Structural and functional characterization of a complex between the acidic transactivation domain of EBNA2 and the Tfb1/p62 subunit of TFIIH

Abstract

Infection with the Epstein-Barr virus (EBV) can lead to a number of human diseases including Hodgkin's and Burkitt's lymphomas. The development of these EBV-linked diseases is associated with the presence of nine viral latent proteins, including the nuclear antigen 2 (EBNA2). The EBNA2 protein plays a crucial role in EBV infection through its ability to activate transcription of both host and viral genes. As part of this function, EBNA2 associates with several host transcriptional regulatory proteins, including the Tfb1/p62 (yeast/human) subunit of the general transcription factor IIH (TFIIH) and the histone acetyltransferase CBP(CREB-binding protein)/p300, through interactions with its C-terminal transactivation domain (TAD). In this manuscript, we examine the interaction of the acidic TAD of EBNA2 (residues 431-487) with the Tfb1/p62 subunit of TFIIH and CBP/p300 using nuclear magnetic resonance (NMR) spectroscopy, isothermal titration calorimeter (ITC) and transactivation studies in yeast. NMR studies show that the TAD of EBNA2 binds to the pleckstrin homology (PH) domain of Tfb1 (Tfb1PH) and that residues 448-471 (EBNA2₄₄₈₋₄₇₁) are necessary and sufficient for this interaction. NMR structural characterization of a Tfb1PH-EBNA2₄₄₈₋₄₇₁ complex demonstrates that the intrinsically disordered TAD of EBNA2 forms a 9-residue α-helix in complex with Tfb1PH. Within this helix, three hydrophobic amino acids (Trp458, Ile461 and Phe462) make a series of important interactions with Tfb1PH and their importance is validated in ITC and transactivation studies using mutants of EBNA2. In addition, NMR studies indicate that the same region of EBNA2 is also required for binding to the KIX domain of CBP/p300. This study provides an atomic level description of interactions involving the TAD of EBNA2 with target host proteins. In addition, comparison of the Tfb1PH-EBNA2₄₄₈₋₄₇₁ complex with structures of the TAD of p53 and VP16 bound to Tfb1PH highlights the versatility of intrinsically disordered acidic TADs in recognizing common target host proteins.

Figure 1. NMR titrations between Tfb1PH and the TAD of EBNA2.

(A) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled Tfb1PH in the absence (black) or presence (red) of 1 mM unlabeled EBNA2_431–487. (B) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled EBNA2_431–487 in the absence (black) or presence (red) of 1 mM (red) unlabeled Tfb1PH. (C) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled Tfb1PH in the absence (black) or presence (red) of 1 mM unlabeled EBNA2_448–471. (D) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled Tfb1PH in the presence of either 1 mM unlabeled EBNA2_448–471 (black) or 1 mM unlabeled EBNA2_431–487 (red).

Figure 2. NMR structure of the Tfb1PH-EBNA2_448–471 complex.

(A) Overlay of the backbone trace of 20 structures of the complex between Tfb1PH (in blue) and EBNA2_448–471 (in orange). The structures were superimposed using the backbone atoms C′, Cα and N of residues 4–63 and 86–112 of Tfb1PH and residues 454–464 of EBNA2_448–471. (B) Ribbon model for the structure of the Tfb1PH-EBNA2_448–471 complex.

Figure 3. Key interactions at the interface of the Tfb1PH-EBNA2_448–471 complex.

(A) Ribbon representation of Tfb1PH (blue) and EBNA_448–471 (orange) highlighting the side chains (shown in sticks) of Tfb1PH (M59, M88 and R61) that interact with the aromatic ring of Phe462 (F462) of EBNA_448–471. (B) Ribbon representation of Tfb1PH (blue) and EBNA_448–471 (orange) highlighting the side chains (shown in sticks) of Tfb1PH (M59, M88 and K57) that interact with the indole ring of Trp458 (W458) and the side chain of Ile461 (I461) of EBNA_448–471.

Figure 4. Electrostatic interactions at the interface of the Tfb1PH-EBNA2_448–471 complex.

(A) The interface of the Tfb1PH-EBNA2_448–471 complex where Tfb1PH is shown as molecular surface with the electrostatic potential mapped on the surface (red negative potential and blue positive potential). EBNA_448–471 is shown as a ribbon (orange) and the side chains of Asp459 (D459) and Glu463 (E463) are shown as sticks with the carboxyl group as a dotted surface. (B) Ribbon representation of Tfb1PH (blue) and backbone trace of the region of EBNA2_448–471 (orange) highlighting the positively charged residues on the surface of Tfb1PH (R61 and R86) and the negatively charged residues of EBNA2 (D459 and E463) in positions to potentially form electrostatic interactions (shown as sticks).

Figure 5. Dissociation constants of Tfb1PH-EBNA2_448–471 and mutants.

(A) Representative ITC thermogram obtained by successive addition of EBNA2_448–471 to Tfb1PH. Experiments are performed at 25°C in a 20 mM Tris pH 7.4 and the results fit to a single-binding site model with 1∶1 stoichiometry. (B) Comparison of the dissociation constant (K_D) values for the binding of Tfb1PH and its mutants (Q49A, K57E, M59A, R61A and M88A) to EBNA2_448–471. (C) Comparison of the K_d values for the binding of EBNA2_448–471 and its mutants (W458T, I461S and F462S) to Tfb1PH. In B–C, n.d. indicates that no heat of interaction was detected under the conditions tested for these mutants.

Figure 6. The hydrophobic residues of the ΦXXΦΦ motif from EBNA2 are important for transactivation.

LexA-EBNA2_431–487 and mutant (W458T, I461 and F462S) fusion proteins were co-transformed in yeast with the reporter LexA operator-Lac-Z fusion plasmid pSH18–34. Results are presented as the percentage of the β-galactosidase units of the tested fusion proteins relative to that of the LexA-GAL4_74–881 positive control (100%). Error bars represent standard error about the mean of a minimum of three independent experiments.

Figure 7. Tfb1PH and CBP KIX bind in a similar manner to EBNA2_448–471.

(A) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled EBNA2_448–471 in the absence (black) or presence (red) of 0.5 mM unlabeled Tfb1PH. (B) Overlay of the ¹H-¹⁵N HSQC spectra for a 0.5 mM sample of ¹⁵N-labeled EBNA2_448–471 in the absence (black) or presence (red) of 0.5 mM unlabeled CBP KIX.

Figure 8. Comparison of the TADs of p53, VP16 and EBNA2 in complex with Tfb1PH.

The structure of Tfb1PH (blue) is shown as either a ribbon (A, C and E) or molecular surface (B, D and F) in complex with the TADs of EBNA2, p53 and VP16, In A–B, the TAD of EBNA2 is shown as a ribbon (orange). The complete helix of EBNA2 is shown in A, whereas the three key hydrophobic residues of the ΦXXΦΦ motif (W458, I461 and F462) of EBNA2 are highlighted in stick (orange) on the surface of Tfb1PH in B. In C–D, the TAD p53 is shown as a ribbon (green). The complete helix of p53 is shown in A, whereas and the three key hydrophobic residues of the ΦXXΦΦ motif (I50, W53 and F54) of p53 are highlighted in stick (green) on the surface of Tfb1PH in D. In E–F, the TAD VP16 is shown as a ribbon (magenta). The complete helix of VP16 is shown in E, whereas the three key hydrophobic residues of the ΦXXΦΦ motif (F475, M478 and F479) of VP16 are highlighted in stick (magenta) on the surface of Tfb1PH in F.