Follow the leader: collective cell migration in cancer and developmental biology

Thesis

Collective cell migration is an essential phenomenon in biology, occurring in diverse processes such as neural crest stem cell migration and angiogenesis (the formation of new blood vessels from pre-existing ones). One particular mode of collective cell migration involves two distinct phenotypes: leader cells, which migrate towards external biophysical signals, and follower cells, which move in response to leader-created cues. Mechanisms underpinning leader-follower migration remain poorly characterised in general, partly because they occur over multiple spatial and temporal scales. Mathematical modelling can address such issues, as it can be used to tractably evaluate different biological hypotheses, connect individual cell interactions to the behaviour of collectives, and guide the design of in vivo experiments. Open questions remain, however, regarding the validity of certain mathematical models and the conditions for which they accurately capture data from biological experiments.

In this thesis, we address different questions relating to leader-follower migration within the contexts of angiogenesis and neural crest cell migration. By comparing agent-based model simulations for angiogenesis to solutions from both a classical phenomenological framework and an alternative coarse-grained model, we identify parameter regimes for which the phenomenological framework accurately describes results of the other two systems. In doing so, we uncover implicit physical assumptions that underpin the continuum models. We also collaborate with experimental colleagues to develop a new agent-based model for cranial neural crest cell migration that incorporates cell remodelling of, and movement through, a naive extracellular matrix. We perform a global sensitivity analysis on the model and rank the mechanisms that most dominate collective cell behaviour. Further tests of this model suggest that migration is most efficient when leader and follower cell phenotypes are not identical. We conclude by providing, to our knowledge, the first detailed investigation of mechanisms that ensure robust neural crest cell migration.

Authors: Martinson, WD

• DOI: 10.5287/ora-dxzonajj8
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: agent based model; coarse graining; contact guidance; extracellular matrix remodelling; multiscale modelling; sensitivity analysis; neural crest cell; asymptotic analysis; collective cell migration; angiogenesis
• Subjects: Biology--Mathematical models