Feedback loops involving ERK, AMPK and TFEB generate non-genetic heterogeneity that allows cells to evade anoikis

Cancer cells metastasize by evading apoptosis induced due to the detachment from the extracellular matrix. Deciphering the adaptive strategies employed by cancer cells in matrix-deprived condition can help design novel therapeutic strategies to tackle metastasis. Here, we provide evidence for non-genetic heterogeneity in matrix-detached cells enabled by feedback loop among AMPK, ERK and TFEB that determines autophagy maturation and cell survival. The subpopulation of matrix-detached cells with pAMPK low /pERK high /TFEB low state show autophagy maturation arrest and elevated cell death. Conversely, pAMPK high /pERK low / TFEB high cells show high autophagy maturation and better cell survival. We show that AMPK inhibits ERK activity in suspension; ERK negatively regulates TFEB which promotes autophagy maturation and re-enforces AMPK. Inhibition of ERK promotes autophagy maturation, cell survival and metastasis in vivo , while AMPK inhibition (and TFEB depletion) renders the population homogeneous by depleting the pAMPK high /pERK low /TFEB high subpopulation, driving detachment-induced death. Such non-genetic heterogeneity is further deciphered by mathematical modelling and RNA-sequencing data of circulating tumor cells (CTCs) isolated from breast cancer patients. Altogether, our work unravels a contextual feedback loop involving two kinases and a transcription factor that helps a subpopulation to evade cell death in matrix-deprived condition. Disrupting such feedback loops may offer improved therapeutic efficacy and a novel approach to constrain metastasis. Significance of the Study Attachment to the extracellular matrix is pivotal for the growth and survival of normal epithelial cells. In contrast, cancer cells acquest the ability to survive matrix-deprivation and cause cancer spread, or metastasis, which is a leading cause of cancer-related deaths. Non-genetic heterogeneity within cancer cell populations is increasingly being recognized as a major cause of treatment failure. However, the implications of such heterogeneity in the survival of matrix-detached cancer cells remain poorly understood. In this study, we demonstrate feedback loops involving kinases and transcription factor to maintain non-genetic heterogeneous population, where population harbouring pAMPK high /pERK low /TFEB high status shows survival advantage. Targeting such feedback loops that generate non-genetic heterogeneity can open newer therapeutic means to restrict cancer spread.