Heat Shock Protein (Hsp90) is a "chaperone" protein, essential in eukaryotic cells as it enables many other important regulatory proteins to attain their biologically-active conformation. Hsp90 is also now attracting attention as one of the most promising targets for new anticancer drug development.
We are unravelling how Hsp90 operates in molecular detail in collaborative studies with the Institute of Cancer Research. To this end we have also developed a way of screening the thousands of proteins in a cell for the subset that interact with Hsp90 [1](Figure 1). We are also combining protein structure, biochemical and genetic approaches in order to unravel the structure/function relationships of this important chaperone [2,3].
Figure 1 A. Schematic of Hsp90, a protein that is constitutively dimerised at its C-terminus (C) and transiently dimerised at its N-terminus (N); which uses ATP binding and hydrolysis to assist the final folding step of a target, or "Hsp90 client" protein (green). The E33A mutation in Hsp90 arrests the final ATPase step, causing Hsp90-client complexes to accumulate in vivo.
1 B. Three sample plates from a 16-plate, 384 colony format array of 6000 yeasts, each expressing just one of the 6000 proteins of yeast as a two hybrid "prey" fusion; together with either the control two hybrid "bait" containing Hsp90 (Hsp82-BD) or the E33A mutant version of this "bait" (Hsp82(E33A)-BD). Protein-protein interaction is monitored as growth in the presence of 3 aminotriazole (3-AT). Note that just a few of the 384 colonies on each plate are exhibiting growth dependent on the E33A mutation in the two-hybrid "Bait" which acts to "freezes" Hsp90/client complexes at a late stage of the chaperone cycle (from [2]).
The cancerous state of the most intractable tumours is driven by multiple mutant proteins almost all of which need Hsp90 for their activity. The Hsp90 of cancer cells can be inhibited by highly selective drugs, whereupon these oncogenic proteins are all - in a concerted manner - inactivated and targeted for degradation. In this way, Hsp90 inhibitor drugs can cause combinatorial depletion of many cancer-causing pathways and a modulation of all the hallmark traits of malignancy.
Phase 2 clinical trials have revealed these highly-selective Hsp90 inhibitors display a high selectivity for cancer versus normal cells and a therapeutic activity at doses that are well tolerated in patients. However cancer drug therapy is often eventually compromised by the development of drug resistance. With Cancer Research UK support, are using the yeast model system to investigate whether it is possible for drug resistance to arise by mutation to Hsp90. Such resistance should not arise readily by mutation to those amino acids of Hsp90 that facilitate drug binding as these are required for the essential ATP binding/ATPase steps of the chaperone cycle and tightly conserved. Despite this, we have recently identified a mutation causing resistance to one major class of Hsp90 inhibitor [4](Figure 2)
Figure 2. How a leucine to valine change makes the Hsp90 from a mycoparasitic fungus resistant to the most potent natural product inhibitor of Hsp90, radicicol (RAD). The conformation and binding interactions of RAD within the N-terminal domain of wild type and L34I,I35V mutant Hsp90 are shown superimposed. RAD is in cyan; wild-type structure or in yellow; LI to IV mutant structure. Amino acid residues from the wild type Hsp82 are in cyan while those from the LI to IV mutant protein are in green, except for the residues representing the LI to IV mutations which are in yellow. Hydrogen bonds are shown as red (wild type) and blue (LI to IV mutant) dotted lines; water molecules as red (wild type) and cyan (LI to IV mutant) spheres. Note that the LI to IV mutant structure has 3 extra tightly coordinated water molecules which act to weaken binding of RAD (taken from [4]).
Other recent studies in the laboratory have focussed on the mechanisms whereby yeast ages [5,6] and acquires resistance to weak organic acid preservatives [7].
Selected Publications
[1] Millson, S.H., Truman, A.W., King, V., Prodromou, C., Pearl, L.H. and Piper, P.W. (2005) Two-hybrid screening the yeast proteome for Hsp90 chaperone interactors uncovers a novel Hsp90 requirement in the activity of a stress-activated MAP kinase, Slt2p(Mpk1p). Eukaryot. Cell 4, 849-860.
[2] Ali, M.M., Roe, S.M., Vaughan, C.K., Meyer, P., Panaretou, B., Piper, P.W., Prodromou, C. and Pearl, L.H. (2006) Crystal structure of an Hsp90-nucleotide-p23/Sba1 closed chaperone complex. Nature 440, 1013-7.
[3] Truman, A.W., Millson, S.H., Nuttall. J.M., King, V., Mollapour, M., Prodromou, C., Pearl, L.H. and Piper, P.W. (2006) Expressed in yeast, human ERK5 is a client of the Hsp90 chaperone and complements loss of the Slt2(Mpk1)p cell integrity stress-activated protein kinase. Eukaryot. Cell 5, 1914-24.
[4] Prodromou, C, Nuttall, J.M., Millson, S.H., Roe, S.M., Sim, T.S., Tan, D., Workman, P., Pearl, L.H. and Piper, P.W. (2009) Structural basis of the radicicol resistance displayed by a fungal Hsp90. ACS Chem. Biol. 17, 289-297.
[5] Harris, N., Costa, V., MacLean, M., Mollapour, M., Moradas-Ferreira, P. and Piper, P.W. (2003) Mn-Sod overexpression extends the yeast chronological (G0) life span, but acts independently of Sir2p histone deacetylase to shorten the replicative life span of dividing cells. Free Rad. Biol. Med. 34, 1594-1606.
6] Harris, N., Bachler, M., Costa, V., Mollapour, M., Moradas-Ferreira, P. and Piper, P.W. (2005) Overexpressed Sod1p acts either to reduce or to increase the lifespans and stress resistance of yeast, depending on whether it is Cu2+-deficient or an active Cu,Zn-superoxide dismutase. Aging Cell 4, 41-52.
[7] Mollapour, M. and Piper, P.W. (2007) Hog1 MAP kinase phosphorylation targets the yeast Fps1 aquaglyceroporin for endocytosis, thereby rendering cells resistant to acetic acid. Mol. Cell. Biol. 27, 6446-6456.
