In Silico Identification of Vdr As A New Potential Target of Thymoquinone

Background and aim: in recent decades, nigella sativa and its main compound, thymoquinone, have been attracted much attention among civilizations and medical researchers with their therapeutic effects on many diseases including asthma, fatty liver, hypertension, diabetes, inflammation, cough, bronchitis, headaches, eczema, fever, dizziness and flu and a variety of cancer types, but many of thymoquinone mechanisms remain unclear. in the present study, computational chemical and biological approaches have been used to identify the potential targets of thymoquinone. in silico target fishing, gene expression data analysis, quantitative structure–activity relationship (qsar) and docking studies were the methods used in this study. the results showed that thymoquinone is a potent antagonist of vitamin d3 receptor (vdr). methods: 1. gene expression data analysis: in this section, the gene expression data information for tq-treated hct116 colorectal cancer cells, with the name of gse122860_fold-change_data.txt.gz was downloaded from the gene expression omnibus (geo) database available at https://www.ncbi.nlm.nih.gov/geo/ with the accession number of gse122860 and the data were adjusted based on p-value (less than 0.05) for tq treated cells and then the symbols of negatively regulated gene (fold change < 0) were submitted to enrichr available at http://amp.pharm.mssm.edu/enrichr for enrichment analysis. 2. target fishing by pharmamapper: the pharmmapper online tool (available at http://www.lilab-ecust.cn/pharmmapper/) is a web server for potential drug target identification by reversed pharmacophore matching the query compound against an in-house pharmacophore model database (wang et al., 2017). pharmacophore modeling is a broadly used ligand-based method in drug target identification. it refers to a protein−ligand interaction pattern corresponding to a desired pharmacological effect and can be considered as the largest common denominator shared by a set of active molecules (wang et al., 2017). the sdf file of tq was downloaded from pubchem database (cid:10281) and uploaded to the pharmmapper server. the submission started by selecting “human protein targets only (v2010, 2241)” as targets set and other parameters were not changed. 3. analyzing protein-ligand interaction using autodock4: autodock is a suite of c programs used to predict the bound conformations of a small, flexible ligand to a macromolecular target of known structure. the technique combines simulated annealing for conformation searching with a rapid grid‐based method of energy evaluation (goodsell et al., 1996). in the beginning the sdf file of tq was converted to pdb file using openbabel and the crystal structure of vitamin d3 receptor (vdr) was downloaded from protein data bank (pdb) database (pdb id: 1db1). the structure was adjusted and all non-standard elements were deleted using viewerlite software. autodock4 and autodocktools were employed to perform docking with flexible ligand and rigid protein (huey et al., 2011). the results were analyzed with ligplot+ and notepad++ after getting the ligand-protein complex written by autodocktools and converted to pdb using discovery studio 4.5 software. 4. prediction of ic50 by qsar modeling: quantitative structure-activity relationship (qsar) describes how a known biological activity can be considered as a function of the molecular descriptors derived from the chemical structure of a set of molecules (bajorath, 2011). many physiological activities of a molecule can be related to their chemical structure. molecular descriptors as numerical values associated with the chemical constitution for correlation of chemical structure with various physical properties, chemical reactivity, or biological activity are termed as descriptors for a qsar model (roy et al., 2015). half maximum inhibitory concentration (ic50) is the concentration of an inhibitor that is required to inhibit 50% of the enzyme activity in vitro. in this section, the structures and information of compounds with known ic50 values against vdr were downloaded from binding database (available at https://www.bindingdb.org/bind/index.jsp) in sdf and tsv formats, respectively. padel-descriptor software was used to define molecular descriptors of downloaded structures and thymoquinone sdf file, and then the data were adjusted based on target source organisms (homo sapiens) by excel. s-mlr software was employed to select descriptors for qsar equations and finally, building the qsar model and ic50 value prediction for tq were performed by chemoface software using “partial least square” as regression method. results: 1. gene expression data analysis: based on trrust transcription factors 2019, human vdr was identified as the transcription factor for the four down regulated genes dkk4, erbb2, ddit4 and egfr with the p-value, adjusted p-value, odd ratio and combined score of 0.0001440, 0.005481, 15.00 and 132.67 respectively. 2. target fishing by pharmamapper: vdr and the pharmacophore model of 1db1 (vdr complexed to 1,25 dihydroxy vitamin d3 structure at pdb database) were predicted by pharmmapper as potential target of tq with 10 similar pharmacophoric features and fit score of 3.814. 1,25 dihydroxy vitamin d3 or calcitriol has been identified as vdr antagonist (reinhart, 2004). 3. analyzing protein-ligand interaction using autodock4: the docking results showed that vdr was targeted by tq by making hydrogen bonds at tyra:143 and sera:237 positions with respectively the estimated free energy of binding, final intermolecular energy and final total internal energy of -5.31, -5.61 and -0.23 kcal/mol. 4. prediction of ic50 by qsar modeling: the predicted ic50 of tq against was 267.9915 nm, while the reported ic50 values for calcitriol against vdr on binding database range from 0.0600 nm to 13 nm. conclusion: in recent decades, computational chemogenomics has emerged as an interdisciplinary field of biomedical and pharmaceutical research aimed at identifying ligands or potential regulators for a wide range of gene products. this approach has accelerated studies of the function of gene products in living things and pharmaceutical discoveries (brown and bajorath, 2019). the basic assumption that guides all computational approaches in chemogenomics is that similar compounds bind to similar targets and therefore show a similar binding profile and conversely targets that bind similar ligands have similar binding sites (jacoby, 2013). a wide range of vegetables, fruits, spices and herbs from long time ago have been known for a number of therapeutic and preventative uses and contain various bioactive compounds. plants are a natural source of many available drugs that have found their place among modern treatments (kaur et al., 2018). in recent years, using bioinformatics and combining computational simulations and predictions with laboratory research results, has made it possible to more effectively explore pharmaceutical compounds and predict their role in activity regulation of important therapeutic targets (zheng et al., 2018). synergistic effect of both vitamin d and tq as anticancer agents has been reported in both in vitro and in vivo studies (kaseb et al., 2007, yi et al., 2008, ordonez-moran et al., 2005, hummel et al., 2013, mohamed et al., 2017). in this study we identified vdr as the potential target of tq with similar binding profile to calcitriol as a synthetic physiologically active analog of vitamin d, specifically the vitamin d3 and vdr complex. these results help us to use chemical similarity as a probe of protein function, predict off-target effects of known drug calcitriol and predict new biological targets of the both compounds.