Understanding heterogeneity of gene expression in the innate immune system through mathematical modelling

Abstract

Mammalian gene transcription is rigorously regulated through various complex mechanisms to ensure accuracy and prevent errors in the synthesis of RNA molecules. Paradoxically, transcriptional cell-to-cell variability, despite genetic homogeneity, has been broadly documented, including in the innate immune system where precise responses are crucial upon encountering pathogens. The stochastic nature of transcription is believed to be a driver of this variability; the transcriptional process of most genes involves random transitions between inactive and active gene states, leading to the production of messenger RNA (mRNA) in a burst-like manner, giving rise to inherent heterogeneity in gene expression at the single-cell level. While this phenomenon has been studied for decades, it remains unclear whether, and how, single-cell variability in the innate immune system is controlled in response to different environmental conditions. Combining data analysis, statistical inference, and state-of-the art mathematical modelling with data from various wet lab techniques, this thesis presents interdisciplinary research on characterising cellular variability in the innate immune Toll-like receptor system and provides new understanding of the underlying control mechanisms. The first chapter introduces the notion of gene expression heterogeneity with an overview of the existing relevant literature and discusses various approaches from both the biological and mathematical fields that have been or can be potentially employed to study this phenomenon. Chapter two focuses on analysis and mathematical modelling of single molecule fluorescence in situ hybridisation count data of inducible immune genes. Gene-specific linear mean-variance relationships of mRNA transcript counts across a range of immune conditions, and their corresponding bursting characteristics, are established. Chapter three validates the linear constraints, and their underlying transcriptional bursting modulation, globally and demonstrates, through stochastic modelling of single-cell RNA-seq counts of 96 immune genes, an association between high variability levels and increased complexity of transcriptional regulation. In addition, evolutionary differences in response variability across several species are characterised. Chapter four provides evidence that heterogenous single cell innate immune responses are in part imprinted over multiple cell divisions. Overall, this thesis offers novel tools and findings that take us a step forward in understanding cell-to-cell variability in the innate immune system with broader implications for other biological systems.

Date of Award	1 Aug 2024
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Mark Muldoon (Supervisor) & Pawel Paszek (Supervisor)