Stochastic analysis of a miRNA-protein toggle switch

Enrico Giampieri, Daniel Remondini, L. de Oliveira, Gastone Castellani, Píetro Lió

Abstract

Within systems biology there is an increasing interest in the stochastic behavior of genetic and biochemical reaction networks. An appropriate stochastic description is provided by the chemical master equation, which represents a continuous time Markov chain (CTMC). In this paper we consider the stochastic properties of a biochemical circuit, known to control eukaryotic cell cycle and possibly involved in oncogenesis, recently proposed in the literature within a deterministic framework. Due to the inherent stochasticity of biochemical processes and the small number of molecules involved, the stochastic approach should be more correct in describing the real system: we study the agreement between the two approaches by exploring the system parameter space. We address the problem by proposing a simplified version of the model that allows analytical treatment, and by performing numerical simulations for the full model. We observed optimal agreement between the stochastic and the deterministic description of the circuit in a large range of parameters, but some substantial differences arise in at least two cases: 1) when the deterministic system is in the proximity of a transition from a monostable to a bistable configuration, and 2) when bistability (in the deterministic system) is "masked" in the stochastic system by the distribution tails. The approach provides interesting estimates of the optimal number of molecules involved in the toggle. Our discussion of the points of strengths, potentiality and weakness of the chemical master equation in systems biology and the differences with respect to deterministic modeling are leveraged in order to provide useful advice for both the bioinformatician practitioner and the theoretical scientist.

Introduction: The Role of Molecular Noise

In systems biology, biochemical networks are traditionally modeled using Ordinary Differential Equations (ODEs) based on the Law of Mass Action. While useful, these deterministic models assume high concentrations of molecules. However, inside a cell, key players like transcription factors often exist in low copy numbers—frequently fewer than 1,000 molecules. In this regime, "molecular noise"—the randomness of individual discrete events—becomes significant.

This paper explores the E2F-MYC-miR-17-92 toggle switch, a critical regulator of the cell cycle and oncogenesis. By comparing deterministic (ODE) and stochastic (Chemical Master Equation, or CME) modeling approaches, the authors demonstrate that while the two models often align, stochasticity reveals behaviors that deterministic models simply cannot see.

The E2F-MYC-miR-17-92 Circuit

The circuit under investigation acts as a biological "toggle switch" that determines whether a cell remains quiescent, enters a proliferative state, or undergoes apoptosis. The core of this switch involves the interaction between the E2F/Myc protein complex and the miR-17-92 microRNA cluster.

Schematic of the E2F-MYC-miR-17-92 regulatory network and its downstream effectors.

As shown in Figure 1, the network is complex, involving feedback loops between Myc, E2F, and the miRNA cluster. The authors focus on whether the standard deterministic view accurately captures the behavior of this switch or if the inherent noise of the cellular environment creates novel dynamics.

Modeling Methodology

The researchers employed a multi-tiered modeling strategy to bridge the gap between theory and simulation:

  • Deterministic Modeling: A 2D system of ODEs described the concentrations of the protein complex ($p$) and the miRNA cluster ($m$).
  • Stochastic Modeling (CME): The system was formulated as a continuous-time Markov chain to capture the probability distribution of molecule counts over time.
  • Analytical Simplification: By assuming "time-scale separation"—specifically that miRNA dynamics are much faster than protein dynamics—the authors derived an exact analytical solution for the steady-state probability distribution in a reduced 1D model.
  • Numerical Simulation: For the full 2D stochastic model, they used a modified Gillespie algorithm (the Logarithmic Direct Method) to simulate single trajectories and estimate the stationary distribution.

Parameters for these simulations were derived from previously published biological models, specifically those by Aguda et al. (2008).

Key Insights: Ghosts and Masking

The study revealed two critical phenomena where stochasticity diverges from deterministic predictions:

The "Ghost" Effect

One of the most striking findings is that the stochastic system can exhibit bistability (two distinct peaks in probability) even when the deterministic model predicts monostability. This occurs near "bifurcation points." The system's dynamics slow down near a recently disappeared fixed point, "trapping" the system and creating a "ghost" peak in the distribution.

Peak Masking

Conversely, a system that is deterministically bistable might appear monostable in a stochastic simulation. If the variance of the primary state is high enough, its "tail" can overlap and completely mask the second peak, making a proliferative state look like a rarely visited metastable state.

The 100-Molecule Rule

The authors note that in a typical cell volume ($10^{-13}$ liters), a 1 nM concentration corresponds to only about 100 molecules. At this scale, uncertainty is roughly 10%, making the stochastic approach a necessity for accuracy. A "metastable" vs. a "fully bimodal" distribution could represent the difference between a healthy cell and one primed for tumoral transition.

Results and Limitations

The authors categorized the system's behavior into four cases. They found high agreement when time-scale separation is strong, but significant failures in the 1D approximation when protein and miRNA dynamics occur at similar speeds. Furthermore, the "ghosting" and "masking" effects highlight the nuanced way stochastic models discriminate between healthy and cancerous states.

Future work will need to address the assumption of constant molecular influx, as real-world gene expression is often "bursty." Additionally, as the state space for larger networks explodes computationally, more efficient approximation methods will be required. The ultimate goal is fitting these stochastic models to single-cell experimental data, such as flow cytometry, to identify which toggle switch modes correspond to specific cancer types.

Summary updated

Abstract

Within systems biology there is an increasing interest in the stochastic behavior of genetic and biochemical reaction networks. An appropriate stochastic description is provided by the chemical master equation, which represents a continuous time Markov chain (CTMC). In this paper we consider the stochastic properties of a biochemical circuit, known to control eukaryotic cell cycle and possibly involved in oncogenesis, recently proposed in the literature within a deterministic framework. Due to the inherent stochasticity of biochemical processes and the small number of molecules involved, the stochastic approach should be more correct in describing the real system: we study the agreement between the two approaches by exploring the system parameter space. We address the problem by proposing a simplified version of the model that allows analytical treatment, and by performing numerical simulations for the full model. We observed optimal agreement between the stochastic and the deterministic description of the circuit in a large range of parameters, but some substantial differences arise in at least two cases: 1) when the deterministic system is in the proximity of a transition from a monostable to a bistable configuration, and 2) when bistability (in the deterministic system) is "masked" in the stochastic system by the distribution tails. The approach provides interesting estimates of the optimal number of molecules involved in the toggle. Our discussion of the points of strengths, potentiality and weakness of the chemical master equation in systems biology and the differences with respect to deterministic modeling are leveraged in order to provide useful advice for both the bioinformatician practitioner and the theoretical scientist.
BibTeX
@inproceedings{giampieri2011stochastic,
  author = {Enrico Giampieri and Daniel Remondini and L. de Oliveira and Gastone Castellani and Píetro Lió},
  title = {{Stochastic analysis of a miRNA-protein toggle switch}},
  booktitle = {Molecular},
  year = {2011},
  doi = {10.1039/c1mb05086a},
}
Figure