Aim: Colorectal cancer (CRC) ranks among the most prevalent malignancies across the globe, with treatment resistance often closely linked to the complexity of the tumor microenvironment (TME). This study aims to establish a gene signature associated with epithelial-mesenchymal transition (EMT) that integrates TME dynamics, prognosis prediction, and drug resistance assessment in CRC. Methods: We employed the Cancer Genome Atlas (TCGA) resource and bulk RNA-sequencing profiles linked to EMT to identify common differentially expressed genes (DEGs). An eight-gene signature was constructed using multivariable Cox regression analysis. The correlations of risk groups (scores) with overall survival, biological characteristics, and drug sensitivity were analyzed in CRC patients. Results: Patients in the high-risk group exhibited significantly worse clinical outcomes than those in the low-risk group. Moreover, a lower risk score was significantly correlated with increased responsiveness to both immune checkpoint inhibitors and 5-fluorouracil in CRC patients. Additionally, experiments in vitro and in vivo confirmed that FABP4 functions as an oncogene in CRC by facilitating cell growth, migration, and chemotherapy resistance. Conclusion: The EMT-associated gene signature holds significant value for predicting clinical prognosis, immunotherapy responsiveness, and chemotherapy sensitivity in CRC.
Xiaonan Shen, Hao Zhong, Wenqing Jia et al.· Cancer Drug Resistance· 0 citations
BACKGROUND
Colorectal cancer (CRC) exhibits pronounced biological diversity, a feature increasingly attributed to alterations in cellular metabolic reprograms. Serine metabolism supports nucleotide synthesis, redox balance, and epigenetic regulation via one-carbon metabolism, yet its role in shaping the tumor cellular interactions and immune landscape at single-cell level remains unclear.
METHODS
Single-cell transcriptomic profiles (GSE284449) were jointly analyzed with bulk expression datasets (GSE271719), together with Mendelian randomization (MR)-based inference, to dissect serine metabolism in CRC. High-serine-metabolism (HSM) cell populations were identified using scMetabolism, and robust marker genes were selected through Lasso, random forest, XGBoost, and SVM-RFE machine learning approaches. MR was applied to evaluate causal associations with CRC risk. Functional validation included IHC, qRT-PCR, western blot, and LC-MS metabolomics, while CellChat analysis characterized HSM cell interactions with immune and stromal cells.
RESULTS
ZBTB21 and TRPM2 were identified as core regulators of HSM cells, with ZBTB21 predominantly expressed in monocytes and pro-B cells. MR analysis suggested a potential inverse association between genetically predicted ZBTB21 expression and CRC risk, indicating that ZBTB21 may exert context-dependent effects at the population level. Cell-cell communication analysis suggested that HSM monocytes may interact with fibroblasts through signaling pathways including the MIF-SPP1 axis; however, these interactions are computationally inferred and require further experimental validation. Functional assays showed that ZBTB21 overexpression upregulated PHGDH and SHMT1, increased intracellular NADPH/NADP⁺ ratios, and influenced monocyte-related phenotypes. ZBTB21 levels were markedly increased in CRC samples relative to matched non-tumorous mucosal tissues.
CONCLUSIONS
At single-cell resolution, ZBTB21 emerges as a metabolic regulator that strengthens serine biosynthesis and redox homeostasis. While integrative analyses suggest a potential link between ZBTB21-associated metabolic states and immune interactions, further experimental validation is required to establish causal relationships. This integrative framework connects genetic causality, metabolism, and immune interactions, providing mechanistic insights and potential strategies for metabolic-targeted therapies in CRC.
TRIAL REGISTRATION
Not applicable.
Yi-Mei Jiang, Haiyan Huang, Haoran Feng et al.· Cancer Immunology and Immuno...· 0 citations
It is suggested that Hashimoto’s thyroiditis reshapes both epithelial transcriptional states and the intratumoral microbial environment, potentially contributing to a less aggressive tumor phenotype.
Yu Che, W. Kou, Lianghui Xu et al.· iScience· 0 citations