The wing in yeast heat shock transcription factor (HSF) DNA-binding domain is required for full activity.

The yeast heat shock transcription factor (HSF) belongs to the winged helix family of proteins. HSF binds DNA as a trimer, and additional trimers can bind DNA co-operatively. Unlike other winged helix-turn-helix proteins, HSF's wing does not appear to contact DNA, as based on a previously solved crystal structure. Instead, the structure implies that the wing is involved in protein-protein interactions, possibly within a trimer or between adjacent trimers. To understand the function of the wing in the HSF DNA-binding domain, a Saccharomyces cerevisiae strain was created that expresses a wingless HSF protein. This strain grows normally at 30 degrees C, but shows a decrease in reporter gene expression during constitutive and heat-shocked conditions. Removal of the wing does not affect the stability or trimeric nature of a protein fragment containing the DNA-binding and trimerization domains. Removal of the wing does result in a decrease in DNA-binding affinity. This defect was mainly observed in the ability to form the first trimer-bound complex, as the formation of larger complexes is unaffected by the deletion. Our results suggest that the wing is not involved in the highly co-operative nature of HSF binding, but may be important in stabilizing the first trimer bound to DNA.

Substitutions in bacteriophage T4 AsiA and Escherichia coli sigma(70) that suppress T4 motA activation mutations.

Bacteriophage T4 middle-mode transcription requires two phage-encoded proteins, the MotA transcription factor and AsiA coactivator, along with Escherichia coli RNA polymerase holoenzyme containing the sigma(70) subunit. A motA positive control (pc) mutant, motA-pc1, was used to select for suppressor mutations that alter other proteins in the transcription complex. Separate genetic selections isolated two AsiA mutants (S22F and Q51E) and five sigma(70) mutants (Y571C, Y571H, D570N, L595P, and S604P). All seven suppressor mutants gave partial suppressor phenotypes in vivo as judged by plaque morphology and burst size measurements. The S22F mutant AsiA protein and glutathione S-transferase fusions of the five mutant sigma(70) proteins were purified. All of these mutant proteins allowed normal levels of in vitro transcription when tested with wild-type MotA protein, but they failed to suppress the mutant MotA-pc1 protein in the same assay. The sigma(70) substitutions affected the 4.2 region, which binds the -35 sequence of E. coli promoters. In the presence of E. coli RNA polymerase without T4 proteins, the L595P and S604P substitutions greatly decreased transcription from standard E. coli promoters. This defect could not be explained solely by a disruption in -35 recognition since similar results were obtained with extended -10 promoters. The generalized transcriptional defect of these two mutants correlated with a defect in binding to core RNA polymerase, as judged by immunoprecipitation analysis. The L595P mutant, which was the most defective for in vitro transcription, failed to support E. coli growth.

The MotA transcriptional activator of bacteriophage T4 binds to its specific DNA site as a monomer.

During bacteriophage T4 middle mode gene expression, the MotA transcription factor binds to T4 middle promoters at a -30 mot box consensus sequence to allow activation. Previous binding studies showed that MotA forms multiple gel-shifted complexes with DNA, and structural evidence suggested that MotA dimerizes upon DNA binding. We have shown that a short (13 bp) mot box DNA substrate binds MotA protein but fails to form slower migrating complexes. Therefore, the slower migrating complexes in gel shift assays are caused by DNA-mediated binding events. Competition experiments indicate that the slower migrating complexes are formed by nonspecific binding events, while the first-shifted complex is caused by specific binding to the mot box. Saturation binding experiments revealed that the stoichiometry of MotA binding to DNA is 1:1 in the first-shifted complex, while the slower complexes apparently contain MotA multimers. Gel shift assays using mixtures of MotA and a GST-MotA fusion protein supported the conclusion that the first-shifted complex contains one protein molecule bound to DNA. Furthermore, MotA monomers were cross-linked by glutaraldehyde under conditions where slower complexes exist, but not under conditions that lead to only the first-shifted complex. We conclude that MotA binds specifically to the mot box as a monomer and that additional nonspecific binding events require flanking DNA.

The activation domain of the MotA transcription factor from bacteriophage T4.

Bacteriophage T4 encodes a transcription factor, MotA, that binds to the -30 region of middle-mode promoters and activates transcription by host RNA polymerase. We have solved the structure of the MotA activation domain to 2.2 A by X-ray crystallography, and have also determined its secondary structure by NMR. An area on the surface of the protein has a distinctive patch that is populated with acidic and hydrophobic residues. Mutations within this patch cause a defective T4 growth phenotype, arguing that the patch is important for MotA function. One of the mutant MotA activation domains was purified and analyzed by NMR, and the spectra clearly show that the domain is properly folded. The mutant full-length protein appears to bind DNA normally but is deficient in transcriptional activation. We conclude that the acidic/hydrophobic surface patch is specifically involved in transcriptional activation, which is reminiscent of eukaryotic acidic activation domains.