Sequence-specific dynamics of DNA response elements and their flanking sites regulate the recognition by AP-1 transcription factors

Abstract

Activator proteins 1 (AP-1) comprise one of the largest families of eukaryotic basic leucine zipper transcription factors. Despite advances in the characterization of AP-1 DNA-binding sites, our ability to predict new binding sites and explain how the proteins achieve different gene expression levels remains limited. Here we address the role of sequence-specific DNA dynamics for stability and specific binding of AP-1 factors, using microseconds long molecular dynamics simulations. As a model system, we employ yeast AP-1 factor Yap1 binding to three different response elements from two genetic environments. Our data show that Yap1 actively exploits the sequence-specific plasticity of DNA within the response element to form stable protein-DNA complexes. The stability also depends on the four to six flanking nucleotides, adjacent to the response elements. The flanking sequences modulate the conformational adaptability of the response element, making it more shape-efficient to form specific contacts with the protein. Bioinformatics analysis of differential expression of the studied genes supports our conclusions: the stability of Yap1-DNA complexes, modulated by the flanking environment, influences the gene expression levels. Our results provide new insights into mechanisms of protein-DNA recognition and the biological regulation of gene expression levels in eukaryotes.

The Biologist’s Guide to the Glucocorticoid Receptor’s Structure

“…The head-to-head arrangement was also observed in the recent hGR multidomain crystal structure [122]. The recognition helix of the first monomer involving rGR Lys461, Val462 and Arg466 (see Figure 4 for cross-species sequence alignment) makes specific major groove contacts [54,113]. This is similar for the second monomer but with a stronger Lys461 contact and lacking the Val462 contact [54,113].…”

“…The recognition helix of the first monomer involving rGR Lys461, Val462 and Arg466 (see Figure 4 for cross-species sequence alignment) makes specific major groove contacts [54,113]. This is similar for the second monomer but with a stronger Lys461 contact and lacking the Val462 contact [54,113]. Residues 509-515 form the C-terminal helix 4 (H4, residues hGR 489-495, mGR 506-512) which lies across the minor groove [54].…”

“…A more recent study found that mutations in the GREM had no effects on GR binding and proposed instead that GR binds in a structure-specific rather than sequence-specific fashion [112]. This was supported by in vitro and in silico observations showing that a loss of the Gas5 RNA terminal loop resulted in a drastic decrease in GR binding [112,113]. A 4 bp stem-loop proved sufficient for binding, and increasing the loop length or altering its sequence had no significant impact on binding [112].…”

“…The GR DBD binds the consensus GRE and Gas5 with comparable affinities in the 67-125 nM range [100,112]. Although GR binds GREs as a dimer with slightly positive cooperativity (Hill coefficient of 1.3-1.4), this is not the case for GR-RNA binding (Hill coefficient of 0,93) [100,112,113]. NMR spectra revealed that the GR dimerization loop was not affected by binding to a 33-nucleotide Gas5 RNA hairpin.…”

“…For the latter, GR binds a cryptic response element (AATTY, Y = pyrimidine base) between the binding footprints of NF-κB subunits within κBREs [102]. In addition, the GR DBD is capable of binding biological and synthetic RNAs of which Gas5 is the most thoroughly researched [109][110][111][112][113]. The final mechanism, referred to as tethering, does not involve direct DNA contacts but is mediated by various protein-protein interactions [114][115][116][117][118].…”