TATA box-binding protein (TBP)-related factor 2 (TRF2), a third member of the TBP family.

The TATA box-binding protein (TBP) is an essential component of the RNA polymerase II transcription apparatus in eukaryotic cells. Until recently, it was thought that the general transcriptional machinery was largely invariant and relied on a single TBP, whereas a large and diverse collection of activators and repressors were primarily responsible for imparting specificity to transcription initiation. However, it now appears that the "basal" transcriptional machinery also contributes to specificity via tissue-specific versions of TBP-associated factors as well as a tissue-specific TBP-related factor (TRF1) responsible for gene selectivity in Drosophila. Here we report the cloning of a TBP-related factor (TRF2) that is found in humans, Drosophila, Caenorhabditis elegans, and other metazoans. Like TRF1 and TBP, TRF2 binds transcription factor IIA (TFIIA) and TFIIB and appears to be part of a larger protein complex. TRF2's primary amino acid structure suggests divergence in the putative DNA binding domain, and not surprisingly, it fails to bind to DNA containing canonical TATA boxes. Most importantly, TRF2 is associated with loci on Drosophila chromosomes distinct from either TBP or TRF1, so it may have different promoter specificity and regulate a select subset of genes. These findings suggest that metazoans have evolved multiple TBPs to accommodate the vast increase in genes and expression patterns during development and cellular differentiation.

Biochemical and biophysical characterization of the trimerization domain from the heat shock transcription factor.

Previously, we had characterized a 91 amino acid fragment of the heat shock transcription factor from the yeast Kluyveromyces lactis and had shown it to be highly alpha-helical and sufficient for formation of homotrimers [Peteranderl, R., and Nelson, H. C. M. (1992) Biochemistry 31, 12272-12276]. Based on those data, as well as the presence of hydrophobic heptad repeats, we postulated that the trimerization domain contains a three-stranded coiled-coil and that it might resemble the trimerization domain found in influenza hemagglutinin. Here, we further characterize the trimerization domain and show that the minimal domain needs 71 residues to remain trimeric and highly alpha-helical. 19F NMR spectroscopy suggests that the structure contains three parallel strands that are in register along the long axis of the coiled-coil. Electron paramagnetic resonance spectroscopy studies show that the C-termini of the subunits are in close proximity; this is in contrast to the topology of the hemaglutinin trimerization domain where the C-termini form buttressing helices. Analytical ultracentrifugation also confirms that the structure is elongated and unlikely to have buttressing helices. Additional experiments suggest that the trimerization domain has at least two subdomains. The first subdomain has the potential to form trimers independently, though not as stably as the complete domain. The second subdomain is quite helical, forms large oligomers, and appears to provide stability to the complete domain. Our current model for the heat shock transcription factor trimerization domain is a highly elongated coiled-coil structure, with a potential break in the coiled-coil region located between the two subdomains.

HIV-1 gp41 tertiary structure studied by EPR spectroscopy.

HIV gp41 is the transmembrane glycoprotein responsible for fusion of viral and cellular membranes, enabling viral entry. The structure of gp41 was studied using two synthetic peptides derived from the ectodomain of gp41: a 38-residue peptide from the "heptad repeat" region (hr.wt), and a 34-residue peptide from a region closer to the C-terminus (bt wt). These peptides were found to form a trimer of heterodimers with approximately 80% alpha-helicity. To study their alignment, distances between spin-labels attached to Cys residues on Cys-substituted peptides were measured using a recently-developed electron paramagnetic resonance method [Rabenstein, M.D., & Shin, Y.-K. (1995) Proc Natl. Acad. Sci. U.S.A. 92, 8239-8243]. The heterotrimeric peptides were found to be antiparallel, consistent with a study on proteolytically cleaved peptide fragments of gp41 [Lu, M., Blacklow, S.C., & Kim, P.S. (1995) Nat. Struct. Biol. 2, 1075-1082]. Furthermore, the C-terminal 19 residues of hr.wt are not apposed to bt.wt, and 15 residues of bt.wt extend beyond the end of br.wt. Consistent with this alignment are tertiary interactions between specific sites of these peptides probed by spin-label mobility. Additionally, a second pair of peptides was studied. From the model, these are expected to align with complete overlap. Alone, neither was helical, but when mixed, they were 83% helical. Based on the alignment of the peptides, a model of the prefusogenic form of gp41 was constructed which is significantly different from the structure of influenza hemagglutinin.

A peptide from the heptad repeat of human immunodeficiency virus gp41 shows both membrane binding and coiled-coil formation.

The envelope glycoprotein gp41 from human immunodeficiency virus type 1 (HIV-1) is involved in membrane fusion and virus entry. It contains a functionally important leucine zipper-like heptad repeat region (residues 553-590). To investigate the solution structure and membrane-binding properties of this region, cysteine-substituted variants of a 38-residue peptide derived from the heptad repeat were synthesized and modified with nitroxide spin labels. Analytical equilibrium ultracentrifugation studies indicated it is primarily tetrameric in solution, in contrast to the protein gp160 which is a mixture of trimers and tetramers. Electron paramagnetic resonance (EPR) measurements indicated that the peptide was bound to vesicles containing 10 mol % negatively charged lipids. The peptides were bound parallel to the membrane surface, near the water-membrane interface, in a structure different from the solution structure, most likely as monomers. When Asp, Pro, or Ser was substituted for Ile at the core "a" position of the heptad repeat in the middle of the peptide, the coiled coil was destabilized. In addition, these peptides showed reduced membrane-binding affinities. Thus, mutations that destabilized coiled-coil formation also decreased membrane-binding propensity. These experimental results, taken with previous evidence, suggest two functions for the heptad repeat of gp41 after CD4 binding: (1) to form an extended coiled coil; (2) to provide a hydrophobic face that binds to the host-cell membrane, bringing the viral and cellular membranes closer and facilitating fusion.

Determination of the distance between two spin labels attached to a macromolecule.

An EPR "spectroscopic ruler" was developed using a series of alpha-helical polypeptides, each modified with two nitroxide spin labels. The EPR line broadening due to electron-electron dipolar interactions in the frozen state was determined using the Fourier deconvolution method. These dipolar spectra were then used to estimate the distances between the two nitroxides separated by 8-25 A. Results agreed well with a simple alpha-helical model. The standard deviation from the model system was 0.9 A in the range of 8-25 A. This technique is applicable to complex systems such as membrane receptors and channels, which are difficult to access with high-resolution NMR or x-ray crystallography, and is expected to be particularly useful for systems for which optical methods are hampered by the presence of light-interfering membranes or chromophores.