ETD

• cell mechanics
• correlative microscopy
• super-resolution microscopy
• traction force microscopy

Recent evidence has highlighted the crucial role of cellular mechanical properties in processes such as differentiation, migration, embryonic development, tissue homeostasis, and cancer metastasis. In parallel, high-dose FLASH radiotherapy has revealed potential benefits in cancer treatment, due to its increased sparing effect on healthy tissues compared to conventional radiotherapy. However, the under lying mechanisms remain poorly understood, as most research has focused on biochemical responses, leaving cellular mechanical properties largely unexplored.
Preliminary data suggest a differential post-irradiation rearrangement of cytoskeletal organization in healthy and cancer cells, which can be efficiently investigated at the molecular level using super resolution fluorescence techniques. Since cytoskeletal structural organization is tightly linked to cellular mechanical properties and active force generation, it also becomes relevant to investigate potential modifications in the mechanical behavior of irradiated cells. Unfortunately, combining super-resolution microscopy and force measurements remains challenging due to the experimental constraints. This work aims to bridge the gap by developing a correlative approach integrating AI-assisted super-resolution imaging and traction force microscopy (TFM), enabling the simultaneous reconstruction of high resolution cytoskeletal features and cellular force generation.
The proposed experimental pipeline aims to overcome the intrinsic imaging limitations of single-molecule super-resolution microscopy for live-cell imaging on TFM gel substrates, exploiting the possibility to acquire high-resolution molecular and mechano-structural information directly from widefield imaging.
Despite being at a preliminary stage of development, the methodology allowed the simultaneous reconstruction of super-resolution images of microtubules and traction fields in living HeLa cells. This tool represents a combination of techniques capable of providing impactful insights into radiation-induced mechanical changes by correlating cytoskeletal organization and force generation. As a future perspective, the remaining limitations, mainly arising from AI reconstruction artifacts, could be further mitigated through additional transfer learning training.