Correlation between glutathione oxidation and trimerization of heat shock factor 1, an early step in stress induction of the Hsp response

Abstract

The heat shock protein (Hsp) response is induced by heat shock and a large variety of different chemicals. Searching for a common denominator of these different inducers, we and others developed the notion that all inducers may generate abnormally folded, i.e. non-native, proteins, and that such non-native proteins may trigger the Hsp response. Experimentation prompted by this notion resulted, for example, in the demonstration that chemically denatured proteins, introduced in cells by microinjection, can activate the response. Based on the chemical nature of inducers and on results reported from several studies, we hypothesized that inducers of the Hsp response may be generally capable of triggering oxidation of non-protein thiols, particularly glutathione. Such oxidation is known to lead to formation of glutathione-protein mixed disulfides and protein-protein disulfides. Presumably, thiol adduction and cross-linking would affect the structure of proteins involved, resulting in unfolding of a fraction of these proteins, causing heat shock factor (Hsf) activation. To test the feasibility of this hypothesis, thirteen different inducers were selected, and it was shown that all chemical inducers as well as heat shock cause drastic oxidation of glutathione under conditions under which they induce HSE DNA-binding activity. Under the same conditions, ail chemical inducers and heat shock also cause trimerization of Hsf1. For several inducers, it was also shown that they enhance thiol oxidation of proteins. Finally, in vitro experiments support the notion that activation of Hsf1 does not require oxidation of the factor itself or of its coregulators. These results are in complete agreement with the above hypothesis.