[Paper Review] Ubiquity of Log-normal Distributions in Intra-cellular Reaction Dynamic

Chikara Furusawa, Takao Suzuki+3 more|ArXiv.org|Mar 29, 2005

Gene Regulatory Network AnalysisBiochemistry, Genetics and Molecular Biology15 references3 citations

TL;DR

This paper proposes that intracellular chemical abundances in growing cells follow a log-normal distribution due to multiplicative stochastic fluctuations in catalytic reaction networks, with a linear relationship between mean and standard deviation. The authors demonstrate this via a theoretical model near critical self-replication states and confirm experimentally using fluorescent protein distributions in E. coli, revealing a universal statistical law underlying cellular heterogeneity and plasticity.

ABSTRACT

The discovery of two fundamental laws concerning cellular dynamics with recursive growth is reported. First, the chemical abundances measured over many cells are found to obey a log-normal distribution and second, the relationship between the average and standard deviation of the abundances is found to be linear. The ubiquity of the laws is explored both theoretically and experimentally. First by means of a model with a catalytic reaction network, the laws are shown to appear near the critical state with efficient self-reproduction. Second by measuring distributions of fluorescent proteins in bacteria cells the ubiquity of log-normal distribution of protein abundances is confirmed. Relevance of these findings to cellular function and biological plasticity is briefly discussed.

Motivation & Objective

• To identify universal statistical laws governing fluctuations in intracellular chemical abundances across cell populations.
• To explain why log-normal distributions emerge in cellular dynamics despite the expectation of sharp, low-fluctuation distributions.
• To investigate the relationship between mean abundance and standard deviation in cellular protein expression.
• To validate theoretical predictions with experimental measurements of fluorescent protein levels in E. coli.
• To explore the implications of these laws for cellular function, adaptation, and evolutionary plasticity.

Proposed method

• Modeling intracellular dynamics using a catalytic reaction network with stochastic fluctuations, where reaction rates are subject to multiplicative noise.
• Deriving the Langevin equation for log-abundance dynamics: d(log n_m)/dt = a̅ + η(t), leading to a normal distribution of log-abundance and thus a log-normal distribution of abundance.
• Simulating the reaction network near the critical state (D ≈ D_c) to observe emergence of log-normal distributions and linear mean-standard deviation scaling.
• Experimentally measuring fluorescent protein abundances in E. coli under various promoters and genomic locations to test distribution universality.
• Analyzing the data using logarithmic scaling and statistical tests to confirm log-normal fit across diverse expression conditions.
• Comparing results with existing gene expression data and contrasting findings with studies under non-steady growth conditions.

Experimental results

Research questions

• RQ1 Why do intracellular chemical abundances across a population of cells follow a log-normal distribution rather than a Gaussian distribution?
• RQ2 What is the origin of the linear relationship between the mean and standard deviation of protein abundances in growing cells?
• RQ3 Does the log-normal distribution of protein abundances persist across different promoters and genomic integration sites in E. coli?
• RQ4 How do fluctuations in catalytic reaction networks give rise to universal statistical laws in cellular dynamics?
• RQ5 To what extent do these statistical laws break down during non-steady growth phases or cell differentiation?

Key findings

• The distribution of fluorescent protein abundances in E. coli across a population of cells follows a log-normal distribution, independent of promoter strength or genomic location.
• A linear relationship between the mean and standard deviation of protein abundances was experimentally confirmed in E. coli under exponential growth conditions.
• Theoretical modeling shows that log-normal distributions emerge due to multiplicative noise propagation in catalytic reaction networks near the critical self-replication state (D ≈ D_c).
• The log-normal distribution implies significantly larger relative fluctuations than Gaussian distributions, challenging the assumption of small fluctuations in current models of cellular heterogeneity.
• The laws break down in non-steady growth phases, where distributions may become bimodal or deviate from log-normality, indicating a link to biological plasticity.
• The findings suggest that the observed statistical laws are a consequence of efficient, self-replicating dynamics near criticality, providing a universal basis for understanding cellular fluctuations.