Fig 1: Epigenetic library screen to identify compounds with radiosensitizing and CSC-targeting effects in HNSCC cell lines. (A) Schematic illustration of the experimental screening setup including cell plating, treatment, and readout of clonogenicity after 10 days, DNA repair capacity with the γH2AX assay after 24 h, and the self-renewal potential with the sphere formation assay after 14 days. The HNSCC cell lines FaDu and Cal33 were seeded in 96-well plates followed by 24 h of pre-treatment with 5 μM of the Epigenetics Screening Library (Cayman Chemical, cat. no. 11076), consisting of 146 compounds before irradiation with 4 Gy (n = 2). (B) Combined analysis of screening results obtained from FaDu and Cal33 cell were normalized to DMSO and shown as fold change. Out of 146 compounds, 23 (15.8%) exhibited radiosensitizing effects in the colony formation assay (CFA) readout, 24 compounds (16.4%) affected self-renewal, and 7 compounds affected DNA damage repair (DDR) after irradiation (4.8%) (n = 4, * p < 0.05, one-sided paired t-test). Compounds with plating efficiency < 0.05 in CFA were regarded as cytotoxic and eliminated from further analysis. (C) The illustrative heatmap summarizes the six identified compounds with a significant radiosensitizing effect in at least two readouts. The mean values are normalized to vehicle control. The positive control 3-deazaneplanocin A (DZNeP), an inhibitor of S-adenosylmethionine-dependent methyltransferase that targets the degradation of EZH2, has the highest radiosensitizing potential. GSK-J1 targets the H3K27me3- and H3K4me3/2/1-specific demethylases UTX andJMJD3, UNC0321 the histone methyltransferase G9a, Sinefungin is a natural analog of s-adenosyl-L-methionine (SAM) and is a pan-DNA methyltransferase (DNMT) inhibitor. AK-7 targets the ADP-ribosyl transferase and deacetylase SIRT2 (sirtuin), and trans-resveratrol is a polyphenolic natural compound found in grapes and berries (* p < 0.05; ** p < 0.01). (D) Nine epigenetic targeting compounds with high radiosensitizing effects in the screening setup were further validated in the HNSCC cell lines Cal33, FaDu, UT-SCC-5, and SAS. For this, the cells were treated for 72 h with increasing concentrations of the inhibitors to determine the half maximal inhibitory concentration (IC50) with the CellTiterGlo-based cell viability assay. The exemplary curves show the results for the EZH2 inhibitor DZNeP and the JMJD3/UTX inhibitor GSK-J1. (E) The cell viability and IC50 assay were also combined with 4 Gy irradiation after a one-hour pre-treatment with the compounds. Data are presented as mean IC50 values of the pooled five HNSCC cell lines (n = 3). (F) Six out of nine previously selected compounds were further tested using 3D colony formation assay. The shown data are normalized to sham and DMSO control (n = 3, mean ± SD).