Decellularized fennel and dill leaves as possible 3D channel network in GelMA for the development of an adipose tissue model.

The development of 3D scaffold-based models would represent a great step forward in cancer research, offering the possibility of predicting the potential response to targeted anticancer or anti-angiogenic therapies. As regards, 3D models require proper materials, which faithfully recapitulated extracellular matrix (ECM) properties, adequate cell lines, and an efficient vascular network. The aim of this work is to investigate the possible realization of an 3D scaffold-based model of adipose tissue, by incorporating decellularized 3D plant structures within the scaffold. In particular, in order to obtain an adipose matrix capable of mimicking the composition of the adipose tissue, methacrylated gelatin (GelMA), UV photo-crosslinkable, was selected. Decellularized fennel, wild fennel and, dill leaves have been incorporated into the GelMA hydrogel before crosslinking, to mimic a 3D channel network. All leaves showed a loss of pigmentation after the decellularization with channel dimensions ranging from 100 to 500 µm up to 3 μm, comparable with those of human microcirculation (5-10 µm). The photo-crosslinking process was not affected by the embedded plant structures in GelMA hydrogels. In fact, the weight variation test, performed on hydrogels with or without decellularized leaves showed a weight loss in the first 96 h, followed by a stability plateau up to 5 weeks. No cytotoxic effects were detected comparing the three prepared GelMA/D-leaf structures; moreover, the ability of the samples to stimulate differentiation of 3T3-L1 preadipocytes in mature adipocytes was investigated, and cells were able to grow and proliferate in the structure, colonizing the entire microenvironment and starting to differentiate. The developed GelMA hydrogels mimicked adipose tissue together with the incorporated plant structures seem to be an adequate solution to ensure an efficient vascular system for a 3D model. The obtained results showed the potentiality of the innovative proposed approach to mimic the tumoral microenvironment in 3D scaffold-based models.