Quantitative analysis of HDAC inhibition using the TGx-HDACi transcriptomic biomarker in human TK6 cells
Quantitative analysis of HDAC inhibition using the TGx-HDACi transcriptomic biomarker in human TK6 cells
Cho, E.; Williams, A.; Yauk, C. L.
AbstractTranscriptomic biomarkers provide an objective approach for extracting mechanistic information from complex gene expression datasets and may support quantitative applications in chemical assessment. TGx-HDACi is an 81-gene transcriptomic biomarker designed to identify histone deacetylase inhibitors (HDACi) in human TK6 cells. We evaluated whether benchmark concentrations (BMCs) derived from TGx-HDACi responses are concordant with BMCs based on reductions in HDAC enzyme activity. Human TK6 cells were exposed to concentration ranges of three HDAC inhibitors from different representative classes: the pan-HDAC inhibitors trichostatin A (TSA) and valproic acid (VPA), and the class I-selective inhibitor entinostat. HDAC enzyme activity and TGx-HDACi responses were assessed, and concentration-response modeling was performed to derive BMCs. TGx-HDACi correctly identified all three compounds as HDAC inhibitors and showed concentration-dependent responses. For TSA and VPA, TGx-HDACi and HDAC activity BMCs had overlapping confidence intervals, indicating good agreement between potency measures. In contrast, entinostat produced higher TGx-HDACi BMCs than the enzyme assay. This reduced quantitative concordance may reflect differences in HDAC selectivity, as TGx-HDACi was developed primarily using pan-HDAC inhibitors and appears less responsive to the more limited transcriptional changes induced by class I-selective inhibitors. Overall, these results demonstrate that TGx-HDACi reliably identifies HDAC inhibition and can provide potency estimates that align with enzyme activity measures for some HDAC inhibitor classes. However, the findings highlight the importance of considering biomarker applicability domain when interpreting potency estimates across compounds with differing HDAC selectivity.