TAK-901

Prostate cancer is the most common malignant tumor in males, which frequently develops into castration-resistant prostate cancer (CRPC). CRPC metastasis is the main reason for its high mortality rate. At present, it lacks effective treatment for patients with CRPC. Raltitrexed (RTX) has been shown to be effective in the treatment of colorectal cancer. However, the effect of RTX on prostate cancer and the underlying mechanism remain unknown. In the current study, we found that RTX could dose-dependently inhibit proliferation, migration, colony formation and induce apoptosis in DU145 and PC-3 cells. RTX also increased ROS generation in prostate cancer cells. Pretreatment with N-acetyl-L-cysteine (NAC) significantly prevented RTX-induced cell apoptosis and endoplasmic reticulum (ER) stress signaling activation in prostate cancer cells. Additionally, we found RTX-induced ROS generation and ER stress activation depended on the expression of heat shock protein family A member 8 (HSPA8). Over-expression of HSPA8 could alleviate RTX-induced cell apoptosis, ROS generation and ER stress signaling activation. Finally, our study also showed that RTX attenuated the tumor growth of prostate cancer in the DU145 xenograft model and significantly downregulated HSPA8 expression and activated ER stress signaling pathway in tumor tissues. Our study is the first to reveal that RTX induces prostate cancer cells apoptosis through inhibiting the expression of HSPA8 and further inducing ROS-mediated ER stress pathway action. This study suggests that RTX may be a novel promising candidate drug for prostate cancer therapy.

The mechanism of action (MoA) of a compound describes the biol. interaction through which it produces a pharmacol. effect.Multiple data sources can be used for the purpose of predicting MoA, including compound structural information, and various assays, such as those based on cell morphol., transcriptomics and metabolomics.In the present study we explored the benefits and potential additive/synergistic effects of combining structural information, in the form of Morgan fingerprints, and morphol. information, in the form of five-channel Cell Painting image data.For a set of 10 well represented MoA classes, we compared the performance of deep learning models trained on the two datasets sep. vs. a model trained on both datasets simultaneously.On a held-out test set we obtained a macro-averaged F1 score of 0.58 when training on only the structural data, 0.81 when training on only the image data, and 0.92 when training on both together.Thus indicating clear additive/synergistic effects and highlighting the benefit of integrating multiple data sources for MoA prediction.