Multiscale Modelling of Stem Cell Population Dynamics

  • Daniel Ward

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

Stem cells, with their propensity for pluripotency and self-regulation, have been the focus of intense research over the past decades. Understanding how the pluripotent phenotype is governed, across scales ranging from the cells' underlying genetic components to population-level interactions, is key to progressing from cultures of thousands of stem cells to maximal cell cultures, or entire tissues, from a small starting population. Stem cells are the precursors to all living cell types, existing during the embryonic development stage, but they can also be found throughout the bodies of all developed living creatures, for example within the bone marrow or residing at the base of an intestinal crypt. The differential behaviours of embryonic and adult stem cells, combined with complex underlying genetic regulation and environment specific dynamics, present a number of open and pressing questions. In this thesis, we present mathematical and computational modelling approaches which allow us to elucidate the key dynamics of three different stem cell populations, to shine a light on the differential effects of culture conditions on each cell type, as well as to highlight the usefulness of modelling for driving experimental research and development of culture protocols. In the first part of the thesis, we develop a delay differential equation model to capture the culture-dependent dynamics of a homogeneous population of human haematopoietic stem cells, and show that a key culture protocol component, epo, has two response phases in which it differentially affects cell proliferation and differentiation. In the second part, we introduce agent-based modelling as a method for capturing the population dynamics of mouse embryonic stem cells (mESCs) in different culture conditions. We showed that linking the MycN component of the mESC gene regulatory network to the cell cycle captures both the subcellular distributions of key proteins and growth dynamics in vitro. Finally, we develop a multiscale agent-based model of intestinal crypts, that couples ordinary differential equation modelling of subcellular kinetics to a cell-based description of cell movement, proliferation, and contact inhibition (CI). This enables us to recapitulate tissue level dynamics of intestinal crypts, as well as to present an alternative approach to describing the formation of the Wnt expression gradient. We showed that cross-talk between the Hippo and Wnt signalling pathways is able to affect CI and that CI is likely significantly reduced in intestinal crypt mutations. Together, this research shows the effectiveness of modelling, across physical and temporal scales, to recapitulate in vitro and in vivo stem cell dynamics, as well as to capture the contributions of key behaviours such as proliferation and differentiation to healthy and dysplastic population growth.
Date of Award19 Mar 2019
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorLucia Marucci (Supervisor) & Martin E Homer (Supervisor)

Keywords

  • Multiscale modelling
  • Agent-based modelling
  • Computational biology
  • Mathematical biology

Cite this

'