Polymethionine(oxide)s were readily synthesized through polycondensation with amines, accompanying the elimination of phenol and CO2 and used as antifouling polymer against biological matters.

No abstract

We presented here design, syntheses and inhibitory activities of novel hypoxia-targeting IDO hybrid inhibitors conjugated with an unsubstituted L-Trp as an IDO affinity moiety without inhibitor 1MT, such as L-Trp-TPZ hybrids 1 (TX-2274), 2 (UTX-3), 3 (UTX-4), and 4 (UTX-2). TPZ-monoxide hybrids 1 and 3 were good competitive IDO inhibitors, while TPZ hybrids 2 and 4 were uncompetitive IDO inhibitors. Among them TPZ-monoxide hybrid 1 have the strongest IDO inhibitory activity. It suggests that TPZ-monoxide hybrids 1 and 3 are able to bind the active site of IDO, TPZ hybrids 2 and 4 are able to bind the enzyme-substrate complex. We proposed the possible mechanism of action of TPZ hybrid 2 that may first affect as a hypoxic cytotoxin, and then metabolized to TPZ-monoxide hybrid 1, which may do as an IDO inhibitor more effectively than its parent TPZ hybrid 2.

Background Neoepitopes are cancer-specific antigens and significant therapeutic cancer vaccine candidates. Tumor neoepitopes induce an immune response to eliminate cancer cells. This immune activation depends on the binding affinity between antigen peptide and the major histocompatibility complex (MHC), which is an immune receptor. The epitope-MHC binding assay is a technologically difficult, time-consuming, and expensive experiment because it involves HLA protein expression and epitope peptide synthesis. Therefore, prediction methods of these binding affinities have been developed using computational prediction approaches. In particular, because of the wide variety of MHC class II subtypes, there is a need to improve the performance of MHC class II prediction. Here, we propose a novel deep learning model that can predict epitope-MHC class II binding from limited training data. Results MTL4MHC2 consists of multi-task Bi-LSTM models, an antigen peptide learning model and an MHC peptide learning model. Each multi-task model shares the MHC class I and II learning parameters. MTL4MHC2 achieves an AUC-ROC score of 82.2%, outperforming state-of-the-art models while maintaining generalization performance. Conclusions We have demonstrated the effectiveness of multi-task learning for improving prediction performance from limited training data. MTL4MHC2 can be applied to develop novel cancer vaccines.