Cell fate choice

Understanding how heterogeneity, stochastic differentiation and cell sorting can result in robust developmental patterning

Slide1Embryonic development is a remarkable feat of biological reproducibility. It is generally assumed that such robustness must reflect highly stereotypical or deterministic underlying mechanisms, with little room for disorder or heterogeneity. It is even intuitive to imagine heterogeneity should be suppressed. However, recent observations of cell fate choice in simple microbial systems challenge this idea, and even suggest heterogeneity in cell signalling and responses (also termed noise and stochasticity) can be evolutionarily advantageous (e.g. bet-hedging). These findings therefore have potentially field changing implications, because they raise the question of whether heterogeneity can also be harnessed or regulated to play a role in cell fate choice and the generation of developmental pattern in multicellular organisms.

Slide1To address this, we use Dictyostelium discoideum as a model system, because it permits a uniquely powerful combination of approaches to be applied to the question. Firstly, developmental patterning in Dictyostelium is based on ‘salt and pepper’ differentiation followed by sorting out, and therefore heterogeneity has been proposed to play a pivotal role. Secondly, Dictyostelium is amenable to forward and reverse genetic manipulation, is easily and rapidly grown in the lab to biochemical scales, whilst its relatively small number of defined cell types can be tracked in vivo by live cell imaging during development. We are thus in a unique position to take this opportunity and generate the first integrative ‘top to bottom’ understanding of how heterogeneity, stochastic differentiation and cell sorting result in robust developmental patterning.

For example, we recently employed a novel forward genetic approach in DictyosteliumSlide1 discoideum. These studies revealed that the Ras-GTPase regulator gefE is required for normal lineage priming and salt and pepper differentiation. This is because Ras-GTPase activity sets the intrinsic response threshold to lineage specific differentiation signals. Importantly, we found that although gefE expression is uniform, transcription of its target, rasD, is both heterogeneous and dynamic, thus providing a novel mechanism for heterogeneity generation and position-independent differentiation.

Current research topics include:

1. How are the gene networks that regulate differentiation affected by noise and stochasticty?

We are using forward genetics, single cell RNA-seq and reverse genetics  to identify novel heterogeneously expressed genes that affect cell fate choice and determine the molecular source of heterogeneity.

2. Do differentiating cells all follow the same path towards the differentiated state?

Until recently, however, it has been impossible to follow the behaviour of entire gene networks in individual cells, or to follow their temporal changes in activity in individual cells as they differentiate along different linages. Single cell gene expression analysis, together with novel computational reconstruction of gene network dynamics provides this opportunity.


3. How does heterogeneity affect responses to regulators of cell fate?

We are using forward genetic and next generation sequencing approaches to define a small molecule regulated signalling pathway and determine how it is affected by heterogeneity.

4. How can sorting out can generate robust developmental pattern from heterogeneous differentiation?

RNA-seq and ChIP-seq is being used to identify genes that are switch on or off during early differentiation. Functional reverse genetic analyses and live cell imaging are being used to determine how gene expression changes affects differential chemotaxis, adhesion and cell sorting.

5. Functional analyses of gene networks identified.

Gene expression, in vivo imaging, signalling, chemotaxis, adhesion, subcellular localisation and biochemical function data are being integrated from wild type and knockout mutants to define robust gene networks.