Chronic metabolic stress drives developmental programs and loss of tissue functions in non-transformed liver that mirror tumor states and stratify survival

Biology Biology
Alex K. Shalek Alex K. Shalek
Ben Mead Ben Mead
Constantine Tzouanas Constantine Tzouanas
Evelyn Yuzhou Tong Evelyn Yuzhou Tong
Jay Prakadan Jay Prakadan
Kellie Kolb Kellie Kolb
Michelle Ramseier Michelle Ramseier
Tyler Dao Tyler Dao

Tzouanas et al.▾ Tzouanas, C. N., Sherman, M. S., Shay, J. E. S., Rubin, A. J., Mead, B. E., Dao, T. T., Butzlass, T., Mana, M. D., Kolb, K. E., Walesky, C., Pepe-Mooney, B. J., Smith, C. J., Prakadan, S. M., Ramseier, M. L., Tong, E. Y., Joung, J., Chi, F., McMahon-Skates, T., Winston, C. L., Jeong, W., Aney, K. J., Chen, E., Nissim, S., Zhang, F., Deshpande, V., Lauer, G. M., Yilmaz, Ö. H., Goessling, W., Shalek, A. K.


December, 2023


Under chronic stress, cells must balance competing demands between cellular survival and tissue function. In metabolic dysfunction-associated steatotic liver disease (MASLD, formerly NAFLD/NASH), hepatocytes cooperate with structural and immune cells to perform crucial metabolic, synthetic, and detoxification functions despite nutrient imbalances. While prior work has emphasized stress-induced drivers of cell death, the dynamic adaptations of surviving cells and their functional repercussions remain unclear. Namely, we do not know which pathways and programs define cellular responses, what regulatory factors mediate (mal)adaptations, and how this aberrant activity connects to tissue-scale dysfunction and long-term disease outcomes. Here, by applying longitudinal single-cell multi-omics to a mouse model of chronic metabolic stress and extending to human cohorts, we show that stress drives survival-linked tradeoffs and metabolic rewiring, manifesting as shifts towards development-associated states in non-transformed hepatocytes with accompanying decreases in their professional functionality. Diet-induced adaptations occur significantly prior to tumorigenesis but parallel tumorigenesis-induced phenotypes and predict worsened human cancer survival. Through the development of a multi-omic computational gene regulatory inference framework and human in vitro and mouse in vivo genetic perturbations, we validate transcriptional (RELB, SOX4) and metabolic (HMGCS2) mediators that co-regulate and couple the balance between developmental state and hepatocyte functional identity programming. Our work defines cellular features of liver adaptation to chronic stress as well as their links to long-term disease outcomes and cancer hallmarks, unifying diverse axes of cellular dysfunction around core causal mechanisms.