Deciphering the self-organizing principles governing cellular heterogeneity in brain tumors
Unlike developing tissues, tumors have long been considered as highly disorganized tissues. However, recent studies now suggest that some of them may be more organized than initially envisaged, composed of different cell types produced by the recapitulation of embryonic genetic programs. It remains unclear how embryonic genetic programs are coopted to promote tumor growth. Using simple brain tumors in Drosophila, we have demonstrated that the growth and cellular composition is predictable and driven by a fine-tuned hierarchical scheme of cell divisions deployed along an early embryonic-like genetic program (DOI: 10.7554/eLife.50375). Moreover, we have observed a typical spatial organization of tumor cells possibly suggesting robust underlying self-organizing rules. Our aim is to investigate whether spatial cellular organization, predictable cell composition and tumor growth rate result from self-organizing rules appearing upon tumor initiation leading to the cooption of the embryonic-like program.
Tumor self-organization, numerical simulation, Monte-Carlo simulations, confocal microscopy, Drosophila genetics
Our objective is to investigate whether the spatial organization of cells observed in tumors reflects self-organizing principles that determine the cellular composition and growth rate of brain tumors. A 3D numerical model that we have recently developed has generated predictions that need to be investigated experimentally. For this purpose, the candidate will generate new quantitative data from the live imaging of growing tumors and integrate them into the model until the overarching self-organizing rules can be deciphered.
Proposed approach (experimental / theoretical / computational)
The candidate will perform both experimental and computational work. Recent simulations predict that the observed spatial organization and composition of tumor cells can only be achieved if the various cells of the tumor exhibit differential adhesion combined with negative feedback mechanisms between cells.
To investigate and validate these predictions, the candidate will perform a series of measurements obtained from the live imaging of Drosophila tumors using confocal microscopy to gain quantitative insights of cell dynamics and cellular organization.
With the new quantitative data, the candidate should be able to refine the in silico model and generate new predictions. He/she will thus need to get acquainted with the recently developed numerical model that uses a cell-based Monte Carlo scheme to simulate the 3D growth of hierarchical tumors.
The candidate will be co-supervised by a biologist (Cédric Maurange) and a physicist (Raphaël Clément), both localized in different teams of the Institute of Developmental Biology of Marseille (IBDM). He/she will share the time between the two labs.
In the biology lab, the candidate will familiarize with basic rules of tumor biology and Drosophila genetics, perform live imaging using confocal microscopy, and data analysis.
In the modeling lab, he will get acquainted to the 3D numerical model of tumor growth, learn how to modify the script, implement novel functions, and generate novel predictions.
Thus, the candidate will be confronted to a multidisciplinary environment and will himself gain expertise in experimental biology, imaging and numerical modeling.
The candidate should have a Master (or international equivalent) in physics or biophysics, and a robust background in numerical simulations. A strong interest for life sciences is mandatory. Although basic notions of cell biology will be an asset, they are not essential. However, the candidate should be ready to embark on experimental work involving Drosophila genetics and live imaging by confocal microscopy.