No abstract

Expression of the immediate-early response gene IER2 has been associated with the progression of several types of cancer, but its functional role is poorly understood. We found that increased IER2 expression in human melanoma is associated with shorter overall survival, and subsequently investigated the mechanisms through which IER2 exerts this effect. In experimental melanoma models, sustained expression of IER2 induced senescence in a subset of melanoma cells in a p53/MAPK/AKT-dependent manner. The senescent cells produced a characteristic secretome that included high levels of the extracellular phosphoglycoprotein osteopontin. Nuclear localization of the IER2 protein was critical for both the induction of senescence and osteopontin secretion. Osteopontin secreted by IER2-expressing senescent cells strongly stimulated the migration and invasion of non-senescent melanoma cells. Consistently, we observed coordinate expression of IER2, p53/p21, and osteopontin in primary human melanomas and metastases, highlighting the pathophysiological relevance of IER2-mediated senescence in melanoma progression. Together, our study reveals that sustained IER2 expression drives melanoma invasion and progression through stimulating osteopontin secretion via the stochastic induction of senescence.

Glioblastoma multiforme (GBM) is the most commonly diagnosed brain cancer in adults, characterized by rapid proliferation and aggressive invasion into the stroma. Advances in our understanding of the molecular subtypes of GBM have provided attractive druggable targets. However, the high degree of heterogeneity both among patients and within individual tumors has proven a significant challenge for the development of effective therapies. We hypothesized that this heterogeneity is also represented in the mechanical phenotypes of GBM, as the physical properties of tumor tissue strongly influence elements of tumor progression including cell cycle regulation, migration, and therapeutic resistance. To assess these phenotypes, we employed optical trap-based active microrheology to determine the viscoelastic properties of patient-derived GBM cells in 3D hydrogels mimicking the brain ECM. We found that each GBM cell line had a distinct rheological profile as a function of treatment status, and cell lines could be further characterized by strong power law dependence describing intracellular viscoelastic behavior. Single-cell phenotyping according to power law dependence was able to identify subpopulations of cells within the treatment-resistant line. Finally, proteomic analysis indicated that altered mechanical profiles were associated with differential cytoskeletal regulation, particularly in actin- and myosin-binding pathways. This work suggests that evaluating mechanical properties may serve as a valuable strategy for the further stratification of these tumors, and encourages the investigation of cytoskeleton regulation as a potential therapeutic target for GBM.