A Thoroughly Validated Virtual Screening Strategy for Discovery of Novel HDAC3 Inhibitors

Abstract

Histone deacetylase 3 (HDAC3) has been recently identified as a potential target for the treatment of cancer and other diseases, such as chronic inflammation, neurodegenerative diseases, and diabetes. Virtual screening (VS) is currently a routine technique for hit identification, but its success depends on rational development of VS strategies. To facilitate this process, we applied our previously released benchmarking dataset, i.e., MUBD-HDAC3 to the evaluation of structure-based VS (SBVS) and ligand-based VS (LBVS) combinatorial approaches. We have identified FRED (Chemgauss4) docking against a structural model of HDAC3, i.e., SAHA-3 generated by a computationally inexpensive “flexible docking”, as the best SBVS approach and a common feature pharmacophore model, i.e., Hypo1 generated by Catalyst/HipHop as the optimal model for LBVS. We then developed a pipeline that was composed of Hypo1, FRED (Chemgauss4), and SAHA-3 sequentially, and demonstrated that it was superior to other combinations in terms of ligand enrichment. In summary, we present the first highly-validated, rationally-designed VS strategy specific to HDAC3 inhibitor discovery. The constructed pipeline is publicly accessible for the scientific community to identify novel HDAC3 inhibitors in a time-efficient and cost-effective way.

Sequential ligand- and structure-based virtual screening approach for the identification of potential G protein-coupled estrogen receptor-1 (GPER-1) modulators

“…Recently, Huabin Hu et al identied FRED (Chemgauss4 score) as the best docking score among three different docking score functions. 51 A maximum of 200 conformers per ligand was generated and used as a set of input. Docking of reported GPER-1 ligands and selected virtual hit compounds were performed using the same protocol as described in method 2.3.3.…”

The discovery of novel HDAC3 inhibitors via virtual screening and in vitro bioassay

“…To facilitate cherry-picking of compounds from the set of 744 potential hits, we constructed a knowledge-based PF specific to HDAC3 by using our unique cheminformatics method 40 . The PF was trained based on our model of HDAC3 bound to SAHA 39 , and it was able to classify poses in a CV accuracy of 86.75% and a prediction accuracy of 87%. We applied the PF by coupling it with docking program(s).…”

“…MUBD-HDAC3 facilitated the rational design of the VS strategy/pipeline, i.e. Hypo1_FRED_SAHA-3 for HDAC3Is discovery 39 . Another important issue to which we have been dedicated was the construction of target-specific pose filters (PFs)/classifiers, which aimed to replace the role of human experts in cherry-picking of compounds for bioassay by automatically inspecting binding modes.…”

Decoys Selection in Benchmarking Datasets: Overview and Perspectives

“…Indeed, MUV relies on the availability of experimental data and is restricted to well-studied targets. The authors subsequently proposed the Maximum Unbiased Benchmarking Data sets (MUBD, see section Benchmarking Databases) that was applied to GPCRs (Xia et al, 2014 ), HDACs (Xia et al, 2015 ; Hu et al, 2017 ) and Toll-like receptor 8 (Pei et al, 2015 ). The MUBD-DecoyMaker algorithm relies on both a minimal and required topological dissimilarity ( sims ) between decoy and active compounds, but makes use of an additional criterion that minimizes the simsdiff parameter, i.e., ensures that decoy and active compounds are as similar as possible.…”