Restricting direct interaction of CDC37 with HSP90 does not compromise chaperoning of client proteins.

The HSP90 molecular chaperone plays a key role in the maturation, stability and activation of its clients, including many oncogenic proteins. Kinases are a substantial and important subset of clients requiring the key cochaperone CDC37. We sought an improved understanding of protein kinase chaperoning by CDC37 in cancer cells. CDC37 overexpression in human colon cancer cells increased CDK4 protein levels, which was negated upon CDC37 knockdown. Overexpressing CDC37 increased CDK4 protein half-life and enhanced binding of HSP90 to CDK4, consistent with CDC37 promoting kinase loading onto chaperone complexes. Against expectation, expression of C-terminus-truncated CDC37 (ΔC-CDC37) that lacks HSP90 binding capacity did not affect kinase client expression or activity; moreover, as with wild-type CDC37 overexpression, it augmented CDK4-HSP90 complex formation. However, although truncation blocked binding to HSP90 in cells, ΔC-CDC37 also showed diminished client protein binding and was relatively unstable. CDC37 mutants with single and double point mutations at residues M164 and L205 showed greatly reduced binding to HSP90, but retained association with client kinases. Surprisingly, these mutants phenocopied wild-type CDC37 overexpression by increasing CDK4-HSP90 association and CDK4 protein levels in cells. Furthermore, expression of the mutants was sufficient to protect kinase clients CDK4, CDK6, CRAF and ERBB2 from depletion induced by silencing endogenous CDC37, indicating that CDC37's client stabilising function cannot be inactivated by substantially reducing its direct interaction with HSP90. However, CDC37 could not compensate for loss of HSP90 function, showing that CDC37 and HSP90 have their own distinct and non-redundant roles in maintaining kinase clients. Our data substantiate the important function of CDC37 in chaperoning protein kinases. Furthermore, we demonstrate that CDC37 can stabilise kinase clients by a mechanism that is not dependent on a substantial direct interaction between CDC37 and HSP90, but nevertheless requires HSP90 activity. These results have significant implications for therapeutic targeting of CDC37.

Malignant progression of a pleomorphic xanthoastrocytoma in a child.

Pleomorphic xanthoastrocytoma (PXA) is a recently recognized rare cerebral neoplasm that predominantly affects young patients. We report on the case of a 3-year-old boy who presented with a 2-week history of headaches and seizures. Radiological investigation revealed a lesion in the right parietal-occipital lobe. The lesion was excised and histology disclosed the presence of a PXA with anaplastic features. 1 year later follow-up magnetic resonance imaging (MRI) revealed tumor relapse. An MRI of the spine was also performed and demonstrated leptomeningeal dissemination. The patient underwent a second operation. Histology revealed that the presence of a malignant PXA with anaplastic features. The patient received radiotherapy and 9 months later on follow-up MRI a new tumor recurrence was noted. A third craniotomy was performed and the tumor removed. Histological examination revealed dedifferentiation to glioblastoma multiforme. The patient was referred to the oncology department and received chemotherapy with temozolamide. 8 months later the patient was stable without tumor recurrence. PXAs require close follow-up because of their unpredictable biological behaviour.

The ATPase cycle of Hsp90 drives a molecular 'clamp' via transient dimerization of the N-terminal domains.

How the ATPase activity of Heat shock protein 90 (Hsp90) is coupled to client protein activation remains obscure. Using truncation and missense mutants of Hsp90, we analysed the structural implications of its ATPase cycle. C-terminal truncation mutants lacking inherent dimerization displayed reduced ATPase activity, but dimerized in the presence of 5'-adenylamido-diphosphate (AMP-PNP), and AMP-PNP- promoted association of N-termini in intact Hsp90 dimers was demonstrated. Recruitment of p23/Sba1 to C-terminal truncation mutants also required AMP-PNP-dependent dimerization. The temperature- sensitive (ts) mutant T101I had normal ATP affinity but reduced ATPase activity and AMP-PNP-dependent N-terminal association, whereas the ts mutant T22I displayed enhanced ATPase activity and AMP-PNP-dependent N-terminal dimerization, indicating a close correlation between these properties. The locations of these residues suggest that the conformation of the 'lid' segment (residues 100-121) couples ATP binding to N-terminal association. Consistent with this, a mutation designed to favour 'lid' closure (A107N) substantially enhanced ATPase activity and N-terminal dimerization. These data show that Hsp90 has a molecular 'clamp' mechanism, similar to DNA gyrase and MutL, whose opening and closing by transient N-terminal dimerization are directly coupled to the ATPase cycle.

Structure and in vivo function of Hsp90.

Until recently, Hsp90 was one of the least well understood of the molecular chaperones, but considerable progress is now being made in unravelling its biochemistry. Hsp90 has now been shown to possess an inherent ATPase that is essential for the activation of authentic 'client' proteins in vivo and in vitro. The molecular detail of Hsp90's interactions with co-chaperones is also becoming clearer and the identification of key roles in assembling regulatory and signalling pathways has made it a target for anticancer drug development. Despite this, a clear understanding of how Hsp90 contributes to the folding and/or activation of its client proteins remains some way off.

Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin.

The cellular activity of several regulatory and signal transduction proteins, which depend on the Hsp90 molecular chaperone for folding, is markedly decreased by geldanamycin and by radicicol (monorden). We now show that these unrelated compounds both bind to the N-terminal ATP/ADP-binding domain of Hsp90, with radicicol displaying nanomolar affinity, and both inhibit the inherent ATPase activity of Hsp90 which is essential for its function in vivo. Crystal structure determinations of Hsp90 N-terminal domain complexes with geldanamycin and radicicol identify key aspects of their nucleotide mimicry and suggest a rational basis for the design of novel antichaperone drugs.

Regulation of Hsp90 ATPase activity by tetratricopeptide repeat (TPR)-domain co-chaperones.

The in vivo function of the heat shock protein 90 (Hsp90) molecular chaperone is dependent on the binding and hydrolysis of ATP, and on interactions with a variety of co-chaperones containing tetratricopeptide repeat (TPR) domains. We have now analysed the interaction of the yeast TPR-domain co-chaperones Sti1 and Cpr6 with yeast Hsp90 by isothermal titration calorimetry, circular dichroism spectroscopy and analytical ultracentrifugation, and determined the effect of their binding on the inherent ATPase activity of Hsp90. Sti1 and Cpr6 both bind with sub-micromolar affinity, with Sti1 binding accompanied by a large conformational change. Two co-chaperone molecules bind per Hsp90 dimer, and Sti1 itself is found to be a dimer in free solution. The inherent ATPase activity of Hsp90 is completely inhibited by binding of Sti1, but is not affected by Cpr6, although Cpr6 can reactivate the ATPase activity by displacing Sti1 from Hsp90. Bound Sti1 makes direct contact with, and blocks access to the ATP-binding site in the N-terminal domain of Hsp90. These results reveal an important role for TPR-domain co-chaperones as regulators of the ATPase activity of Hsp90, showing that the ATP-dependent step in Hsp90-mediated protein folding occurs after the binding of the folding client protein, and suggesting that ATP hydrolysis triggers client-protein release.

ATP binding and hydrolysis are essential to the function of the Hsp90 molecular chaperone in vivo.

Hsp90 is an abundant molecular chaperone essential to the establishment of many cellular regulation and signal transduction systems, but remains one of the least well described chaperones. The biochemical mechanism of protein folding by Hsp90 is poorly understood, and the direct involvement of ATP has been particularly contentious. Here we demonstrate in vitro an inherent ATPase activity in both yeast Hsp90 and the Escherichia coli homologue HtpG, which is sensitive to inhibition by the Hsp90-specific antibiotic geldanamycin. Mutations of residues implicated in ATP binding and hydrolysis by structural studies abolish this ATPase activity in vitro and disrupt Hsp90 function in vivo. These results show that Hsp90 is directly ATP dependent in vivo, and suggest an ATP-coupled chaperone cycle for Hsp90-mediated protein folding.

Identification and structural characterization of the ATP/ADP-binding site in the Hsp90 molecular chaperone.

Hsp90 molecular chaperones in eukaryotic cells play essential roles in the folding and activation of a range of client proteins involved in cell cycle regulation, steroid hormone responsiveness, and signal transduction. The biochemical mechanism of Hsp90 is poorly understood, and the involvement of ATP in particular is controversial. Crystal structures of complexes between the N-terminal domain of the yeast Hsp90 chaperone and ADP/ATP unambiguously identify a specific adenine nucleotide binding site homologous to the ATP-binding site of DNA gyrase B. This site is the same as that identified for the antitumor agent geldanamycin, suggesting that geldanamycin acts by blocking the binding of nucleotides to Hsp90 and not the binding of incompletely folded client polypeptides as previously suggested. These results finally resolve the question of the direct involvement of ATP in Hsp90 function.

A molecular clamp in the crystal structure of the N-terminal domain of the yeast Hsp90 chaperone.

Hsp90 is a highly specific chaperone for many signal transduction proteins, including steroid hormone receptors and a broad range of protein kinases. The crystal structure of the N-terminal domain of the yeast Hsp90 reveals a dimeric structure based on a highly twisted sixteen stranded beta-sheet, whose topology suggests a possible 30-domain-swapped structure for the intact Hsp90 dimer. The opposing faces of the beta-sheets in the dimer define a potential peptide-binding cleft, suggesting that the N-domain may serve as a molecular 'clamp' in the binding of ligand proteins to Hsp90.

Expression and crystallization of the yeast Hsp82 chaperone, and preliminary X-ray diffraction studies of the amino-terminal domain.

Expression of the Saccharomyces cerevisiae Hsp82 chaperone in a pep4-3- and hsc82-deficient strain of S. cerevisiae yielded over 25% of the total cell protein as intact Hsp82. Similarly, the amino-terminal domain (residues 1-220) of Hsp82 was expressed to 18% of the total cell protein. Crystals of the intact Hsp82 were readily obtained. The crystals were very fragile, suggesting a high solvent content, and diffracted to approximately 8 A. Tetragonal bipyrimidal crystals of the amino-terminal domain of Hsp82 were readily obtained under a variety of different conditions. The crystals have primitive tetragonal space group (P422, P4(1)22, or its enantiomorph P4(3)22) with unit cell dimensions of a = 75.1 A and c = 111.3 A, contain 60% by volume solvent, and diffract to 2.5 A resoltuion. Addition of 25% glycerol to the mother liquor gave rise to large rod-shaped crystals. The crystals diffract to 2.8 A resolution, have an orthorhombic space group (P222(1), P2(1)2(1)2, or P2(1)2(1)2(1)) with cell dimensions of a = 45.2 A, b = 115.4 A, and c = 116.9 A, and a solvent content of 58% by volume.

Synthesis of a modified gene encoding human ornithine transcarbamylase for expression in mammalian mitochondrial and universal translation systems: a novel approach towards correction of a genetic defect.

The mitochondrial (MT) genome is a potential means of gene delivery to human cells for therapeutic expression. As a first step towards this, we have synthesized a gene coding for mature human ornithine transcarbamylase (OTC) by recursive PCR using 18 oligodeoxyribonucleotides, each 70-80 nucleotides in length, using codons which should allow translation in accordance with both mammalian mt and universal codon usage. Flanking mt DNA sequences were incorporated which are designed to facilitate site-specific cloning into the mt genome. Expression of this human gene in Escherichia coli leads to an immunoreactive OTC product of the correct size and N-terminal amino-acid sequence, but which forms inclusion bodies and lacks enzymatic activity.

The aconitase of Escherichia coli. Nucleotide sequence of the aconitase gene and amino acid sequence similarity with mitochondrial aconitases, the iron-responsive-element-binding protein and isopropylmalate isomerases.

The nucleotide sequence of the aconitase gene (acn) of Escherichia coli was determined and used to deduce the primary structure of the enzyme. The coding region comprises 2670 bp (890 codons excluding the start and stop codons) which define a product having a relative molecular mass of 97,513 and an N-terminal amino acid sequence consistent with those determined previously for the purified enzyme. The acn gene is flanked by the cysB gene and a putative riboflavin biosynthesis gene resembling the ribA gene of Bacillus subtilis. The 1004-bp cysB--acn intergenic region contains several potential promoter and regulatory sequences. The amino acid sequence of the E. coli aconitase is similar to the mitochondrial aconitases (27-29% identity) and the isopropylmalate isomerases (20-21% identity) but it is most similar to the human iron-responsive-element-binding protein (53% identity). The three cysteine residues involved in ligand binding to the [4Fe-4S] centre are conserved in all of these proteins. Of the remaining 17 active-site residues assigned for porcine aconitase, 16 are conserved in both the bacterial aconitase and the iron-responsive-element-binding protein and 14 in the isopropylmalate isomerases. It is concluded that the bacterial and mitochondrial aconitases, the isopropylmalate isomerases and the iron-responsive-element-binding protein form a family of structurally related proteins, which does not include the Fe-S-containing fumarases. These relationships raise the possibility that the iron-responsive-element-binding protein may be a cytoplasmic aconitase and that the E. coli aconitase may have an iron-responsive regulatory function.

The aconitase of Escherichia coli: purification of the enzyme and molecular cloning and map location of the gene (acn).

The aconitase of Escherichia coli was purified to homogeneity, albeit in low yield (0.6%). It was shown to be a monomeric protein of Mr 95,000 or 97,500 by gel filtration and SDS-PAGE analysis, respectively. The N-terminal amino acid sequence resembled that of the Bacillus subtilis enzyme (citB product), but the similarity at the DNA level was insufficient to allow detection of the E. coli acn gene using a 456 bp citB probe. Phages containing the acn gene were isolated from a lambda-E. coli gene bank by immunoscreening with an antiserum raised against purified bacterial enzyme. The acn gene was located at 28 min (1350 kb) in the physical map of the E. coli chromosome by probing Southern blots with a fragment of the gene. Attempts to locate the gene using the same procedure with oligonucleotide probes encoding segments of the N-terminal amino acid sequence were complicated by the lack of probe specificity and an inaccuracy in the physical map of Kohara et al. (Cell 50, 495-508, 1987). Aconitase specific activity was amplified some 20-200-fold in cultures transformed with pGS447, a derivative of pUC119 containing the acn gene, and an apparent four-fold activation-deactivation of the phagemid-encoded enzyme was observed in late exponential phase. The aconitase antiserum cross-reacted with both the porcine and Salmonella typhimurium (Mr 120,000) enzymes.

Cloning of the HIS3 gene of Yarrowia lipolytica.

The HIS3 gene of the yeast Yarrowia lipolytica has been cloned from a genomic library by complementation of the his3 mutation of Saccharomyces cerevisiae. The gene was subsequently subcloned in Escherichia coli and characterized by restriction enzyme mapping.