Harnessing the TAF1 Acetyltransferase for Targeted Acetylation of the Tumor Suppressor p53

Abstract

Pharmacological reactivation of the tumor suppressor p53 remains a key challenge for the treatment of cancer. Acetylation Targeting Chimera (AceTAC), a novel technology is previously reported that hijacks lysine acetyltransferases p300/CBP to acetylate the p53Y220C mutant. However, p300/CBP are the only acetyltransferases harnessed for AceTAC development to date. In this study, it is demonstrated for the first time that the TAF1 acetyltransferase can be recruited to acetylate p53Y220C. A novel TAF1-recruiting AceTAC, MS172 is discovered, which effectively acetylates p53Y220C lysine 382 in a concentration-, time- and TAF1-dependent manner via inducing the ternary complex formation between p53Y220C and TAF1. Notably, MS172 suppresses the proliferation in multiple p53Y220C-harboring cancer cell lines more potently than the previously reported p300/CBP-recruiting p53Y220C AceTAC MS78 with little toxicity in p53 WT and normal cells. Additionally, MS172 is bioavailable in mice and suitable for in vivo efficacy studies. Lastly, novel upregulation of metallothionine proteins by MS172-induced p53Y220C acetylation is discovered using RNA-seq and RT-qPCR studies. This work demonstrates that TAF1 can be harnessed for AceTAC development and expands the very limited repertoire of the acetyltransferases that can be leveraged for developing AceTACs, thus advancing the targeted protein acetylation field.

Keywords: TAF1; acetac; p53y220c; targeted protein acetylation.

Figure 1

Design and evaluation of TAF1‐recruiting p53Y220C‐targeting AceTACs. A) Co‐crystal structure of the p53Y220C‐PK9328 complex (PDB ID: 6GGF).^{[ 14 ]} Left: the cross‐section of the p53Y220C binding pocket (in gray) occupied by PK9328 (in magenta). The solvent‐exposed region of PK9328 is highlighted by the red dashed cycle. Right: the chemical structure of PK9328. B) Co‐crystal structure of the TAF1 (BD2)‐GNE‐371 complex (PDB ID: 6DF7).^{[ 19 ]} Left: the cross‐section of the TAF1‐BD2 binding pocket (in gray) occupied by GNE‐371 (in magenta). The solvent‐exposed region of GNE‐371 is highlighted by the red dashed cycle. Right: the chemical structure of GNE‐371. C) Chemical structures of alkylene linker‐based p53Y220C AceTAC compounds 1–8. D) Chemical structures of PEG linker‐based p53Y220C AceTAC compounds 9–12. E) Representative western blot (WB) results of PK9328, compounds 1–12, and GNE‐371 in H1299‐ p53Y220C stable cells treated with the indicated compound at 5 µm for 8 h (from two independent experiments). Total cell lysate was used for WB and vinculin was used as a loading control. F) Quantification of the fold change of the p53K382ac level (p53K382ac level over total p53 protein level) for the WB results shown in panel E and its biological repeat.

Figure 2

MS172 induces p53Y220C acetylation in a concentration‐ and time‐dependent manner in BxPC3 cells. A) Chemical structure of MS172. B) WB results of the p53K382ac level in BxPC3 cells treated with PK9328, GNE‐371, or MS172 at 0, 100, 300, or 1000 nm for 24 h. The results shown are representative of three independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. C) Left: WB results of the p53K382ac level in BxPC3 cells treated with MS172 at 0, 3, 10, 30, 100, 300 or 1000 nm for 24 h. The results shown are representative of three independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. Right: quantification of the fold change of the p53K382ac level shown on the left and its biological repeats. The results shown are the mean values ± SD from three independent experiments. D) Total p53 lysine acetylation in BxPC3 cells treated with PK9328 or MS172 at 100, 300, or 1000 nm for 24 h. A p53 mouse monoclonal antibody (DO‐1) was coated onto the microwells of an ELISA plate. After incubation with total cell lysates for 2 h, p53 was captured by the coated antibody. Following extensive washing, acetylated‐lysine rabbit monoclonal antibody was added to detect the total acetylated lysine on p53. An anti‐rabbit IgG with horse‐radish‐peroxidase (HRP) linked antibody was then used to recognize the bound detection antibody. Vehicle treatment was used to normalize the fold change of the total p53 lysine acetylation level. The results shown are the mean values ± SD from three independent experiments. ^* p <.05, ^*** p <.001. E) WB results of the p53K382ac level in BxPC3 cells treated with 1000 nm of MS172 at the indicated time point. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. F) Quantification of the fold change of the p53K382ac level from the WB results shown in panel E and its biological repeat. Vehicle treatment from each specific time point was used to normalize the fold change of the p53K382ac level.

Figure 3

MS172 induces ternary complex formation between p53Y220C and TAF1. Thermal shift assay (TSA) results of p53Y220C‐DBD A) and TAF1‐BD2 B) induced by MS172. The results shown are the mean values ± SD from three biological repeats. C) FRET‐based biochemical assay results for GNE‐371‐ and MS172‐mediated ternary complex formation between GST‐TAF1‐BD2 and p53Y220C‐DBD‐His. Briefly, anti‐GST and anti‐His HTRF antibody pairs bind to GST‐tagged TAF1‐BD2 and His‐tagged p53Y220C‐DBD. A HTRF signal is generated only if the compound binds and the ternary complex is formed and a signal is normalized to the DMSO treatment. The results shown are the mean values ± SD from two biological repeats. D) Representative WB results of MS172‐mediated p53Y220C‐TAF1 interaction via p53‐FLAG pull‐down in H1299‐p53‐null and H1299‐p53Y220C cells treated with MS172 at 0, 1, or 10 µm for 24 h. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. E) Representative WB results of p53‐FLAG pull‐down after treatment of H1299‐p53Y220C cells with DMSO or MS172 at 10 µm for 24 h with either control shRNA or TAF1 shRNA. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. F) WB results of p53‐FLAG pulldown after treatment of H1299‐p53Y220C cells with GNE‐371 alone at 10 µm, MS172 alone at 3 µm, or GNE‐371 pretreatment at 10 µm for 4 h, followed by MS172 treatment at 3 µm for 20 h. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control.

Figure 4

MS172 effectively acetylates p53Y220C and inhibits the growth in p53Y220C‐harboring cancer cell lines and is non‐toxic in p53 WT cells. WB results of PK9328, GNE‐371, MS78, and MS172 in A) BxPC3, B) NUGC3, and C) Huh7 cells treated with the indicated compound at 0, 100, 300, or 1000 nm for 24 h. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. D) Total p53 lysine acetylation in BxPC3, NUGC3 and Huh7 cells treated with PK9328, GNE‐371, MS78, or MS172 at 1000 nm for 24 h. Vehicle treatment was used to normalize the fold change of the total p53 lysine acetylation level. The results shown are the mean values ± SD from three independent experiments. ^* p <.05, ^** p <.01. E) Cell viability of PK9328, GNE‐371, MS78 and MS172 in BxPC3, NUGC3, and Huh7 cells. The cells were treated with DMSO or the indicated compound for 72 h. The mean values ± SD from three biological experiments (each in technical triplicates) are shown. GraphPad Prism 8 was used in the analysis of raw data. F) WB results of PK9328, GNE‐371, and MS172 in U2OS cells treated with the indicated compound at 0, 100, 300, or 1000 nm for 24 h. The results shown are representative of two independent experiments. Total cell lysate was used for WB and vinculin was used as a loading control. G) Cell viability of PK9328, GNE‐371, and MS172 in U2OS cells. The cells were treated with the indicated compound for 72 h. The mean value ± SD for each concentration (in technical triplicates from two biological experiments) is shown in the curves. GraphPad Prism 8 was used in the analysis of raw data.

Figure 5

MS172‐mediated p53Y220C acetylation induces regulation of metallothionine proteins in BxPC3 cells. A) Heatmap enrichment in high confidence p53‐target genes. BxPC3 cells were treated with DMSO, PK9328 or MS172 at 5 µm for 4 h in triplicates. B) Volcano plot of differential gene expression (DGE) of upregulated proteins and pathways upon MS172 treatment compared to PK9328 treatment after normalization with DMSO treatment. C) Enrichment plot of significant genes upregulated from the 343 high‐confidence p53‐target genes (q‐value <.01, normalized enrichment score (NES) = 1.49) in BxPC3 cells treated with 5 µm of MS172 compared to 5 µm of PK9328 treatment for 4 h in triplicates. D) KEGG pathway analysis of upregulated pathways upon 5 µm of MS172 treatment compared to 5 µm of PK9328 for 4 h in triplicates. E) RT‐qPCR of MT1X, MT1N, ZNF391, and ZNF778 expression in BxPC3 cells treated with DMSO, PK9328, or MS172 at 0.2, 1, or 5 µm for 4 h from three independent experiments. The mRNA expression for each gene was first normalized to internal GAPDH and then calculated relative to the DMSO control. ^* p <.05, ^** p <.01, ^*** p <.001.