROS in lOμM4-HPR group was about twice more than5μM4-HPR group. We pretreated HeLa cells with ROS scavenger, NAC and GSH, followed by4-HPR exposure.2mM NAC and GSH almost totally blocked5μM4-HPR caused-autophagy as well as10μM4-HPR caused-apoptosis, but could not inhibit lOμM4-HPR caused-autophagy. In contrast,4mM NAC and GSH could partially inhibit lOμM4-HPR caused-autophagy when comparing to2mM NAC. These data suggest the pivotal roles of ROS in4-HPR triggered autophagy and apoptosis. Of interest is the factor that although both5and10μM4-HPR evoke the different intracellular oxidation status and decrease the intracellular GSH levels, and lOμM4-HPR was more powerful to induce the oxidation of free sulfhydryl groups on intracellular proteins. These resultsstrongly suggest that the oxidation of free sulfhydryl groups regulated by ROS may be the key factor involved in the transition of autophagy and apoptosis.2)4-HPR-driven ROS-regulated JNK1and p38signaling pathways are response for4-HPR induced autophagy and apoptosis respectively5and10μM4-HPR induced a sustained activation of JNK1in HeLa cells, but the activation of p38was only observed in10μM4-HPR treated cells. When JNK1were inhibited by specific siRNA or pharmacological inhibitor, the induction of LC3-II caused by4-HPR (5and10μM) was almost completely blocked, along with an increase of apoptosis. By contrast, when the HeLa cells were pretreated with either siRNA against p38or pharmacological inhibitor, the apoptosis were dramatically inhibited without affecting the autophagy upon4-HPR treatment. Moreover, when we used specific siRNA against ASK1,4-HPR-induced activation of JNK1as well as p38were greatly attenuated. Collectively, these data further demonstrate that the JNK1and p38signaling activated by ASK1are response for4-HPR-induced autophagy and apoptosis respectively.3) ASK1formed a complex with DJ-1under mild levels of ROS, but not under excessive levels of ROSWe applied co-immunoprecipitation with anti-ASK1antibody, followed by LC-MS/MS analysis to search for proteins that specifically bind to ASK1under different oxidant states. More than20proteins visible by silver staining,5unique bands at approximately70,40,30,25and20kDa were identified to show different trends among all three groups. These5bands were HSPA1B, HNRNPAB, RPS3A, NUDT21and PARK7. We further use the immunoprecipitation assay to validate these candidate proteins and found that DJ-1showed different interaction status with ASK1under control or4-HPR treatment. Thus, our results indicate that DJ-1may be conditionally associated with ASK1, which may relate to4-HPR caused oxidant states. As expected, NAC pretreatment could significantly modulate the interaction between DJ-1and ASK1. Consistent with the immunoprecipitation results, immunofluorescence results also confirmed that only under5μM4-HPR treatment, direct co-localization of ASKland DJ-1in the cytoplasm could be clearly demonstrated by merged immunofluorescence images.4) The oxidation states of DJ-1mediate the activation of ASK1by regulating DJ-1-ASK1complex formation and dictate the cellular response to4-HPR triggered ROSIsoelectric focusing experiment revealed that two different PI values of DJ-1were observed in cells treated with5or10μpM4-HPR respectively. Importantly, the PI value of DJ-1protein isolated from10μM4-HPR-treated cells was much lower than the protein isolated from5μM4-HPR-treated cells, indicating this PI shift may be caused by increasing intracellular oxidative stress induced by4-HPR. We next examined the effect of different concentrations of4-HPR on its dimer formation by a chemical crosslinking analysis, the results showed that the levels of dimer DJ-1observed in DSS-treated cells were only decreased in the presence of10μM4-HPR, indicating that only high concentration of4-HPR affect dimer formation of DJ-1. We further applied compound23, which directly binds to DJ-1and prevents DJ-1’s excess oxidation, DJ-1segregated from ASK1under10μM4-HPR treatment by immunoprecipitation assay, however, this dissociation was totally blocked by pretreatment of compound23, suggesting that only mild oxidant state of DJ-1can bind to ASK1and excessive oxidized form of DJ-1will dislodge from ASK1. Furthermore, pretreatment of compound23also blocked ASK1activated p38signaling under10μM4-HPR treatment, and attenuated the apoptosis-induction ability of4-HPR, without affecting the autophagy, indicating that ASKl-DJ-1binding is required for p38activation and apoptosis induction..We next generated two artificial DJ-1mutants at position106and the results implied the oxidation state of DJ-1Cys-106determine the autophagy and apoptosis induction caused by4-HPR. Once again, our data demonstrate that while mild oxidative state of DJ-1is closely correlated with cytoprotection process-autophagy, extensive oxidation of DJ-1is controlling the initiation of apoptosis.5) DJ-1depletion enhance the sensitivity of tumor cells to4-HPR in vitro and in vivoWe observed that the In vitro anti-proliferation activity of4-HPR (IC50value) was negatively correlated with the protein expression level of DJ-1among a panel of human cancer cell lines (R=0.815). Notably, DJ-1depletion increased4-HPR-induced apoptosis in5μM4-HPR treated HeLa cells. To test whether this enhancement effect could be reproduced in vivo, we evaluated antitumor activity of4-HPR in a xenograft nude mouse model generated by HeLa-shRNA-Con or HeLa-shRNA-DJ-1cells. The data indicate that although the expansion of DJ-1-depleted-xenograft is lower than control group, DJ-1depletion indeed enhance the sensitivity of tumor cells to4-HPR induced growth arrest and apoptosis both in vitro and in vivo.6) ER stress signaling participated in4-HPR-driven ROS-caused autophagyWe examined the IRE1and eIF2a signaling pathways, two branchs of the unfold protein response, and found that IRE1and eIF2a signalings mediate autophagy and apoptosis respectively. Activation of endogenous JNK1by4-HPR (5and10μM) was almost completely eliminated in the IRE1siRNA group, while p38has no change, suggesting that ASK1-TRAF2-IRE1complex is only required for4-HPR-triggereed activation of JNK1but not p38. The formation of ASK1-TRAF2-IRE1complex in the cells treated with5or10μM4-HPR was also confirmed by immunoprecipitation assay. Thus, our immunoprecipitation results demonstrate that both the DJ-1-ASK1complex and IRE1-TRAF2-ASK1complex are forming in the cells treated with5μM4-HPR, but possibly in a mutually exclusive manner, suggesting a competition manner for these two protein complexes.ConclusionIn present study we investigate the involvement of protein posttranslational modification in4-HPR-mediated cell death. We, for the first time, demonstrate (1) the potential involvement of ROS-mediated conformational change of Akt in4-HPR-induced apoptosis, this change, as a consequence, mediates the dephosphorylation of Akt via regulating Akt-Hsp90or Akt-PP2A complex formation;(2) the potential involvement of ROS-mediated oxidation state change of DJ-1in4-HPR-induced autophagy-apoptosis transition. These observations are meaningful for further understanding of the mechanism of anticancer drugs-driven ROS, and provide the direct evidence to support Akt and DJ-1as a potential biomarker or novel target for cancer therapy. |
PDF Full Text Request