Intrinsic noise alters the frequency spectrum of mesoscopic oscillatory chemical reaction systems

Rajesh Ramaswamy, Ivo F. SbalzariniScientific 2011 · DOI

Abstract

Mesoscopic oscillatory reaction systems, for example in cell biology, can exhibit stochastic oscillations in the form of cyclic random walks even if the corresponding macroscopic system does not oscillate. We study how the intrinsic noise from molecular discreteness influences the frequency spectrum of mesoscopic oscillators using as a model system a cascade of coupled Brusselators away from the Hopf bifurcation. The results show that the spectrum of an oscillator depends on the level of noise. In particular, the peak frequency of the oscillator is reduced by increasing noise, and the bandwidth increased. Along a cascade of coupled oscillators, the peak frequency is further reduced with every stage and also the bandwidth is reduced. These effects can help understand the role of noise in chemical oscillators and provide fingerprints for more reliable parameter identification and volume measurement from experimental spectra.

Introduction

In macroscopic chemistry, reaction dynamics are typically described by Reaction Rate Equations (RRE), which treat concentrations as continuous and deterministic. However, in mesoscopic environments like biological cells or nano-reactors, the small number of molecules leads to significant fluctuations known as intrinsic noise. While it is well-established that noise can induce "stochastic oscillations" in systems that are macroscopically stable, its specific impact on the frequency characteristics of these oscillators has remained less clear. This study investigates whether intrinsic noise acts as a systematic "tuner," altering the fundamental timing and spectral properties of chemical oscillators.

Methodology: The Brusselator Cascade

To explore these effects, the researchers utilized the Brusselator model, a standard theoretical framework for autocatalytic chemical oscillators. They specifically examined a cascade of $N$ downstream-coupled Brusselators, where the output of one stage serves as the input for the next.

Schematic of a cascade of $N$ coupled Brusselator oscillators in a mesoscopic reactor volume $\Omega$.

The system was analyzed using Gillespie’s Stochastic Simulation Algorithm (SSA) to capture the full impact of molecular discreteness. The primary variable was the reactor volume ($\Omega$), which inversely controls the level of intrinsic noise—smaller volumes correspond to higher noise levels. The researchers calculated the Power Spectral Density (PSD) to identify the peak frequency ($\omega_n^m$) and bandwidth ($\omega_n^b$) at each stage $n$.

Key Insights and Results

The study reveals that intrinsic noise fundamentally reshapes the oscillator's frequency spectrum in several ways:

  • Frequency Shifting: Contrary to the Linear-Noise Approximation (LNA), which assumes the peak frequency is independent of volume, the stochastic simulations show that increasing noise (decreasing $\Omega$) systematically lowers the peak oscillation frequency. For a single-stage oscillator, reducing the volume to $\Omega=1$ resulted in a frequency drop from 0.86 to 0.45—a nearly 50% reduction.
  • The Cascade Effect: In a multi-stage chain, this frequency shift is cumulative. In a 20-stage cascade at high noise levels ($\Omega=1$), the frequency plummeted further to 0.28.
  • Bandwidth Dynamics: While noise generally broadens the spectrum (increasing bandwidth) in a single stage, moving down the cascade actually decreases the bandwidth for a fixed volume. This "sharpening" effect means that oscillations become more distinct at later stages of a signaling chain, even as their frequency decreases.

Conclusion and Implications

These findings demonstrate that traditional macroscopic models and linear approximations fail to capture critical spectral shifts in mesoscopic systems. The sensitivity of frequency and bandwidth to noise levels suggests that these spectral features can serve as "fingerprints" for parameter identification. For instance, experimental data from techniques like Fluorescence Correlation Spectroscopy (FCS) could be used to back-calculate reactor volumes or reaction rates more accurately by accounting for noise-induced shifts. Future work may explore how biological systems, such as circadian clocks or signaling pathways, might have evolved to leverage or mitigate these noise-driven frequency changes.

Summary updated

Abstract

Mesoscopic oscillatory reaction systems, for example in cell biology, can exhibit stochastic oscillations in the form of cyclic random walks even if the corresponding macroscopic system does not oscillate. We study how the intrinsic noise from molecular discreteness influences the frequency spectrum of mesoscopic oscillators using as a model system a cascade of coupled Brusselators away from the Hopf bifurcation. The results show that the spectrum of an oscillator depends on the level of noise. In particular, the peak frequency of the oscillator is reduced by increasing noise, and the bandwidth increased. Along a cascade of coupled oscillators, the peak frequency is further reduced with every stage and also the bandwidth is reduced. These effects can help understand the role of noise in chemical oscillators and provide fingerprints for more reliable parameter identification and volume measurement from experimental spectra.
BibTeX
@inproceedings{ramaswamy2011intrinsic,
  author = {Rajesh Ramaswamy and Ivo F. Sbalzarini},
  title = {{Intrinsic noise alters the frequency spectrum of mesoscopic oscillatory chemical reaction systems}},
  booktitle = {Scientific},
  year = {2011},
  doi = {10.1038/srep00154},
}
Figure