Probabilistic regulation of cellular heterogeneity
The massive inventory of cell types and regulatory signals uncovered by experiment has fostered the ‘deterministic’ view that each different cell type has its own unique guiding set of signals. Many experiments support the idea that if two cells have a different fate they must have had a different history that can be determined. In contradiction to this view we have proposed a probabilistic explanation for cell heterogeneity that is supported by experiment and promises to greatly simplify our understanding of how the myriad alternative cell fates arise.
Cells of the immune system are highly regulated by signals received from numerous cell surface receptors. These signals are often received simultaneously and regulate multiple aspects of behaviour including proliferation, survival and differentiation decisions. Other influences include the affinity and concentration of antigen. We are using the quantitative methods mentioned above to dissect the manner in which such signals are integrated by both B and T cells. We find that lymphocytes behave as if composed of separate independent ‘machines’ with stochastic features that govern times to divide and times to die and the rate of differentiation in association with division number. As a consequence of the stochastic variation displayed by these ‘machines’ many alternative outcomes are possible for individual cells without explicit coding. However, the net response of the population is highly predictable. Cytokines that affect proliferation rate and survival often ‘add’ together in quantitative manner yielding surprisingly large effects on final cell number that can explain costimulation, and place the classic two signal theory of T cell activation on a quantitative basis. Related rules can be applied to differentiation.
Our modeling framework illustrates how a small number of simple cellular rules built on probabilistic principles can explain a broad range of complex immune behaviour.