Quantitative analysis and mathematical modelling of polarity establishment in C. elegans embryos

Simon Blanchoud

Cell polarity is crucial for the development and homeostasis of metazoan organisms. Whereas proteins required for cell polarity have been identified in several systems, how their function evolves in space and time remains incompletely understood. During my Doctoral thesis, I will address this important question using a multidisciplinary systems biology investigating polarity establishment in the one-cell stage embryo of C. elegans. My objective is to perform a thorough quantitative analysis of the dynamic patterns of the main players of polarity establishment coupled with mathematical modelling of this process.

In order to be able to consistently analyse large datasets, I have been developing, during my first year, a REference COordinate System (RECOS). This tool will allow me to automatically detect, from time-lapse recordings, the cortex of the embryo, standardize it and quantify the fluorescence of tagged proteins with high spatial and temporal resolution. While the detection based only on Differential Interference Contrast (DIC) images still requires some improvements to be comparable to manual tracking, a robust methodology using fluorescent markers has been developed. Even if the accuracy of both methods still needs to be more exhaustively assessed, preliminary results indicate that it can serve as a general framework for accurate and consistent analysis of C. elegans early development.

Using this tool and the relevant mutant strains, I will perform, during the next three years, a large-scale analysis of polarity establishment coupled with mathematical modelling in order to further challenge our understandings of this process.

Figure 1 : Different segmentation steps of the cortex when using the mCherry-PH channel. (A) Initial image (B) Elliptical projection of the attractive term (C) Elliptical projection of the repulsive term (D) Result of the segmentation (blue) on top of the filtered image (E) Magnification of the red area in D. Attractive pixels are blue and repulsive ones red.

Figure 2 : (A) A superposition of the different segmentations used to measure the segmentation performance of the cortex using the mCherry channel. The colour code for the curves is the same as in panel C, the mean of the manual segmentations is in green. (B) Magnification of the area defined by the red area in panel A. (C) Quantification of the mean distance of every segmentation over the whole movie. Note that given the size of the embryo, an error of 3 pixels corresponds to ~2% of the embryo’s radius.