基于平滑肌定向TP53敲除iPSC的平滑肌肉瘤细胞模型鉴定出新的可干预通路

Leiomyosarcoma (LMS) is a mesenchymal tumor of smooth muscle origin, representing one of the most common types of soft tissue sarcoma. The genetic landscape of LMS is driven by TP53 mutation that recurs in more than 60% of patients. Leiomyosarcoma prognosis is still disappointing, since up to now no drug combination has shown relevant clinical activity in the advanced or metastatic setting and the role of adjuvant therapy remains controversial. The aim of this project was to dissect the vulnerability of the most recurrent genetic lesion in a LMS model to identify novel actionable pathways. To this end we developed a cell model system recapitulating LMS molecular and lineage phenotypes by engineering donor-derived induced pluripotent stem cells (iPSC) at the TP53 locus using Crispr/Cas9 system. The model was then differentiated towards smooth muscle and characterized by gene expression profiling, protein and transcript biomarker expression, in vitro 2D and 3D cell growth and drug response testing. TP53 -edited cell pool was subcloned to select lines carrying either homozygous or compound heterozygous frameshift mutations ( TP53 Knockout, KO). Mesoderm lineage commitment upregulated specific mesoderm markers (T, Mixl1, CD56) in both parental (iPSC WT) and TP53 KO lines, while only TP53 KO cells significantly downregulated p53-target genes. Differentiation towards smooth muscle showed upregulation of terminal smooth muscle markers at the transcript (ACTA2, ACTG2, TAGLN) and protein level (alpha-SMA, Cnn1) in iPSC WT and TP53 KO clones. At the phenotypic level TP53 loss promoted cell growth at early and late stages of smooth muscle differentiation (2 - 2.5-fold induction, p<0.001), and significantly induced 3D spheroid cell growth with respect to the wild type counterpart (3.6-fold, p<0.001). Gene expression profiling during smooth muscle differentiation uncovered the malignant phenotype driven by TP53 knockout, shown by the upregulation of the CINSARC signature that is associated to tumor aggressiveness in sarcomas. Moreover, it identified FBXW7 as the most significantly silenced gene immediately after TP53 (p<0.001). Analysis of the TCGA Firehose Legacy dataset confirmed that FBXW7 was significantly downregulated in TP53 -mutant or deleted sarcomas with respect to WT tumors. Since it is known that FBXW7 -deficient cells are more sensitive to Tigecycline, a drug that targets mitochondrial translation, we tested the drug in TP53 KO LMS model, showing that smooth muscle-committed TP53 KO cells are three times more sensitive to this compound than wild type cells. This study shows that TP53 KO LMS iPSC model is an effective tool for target discovery, and that FBXW7 pathway could be an amenable target in TP53 -mutant LMS. The research leading to these results has received funding from AIRC under IG 2024 - ID. 30644 project - P.I. Astolfi Annalisa.