MuCoCP: A priori chemical knowledge-based multimodal contrastive learning pre-trained neural network for the prediction of cyclic peptide membrane penetration ability.

MOTIVATION: There has been a burgeoning interest in cyclic peptide therapeutics due to their various outstanding advantages and strong potential for drug formation. However, it is undoubtedly costly and inefficient to use traditional wet lab methods to clarify their biological activities. Using Artificial Intelligence instead is a more energy-efficient and faster approach. MuCoCP aims to build a complete pre-trained model for extracting potential features of cyclic peptides, which can be fine-tuned to accurately predict cyclic peptide bioactivity on various downstream tasks. To maximize its effectiveness, we use a novel data augmentation method based on a priori chemical knowledge and multiple unsupervised training objective functions to greatly improve the information-grabbing ability of the model. RESULTS: To assay the efficacy of the model, we conducted validation on the membrane-permeability of cyclic peptides which achieved an accuracy of 0.87 and R-squared of 0.503 on CycPeptMPDB using semi-supervised training and obtained an accuracy of 0.84 and R-squared of 0.384 using a model with frozen parameters on an external dataset. This result has achieved state-of-the-art (SOTA), which substantiates the stability and generalization capability of MuCoCP. It means that MuCoCP can fully explore the high-dimensional information of cyclic peptides and make accurate predictions on downstream bioactivity tasks, which will serve as a guide for the future de novo design of cyclic peptide drugs and promote the development of cyclic peptide drugs. AVAILABILITY: All code used in our proposed method can be found at https://github.com/lennonyu11234/MuCoCP. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics.

Effect of plasma exosome lncRNA on isoproterenol hydrochloride-induced cardiotoxicity in rats.

Isoprenaline hydrochloride (IH) is a ß-adrenergic receptor agonist commonly used in the treatment of hypotension, shock, asthma, and other diseases. However, IH-induced cardiotoxicity limits its application. A large number of studies have shown that long noncoding RNA (lncRNA) regulates the occurrence and development of cardiovascular diseases. This study aimed to investigate whether abnormal lncRNA expression is involved in IH-mediated cardiotoxicity. First, the Sprague-Dawley (SD) rat myocardial injury model was established. Circulating exosomes were extracted from the plasma of rats and identified. In total, 108 differentially expressed (DE) lncRNAs and 150 DE mRNAs were identified by sequencing. These results indicate that these lncRNAs and mRNAs are substantially involved in chemical cardiotoxicity. Further signaling pathway and functional studies indicated that lncRNAs and mRNAs regulate several biological processes, such as selective mRNA splicing through spliceosomes, participate in sphingolipid metabolic pathways, and play a certain role in the circulatory system. Finally, we obtained 3 upregulated lncRNAs through reverse transcription-quantitative PCR (RT-qPCR) verification and selected target lncRNA-mRNA pairs according to the regulatory relationship of lncRNA/mRNA, some of which were associated with myocardial injury. This study provides valuable insights into the role of lncRNAs as novel biomarkers of chemical-induced cardiotoxicity.

The role of exosomal lncRNAs in acetaminophen-induced induced liver injury in SD rats.

Background: Drug-induced liver injury (DILI) is a leading cause of drug development failures during clinical trials and post-market introduction. Current biomarkers, such as ALT and AST, lack the necessary specificity and sensitivity needed for accurate detection. Exosomes, which protect LncRNAs from RNase degradation, could provide reliable and easily accessible options for biomarkers. Materials and methods: RNA-sequencing was used to identify differentially expressed LncRNAs (DE-LncRNAs), followed by isolation of LncRNAs from plasma exosomes in this study. Exosome characterization was conducted by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and Western blot (WB). Bioinformatics analysis included functional enrichment and co-expression network analysis. Five rat models were established, and quantitative real-time PCR was used to verify the specificity and sensitivity of two candidate exosomal LncRNAs. Results: The APAP-induced hepatocellular injury model was successfully established for RNA-sequencing, leading to the identification of several differentially expressed exosomal LncRNAs. Eight upregulated exosomal DE-LncRNAs were selected for validation. Among them, NONRATT018001.2 (p < 0.05) and MSTRG.73954.4 (p < 0.05) exhibited a more than 2-fold increase in expression levels. In hepatocellular injury and intrahepatic cholestasis models, both NONRATT018001.2 and MSTRG.73954.4 showed earlier increases compared to serum biomarkers ALT and AST. However, no histological changes were observed until the final time point. In the fatty liver model, NONRATT018001.2 and MSTRG.73954.4 increased earlier than ALT and AST at 21 days. By the 7th day, minor steatosis was evident in liver tissue, while the expression levels of the two candidate exosomal LncRNAs exceeded 2 and 4 times, respectively. In the hepatic fibrosis model, NONRATT018001.2 and MSTRG.73954.4 showed increases at every time point. By the 49th day, hepatocellular necrosis and fibrosis were observed in the liver tissue, with NONRATT018001.2 showing an increase of more than 8 times. The specificity of the identified exosomal DE-LncRNAs was verified using a myocardial injury model and they showed no significant differences between the case and control groups. Conclusion: NONRATT018001.2 and MSTRG.73954.4 hold potential as biomarkers for distinguishing different types of organ injury induced by drugs, particularly enabling early prediction of liver injury. Further experiments, such as siRNA interference or gene knockout, are warranted to explore the underlying mechanisms of these LncRNAs.