Chemotherapy-Induced Senescence Reprograms Lymphoma and Leukemia Cells into Latent Cancer Stem Cells That Are Susceptible to Conceptually Novel Treatments

Cellular senescence is a stress-responsive cell-cycle arrest program that terminates further expansion of (pre-)malignant cells. Senescence imposes a tumor-suppressive barrier in lymphomagenesis, and acts as an effector program in response to chemotherapy in various hematological malignancies. Interestingly, many key signaling components of the senescence machinery also operate as critical regulators of stem cell functions (collectively termed ‘stemness’), among them the p53 axis and control of lysine 9 trimethylation at histone H3 (H3K9). We investigated here in vitro and in vivo whether chemotherapy-induced senescence may change stem cell-related functionalities in aggressive B-cell lymphomas and acute leukemias of murine and human origin.

Gene expression and functional analyses comparing senescent vs. non-senescent Eµ-myc transgenic B-cell lymphomas unveiled massive upregulation of an adult tissue stem cell signature, activated Wnt signaling, and de novo expression of distinct stem cell markers in senescence. Utilizing Suv39h1- (an H3K9-targeting methyltransferase) and p53-based genetically switchable ‘matched pair’ on/off models of senescence to mimic spontaneous escape (‘previously senescent’, PS), we found PS cells to re-enter the cell-cycle with strongly enhanced and Wnt-dependent clonogenic growth when compared to their equally chemotherapy-exposed but never senescent (NS) counterparts.In vivo, these PS lymphoma cells presented with a much higher tumor initiation potential, which was neutralized upon pharmacological or genetic Wnt inhibition. Strikingly, temporary enforcement of senescence in a p53-regulatable leukemia model reprogrammed non-stem bulk PS leukemia cells into leukemia-initiating stem cells, whose de novo self-renewing potential was also Wnt-dependent. In contrast, equally chemotherapy-exposed NS bulk cells (i.e. p53 always ‘off’) did not acquire stemness potential.

Our data, further supported by consistent findings in various human cancer cell lines and primary patient-derived lymphoma and leukemia samples, characterize senescence as a fundamentally reprogrammed cellular condition, and uncover senescence-associated stemness as an unexpected, cell-autonomous feature that exerts its detrimental potential upon escape from the cell-cycle block. These findings raise concerns about the long-term benefit of senescence-inducing cancer therapies, and provide new mechanistic insights into the plasticity of the “cancer stem cell” condition. In turn, we present synthetic lethal targeting of senescence-associated stemness as a conceptually novel, outcome-improving treatment strategy in lymphoma, leukemia and possibly other cancer entities as well.