Integrative single-cell transcriptomic analyses reveal the cellular ontological and functional heterogeneities of primary and metastatic liver tumors.

BACKGROUND: The global cellular landscape of the tumor microenvironment (TME) combining primary and metastatic liver tumors has not been comprehensively characterized. METHODS: Based on the scRNA-seq and spatial transcriptomic data of non-tumor liver tissues (NTs), primary liver tumors (PTs) and metastatic liver tumors (MTs), we performed the tissue preference, trajectory reconstruction, transcription factor activity inference, cell-cell interaction and cellular deconvolution analyses to construct a comprehensive cellular landscape of liver tumors. RESULTS: Our analyses depicted the heterogeneous cellular ecosystems in NTs, PTs and MTs. The activated memory B cells and effector T cells were shown to gradually shift to inhibitory B cells, regulatory or exhausted T cells in liver tumors, especially in MTs. Among them, we characterized a unique group of TCF7+ CD8+ memory T cells specifically enriched in MTs that could differentiate into exhausted T cells likely driven by the p38 MAPK signaling. With regard to myeloid cells, the liver-resident macrophages and inflammatory monocyte/macrophages were markedly replaced by tumor-associated macrophages (TAMs), with TREM2+ and UBE2C+ TAMs enriched in PTs, while SPP1+ and WDR45B+ TAMs in MTs. We further showed that the newly identified WDR45B+ TAMs exhibit an M2-like polarization and are associated with adverse prognosis in patients with liver metastases. Additionally, we addressed that endothelial cells display higher immune tolerance and angiogenesis capacity, and provided evidence for the source of the mesenchymal transformation of fibroblasts in tumors. Finally, the malignant hepatocytes and fibroblasts were prioritized as the pivotal cell populations in shaping the microenvironments of PTs and MTs, respectively. Notably, validation analyses by using spatial or bulk transcriptomic data in clinical cohorts concordantly emphasized the clinical significance of these findings. CONCLUSIONS: This study defines the ontological and functional heterogeneities in cellular ecosystems of primary and metastatic liver tumors, providing a foundation for future investigation of the underlying cellular mechanisms.

PTBP1 promotes the progression of hepatocellular carcinoma by enhancing the oncogenic splicing switch of FGFR2.

Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer accounting for 90% of cases. It is a highly invasive and deadly cancer with a gradual onset. Polypyrimidine tract-binding protein 1 (PTBP1) is an important RNA-binding protein involved in RNA metabolism and has been linked to oncogenic splicing events. While the oncogenic role of PTBP1 in HCC cells has been established, the exact mechanism of action remains unclear. This study aimed to investigate the functional connection between PTBP1 and dysregulated splicing events in HCC. Through immunoprecipitation-mass spectrometry analyses, we discovered that the proteins bound to PTBP1 were significantly enriched in the complex responsible for the alternative splicing of FGFR2 (fibroblast growth factor receptor 2). Further RNA immunoprecipitation and quantitative PCR assays confirmed that PTBP1 down-regulated the FGFR2-IIIb isoform levels and up-regulated the FGFR2-IIIc isoform levels in HCC cells, leading to a switch from FGFR2-IIIb to FGFR2-IIIc isoforms. Subsequent functional evaluations using CCK-8, transwell, and plate clone formation assays in HCC cell lines HepG2 and Huh7 demonstrated that FGFR2-IIIb exhibited tumor-suppressive effects, while FGFR2-IIIc displayed tumor-promoting effects. In conclusion, this study provides insights into the PTBP1-mediated alternative splicing mechanism in HCC progression, offering a new theoretical basis for the prevention and treatment of this malignancy. Mechanistically, the isoform switch from FGFR2-IIIb to FGFR2-IIIc promoted epithelial-mesenchymal transformation (EMT) of HCC cells and activated the FGFR cascades ERK and AKT pathways.