Gene regulatory interactions limit the gene expression diversity

Orr Levy, Shubham Tripathi, Scott D. Pope, Yang Y. Liu, Ruslan MedzhitovarXiv 2023 · DOI · arXiv

TL;DR

This paper reveals a fundamental trade-off in biological systems: the complexity of gene regulatory networks (GRNs) is limited by stability constraints, which in turn restricts the diversity of gene expression within cells. This finding helps explain why cells, despite their diverse functions, operate within specific expression profiles.

What Problem Does This Paper Address?

Cells achieve their specialized functions through intricate gene regulatory networks (GRNs) that control gene expression. While we know GRNs are crucial for cellular diversity and adaptation, the underlying principles that shape them and the forces that govern their complexity have remained elusive. Specifically, researchers have wondered if there's a limit to how diverse gene expression can be within a cell, and how GRN interactions contribute to this. This paper tackles this question by investigating a potential trade-off between the intensity of GRN interactions and the diversity of gene expression.

Methods

The authors developed a computational framework to estimate GRN interaction intensity directly from single-cell RNA sequencing (scRNA-seq) data, without needing to reconstruct the entire GRN. This method, termed Global Coordination Level (GCL), quantifies the multivariate interdependence among genes. They used a bias-corrected distance correlation measure to assess this interdependence. To validate their approach, they employed a classical GRN dynamical model to simulate gene expression data and then applied their GCL method to both simulated and real scRNA-seq datasets.

Datasets

The study utilized scRNA-seq data from various sources, including:

  • Simulated Data: Generated using a co-regulation GRN dynamics model.
  • Real Data:
    • Human: Tabula Sapiens dataset [31] (multiple tissues and cell types).
  • Mice: Tabula Muris dataset [32] (multiple tissues and cell types).
  • Human Gut: Data from adult, pediatric, and fetal donors [33].
  • Fruit Fly: Fly Cell Atlas dataset [34].
  • C. elegans: Atlas dataset [35].
  • Hematopoietic Stem Cell (HSC) differentiation data from both mice and humans [32, 36].
  • Macrophage stimulation data (LPS and IL-4) from mice.

Key Insights and Ideas

  • The GRN Complexity Limit: The core idea is that GRNs have an inherent limit to their complexity. This limit is not just about the number of genes or interactions, but a product of both gene expression diversity and interaction intensity. Pushing beyond this limit leads to instability.
  • Stability as a Driver: The observed trade-off is driven by the need for cellular stability. GRNs must evolve to be either highly interconnected with few expressed genes, or have many expressed genes with weaker interactions to maintain stability.
  • Transition States are Complex: Cells undergoing differentiation or other state transitions (like epithelial-to-mesenchymal) exhibit higher complexity and are less stable than their more differentiated or stem cell counterparts. This suggests these transient states are operating closer to the GRN complexity limit.
  • GCL as a Proxy for Interaction Intensity: The developed GCL metric effectively captures the underlying GRN interaction intensity, providing a valuable tool for analyzing gene expression data without full GRN reconstruction.

Key Results

  • Simulated Trade-off: Simulations using a GRN dynamics model showed a clear negative correlation between gene expression diversity (measured by Neff) and GRN interaction intensity (measured by GCL or direct interaction parameters).
  • Real Data Validation: This trade-off was consistently observed across diverse real scRNA-seq datasets from humans, mice, fruit flies, and C. elegans, spanning various tissues and cell types.
  • Differentiation Dynamics: During HSC differentiation, partially differentiated (transition) cells showed higher complexity (GCL * Neff) compared to stem cells and fully differentiated cells.
  • Stimulation Effects: Macrophage stimulation led to a decrease in gene expression diversity (Neff) and an increase in GRN interaction intensity (GCL), supporting the trade-off hypothesis.

Limitations and Future Work

  • Model Assumptions: The study acknowledges assumptions made in their modeling, such as simulating cells within a designated class (e.g., network topology) and assuming all cells of a given type originate from the same GRN. Future work could explore intermingling these classes.
  • GRN Reconstruction: While GCL bypasses the need for full GRN reconstruction, understanding the specific network motifs and interactions that drive this trade-off remains an area for future investigation.
  • Broader Applications: The authors suggest their approach for assessing GRN interaction intensity could be valuable for understanding the impact of perturbations on GRNs and for applications in fields like tissue engineering and therapeutic interventions.

Summary updated

Abstract

The diversity of expressed genes plays a critical role in cellular specialization, adaptation to environmental changes, and overall cell functionality. This diversity varies dramatically across cell types and is orchestrated by intricate, dynamic, and cell type-specific gene regulatory networks (GRNs). Despite extensive research on GRNs, their governing principles, as well as the underlying forces that have shaped them, remain largely unknown. Here, we investigated whether there is a tradeoff between the diversity of expressed genes and the intensity of GRN interactions. We have developed a computational framework that evaluates GRN interaction intensity from scRNA-seq data and used it to analyze simulated and real scRNA-seq data collected from different tissues in humans, mice, fruit flies, and C. elegans. We find a significant tradeoff between diversity and interaction intensity, driven by stability constraints, where the GRN could be stable up to a critical level of complexity - a product of gene expression diversity and interaction intensity. Furthermore, we analyzed hematopoietic stem cell differentiation data and find that the overall complexity of unstable transition states cells is higher than that of stem cells and fully differentiated cells. Our results suggest that GRNs are shaped by stability constraints which limit the diversity of gene expression.
BibTeX
@misc{levy2023gene,
  author = {Orr Levy and Shubham Tripathi and Scott D. Pope and Yang Y. Liu and Ruslan Medzhitov},
  title = {{Gene regulatory interactions limit the gene expression diversity}},
  year = {2023},
  eprint = {2311.15503},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2311.15503},
}

Loading related content...

Figure