TAF1 plays a critical role in AML1-ETO driven leukemogenesis

Abstract

AML1-ETO (AE) is a fusion transcription factor, generated by the t(8;21) translocation, that functions as a leukemia promoting oncogene. Here, we demonstrate that TATA-Box Binding Protein Associated Factor 1 (TAF1) associates with K43 acetylated AE and this association plays a pivotal role in the proliferation of AE-expressing acute myeloid leukemia (AML) cells. ChIP-sequencing indicates significant overlap of the TAF1 and AE binding sites. Knockdown of TAF1 alters the association of AE with chromatin, affecting of the expression of genes that are activated or repressed by AE. Furthermore, TAF1 is required for leukemic cell self-renewal and its reduction promotes the differentiation and apoptosis of AE+ AML cells, thereby impairing AE driven leukemogenesis. Together, our findings reveal a role of TAF1 in leukemogenesis and identify TAF1 as a potential therapeutic target for AE-expressing leukemia.

Fig. 1

Depletion of TAF1 blocks the proliferation of AE-expressing cells. a–d Knockdown of TAF1 blocks the growth of Kasumi-1 cells (a) and SKNO-1 cells (b) and has little effect on the growth of K562 cells (c) or CD34+ cells (d). Kasumi-1 cells, SKNO-1 cells, K562 cells, and CD34+ cells were infected with scrambled shRNA or TAF1-directed shRNAs. The levels of TAF1 mRNA and TAF1 protein in each type of cells infected with scrambled shRNA or two different TAF1-directed shRNAs are shown in bar graphs and western blots. The TAF1 expression levels after knockdown are indicated as percentage above each column. The cell numbers between cells infected with scrambled shRNA and cells infected with TAF1 shRNAs at last time point were compared using Student t-test. P-values are displayed. e–h Knockdown of TAF1 reduces the percentage of Kasumi-1 cells in the S phase and has no influence on K562 and CD34+ cells. Cells were infected with scrambled shRNA or TAF1 shRNAs for 4 days and subjected for BrdU assay. Representative flow cytometry pictures are shown in e. f–h The percentages of Kasumi-1 cells (f), K562 cells (g), and CD34+ cells (h) with normal or reduced TAF1 levels in the S phase are shown in bar graphs. All experiments were repeated at least two times independently, a n = 3, b–d n = 2, f n = 3, g–h n = 2. All error bars represent the mean ± SD. The percentage of cells in the S phase in TAF1 shRNA-infected cells was compared with that in scrambled shRNA-infected cells. P values were determined by Student's t-test. ns represents no significant difference, *p < 0.05, **p < 0.01

Fig. 2

TAF1 depletion promotes myeloid differentiation and impairs self-renewal. a, b Knockdown of TAF1 partially reverses the blockade of myeloid differentiation driven by AE in human CD34+ cells (a) and mouse bone marrow cells (b). To monitor myeloid differentiation, CD34+ cells were incubated in myeloid differentiation promoting medium for 4 days. Mac-1 was used as the myeloid marker for the human CD34+ cells, while Mac-1 and Gr-1 were used for mouse bone marrow cells isolated from Mx1-Cre or AE knock-in mice. a, b Representative flow cytometry pictures are shown on the left. Each Bar graph is the summary of at least two independent experiments using either CD34+ cells or bone marrow cells. * represents the comparison of the relative expression levels of differentiation marker(s) between AE group with normal TAF1 level and AE group with reduced TAF1 levels by Student's t-test and p < 0.05. c Knockdown of TAF1 reduces the self-renewal induced by AE. Serial plating assays were performed using bone marrow cells isolated from Mx1-Cre or AE knock-in mice and transduced with scrambled shRNA or TAF1-directed shRNAs. The numbers of colonies at each plating are shown as mean ± SD. d, e Knockdown of TAF1 impairs cobblestone area forming cell (CAFC) frequency in AE+ bone marrow cells (d) and CD34+ cells (e). The numbers of cobblestone area were counted at week 5 and shown as mean ± SD. The colony numbers in cells expressing AE with normal or reduced TAF1 levels were compared. *p < 0.05, ns indicates p > 0.05. P values were determined by Student's t-test. All experiments were repeated at least two times independently. a, b n = 2, c, d n = 3, e n = 2. All error bars (a–e) represent the mean ± SD

Fig. 3

Depletion of TAF1 impairs the proliferation and self-renewal of AE9a+ cells. a Schema of the generation of AE9a+ luciferase+ cell line and the collection of secondary spleen leukemia cells. b TAF1 depletion levels in secondary spleen leukemia cells. Secondary spleen leukemia cells were infected with scrambled shRNA or TAF1 direct shRNAs for 3 days and mRNA was extracted. c Knockdown of TAF1 significantly reduces the proliferation of secondary spleen leukemia cells. All groups started with the same cell numbers; cell numbers were counted on days 3, 5, and 7. The cell numbers between cells infected with scrambled shRNA and cells infected with TAF1 shRNAs at day 7 were compared using Student's t-test. d TAF1 is critical for the colony formation of secondary spleen leukemia cells. The colony number in each TAF1 KD group was compared with that in scrambled group and the relative ratios are displayed. e TAF1 is important for maintaining CAFC frequency in secondary spleen leukemia cells. The number of cobblestone areas was counted at week 5. The CAFC number in each TAF1 KD group was compared with that in scrambled group and the relative ratios are displayed. *p < 0.05, **p < 0.01, p values were determined by Student's t-test. b–e Experiments were repeated three times independently. All bars represent the mean ± SD

Fig. 4

TAF1 knockdown significantly delays leukemia development. a Kaplan–Meier plots show that TAF1 KD significantly extends the survival of recipient mice transplanted with AE9a+ luciferase+ cells. Sublethally irradiated C57BL/6J mice were injected with AE9a+ luciferase+ cells transduced with scrambled shRNA or mouse TAF1 shRNAs (n = 8 in each group). p value was determined using Log-rank (Mantel–Cox) test. b In vivo luciferase imaging indicates that knockdown of TAF1 remarkably impairs leukemia development (n = 8 in each group). Mice were injected with AE9a+ luciferase+ cells expressing wild-type level or reduced levels of TAF1. Twenty days after transplantation, IVIS imaging was performed. c The quantification of total luciferase signal in each mouse of each group as shown in (b). The luciferase signal in the TAF1 KD group was compared to the signal in the scrambled shRNA control group. d KD of TAF1 reduces the presence of GFP+ AE9a+ luciferase+ cells in the peripheral blood. Mice were injected with GFP+ AE9a+ luciferase+ cells infected with scrambled shRNA or TAF1 shRNAs. The percentage of GFP+ AE9a+ luciferase+ cells in peripheral blood of each mouse was measured 3 weeks after transplantation. The percentage of GFP+ AE9a+ in the peripheral blood of the TAF1 KD group was compared with the percentage of GFP+ cells in the scrambled shRNA group. P values were determined by Student's t-test. e Survival curves of mice injected with secondary spleen leukemia cells transduced with scrambled shRNA or TAF1 directed shRNAs. n = 8 mice in each group. P value was determined using Log-rank (Mantel–Cox) test. f The percentage of GFP+ AE9a+ cells in the peripheral blood of each mouse after receiving secondary spleen leukemia cells infected with scrambled shRNA or TAF1-directed shRNAs. Peripheral blood was collected 48 days after transplantation. The percentage of GFP+ AE9a+ cells in peripheral blood in the TAF1 KD group was compared with the percentage for the scrambled shRNA group. P values were determined by Student's t-test. *p < 0.05, **p < 0.01, ****p < 0.0001

Fig. 5

TAF1 associates with acetylated K43 on AE through its bromodomains. a TAF1 physically interacts with AE in Kasumi-1 cells. Co-immunoprecipitation was performed using anti-TAF1 antibody or normal mouse IgG. b Co-immunoprecipitation of TAF1 with AE using anti-ETO antibody or normal goat IgG. c Mutation of lysine-43 to arginine in AE reduces the interaction of AE with TAF1. 293T cells were transfected with p300, TAF1 and AE or AE mutants. Co-immunoprecipitation was performed using an anti-TAF1 antibody. d The deletion of the TAF1 bromodomain regions impairs its binding to AE. 293T cells were transfected with p300, AE, and TAF1 wild type or bromodomain deletion (ΔBr) plasmids. Co-immunoprecipitation was performed using an anti-ETO antibody

Fig. 6

TAF1 knockdown affects the expression of AE target genes. a, b The influence of AE knockdown (a) or TAF1 knockdown (b) on the expression of AE target genes in Kasumi-1 cells. Kasumi-1 cells were transduced with scrambled shRNA or AE shRNAs (a) or TAF1 shRNAs (b) for 5 days. mRNA levels of individual genes were standardized by the level of 18S rRNA. c Depletion of TAF1 impairs the expression of AE target genes in AE9a+ cells. AE9a+ cells were transduced with scrambled shRNA or TAF1 shRNAs for 5 days and the mRNA levels of individual genes were measured and standardized by mouse 18S rRNA levels. d The protein levels of AE target genes in Kasumi-1 cells and K562 cells infected with scrambled shRNA or TAF1 shRNAs. e, f Venn diagrams of the numbers of genes downregulated (e) or upregulated (f) (q < 0.05) in Kasumi-1 cells after TAF1 KD (shTAF1) or AE KD (shAE). Scrambled shRNA-infected cells were used as controls. Kyoto Encyclopedia of Genes and Genomes (KEGG 2016) gene set analysis was performed on genes downregulated (e) or upregulated by both TAF1 KD and AE KD. a–c Experiments were repeated independently three times and all bars represent the mean ± SD

Fig. 7

TAF1 is critical for the association of AE with chromatin. a Knockdown of TAF1 reduces the association of AE with chromatin in Kasumi-1 cells. Kasumi-1 cells were transduced with scrambled shRNA or TAF1 shRNAs for 5 days and then the subcellular fractionations were isolated. Total indicates the whole-cell lysate; cyto indicates the cytoplasm fraction; NS indicates the nuclear soluble fraction; chrom indicates the chromatin fraction. b Pie charts illustrate the distribution of AE peaks or TAF1 peaks across the genome. AE or TAF1 peaks were mapped to closest Refseq gene. Distal intergenic region is the region greater than 3 kb from either upstream or downstream of transcription start sites. The distance in the parentheses was measured from transcription start site. c Venn diagram illustrates the numbers of total AE peaks or TAF1 peaks or AE and TAF1 overlapping peaks in Kasumi-1 cells. Enrichment analysis for KEGG gene sets was performed on genes adjacent to overlapping TAF1 and AE peaks. d, e Venn diagram illustrates the numbers of AE peaks or TAF1 peaks or AE and TAF1 overlapping peaks at the TSS (within 1 kb of the transcription start sites) (d) (p < 1.0e-5) or at non-TSS regions (>1 kb of transcription start site including enhancers) (e) (p < 1.0e-3). P values were estimated using a Monte Carlo simulation of shuffled peaks within either the TSS background or the non-TSS genomic background. The fractions of TAF1 unique peaks, TAF1/AE co-bound peaks, and AE unique peaks at putative enhancers or non-enhancers are plotted (e, right panel). Enhancers were defined as the regions with both H3K4me and H3K27Ac peaks excluding TSS regions. f Venn diagram illustrates the numbers of AE peaks, TAF1 peaks, p300 peaks, and their overlapping peaks. g The representative picture of the peaks of p300, TAF1, AE, polymerase II (pol II), histone H3 lysine 27 acetylation (H3K27Ac), and H3 lysine 4 monomethylation (H3K4me1) at AE-activated gene ID1

Fig. 8

TAF1 bromodomain inhibition reduces the growth of AE-expressing cells. a–c The growth of Kasumi-1 cells, CD34+ cells, and K562 cells in the presence or absence of different concentrations of bromodomain inhibitors Bay-364, Bay-299, or JQ-1. The cell growth was measured by CellTiter-Glow luminescent cell viability assay after 3 days of treatment with the inhibitors or vehicle. Experiments were repeated three times and representative figures are shown. d IC₅₀s (µM) of each inhibitor in Kasumi-1, CD34+, and K562 cells. IC₅₀s were calculated from three independent experiments using GraphPad Prism. Targeted proteins by individual inhibitor are shown in parentheses. e TAF1 inhibitor Bay-364 represses the expression of AE upregulated genes. RNA was extracted from Kasumi-1 cells treated with vehicle or Bay-364 for 72 h. mRNA levels of individual genes were standardized by 18S rRNA level. f TAF1 inhibitor Bay-364 abrogates colony formation in AE9a+ cells. AE9a+ cells were treated with vehicle or 10 μM Bay-364 for 2 days and then 1000 AE9a+ cells were plated on methylcellulose-based medium containing vehicle or Bay-364. Colonies were counted 7 days after plating. The colony numbers from Bay-364-treated cells were compared to the colony numbers from the vehicle-treated cells and p value was determined by Student's t-test. *p < 0.05. g Venn diagrams illustrate the overlapping genes differentially expressed (q < 0.05) after TAF1 KD (shTAF1), AE KD (shAE), or TAF1 inhibitor Bay-364 treatment in Kasumi-1 cells. KEGG analysis was performed on 1022 genes differentially expressed after TAF1 KD, AE KD, and Bay-364 treatment. h TAF1 working model on AE target genes. p300 acetylates AE lysine-43 and the acetylated AE can be recognized by bromodomain of TAF1. TAF1 facilitates the association of AE to either the promoter or enhancer of a subset of AE target genes and then transcription is either activated or repressed. e, f Experiments were repeated independently at least two times. e n = 2, f n = 3. All bars represent the mean ± SD