Crystal structure of the human BRD2 bromodomain: insights into dimerization and recognition of acetylated histone H4.

The BET (bromodomains and extra terminal domain) family proteins recognize acetylated chromatin through their bromodomain and act as transcriptional activators. One of the BET proteins, BRD2, associates with the transcription factor E2F, the mediator components CDK8 and TRAP220, and RNA polymerase II, as well as with acetylated chromatin during mitosis. BRD2 contains two bromodomains (BD1 and BD2), which are considered to be responsible for binding to acetylated chromatin. The BRD2 protein specifically recognizes the histone H4 tail acetylated at Lys12. Here, we report the crystal structure of the N-terminal bromodomain (BD1, residues 74-194) of human BRD2. Strikingly, the BRD2 BD1 protein forms an intact dimer in the crystal. This is the first observation of a homodimer among the known bromodomain structures, through the buried hydrophobic core region at the interface. Biochemical studies also demonstrated BRD2 BD1 dimer formation in solution. The two acetyllysine-binding pockets and a negatively charged secondary binding pocket, produced at the dimer interface in BRD2 BD1, may be the unique features that allow BRD2 BD1 to selectively bind to the acetylated H4 tail.

The neural repressor NRSF/REST binds the PAH1 domain of the Sin3 corepressor by using its distinct short hydrophobic helix.

In non-neuronal cells and neuronal progenitors, many neuron-specific genes are repressed by a neural restrictive silencer factor (NRSF)/repressor element 1 silencing transcription factor (REST), which is an essential transcriptional repressor recruiting the Sin3-HDAC complex. Sin3 contains four paired amphipathic helix (PAH) domains, PAH1, PAH2, PAH3 and PAH4. A specific target repressor for Sin3 is likely to bind to one of them independently. So far, only the tertiary structures of PAH2 domain complexes, when bound to the Sin3-interacting domains of Mad1 and HBP1, have been determined. Here, we reveal that the N-terminal repressor domain of NRSF/REST binds to the PAH1 domain of mSin3B, and determine the structure of the PAH1 domain associated with the NRSF/REST minimal repressor domain. Compared to the PAH2 structure, PAH1 holds a rather globular four-helix bundle structure with a semi-ordered C-terminal tail. In contrast to the amphipathic alpha-helix of Mad1 or HBP1 bound to PAH2, the short hydrophobic alpha-helix of NRSF/REST is captured in the cleft of PAH1. A nuclear hormone receptor corepressor, N-CoR has been found to bind to the PAH1 domain with a lower affinity than NRSF/REST by using its C-terminal region, which contains fewer hydrophobic amino acid residues than the NRSF/REST helix. For strong binding to a repressor, PAH1 seems to require a short alpha-helix consisting of mostly hydrophobic amino acid residues within the repressor. Each of the four PAH domains of Sin3 seems to interact with a characteristic helix of a specific repressor; PAH1 needs a mostly hydrophobic helix and PAH2 needs an amphipathic helix in each target repressor.

Crystal structure of an enhancer of rudimentary homolog (ERH) at 2.1 Angstroms resolution.

The enhancer of rudimentary gene, e(r), of Drosophila melanogaster encodes an enhancer of rudimentary (ER) protein with functions implicated in pyrimidine biosynthesis and the cell cycle. The ER homolog (ERH) is highly conserved among vertebrates, invertebrates, and plants. Xenopus laevis ERH was reported to be a transcriptional repressor. Here we report the 2.1 Angstroms crystal structure of murine ERH (Protein Data Bank ID 1WZ7), determined by the multiwavelength anomalous dispersion (MAD) method. The monomeric structure of ERH comprises a single domain consisting of three alpha-helices and four beta-strands, which is a novel fold. In the crystal structure, ERH assumes a dimeric structure, through interactions between the beta-sheet regions. The formation of an ERH dimer is consistent with the results of analytical ultracentrifugation. The residues at the core region and at the dimer interface are highly conserved, suggesting the conservation of the dimer formation as well as the monomer fold. The long flexible loop (44 approximately 53) is also significantly conserved, suggesting that this loop region may be important for the functions of ERH. In addition, the putative phosphorylation sites are located at the start of the beta2-strand (Thr18) and at the start of the alpha1-helix (Ser24), implying that the phosphorylation might cause some structural changes.