Identification of a Histone Deacetylase 8 Inhibitor through Drug Screenings Based on Machine Learning.

Histone deacetylase 8 (HDAC8) is a zinc-dependent HDAC that catalyzes the deacetylation of nonhistone proteins. It is involved in cancer development and HDAC8 inhibitors are promising candidates as anticancer agents. However, most reported HDAC8 inhibitors contain a hydroxamic acid moiety, which often causes mutagenicity. Therefore, we used machine learning for drug screening and attempted to identify non-hydroxamic acids as HDAC8 inhibitors. In this study, we established a prediction model based on the random forest (RF) algorithm for screening HDAC8 inhibitors because it exhibited the best predictive accuracy in the training dataset, including data generated by the synthetic minority over-sampling technique (SMOTE). Using the trained RF-SMOTE model, we screened the Osaka University library for compounds and selected 50 virtual hits. However, the 50 hits in the first screening did not show HDAC8-inhibitory activity. In the second screening, using the RF-SMOTE model, which was established by retraining the dataset including 50 inactive compounds, we identified non-hydroxamic acid 12 as an HDAC8 inhibitor with an IC50 of 842 nM. Interestingly, its IC50 values for HDAC1 and HDAC3-inhibitory activity were 38 and 12 µM, respectively, showing that compound 12 has high HDAC8 selectivity. Using machine learning, we expanded the chemical space for HDAC8 inhibitors and identified non-hydroxamic acid 12 as a novel HDAC8 selective inhibitor.

Strategies for the Discovery of Target-Specific or Isoform-Selective Modulators.

Currently, the creation of class- and isoform-selective modulators of biologically important targets is a particularly challenging problem because different isoforms within a protein family often show striking similarity in spatial quaternary structure, especially at the catalytic sites or binding pockets. Therefore, an understanding of both the precise three-dimensional structure of the target protein and the mechanisms of action of modulators is important for developing more effective and selective agents. In this Perspective, we discuss currently available rational design strategies for obtaining class- and isoform-selective inhibitors and we illustrate these strategies with the aid of specific examples from the recent literature. The strategies covered include: (1) target-derived (-dependent) de novo drug discovery methodologies, and (2) follow-on derivatization approaches from initially identified active molecules (hit-to-lead and lead-to-candidate efforts). We also comment on prospects for further development and integration of strategies to achieve target-specific or isoform-selective inhibition.

Identification of highly selective and potent histone deacetylase 3 inhibitors using click chemistry-based combinatorial fragment assembly.

To find histone deacetylase 3 (HDAC3)-selective inhibitors, a series of 504 candidates was assembled using "click chemistry", by reacting nine alkynes bearing a zinc-binding group with 56 azide building blocks in the presence of Cu(I) catalyst. Screening of the 504-member triazole library against HDAC3 and other HDAC isozymes led to the identification of potent and selective HDAC3 inhibitors T247 and T326. These compounds showed potent HDAC3 inhibition with submicromolar IC50s, whereas they did not strongly inhibit other isozymes. Compounds T247 and T326 also induced a dose-dependent selective increase of NF-κB acetylation in human colon cancer HCT116 cells, indicating selective inhibition of HDAC3 in the cells. In addition, these HDAC3-selective inhibitors induced growth inhibition of cancer cells, and activated HIV gene expression in latent HIV-infected cells. These findings indicate that HDAC3-selective inhibitors are promising candidates for anticancer drugs and antiviral agents. This work also suggests the usefulness of the click chemistry approach to find isozyme-selective HDAC inhibitors.