No abstract

BackgroundAberrant activation of β-catenin and Yes-associated protein (YAP) signaling pathways has been associated with hepatocellular carcinoma (HCC) progression. The LIM domain protein Ajuba regulates β-catenin and YAP signaling and is implicated in tumorigenesis. However, roles and mechanism of Ajuba expression in HCC cells remain unclear. The E3 ligase Hakai has been shown to interact with other Ajuba family members and whether Hakai interacts and regulates Ajuba is unknown.MethodsHCC cell lines stably depleted of Ajuba or Hakai were established using lentiviruses expressing shRNAs against Ajuba or Hakai. The effects of Ajuba on HCC cells were determined by a number of cell-based analyses including anchorage-independent growth, three dimension cultures and trans-well invasion assay. In vivo tumor growth was determined in a xenograft model and Ajuba expression in tumor sections was examined by immunohistochemistry. Co-immunoprecipitation, confocal microscopy and immunoblot assay were used to examine the expression and interaction between Ajuba and Hakai.ResultsDepletion of Ajuba in HCC cells significantly enhanced anchorage-independent growth, invasion, the formation of spheroids and tumor growth in a xenograft model, suggesting a tumor suppressor function for Ajuba in HCC. Mechanistically, Ajuba depletion triggered E-cadherin loss and β-catenin translocation with increased Cyclin D1 levels. In addition, depletion of Ajuba upregulated the levels of YAP and its target gene CYR61. Furthermore, siRNA-mediated knockdown of either β-catenin or YAP attenuated the pro-tumor effects by Ajuba depletion on HCC cells. Notably, Ajuba stability in HCC cells was regulated by Hakai, an E3 ligase for E-cadherin. Hakai interacted with Ajuba via its HYB domain and induced Ajuba neddylation, which was antagonized by the neddylation inhibitor, MLN4924, but not MG132. We further show that overexpression of Hakai in HCC cells markedly increased anchorage-independent growth, spheroid-formation ability and tumor growth in xenografts whereas Hakai depletion resulted in these opposite effects, indicating an oncogenic role for Hakai in HCC. Hakai also induced β-catenin translocation with increased levels of Cyclin D1.ConclusionsOur data suggest a role for Ajuba and Hakai in HCC, and uncover the mechanism underlying the regulation of Ajuba stability.Electronic supplementary materialThe online version of this article (10.1186/s13046-018-0806-3) contains supplementary material, which is available to authorized users.

Matrine, an alkaloid compound isolated from the medicinal plant Sophora flavescens, inhibits many types of cancer proliferation. However, the precise mechanism of the matrine antihuman chronic myeloid leukemia remains unclear. In this study, we showed that matrine significantly inhibited the cell proliferation and induced apoptosis by regulating Warburg effect through controlling hexokinases 2 (HK2) expression in myeloid leukemia cells. Interestingly, matrine inhibited the expression of HK2 mediated by reduction in c-Myc binding to HK2 gene intron and led to downregulation of HK2, which upregulated proapoptotic protein Bad and then induced apoptosis. We further demonstrated that matrine could synergize with lonidamine, an inhibitor of HK2, for the treatment of myeloid leukemia, both in vitro and in vivo. Taken together, our findings reveal that matrine could promote human myeloid leukemia cells apoptosis via regulating Warburg effect by controlling HK2.

The tumor suppressor Spred2 (Sprouty-related EVH1 domain-2) induces cell death in a variety of cancers. However, the underlying mechanism remains to be elucidated. Here we show that Spred2 induces caspase-independent but autophagy-dependent cell death in human cervical carcinoma HeLa and lung cancer A549 cells. We demonstrate that ectopic Spred2 increased both the conversion of microtubule-associated protein 1 light chain 3 (LC3), GFP-LC3 puncta formation and p62/SQSTM1 degradation in A549 and HeLa cells. Conversely, knockdown of Spred2 in tumor cells inhibited upregulation of autophagosome maturation induced by the autophagy inducer Rapamycin, which could be reversed by the rescue Spred2. These data suggest that Spred2 promotes autophagy in tumor cells. Mechanistically, Spred2 co-localized and interacted with LC3 via the LC3-interacting region (LIR) motifs in its SPR domain. Mutations in the LIR motifs or deletion of the SPR domain impaired Spred2-mediated autophagosome maturation and tumor cell death, indicating that functional LIR is required for Spred2 to trigger tumor cell death. Additionally, Spred2 interacted and co-localized with p62/SQSTM1 through its SPR domain. Furthermore, the co-localization of Spred2, p62 and LAMP2 in HeLa cells indicates that p62 may be involved in Spred2-mediated autophagosome maturation. Inhibition of autophagy using the lysosomal inhibitor chloroquine, reduced Spred2-mediated HeLa cell death. Silencing the expression of autophagy-related genes ATG5, LC3 or p62 in HeLa and A549 cells gave similar results, suggesting that autophagy is required for Spred2-induced tumor cell death. Collectively, these data indicate that Spred2 induces tumor cell death in an autophagy-dependent manner.