Metastasis Related Epithelial-Mesenchymal Transition Signature Predicts Prognosis and Response to Chemotherapy in Acute Myeloid Leukemia

The Role of Epithelial-to-Mesenchymal Transition Transcription Factors (EMT-TFs) in Acute Myeloid Leukemia Progression

Acute myeloid leukemia (AML) is a diverse malignancy originating from myeloid progenitor cells, with significant genetic and clinical variability. Modern classification systems like those from the World Health Organization (WHO) and European LeukemiaNet use immunophenotyping, molecular genetics, and clinical features to categorize AML subtypes. This classification highlights crucial genetic markers such as FLT3, NPM1 mutations, and MLL-AF9 fusion, which are essential for prognosis and directing targeted therapies. The MLL-AF9 fusion protein is often linked with therapy-resistant AML, highlighting the risk of relapse due to standard chemotherapeutic regimes. In this sense, factors like the ZEB, SNAI, and TWIST gene families, known for their roles in epithelial-mesenchymal transition (EMT) and cancer metastasis, also regulate hematopoiesis and may serve as effective therapeutic targets in AML. These genes contribute to cell proliferation, differentiation, and extramedullary hematopoiesis, suggesting new possibilities for treatment. Advancing our understanding of the molecular mechanisms that promote AML, especially how the bone marrow microenvironment affects invasion and drug resistance, is crucial. This comprehensive insight into the molecular and environmental interactions in AML emphasizes the need for ongoing research and more effective treatments.

Epithelial-mesenchymal plasticity in cancer: signaling pathways and therapeutic targets

Currently, cancer is still a leading cause of human death globally. Tumor deterioration comprises multiple events including metastasis, therapeutic resistance and immune evasion, all of which are tightly related to the phenotypic plasticity especially epithelial-mesenchymal plasticity (EMP). Tumor cells with EMP are manifest in three states as epithelial-mesenchymal transition (EMT), partial EMT, and mesenchymal-epithelial transition, which orchestrate the phenotypic switch and heterogeneity of tumor cells via transcriptional regulation and a series of signaling pathways, including transforming growth factor-β, Wnt/β-catenin, and Notch. However, due to the complicated nature of EMP, the diverse process of EMP is still not fully understood. In this review, we systematically conclude the biological background, regulating mechanisms of EMP as well as the role of EMP in therapy response. We also summarize a range of small molecule inhibitors, immune-related therapeutic approaches, and combination therapies that have been developed to target EMP for the outstanding role of EMP-driven tumor deterioration. Additionally, we explore the potential technique for EMP-based tumor mechanistic investigation and therapeutic research, which may burst vigorous prospects. Overall, we elucidate the multifaceted aspects of EMP in tumor progression and suggest a promising direction of cancer treatment based on targeting EMP.

A variational autoencoder trained with priors from canonical pathways increases the interpretability of transcriptome data

Interpreting transcriptome data is an important yet challenging aspect of bioinformatic analysis. While gene set enrichment analysis is a standard tool for interpreting regulatory changes, we utilize deep learning techniques, specifically autoencoder architectures, to learn latent variables that drive transcriptome signals. We investigate whether simple, variational autoencoder (VAE), and beta-weighted VAE are capable of learning reduced representations of transcriptomes that retain critical biological information. We propose a novel VAE that utilizes priors from biological data to direct the network to learn a representation of the transcriptome that is based on understandable biological concepts. After benchmarking five different autoencoder architectures, we found that each succeeded in reducing the transcriptomes to 50 latent dimensions, which captured enough variation for accurate reconstruction. The simple, fully connected autoencoder, performs best across the benchmarks, but lacks the characteristic of having directly interpretable latent dimensions. The beta-weighted, prior-informed VAE implementation is able to solve the benchmarking tasks, and provide semantically accurate latent features equating to biological pathways. This study opens a new direction for differential pathway analysis in transcriptomics with increased transparency and interpretability.

RPF2 mediates the CARM1‑MYCN axis to promote chemotherapy resistance in colorectal cancer cells

Ribosome production factor 2 homolog (RPF2) plays an important role in the life processes of ribosomal biogenesis; however, the function and mechanism of RPF2 in tumors are unclear. The present study demonstrated that RPF2 expression is involved in chemoresistance in colorectal cancer (CRC) cells. The current study demonstrated that upregulation of RPF2 expression in CRC promoted resistance to chemotherapeutic agents in CRC cells, whereas knockdown of RPF2 leads to increased sensitivity of CRC to chemotherapy. In addition, it was found that overexpression of RPF2 led to an increase in ATP‑binding cassette (ABC)B1 expression in CRC cells; accordingly, inhibition of RPF2 reduced the level of ABCB1 in CRC cells, thus suggesting that ABCB1 may be a downstream factor of RPF2 in the promotion of chemotherapy resistance to CRC. The results also suggested that the expression of N‑myc proto‑oncogene protein (MYCN), an upstream regulator of ABCB1, was affected by RPF2 in CRC cells. In addition, it was also found that the downstream protein coactivator‑associated arginine methyltransferase 1 (CARM1) of RPF2 existed in direct binding to MYCN and this interaction was regulated by RPF2. The above results suggested that RPF2 is probably regulated ABCB1 expression in CRC through the CARM1‑MYCN pathway, thereby promoting CRC drug resistance.

Bone marrow mesenchymal/fibroblastic stromal cells induce a distinctive EMT-like phenotype in AML cells

The development of epithelial-to-mesenchymal transition (EMT) like features is emerging as a critical factor involved in the pathogenesis of acute myeloid leukaemia (AML). However, the extracellular signals and the signalling pathways in AML that may regulate EMT remain largely unstudied. We found that the bone marrow (BM) mesenchymal/fibroblastic cell line HS5 induces an EMT-like migratory phenotype in AML cells. AML cells underwent a strong increase of vimentin (VIM) levels that was not mirrored to the same extent by changes of expression of the other EMT core proteins SNAI1 and SNAI2. We validated these particular pattern of co-expression of core-EMT markers in AML cells by performing an in silico analysis using datasets of human tumours. Our data showed that in AML the expression levels of VIM does not completely correlate with the co-expression of core EMT markers observed in epithelial tumours. We also found that vs epithelial tumours, AML cells display a distinct patterns of co-expression of VIM and the actin binding and adhesion regulatory proteins that regulate F-actin dynamics and integrin-mediated adhesions involved in the invasive migration in cells undergoing EMT. We conclude that the BM stroma induces an EMT related pattern of migration in AML cells in a process involving a distinctive regulation of EMT markers and of regulators of cell adhesion and actin dynamics that should be further investigated. Understanding the tumour specific signalling pathways associated with the EMT process may contribute to the development of new tailored therapies for AML as well as in different types of cancers.