Drug repositioning by prediction of drug's anatomical therapeutic chemical code via network-based inference approaches.

Drug discovery and development is a time-consuming and costly process. Therefore, drug repositioning has become an effective approach to address the issues by identifying new therapeutic or pharmacological actions for existing drugs. The drug's anatomical therapeutic chemical (ATC) code is a hierarchical classification system categorized as five levels according to the organs or systems that drugs act and the pharmacology, therapeutic and chemical properties of drugs. The 2nd-, 3rd- and 4th-level ATC codes reserved the therapeutic and pharmacological information of drugs. With the hypothesis that drugs with similar structures or targets would possess similar ATC codes, we exploited a network-based approach to predict the 2nd-, 3rd- and 4th-level ATC codes by constructing substructure drug-ATC (SD-ATC), target drug-ATC (TD-ATC) and Substructure&Target drug-ATC (STD-ATC) networks. After 10-fold cross validation and two external validations, the STD-ATC models outperformed the SD-ATC and TD-ATC ones. Furthermore, with KR as fingerprint, the STD-ATC model was identified as the optimal model with AUC values at 0.899 ± 0.015, 0.916 and 0.893 for 10-fold cross validation, external validation set 1 and external validation set 2, respectively. To illustrate the predictive capability of the STD-ATC model with KR fingerprint, as a case study, we predicted 25 FDA-approved drugs (22 drugs were actually purchased) to have potential activities on heart failure using that model. Experiments in vitro confirmed that 8 of the 22 old drugs have shown mild to potent cardioprotective activities on both hypoxia model and oxygen-glucose deprivation model, which demonstrated that our STD-ATC prediction model would be an effective tool for drug repositioning.

Pathway-Based Drug Repurposing with DPNetinfer: A Method to Predict Drug-Pathway Associations via Network-Based Approaches.

Identification of drug-pathway associations plays an important role in pathway-based drug repurposing. However, it is time-consuming and costly to uncover new drug-pathway associations experimentally. The drug-induced transcriptomics data provide a global view of cellular pathways and tell how these pathways change under different treatments. These data enable computational approaches for large-scale prediction of drug-pathway associations. Here we introduced DPNetinfer, a novel computational method to predict potential drug-pathway associations based on substructure-drug-pathway networks via network-based approaches. The results demonstrated that DPNetinfer performed well in a pan-cancer network with an AUC (area under curve) = 0.9358. Meanwhile, DPNetinfer was shown to have a good capability of generalization on two external validation sets (AUC = 0.8519 and 0.7494, respectively). As a case study, DPNetinfer was used in pathway-based drug repurposing for cancer therapy. Unexpected anticancer activities of some nononcology drugs were then identified on the PI3K-Akt pathway. Considering tumor heterogeneity, seven primary site-based models were constructed by DPNetinfer in different drug-pathway networks. In a word, DPNetinfer provides a powerful tool for large-scale prediction of drug-pathway associations in pathway-based drug repurposing. A web tool for DPNetinfer is freely available at http://lmmd.ecust.edu.cn/netinfer/.

In silico prediction of mitochondrial toxicity of chemicals using machine learning methods.

Mitochondria are important organelles in human cells, providing more than 95% of the energy. However, some drugs and environmental chemicals could induce mitochondrial dysfunction, which might cause complex diseases and even worsen the condition of patients with mitochondrial damage. Some drugs have been withdrawn from the market due to their severe mitochondrial toxicity, such as troglitazone. Therefore, there is an urgent need to develop models that could accurately predict the mitochondrial toxicity of chemicals. In this paper, suitable data were obtained from literature and databases first. Then nine types of fingerprints were used to characterize these compounds. Finally, different algorithms were used to build models. Meanwhile, the applicability domain of the prediction models was defined. We have also explored the structural alerts of mitochondrial toxicity, which would be helpful for medicinal chemists to better predict mitochondrial toxicity and further optimize lead compounds.

NetInfer: A Web Server for Prediction of Targets and Therapeutic and Adverse Effects via Network-Based Inference Methods.

In this study, we developed a web server named NetInfer for prediction of targets and therapeutic and adverse effects via network-based inference methods. Compared with our previously developed standalone version of NetInfer, this web server provides a user-friendly interface. With the web server, users can easily predict potential target proteins, microRNAs, Anatomical Therapeutic Chemical (ATC) classification codes, or adverse drug events for small molecules of their interests in a few steps. Most of the prediction models were constructed on the basis of our previous studies, where those models have been evaluated systematically and demonstrated high performance. The high-quality models can generate accurate predictions. As a case study, we predicted ATC codes and target proteins for several drugs. The predicted therapeutic effects of these drugs on cardiovascular diseases and their potential molecular mechanisms were validated by the literature. This successful case study demonstrated that our web server would be a powerful tool in drug repositioning and systems pharmacology. The web server of NetInfer is freely available at http://lmmd.ecust.edu.cn/netinfer/.

SPH7854, a gut-limited RORγt antagonist, ameliorates TNBS-induced experimental colitis in rat.

Multiple lines of evidence suggest that Retinoic Acid Related Orphan Nuclear Receptor gamma t (RORγt) is a potent therapeutic target for inflammatory bowel disease (IBD). However, systemic blockade of RORγt easily leads to thymic lymphoma and aberrant liver function. Therefore, the development of gut-limited RORγt antagonists may lead to the development of innovative IBD therapeutics that improve safety and retain effectiveness. We discovered SPH7854, a potent and selective RORγt antagonist. The effect of SPH7854 on the differentiation of T helper 1 (Th1)/Th17/regulatory T (Treg) cells was evaluated in mouse and human primary cells. SPH7854 (2-(4-(ethylsulfonyl)phenyl)-N- (6-(2-methyl-2-(pyridin-2-yl) propanoyl)pyridin-3-yl)acetamide) dose-dependently inhibited interleukin-17A (IL-17A) secretion from mouse CD4 + T cells and human peripheral blood mononuclear cells (PBMC). Additionally, SPH7854 strongly suppressed Th17 cell differentiation and considerably promoted Treg cell differentiation while slightly affected Th1 cell differentiation from mouse CD4 + T cells. The pharmacokinetic (PK) studies indicated that SPH7854 was restricted to the gut: the bioavailability and maximal plasma concentration of SPH7854 after oral administration (6 mg/kg) were 1.24 ± 0.33 % and 4.92 ± 11.81 nM, respectively, in rats. Strikingly, oral administration of SPH7854 (5 mg/kg and 15 mg/kg) twice daily significantly alleviated 2, 4, 6-trinitrobenzensulfonic acid (TNBS)-induced colitis in rats. SPH7854, especially at 15 mg/kg, significantly alleviated symptoms and improved macroscopic signs and microscopic structure in rat colitis, with decreased colonic mucosal levels of IL-17A, IL-6, tumor necrosis factor α (TNFα), monocyte chemoattractant protein-1 (MCP-1) and myeloperoxidase (MPO). These evidences indicated that blockade of RORγt activity via a gut-limited antagonist may be an effective and safe therapeutic strategy for IBD treatment.

Development of a Multi-Target Strategy for the Treatment of Vitiligo via Machine Learning and Network Analysis Methods.

Vitiligo is a complex disorder characterized by the loss of pigment in the skin. The current therapeutic strategies are limited. The identification of novel drug targets and candidates is highly challenging for vitiligo. Here we proposed a systematic framework to discover potential therapeutic targets, and further explore the underlying mechanism of kaempferide, one of major ingredients from Vernonia anthelmintica (L.) willd, for vitiligo. By collecting transcriptome and protein-protein interactome data, the combination of random forest (RF) and greedy articulation points removal (GAPR) methods was used to discover potential therapeutic targets for vitiligo. The results showed that the RF model performed well with AUC (area under the receiver operating characteristic curve) = 0.926, and led to prioritization of 722 important transcriptomic features. Then, network analysis revealed that 44 articulation proteins in vitiligo network were considered as potential therapeutic targets by the GAPR method. Finally, through integrating the above results and proteomic profiling of kaempferide, the multi-target strategy for vitiligo was dissected, including 1) the suppression of the p38 MAPK signaling pathway by inhibiting CDK1 and PBK, and 2) the modulation of cellular redox homeostasis, especially the TXN and GSH antioxidant systems, for the purpose of melanogenesis. Meanwhile, this strategy may offer a novel perspective to discover drug candidates for vitiligo. Thus, the framework would be a useful tool to discover potential therapeutic strategies and drug candidates for complex diseases.

A novel derivative of artemisinin inhibits cell proliferation and metastasis via down-regulation of cathepsin K in breast cancer.

Breast cancer is one of the main diagnosis cancers annually worldwide. It is difficult to thorough cure due to drug resistance and the high possibility of metastasis. SM934 is a novel water-soluble artemisinin analog, and has been reported to have a promising therapeutic effect on multiple autoimmune diseases. In this study, SM934 was combined with Testosterone, which is related to prostate cancer, and the reaction product was SM934-Testosterone. We aimed to explore whether SM934, Testosterone and SM934-Testosterone could inhibit tumorigenesis and metastasis of breast cancer cells. Moreover, the mechanism also remains to be clarified. The results of our study showed that among the three compounds, only SM934-Testosterone treatment could lead to the suppression of cell proliferation and metastasis with IC50 = 30.66 ±â€¯2.13 µM at 24 h in MDA-MB-231 and IC50 = 31.11 ±â€¯1.79 µM at 24 h in SK-BR-3, where apoptosis was induced. But SM934-Testosterone showed little effects on breast cancer in vivo due to its poor water-solubility. Furthermore, computational target prediction and experimental validation demonstrated that Cathepsin K was the target of SM934-Testosterone. SM934-Testosterone inhibited the expression of Cathepsin K in breast cancer cells. Then, down-regulation of Cathepsin K in cancer cells by transfected with Cathepsin K shRNA inhibited cell proliferation and metastasis of breast cancer cells. Moreover, pathway enrichment was performed to understand the mechanism of action that Cathepsin K could adjust apoptosis regulator Bcl-X, and knockdown of Cathepsin K by SM934-Testosterone resulted in the reduction of Bcl-xL, which has been reported to be related to the proliferation and metastasis of cells. Collectively, SM934-Testosterone inhibited proliferation and metastasis ability of breast cancer cells via inhibiting the expression of Cathepsin K followed by the inhibition of Bcl-xL.

Insights into mechanisms and severity of drug-induced liver injury via computational systems toxicology approach.

Liver is the central place for drug metabolism. Drug-induced liver injury (DILI) is hence inevitable, and has become one of the leading causes for drug failure in development and drug withdrawal from the market. Due to lack of reliable preclinical and in vivo toxicology test conditions, it is time-consuming, laborious and costly to interpret the mechanisms of DILI through bioassays. In this paper, we developed a computational systems toxicology approach to investigate the molecular mechanisms of DILI. Totally 1478 DILI compounds were collected, together with 1067 known targets for 896 DILI compounds. Then, 173 new potential targets of these compounds were predicted by our bSDTNBI (balanced substructure-drug-target network-based inference) method. After network analysis, 145 primary genes were found to relate with hepatotoxicity and have higher expression in liver, among which 26 genes were predicted by our method, such as CYP2E1, GSTA1, EPHX1, ADH1B, ADH1C, ALDH2, F7, and IL2. A scoring function, DILI-Score, was further proposed to assess the hepatotoxic severity of a given compound. Finally, as case studies, we analyzed the mechanisms of DILI from the perspective of off-targets, and found out the pivotal genes for liver injuries induced by tyrosine kinase inhibitors and TAK-875. This work would be helpful for better understanding mechanisms of DILI and provide clues for reducing risk of DILI.