Hong Kang

HIT: linking herbal active ingredients to targets

The information of protein targets and small molecule has been highly valued by biomedical and pharmaceutical research. Several protein target databases are available online for FDA-approved drugs as well as the promising precursors that have largely facilitated the mechanistic study and subsequent research for drug discovery. However, those related resources regarding to herbal active ingredients, although being unusually valued as a precious resource for new drug development, is rarely found. In this article, a comprehensive and fully curated database for Herb Ingredients’ Targets (HIT, http://lifecenter.sgst.cn/hit/) has been constructed to complement above resources. Those herbal ingredients with protein target information were carefully curated. The molecular target information involves those proteins being directly/indirectly activated/inhibited, protein binders and enzymes whose substrates or products are those compounds. Those up/down regulated genes are also included under the treatment of individual ingredients. In addition, the experimental condition, observed bioactivity and various references are provided as well for users reference. Derived from more than 3250 literatures, it currently contains 5208 entries about 1301 known protein targets (221 of them are described as direct targets) affected by 586 herbal compounds from more than 1300 reputable Chinese herbs, overlapping with 280 therapeutic targets from Therapeutic Targets Database (TTD), and 445 protein targets from DrugBank corresponding to 1488 drug agents. The database can be queried via keyword search or similarity search. Crosslinks have been made to TTD, DrugBank, KEGG, PDB, Uniprot, Pfam, NCBI, TCM-ID and other databases.

Insight into potential toxicity mechanisms of melamine: an in silico study.

The toxicity of melamine has attracted much attention since the outbreak of nephrolithiasis among children ingesting melamine-contaminated infant formula in China. However, there is little knowledge about the molecular mechanisms underlying the melamine-induced toxicity. In this paper, a ligand-protein docking method (INVDOCK) was applied to predict the toxicity-related target proteins for melamine and its metabolite, cyanuric acid. As a result, 23 and 35 proteins were finally identified as the potential target proteins for melamine and cyanuric acid, respectively. Through an enrichment analysis, it was found that nephrotoxicity and lung toxicity might be the most significant toxicities induced by melamine and cyanuric acid. Four target proteins (glutathione peroxidase 1, beta-hexosaminidase subunit beta, L-lactate dehydrogenase and lysozyme C) may be related to the molecular basis of the nephrotoxicity induced by melamine except for known kidney crystals formation. After mapping all these toxicity-related target proteins onto cellular pathways, it was indicated that the toxicities of melamine and cyanuric acid might also be caused by breaking down redox balance, perturbing the arginine and proline metabolism and damaging the homeostasis of energy production system. To further explore the mechanisms underlying the toxicities of melamine and cyanuric acid, a biological signal cascades network constructed by some of the toxicity-related target proteins was discussed.

Proteochemometric Modeling of the Bioactivity Spectra of HIV-1 Protease Inhibitors by Introducing Protein-Ligand Interaction Fingerprint

HIV-1 protease is one of the main therapeutic targets in HIV. However, a major problem in treatment of HIV is the rapid emergence of drug-resistant strains. It should be particularly helpful to clinical therapy of AIDS if one method can be used to predict antivirus capability of compounds for different variants. In our study, proteochemometric (PCM) models were created to study the bioactivity spectra of 92 chemical compounds with 47 unique HIV-1 protease variants. In contrast to other PCM models, which used Multiplication of Ligands and Proteins Descriptors (MLPD) as cross-term, one new cross-term, i.e. Protein-Ligand Interaction Fingerprint (PLIF) was introduced in our modeling. With different combinations of ligand descriptors, protein descriptors and cross-terms, nine PCM models were obtained, and six of them achieved good predictive abilities (Q2 test>0.7). These results showed that the performance of PCM models could be improved when ligand and protein descriptors were complemented by the newly introduced cross-term PLIF. Compared with the conventional cross-term MLPD, the newly introduced PLIF had a better predictive ability. Furthermore, our best model (GD & P & PLIF: Q2test = 0.8271) could select out those inhibitors which have a broad antiviral activity. As a conclusion, our study indicates that proteochemometric modeling with PLIF as cross-term is a potential useful way to solve the HIV-1 drug-resistant problem.

Virtual drug screen schema based on multiview similarity integration and ranking aggregation.

The current drug virtual screen (VS) methods mainly include two categories. i.e., ligand/target structure-based virtual screen and that, utilizing protein-ligand interaction fingerprint information based on the large number of complex structures. Since the former one focuses on the one-side information while the later one focuses on the whole complex structure, they are thus complementary and can be boosted by each other. However, a common problem faced here is how to present a comprehensive understanding and evaluation of the various virtual screen results derived from various VS methods. Furthermore, there is still an urgent need for developing an efficient approach to fully integrate various VS methods from a comprehensive multiview perspective. In this study, our virtual screen schema based on multiview similarity integration and ranking aggregation was tested comprehensively with statistical evaluations, providing several novel and useful clues on how to perform drug VS from multiple heterogeneous data sources. (1) 18 complex structures of HIV-1 protease with ligands from the PDB were curated as a test data set and the VS was performed with five different drug representations. Ritonavir ( 1HXW ) was selected as the query in VS and the weighted ranks of the query results were aggregated from multiple views through four similarity integration approaches. (2) Further, one of the ranking aggregation methods was used to integrate the similarity ranks calculated by gene ontology (GO) fingerprint and structural fingerprint on the data set from connectivity map, and two typical HDAC and HSP90 inhibitors were chosen as the queries. The results show that rank aggregation can enhance the result of similarity searching in VS when two or more descriptions are involved and provide a more reasonable similarity rank result. Our study shows that integrated VS based on multiple data fusion can achieve a remarkable better performance compared to that from individual ones and, thus, serves as a promising way for efficient drug screening, taking advantages of the rapidly accumulated molecule representations and heterogeneous data in the pharmacological area.

HIM-herbal ingredients in-vivo metabolism database

BackgroundHerbal medicine has long been viewed as a valuable asset for potential new drug discovery and herbal ingredients’ metabolites, especially the in vivo metabolites were often found to gain better pharmacological, pharmacokinetic and even better safety profiles compared to their parent compounds. However, these herbal metabolite information is still scattered and waiting to be collected.DescriptionHIM database manually collected so far the most comprehensive available in-vivo metabolism information for herbal active ingredients, as well as their corresponding bioactivity, organs and/or tissues distribution, toxicity, ADME and the clinical research profile. Currently HIM contains 361 ingredients and 1104 corresponding in-vivo metabolites from 673 reputable herbs. Tools of structural similarity, substructure search and Lipinski’s Rule of Five are also provided. Various links were made to PubChem, PubMed, TCM-ID (Traditional Chinese Medicine Information database) and HIT (Herbal ingredients’ targets databases).ConclusionsA curated database HIM is set up for the in vivo metabolites information of the active ingredients for Chinese herbs, together with their corresponding bioactivity, toxicity and ADME profile. HIM is freely accessible to academic researchers at http://www.bioinformatics.org.cn/.

Integrating Clinical Indexes into Four-Diagnostic Information Contributes to the Traditional Chinese Medicine (TCM) Syndrome Diagnosis of Chronic Hepatitis B

Traditional Chinese Medicine (TCM) treatment has been commonly used to treat Chronic Hepatitis B (CHB) in Asian countries based on TCM syndrome diagnosis, also called “ZHENG”. The syndrome is identified through the four-diagnostic methods, with certain degree of subjectivity and ambiguity from individual doctors. Normally those CHB patients also receive series of parameters from modern clinical examination, while they are routinely believed to be unrelated with the TCM syndrome diagnosis. In this study, we investigated whether these biomedical indexes in modern medicine could be beneficial to TCM syndrome diagnostics in an integrative way. Based on 634 patient samples from health controls and three subtypes of CHB syndromes, a two-view based hierarchical classification model was tested for TCM syndromes prediction based on totally 222 parameters integrated from both TCM practice and modern clinical tests. The results indicated that the performance of syndrome classification based on a proper integration of TCM and modern clinical indexes was significantly higher than those based on one view of parameters only. Furthermore, those indexes correlated with CHB syndrome diagnosis were successfully identified for CM indexes and biochemical indexes respectively, where potential associations between them were hinted to the MAPK signaling pathway.

Identification of Toxic Pyrrolizidine Alkaloids and Their Common Hepatotoxicity Mechanism.

Pyrrolizidine Alkaloids (PAs) are currently one of the most important botanical hepatotoxic ingredients. Glutathion (GSH) metabolism is the most reported pathway involved in hepatotoxicity mechanism of PAs. We speculate that, for different PAs, there should be a common mechanism underlying their hepatotoxicity in GSH metabolism. Computational methods were adopted to test our hypothesis in consideration of the limitations of current experimental approaches. Firstly, the potential targets of 22 PAs (from three major PA types) in GSH metabolism were identified by reverse docking; Secondly, glutathione S-transferase A1 (GSTA1) and glutathione peroxidase 1 (GPX1) targets pattern was found to be a special characteristic of toxic PAs with stepwise multiple linear regressions; Furthermore, the molecular mechanism underlying the interactions within toxic PAs and these two targets was demonstrated with the ligand-protein interaction analysis; Finally, GSTA1 and GPX1 were proved to be significant nodes in GSH metabolism. Overall, toxic PAs could be identified by GSTA1 and GPX1 targets pattern, which suggests their common hepatotoxicity mechanism: the interfering of detoxication in GSH metabolism. In addition, all the strategies developed here could be extended to studies on toxicity mechanism of other toxins.

Screening Ingredients from Herbs against Pregnane X Receptor in the Study of Inductive Herb-Drug Interactions: Combining Pharmacophore and Docking-Based Rank Aggregation

The issue of herb-drug interactions has been widely reported. Herbal ingredients can activate nuclear receptors and further induce the gene expression alteration of drug-metabolizing enzyme and/or transporter. Therefore, the herb-drug interaction will happen when the herbs and drugs are coadministered. This kind of interaction is called inductive herb-drug interactions. Pregnane X Receptor (PXR) and drug-metabolizing target genes are involved in most of inductive herb-drug interactions. To predict this kind of herb-drug interaction, the protocol could be simplified to only screen agonists of PXR from herbs because the relations of drugs with their metabolizing enzymes are well studied. Here, a combinational in silico strategy of pharmacophore modelling and docking-based rank aggregation (DRA) was employed to identify PXRs agonists. Firstly, 305 ingredients were screened out from 820 ingredients as candidate agonists of PXR with our pharmacophore model. Secondly, DRA was used to rerank the result of pharmacophore filtering. To validate our prediction, a curated herb-drug interaction database was built, which recorded 380 herb-drug interactions. Finally, among the top 10 herb ingredients from the ranking list, 6 ingredients were reported to involve in herb-drug interactions. The accuracy of our method is higher than other traditional methods. The strategy could be extended to studies on other inductive herb-drug interactions.

Binding Mechanism of Inhibitors to CDK5/p25 Complex: Free Energy Calculation and Ranking Aggregation Analysis

Cyclin‐dependent kinase‐5 (CDK5) plays an indispensable role in the central nervous system. Competitive inhibition of the ATP‐binding pocket of CDK5 is involved in fighting with neurodegenerative diseases, diabetes, tumors, inflammations etc. To better design ATP‐binding competitive inhibitors, the binding mechanism of three important inhibitors of kinases, (R)‐roscovitine (RRC), aloisine‐A (ALH) and indirubin‐3′‐oxime (IXM), together with their receptor CDK5, were studied by molecular dynamics simulations. The H‐bond analysis demonstrated that a strong bond was formed between the CO or NH groups in the backbone of Cys83 and the N or NH groups on the nitrogen‐containing ring of inhibitors. These hydrogen bonds significantly increase the binding and inhibitory efficiency. The free energy analysis show that the order of predicted binding affinities of these three inhibitors toward CDK5/p25 is IXM>ALH>RRC, which is consistent with the experimental data. Besides the hydrogen bond formation, the van der Waals interactions between residues Ile10, Val18, and Leu133 of CDK5 and inhibitors were discovered to constitute another substantial component of their binding mode. Worth mentioning is that the conformational turnover of the inhibitor RRC was observed during the course of molecular dynamics simulations. We believe that this is the reason why RRC has the lower H‐bond occupancy and binding affinity than the other two inhibitors. Furthermore, during the analysis of the per‐residue decomposition, the ranking aggregation method was firstly employed to rank the contribution of different residues. The results demonstrated that the top five residues in the active pocket of CDK5 were Cys83, Leu133, Ile10, Phe82, and Glu81, which is in good agreement with the results of H‐bond analysis and binding free energy analysis. These findings should provide insights into the inhibition mechanism of the CDK5/p25 complex and be useful for the rational design of novel ATP‐binding competitive inhibitors in the near future.

Comparison of Different Ranking Methods in Protein-Ligand Binding Site Prediction

In recent years, although many ligand-binding site prediction methods have been developed, there has still been a great demand to improve the prediction accuracy and compare different prediction algorithms to evaluate their performances. In this work, in order to improve the performance of the protein-ligand binding site prediction method presented in our former study, a comparison of different binding site ranking lists was studied. Four kinds of properties, i.e., pocket size, distance from the protein centroid, sequence conservation and the number of hydrophobic residues, have been chosen as the corresponding ranking criterion respectively. Our studies show that the sequence conservation information helps to rank the real pockets with the most successful accuracy compared to others. At the same time, the pocket size and the distance of binding site from the protein centroid are also found to be helpful. In addition, a multi-view ranking aggregation method, which combines the information among those four properties, was further applied in our study. The results show that a better performance can be achieved by the aggregation of the complementary properties in the prediction of ligand-binding sites.