Abstract
Chronic lymphocytic leukemia (CLL) is a mature B cell malignancy and is the most prevalent type of leukemia in adults. There is no curative therapy for this disease; however, several new agents have shown very promising results. Autophagy has not been studied in CLL and in this study we first sought to determine if autophagy was functional in CLL with classic inducers, and if this contributes to direct cytotoxicity or protection from cell death. While autophagy is activated with all classic stimuli of this process, only unfolded protein endoplasmic reticulum (ER) stress-mediated autophagy protects from cell death. Interestingly, select therapeutic agents (fludarabine, GS-1101, flavopiridol), which are active in CLL, also induce autophagy. Of interest, only the broad cyclin-dependent kinase inhibitor flavopiridol has improved efficacy when autophagy is antagonized biochemically (chloroquine) or by siRNA. This promoted an investigation which demonstrated unexpectedly that flavopiridol mediates ER stress and downstream activation of MAP3K5/ASK1, which ultimately is responsible for cell death. Similarly, autophagy activated in part via ER stress and also CDK5 inhibition is protective against cell death induced by this process. Collectively, our studies demonstrate that in CLL, autophagy is induced by multiple stimuli but only acts as a mechanism of resistance against ER stress-mediating agents. Similarly, flavopiridol mediates ER stress as a primary mechanism of action in CLL, and autophagy serves as a mechanism of resistance to this agent.
Chronic lymphocytic leukemia is an incurable B-cell malignancy, despite the numerous undergoing investigations focused on identifying new therapeutic approaches for this disease. Flavopiridol is a CDK inhibitor, which shows significant activity as a therapeutic agent in CLL, although the mechanism of action is still unclear. Previous work performed by our group suggested that flavopiridol causes changes indicative of autophagy in CLL patient cells. Given that autophagy has been described in the literature as an intracellular process that can act as a survival strategy or a cell death mechanism depending upon cell type and circumstances, we performed for the first time a detailed analysis of this pathway in CLL cells. We investigated the autophagy machinery in CLL by testing classic autophagy inducers (starvation and rapamycin), an ER stress inducer (thapsigargin), the CLL therapeutic agents fludarabine and GS-1101, and the CDK inhibitor flavopiridol. Analyses of LC3 cytoplasmic accumulation by immunofluorescence and immunoblot and of SQSTM1/p62 expression by immunoblot demonstrate significant autophagy induction in the presence of flavopiridol, thapsigargin, and fludarabine that is similar to rapamycin and starvation. Supportive of our in vitro experiments, we observed a significant induction of autophagy in serial pharmacodynamic samples obtained from CLL patients treated with flavopiridol. To further understand the role of autophagy in this context, we inhibited the process using chloroquine (an antimalarial drug that blocks late steps of autophagy by preventing autophagosome-lysosome fusion and enzymatic digestion of autophagosomal cargo), or siRNA directed against ATG5 and ATG7, two essential genes in autophagy initiation. Chloroquine dosages used in our experiments are equivalent to plasma concentrations attained in malaria patients.
Interestingly, when we tested CLL viability in the presence of chloroquine or siRNA directed against ATG5 or ATG7, we observed a significant increase in cytotoxicity for the ER stress inducer thapsigargin and the CDK inhibitor flavopiridol, but not for the others. This observation led us to hypothesize that flavopiridol might also induce ER stress similar to thapsigargin, and that autophagy caused by ER stress acts as a mechanism of drug resistance in CLL cells.
To test ER stress in CLL cells exposed to flavopiridol, we first measured gene expression for XBP1, HSPA5/GRP78, ERN1/IRE1 and DDIT3/CHOP using RT-PCR. We observed a significant increase of these markers in CLL cells treated in vitro with flavopiridol as well as in serial samples collected from CLL patients treated with flavopiridol. Further testing of this pathway revealed that the ER stress response is defective in CLL cells, as suggested by the absence of EIF2AK3/PERK activation and XBP1 splicing. However, we observed that CLL cells exposed to flavopiridol show ERN1-TRAF2-MAP3K5 complex formation by co-immunoprecipitation, ATF6 nuclear localization by immunofluorescence and ATF6 binding to the HSPA5 promoter region by chromatin immunoprecipitation assay. Investigation of events downstream from ERN1-TRAF2-MAP3K5 complex formation revealed induction of MAPK8/JNK and MAPK14/p38α phosphorylation and CASP4 activation in the in vivo and the in vitro samples. MAP3K5 siRNA knockdown and CASP4 inhibition antagonize flavopiridol-induced cell death. Another interesting point is that CDK5, which is inhibited by flavopiridol, inhibits autophagy by phosphorylating PIK3C3/VPS34, an essential initiator of this process. Knockdown of CDK5 in our system results in both induction of autophagy and increase in ER stress marker genes.
Collectively, these findings suggest that, at least partly, flavopiridol-induced cell death in CLL cells is due to ER stress through MAP3K5 and CASP4 activation and that the activation of ER stress response leads also to induction of autophagy as a mechanism of resistance. Therefore, therapeutics such as chloroquine that block the autophagic process may convert a case resistant to an ER stress-inducing therapeutic agent to a responsive one. Moreover, the results presented here open the possibility to further explore combinatorial therapeutic strategies using ER stress inducer agents and autophagy inhibitors to improve the therapeutic benefit not just in CLL but also in other types of cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.