Infoscience

Thesis

Acquisition et recalage informatique d'images confocales de tissus denses: application à l'étude de la papille folliculaire de vibrisse de souris

Analysis of the 3D structure of a biological sample is an important step to model cell and tissue dynamics. As a model system, we set up to determine the number of nuclei in a follicular papilla using confocal imaging and 3D computer reconstruction. The follicular papilla is a small mesenchymal ovalshape structure localized at the basis of the hair follicle that is particularly important for hair follicle morphogenesis, and the size of which varies during hair growth. However, little is known on the influence of the hair cycle on the number of cells in a follicular papilla. Fgf5 -/- mice have long hairs (angora phenotype) due to an extended hair cycle, and small dermal papilla. There are several hypotheses to explain the latter observation : there are less cells in Fgf5 -/- dermal papilla than in wild type papilla mutant papilla produces less extra cellular matrix than wild type papilla it is a combination of both To discriminate between these hypotheses, dermal papilla were microdissected from vibrissal follicles occupying the same position on the mystacial pad of same age Fgf5 -/- and wild type mice. Nuclei were stained with a fluorescent dye and the papilla examined under a confocal microscope. Because of dermal papilla thickness, two stacks of images were acquired for each sample (recto and verso acquisitions), registrated using especially implemented computer science program and analyzed. This whole process turned out to be far more complex and cumbersome than anticipated. First, we had to design a special setup in order to be able to make two side scans of samples, second the acquisition time of two stacks of images to cover the entire dermal papilla was rather long, third repositioning and analysis of the images were complex. Nonetheless, the number of nuclei in a dermal papilla was determined with precision. Analysis of dermal papilla obtained at different time of the hair cycle is ongoing. Our work paves the way to the analysis of other biological structures important during skin morphogenesis or regeneration, i.e. the ectodermal placodes. An important lesson of our work is that there is no straightforward imaging and computer science approach when it comes to the biology of complex structures and that the technology needs to be customized to the biological question.

Related material