Epigenetics as a mediator of plasticity in cancer

Abstract

The concept of an epigenetic landscape describing potential cellular fates arising from pluripotent cells, first advanced by Conrad Waddington, has evolved in light of experiments showing nondeterministic outcomes of regulatory processes and mathematical methods for quantifying stochasticity. In this Review, we discuss modern approaches to epigenetic and gene regulation landscapes and the associated ideas of entropy and attractor states, illustrating how their definitions are both more precise and relevant to understanding cancer etiology and the plasticity of cancerous states. We address the interplay between different types of regulatory landscapes and how their changes underlie cancer progression. We also consider the roles of cellular aging and intrinsic and extrinsic stimuli in modulating cellular states and how landscape alterations can be quantitatively mapped onto phenotypic outcomes and thereby used in therapy development.

Plasticity of cancer cell phenotypes

During differentiation, cells adopt phenotypic states of progressive specificity. Cancer cells violate this property, instead adopting increased plasticity of structure and function. Epigenetic change has been considered a developmental landscape that can channel specific differentiation events and define and constrain distinct phenotypic and gene expression states. In a Review, Feinberg and Levchenko discuss how cancer epigenetic landscapes can be defined quantitatively, borrowing from theory used in physical sciences to define potential energy and its relationship to physical or chemical states. This strategy has yielded new insights whereby stochastic changes in the epigenetic landscape of cancer cells drive oncogenic phenotypes. Such analyses can also reveal pathogenic signaling and therapeutic targets. —GKA

Abstract

A review discusses how epigenetic stochasticity can connect mutations and environmental perturbations to cancer progression and treatment.

Abstract

BACKGROUND

During differentiation, living cells within complex organisms adopt phenotypic states of progressive specificity. Cancerous cells and tissues violate this property, adopting increased plasticity of cell states, tissue structure, and function during their progression. The information about the repertoire of normal differentiation outcomes is genetically encoded, but the information about the particular realization of this potential and cell regulation in response to the environment is encoded epigenetically in DNA methylation and biochemical modification of chromatin. Dating back to Conrad Waddington’s prescient work, epigenetic change has been viewed schematically as a developmental landscape that can channel specific differentiation events and define and constrain distinct phenotypic and gene expression states. More recently, cancer onset and progression have been viewed as a reversal or deformation of this landscape. In the physical sciences, potential energy landscapes and their relationships to the probability distribution of physical or chemical states have been developed and refined for decades, but they have only recently been applied to more quantitatively realize Waddington’s classical landscape idea. Such approaches are particularly appealing in describing the cancer epigenetic landscape given that the plasticity of cell states realized on such a landscape lies at the functional core of the disease.

ADVANCES

Recent developments in experimental technologies, including single cell–resolution analysis of mRNA and protein expression as well as molecular assays of epigenetic modifications of DNA and histones, have enriched our understanding of the diversity of phenotypic states defined by genomic information and epigenetic control. In this work, we expand on the emerging view that there is considerable variability in the expression of biological molecules even within presumably isogenic cells in normal homeostatic tissues or in well-defined cell lines in cell culture. This revelation suggests that the biological processes may be essentially stochastic and that biological variability on the cellular level can be indicative of—or even drive—important aspects of biological function. This analysis has also enabled assessment of the probability distributions of different cellular states, or quasipotential energy, and the use of these to determine the associated entropy, a measure of informational uncertainty. These measures, which we define in detail, enable a precise and quantitative definition of the underlying epigenetic landscapes, coordinately reflected by gene expression landscapes. Cancer-related genetic and epigenetic alterations can increase the entropy of the landscape as a whole and result in higher variability and occupancy of otherwise cryptic attractors. An increase in entropy and thus heterogeneity of the responses—rather than alteration of the average response—is emerging as a key and often overlooked feature of the landscape deformation in cancer pathogenesis. Changes in entropy can also accompany cell differentiation and aging in ways that further inform cancer etiology. They also permit distinguishing phenotypic plasticity from phenotypic heterogeneity. Using recent observations and landscape conceptualization, we outline several scenarios that can occur during precancerous and cancerous progression. We also discuss the molecular mechanisms enabling these scenarios, relating them to specific landscape transformations. We suggest how the relationship between the epigenetic landscape alterations and corresponding phenotypic changes can be quantitatively assessed and used to further understand the information transfer in signaling pathways and to develop new therapeutic interventions. This approach can also incorporate recently introduced ideas of the archetypical states of cells within normal and cancerous tissues.

OUTLOOK

New integrated theoretical and experimental methods in quantitative analyses of the cancer epigenetic landscape provide the tools to understand the connections between genetic and environmental drivers of cancer evolution and the relationships between epigenetic regulatory networks that mediate the landscape. Continued advances in single-cell measurements, including assessment of DNA methylation, genomic sequencing, and chromatin analysis, will allow further understanding of the dynamics of the landscapes having progressively increasing complexity and accounting for tumor evolution, progression to invasive and metastatic spread, and associated alterations in anatomical organization and structure. Moreover, a greater understanding of biological stochasticity, defined mathematically as epigenetic and gene expression entropy, can uncover the cellular actors and mechanisms by which cancer plasticity enables escape from natural defenses and therapeutic interventions.