Dynamic cell differentiation in multicellularity with specialized cell types

Abstract

The specialization of cells is a hallmark of complex multicellularity. Cell differentiation enables the emergence of specialized cell types that carry out separate functions previously executed by a multifunctional ancestor cell. One view is that initial cell differentiation occurred randomly, especially for genetically identical cells, exposed to the same life history environment. How such a change in differentiation probabilities can affect the evolution of differentiation patterns is still unclear. We develop a theoretical model to investigate the effect of stage-dependent cell differentiation - cells change their developmental trajectories during a single round of development via cell divisions - on the evolution of optimal differentiation patterns. We found that irreversible differentiation - a cell type gradually losing its differentiation capability to produce other cell types - is more favored under stage-dependent than stage-independent cell differentiation in relatively small organisms with limited differentiation probability variations. Furthermore, we discovered that irreversible differentiation of germ cells, which is the gradual loss of germ cells ability to differentiate, is a prominent pattern among irreversible differentiation patterns under stage-dependent cell differentiation. In addition, large variations in differentiation probabilities prohibit irreversible differentiation from being the optimal differentiation pattern. Author summaryThe differentiation of cells into different branches is a characteristic feature of multicellular organisms. To understand its origin, the mechanism of division of labour was proposed, where cells are specialized at distinct tasks. In previous models, a cell type is usually assumed to produce another cell type with a fixed probability which is referred to as stage-independent differentiation. However, it has been argued that cell differentiation is a dynamic process in which cells possess changing differentiation capabilities during the different stages of an organisms development. Stage-dependent differentiation exhibits more diverse patterns of development than differentiation with fixed probabilities, thus it can lead to novel targets of selection. How does stage-dependent differentiation impact the evolution of optimal differentiation patterns compared with stage-independent one? To address this question, we built a stage-dependent cell differentiation model and classified differentiation patterns based on the cells differentiation capability in their last cell division. We investigate how stage-dependent differentiation probabilities impact the evolution of the optimal differentiation pattern, which acts on the fitness of an organism. As we take the growth rate as a proxy of an organisms fitness, we seek the "optimal strategy" that leads to the fastest growth. Our numerical results show that irreversible differentiation which gradually loses its differentiation capability, is favored over stage-independent differentiation in small organisms. Meanwhile, irreversible differentiation wont be optimal when there are no constraints on the changes of stage-dependent differentiation probabilities between successive cell divisions.