Protoplasts under stress: Investigating the effect of compression on protoplasts using microfluidics

Protoplasts are plant cells which have had their cell walls enzymatically removed. They can be isolated from almost any plant tissue of many different plant species. The key feature of protoplasts is that they can be induced to regenerate from a single cell back into a full plant. The regeneration of a protoplast into a specific tissue type or a whole plant is a complex, time and labor-consuming task with a yield that is not easy to predict, as all the factors affecting the regeneration efficiency, both internal and external, are not yet fully understood.

The complexity of the regeneration process makes it difficult to investigate the sub-processes happening therein. There are phenomena that still bear exploration even in the early, single-cell stage of regeneration.

In this thesis, we investigated whether the application of mechanical compression on a single protoplast in a finely controlled microfluidic environment can affect their divisions and cell wall regeneration. Microfluidics served here as an engineering tool, that enabled us to reproducibly apply stimuli to cultured cells and closely observe the result. In this thesis, microfluidic platforms were used to investigate the initial divisions of freshly isolated protoplasts of Nicotiana tabacum, a widely used model plant species. Uniquely, our devices allowed us to affect these protoplasts with mechanical compression without also introducing other known factors of influence like osmotic stress. Our microfluidic device was designed to be compatible with high-content imaging platforms, allowing us to generate large time-resolved image datasets of cell behavior. Due to the large quantities of data generated on this platform and the sometimes subtle effects that compression has on the regeneration process, we recognized a strong need for automated tools for data analysis. Additionally, such a rich dataset could serve as the input for a prediction model that would make it possible to predict the outcome of the regeneration of a single protoplast based on its initial image. Such a model would be of great value both for fundamental biology, as it can help identify specific features crucial for the regeneration success, and the crop industry, as then it can significantly improve the efficiency of the regeneration process.