Human pluripotent stem cell-derived model of perineural tumor microenvironment

Melanoma is a tumor of the melanocytic lineage and the most aggressive skin cancer because of its high metastatic potential. The tumor microenvironment of the primary melanoma is composed of keratinocytes, adipocytes, fibroblasts, immune cells, blood vessels and nerves. Nowadays, it is well known that each cellular compartment of the tumor microenvironment reacts to the cancer cells and either acquires a pro-or anti-tumorigenic state. Tumor-associated macrophages are known, for instance, to play an opposite role depending on the microenvironment. They can promote angiogenesis, tumor growth and dissemination, or cytotoxicity and apoptosis. Peripheral nerves have also recently caught a lot of attention for their role in tumor formation. Cancer cells infiltrate nerves, a process called perineural invasion (PNI), or secrete molecular cues to guide peripheral nerve infiltration. In turn, nerves promote tumor growth and metastasis, and sympathetic and sensory nerves play different roles in tumor formation, including melanoma. Finally, the cells of the tumor niche not only communicate with the cancer cells, but clearly also among each other, as nicely demonstrated by the observation that macrophages support pain resolution by transferring mitochondria to sensory neurons. The tumor microenvironment has been extensively investigated using genetically engineered mouse models. However, some aspects of human biology cannot be recapitulated in the mouse. Thanks to the human pluripotent stem cell (hPSC) technologies, we can now generate otherwise not accessible human cells. This study will build a melanoma patient specific 3D hPSC-derived system to study cellular cross-talk among melanoma cells, macrophages, and the peripheral nervous system. Aim A) Derivation of a melanoma patient induced pluripotent stem cells (iPSC) line and melanoma cell line. To be able to recreate some aspects of the tumor microenvironment that are specific and relevant for melanoma patients, we will derive a patient-specific iPSC line. The project will first start with one patient as proof of principle. The long-term goal will be then to expand this work and build a platform that allows the modeling of the tumor microenvironment of multiple patients, which will eventually promote personalized therapies.-A.1: Fibroblast reprogramming: patient-derived fibroblasts will be reprogrammed using Sendai virus technologies.-A.2: iPSC lines characterization: patient-derived iPSC clones will be karyotyped, tested for mycoplasma and validated for expression of pluripotency markers, as OCT4, NANOG and SOX2. We will perform whole genome sequencing to detect any mutations that might affect the readouts. The melanoma patient-specific iPSC clones will be tested for their differentiation potential into sensory neurons, Schwann cells and macrophages, and their differentiation efficiency will be compared with that of at least three additional control iPSC lines (J1, BJ1, 348).-A.3: Isolation of a patient derived melanoma cell line: primary tumor will be mechanically and enzymatically dissociated. The culture will be depleted of erythrocytes. Expansion of the melanoma culture and eventual FACS sorting for CD117 (cKIT).Aim B) Melanoma patient iPSC-derived 3D perineural tumor microenvironment (PTM): iPSC will be coaxed into pure populations of sensory neurons, Schwann cells and macrophages. We will then use those cells to build assembloids and create a 3D platform to study and manipulate the cellular cross-talk between melanoma cells and the cells of the perineural tumor microenvironement.-B.1: Nerve-tumor cross-talk: iPSC will be differentiated into neural crest cells and then further differentiated into sensory neurons, which will be validated for BRN3A and TUJI expression. Neuronal activity will be tested using a multielectrode array (MEA). Sensory neurons will be produced following a protocol that gives rise to a mixed population of nociceptive, mechanoreceptive and proprioceptive sensory neurons. iPSC will again be differentiated into neural crest cells and further differentiated into Schwann cells. Pure populations of sensory neurons and Schwann cells will be used to make assembloids, hereafter called peripheral nerve (PN) assembloids. The capability of Schwann cells to myelinate will be validated using electron microscopy (EM). Patient-derived melanoma cells will be applied on the surface of the PN assembloids. We will investigate the perineural invasion (PNI) capacity of melanoma cells, and the features of cellular cross-talk will be analyzed by TUNEL, RT-qPCR (upon FACS sorting), IF, ELISA, EM and MEA. Finally, we will model the nerve-tumor cross-talk by treating the PN and melanoma assembloid with a panel of neurotransmitter antagonists.-B.2: Nerve-macrophage-tumor cross-talk: iPSC will be differentiated into erythro-myeloid progenitors and then further into macrophages. iPSC-derived macrophages will be tested for IBA1 expression, phagocytic properties (phagocytosis assay kit), release of C3 and inflammatory cytokines (ELISA). Media composition and the number of macrophages to be added to the PN assembloids (peripheral nervous system and macrophage assembloids, hereafter called PTM assembloids) will be optimized to have the minimal induction of macrophage reactivity if not stimulated. Patient-derived melanoma cells and melanoma cell lines will be added to the tricellular PTM assembloids and their PNI capacity will be assessed in the presence of macrophages. Aim C) Study of the mechanisms responsible for the instruction and the remodeling of the PTM: melanoma cell instruct and remodel the niche during tumor initiation and progression. Thanks to the PTM & melanoma assembloids we will be able to study the transcriptional changes occurring during tumor growth and genetically manipulate driver candidates in any specific cellular compartment.-C.1: Analysis of the niche remodeling during perineural invasion: PTM assembloids will be collected at 4 different time points: t0 (baseline, PTM assembloid without melanoma cells), t48h (48 hours post PTM assembloid + melanoma cells), t7days (7 days post PTM assembloid + melanoma cells), t14days (14 days post PTM assembloid + melanoma cells). At each time point, PTM composition and cellular states will be analyzed by IF and FACS. -C.2: Transcriptional changes occurring during perineural invasion and tumor growth: Spatial transcriptomics of the PTM assembloids collected at the 4 different time points above mentioned will highlight the transcriptional changes occurring in each cellular compartment of the assembloid: sensory neurons, Schwann cells, macrophages, and melanoma cells. This will clarify the transcriptional changes occurring during perineural invasion and niche remodeling-C.3: Genetic perturbation studies during perineural invasion: Spatial transcriptomics will highlight the transcriptional changes related to perineural invasion. Thanks to the stem cell technologies, we will genetically manipulate each separate cellular compartment of the PTM and melanoma assembloids. We will particularly focus on the mechanisms that promotes the acquisition of a protective niche and anti-tumorigenic cellular cross-talks. Finally, the PTM and melanoma assembloids will be injected subcutaneously into immunodeficient NSG mice to validate the melanoma tumorigenic and metastatic potential depending on the tumor microenvironment in vivo.Our work will provide mechanistic insights into the process of perineural invasion and cellular cross-talk in the primary melanoma microenvironment. Furthermore, this study will boost the development of novel therapeutic approaches aimed at promoting an anti-tumorigenic microenvironment.