Prediction of drug-target interactions and drug repositioning via network-based inference

Abstract

Drug-target interaction (DTI) is the basis of drug discovery and design. It is time consuming and costly to determine DTI experimentally. Hence, it is necessary to develop computational methods for the prediction of potential DTI. Based on complex network theory, three supervised inference methods were developed here to predict DTI and used for drug repositioning, namely drug-based similarity inference (DBSI), target-based similarity inference (TBSI) and network-based inference (NBI). Among them, NBI performed best on four benchmark data sets. Then a drug-target network was created with NBI based on 12,483 FDA-approved and experimental drug-target binary links, and some new DTIs were further predicted. In vitro assays confirmed that five old drugs, namely montelukast, diclofenac, simvastatin, ketoconazole, and itraconazole, showed polypharmacological features on estrogen receptors or dipeptidyl peptidase-IV with half maximal inhibitory or effective concentration ranged from 0.2 to 10 µM. Moreover, simvastatin and ketoconazole showed potent antiproliferative activities on human MDA-MB-231 breast cancer cell line in MTT assays. The results indicated that these methods could be powerful tools in prediction of DTIs and drug repositioning.

Figure 1. Schematic diagram of (A) drug-based similarity inference (DBSI), (B) target-based similarity inference (TBSI) and (C) network-based inference (NBI) methods.

The entire workflow includes five steps: (i) collection of known drug-target interaction data and construction of bipartite drug-target graphs; (ii) calculation of drug-drug two dimensional structural similarity (S_C), target-target genomic sequence similarity (S_g) and drug-target topology network similarity; (iii) application of new methods in prediction of new drugs for a given target (pink square) or new targets for a given drug (pink circle); (iv) validation of new drug-target interactions by experimental assays (D); (v) visualization of experimental results using drug-target-disease associations network analysis (E). In A–C, given drug node (pink circle) denotes the drug which we want to predict new target for, given target node (pink square) denotes the target which we want to predict new drug for, drug with resource (green circle) denotes that this drug have resource, target with resource (green square) denotes that this target have resource, the more resource a node possesses, the darker the color is, blue edges denote the drug-target interactions with known experimental evidence, black arrows denote the resource diffusion direction. In E, green circle: drug node, red square: on-target node, blue square: off-target node, yellow square: new off-target node, violet square: disease node.

Figure 2. The receiver operating characteristic (ROC) curves of the three different methods to predict new known drugs for a given target on the four benchmark data sets by simulation 30 times of 10-fold cross validation test, (a) enzymes, (b) ion channels, (c) GPCRs and (d) nuclear receptors, drug-based similarity inference (DBSI): dot dash curve, target-based similarity inference (TBSI): solid curve, network-based inference (NBI): dash curve, FPR: false positive rate and TPR: true positive rate.

Figure 3. The drug–target (DT) bipartite network, in which a drug node (circle) and a target node (square) are connected to each other by grey edge if the target is annotated to have known experimental interactions with the drug in DrugBank.

The DT network was generated using known FDA-approved small molecule DT interactions. The size of the drug node is the fraction of the number of targets that the drug linked in DrugBank. The size of the target node is the fraction of the number of drugs that the target linked in DrugBank. Color codes are given in the legend. Drug nodes (circles) are colored according to their Anatomical Therapeutic Chemical Classification. The graph was prepared by Cytoscape (http://www.cytoscape.org/).

Figure 4. Predicted and bioassay results of new identified drug-target indications for five known approved drugs.

Data shown are the mean for at least triplicate measurements. ^aOriginal pharmacological target information was extracted from DrugBank (http://www.DrugBank.ca/). ^b50% relatively effective concentration is the concentration of the tested chemical showing 50% of agonistic activity of the maximum activity of E2. REC₅₀ provides the estrogenic activity relative to that of E2. ^cAntiproliferative activities were assayed on human MDA-MB-231 breast cancer cell line by MTT assays. IC₅₀: half maximal inhibitory concentration, EC₅₀: half maximal effective concentration, ER: Estrogen Receptors, DPP-IV: dipeptidyl peptidase-IV.

Figure 5. Dose-response curves of experimentally validated polypharmacology activities on estrogen receptor (ER) and dipeptidyl peptidase-IV (DPP-IV).

Dose-response curves for inhibitive activation: montelukast to DPP-IV (A), for transcriptional activation: tamoxifen (Tam) to ERα (black solid line) and ERβ (red dash line) (B), diclofenac to ERα (black solid line) and ERβ (red dash line) (C), simvastatin to ERβ (D), ketoconazole to ERβ (E), itraconazole to ERα (F), itraconazole to ERβ (G). In A–G, error bars were presented as the mean SD (standard deviation) of three duplicate determinations.

Figure 6. Dose-response curves of the antiproliferative potencies for ketoconazole (A) and simvastatin (B) on human MDA-MB-231 breast cancer cell lines by MTT assay.

Error bars are presented as the SD (standard deviation) of three duplicate determinations.

Figure 7. Discovered drug-target, target-disease and disease-gene associations network.

Grey arrows denote the old drug-target interactions, grey edges denote the old target-disease associations and blue edges denote the known disease-gene associations, which were extracted from DrugBank, Online Mendelian Inheritance in Man (OMIM) Morbid Map and literature reports (The further data were given in Table S7). Red arrows among approved drug nodes (cyan circle) and target nodes (yellow squares) denote the new discovered drug-target interactions in this study. Red dotted edges denote new target-disease associations discovered in this study. Cyan circle: drug node, red square: on-target (Primary targets annotated in DrugBank), grey square: off-target, yellow square: new off-target (new discovered target for a given drug validated in this study), violet square: disease node, green regular hexagon: gene. The graph was prepared by Cytoscape (http://www.cytoscape.org/).