Computing the threshold of the influence of intercellular nanotubes on cell-to-cell communication integrity

Dragutin T. Mihailović, Vladimir Kostić, Igor Balaž, D. KaporChaos 2016 · DOI · arXiv

Introduction

In multicellular organisms, functional integrity relies on robust intercellular communication. Traditionally, this exchange is attributed to stable, localized structures known as gap junctions (GJs). However, recent biological discoveries have highlighted the role of tunneling nanotubes (TNTs)—long-distance, transient bridges that facilitate the transfer of organelles, proteins, and signaling molecules. Unlike GJs, TNTs are highly dynamic, forming and dissolving within minutes or hours, which introduces a significant layer of uncertainty or "noise" into the cellular network.

Scanning electron microscopy (SEM) image of intercellular connections. The green arrow indicates a tunneling nanotube (TNT), while the dotted circle highlights a cluster of gap junctions (GJs).

This paper addresses a fundamental question: at what point does the dynamic perturbation introduced by TNTs shift a stable biological system into a state of communicational disintegration? By employing the mathematical framework of matrix pseudospectra, the researchers quantify the "distance to instability," establishing a threshold for the influence of these transient connections.

Methodology

The researchers developed a deterministic model using ordinary differential equations (ODEs) to simulate substance exchange across a network of 100 cells. The internal dynamics of each cell are governed by Michaelis-Menten kinetics, modeling metabolic processes.

The core of the analysis lies in the communication matrix $A$, defined as: $$A = \hat{A} + \xi\Delta$$ Where:

  • $\hat{A}$ represents the stable, observable communication mediated by gap junctions.
  • $\xi\Delta$ represents the unobservable perturbation caused by the dynamic formation of TNTs, with $\xi$ serving as a scaling parameter.

Rather than relying solely on standard eigenvalue analysis—which assumes a perfectly known system—the authors utilize pseudospectra. This approach evaluates the sensitivity of the system's stability to perturbations. Specifically, they calculate the smallest perturbation $\xi$ required to push an eigenvalue into the unstable right-half of the complex plane, defining the "distance to instability."

Network Architectures and Analysis

The model was evaluated on two distinct synthetic network architectures to simulate different biological conditions:

  1. Newman-Gastner Spatial Networks: These mimic healthy tissue where connections are primarily determined by physical proximity.
  2. Erdős-Rényi Modular Networks: These feature clusters of highly connected cells, used to represent both healthy modularity and "pathological" states where specific modules exhibit impaired signaling.

Key Results and Stability Phases

The study identified a critical threshold ($\xi_{crit}$) for system stability. For a 100-cell Newman-Gastner network, this threshold was calculated at $2.38 \times 10^{-3}$. The system's behavior can be categorized into four distinct phases based on the level of TNT-mediated perturbation:

  1. Pure GJ ($\xi=0$): The system exhibits a rapid return to equilibrium.
  2. Stable ($\xi < \xi_{crit}$): Communication integrity is maintained despite the "noise" from TNTs.
  3. Critical ($\xi = \xi_{crit}$): The system enters an oscillatory mode, teetering on the edge of instability.
  4. Unstable ($\xi > \xi_{crit}$): The network undergoes "communicational disintegration," where substance concentrations diverge.
Stability analysis of a 100-cell Newman-Gastner network. (a) Spatial network architecture. (b) Pseudospectra showing the sensitivity of eigenvalues. (c) Time-evolution of the system norm for different perturbation levels, illustrating the transition from stability to disintegration.

A significant finding was the increased fragility of diseased systems. In the Erdős-Rényi modular network, a "pathological" state (where one module has degraded reception) showed a threshold nearly an order of magnitude lower than the healthy state ($\xi_{crit} = 6.55 \times 10^{-4}$ vs $3.15 \times 10^{-3}$). This suggests that pre-existing cellular damage makes a network far more vulnerable to the perturbations introduced by dynamic nanotubes.

Conclusion

This research provides a robust mathematical tool for quantifying the impact of transient intercellular connections. By treating TNTs as functional perturbations, the authors move beyond static architectural models to a more dynamic understanding of biological stability.

Future work aims to incorporate the stochastic nature of cellular environments using Bregman divergences and more complex modeling of active cargo transport. Such advancements will further refine our ability to predict the tipping points of biological systems in both healthy and diseased states.

Summary updated

BibTeX
@inproceedings{mihailovic2016computing,
  author = {Dragutin T. Mihailović and Vladimir Kostić and Igor Balaž and D. Kapor},
  title = {{Computing the threshold of the influence of intercellular nanotubes on cell-to-cell communication integrity}},
  booktitle = {Chaos},
  year = {2016},
  doi = {10.1016/j.chaos.2016.06.001},
}
Figure