Structure of the AML1-ETO eTAFH domain-HEB peptide complex and its contribution to AML1-ETO activity

Abstract

AML1-ETO is the chimeric protein product of the t(8;21) in acute myeloid leukemia. The ETO portion of the fusion protein includes the eTAFH domain, which is homologous to several TATA binding protein-associated factors (TAFs) and interacts with E proteins (E2A and HEB). It has been proposed that AML1-ETO-mediated silencing of E protein function might be important for t(8;21) leukemogenesis. Here, we determined the solution structure of a complex between the AML1-ETO eTAFH domain and an interacting peptide from HEB. On the basis of the structure, key residues in AML1-ETO for HEB association were mutated. These mutations do not impair the ability of AML1-ETO to enhance the clonogenic capacity of primary mouse bone marrow cells and do not eliminate its ability to repress proliferation or granulocyte differentiation. Therefore, the eTAFH-E protein interaction appears to contribute relatively little to the activity of AML1-ETO.

Figure 1

Binding mode is retained in the covalent eTAFH-HEB fusion protein. (A) Overlay of ¹⁵N-¹H HSQC spectra of apo-eTAFH (red), eTAFH/HEB peptide complex (blue), and eTAFH-HEB fusion protein (green). The amides observed only when eTAFH is bound to HEB (), and the amides undergoing large chemical shift changes on HEB binding (→) are labeled with the residue number. *The peak for T321 is folded in the spectrum of eTAFH-HEB (green). (B) Backbone dynamics of eTAFH-HEB represented by plots of {¹H}¹⁵N heteronuclear NOE and R₂ relaxation rate versus residue number.

Figure 2

Solution structure of the eTAFH domain–HEB peptide complex. (A) Stereoview of an ensemble of 20 lowest energy NMR solution structures. The backbone of residues G267 to Q353 of the eTAFH domain of AML1-ETO (turquoise) and residues I12 to M26 of the AD1 domain from HEB (orange) are displayed after superimposing the structures using residues Q269 to L350 of eTAFH and E16-L21 of HEB. (B) Ribbon representation of the lowest energy structure. (C) Surface of the eTAFH domain with the side chains of the binding site displayed and labeled (white indicate nonpolar residues; red, acidic residues; blue, basic residues; green, polar residues). (D) HEB backbone represented as a ribbon with the side chains interacting with eTAFH domain displayed and labeled. Vmd-Xplor was used to generate the figures.

Figure 3

eTAFH mutations impair HEB binding. (A) ITC measurements of the binding of HEB peptide to wild-type and mutant eTAFH domains. In each panel, the top portion is the raw data, and the bottom portion is a plot of the binding corrected for dilution enthalpy (squares indicate experimental data; line, fit to a one-site binding model). The average N (stoichiometry) and K_d from 2 independent experiments (± SD) for each protein are shown in the box. The stoichiometry was fixed to 1 to fit the F332A mutant data. (B) Overlay of ¹⁵N-¹H HSQC spectra of wild-type eTAFH domain (red) and the F277A mutant (blue). Most peaks in the F277A mutant are absent or shifted, indicating the structure is disrupted.

Figure 4

eTAFH domain binds peptides with a conserved (D/E)LXXLL motif from HEB, cMyb, N-CoR, and STAT6. (A) Overlays of selected amides in the ¹⁵N-¹H HSQC spectra of the eTAFH domain recorded as a function of the concentration of each peptide. Each column represents titration data for 3 separate amides with one particular peptide. (B) K_d determination using chemical shift changes resulting from titrations of each peptide. The 3 best fits from changes in ¹H^N shifts for each peptide are displayed. (C) Results of a K_d determination using fluorescence polarization with 0.2 μM fluorescein-labeled HEB peptide (FLSN-TDKELSDLLD) and increasing concentrations of eTAFH. Results of one titration are shown. Two independent experiments were carried out resulting in K_d = 12.5 ± 2.1 μM. (D) Sequence alignment of peptides examined for eTAFH binding. Consensus amino acids are colored in blue. Asterisks indicate residues in HEB whose side chains contact the eTAFH domain.

Figure 5

In vivo activities of AML1-ETO are relatively unaffected by mutations that impair HEB binding. (A) Schematic diagram of AML1-ETO and the location of mutations (asterisks). (B) Effect of eTAFH mutations on AML1-ETO's repression of granulocyte differentiation after 7 days of culture in the presence of IL-3, IL-6, SCF, and G-CSF. Cells within the forward and side scatter gates were further gated for GFP expression, and GFP-positive cells were examined for Mac-1 and Gr-1 expression. Data represent triplicate samples from 2 independent experiments. Error bars represent 95% confidence intervals. Significant differences from AML1-ETO (#) are indicated with asterisks (ANOVA and Dunnett multiple comparison test, **P < .01). (C) Representative flow of Lin⁻ bone marrow cells infected with MigR1 retroviruses expressing AML1-ETO, the m7 oligomerization mutant, and the m7+eTAFH mutants. (D) Average percentages of Gr-1⁺Mac-1⁺ cells. Significant differences relative to the m7 mutant are indicated with asterisks (ANOVA and Dunnett multiple comparison test, **P < .01, *P < .05). (E) Cos7 cells were cotransfected with AML1/ETO and its mutated derivatives and FLAG-tagged HEB. (Top) Cell lysates immunoprecipitated (IP) with anti-FLAG and blotted with antibody to the Runt domain (RD) in AML1/ETO. (Middle) Input lysate (1%) was blotted with anti-RD to detect AML1/ETO proteins. (Bottom) Membranes from the top panel were reprobed with anti-FLAG antibodies. The percentages of immunoprecipitated AML1-ETO proteins relative to input were 3.7% (AML1/ETO), 2.3% (AML1-ETO F332A), 1.1% (AML1-ETO m7), and 0.7% (AML1-ETO m7+F332A). (F) Representative flow of BrdU incorporation 48 hours after transduction of Lin⁻ bone marrow cells with MigR1 expressing GFP, AML1-ETO, or the AML1-ETO eTAFH mutants. (G) Percentage of GFP⁺ cells that had incorporated BrdU after an 1-hour BrdU pulse. Data are from 2 experiments each with triplicate samples (error bars = 95% confidence intervals; significant differences from AML1-ETO indicated with an asterisk, ANOVA and Dunnett multiple comparison test, **P < .01). (H) Serial replating of bone marrow cells. Graphs represent the average number of colonies from each round of replating in the presence of IL-3, IL-6, and SCF. Day 7 represents colony numbers per 10³ cells plated and days 14, 21, and 28 from 10⁴ plated cells. Numbers are averaged from 3 experiments, each containing triplicate samples. The numbers of colonies derived from F277A- and F332A-transduced cells were significantly lower than those from AML1-ETO–transduced cells at day 14 and day 28 (P < .01). Error bars represent 95% confidence intervals.

Figure 6

Cartoon representations of the orientation of LXXLL motifs found in 2 different classes. (A) The orientation of LXXLL motifs of coactivators on the ligand-binding domain of nuclear receptors (NRs). (B-D) The binding orientation of LXXLL motifs of transcription factors: (B) STAT6 binding the PAS-B domain of NcoA-1, (C) cMyb binding the KIX domain of CBP, (D) HEB binding the eTAFH domain of AML1-ETO. The glutamic acids of STAT6 and HEB and the arginine of cMyb providing side chain interactions are in light gray and labeled.