Stem cells: a window of opportunity in low-dimensional EMT space

Qing NieOncotarget 2018 · DOI · OpenAlex

Abstract

Cancer progression is driven by the complex interplay between a cell's ability to self-renew (stemness) and its ability to migrate (Epithelial-Mesenchymal Transition, or EMT). This paper introduces a mathematical framework that links these two processes, defining a "window of opportunity" for stemness within the EMT spectrum. By modeling regulatory circuits like Notch and NF-κB, the research demonstrates how the location and size of this window determine cancer aggressiveness and heterogeneity across different tumor types.

Introduction

In developmental biology and oncology, cells frequently navigate a trade-off between proliferation and motility. "Cancer Stem Cells" (CSCs) are responsible for tumor initiation and growth, while the "Epithelial-Mesenchymal Transition" (EMT) enables these cells to detach and metastasize. Historically, these phenomena were studied independently. However, understanding how cells coordinate these tasks is vital for predicting tumor behavior. This study by Bocci et al. utilizes computational modeling to bridge this gap, proposing that stemness is not a fixed trait but a "window of opportunity" that shifts along a low-dimensional EMT space.

The "Window of Opportunity" Framework

The researchers employed mechanism-based computational modeling to simplify complex gene expression profiles into a one-dimensional "EMT score." This spectrum ranges from the stationary Epithelial (E) state to the highly mobile Mesenchymal (M) state.

The 'window of opportunity' for stemness moves along the Epithelial-Mesenchymal spectrum, controlled by specific regulatory signals.

A central finding is that stemness is typically tied to an "intermediate cell state" (ICS)—a hybrid phenotype that possesses both epithelial and mesenchymal characteristics. As shown in the figure above, this window is dynamic; its position and size are governed by "location regulators" and "window size control" mechanisms.

Regulatory Dynamics and Notch Signaling

The study highlights the role of the Notch-Jagged signaling pathway as a critical "matcher." Notch signaling facilitates cell-to-cell communication that aligns the intermediate EMT state with the gene expression patterns required for stemness.

Furthermore, the size of the stemness window is determined by a "signal tug-of-war." For example, NF-κB signaling acts as an expander, increasing the likelihood of a cell achieving stemness. Conversely, intense EMT induction can paradoxically shrink this window, limiting the potential for stem-like behavior even as cells become more migratory.

Clinical Validation and Therapeutic Implications

The model's predictions were validated using gene expression data from lung, breast, thyroid, and glioblastoma cancers. Each cancer type exhibited a unique "window" profile:

  • Aggressiveness: The location of the window (closer to the M end) correlates with higher invasive potential.
  • Heterogeneity: The model explains why some tumors contain diverse populations of CSCs.
  • Drug Response: The researchers successfully modeled the effects of metformin, a common anti-diabetic drug with known anti-cancer properties. Metformin appears to work by manipulating the stemness window, effectively "shrinking" it or shifting it toward a less aggressive state.

Conclusion

This low-dimensional framework provides a powerful tool for predicting cancer behavior and identifying new therapeutic targets. By understanding the mathematical rules governing the "window of opportunity," clinicians may eventually be able to "tune" cellular states to reduce tumor aggressiveness. Future work will focus on identifying the specific molecular "knobs" that control these shifts and addressing the complexities of tissues that may harbor multiple intermediate states simultaneously.

Summary updated

BibTeX
@inproceedings{nie2018stem,
  author = {Qing Nie},
  title = {{Stem cells: a window of opportunity in low-dimensional EMT space}},
  booktitle = {Oncotarget},
  year = {2018},
  doi = {10.18632/oncotarget.25852},
}
Figure