ORGANOID MODELS TO STUDY THE CROSSTALK BETWEEN TUMOR CELLS AND TUMOR INFILITRATING LYMPHOCYTES

Recent advances in 3D culture technology allow embryonic and adult mammalian stem cells to generate organoids in vitro, which reflect key structural and functional properties of organs they originate (Clevers H., Cell, 2016). For this reason, they represent a powerful tool to study human physiological and pathological processes, in particular to investigate complex processes like tumorigenesis and tumor growth, resembling the in vivo mechanisms. In particular, in tumor contest neoplastic cells activate several strategies to escape from immunesurveillance, such as the recruitment of immune cells with immunosuppressive functions. In this regard, CD4+ T regulatory cells (Tregs), physiologically engaged in the maintenance of immunological self-tolerance and immune homeostasis, are potent suppressors of effector cells and found at high frequencies in various types of cancer. A recent transcriptome analysis performed in our lab (De Simone M. et al., Immunity, 2016) revealed that tumor-infiltrating Tregs, isolated from CRC (colorectal cancer) and NSCLC (non-small cell lung cancer) patients, expressed a unique and specific gene signature, correlated with patients\u2019 survival. In line with our findings, non-lymphoid tissue infiltrating Tregs can exhibit specific phenotypes and transcriptional profile involved in glucose metabolism, tissue repair and muscle regeneration, far from their well-established suppressive roles (Cipolletta D. et al., 2012; Arpaia N. et al., 2015). Thus, our work wants to evaluate the immune dependent and independent function of tissue- infiltrating Tregs, exploiting a co-culture model with normal and colon cancer- derived organoids. This approach could be suitable to recapitulate primary tumorigenesis, cancer microenvironment effect on Tregs recruitment and phenotype and, vice versa, infiltrating Tregs influence on tumor onset, growth and tissue homeostasis. In order to answer our biological questions by using organoid model, we first of all derived organoids starting from CRC patients\u2019 biopsies, according to protocol published by Sato T. et al. (2011). We generated a biobank of 20 and 28 human- normal and tumoral colon- derived organoid lines, respectively, from tumoral biopsies and the adjacent normal mucosa of patients affected by colorectal cancer. In particular, these organoid lines can be propagated, frozen and defrosted like any tumour cell line, morphologically recapitulating the cellular composition and architecture of colon primary tissue and, importantly, representative of all CRC molecular subtypes. Moreover, our RNA-seq bulk analysis on our organoid lines unveiled that they are very stable from the transcriptional point of view during passages and preserve the inter-individual heterogeneity. Furthermore, our ChIP-seq data showed that our library resembled the same epigenetic landscape of colorectal primary tumour; in this context our study represents an innovative approach to investigate epigenetic processes leading CRC development and progression. Considering that Tumor-infiltrating- Tregs (TI- Tregs) were found to express a peculiar gene signature (De Simone et al., 2016; Plitas et al., 2016), we decided to exploit organoid model to better elucidate processes leading the development and the recruitment of these cells at the tumour site. To this end, we proceeded with the establishment of Tregs- PDO co-culture, assessing the feasibility of this system and evaluating Tregs viability in organoid culture conditions, without observing any significant differences in their survival. Moreover, we evaluated their ability to migrate into 3D organoid structure, revealing their capacity to overcome firstly the obstacle represented by Matrigel (which mimics extracellular matrix) and then creeping into the 3D architecture of organoids, recapitulating the same behaviour they have when recruited at the tumour site. An important evidence about the possible influence of tumoroids on Treg phenotype came from our preliminary co-culture experiment, revealing that PDO-Tregs co-culture upregulated the expression of PDL1 (one of the genes belonging to TI-Treg signature (De Simone et al., 2016)) specifically on Treg but not on Tconv cells. On the other hand, with our co-culture preliminary experiment we observed that expanded TI-Tregs enhanced PDL1 expression on tumoroid cells, which not happened when organoids were co-cultivated with Tconv cells, suggesting the possible influence of Tregs on the expression of specific molecules on cancer cell surface. In conclusion, the exploitation of TI-Treg-PDO co-culture could shed light on the interplay between tumor and TI-Tregs, giving the opportunity to potentially interfere in this crosstalk and design a peculiar anticancer therapy targeting TI-Tregs in a personalized manner