Anticipating critical transitions in epithelial-hybrid-mesenchymal cell-fate determination

Abstract

13In the vicinity of a tipping point, critical transitions occur when small changes in an input condition causes 14 sudden, large and often irreversible changes in the state of a system. Many natural systems ranging from 15 ecosystems to molecular biosystems are known to exhibit critical transitions in their response to stochastic 16 perturbations. In diseases, an early prediction of upcoming critical transitions from a healthy to a dis-17 ease state by using early warning signals is of prime interest due to potential application in forecasting 18 disease onset. Here, we analyze cell-fate transitions between different phenotypes (epithelial, hybrid epithe-19 lial/mesenchymal (E/M) and mesenchymal states) that are implicated in cancer metastasis and chemoresis-20 tance. These transitions are mediated by a mutually inhibitory feedback loop microRNA-200/ZEB driven 21 by the levels of transcription factor SNAIL. We find that the proximity to tipping points enabling these 22 transitions among different phenotypes can be captured by critical slowing down based early warning sig-23 nals, calculated from the trajectory of ZEB mRNA level. Further, the basin stability analysis reveals the 24 unexpectedly large basin of attraction for a hybrid E/M phenotype. Finally, we identified mechanisms that 25 can potentially elude the transition to a hybrid E/M phenotype. Overall, our results unravel the early warn- 26 ing signals that can be used to anticipate upcoming epithelial-hybrid-mesenchymal transitions. With the 27 emerging evidence about the hybrid E/M phenotype being a key driver of metastasis, drug resistance, and 28 tumor relapse; our results suggest ways to potentially evade these transitions, reducing the fitness of cancer 29 cells and restricting tumor aggressiveness. 30 Keywords: critical transition | indicators of critical slowing down | alternative stable states | epithelial-31 hybrid-mesenchymal transition | cancer biology 32 1 This article contains supplementary materials. 2 sudipta@iitrpr.ac.in 3 h.levine@northeastern.edu 4 mkjolly@iisc.ac.in 5 parthasharathi@iitrpr.ac.in Significance Statement 33Epithelial-hybrid-mesenchymal transitions play critical roles in cancer metastasis, drug resistance, and tumor 34 relapse. Recent studies have proposed that cells in a hybrid epithelial/mesenchymal phenotype may be more 35 aggressive than those on either end of the spectrum. However, no biomarker to predict upcoming transitions 36 has been identified. Here, we show that critical slowing down based early warning signals can detect sudden 37 transitions among epithelial, hybrid E/M, and mesenchymal phenotypes. Importantly, our results highlight 38 how stable a hybrid E/M phenotype can be, and how can a transition to this state be avoided. Thus, our 39 study provides valuable insights into restricting cellular plasticity en route metastasis. 40 Introduction 41Biological systems often display nonlinear dynamics and emergent complex behavior, and consequent multi-42 stability [1, 2]. This nonlinear behavior in many ...

Machine learning methods trained on simple models can predict critical transitions in complex natural systems

“…Transition from one steady state to another occur in many complex systems, such as financial markets (May et al 2008), human societies (Scheffer 2009, Downey et al 2016, Scheffer 2016), climate systems (Dakos et al 2008, Lenton 2011, Lenton et al 2019, systems biology (Sarkar et al 2019), and ecosystems (Scheffer and Carpenter 2003, Veraart et al 2012, Rindi et al 2017. Such transitions can be abrupt and irreversible (i.e., catastrophic), or smooth and reversible (i.e., non-catastrophic), and can occur due to gradual external forcing or random fluctuations in the system.…”

“…Consequently, there remains a critical need to develop a robust toolkit to identify tipping points using widely available time series data. Were this to be achieved, it could have significant implications for the management of a host of systems, from financial markets to species at risk of extinction (Sarkar et al 2019, Scheffer and Carpenter 2003, Veraart et al 2012.…”

Anticipating the novel coronavirus disease (COVID-19) pandemic

“…In particular, the phenomenon of CSD can be captured as a large time taken by a system to return to its previous states due to which the rate of return of a system decreases prior to a transition. Moreover, it leads to an increase in the short term memory of a system, this feature can be identified by the changes in the correlation structure of a time-series preceding a critical transition [13,14,[21][22][23]. Thus, it is crucial to understand whether these indicators capture the characteristics of slowing down in the epidemic growth curves.…”

“…Theory suggests applicability of a variety of leading generic indicators, widely known as Early Warning Signals (EWSs) (e.g. variance, autocorrelation, skewness, and kurtosis) to identify the proximity of a system to such a critical transition [22, 23, 25, 26]. For instance, in time-series data following ancient abrupt climate shifts, EWSs can be identified before the critical transition took place [27].…”

Tipping-point analysis uncovers critical transition signals from gene expression profiles

“…Our work implicates a theoretical debate in the field of computational biology regarding significant CTS identification: whether TFs can play key roles in critical transition (Sarkar et al, 2019, Zhou, Yui et al, 2019), or whether gene signals during transition are entirely stochastic (Mojtahedi et al, 2016, Wu, Chen et al, 2018). The novel conceptualization is that the lineage specifications, at least partly, are not just a simple gradual silencing of the progenitor cells with simple upregulation of lineage marker genes (Zhou et al, 2019).…”