Abstract
Abstract 691
Autophagy is an active process of intracellular degradation that occurs via sequestration of intracytoplasmic components (proteins, organelles) inside double membrane vesicles termed “autophagosomes”, followed by enzymatic degradation upon fusion with lysosomes. Autophagy is an adaptive process induced by cellular starvation but also contributes to development, aging, and pathogenesis of infections, neurodegenerative disease, atherosclerosis and cancer. In cancer, autophagy can function either as a cell survival signal or death response (i.e. type II cell death). To date, autophagy has not been characterized in chronic lymphocytic leukemia (CLL). The aim of our study is to investigate the presence and the role of autophagy in CLL and the potential contribution it may have to cell death or drug resistance with common therapeutics used in CLL. Our initial studies demonstrated that the essential cellular machinery for autophagy was present. CLL cells have a basal low level of autophagy when cultured in standard media that is actively promoted by cellular starvation or treatment with rapamycin. Given the potential contribution of autophagy in the death process or mechanism of drug resistance, we examined the influence of different therapeutics currently approved for CLL including chlorambucil, fludarabine, rituximab and dexamethasone. Cells were exposed to the drugs for four hours and examined by immunofluorescent staining of endogenous LC3 protein, an established marker of autophagy that becomes conjugated with phosphatydylethanolamine and recruited to the autophagosome membrane. Similar to rapamycin, fludarabine promotes autophagosome accumulation in all CLL samples tested, whereas dexamethasone caused this effect only in a subset. In contrast, no autophagosome accumulation was observed with chlorambucil or rituximab plus cross-linking.To determine the influence of this autophagic response on CLL cell survival, we next examined if chloroquine, which inhibits autophagy via preventing fusion of autophagosomes with lysosomes, affected fludarabine-mediated cell death at concentrations attainable in patients. Although chloroquine inhibited autophagy under these conditions as noted by confocal microscopy showing lack of co-localization between LC3 (autophagosome marker) and LAMP-2 (lysosome marker), it had no effect on fludarabine-mediated cell death. Subsequent studies showed that investigational agents including inhibitors of PI3-kinase, HSP-90, and cyclin-dependent kinases (CDK) each induced autophagy, while inhibitors of histone deacetylases (HDACs) did not. Notably, chloroquine enhanced cytotoxicity mediated by the CDK inhibitor flavopiridol in all patient samples examined, while it produced no effect on the cytotoxicity of the remaining agents. This sensitization to flavopiridol was most striking in CLL samples with the greatest viability after four hours of in vitro treatment with flavopiridol, as assessed by annexin/PI flow cytometry. Given that endoplasmic reticulum (ER) stress induces autophagy in normal cells, we examined this pathway in flavopiridol-treated CLL patient cells. These studies demonstrated that in a subset of samples, flavopiridol, but not fludarabine, promotes early intracellular ER-derived calcium flux, concomitant with the appearance of ER stress evidenced by quantitative real-time PCR showing increased gene expression of specific markers (XBP1, IRE1, Grp78) and standard PCR showing XBP1 splicing. To determine if this finding was relevant to the in vivo setting, we subsequently examined samples obtained serially from CLL patients during treatment with flavopiridol as part of two completed clinical trials at our institution. Similar to our in vitro studies, we found that flavopiridol actively induces autophagy in vivo through an ER stress-directed pathway. Collectively, our data demonstrate that autophagy is relevant to CLL biology and may serve as a pharmacodynamic marker of targeted therapy in CLL. Furthermore, induction of autophagy appears to contribute to flavopiridol resistance in CLL, whereas its role with other therapeutics is unclear.
This work is supported by the Leukemia and Lymphoma Society, the D. Warren Brown Foundation, and the National Cancer Institute (CLL Research Consortium and OSU Leukemia SPORE).
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.