HDAC6 inhibition disrupts HDAC6-P300 interaction reshaping the cancer chromatin landscape

Abstract

Background: Histone deacetylases (HDACs) are crucial regulators of gene expression, DNA synthesis, and cellular processes, making them essential targets in cancer research. HDAC6, specifically, influences protein stability and chromatin dynamics. Despite HDAC6's potential therapeutic value, its exact role in gene regulation and chromatin remodeling needs further clarification. This study examines how HDAC6 inactivation influences lysine acetyltransferase P300 stabilization and subsequent effects on chromatin structure and function in cancer cells.

Methods and results: We employed the HDAC6 inhibitor ITF3756, siRNA, or CRISPR/Cas9 gene editing to inactivate HDAC6 in different epigenomic backgrounds. Constantly, this inactivation led to significant changes in chromatin accessibility, particularly increased acetylation of histone H3 lysines 9, 14, and 27 (ATAC-seq and H3K27Ac ChIP-seq analysis). Transcriptomics, proteomics, and gene ontology analysis revealed gene changes in cell proliferation, adhesion, migration, and apoptosis. Significantly, HDAC6 inactivation altered P300 ubiquitination, stabilizing P300 and leading to downregulating genes critical for cancer cell survival.

Conclusions: Our study highlights the substantial impact of HDAC6 inactivation on the chromatin landscape of cancer cells and suggests a role for P300 in contributing to the anticancer effects. The stabilization of P300 with HDAC6 inhibition proposes a potential shift in therapeutic focus from HDAC6 itself to its interaction with P300. This finding opens new avenues for developing targeted cancer therapies, improving our understanding of epigenetic mechanisms in cancer cells.

Keywords: Apoptosis; Cancer; Cell cycle; Deacetylase inhibitors; HAT; HDAC; Histone acetylation; Proliferation; Tumorigenesis.

Fig. 1

HDAC6 inhibitor ITF3756 and HDAC6 inactivation by CRISPR/CAS9 modulate lysine acetylation. A, B Immunofluorescence staining after ITF3756 treatment of HCC1806 cell line. Left: representative images showing the acetylation of H3K27 (A) and H3K9 (B) (green signal) after 16 h treatment with ITF3756 1 µM. DAPI (blue signal) and α-Tubulin (red signal) were used as normalizers. Images were acquired at 20X magnification. Scale bar: 100 µm. Right: densitometric analysis of the mean fluorescent intensity (MFI) of the ITF3756-treated cells (green bar) compared to solvent (black bar). Error bars indicate SEM. Data were analyzed by paired t test, N = 3; *p < 0.05; **p < 0.005. C, D Immunofluorescence staining of B16F10 KO_HDAC6 vs WT. Left: representative images showing the acetylation of H3K27 (C) and H3K9 (D) (green signal). DAPI (blue signal) and α-Tubulin (red signal) were used as normalizers. Images acquired at 40X magnification. Scale bar: 50 µm. Right: densitometric analysis of the mean fluorescent intensity (MFI) of the KO_HDAC6 condition (green bar) compared to the control (black bar). Error bars indicate SEM. Data were analyzed by unpaired t test, N = 3; *p < 0.05; **p < 0.005. E, F The graphs show the results of capillary electrophoresis experiments performed to detect H3K9 and H3K27 acetylation in the liver (E) and spleen (F) of HDAC6_KO mice (blue dots) compared to the control (WT, black dots). The β-Actin was used as a normalizer. Error bars indicate SD. Data were analyzed by unpaired t test, N = 10 for each condition; *p < 0.05

Fig. 2

HDAC6 inhibition/inactivation increases α-Tubulin K40 acetylation. A Immunofluorescence analysis. Left: representative images showing the acetylation of α-Tubulin K40 (green signal) in the HCC1806 cell line. DAPI (blue signal) was used as a normalizer. Images were acquired at 20X magnification. Scale bar: 100 µm. Right: densitometric analysis of the mean fluorescent intensity (MFI). Error bars indicate SEM. Statistical analysis was performed by paired t test, N = 3; **p < 0.005. B Quantification of α-Tubulin K40 acetylation in HCC1806 cell line. Left: the panels show representative capillary electrophoresis experiments. The right graph represents the mean protein levels of α-Tubulin K40Ac, comparing the ITF3756-treated (green bar) to the vehicle-treated cells (black bar). The GAPDH was used as a normalizer. Error bars indicate SEM. Statistical analysis was performed by paired t test, N = 3; **p < 0.005. C Immunofluorescence staining: α-Tubulin K40Ac in B16F10 cells. Left: representative images showing the acetylation of α-Tubulin K40 (green signal) in the B16F10 cell line. DAPI (blue signal) was used as a normalizer. Images acquired at 20X magnification. Scale bar: 100 µm. Right: densitometric analysis of the mean fluorescent intensity (MFI). Error bars indicate SEM. Statistical analysis was performed by paired t test, N = 3; ***p < 0.0005. (D) Quantification of α-Tubulin K40 acetylation in B16F10 cells. Left: the panel shows a representative capillary electrophoresis experiment performed on the B16F10 cell line. The graph on the right represents the mean protein level of α-Tubulin K40Ac, comparing the KO_HDAC6 (green bar) and the WT (black bar) conditions. The β-Actin was used as a normalizer. Error bars indicate SEM. Statistical analysis was performed by paired t test, N = 3; ***p < 0.0005. E The graphs show the results of capillary electrophoresis experiments, detecting acetylation of α-Tubulin: K40 in HDAC6_KO mice (blue dots) compared to controls (WT, black dots) in liver, spleen, brain, and heart tissues. Error bars indicate SD. Statistical analysis was performed by unpaired t test, N = 10; ***p < 0.0005

Fig. 3

HDAC6 inhibition/inactivation increases total HAT activity and P300 protein level. A Total HAT activity assay was performed in the HCC1806 cell line, comparing ITF3756 1 µM 16 h treatment (green bar) and the control condition (black bar). Error bars indicate SEM. Data were analyzed by paired t test, N = 3; **p < 0.005. B P300 and HDAC6 protein expression. Left: representative image showing P300 and HDAC6 proteins in the HCC1806 cell line. Right: The graph reports the mean protein levels. Vinculin was used as a normalizer. Error bars indicate SEM. Multiple comparison two-way ANOVA was used to analyze the data, N = 5; **p < 0.005. C Immunofluorescence staining. Left: representative images showing P300 (green signal) and HDAC6-Flag (red signal) in the HCC1806 cell line after 24 h of pCDNA3_HDAC6_Flag transfection and 16 h treatment with 1 µM ITF3756. DAPI (blue signal) was used as a normalizer. Images were acquired at 20X magnification. Scale bar: 100 µm. Right: densitometric analysis of the mean fluorescent intensity (MFI). Error bars indicate SEM. Data were analyzed by paired t test, N = 3; *p < 0.05. D Protein expression analysis after HDAC6 silencing and P300 inhibitor EML425 treatment. Left: the panel indicates representative image of capillary electrophoresis experiments showing P300 and HDAC6 protein levels after HDAC6 silencing, quantifying P300, HDAC6, H3K9, and H3K27 acetylation on HCC1806 cell line after HDAC6 silencing and subsequent administration of 100 µM EML425 for 16 h. Right: the graph shows the mean protein signals of P300, HDAC6, H3K9Ac, and H3K27Ac in EML425-treated cells (orange bars) compared to vehicle-treated cells (gray bars). Scramble treated (red bars) and untreated cells (black bars) were used as controls. β-Actin and Vinculin were used as normalizers. Error bars indicate SEM. Statistical analysis was performed by multiple paired t tests, N = 3; *p < 0.05; **p < 0.005. E Protein expression analysis after P300 silencing and HDAC6 inhibitor ITF3756 treatment, quantifying P300, HDAC6, H3K9, and H3K27 acetylation on HCC1806 cell line after P300 silencing and subsequent administration of 1 µM ITF3756 for 16 h. Left: representative capillary electrophoresis experiments. Right: the graph shows the mean protein signals of P300, HDAC6, H3K9Ac, and H3K27Ac in ITF3756-treated cells (blue bars) compared to vehicle-treated cells (gray bars). Scramble treated (green bars) and untreated cells (black bars) were used as controls. β-Actin and Vinculin were used as normalizers. Error bars indicate SEM. Statistical analysis was performed by multiple paired t tests, N = 3; *p < 0.05, **p < 0.005. F Total HAT activity in the B16F10 cell line (HDAC6_KO, green bar, vs WT, black bar). Error bars indicate SEM. Data were analyzed by unpaired t test, N = 4; ***p < 0.0005. G Left: representative capillary electrophoresis experiment showing P300 and HDAC6 proteins in B16F10 cells. Right: mean protein levels of P300 and HDAC6. Vinculin was used as a normalizer. Error bars indicate SEM. Multiple comparison two-way ANOVA was used to analyze the data, N = 4; *p < 0.05; ***p < 0.0005

Fig. 4

ITF3576 interferes with HDAC6/P300 association. A Left: Co-immunoprecipitation capillary electrophoresis experiments were performed on HEK293T cells transfected with pCDNA3.1 + _HDAC6_Flag and pCDNA3.1 + as a control for 24 h and treated with 1 µM ITF3756 for an additional 16 h; DMSO was used as solvent control. The arrows indicate a decreased P300-HDAC6 interaction upon ITF3756 treatment. Right, top: mean protein levels of free P300 upon ITF3576 or vehicle treatment and immunoprecipitation with an anti-P300 antibody in pCDNA3.1 + _HDAC6_Flag-transfected cells. Right, bottom: HDAC6-Flag protein levels upon ITF3576 or vehicle treatment and immunoprecipitation with an anti-P300 antibody in pCDNA3.1 + _HDAC6_Flag-transfected cells-, normalized on respective P300 content. Error bars indicate SEM. Data were analyzed by unpaired t test, N = 3; *p < 0.05. B Immunofluorescence. Representative images showing P300 (green signal) and HDAC6-Flag (red signal) in the HCC1806 cell line after 24 h of pCDNA3_HDAC6_Flag transfection and 16 h of treatment with 1 µM ITF3756. DAPI (blue signal) was used as a normalizer. Images were acquired at 20X magnification. Scale bar: 100 µm. C ITF3576 impairs P300 ubiquitination. Top: capillary electrophoresis experiments were performed on HCC1806 cells treated with 1 µM ITF3756 for 16 h compared to solvent control. Samples were immunoprecipitated with an anti-P300 antibody, blotted with an anti-P300 antibody (on the middle left), and an anti-polyubiquitin antibody (on the right). Arrows on the left indicate P300 (red) and the normalizer Vinculin (blue). The red arrows on the right show the P300 polyubiquitinated species. Bottom, left: P300 mean protein levels in ITF3756-treated cells (green bar) compared to solvent (black bar), measuring immunoprecipitated and blotted with anti-P300 antibody samples. Bottom, right: quantification of the mean content of the P300 ubiquitinated species, measuring samples immunoprecipitated with anti-P300 antibody and blotted with anti-polyubiquitin antibody, in cells treated with ITF3756 (red bar) compared to the control (black bar), and normalized on the respective P300 amount. Error bars indicate SEM. Data were analyzed by unpaired t test, N = 3; *p < 0.05

Fig. 5

Chromatin accessibility landscape of KO_HDAC6 vs. WT B16F10 cells revealed by ATAC-Seq. A The principal component analysis results on the whole dataset in KO_HDAC6 B16F10 cells (green dots) compared to WT control (black dots). N = 3. B Volcano plot of differentially accessible regions. The accessible regions are toward the right, the inaccessible are toward the left, and the most statistically significant are toward the top. Red dots represent significant open and closed regions with − 0.5 ≤ shrunken LogFC ≤ 0.5 and a p-adjusted value ≤ 0.05. Black dots represent non-significant changes. N = 3. C The graph illustrates the distribution of the differentially accessible regions of specific genomic features depicted as follows: distal intergenic regions (green), downstream regions (yellow) encompassing 3000 bp from transcription end site (TES), introns (red), coding exons (orange), 3′ UTRs (light blue) and 5′ UTRs (pink), proximal (0–1000 bp, blue), and distal promoter (violet) encompassing 3000 bp upstream of the transcription start site (TSS). D Gene ontology analysis of differentially accessible regions between KO_HDAC6 and WT cells: inaccessible regions (ICR, blue bar graph, upper panel) and accessible chromatin regions (ACR, red bar graph, lower panel). N = 3

Fig. 6

Chromatin accessibility landscape of KO_HDAC6 vs WT B16F10 cells revealed by H3K27Ac-ChIP-seq. A The principal component analysis results on the whole dataset in KO_HDAC6 B16F10 cells (green dots) compared to WT control (black dots). N = 3. B Volcano plot of differentially accessible regions. The H3K27Ac enriched regions are toward the right, the H3K27Ac decreased regions are toward the left, and the most statistically significant are toward the top. Red dots represent significant H3K27Ac enriched and decreased regions with − 0.5 ≤ shrunken LogFC ≤ 0.5 and a p-adjusted value ≤ 0.05. Black dots represent non-significant changes. N = 3. C The graph illustrates the distribution of the differentially accessible regions of specific genomic features depicted as follows: distal intergenic regions (green), downstream regions (yellow) encompassing 3000 bp from transcription end site (TES), introns (red), coding exons (orange), 3′ UTRs (light blue) and 5′ UTRs (pink), proximal (0–1000 bp, blue), and distal promoter (violet) encompassing 3000 bp upstream of the transcription start site (TSS). D Gene ontology analysis of differentially accessible regions between KO_HDAC6 and WT cells revealed by H3K27Ac-ChIP-seq: inaccessible regions (ICR, blue bar graph, upper panel) and accessible chromatin regions (ACR, red bar graph, lower panel). N = 3

Fig. 7

ATAC-seq and H3K27Ac-ChIP-seq integrated analysis in KO_HDAC6 vs WT B16F10 cells. A Venn diagram (left) and gene network analysis (right) show the intersection between downregulated genes identified by ATAC-seq and H3K27Ac-ChIP-seq analysis associated with cell proliferation, cell adhesion, and apoptosis. B Venn diagram (left) and gene network analysis (right) show the intersection between upregulated genes identified by ATAC-seq and H3K27Ac-ChIP-seq analysis associated with cell proliferation, cell adhesion, and apoptosis. C GO target validation: mRNA levels assayed by RT-qPCR in B16F10 HDAC6_KO (green bars) vs WT (black bars). Gapdh was used as a normalizer, and relative mRNA expression was calculated using the 2^−ΔΔCt method. Error bars indicate SEM. Multiple t tests, N = 5, performed statistical analysis; ***p < 0.0005