Fig. 6: IL-8 signaling alters histone marks, promoting post-therapy epigenetic plasticity.

a Correlation between IL-8 and 12042 genes from TCGA was determined by Pearson correlation coefficients. Sixty-eight genes with coefficients > 0.5 or < −0.5 and false discovery rate (FDR) < 0.05 were selected. Unsupervised hierarchical clustering of those genes found two clusters with the highest cophenetic coefficient at 0.95 and an average silhouette width at 0.46. Principal component analysis was used to validate these clusters. Enrichment analysis found one cluster of genes was enriched at cell chemotaxis (GO: 0060326, adjusted p-value 1.058e-11) and cytokine activity (GO: 0005125, adjusted p-value 2.495e-9), while another cluster of genes enriched at wounding (GO: 009611, adjusted p-value 0.002) and hypoxia (GO: 0001666, adjusted p-value 0.004). b Representative immunoblot of different histone marks. A panel of PDX lines from a different subtype of GBMs was exposed to IL-8 (50 ng/ml) for 24 h. Nuclei were extracted from the harvested cells and subjected to immunoblot analysis for suppressive histone marks H3K27 and H3K9 trimethylation (me3) and activating mark H3K27 acetylation(ac). Immunoblotting for total histone three was performed to confirm the equal loading. c The extracted nuclei from the U251 IL-8 knockdown cells as described in Fig. 5f were subjected to immunoblot analysis for various histone marks as described above. d GBM43 cells were treated with IL-8 (50 ng/ml) for 2 and 24 h, cytoplasmic and nuclear extract was prepared, cells were harvested, and immunoprecipitation assays were performed with the anti-EZH2 antibody. Immunoprecipitated protein was subjected to immunoblot analysis with antibodies against phosphor-EZH2 (S21, and Thr345), and SUZ12. β-Actin and histone 3 were used as loading controls. e IL-8 (50 ng/ml)-treated GBM43 PDX lines were harvested at 6, 12, and 24 h post IL-8 exposure. mRNA was extracted and subjected to reverse-transcription polymerase chain reaction (RT-PCR) analysis of OLIG2, SERPINB2, IL6R, and BMP2K transcripts. Bars represent means from two experiments in triplicate and error bars represent the standard deviation. Multiple Student's t tests were performed. **p < 0.01, ****p < 0.0001. f A panel of GBM PDX lines was treated with TMZ (50 µM) for 48 h. Nuclei were extracted from the harvested cells and subjected to immunoblot analysis for suppressive histone marks H3K27me3 and H3K9me2 and activating mark H3K27ac and H3K4me3. Immunoblotting for total histone three was performed to confirm equal loading. g The U251 IL-8 control and knockdown cells as described in Fig. 5a were treated with TMZ(50 µM) for 4 days. Nuclei were extracted from the harvested cells and subjected to immunoblot analysis for suppressive histone marks H3K27me3 and H3K9me2 and activating mark H3K27ac. Left, representative densitometry analysis is expressed as percent of control shRNA (shCtrl). h The GBM43 PDX line was treated with TMZ(50 µM) in the presence of 3-deazaneplanocin A (DZNep, EZH2-I, 5 µmol/L), a histone methyltransferase EZH2 inhibitor for 8 days. Cells were harvested and the GIC population was analyzed by FACS analysis of the CD133 and CD15 positive cells. Bars represent means from two experiments in triplicate and error bars represent the standard deviation. Multiple Student's t tests were performed. ***p < 0.001