Fibulins (FBLNs), a family of extracellular matrix proteins, have recently been shown to act as tumor suppressors or activators in different cancers, and the underlying molecular mechanisms of their action in cancer remain unclear. We have previously shown that the expression of FBLN3 is suppressed by promoter hypermethylation and is associated with invasiveness in aggressive non-small cell lung cancer. In this study, we evaluated the roles and signaling mechanism of FBLN3 in lung cancer stem cells (CSCs). Forced expression of FBLN3 suppressed invasion and migration of lung adenocarcinoma cells and decreased the expression of epithelial-to-mesenchymal transition (EMT) activators, including N-cadherin and Snail. Stemness activities of lung adenocarcinoma cells were also suppressed by FBLN3 as indicated by a decrease in spheroid formation and the levels of stemness markers such as Sox2 and β-catenin. These effects of FBLN3 were mediated by the glycogen synthase kinase-3β, GSK3β/β-catenin pathway, and the upstream regulators of GSK3β, including phosphoinositide 3-kinase (PI3K)/AKT and insulin-like growth factor-1 receptor (IGF1R), were inactivated by FBLN3. Moreover, IGF1R was shown to be a direct target of FBLN3, which competitively inhibited insulin-like growth factor (IGF) action. To confirm the effect of FBLN3 on lung CSCs, aldehyde dehydrogenase-positive (ALDH+) A549 lung CSCs were sorted and treated with recombinant FBLN3 protein. FBLN3 clearly suppressed EMT, stemness activity and the over-activated IGF1R/PI3K/AKT/GSK3β pathway of the ALDH+ CSC subpopulation. In addition, injection of recombinant FBLN3 protein around subcutaneous xenografts established with ALDH+ CSCs in athymic nude mice significantly suppressed tumor growth and progression. Overall, our results show that FBLN3 suppresses both EMT and self-renewal of the lung CSCs by modulating the IGF1R/PI3K/AKT/GSK3β pathway and that FBLN3 would be useful as an alternative CSC therapy.

No abstract

Recent studies indicate that reactive oxygen species, such as H2O2, can be generated by anti-cancer drugs, can damage cells, and then induce apoptotic cell death. In this study, we reported whether polyamines were capable of affecting apoptotic cell death triggered by H2O2 in leukemia cells or not. Alpha-difluoromethylornithine treatment (DFMO, 3 mmol/L, 48 h), which depletes intracellular putrescine by inhibiting ornithine decarboxylase, reduced H2O2-induced cell death in the HL-60 leukemia cells. Cytotoxicity caused by H2O2 in putrescine-depleted cells was 50% lower than that in the control cells, as determined by propidium iodide, the annexin V and DNA fragmentation assays. Following putrescine (1 mmol/L) supplement, cell death induction caused by H2O2 was restored to a similar level as the DFMO-untreated control cells. It seems that this partly resulted from the intralysosomal iron-dependent oxidation of the cells because DFMO did not significantly affect the increment of enzymes related to oxidative-stress resistance. Putrescine depletion by DFMO treatment reduced the cellular iron uptake of the cells by about 70%. In parallel to the reduction of iron uptake, lysosomal damage (assayed by acridine orange relocalization or uptake test) in the DFMO-treated cells was far less than that in the control cells. Moreover, putrescine supplement also restored the iron uptake to the control cell levels. Pre-incubation with desferrioxamine (DFO), which chelates iron and forms a non-reactive Fe-DFO complex that is localized in the lysosomal compartment, inhibited H2O2-induced cell death. This work suggests that polyamines may play a critical role in apoptotic cell death triggered by H2O2 via the regulation of the iron-dependent instability of the lysosome.

In previous studies, polyamine depletion by DFMO (alpha-difluoromethylornithine)-treatment reduced H(2)O(2)-induced apoptotic cell death by reduction of ferric ion uptake. In the present study, we analyzed the reduction of radiation-induced cell death by polyamine depletion. Exposure of HT29 cells to radiation induced severe cell death, but when cells were pretreated with DFMO, a specific inhibitor of polyamine biosynthesis, radiation-induced cell death was reduced to 50-60% of control. Cell cycle analysis showed that, in these cells, the time to reach the G(2)/M phase arrest was delayed for 20-24 h compared to the control cells, at which stage the fate of cells exposed to ionizing radiation is determined. DFMO-treated cells also showed a low level of thioredoxin, which is a high-level determinant of the cellular fate. To investigate the relationship between the G(2)/M phase arrest and the reduction of thioredoxin caused by polyamine depletion, we also analyzed thioredoxin-antisensed (asTRX) HT29 cells as for DFMO-treated cells. In asTRX-transfected cells, the gamma-irradiation-induced G(2)/M phase arrest was also significantly delayed and radiation-induced cell death was profoundly reduced, as in the DFMO-treated cells. Both sets of cells showed a decrease of cyclin D1 and an increment of HSP25, which are involved in radiation-induced cell cycle progress. Overall, these results suggest that polyamines are essential for normal cell death of HT29 cells triggered by gamma-radiation and that this is partially mediated by the regulation of thioredoxin expression.