|Home | About | Journals | Submit | Contact Us | Français|
Hsp90 is an environmentally contingent molecular chaperone that influences the form and function of diverse regulators of cellular signaling. Hsp90 potentiates the evolution of fungal drug resistance by enabling crucial cellular stress responses. Here we demonstrate that in the leading fungal pathogen of humans, Candida albicans, Hsp90 governs cellular circuitry required not only for drug resistance but also for the key morphogenetic transition from yeast to filamentous growth that is crucial for virulence. This transition is normally regulated by environmental cues, such as exposure to serum, that are contingent upon elevated temperature to induce morphogenesis. The basis for this temperature dependence has remained enigmatic.
We show that compromising Hsp90 function pharmacologically or genetically induces a transition from yeast to filamentous growth in the absence of external cues. Elevated temperature relieves Hsp90-mediated repression of the morphogenetic program. Hsp90 regulates morphogenetic circuitry by repressing Ras1-PKA signaling. Modest Hsp90 compromise enhances the phenotypic effects of activated Ras1 signaling while deletion of positive regulators of the Ras1-PKA cascade blocks the morphogenetic response to Hsp90 inhibition. Consistent with the requirement for morphogenetic flexibility for virulence, depletion of C. albicans Hsp90 attenuates virulence in a murine model of systemic disease.
Hsp90 governs the integration of environmental cues with cellular signaling to orchestrate fungal morphogenesis and virulence, suggesting new therapeutic strategies for life-threatening infectious disease. Hsp90’s capacity to govern a key developmental program in response to temperature change provides a new mechanism that complements the elegant repertoire that organisms utilize to sense temperature.
Hsp90 is an essential molecular chaperone that regulates the form and function of diverse client proteins in all eukaryotes [1–3]. Many Hsp90 clients are key regulators of cell signaling and dwell in incompletely folded or aggregation prone states. They dynamically cycle through complexes with Hsp90 and co-chaperones until the proper signal engenders their activation. As a consequence of its function in chaperoning regulators of cell circuitry, Hsp90 has a capacity to modulate the translation of genotype to phenotype . Hsp90 can buffer the expression of genetic and epigenetic variation, allowing it to be revealed in response to environmental stress [4–6]. Although Hsp90 is induced under conditions of stress, such as increased temperature, global problems in protein folding that ensue can overwhelm its functional capacity. As a capacitor for the storage and release of genetic variation, Hsp90 can influence the emergence of new traits.
In fungi, Hsp90 potentiates the rapid evolution of drug resistance . Fungal pathogens are a leading cause of human mortality worldwide, especially among immunocompromised individuals. Fungal infections are notoriously difficult to treat with the limited number of antifungal drugs available that have selective toxicity to fungi, which share close evolutionary relationships with their human hosts . Therapeutic challenges are exacerbated as resistance has emerged for all clinically useful antifungals. Hsp90 enables the emergence of resistance to the most widely deployed antifungals, the azoles, which exert cell membrane stress . Genetic or pharmacological inhibition of Hsp90 function blocks the emergence of azole resistance in the model yeast Saccharomyces cerevisiae and the pathogenic yeast Candida albicans . Compromising Hsp90 function enhances the therapeutic efficacy of azoles in two metazoan models of disseminated C. albicans disease .
C. albicans reigns as the leading fungal pathogen of humans and its virulence is intimately coupled with morphogenetic plasticity. Growth forms range from unicellular budding yeast to true hyphae composed of filaments with parallel-sided walls lacking constrictions between cells [9, 10]. Phenotypically intermediate pseudohyphal forms are composed of filaments of elongated cells demarked by constrictions. Due to phenotypic and regulatory similarities [11, 12], hyphal and pseudohyphal forms are often referred to collectively as filamentous forms. Morphogenetic plasticity is implicated in virulence as mutants defective in the yeast to filament transition are attenuated in virulence [9, 13]. Filaments have invasive properties that can promote tissue penetration and escape from immune cells, while yeast are suited for dissemination in the bloodstream. The coupling between morphogenesis and virulence is reinforced as genes that govern morphogenesis are co-regulated with those encoding virulence factors such as proteases and adhesins .
C. albicans morphogenesis is induced by distinct environmental cues and is governed by complex regulatory networks. Morphogenesis is induced by signals such as nutrient limitation, pH, CO2, and serum . Ras1, a small GTPase, regulates two key signal transduction pathways that control morphogenesis, the mitogen activated protein (MAP) kinase pathway and the cAMP-protein kinase A (PKA) pathway . The cAMP-PKA pathway has the most prominent role in C. albicans morphogenesis such that its inactivation blocks filamentation in diverse conditions, including exposure to serum. The active component of serum that induces filamentation is bacterial cell wall fragments that directly stimulate the production of cAMP . Notably, the induction of filamentous growth by the vast majority of environmental cues, including serum, requires a concomitant increase in temperature and the basis for this requirement has remained enigmatic.
We report here that Hsp90 orchestrates C. albicans morphogenesis and that it is relief of Hsp90-mediated repression of the morphogenetic program that underpins the requirement for elevated temperature to induce filamentous growth. Compromising Hsp90 function induces a yeast to filament transition via the Ras1-PKA cascade and attenuates virulence in a murine model of systemic disease. Our results establish an entirely new role for Hsp90 in this central developmental process.
We initially compromised Hsp90 function pharmacologically with structurally unrelated inhibitors geldanamycin (GdA) or radicicol (RAD) that bind with high affinity to Hsp90’s unusual ATP binding pocket . In liquid rich medium at 30°C, C. albicans normally grows as a budding yeast. Inhibition of Hsp90 induced a profound morphogenetic switch to filamentous growth (Figure 1A and Figure S1A). Next, we engineered strains in which the only HSP90 allele was regulated by one of two repressible promoters. In our system, transcription from tetO is repressed by tetracycline or the analog doxycyline and induced in their absence; transcription from MAL2p is repressed by glucose and induced by maltose (Figure S1B). Transcriptional repression of HSP90 from either promoter induced a yeast to filament transition (Figure 1B). This transition was also observed on solid medium, coupled with robust invasion into the agar (Figure 1C). Neither the addition of doxycycline nor the change in carbon source affected growth or morphogenesis of strains in which HSP90 was not subject to transcriptional repression (Figure 1C and Figure S1C, D, and E). The magnitude of Hsp90 depletion required to induce filamentation reduced growth rates (Figure S1C) but had little effect on stationary phase cell density (Figure S1D). Further, cells were able to revert to yeast upon relief of Hsp90 inhibition (Figure S2). Thus, Hsp90 reversibly governs the morphogenetic transition from yeast to filamentous growth.
A striking requirement for most conditions that induce filamentation, including serum, is an elevated temperature of 37°C, yet the basis for this requirement has remained enigmatic . Given that Hsp90 function can be compromised at elevated temperatures due to global problems in protein folding and given its role in morphogenesis (Figure 1), we took four complementary approaches to determine if the temperature requirement is to relieve Hsp90-mediated repression of the morphogenetic program.
First, we tested whether a modest reduction of Hsp90 could bypass the temperature requirement for serum-induced filamentation. Exposure to serum at 30°C did not induce filamentation in a strain with wild-type or heterozygous HSP90 levels nor did the moderate reduction of Hsp90 achieved when the native HSP90 promoter was replaced with MAL2p and the MAL2p-HSP90/hsp90Δ strain was grown in maltose without serum (Figures S1B and 2A). However, the reduction of Hsp90 in the MAL2p-HSP90/hsp90Δ strain did enable serum to induce morphogenesis at 30°C (Figure 2A), consistent with the temperature requirement being rooted in relief of Hsp90-mediated repression.
Second, we postulated that if elevated temperature were required to relieve Hsp90 repression of filamentation, then a lower dose of Hsp90 inhibitor would suffice to induce filamentous growth at 37°C compared to 30°C. We found that a high dose of Hsp90 inhibitor induced morphogenesis at either temperature, while a lower dose only induced morphogenesis at 37°C (Figure 2B). That less Hsp90 inhibitor is required to induce filamentation at elevated temperature suggests that Hsp90’s functional capacity may be overwhelmed.
Third, to test whether Hsp90’s functional capacity is overwhelmed in conditions favoring filamentation, we exploited activation of the heat shock response. In S. cerevisiae, Hsf1 governs transcriptional responses to stress by remodeling expression of ~3% of the genome . Hsp90 exerts a repressive effect on Hsf1 such that compromising Hsp90 leads to induction of the heat shock response, upregulating heat shock proteins among other targets . We monitored expression of a heat shock reporter using a construct with the HSP70 promoter containing three heat shock elements fused to lacZ, integrated at the native HSP70 locus. As expected, inhibition of Hsp90 with GdA induced expression from the HSP70 promoter (Figure S3A). Either an increase in temperature from 30°C to 37°C or exposure to serum at 30°C yielded an intermediate induction (P < 0.001, ANOVA, Figure 2C). The combination of serum and elevated temperature resulted in further induction of the heat shock response (P < 0.01), consistent with Hsp90’s functional capacity being compromised by conditions that induce filamentous growth.
Fourth, to validate that Hsp90’s functional capacity is overwhelmed by conditions favoring filamentation, we monitored activation of the Hsp90 client protein calcineurin, which regulates stress responses [4, 20]. Hsp90 physically interacts with calcineurin and keeps it poised for activation in response to calcium. Upon activation, calcineurin dephosphorylates target proteins including the transcription factor Crz1 [21, 22]. De-phosphorylated Crz1 translocates to the nucleus and binds calcineurin-dependent response elements (CDREs) to induce transcription. We used a construct with the UTR2 promoter, which contains a CDRE and is regulated by calcineurin , fused to lacZ. Inhibition of Hsp90 with GdA reduced calcineurin activation in response to CaCl2 (Figure S3B). Filament-inducing conditions of serum at 37°C reduced calcineurin activation in response to CaCl2 relative to serum at 30°C (P < 0.001, Figure 2D), consistent with compromised Hsp90 function.
We screened a panel of previously characterized mutants to identify those blocked in filamentation in response to Hsp90 inhibition. Positive regulators of the Ras1-PKA pathway were blocked in the morphogenetic response (Figure 3A and B and Figure S4). Ras1 cycles between an inactive GDP-bound and an active GTP-bound state and regulates filamentation in response to serum and other stimuli [16, 23, 24]. Ras1 is activated by the guanine exchange factor Cdc25 and stimulates the adenylyl cyclase activity of Cdc35, which works with cyclase associated protein Cap1 to produce cAMP [25, 26]. Elevated intracellular cAMP levels activate the catalytic subunits of PKA (Tpk1 and Tpk2). We found that ras1, cdc25, and cdc35 null mutants were completely blocked in the morphogenetic response while cap1 mutants showed a partial block in the presence of Hsp90 inhibitors. Neither tpk1 nor tpk2 null mutants were blocked, consistent with their partial redundancy ; however, transcriptional repression of the only TPK1 allele in a strain lacking TPK2 completely blocked filamentation. These mutants showed comparable blocks in serum-induced filamentation, consistent with both Hsp90 and serum acting through this pathway (Figure S5). PKA activates Efg1, which activates a transcriptional program governing morphogenesis in response to many cues [27, 28]. While efg1 null mutants were blocked in response to serum (Figure S5), they filament in response to Hsp90 inhibition (Figure 3A), suggesting that Hsp90 signals through other downstream targets of PKA.
To assess the morphogenetic response to Hsp90 inhibition by an independent and quantitative method, we analyzed transcript levels of the well-characterized hyphal wall protein 1, HWP1. HWP1 is expressed exclusively in filamentous forms and is regulated by Ras1-PKA signaling via Efg1 in concert with other regulators . In the wild-type, inhibition of Hsp90 caused more than 250-fold induction of HWP1, measured by quantitative real time RT-PCR relative to the constitutive GPD1 control (Figure 3C). In contrast, induction of HWP1 was blocked in the ras1, cdc25, or cdc35 mutants, consistent with the morphological block as yeast. There was a moderate 18-fold induction of HWP1 in the cap1 mutant upon Hsp90 inhibition, reflecting its partial defect in filamentation. High-level HWP1 induction was evident in the tpk1 and tpk2 mutants while depletion of PKA blocked induction. There was no significant induction of HWP1 in the efg1 mutant, consistent with Efg1’s established role in HWP1 regulation .
Next, we tested for genetic interactions in mutants with activated Ras1-signaling or reduced Hsp90. A strain harboring one dominant active RAS1 allele (Ras1V13) demonstrated increased filamentation under permissive conditions [23, 30], but remained as yeast with smooth colonies on solid rich medium at 30°C (Figure 3D). A strain heterozygous for HSP90 remained as yeast with smooth colonies under these conditions. However, the combination of Ras1V13 and reduced Hsp90 levels yielded enhanced filamentation in both solid and liquid medium (Figure 3D and Figure S6A). Deletion of IRA2, which encodes the GTPase-activating protein that cycles Ras1 to the inactive state, also stimulates Ras1-signaling . Deletion of one IRA2 allele had no effect in liquid in a strain with wild-type Hsp90, however, deleting one HSP90 allele in the ira2Δ heterozygote enhanced filamentation (Figure S6B). Thus, genetic epistasis suggests that Hsp90 regulates morphogenesis by repressing Ras1-PKA signaling until the appropriate cues compromise Hsp90, relieving repression of the morphogenetic program.
There are two models by which Hsp90 could repress Ras1-PKA signaling. Hsp90 could physically interact with a positive regulator and maintain an inactive conformation until Hsp90 is compromised. Alternatively, Hsp90 could interact with a negative regulator and keep it poised for activation. Therefore, compromising Hsp90 function should either phenocopy a gain-of-function allele of a positive signaling component of the Ras1-PKA pathway or a loss-of-function allele of a negative regulator. More complicated models involving Hsp90 regulation of multiple pathway components are also conceivable.
We tested genetically whether some of the Ras1-PKA negative regulators are the sole targets of Hsp90. As mentioned above, Ira2 is a negative regulator of Ras1-signaling. Pde1 and Pde2 are phosphodiesterases that degrade cAMP . Null mutants of these negative regulators remained as yeast in rich medium at 30°C while Hsp90 inhibition induced filamentation (Figure 4). Null mutants of the PKA negative regulatory subunit, Bcy1, are not viable on their own and thus could not be tested. Thus, loss of function of Ira2, Pde1, or Pde2 does not phenocopy Hsp90 inhibition and Hsp90 inhibition can still induce filamentation in their absence, suggesting that Hsp90 does not regulate morphogenesis through any one of these proteins.
To determine where Hsp90 acts in the Ras1-PKA cascade, we exploited the quorum-sensing molecules farnesol and dodecanol . Farnesol is produced by C. albicans and represses filamentation in dense populations. Dodecanol is an alcohol with similar activities to 3-oxo-C12-homoserine lactone, a signaling molecule produced by Pseudomonas aeruginosa. Both farnesol and dodecanol inhibit filamentation induced by serum or Ras1V13 (Figure 5 and ). Since serum directly stimulates Cdc35  and farnesol and dodecanol block serum-induced filamentation , farnesol and dodecanol act at or below Cdc35. In contrast, neither farnesol nor dodecanol inhibited filamentation induced by Hsp90 inhibition (Figure 5). The trivial explanation that farnesol or dodecanol are ineffective because cells are locked as filaments is excluded given that cells can revert to yeast upon relief of Hsp90 inhibition (Figure S2). Thus, Hsp90 is likely to function at a point in the pathway below Cdc35 to regulate morphogenesis.
Hsp90’s role in morphogenesis and its essentiality suggest it as an ideal therapeutic target. To test the impact of genetic depletion of Hsp90 on C. albicans virulence we employed a murine model of systemic disease in which inoculum is delivered by tail vein injection and progresses from the bloodstream to deep-seated infection of major organs such as the kidney . To regulate Hsp90, we exploited the tetracycline-repressible promoter (tetO, Figure S1B). There was no significant difference in kidney fungal burden between mice infected with the wild-type strain or the HSP90 heterozygote (Figure 6). In the absence of tetracycline, the strain with its only HSP90 allele regulated by tetO has comparable Hsp90 levels to a strain in which the only HSP90 allele is under the control of the native stress-inducible promoter (Figure S1B). While HSP90 expression from the native promoter is induced by increased temperature, expression of HSP90 from tetO is not . Even in the absence of tetracycline, mice infected with the strain with its only HSP90 allele regulated by tetO had significantly lower kidney fungal burden than those infected with the strain with its HSP90 allele regulated by the native promoter (Figure 6, P < 0.001, ANOVA). Treating mice infected with the wild-type strain or the HSP90 heterozygote with tetracycline had no significant impact on fungal burden, while tetracycline-induced depletion of HSP90 resulted in sterilization of the kidneys (Figure 6). Thus, Hsp90 provides an attractive therapeutic target for the treatment of life-threatening C. albicans infection.
Our results establish a novel role for Hsp90 in regulating developmental process. In the leading fungal pathogen of humans, C. albicans, Hsp90 regulates the morphogenetic transition from yeast to filamentous growth that is intimately coupled with virulence (Figure 1). Hsp90 orchestrates morphogenesis by repressing specific cellular signaling. Conditions that compromise Hsp90 induce morphogenesis via the Ras1-PKA cascade. Canonical environmental cues that stimulate morphogenesis, such as serum, require elevated temperature to alleviate Hsp90-mediated repression of the morphogenetic program. Thus, Hsp90 links environmental stress with cellular signaling to regulate C. albicans morphogenesis and pathogenesis.
In the human host, where C. albicans resides as a commensal member of the mucosal microbiota, it is poised to respond to environmental cues that induce morphogenesis. Decades of laboratory studies have focused on delineating the stimuli that induce this developmental transition and the regulatory circuitry involved, yet the basis for the elevated temperature requirement to induce filamentation in vitro has remained obscure [9, 10, 16]. Elevated temperature is not required for filamentation in specific conditions such as embedded growth where the Czf1 pathway prevails, but is required for response to serum, nutrient limitation, pH, and CO2 that depend on cellular circuitry regulated by Ras1 or the Rim101-dependent pH pathway. Marshalling four complementary approaches, we demonstrate that compromising Hsp90 can overcome the requirement for elevated temperature for serum-induced filamentation and that Hsp90 function is compromised under conditions that induce morphogenesis (Figure 2). Although Hsp90 is induced in response to heat stress, its functional capacity to regulate client protein activation can be overwhelmed by global problems in protein folding. Together, this implicates Hsp90 as a key regulator of cellular circuitry governing morphogenesis.
Hsp90 regulates the morphogenetic program by repressing Ras1-PKA signaling, as evidenced by both positive and negative epistasis. Our results are consistent with Hsp90 mediating a repressive effect on a positive regulator (Figure 4). There is precedent for such regulation, as Hsp90 represses Hsf1 thereby regulating induction of the heat shock response [18, 19]. Given that the quorum-sensing molecules farnesol and dodecanol block the morphogenetic response to serum , which stimulates Cdc35 , but do not block the response to Hsp90 inhibition (Figure 5) Hsp90 is likely to function downstream of Cdc35. Depletion of PKA or deletion of upstream regulators blocks the morphogenetic response to Hsp90 inhibition, suggesting that PKA activation is dependent upon upstream input. This contingency is consistent with impairment of ras1 mutants in response to serum-induced filamentation, despite serum stimulating the downstream factor Cdc35 . Thus, our results are consistent with a model in which Hsp90 inhibition leads to activation of PKA and induction of filamentous growth.
Both serum and Hsp90 inhibition stimulate filamentation by activating Ras1-PKA signaling, however, marked regulatory differences underpin the responses. While efg1 mutants are blocked in the morphogenetic response to serum, they are not blocked in response to Hsp90 inhibition (Figure 3 and Fig S5), suggesting that Hsp90 works via effectors downstream of PKA distinct from Efg1. That Efg1 is required to regulate filament-specific genes, such as HWP1 (Figure 3C), but is not required for filamentation is reminiscent of the transcription factor Bcr1, which is also required for filament-specific transcription but not for filamentation . Notably, genes implicated in morphogenesis that are dependent on Efg1 in vitro are not dependent on Efg1 in a mouse, suggesting that this regulatory circuitry is more complex than previously appreciated . Other transcriptional regulators that function downstream of PKA signaling and are implicated in C. albicans morphogenesis include Tec1 and Flo8 . In addition to regulatory differences, there are also morphological differences between filaments induced by serum and by Hsp90 inhibition. This likely reflects the pleiotropic impact of Hsp90 on multiple signaling pathways in contrast to the specificity of serum stimulation of Cdc35.
Hsp90’s pleiotropic effects on growth and morphogenesis have broad therapeutic implications. Compromising C. albicans Hsp90 expression results in a significant reduction in virulence in a murine model of systemic disease (Figure 6). Further depletion of Hsp90 results in clearance of the infection. Given that morphogenetic flexibility is required for virulence and compromising Hsp90 drives filamentation, this virulence attenuation could reflect Hsp90’s effect on morphogenesis, virulence regulators, or basal growth. Regardless of mechanism, targeting Hsp90 in the pathogen provides an attractive therapeutic strategy. Drugs structurally related to GdA are in advanced phase clinical trials for treating cancer . At concentrations well tolerated in humans, Hsp90 inhibitors abrogate fungal drug resistance  and enhance therapeutic efficacy of antifungals in a metazoan model of fungal disease , providing another avenue by which Hsp90 could be harnessed to treat of fungal infections.
Strikingly, compromising Hsp90 function induces a developmental program in C. albicans, in contrast to other systems in which compromising Hsp90 blocks developmental transitions. In the slime mold Dictyostelium discoideum, Hsp90 inhibition arrests development at the ‘mound’ stage . In the protozoan parasite Plasmodium falciparum Hsp90 inhibition blocks development from ring stage to trophozoite and in Leishmania donovani it blocks development in the promastigote stage . Given that Hsp90 may interact with up to 10% of the yeast proteome , there will likely be myriad ways in which Hsp90 orchestrates cellular signaling governing environmentally responsive developmental programs.
Hsp90’s capacity to sense temperature change and govern a key developmental program provides a new mechanism that complements the elegant repertoire that organisms utilize to sense temperature. Intriguingly, the systemic dimorphic fungi exhibit temperature-dependent morphogenesis in the inverse direction from C. albicans such that they are filamentous in the soil at ambient temperature and switch to the pathogenic yeast form at human body temperature upon infection . Hsp90’s role in C. albicans temperature-dependent morphogenesis raises the tantalizing possibility that it may impinge on global morphogenetic regulators in dimorphic fungi, such as the hybrid histidine kinase Drk1 . Another distinct mechanism discovered in fungi is temperature-dependent splicing that regulates the circadian clock . In bacteria, thermosensitive nutrient and quorum sensing governs thermotaxis via methylation of Tsr receptors  and an RNA thermosensor controls synthesis of a transcriptional regulator . In mammals, thermosensitive transient receptor potential channels in sensory neurons and skin cells relay temperature information to the brain . The multitude of distinct mechanisms organisms utilize to sense temperature illustrates the stunning complexity of ways in which environmentally contingent responses are transduced in biological systems.
Strains used in this study are listed in Table S1 and their construction is described in the Supplemental Data. C. albicans strains were cultured on YPD (2% bacto-peptone, 1% yeast extract, and 2% glucose) or YPM (as YPD, but with 2% maltose). Overnight cultures were grown in 5 ml of YPD or YPM at 30°C unless otherwise indicated. Cells were diluted to OD600 of 0.2 and grown for the indicated time period for the following treatments: 10% fetal bovine serum (GIBCO), 10 μM geldanamycin (GdA, A.G. Scientific, Inc.), 20 μM radicicol (A.G. Scientific, Inc.), 20μg/mL doxycycline (BD Biosciences), 200 mM dodecanol (Sigma Aldrich Co.), 200 mM farnesol (Sigma Aldrich Co.).
Plasmids used in this study are listed in Table S2 and their construction is described in the Supplemental Data. Plasmids were sequenced to verify the absence of any nonsense mutations. Primers used in this study are listed in Table S3.
Imaging of cells cultured in liquid was performed using Differential Interference Contrast microscopy using a Zeiss Axio Imager.MI and Axiovision software (Carl Zeiss, Inc.). Microscopy of colonies on solid media was performed using a Zeiss Stereo Discovery.V8.
Cells were grown on YPD or YPM plates for 2 days, washed with water, and photographed.
Cells were grown overnight in YPD at 30°C, diluted to OD600 of 0.2 with treatment as indicated, and grown for 1 – 3 hours at 30°C or 37°C. Protein was extracted [46, 47], and protein concentrations were determined by Bradford analysis (Bio-Rad Laboratories, Inc.). β-galactosidase activity was measured using the substrate ONPG (O-nitrophenyl-β-D-galactopyranoside, Sigma Aldrich Co.), as described . β-galactosidase activity is given in units of nanomoles ONPG converted per minute per milligram of protein.
Cells were grown overnight in YPD or 1:1 YPD:YPM at 30°C, were diluted to OD600 of 0.2 with 10 μM GdA as indicated, and grown overnight. Cells were again diluted to OD600 of 0.2 in the same conditions and grown to mid-log phase. RNA was isolated and PCR was performed as described . Reactions were performed in triplicate, using oLC752 and oLC753 (GPD1) and oLC750 and oLC751 (HWP1). Data was analyzed using iQ5 Optical System Software Version 2.0 (Bio-Rad Laboratories, Inc).
Inoculum was prepared as described for injection of 200 μL of a 1 × 106 CFU/mL suspension . We observed discordance between cell counts and CFU measurements for the tetO-HSP90/hsp90Δ strain in initial experiments, such that CFU values were ~60% lower than expected based on cell counts . A 10-fold increase in inoculum concentration for tetO-HSP90/hsp90Δ to 200 μL of 1 × 107 cells/mL produced ~4-fold higher CFU/ml than the inoculum for the wild type and the HSP90 heterozygote and thus was chosen as the dose of tetO-HSP90/hsp90Δ for these experiments. Male CD1 mice (Charles River Laboratories) age 8 weeks (weight 30–34 g) were infected via the tail vein. For the wild type and the HSP90 heterozygote the sample sizes were n = 5 – 13 mice per strain and treatment condition. For the tetO-HSP90/hsp90Δ strain the sample sizes were n = 22 mice for untreated and n = 14 for tetracycline treated. Mice receiving tetracycline were given sterile water supplemented with 2 mg/mL tetracycline (Sigma Biochemicals) beginning 7 days prior to infection and continued until sacrifice. Water was changed once daily. Daily weights during the 7-day lead in period were taken to ensure that mice remained hydrated. Mice were observed three times daily for signs of illness and weighed daily. At day 4 following injection, mice were sacrificed by CO2 asphyxiation and the left kidney was removed aseptically, homogenized in PBS and serial dilutions plated for determination of kidney fungal burden, as described . CFU values were expressed as CFU/g of tissue, log-transformed and compared using an ANOVA with post-hoc testing of significance between groups (GraphPad Prism 4.0). Murine work was performed under a protocol approved by the Institutional Animal Use and Care Committee at Duke University Medical Center.
We thank H. Aziz and W. Schell for assistance with murine experiments; J. Köhler, G. Fink, and D. Hogan for strains and plasmids; B. Larsen for an Hsp90 antibody; G. Fink, J. Köhler, D. Hogan, C. Boone, P. Roy, and Cowen lab members for helpful discussions. R.S.S. is supported by an Ontario Graduate Scholarship; P.U. and J.H. by NIAID grant AI50438; A.K.Z. by NIH/NIAID K08-AI065837-04; J.R.P. by Public Health Service Grants AI73896 and AI28388; L.E.C by a Career Award in the Biomedical Sciences from the Burroughs Wellcome Fund, by a Canada Research Chair in Microbial Genomics and Infectious Disease, and by Canadian Institutes of Health Research Grant MOP-86452.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.