Phd in IINNOVATION IN IMMUNO-MEDIATED AND HEMATOLOGICAL DISORDERS

Thesis title: An organoid-based model to study Innate Lymphoid Cells in Colorectal Cancer

Innate lymphoid cells (ILCs) emerged as central innate immune mediators during gastrointestinal homeostasis, inflammation and tumorigenesis. The role of ILCs in cancer is still controversial since they have been associated with either pro- and antitumor activities. The aim of this study was to perform a comprehensive evaluation of phenotype and functions of natural killer cells (NK) and ILC subsets in the context of colorectal cancer (CRC) focusing on the direct impact of tumour on these lymphocytes. We generated intestinal organoids derived from Apcflox/flox/KrasG12D/+/Trp53R172H/flox/Smad4flox/flox or KrasG12D/+/Trp53loxP mice which we transduced for Cre-recombinase expression using a retroviral vector. The obtained organoids have been named as ‘AKPS’ and ‘KP’ based on the mutations they harbour and used for in vivo and in vitro experiments. We developed an orthotopic animal model suitable for the study of innate immune response at early stages of tumorigenesis because of its capability to develop CRC tumors within few weeks. Single cell RNA-seq analysis on sorted CD45.2+ lymphocytes infiltrating the ‘AKPS’ tumor, 1 week after injection, revealed the presence of NK/ILC1, B cells and an enrichment of T cells within the tumor. CD8 T cells are the major component infiltrating the tumor. We found an effector/memory phenotype based on the expression of CD44 and CD62L, which was confirmed by flow cytometric analysis, but also an enrichment of exhausted CD8+ T cells. Single cell RNA-seq analysis performed in the NKp46+ compartment revealed 9 clusters, of which 7 were defined as NK cells, 1 ILC1 cluster and 1 ILC1-like cluster. NK cells infiltrating the tumor showed different staged of maturation with both mature and functional NK cells and immature NK cells. Flow cytometric analysis on the ‘AKPS’ infiltrate at 1 and 2 weeks after engraftment confirmed that NK cells are the most abundant population of ILCs infiltrating the CRC with an absence of the ILC1, ILC2 and ILC3 populations and we confirmed the presence of ILC1-like cells which populated only the tumor but not the tumor-free large intestine lamina propria. By flow cytometric analysis performed on subcutaneous injection model of the ‘AKPS’ organoids and the more canonical AOM/DSS model revealed the same distribution pattern for the ILCs suggesting that the identity of ILCs populations infiltrating CRC is not affected by both tissue-specific factors and chronic inflammatory stimuli. Flow cytometric analysis performed on the orthotopic injection of the ‘KP’ tumor revealed that ‘KP’ CRC generates an immune infiltrate more enriched of total innate cells and ILCs than ‘AKPS’ CRC thus supporting the idea that tumor mutation burden can impact on innate immune response. To evaluate the possible direct effects of the tumoral component on NK cells, we set up an in vitro co-culture system where WT splenic NK cells were cultured for 24 hours with ‘AKPS’ organoids. Bulk RNA-seq analysis demonstrates the regulation of several genes by ‘AKPS’ organoids-derived soluble factors. Interestingly, among upregulated genes, we found also Irf8 and Irf8 target genes. Moreover, our previous scRNA-seq analysis highlighted the presence of a mature and fully functional NK cell subset also expressing high levels of Irf8 among NK cells infiltrating ‘AKPS’ CRC. IRF8 is an important transcription factor for NK cells functions and maturation and we found that IRF8 protein levels are increased after 48 hours of co-culture with ‘AKPS’ organoids. It is well established that IRF8 expression in immune cells can be induced by several cytokines, including type I interferons, IL-12 and IL-27. Since FACS analysis highlighted the capability of ‘AKPS’ organoids to produce IL-27, we investigated the possible role of this cytokine in the regulation of IRF8 expression in our experimental setting. To this aim, we used an anti-IL-27 antibody which blocks the cytokine binding to its receptor. We found that blockade of IL-27 signalling did not affect basal IRF8 expression in NK cells but it could revert its upregulation by tumor organoids. These data demonstrate that IRF8 and IL-27 signaling might play an important role in the regulation of NK cell activity in CRC. In summary, using tumor intestinal organoids derived from Apcflox/flox/KrasG12D/+/Trp53R172H/flox/Smad4flox/flox or KrasG12D/+/Trp53loxP mice, we developed an orthotopic animal model suitable for the study of innate immune response at early stages of tumorigenesis because of its capability to develop CRC tumors within few weeks. In ‘AKPS’ tumors, we found a lymphocytic infiltrate enriched of exhausted CD8+ T cells and immature and hypofunctional NK cells suggesting a role of these cells in immune surveillance. Moreover, we provided evidence that colorectal cancer cells are able to regulate NK cell differentiation/function via IRF8-upregulation by IL-27.