Inference of Differentiation Decision Times from Blood Stem Cell Genealogies
Final Report Abstract
During differentiation, stem cells and their progeny cascade through multiple lineage decisions, from an initially multipotent state over progenitor states to mature functional cells. Many decisions are assumed to be binary and realized by a genetic toggle switch, a simple cellular memory device consisting of two genes that inhibit each others expression. The most prominent example is the mutual inhibition of Gata1 and PU.1, two transcription factors responsible for the development of erythroid and myeloid blood cells from blood stem cells. While being an intriguing, simple, and well-studied mechanism of cell fate choice, no definite experimental proof of the toggle switch paradigm exists: It is unclear whether a toggle switch actively determines the cell fate or if it merely stabilized a previously chosen fate. Single cell time lapse microscopy allows to monitor cell fate decisions in single cells continuously over time and to address these questions. In this DFG funded project, we first developed a method based on generalized linear models to infer potential cell-extrinsic features (e.g. local cell density) causing differentiation events observed in cellular genealogies. We analyzed the required sample sizes and the influence of cell tracking errors on the results. Furthermore, we utilize genealogical information to validate our model, i.e. to test whether the model is able to explain the correlation structure observed in sister cells. We then combine the ideas of cell-intrinsic and cell-extrinsic processes impacting on cell fate choice into a single model to explain correlated cell fate marker onsets in genealogies. Motivated by our findings in the stochastic toggle switch, we assumed that differentiation is a point process potentially modulated by external factors while the onset of the cell fate marker in response to differentiation is delayed due to an intrinsic stochastic gene expression process. We developed an inference method tailored to this model which allowed us to predict the time point of differentiation from the observed correlations of marker onsets in the genealogies. After testing the method on various synthetic datasets, we applied it to the toggle switch and found that PU.1 dynamics at the predicted time points of differentiation deviate significantly from those predicted by the PU.1/Gata1 toggle switch model . In summary, we rejected the Pu.1/Gata1 toggle switch paradigm of hematopoietic differentiation and provided tools for the analysis of stem cell decisions from time lapse microscopy data. Our interdisciplinary contributions were awarded with the Erwin Schrödinger Prize from the Helmholtz Association and the CSB2 Prize in Systems Biology awarded by the Council for Systems Biology in Boston together with Merrimack Pharmaceuticals, Boston, USA.
Publications
- Inference of spatiotemporal effects on cellular state transitions from time-lapse microscopy. BMC Syst. Biol. 9, 61 (2015)
Strasser, M. K., Feigelman, J., Theis, F. J. & Marr, C.
(See online at https://doi.org/10.1186/s12918-015-0208-5) - Lineage marker synchrony in hematopoietic genealogies refutes the PU.1/GATA1 toggle switch paradigm. Nat. Commun. 9, 2697 (2018)
Strasser, M. K. et al.
(See online at https://doi.org/10.1038/s41467-018-05037-3)
