Landscape of epithelial–mesenchymal plasticity as an emergent property of coordinated teams in regulatory networks
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
In this paper, the authors identify topological metrics in gene-regulatory networks that potentially predict the kinds of phenotypic steady-states that the network allows. In particular, they apply their results to the epithelial-mesenchymal plasticity, showing that the relevant gene regulatory networks are structured as ‘teams' that may be 'strong', yielding stable phenotypes, or 'weak', yielding unstable phenotypes prone to plasticity. The work would be of interest to researchers interested in systems biology and the nonlinear dynamics of biological systems, as well as biologists interested in gene regulatory networks and their (mis)functioning in cancer cells.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Computational and Systems Biology (eLife)
Abstract
Elucidating the design principles of regulatory networks driving cellular decision-making has fundamental implications in mapping and eventually controlling cell-fate decisions. Despite being complex, these regulatory networks often only give rise to a few phenotypes. Previously, we identified two ‘teams’ of nodes in a small cell lung cancer regulatory network that constrained the phenotypic repertoire and aligned strongly with the dominant phenotypes obtained from network simulations (Chauhan et al., 2021). However, it remained elusive whether these ‘teams’ exist in other networks, and how do they shape the phenotypic landscape. Here, we demonstrate that five different networks of varying sizes governing epithelial–mesenchymal plasticity comprised of two ‘teams’ of players – one comprised of canonical drivers of epithelial phenotype and the other containing the mesenchymal inducers. These ‘teams’ are specific to the topology of these regulatory networks and orchestrate a bimodal phenotypic landscape with the epithelial and mesenchymal phenotypes being more frequent and dynamically robust to perturbations, relative to the intermediary/hybrid epithelial/mesenchymal ones. Our analysis reveals that network topology alone can contain information about corresponding phenotypic distributions, thus obviating the need to simulate them. We propose ‘teams’ of nodes as a network design principle that can drive cell-fate canalization in diverse decision-making processes.
Article activity feed
-
-
Author Response
Reviewer #2 (Public Review):
In this paper, the authors identify topological metrics in gene-regulatory networks that have the potential to predict the sub-types of phenotypic steady states that the network can lead to. The results hold great value for the field of Theoretical Systems Biology.
The paper becomes too technical too quickly and assumes a lot of knowledge from the reader. Equations and theoretical concepts are not always well defined. In general, I would recommend connecting the results from the simulations/topology metrics to EMP biology earlier in the paper. Alternatively, rather than investigating 5 networks related to EMP, the generalization of the statements could become stronger if the authors explore the trends of the theoretical analysis in networks modeling other biological processes (such as …
Author Response
Reviewer #2 (Public Review):
In this paper, the authors identify topological metrics in gene-regulatory networks that have the potential to predict the sub-types of phenotypic steady states that the network can lead to. The results hold great value for the field of Theoretical Systems Biology.
The paper becomes too technical too quickly and assumes a lot of knowledge from the reader. Equations and theoretical concepts are not always well defined. In general, I would recommend connecting the results from the simulations/topology metrics to EMP biology earlier in the paper. Alternatively, rather than investigating 5 networks related to EMP, the generalization of the statements could become stronger if the authors explore the trends of the theoretical analysis in networks modeling other biological processes (such as SCLC).
One of the main findings of the paper is that the distance between the matrix of correlation values between nodes in all steady states obtained from simulation and influence matrix indicates that the mean group strength is a good quantity to identify teams of nodes in the network. However, it remains unclear how to identify groups/teams in the networks based on influence: is it unsupervised (hierarchical?) clustering? How do the authors identify the number of teams of nodes in randomized?
The authors also explore whether team structure correlates with the stability of relevant biological phenotypes. To characterize stability, they define static (e.g., frustration and stead state frequency) and dynamic network metrics (e.g., coherence and higher-order perturbations), and correlate them to the mean group strength in both WT and randomized networks. Results are promising: team structure and group mean strength show interesting correlative trends with both the static and dynamic metrics. However, everything relies on the mean group strength, which as mentioned before is not convincingly defined in randomized networks.
Taken together, the conclusions of this paper would be better supported if a better explanation of team identification in gene-regulatory networks would be provided, and if networks related to other biological processes would be investigated.
We thank the referee for their encouraging remarks and valuable suggestions about improving the manuscript. We are excited that the referee finds our results promising and of great value to the field of theoretical systems biology. Following the suggestions given here, we have now included further clarification on various aspects, included results for regulatory networks of melanoma and small cell lung cancer (SCLC, Fig 9, S11), and described in detail the algorithm used to identify teams in a given network (Methods)
-
-
Evaluation Summary:
In this paper, the authors identify topological metrics in gene-regulatory networks that potentially predict the kinds of phenotypic steady-states that the network allows. In particular, they apply their results to the epithelial-mesenchymal plasticity, showing that the relevant gene regulatory networks are structured as ‘teams' that may be 'strong', yielding stable phenotypes, or 'weak', yielding unstable phenotypes prone to plasticity. The work would be of interest to researchers interested in systems biology and the nonlinear dynamics of biological systems, as well as biologists interested in gene regulatory networks and their (mis)functioning in cancer cells.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested …
Evaluation Summary:
In this paper, the authors identify topological metrics in gene-regulatory networks that potentially predict the kinds of phenotypic steady-states that the network allows. In particular, they apply their results to the epithelial-mesenchymal plasticity, showing that the relevant gene regulatory networks are structured as ‘teams' that may be 'strong', yielding stable phenotypes, or 'weak', yielding unstable phenotypes prone to plasticity. The work would be of interest to researchers interested in systems biology and the nonlinear dynamics of biological systems, as well as biologists interested in gene regulatory networks and their (mis)functioning in cancer cells.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
The present manuscript investigates an essential aspect of multicellularity, coordination between gene regulatory networks (GRNs), presented as grouped in so-called teams, and their deregulations in cancer, leading in particular in cancer to epithelial-mesenchymal plasticity (EMP), the propension of plastic cancer cells to go towards an epithelial (MET) or a mesenchymal (EMT) phenotype, according to the needs of the cancer cell population in a given ecosystem.
The authors have a long experience in molecular systems biology studies and publications in EMP (EMT/MET) and they are well aware of publications in the worldwide community on this topic. The present study proposes a structure of GRNs in 'teams' that may be 'strong', yielding stable phenotypes, or 'weak', yielding unstable phenotypes prone to plasticity.
Reviewer #1 (Public Review):
The present manuscript investigates an essential aspect of multicellularity, coordination between gene regulatory networks (GRNs), presented as grouped in so-called teams, and their deregulations in cancer, leading in particular in cancer to epithelial-mesenchymal plasticity (EMP), the propension of plastic cancer cells to go towards an epithelial (MET) or a mesenchymal (EMT) phenotype, according to the needs of the cancer cell population in a given ecosystem.
The authors have a long experience in molecular systems biology studies and publications in EMP (EMT/MET) and they are well aware of publications in the worldwide community on this topic. The present study proposes a structure of GRNs in 'teams' that may be 'strong', yielding stable phenotypes, or 'weak', yielding unstable phenotypes prone to plasticity.
The ideas presented in this manuscript, taking advantage of a well-set methodological analysis of gGRNs and their coordination, include fixation on given physiological phenotypes and their stability by such teams, hybrid states, and plastic transitions between states, all based on their newly introduced concept of teams, which might be an illuminating one to understand normal coordination between tissues in multicellular organisms and its deregulation in cancer.
-
Reviewer #2 (Public Review):
In this paper, the authors identify topological metrics in gene-regulatory networks that have the potential to predict the sub-types of phenotypic stead states that the network can lead to. The results hold great value for the field of Theoretical Systems Biology.
The paper becomes too technical too quickly and assumes a lot of knowledge from the reader. Equations and theoretical concepts are not always well defined. In general, I would recommend connecting the results from the simulations/topology metrics to EMP biology earlier in the paper. Alternatively, rather than investigating 5 networks related to EMP, the generalization of the statements could become stronger if the authors explore the trends of the theoretical analysis in networks modeling other biological processes (such as SCLC).
One of the main …
Reviewer #2 (Public Review):
In this paper, the authors identify topological metrics in gene-regulatory networks that have the potential to predict the sub-types of phenotypic stead states that the network can lead to. The results hold great value for the field of Theoretical Systems Biology.
The paper becomes too technical too quickly and assumes a lot of knowledge from the reader. Equations and theoretical concepts are not always well defined. In general, I would recommend connecting the results from the simulations/topology metrics to EMP biology earlier in the paper. Alternatively, rather than investigating 5 networks related to EMP, the generalization of the statements could become stronger if the authors explore the trends of the theoretical analysis in networks modeling other biological processes (such as SCLC).
One of the main findings of the paper is that the distance between the matrix of correlation values between nodes in all steady states obtained from simulation and influence matrix indicates that the mean group strength is a good quantity to identify teams of nodes in the network. However, it remains unclear how to identify groups/teams in the networks based on influence: is it unsupervised (hierarchical?) clustering? How do the authors identify the number of teams of nodes in randomized?
The authors also explore whether team structure correlates with the stability of relevant biological phenotypes. To characterize stability, they define static (e.g., frustration and stead state frequency) and dynamic network metrics (e.g., coherence and higher-order perturbations), and correlate them to the mean group strength in both WT and randomized networks. Results are promising: team structure and group mean strength show interesting correlative trends with both the static and dynamic metrics. However, everything relies on the mean group strength, which as mentioned before is not convincingly defined in randomized networks.
Taken together, the conclusions of this paper would be better supported if a better explanation of team identification in gene-regulatory networks would be provided, and if networks related to other biological processes would be investigated.
-