Implementation and investigation of artificial skin cancer models using bioimaging and next generation sequencing

In recent years, the scientific and ethical focus has increased on the reduction of animal experiments and their replacement with alternative models. Skin models are 3D in vitro reconstructs developed from one or more skin cell types, and they are important substitutes for animal models that can be used to test drugs and cosmetics as well as to investigate various scientific topics such as anti-cancer treatments. In December 2022, a law change allowed the US Food and Drug Administration (FDA) to approve drugs that have not been tested on animals and this has rendered skin model development more relevant than ever in order to obtain good skin substitutes that reliably mimic native skin. This dissertation is divided into three parts, each concerning a different aspect of skin model research. In the first part, we study an epidermis model known as rat epidermal keratinocyte (REK) organotypic culture (ROC). The ROC model is well-established in the literature, and the aims of our study were to confirm the reproducibility of the ROC model across different laboratories and furthermore to study ROC using live cell imaging by establishing stable fluorescent REK cell lines. Our results show a successfully developed ROC model and demonstrate its reproducibility and robustness by obtaining morphological similarity and barrier functions comparable to those seen in previous ROC studies. We use live cell imaging and a homemade transwell device to record time-lapse videos of ROC developing at the airliquid interface, and we were able to observe signs of early stratification. The transwell device that we present is applicable to analyzing various tissues. In the second part, we develop melanoma skin cancer models and use RNA sequencing and bioimaging techniques to compare them to regular skin models without cancer cells. We find that the melanoma skin models accurately reflect the gene expression patterns observed in native skin and regular skin models. 72-hour treatment of a melanoma skin model with the anti-cancer drug vemurafenib inhibited tumor growth, and altogether, our data indicated that our model might be a reliable model to be used for anti-melanoma drug development. The third part of the dissertation investigates the use of chimeric antigen receptor (CAR) T cell therapy against melanoma in a collagen matrix to provide a 3D environment that better reflects native conditions in comparison to the commonly used 2D cocultures. We use a CD19 CAR as a proof-of-concept and employ live cell imaging to track the CAR T cell interactions with cancer cells. Our tracking data revealed that the CAR T cells tend to have a lower mean track speed and interact with cancer cells for longer durations compared to control T cells. 72-hour CAR T cell treatment of melanoma spheroids in the collagen matrix was partially successful and might be improved by optimization of the protocol. Our 3D model is applicable to any CAR and tumor type of interest and could aid in improving CAR T cells, especially towards better treatment of solid tumors in complicated tumor environments. Collectively, this dissertation presents different areas in which skin models are used as substitutes for animal skin and evaluates how accurate they mimic native human skin. The future development and optimization of 3D models is an important step towards more accurate and ethical conditions for in vitro studies.