The molecular mechanisms underlying cancer resistance remain elusive. One possible explanation is that cancer stem cells (CSCs) elude drug treatment, emerge and reproduce a tumor. Using multiple myeloma as a paradigm, we showed that cancer stem-like cells (CSLCs) appear after genotoxic stress because of their intrinsic properties. However, these properties do not drive the emergence of the CSLCs. Following genotoxic stress, remaining DNA damages lead to a senescence-associated secretory phenotype (SASP). Senescent cells, which are the non-CSLCs, secrete chemokines contributing to the emergence, maintenance and migration of CSLCs. Downregulation of checkpoint protein 2, a key player of SASP, significantly reduced the emergence of CSLCs. Our results unravel a novel molecular mechanism by which SASP might promote malignancy, underlining the dual role of senescence in tumorigenesis. This mechanism, based on mutual cooperation among tumor cells, illustrates how cancer may relapse; its targeting could represent new therapeutic opportunities.

Multiple myeloma (MM) is a malignancy characterized by the accumulation of clonal plasma cells in the bone marrow. Despite extensive efforts to design drugs targeting tumoral cells and their microenvironment, MM remains an incurable disease for which new therapeutic strategies are needed. We demonstrated here that antiestrogens (AEs) belonging to selective estrogen receptor modulators family induce a caspase-dependent apoptosis and trigger a protective autophagy. Autophagy was recognized by monodansylcadaverin staining, detection of autophagosomes by electronic microscopy, and detection of the cleaved form of the microtubule-associated protein light chain 3. Moreover, autophagy was inhibited by drugs such as bafilomycin A1 and 3-methyladenosine. Autophagy was mediated by the binding of AEs to a class of receptors called the antiestrogen binding site (AEBS) different from the classical estrogen nuclear receptors. The binding of specific ligands to the AEBS was accompanied by alteration of cholesterol metabolism and in particular accumulation of sterols: zymostenol or desmosterol depending on the ligand. This was due to the inhibition of the cholesterol-5,6-epoxide hydrolase activity borne by the AEBS. We further showed that the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin pathway mediated autophagy signaling. Moreover, AEBS ligands restored sensitivity to dexamethasone in resistant MM cells. Since we showed previously that AEs arrest MM tumor growth in xenografted mice, we propose that AEBS ligands may have a potent antimyeloma activity alone or in combination with drugs used in clinic.

Mantle cell lymphoma (MCL) is a hematologic neoplasm characterised by the t(11;14)(q13;q32) translocation leading to aberrant cyclin D1 expression. The cell functions of cyclin D1 depend on its partners and/or subcellular distribution, resulting in different oncogenic properties. We observed the accumulation of cyclin D1 in the cytoplasm of a subset of MCL cell lines and primary cells. In primary cells, this cytoplasmic distribution was correlated with a more frequent blastoid phenotype. We performed immunoprecipitation assays and mass spectrometry on enriched cytosolic fractions from two cell lines. The cyclin D1 interactome was found to include several factors involved in adhesion, migration and invasion. We found that the accumulation of cyclin D1 in the cytoplasm was associated with higher levels of migration and invasiveness. We also showed that MCL cells with high cytoplasmic levels of cyclin D1 engrafted more rapidly into the bone marrow, spleen, and brain in immunodeficient mice. Both migration and invasion processes, both in vivo and in vitro, were counteracted by the exportin 1 inhibitor KPT-330, which retains cyclin D1 in the nucleus. Our data reveal a role of cytoplasmic cyclin D1 in the control of MCL cell migration and invasion, and as a true operator of MCL pathogenesis.