The Hsp90 of Candida albicans can confer Hsp90 functions in Saccharomyces cerevisiae: a potential model for the processes that generate immunogenic fragments of this molecular chaperone in C. albicans infections

Barry Panaretou1, Kirsty Sinclair1, Chrisostomos Prodromou1, Jasvinder Johal1, Laurence Pearl1 and Peter W. Piper1

Department of Biochemistry and Molecular Biology, University College London, London WC1E 6BT, UK1

Author for correspondence: Peter W. Piper. Tel: +44 207 504 2212. Fax: +44 207 380 7193. e-mail: piper{at}bsm.bioc.ucl.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
During infections with a number of important eukaryotic pathogens the Hsp90 molecular chaperone of the pathogen is recognized as an immunodominant antigen by the host immune system. Yeast molecular genetics should allow study of the extent of sequence variation within conserved immunodominant epitopes on pathogen Hsp90s that is compatible with essential Hsp90 functions, as well as the processes that generate antigenic subfragments of these Hsp90s. The Hsp90 of the fungal pathogen Candida albicans was shown in this study to provide both essential and nonessential (pheromone signalling and mammalian steroid receptor activation) Hsp90 functions in Saccharomyces cerevisiae cells. Much of the C. albicans Hsp90 expressed in respiratory S. cerevisiae cells was shown to undergo a partial degradation in vivo, a degradation that closely resembles that of the native Hsp82 (one isoform of the homologous Hsp90) in S. cerevisiae. Allowing for the differences in the length of the charged linker region between the N- and C-terminal domains of C. albicans Hsp90 and S. cerevisiae Hsp82, these two proteins expressed in S. cerevisiae appear to give the same major degradation products. These Hsp90 fragments are similar to the products of incomplete Hsp90 degradation found in C. albicans cultures.

Keywords: Candida albicans, Saccharomyces cerevisiae, Hsp90, partial protein degradation, candidiasis

Abbreviations: 5-FOA, 5-fluoroorotic acid; GR, glucocorticoid receptor


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Heat-shock proteins (hsps) are major targets for the immune response to several bacterial, fungal and parasitic infections (Young, 1992 ). Several microbial hsps have been associated with protective immunity (Bonnefoy et al., 1994 ; Gomez et al. , 1995 ; Lowrie et al., 1994 ; Matthews et al., 1991 , 1995 ), the eukaryotic molecular chaperone Hsp90 being important in this respect. Surface-expressed parasite Hsp90 serves as an antigen in Chagas’ disease (Dragon et al., 1987 ), ascariasis (Kumari et al., 1994 ), leishmaniasis (Salotra et al., 1995 ; Skeiky et al., 1995 ; Streit et al., 1996 ) and infections with Schistosoma mansoni (Johnson et al., 1989 ). The successful treatment of children with malaria by the passive transfer of immunoglobin from adults living where malaria is endemic may also be due to antibodies against the Hsp90 of Plasmodium falciparium (Bonnefoy et al., 1994 ). Immunodominant antigens of Candida albicans are present on a 47 kDa subfragment of candidal Hsp90 that circulates in the serum of patients with candidiasis, a fragment that comprises the C-terminal domain of this abundant molecular chaperone (Matthews & Burnie, 1989 ; Matthews et al., 1987 ). Both monoclonal and human recombinant antibodies to a highly conserved epitope on this fragment (LKVIRK) were protective in murine models of candidiasis (Matthews et al., 1991 , 1995 ). These antibodies are also always present in humans who have survived systemic candidiasis, whereas their absence is invariably associated with fatality (Matthews, 1994 ; Matthews et al., 1987 ). So highly conserved is the LKVIRK epitope on eukaryotic Hsp90s that antibodies against this epitope may cross-protect against a number of different microbial pathogens (Matthews, 1994 ). In addition, the essential functions of Hsp90 may not allow large mutational changes to this epitope, whereby microbes could become resistant to LKVIRK-targeting antibodies. There is therefore considerable interest in the potential diagnostic and therapeutic uses of Hsp90-targeting antibodies (Matthews & Burnie, 1996 ; Matthews, 1994 ).

The cellular functions of the abundant Hsp90 molecular chaperone have been most clearly identified in mammalian cells, where this protein is essential for the final activation step of several key proteins of signal transduction (Gething & Sambrook, 1992 ; Pratt & Dittmar, 1998 ; Rutherford & Lindquist, 1998 ; Toft, 1998 ). Hsp90 action in the ligand-dependent activation of mammalian steroid receptor proteins has been particularly well studied (Caplan, 1997 ; Nathan & Lindquist, 1995 ; Pratt & Dittmar, 1998 ). In contrast, although Hsp90 is an important antigen in candidal infections, virtually nothing is known about Hsp90 functions in Candida. Also there is no evidence to suggest how the 40/47 kDa subfragments of C. albicans Hsp90 found circulating in the serum of patients with candidiasis are generated during the pathology of this disease. We show here that the expression of the coding region of the C. albicans HSP90 gene in Saccharomyces cerevisiae can provide Hsp90 functions in the latter yeast. S. cerevisiae can therefore be used to show whether changes to conserved epitopes on C. albicans Hsp90 are compatible with essential Hsp90 functions. We also show that cells of both C. albicans and S. cerevisiae generate Hsp90 fragments, the incomplete degradation of C. albicans Hsp90 expressed in S. cerevisiae resembling partial degradation of this same protein in C. albicans cultures. A similar partial breakdown of the endogenous Hsp90 is also seen in S. cerevisiae cells adapting to respiratory growth. S. cerevisiae molecular genetics can therefore probably be applied to study the events of partial in vivo Hsp90 degradation, the process that may generate the antigenic fragments of this molecular chaperone that appear during candidal infections.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Plasmid constructions.
A vector (pCA90) for the expression of C. albicans Hsp90 in S. cerevisiae was constructed by sequential cloning of three DNA fragments into the S. cerevisiae Escherichia coli shuttle vector YCplac111 (Gietz & Sugino, 1988 ). First, a 414 bp ADH1 terminator Pst I/HindIII fragment from pGBT9 (Clontech) was ligated into PstI/HindIII-cleaved YCplac111 (Gietz & Sugino, 1988 ). Next, the HSC82 promoter was inserted as a BamHI/SalI-cleaved PCR-generated fragment, to give vector pHSCprom. This promoter fragment had previously been amplified from pHSC82 (Panaretou et al., 1998 ) using the primers C GAGCTCGGTACCCGG and GTTTCACCGTCGACTTATTTCAGAATGATTC ( BamHI and SalI sites, respectively, underlined). Finally, the complete C. albicans HSP90 gene coding region was inserted into pHSCprom as a SalI/PstI- cleaved PCR product, producing pCA90. This had been PCR-amplified from C. albicans strain 3153A genomic DNA using PCR primers CCATACAAATGTCGACTTCATTATGGCTGACGC and CTATGTCTAATAGCGGCCGC CTGCAGGATTTTCCTTTTAGC (SalI and PstI sites, respectively, underlined). The S. cerevisiae HSP82 gene coding region was also placed in pHSCprom as a SalI/ PstI-cleaved PCR product, thereby producing a vector (pHSP82b) for expressing Hsp82 from the HSC82 promoter. This gene was PCR-amplified from pHSP82 (Panaretou et al., 1998 ) using PCR primers ACGCAAAGATGTCGACATGGCTAGTGAAAC and ATGATGTTCTACCTGCAGCTAATCTACCTCTTC (SalI and Pst I sites, respectively, underlined).

Strains.
C. albicans Berkhout strain 3153A, from the London Mycology Reference Laboratory Collection, was kindly provided by Alistair Brown, University of Aberdeen, UK. The S. cerevisiae strains used were: PP30(pHSC82) [MATa trp1-289 leu2-3,112 his3-200 ura3-52 ade2-101 oc lys2-801am hsc82:: KANMX4 hsp82::KANMX4(pHSC82)] (Panaretou et al., 1998 ); E929-6C-20 (MAT{alpha} ste11- {Delta}6 can1 leu2-3,112 trp1-{Delta}1 ura3-52) (Louvion et al., 1998 ), SUB61 and SUB63 (Finley et al., 1987 ). PP30(pHSC82) was transformed to leucine prototrophy with the LEU2 vectors pHSP82b or pCA90, then cured of its pHSC82 URA3 vector by growth on 5-fluoroorotic acid (FOA; Melford Laboratories) as described by Panaretou et al. (1998) . This resulted in strains that expressed only the Hsp82 of S. cerevisiae [PP30(pHSP82b)] or only the Hsp90 of C. albicans [PP30(pCA90)].

Steroid hormone receptor assays.
The ligand response of glucocorticoid receptor (GR) was determined by measuring ß-galactosidase activity induced by dexamethasone. Yeast cells were grown overnight to exponential phase at 30 °C in selective medium containing tryptophan and adenine. To ensure that the ligand was added to cultures of equivalent volume and cell density, culture volumes were diluted with fresh medium to 100 ml (1·55x106 cells ml-1). Dexamethasone (10 µM final concentration, from a 20 mM stock in ethanol) was then added to half of the culture; an equivalent amount of ethanol (0·05%) being added to the control. Cultures were incubated for a further 2 h at 30 °C, before cells were collected by centrifugation (3000 g, 5 min). Cell pellets were resuspended in 500 ml 2xZ buffer (60 mM Na2 HPO4 . 7H2O, 60 mM NaH 2PO4 . H2O, 5 mM KCl, 0·5 mM MgSO4, 0·05% ß- mercaptoethanol, pH 7). ß-Galactosidase activity was determined as described by Bohen (1998) .

Antibody preparation and analysis of Hsp90s by Western blotting.
As part of our investigations into the structure of yeast Hsp82 we have expressed several regions of this protein in E. coli so that the fragments could then be used for crystallization trials (Prodromou et al., 1997 ). Two of these fragments (corresponding to regions 1–221 and 300–704 of S. cerevisiae Hsp82) were also employed as antigens in raising polyclonal rabbit antisera. These antisera were shown to be Hsp90- specific (proteins in yeast total cell extracts showing no appreciable cross-reaction to the pre-immune sera). Also, investigations of their cross-reactivity to our different regions of Hsp82 expressed in E. coli revealed they are also domain-specific, in that the antiserum raised against the 1–221 region of Hsp82 recognized an epitope at the extreme N-terminus of full-length Hsp82 and the antiserum raised against the 300–704 region recognized epitopes exclusively within the C-terminal domain of full-length Hsp82 (Sinclair, 1998 ). Preparation of total yeast cell protein extracts was as described earlier (Cheng et al., 1992 ; Panaretou et al. , 1998 ). For the analysis of Hsp90s by Western blotting, these protein extracts were gel-fractionated, immobilized on nitrocellulose and probed at room temperature with the above Hsp82 N- domain-specific or C-domain-specific antisera. Antibody binding was visualized by incubating the blot with peroxidase-conjugated anti- rabbit IgG and then use of the Amersham enhanced chemiluminescence (ECL) kit according to the manufacturer’s instructions.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
C. albicans Hsp90 can confer essential Hsp90 functions in S. cerevisiae
C. albicans appears to have just one gene for Hsp90 (Swoboda et al., 1996 ), whereas S. cerevisiae possesses two, HSC82 and HSP82. In the latter yeast, HSC82 is constitutively expressed at high levels and is only slightly heat-inducible, whereas HSP82 is normally expressed at tenfold lower levels but is strongly induced by heat shock (Borkovich et al., 1989 ). Hsp90 is thought to be an essential protein in eukaryotic organisms, its essential nature having been proven in Drosophila (Rutherford & Lindquist, 1998 ) and S. cerevisiae (a {Delta}hsc82 {Delta}hsp82 S. cerevisiae strain being inviable; Borkovich et al., 1989 ). We recently constructed a S. cerevisiae strain [PP30(pHSC82)] that has the chromosomal coding regions of both HSC82 and HSP82 totally deleted, but which is viable since a copy of the S. cerevisiae HSC82 gene on a URA3 plasmid provides the essential Hsp90 functions (Panaretou et al., 1998 ). By introducing the genes for mutant or heterologous Hsp90s into this strain on a LEU2 vector, then selecting for loss of the URA3 vector on FOA plates, it is possible to determine whether introduced Hsp90 genes can confer the essential Hsp90 functions in S. cerevisiae (Panaretou et al., 1998 ). Absence of growth on FOA shows that the introduced gene is nonfunctional, whereas growth on FOA reveals that it is functional.

The C. albicans HSP90 ORF (Swoboda et al., 1996 ) is devoid of CTG codons, which code for serine in C. albicans but leucine in most other organisms (Santos et al. , 1993 ). It also encodes a product with 84 and 78% amino acid sequence identity, respectively, to the Hsc82 and Hsp82 proteins of S. cerevisiae. We therefore thought it probable that expression of the C. albicans HSP90 ORF in S. cerevisiae would generate functional Hsp90 in the latter yeast. To show that this was indeed the case, we constructed LEU2 vectors (pCA90 and pHSP82b) in which the coding regions of either C. albicans HSP90 or S. cerevisiae HSP82 are expressed from the S. cerevisiae HSC82 promoter (see Methods). These vectors were then used to transform S. cerevisiae PP30(pHSC82) to leucine prototrophy. When subsequently plated onto FOA plates, the cells that contained the pCA90 vector [PP30(pCA90)] grew as well as those expressing the native Hsp82 protein [PP30(pHSP82b)] (Fig. 1a). Since the only Hsp90 coding sequences in S. cerevisiae PP30(pCA90) growing on FOA are those corresponding to C. albicans Hsp90, this Hsp90 can clearly provide essential Hsp90 functions in S. cerevisiae.



View larger version (80K):
[in this window]
[in a new window]
 
Fig. 1. (a) Growth in the absence (left) and presence (right) of FOA of S. cerevisiae PP30(pHSC82) transformed to leucine prototropy with an empty vector (pHSC82prom) or vectors for expression of either the C. albicans Hsp90 (pCA90) or the native S. cerevisiae Hsp82 (pHSP82b). Plates were photographed after 3 d growth at 30 °C. (b) Growth in the absence (left) and presence (right) of 5 µM {alpha}-factor of an S. cerevisiae ste11 mutant strain (E929-6C-20) and S. cerevisiae cells expressing C. albicans Hsp90 or S. cerevisiae Hsp82 [(PP30(pCA90) and PP30(pHSP82b), respectively]. Plates were photographed after 3 d growth at 30 °C.

 
C. albicans Hsp90 can provide the Hsp90 requirement for pheromone signalling in S. cerevisiae
Although Hsp90 is known to be essential for the activation of several important mammalian regulatory proteins, its only targets in S. cerevisiae identified to date are the Ste11 kinase needed for pheromone signalling (Louvion et al., 1998 ) and heat shock transcription factor (Duina et al., 1998 ). Pheromone signalling is defective in S. cerevisiae cells that either have very low Hsp90 levels or which express human Hsp90ß in place of the homologous Hsp90 (Louvion et al. , 1996 ). To determine if S. cerevisiae cells that express only the C. albicans Hsp90 are similarly defective in pheromone signalling, PP30(pCA90) cells were tested for their capacity to arrest growth in response to the S. cerevisiae mating pheromone {alpha}-factor. {alpha}-Factor arrested the growth of these cells (Fig. 1b), revealing that C. albicans Hsp90 can provide the Hsp90 function needed for pheromone signalling in S. cerevisiae.

C. albicans Hsp90 can provide steroid-responsive activation of the mammalian GR expressed in S. cerevisiae
There appears to be a physiological response of C. albicans to corticosteroids (Malloy et al., 1993 ), although the molecular mechanisms of this response and the basis of the enhanced incidence of candidal infections in those patients administered corticosteroids remain unclear. We transformed both PP30(pCA90) and PP30(pHSP82b) with a HIS3 vector (pHCA/rGR; Garabedian & Yamamoto, 1992 ) for S. cerevisiae expression of the F620S mutant form of mammalian GR. In this vector GR expression is under the control of the constitutively expressed glyceraldehyde-3- phosphate dehydrogenase promoter, whilst the F620S mutation seems to increase the affinity of GR for its steroid ligand (Garabedian & Yamamoto, 1992 ). We also transformed these cells with p{Delta}S26X, a URA3 plasmid that contains a reporter gene for GR activity (Schena et al., 1989 ). This reporter comprises three copies of the glucocorticoid response element in a modified CYC1 promoter, elements that cause GR-responsive expression of an adjacent lacZ gene.

Transformation of PP30(pCA90) and PP30(pHSP82b) with both pHCA/rGR and p{Delta}S26X gave HIS+ URA+ transformants that turned blue when replicated onto X-Gal indicator plates containing the GR ligand deoxycorticosterone. The addition of deoxycorticosterone to liquid cultures of these transformants resulted in the rapid induction of GR activity, induction levels being essentially identical irrespective of whether the cells were expressing the heterologous C. albicans Hsp90 or the native S. cerevisiae Hsp82. Two hours after dexamethasone addition to mid-exponential-phase cultures of S. cerevisiae PP30(pHSP82b) and PP30(pCA90) containing the GR receptor and GR-inducible lacZ reporter genes, ß-galactosidase levels had increased above the basal level of 0·1 units to 17·9 and 18·0 (±5%) units, respectively. In contrast, values in the control cultures without steroid addition were only 0·05 and 0·09 (±5%) units ß-galactosidase. C. albicans Hsp90 is therefore fully functional with regard to its ability to participate in steroid responsive activation of mammalian GR protein expressed in S. cerevisiae, an activation that is strictly dependent on Hsp90 function (Caplan, 1997 ; Nathan & Lindquist, 1995 ; Pratt & Dittmar, 1998 ).

C. albicans accumulates Hsp90 fragments, the major forms being of similar size to the N-terminally truncated forms of Hsp90 accumulated by respiratory S. cerevisiae cells
When total soluble protein is extracted from exponentially growing C. albicans cultures not only full-length Hsp90 is detected, but also smaller Hsp90 subfragments (Fig. 2a ). Prominent on the gels are fragments of 72–76 and 47 kDa, the latter most probably being the antigenic C-terminal domain subfragment of Hsp90 detected in earlier investigations (Matthews & Burnie, 1989 ; Matthews et al., 1987 ). A similar series of Hsp90 fragments is also seen in S. cerevisiae cultures under certain conditions of growth (Figs 2a, 3a). Whilst the cell extracts used for these experiments were prepared rapidly in the presence of a cocktail of protease inhibitors, it is still important to determine whether this partial Hsp90 degradation occurs within viable cells or is the result of cell lysis and the subsequent limited proteolysis of released cell contents. Several lines of evidence indicate that the former is the case. Firstly, plating of cells from the cultures used for protein extract preparation in these experiments showed that there had been no appreciable loss of cell viability. At the time of harvesting, practically 100% of the cells were able to form colonies on plates and stained with vital dyes. Secondly, a brief treatment of intact cells with chymotrypsin, prior to the addition of chymostatin, had no effect on these Hsp90 fragmentation patterns. Thirdly, this partial degradation of Hsp90, at least in S. cerevisiae cells, is prevented by a N-terminal (His)6 tag on Hsp82 protein that still permits Hsp90 function (Chang & Lindquist, 1994 ), but is unaffected by a mutation (pep4-3 ) that largely inactivates the vacuolar proteases which constitute the major source of proteolysis in S. cerevisiae cell extracts (Jones, 1991 ) (see Discussion). If the partial degradation of Hsp90 was the result of appreciable cell lysis or incomplete protein degradation after cell breakage, these mutational changes would be expected to exert the converse effects from those actually observed.



View larger version (62K):
[in this window]
[in a new window]
 
Fig. 2. Detection of full-length Hsp90 and Hsp90 subfragments in cell extracts from S. cerevisiae strain SUB61 ( S.c.) and C. albicans (C.a.). Cultures of both yeasts in exponential growth on YPD (glucose) medium at 30 °C were transferred to fresh YP/glycerol medium at time zero. Total protein extracts were then prepared after 0 (lane 1), 18 (lane 2), 24 (lane 3) or 42 (lane 4) h further aerobic maintenance at 30 °C. These extracts were gel-fractionated, blotted and probed with antisera that recognize epitopes either in the C-terminal domain (a) or at the N-terminus (b) of Hsp90. The latter antiserum does not detect the major Hsp90 degradation products of 72–76 and 47 kDa [arrowed in (a)]. Note that the shift to respiratory medium induces substantial Hsp90 partial degradation in S. cerevisiae but does not enhance this degradation in C. albicans.

 


View larger version (92K):
[in this window]
[in a new window]
 
Fig. 3. Detection of full-length Hsp90 and Hsp90 subfragments in cell extracts from S. cerevisiae PP30(pHSP82b) and PP30(pCA90). Cultures of both strains, in exponential growth on YPD (glucose) medium at 30 °C were transferred to fresh YP/glycerol medium. Total protein extracts were then prepared after a further 0 (lane 1), 18 (lane 2), 24 (lane 3) or 42 (lane 4) h aerobic maintenance at 30 °C. These extracts were gel-fractionated, blotted and probed with antisera that recognize epitopes either in the C-terminal domain (a) or at the N-terminus (b) of Hsp90. Major Hsp90 degradation products of 72–76 and 47 kDa are arrowed in (a).

 
We prepared total protein extracts of S. cerevisiae cultured under a number of different conditions and exposed to different stress regimes to investigate whether particular conditions induce the accumulation of relatively stable Hsp90 fragments. Stresses such as sublethal heat shock, osmostress or oxidant exposure did not increase levels of Hsp90 fragments in glucose-grown S. cerevisiae (data not shown). Instead, we found that it was a transfer from fermentative to respiratory media that most effectively caused the appearance of 47 kDa Hsp90 forms in S. cerevisiae cultures (Figs 2a, 3a). With the C. albicans isolate used for this study (3153A) the pattern of these Hsp90 fragments was unaffected by growth at different temperatures and, in contrast to S. cerevisiae, levels of the 47 kDa fragment actually decreased with the transfer of C. albicans cells to a respiratory carbon source (Fig. 2a).

The Hsp90 subfragments from C. albicans cells cross-react with an antiserum that recognizes the C-terminal domain of Hsp90 (Fig. 2a), but not with an antiserum recognizing an epitope at the extreme N-terminus of Hsp90 (Fig. 2b). Therefore, they are probably N-terminally truncated forms of Hsp90, like the Hsp90 fragments that accumulate in respiratory S. cerevisiae cultures (see Discussion). It is noteworthy that the major Hsp90 fragments in cultures of C. albicans and in glycerol-grown S. cerevisiae cultures are remarkably similar in size (Fig. 2a ), an indication that the events of incomplete Hsp90 degradation in these two organisms may be extremely similar.

Partial degradation of the heterologous C. albicans Hsp90 expressed in respiratory S. cerevisiae PP30(pCA90) closely resembles that of the homologous S. cerevisiae Hsp82 protein
We directly compared the accumulation of Hsp90 fragments in S. cerevisiae expressing either the homologous Hsp82 isoform of S. cerevisiae Hsp90 [PP30(pHSP82b)] or the C. albicans Hsp90 [PP30(pCA90)]. With transfer from glucose to a respiratory carbon source (glycerol) the two cultures accumulated an extremely similar, though not quite identical, series of Hsp90 subfragments (Fig. 3a ). As also seen in another S. cerevisiae strain and in C. albicans itself (Fig. 2a), the major accumulated fragments were of 72–76 and 47 kDa (Fig. 3a) and lacked the Hsp90 N-terminus (Fig. 3b). Full-length C. albicans Hsp90 is actually a slightly smaller protein than S. cerevisiae Hsp82 (Figs 3a, b), due primarily to the highly charged linker region between the N-terminal and C-terminal domains of Hsp90 being some 7 amino acids shorter in the former (Swoboda et al., 1996 ). The S. cerevisiae Hsp82 degradation products of 72–76 kDa arise from cleavages within the N-domain (see Discussion). Since a complete linker region is still present in these 72–76 kDa forms, the equivalent fragments derived from the C. albicans Hsp90 are also slightly smaller (Fig. 3a). In contrast, those Hsp90 degradation products in the 47 kDa size range arise from a cleavage at the beginning of the C- terminal domain (Matthews & Burnie, 1989 ) and lack the linker region. These 47 kDa fragments are of the same sizes when derived from the expression of either pHSP82b or pCA90 sequences (although there is a quantitative difference in that pHSP82b expression gave more of these 47 kDa forms than pCA90 expression; Fig. 3a). The blot shown in Fig. 3(a) is therefore a strong indication that the C. albicans Hsp90 expressed in S. cerevisiae cells and the native Hsp82 protein are both being degraded by essentially similar processes. Furthermore, the major 72–76 and 47 kDa subfragments of C. albicans Hsp90 accumulated in respiratory S. cerevisiae PP30(pCA90) cells (Fig. 3a) are identical in size to the fragments isolated from C. albicans (Fig. 2a). It seems probable therefore that S. cerevisiae cells can reasonably faithfully reproduce the steps that generate the major products of incomplete Hsp90 degradation in C. albicans cultures (Fig. 2a). However, more work is needed to precisely define the nature, and also the ordering, of the proteolytic events that lead to these relatively stable chaperone subfragments.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We had two main objectives in constructing a S. cerevisiae strain that had all coding sequences for the native Hsp90s deleted and which expressed instead the C. albicans Hsp90. The first was to show that C. albicans Hsp90 is capable of providing Hsp90 functions in S. cerevisiae. This provides a system in which variants of C. albicans Hsp90 can be tested for functionality. Thus, by altering the sequences corresponding to conserved epitopes on Hsp90 [such as LKVIRK (Matthews, 1994 )], then introducing these mutants into PP30(pHSC82) and plating transformants on FOA (see Methods), future studies will be able to determine the extent to which conserved pathogen Hsp90 epitopes can undergo mutational change. This in turn will indicate whether LKVIRK-targeting antibodies might protect against a number of eukaryotic pathogens (Matthews, 1994 ). Our second objective was to develop a system for investigating Hsp90 turnover, in particular to study whether partial Hsp90 degradation in vivo might be the origin of the antigenic C-domain subfragments of candidal Hsp90 found in the serum of patients with disseminated candidiasis (Matthews et al., 1987 ). There had, prior to this work, been no attempt to address the origins of these pathogen-derived serum Hsp90 fragments.

In parallel with this study we have been investigating the mechanism of the incomplete degradation of Hsp82 protein in respiratory S. cerevisiae cells (unpublished). We found that this degradation is totally abolished by the presence of an N-terminal (His)6 tag on Hsp82 protein. In contrast, an epitope tag placed at the C- terminus of Hsp82 does not abolish processing, but is retained on both the 72–76 and 47 kDa forms of Hsp82. The latter therefore comprise different forms of N-terminally truncated Hsp82 (which also accounts for their lack of detection by an antibody that detects the N-terminus of Hsp90s; Figs 2, 3 ). Whether the 72–76 kDa forms are precursors of the 47 kDa forms is at present unclear. However, the availability in S. cerevisiae of a wide range of mutants defective in either vacuolar proteolysis (Jones, 1991 ) or in the ubiquitination system for intracellular protein turnover (Hershko & Ciechanover, 1998 ) allowed us to probe the question of how Hsp90 fragments arise. We have shown that their production in S. cerevisiae is unaffected by the loss of vacuolar proteases, but altered in cells lacking the Ubc4 ubiquitin-conjugating enzyme. We are therefore currently investigating S. cerevisiae ubiquitination system and proteosome mutants to try to identify other components involved in Hsp82 partial degradation. It should be noted that the vast majority of the proteins targeted for degradation by the ubiquitination system are rapidly and completely degraded (Hershko & Ciechanover, 1998 ). Only a few cases are known where a protein is incompletely degraded, most of them from mammalian systems. The partial degradation of Hsp90 (Figs 2, 3) appears to be the first identified case of an incomplete protein degradation event in yeast and, as such, should lend itself to genetic dissection. Whether loss of N-terminal sequences in a substantial fraction of the Hsp90 of respiratory S. cerevisiae serves any physiological purpose is also at present unclear. Removal of these sequences should lead to the loss of a domain essential for the Hsp90 chaperone function (Panaretou et al., 1998 ).

There are indications that the release of Hsp90 or Hsp90 fragments from the cells of fungal pathogens causes appreciable toxicity to the host, rather analogous to the endotoxin release from Gram-negative bacteria that causes endotoxic shock in bacterial septicaemia. Even S. cerevisiae, a yeast that is normally considered nonpathogenic, becomes much more virulent in mice when its Hsp90 protein levels are elevated (Hodgetts et al., 1996 ). The reasons for this enhancing effect of Hsp90 on yeast pathogenicity are still unclear, but it may involve the malfunctioning of serum proteins that bind Hsp90 fragments (Matthews, 1994 ). Suppression of this effect may be a major ameliorating influence of the presence of anti- Hsp90 antibodies, since there is a close correlation between a good antibody response to Hsp90 (47 kDa antigen) and recovery in candidiasis patients (Matthews & Burnie, 1996 ; Matthews et al. , 1987 ). In addition to Hsp90, other highly abundant cytoplasmic proteins are recognized by the immune system in candidal infections. These include the hsp70 molecular chaperone (Bromuro et al. , 1998 ) and the glycolytic enzyme enolase (Franklin et al., 1990 ). These proteins are also present in the cell wall (Angiolella et al., 1996 ; Lopez-Ribot et al., 1996 ) even though, like Hsp90, they are initially made devoid of secretory signal sequences. It might therefore be assumed that candidal Hsp90 fragments arise in the serum through limited proteolysis after release of Hsp90 from lysed cells of the pathogen, and secretory proteinases are indeed found in the culture supernatants of clinically important Candida species [see Matthews (1994) for literature]. However, candidal secretory proteinases generally have a low pH optimum, unlike the tissues being invaded, which are mostly around neutral pH. Futhermore, our work indicates that the primary events leading to generation of the subfragments of C. albicans Hsp90 may be intracellular, not a process of limited degradation of Hsp90 after its release from lysed cells.

S. cerevisiae has become an extremely useful system for studying the ligand- and Hsp90-dependence of mammalian steroid receptors (Caplan, 1997 ; Pratt & Dittmar, 1998 ; Toft, 1998 ). Strains with conditional mutations in Hsp90 have been used to show that Hsp90 is required, not just for activation of some of these receptors, but also for maintaining them in their activated state (Nathan & Lindquist, 1995 ). In the oomycete water mould Achlya ambisexualis , binding of the fungal steroid antheridiol to its receptor/Hsp90 complex initiates hyphal branching (Brunt et al., 1990 ). However, no steroid-dependent gene activators are known for S. cerevisiae, or indeed for any other ascomycete fungus. Nevertheless, mammalian steroid hormones do have effects on ascomycete yeasts. Oestradiol directly stimulates the dimorphic transition of C. albicans from the yeast to the hyphal form (Madani et al. , 1994 ), whilst corticosteroids and pregnancy predispose humans to candidiasis (Malloy et al., 1993 ). The key targets whereby steroids produce such effects in Candida cells have still to be identified. Those C. albicans steroid-binding proteins that have been characterized to date are flavoproteins, with no obvious homology to the mammalian steroid receptors (Madani et al., 1994 ; Malloy et al., 1993 ). However, Hsp90-dependent steroid receptors may yet be uncovered in certain eukaryotic pathogens, most probably through the sequencing of their genomes. S. cerevisiae strains such as PP30(pCA90) offer the potential for co-expressing such receptors together with the appropriate pathogen Hsp90, a system which should make the ligand- and Hsp90-dependence of the receptors readily amenable to analysis.


   ACKNOWLEDGEMENTS
 
We thank A. J. P. Brown, D. Picard and K. Yamamoto for strains and plasmids. We are also indebted to Ruth Matthews for reading the manuscript. This work was supported by BBSRC grant G06776.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Angiolella, L. , Facchin, M. , Stringaro, A. , Maras, B. , Simonetti, N. & Cassone, A. (1996). Identification of a glucan- associated enolase as a main cell wall protein of Candida albicans and an indirect target of lipopeptide antimycotics. J Infect Dis 173, 684-690.[Medline]

Bohen, S. P. (1998). Genetic and biochemical analysis of p23 and ansamycin antibiotics in the function of Hsp90-dependent signalling proteins. Mol Cell Biol 18, 3330-3339 .[Abstract/Free Full Text]

Bonnefoy, S. , Gysin, J. , Blisnick, T. , Guillotte, M. , Carcy, B. , Pereira da Silva, L. & Mercereau-Puijalon, O. (1994). Immunogenicity and antigenicity of a Plasmodium falciparum protein fraction (90–110 kDa) able to protect squirrel monkeys against asexual blood stages. Vaccine 12, 32-40.[Medline]

Borkovich, K. A. , Farrely, F. W. , Finkelstein, D. B. , Taulein, J. & Lindquist, S. (1989). Hsp82 is an essential protein that is required in higher concentrations for growth of cells at higher temperature. Mol Cell Biol 9, 3919-3930.[Medline]

Bromuro, C. , La Valle, R. , Sandini, S. , Urbani, F. , Ausiello, C. M. , Morelli, L. , Fe d’Ostiani, C. , Romani, L. & Cassone, A. (1998). A 70-kilodalton heat shock protein of Candida albicans is highly immunogenic and enhances systemic murine candidiasis. Infect Immun 66, 2154-2162 .[Abstract/Free Full Text]

Brunt, S. A. , Riehl, R. & Silver, J. C. (1990). Steroid hormone regulation of the Achlya ambisexualis 85-kilodalton heat shock protein, a component of the Achlya steroid receptor complex. Mol Cell Biol 10, 273-281.[Medline]

Caplan, A. J. (1997). Yeast molecular chaperones and the mechanism of steroid hormone action. Trends Endocrinol Metab 8, 271-276.

Chang, H. C. J. & Lindquist, S. (1994). Conservation of Hsp90 macromolecular complexes in Saccharomyces cerevisiae. J Biol Chem 269, 24983-24988 .[Abstract/Free Full Text]

Cheng, L. , Hirst, K. & Piper, P. W. (1992). Authentic temperature- regulation of a heat shock gene inserted into yeast on a high copy number vector. Influences of overexpression of HSP90 protein on high temperature growth and thermotolerance. Biochim Biophys Acta 1132, 26-34.[Medline]

Dragon, E. A. , Sias, S. R. , Kato, E. A. & Gabe, J. D. (1987). The genome of Trypanosoma cruzi contains a constitutively expressed, tandemly arranged multicopy gene homologous to a major heat shock protein. Mol Cell Biol 7, 1271-1275.[Medline]

Duina, A. A. , Kalton, H. M. & Gaber, R. F. (1998). Requirement for Hsp90 and a Cyp-40-type cyclophilin in negative regulation of the heat shock response. J Biol Chem 273, 18974-18978 .[Abstract/Free Full Text]

Finley, D. , Ozkaynak, E. & Varshavsky, A. (1987). The yeast polyubiquitin gene is essential for resistance to high temperatures, starvation and other stresses. Cell 48, 1035-1046 .[Medline]

Franklin, K. M. , Warmington, J. R. , Ott, A. K. & Ashman, R. B. (1990). An immunodominant antigen of Candida albicans shows homology to the enzyme enolase. Immunol Cell Biol 68, 173-178.[Medline]

Garabedian, M. J. & Yamamoto, K. R. (1992). Genetic dissection of the signalling domain of a mammalian steroid receptor in yeast. Mol Biol Cell 3, 1245-1257.[Abstract]

Gething, M.-J. & Sambrook, J. (1992). Protein folding in the cell. Nature 355, 33-45.[Medline]

Gietz, R. D. & Sugino, A. (1988). New yeast–Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base-pair restriction sites. Gene 74, 527-534.[Medline]

Gomez, F. J. , Allendoerfer, R. & Deepe, G. S. (1995). Vaccination with recombinant heat shock protein from Histoplasma capsulatum protects mice against pulmonary histoplasmosis. Infect Immun 63, 2587-2595 .[Abstract]

Hershko, A. & Ciechanover, A. (1998). The ubiquitin system. Annu Rev Biochem 67, 425-479.[Medline]

Hodgetts, S. , Matthews, R. , Morrissey, G. , Mitsutake, K. , Piper, P. & Burnie, J. (1996). Over-expression of Saccharomyces cerevisiae hsp90 enhances the virulence of this yeast in mice. FEMS Immunol Med Microbiol 16, 229-234.[Medline]

Johnson, K. S. , Wells, K. , Bock, J. V. , Nene, V. , Taylor, D. W. & Cordingley, J. S. (1989). The 86-kilodalton antigen from Schistosoma mansoni is a heat-shock protein homologous to yeast HSP-90. Mol Biochem Parasitol 36, 19-28.[Medline]

Jones, E. W. (1991). Three proteolytic systems in the yeast Saccharomyces cerevisiae. J Biol Chem 266, 7963-7966 .[Free Full Text]

Kumari, S. , Lillibridge, C. D. , Bakeer, M. , Lowrie, R. C. , Jayaraman, K. & Philipp, M. T. (1994). Brugia malayi : the diagnostic potential of recombinant excretory/secretory antigens. Exp Parasitol 79, 489-505.[Medline]

Lopez-Ribot, J. L. , Alloush, H. M. , Masten, B. J. & Chaffin, W. L. (1996). Evidence for presence in the cell wall of Candida albicans of a protein related to the hsp70 family. Infect Immun 64, 3333-3340 .[Abstract]

Louvion, J.-F. , Warth, R. & Picard, D. (1996). Two eukaryote-specific regions of Hsp82 are dispensible for its viability and signal transduction functions in yeast. Proc Natl Acad Sci USA 93, 13937-13942 .[Abstract/Free Full Text]

Louvion, J.-F. , Abbas-Terki, T. & Picard, D. (1998). Hsp90 is required for pheromone signalling in yeast. Mol Biol Cell 9, 3071-3083.[Abstract/Free Full Text]

Lowrie, D. B. , Jascon, R. E. , Colston, M. J. & Silva, C. L. (1994). Towards a DNA vaccine against tuberculosis. Vaccine 12, 1537-1540 .[Medline]

Madani, N. D. , Malloy, P. J. , Rodriguez-Pombo, P. , Krishnan, A. V. & Feldman, D. (1994). Candida albicans estrogen- binding protein gene encodes an oxidoreductase that is inhibited by estradiol. Proc Natl Acad Sci USA 91, 922-926.[Abstract]

Malloy, P. J. , Zhao, X. , Madani, N. D. & Feldman, D. (1993). Cloning and expression of the gene from Candida albicans that encodes a high-affinity corticosteroid-binding protein. Proc Natl Acad Sci USA 90, 1902-1906 .[Abstract]

Matthews, R. C. (1994). Pathogenicity determinants of Candida albicans: potential targets for immunotherapy? Microbiology 140, 1505-1511 .[Medline]

Matthews, R. C. & Burnie, J. P. (1989). Cloning of a DNA sequence encoding a major fragment of the 47 kilodalton stress protein homologue of Candida albicans. FEMS Microbiol Lett 60, 25-30.

Matthews, R. & Burnie, J. (1996). Antibodies against Candida: potential therapeutics? Trends Microbiol 4, 354-358.[Medline]

Matthews, R. C. , Burnie, J. P. & Tabaqchali, S. (1987). Isolation of immunodominant antigens from sera of patients with systemic candidasis and characterization of serological response to Candida albicans. J Clin Microbiol 25, 230-237.[Medline]

Matthews, R. C. , Burnie, J. P. , Howat, D. T. R. & Walton, F. (1991). Autoantibody to Hsp90 can mediate protection against systemic candidosis. Immunol 74, 20-24.[Medline]

Matthews, R. C. , Hodgetts, S. & Burnie, J. P. (1995). Preliminary assessment of a human recombinant antibody fragment to hsp90 in murine invasive candidiasis. J Infect Dis 171, 1668-1671 .[Medline]

Nathan, D. F. & Lindquist, S. (1995). Mutational analysis of Hsp90 function: interactions with a steroid receptor and a protein kinase. Mol Cell Biol 15, 3917-3925 .[Abstract]

Panaretou, B. , Prodromou, C. , Roe, S. M. , O’Brien, R. , Ladbury, J. E. , Piper, P. W. & Pearl, L. H. (1998). ATP binding and hydrolysis are essential to the function of the Hsp90 molecular chaperone in vivo. EMBO J 17, 4829-4836 .[Abstract/Free Full Text]

Pratt, W. B. & Dittmar, K. D. (1998). Studies with purified chaperones advance the understanding of the mechanism of glucocorticoid receptor–hsp90 heterocomplex assembly. Trends Endocrinol Metab 9, 244-252.

Prodromou, C. , Roe, S. M. , O’Brien, R. , Ladbury, J. E. , Piper, P. W. & Pearl, L. H. (1997). Identification and structural characterization of the ATP/ADP-binding site in the Hsp90 molecular chaperone. Cell 90, 65-75.[Medline]

Rutherford, S. L. & Lindquist, S. (1998). Hsp90 as a capacitor for morphological evolution. Nature 396, 336-342.[Medline]

Salotra, P. , Chauhan, D. , Ralhan, R. & Bhatnagar, R. (1995). Tumour necrosis factor-{alpha} induces preferential expression of stress proteins in virulent promastigotes of Leishmania donovani. Immunol Lett 44, 1-5.[Medline]

Santos, M. A. , Keith, G. & Tuite, M. F. (1993). Non-standard translational events in Candida albicans mediated by an unusual seryl-tRNA with a 5'-CAG-3' (leucine) anticodon. EMBO J 12, 607-616.[Abstract]

Schena, M. , Freedman, L. P. & Yamamoto, K. R. (1989). Mutations in the glucocorticoid receptor zinc finger region that distinguish interdigitated DNA binding and transcriptional enhancement activities. Genes Dev 3, 1590-1601.[Abstract]

Sinclair, K. E. (1998). Effect of the yeast ubiquitination system on proteins accumulated at the entry to stationary phase. PhD thesis, University of London.

Skeiky, Y. A. , Benson, D. R. B. , Guderian, J. A. , Whittle, J. A. , Bacelar, O. , Carvalho, E. M. & Reed, D. G. (1995). Immune responses of leishmaniasis patients to heat shock proteins of Leishmania species and humans. Infect Immun 63, 4105-4114 .[Abstract]

Streit, J. A. , Donelson, J. E. , Agey, M. W. & Wilson, M. E. (1996). Developmental changes in the expression of Leishmania chagasi pg63 and heat shock protein in a human macrophage cell line. Infect Immun 64, 1810-1818 .[Abstract]

Swoboda, R. K. , Bertram, G. , Budge, S. , Gooday, G. W. , Gow, N. A. R. & Brown, A. J. P. (1996). Structure and function of the HSP90 gene from the pathogenic fungus Candida albicans. Infect Immun 63, 4506-4514 .[Abstract]

Toft, D. O. (1998). Recent advances in the study of hsp90 structure and mechanism of action. Trends Endocrinol Metab 9, 238-243.

Young, D. B. (1992). Heat shock proteins – immunity and autoimmunity. Curr Opin Immunol 4, 396-400.[Medline]

Received 6 July 1999; revised 26 August 1999; accepted 10 September 1999.