Interactions between DNA-bound trimers of the yeast heat shock factor.

Abstract

The heat shock transcription factor (HSF) is a trimer that binds to DNA containing inverted repeats of the sequence nGAAn. HSF can bind DNA with the sequence nGAAnnTTCn or with the sequence nTTCnnGAAn, with little preference for either sequence over the other. However, (nGAAnnTTCn)2 is considerably less active as a heat shock response element (HSE) than is (nTTCnnGAAn)2. The electrophoretic mobilities of DNA-protein complexes and chemical cross-linking between protein monomers indicate that the sequence (nGAAnnTTCn)2 is capable of binding a single HSF trimer. In contrast, the sequence with higher biological activity, (nTTCnnGAAn)2, is capable of binding two trimers. Thus, the ability of four-nGAAn-element HSEs to bind one or two trimers depends on the permutation with which the elements are presented. A survey of naturally occurring HSEs shows the sequence (nTTCnnGAAn)2 to be the more prevalent. We suggest that the greater ability of one permutation over the other to bind two HSF trimers accounts for the initial identification of the naturall occurring heat shock consensus sequence as a region of dyad symmetry.The eukaryotic heat shock transcription factor (HSF) is the main transcription factor responsible for expression of heat shock protein (hsp) genes during periods of stress. The activity of HSF is exquisitely sensitive to the presence of unfolded, denatured, or aberrant proteins, which appear to act as an intracellular signal to induce hsp gene transcription. It is likely that this signal acts on the posttranslational regulation of HSF activity, at least in part, through one of the heat shock proteins, hsp7O (reviewed in reference 16). Association of hsp70 with HSF appears to be necessary to sequester HSF in a form that is unable to activate transcription effectively. Any stress-induced increase in the concentration of aberrant proteins is thought to result in the titration of hsp70 and the accumulation of HSF in an active form.In vertebrate and Drosophila cells, the inactive form of HSF is a non-DNA-binding monomer (5,20,29). Heat shock causes the assembly of HSF monomers into trimers, which are then competent to bind DNA (5,20,29). In the yeast Saccharomyces cerevisiae, HSF is constitutively competent to bind DNA (11) and is trimeric even before heat shock (25).The observation that HSF is trimeric raises some interesting questions with regard to its DNA interaction. The initial analyses of natural heat shock elements (HSEs) resulted in the description of a consensus sequence, CTnGAAnnTTC nAG (17,18,27). This consensus HSE is palindromic, reminiscent of binding sites for dimeric proteins. It has more recently been suggested that functional HSEs are more properly described as inverted repeats of the sequence module nGAAn (2, 31). Within an array of nGAAn repeats, it is likely that each nGAAn element is contacted by a single HSF monomer; Perisic et al. (19) observed that repeats of two to three nGAAn elements are bound by a single trimer of Drosophila HSF, repeats of four to six elements are bound by two tri...

Complex Regulation of the Yeast Heat Shock Transcription Factor

“…We will first consider the gel mobility shift. Yeast HSF, when provided with 32 P-labeled DNA containing the sequence GAAnnTTCnnGAAnnTTC, forms a DNA-HSF complex containing a single HSF trimer (Bonner et al, 1994). As shown in Figure 1, lane 1 (on left), this is the sole complex that is visible when an HSF-containing cell extract is incubated with the probe in the presence of 1 mM ATP.…”

“…To examine the activity of HSF in the absence of additional, potentially interacting transcription factors, we prepared a ␤-galactosidase reporter gene that has, as its sole regulatory element, a synthetic heat shock element (HSE-lacZ [Bonner et al, , 1994). The lacZ vector pLG⌬178, originally built by Guarente and Mason (1983), has been used by many laboratories and shown to have essentially no activity without the insertion of an appropriate UAS; to the best of our knowledge, this reporter gene responds only to HSF.…”

“…For coimmunoprecipitation, extract (600 g of protein) prepared as described previously (Bonner et al, 1994) was diluted into 1 ml of XLB buffer (175 mM NaCl, 15 mM Tris, pH 8.0, 1 mM EDTA, 1.5 mM MgCl 2 ). Cross-linking was initiated by the addition of 15 l of 10 mM dithiobis[succinimidylproprionate] (DSSP; Pierce Chemical, Rockford, IL).…”

“…Extracts were prepared from YJB371 as described previously (Bonner et al, 1994). For each binding assay, extract (35 g of protein) was suspended in 15 l of binding reaction (20 mM N-2-hydroxyethylpiperazine-NЈ-2-ethanesulfonic acid [HEPES], pH 7.9, 1 mM EDTA, 60 mM KCl, 12% glycerol, 1 mM dithiothreitol, 1 g/l bovine serum albumin [BSA], 0.1 g/ml sonicated E. coli DNA).…”

A Novel Non-conventional Heat Shock Element Regulates Expression of MDJ1 Encoding a DnaJ Homolog in Saccharomyces cerevisiae

“…In fission yeast, D. melanogaster, and mammals, the binding activity of HSFs to HSEs is dramatically stimulated by heat shock as HSF monomers are converted to trimers (9 -11). In contrast, in the budding yeast, HSF constitutively binds to HSE, but the binding activity seems to increase after heat shock-induced hyperphosphorylation (7,(12)(13)(14)(15)(16).The HSE is composed of several contiguous inverted repeats of the 5-base pair sequence nGAAn (where n is any nucleotide) (1,17,18). The number of the pentameric units in HSE varies, but at least three units are thought to be the minimum required for heat regulation in vivo.…”