Molecular docking and ADMET analysis of hydroxamic acids as HDAC2 inhibitors

Histone deacetylase (HDAC2) belongs to the hydrolase family and a promising target for cancers. We reported 96 hydroxamic compounds optimized using hydrogen-donors, hydrophobic and electron withdrawing groups followed by molecular docking studies. The optimized compounds show good LibDock score and H-bond interaction in the active site of HDAC2. We selected 20 compounds as the best HDAC2 inhibitors based on the LibDock score, binding energy and hydrogen bonding. ADMET predictions on these compounds show good absorption, BBB penetration and no liver toxicity. We subsequently report four compounds selected as best HDAC2 inhibitors based on the LibDock, binding energy, H-bonding and ADMET properties.

belongs to the hydrolase family and classified into two classes on the basis of sequence similarity, class I has four isomers of HDAC1-3, and HDAC8 and are related to yeast Rpd3 gene, class II has six isomers of HDAC4-7 and HDAC9-10 and are related to Hda1 and class I and II operated by zinc dependent mechanism [4]. Histone deacetylases (HDACs) control the gene expression and cellular signalling and histone deacetylases 2 (HDAC2) is over expressed in solid tumors including colon cancer, lung cancer, cervical carcinoma, breast cancer, and kidney/cervix cancer and also in Alzheimer's disease [5][6][7]. Several natural and synthetic derivatives have been identified to be able to inhibit the activity of the HDACs. 381 ©Biomedical Informatics (2019) HDAC inhibitors (HDACi) arrest cell growth and leads to differentiation and apoptosis in tumor cells. HDACi can be divided into several structural classes including hydroxamic acids, cyclic peptides, aliphatic acids and benzamides etc. [8][9]. Naturally identified Hydroxamic acid HDAC inhibitor was Trichostatin A (TSA) and SAHA (Suberoylanilide hydroxamic acid or Vorinostat (Zolinza®)) is structurally similar to TSA was first HDAC inhibitor approved for the treatment of refractory cutaneous T-cell lymphoma by Food and Drug Administration (FDA) in October 2006 [10][11]. The compounds with radio sensitizing properties were found to be effective in the clinical application as they are cell specific [12]. Research on the SAHA as HDAC inhibitor for the treatment of hematologic and solid tumors is found to be efficient [13]. Some studies found that HDAC inhibitors can be used for targeting the radio resistant cancers [14].  Table 1. Molecular docking analysis performed on these molecules to investigate for better HDAC2 inhibitors. All ligands were sketched using ISIS draw and given as input file in prepare ligand module in Discovery studio (DS). This generated 3Dstructures, tatuomers, and isomers and filtered the ligands by Lipinski rule of five. After applying the force fields on ligands the structures were minimized for lowest energy.

Protein preparation and docking:
The crystal structure of HDAC2 (PDB ID: 3MAX) was downloaded from protein database (http://www.rcsb.org/pdb). The protein preparation was carried out in DS by removing water molecules and co-crystallized ligand further applying force filed parameter CHARMm to protein. The receptor binding sites were searched using flood filling algorithm. Docking calculations carried out using LibDock program implemented in discovery studio [23]. The 15 Å site sphere was selected using coordinates in predefined binding site for docking studies. The 500 binding site features, so call "HotSpots" in binding site spheres were determined using a grid placed into the binding site with polar and apolar probes. The conformations of ligands poses were generated using FAST method and then placed into the binding site sphere. The docking poses were pruned and optimized. Final best optimized compounds were selected based on the LibDock score and H-bonds and the results were compared with the SAHA compound.
In 3D molecular docking studies, the candidate compound docked into the target protein and provides a variety of structural information such as hydrogen bonding interaction, electrostatic interaction, and molecular surface complementary and so on. The binding energy of complex calculated using Eq. 1, which gives the better understanding of binding affinity of the docked complex.

ADMET:
ADMET (Absorption, Distribution, Metabolism, and Excretion): In drug discovery many potential drugs failed in clinical trials or late drug discovery process, due to poor drug like properties and adverse side effects. In the current investigation, all the optimized hydroxamic acid compounds were subjected to ADMET studies to make sure toxicity risks and drug-relevant properties of molecules which are key factors, to determine drug-likeness of lead molecules. ADMET studies were conducted on selected lead compounds using Discovery Studio (Accelrys, San Diego, CA, USA). This module uses six mathematical models, to quantitatively predict properties by a set of rules/keys that specify threshold ADMET characteristics for the chemical structure of the molecules based on the available drug information.      (2)

Binding Energy calculation:
Binding energy calculations were performed on the best 20 Compounds which have good docking score and H-bond interaction and results are listed in Table 2. In order to compare the obtained Binding energy (ΔE), calculations also performed on active HDAC2 inhibitors (SAHA). Table 2 shows the binding energy calculation (PM6) of HDAC2-inhibitor complexes. The binding energy of active HDAC2 inhibitors SAHA is -33.25 kcal/mol, and the binding energy of selected four compounds as shows these compounds have smaller binding energy than active HDAC2 inhibitors and were suggesting an inhibitors of HDAC2.The compounds, which are having hydrogen bond interactions with ARG, HIS, TYR active residues shows smaller binding energies. This implies that the active site residues ARG, HIS, TYR are become more favourable to the binding of HDAC2 inhibitors.

ADMET:
ADMET predictions were carried out to evaluate drug likeness of top 20 selected compounds and the properties were reported in Table 3 together with biplot Figure 3. The pharmacokinetic profiles of selected compounds were predicted by means of six precalculated ADMET model provided by Discovery studio. Figure 3 bi plot shows two analogous 95% and 99% confidence ellipse corresponding to HIA and BBB models. PSA have inverse relationship with human intestinal absorption and thus cell wall permeability. The log P used to estimate the lipophilicity, thus the information of H-bonding characteristics as obtained by calculating PSA could be taken into consideration along with logP calculation. The model with descriptors AlogP98 and PSA 2D with a bi-plot comprising 95% and 99% confidence ellipses was considered for the accurate prediction for the cell permeability of compounds. Selected 20 compounds had a good adsorption prediction for metabolism. In toxicity evaluation except H32 all compounds displayed CYP2D6 inhibiting and hepatotoxicity, suggesting that these compounds have no toxicity in the liver. Blood brain barrier (BBB) penetration showed that 10 compounds have good penetration; 8 compounds have low penetration and 2 compounds have undefined penetration;10 compounds may suitable for central nerve system therapy.  0.649 a Absorption: good absorption = 0;moderate absorption = 1; low absorption = 2; b BBB level (blood brain barrier): very high penetration = 0; high penetration = 1; medium penetration = 2; low penetration = 3; undefined penetration = 4. c CYP2D6: noninhibitor = 0, inhibitor = 1. d Hepatotoxicity: noninhibitor = 0, inhibitor = 1.

Conclusion:
It is of interest to identify better inhibitors for HDAC2. Here, we report the binding of 4 HDAC2 inhibitors with optimal LibDock 386 ©Biomedical Informatics (2019) score, binding energy and hydrogen-bonds. It is further noted by ADMET analysis that these compounds have good absorption, less toxic in the human liver and BBB penetration and may therefore suggest as HDAC2 inhibitors.