|Home | About | Journals | Submit | Contact Us | Français|
The echinocandin caspofungin is a new antifungal drug that blocks cell wall synthesis through inhibition of β-(1-3)-glucan synthesis. Saccharomyces cerevisiae cells are able to tolerate rather high caspofungin concentrations, displaying high viability at low caspofungin doses. To identify yeast genes implicated in caspofungin tolerance, we performed a genome-wide microarray analysis. Strikingly, caspofungin treatment rapidly induces a set of genes from the protein kinase C (PKC) cell integrity signaling pathway, as well as those required for cell wall maintenance and architecture. The mitogen-activated protein kinase Slt2p is rapidly activated by phosphorylation, triggering signaling through the PKC pathway. Cells lacking genes such as SLT2, BCK1, and PKC1, as well as the caspofungin target gene, FKS1, display pronounced hypersensitivity, demonstrating that the PKC pathway is required for caspofungin tolerance. Notably, the cell surface integrity sensor Wsc1p, but not the sensors Wsc2-4p and Mid2p, is required for sensing caspofungin perturbations. The expression modulation of PKC target genes requires the transcription factor Rlm1p, which controls expression of several cell wall synthesis and maintenance genes. Thus, caspofungin-induced cell wall damage requires Wsc1p as a dedicated sensor to launch a protective response through the activated salvage pathway for de novo cell wall synthesis. Our results establish caspofungin as a specific activator of Slt2p stress signaling in baker's yeast.
The fungal cell wall is the essential cellular boundary, controlling many transport processes, cellular metabolism, as well as all communication with the extracellular world. Because of its mechanical strength, it allows cells to withstand turgor pressure and consequently prevents cell lysis. Proper cell wall architecture requires cell wall components such as β-1,3-glucan, chitin, and mannoproteins, all of which form a large complex (21, 22, 25). Their coordinated synthesis represents an essential step for the assembly of a functional cell wall to ensure cell integrity (10). Certain antifungal drugs, such as caspofungin, a semisynthetic derivative of the secondary metabolite pneumocandin Bo from the fungus Glarea lozoyensis (1), specifically block cell wall synthesis. Caspofungin acts fungicidal, since it is a noncompetitive inhibitor of the 1,3-glucan synthases Fks1p and Fks2p (30), both of which are believed to catalyze the polymerization of UDP-glucose into β-1,3-glucan during cell wall biogenesis (39). When caspofungin is combined with other antifungal drugs, such as fluconazole or amphotericin B, synergistic or additive effects against a variety of clinically important fungal pathogens have been demonstrated in vitro and in vivo (56). Cells lacking Fks1p display increased chitin content, elevated levels of the second 1,3-β-glucan synthase, Fks2p (42), as well as altered expression of glycosylphosphatidylinositol (GPI)-anchored cell surface proteins (57). These changes may reflect a compensatory response to maintain cell wall integrity.
The intracellular protein kinase C (PKC) signal transduction pathway is essential for sensing cell integrity under a variety of environmental conditions or morphogenetic events. The PKC response regulates cell wall and actin cytoskeleton dynamics (13), and it is activated during polarized growth such as budding and mating (64). In addition, PKC signaling is activated by environmental conditions that jeopardize cell wall stability, including high temperature (19), hypotonic shock (8), or impaired cell wall synthesis (24). Accordingly, the absence of PKC signaling causes cell lysis when yeast is exposed to any of these inducing conditions. Osmotic stabilization can prevent cell lysis, which also indicates defective maintenance of a functional cell wall (34, 58).
Sensing of cell wall perturbations requires dedicated surface sensors. Genetic studies place the WSC (for cell wall integrity and stress response component) genes upstream of the mitogen-activated protein (MAP) kinase cascade. The WSC family comprises four genes: WSC1/HCS77/SLG1, WSC2, WSC3, and WSC4/YFW1 (12, 17, 37, 59, 63). The Wsc1-4p proteins are highly O glycosylated plasma membrane proteins that contain a extracellular domain with a cysteine motif, and an S/T-rich domain that carries glycosylation sites (36, 49, 59). Additional cell wall stress sensors are the partially redundant Mid2p and Mtl1p cell surface proteins. These proteins act as mechanosensors of cell wall stress during budding or pheromone-induced morphogenesis, high temperature, or other cell wall disturbances (12, 24, 49, 59).
The activation of the PKC pathway proceeds through the small G protein Rho1p, via Pkc1p (35), and a downstream MAP kinase cascade. Although the molecular mechanisms by which sensors transmit the signal to downstream effectors remain ill defined, the Rho1-GDP/GTP exchange factor Rom2p may mediate Rho1p activation (3). Rho1p is a small GTPase upregulated by the GDP/GTP exchange factors Rom1p and Rom2p (46, 48) and downregulated by the GTPase-activating proteins Sac7p and Bem2p (47, 52). Among other functions, Rho1p binds and activates Pkc1p (20, 45), which in turn activates the MAP kinase kinase kinase Bck1p/Slk1p (6, 33), the functionally redundant MAP kinase kinase kinases Mkk1p and Mkk2p (15), and the MAP kinase Slt2p/Mpk1p (32, 58). PKC signaling is constantly guarding cell integrity, and the expression of many cell wall biosynthesis genes requires PKC (14, 65). Nevertheless, a parallel cell integrity signaling mechanism involves the Ykr2p and Ypk1p kinases, since the absence of both of these kinases also leads to cell lysis at elevated temperatures (50).
A previous genome-wide survey of genes whose expression was altered in response to Mpk1p/Slt2p activation indicated that about 20 genes were upregulated (18). This set contained five genes encoding GPI-anchored proteins, at least four of which (Ylr194c, Crh1p, Pst1p, and Cwp1p) are also induced upon loss of Fks1p (57). The PKC pathway is also important for other fungal pathogens, including human commensal pathogens such as Candida albicans or Cryptococcus neoformans. Mkc1p is the C. albicans homologue of Slt2p, and mkc1Δ/mkc1Δ mutant strains display cell surface alterations, an increase in O-glycosylated mannoproteins, hypersensitivity to antifungal agents that inhibit β-1,3-glucan and chitin synthesis (43, 44), as well as reduced virulence in vivo (11). Likewise, PKC signaling mediates response to caspofungin-imposed cell wall perturbations and high temperature in Cryptococcus neoformans (28).
During our efforts to characterize the molecular mechanisms of caspofungin resistance in fungi (54), we noticed that baker's yeast displays much higher tolerance to this new antifungal drug than did the fungal pathogen C. albicans. Hence, we investigated the global response of baker's yeast to caspofungin, exploiting a genome-wide microarray analysis to identify genes or pathways implicated in caspofungin susceptibility. Interestingly, our data show that caspofungin rapidly and specifically activates Slt2p, leading to PKC pathway activation. We also identify the Wsc1p cell surface protein as the dedicated sensor for caspofungin stress and show that the appropriate response to caspofungin-induced cell wall damage requires a functional PKC pathway.
The Saccharomyces cerevisiae strains used in the present study were isogenic derivatives of BY4741 (MATa ura3-Δ0 his3-Δ1 leu2-Δ0 met15-Δ0), all of which were kanamycin cassette deletions from the EUROSCARF knockout collection (http://www.uni-frankfurt.de/fb15/mikro/euroscarf/). Strain DL376 (MATa pkc1Δ::LEU2), a derivative of EG123, was kindly provided by David Levin (18). Unless otherwise indicated, yeast strains were grown routinely at 30°C in YPD medium (1% yeast extract, 2% peptone, 2% glucose). In the case of pkc1Δ, the media were supplemented with d-sorbitol at a final concentration of 1 M. Caspofungin (Merck & Co., Whitehouse Station, N.J.), Congo Red (CR; Sigma, St. Louis, Mo.), and caffeine (Merck & Co., Darmstadt, Germany) were prepared as stock solutions in sterile water and added to the medium at the desired concentrations. Sensitivity phenotypes were assayed with cells grown to the exponential growth phase and diluted to an optical density at 600 nm (OD600) of 0.2. Identical volumes of cultures, as well as 1:10, 1:100, and 1:1,000 serial dilutions, were spotted onto agar plates containing various concentrations of drugs (29). Colony growth was inspected and recorded after a 48-h incubation at 30°C. To record growth curves in liquid culture, overnight cultures of wild-type S. cerevisiae cells (BY4741) were diluted to an OD600 of 0.2 and then grown to an OD600 of 1 in YPD medium at 30°C for additional 2 h to allow cells to adapt to the medium, followed by the addition of caspofungin. OD600 values were recorded in a multilabel counter (Wallac 1420; Perkin-Elmer, Turku, Finland).
Total yeast RNA was isolated exactly as described previously (53). About 20 μg of glyoxal-treated total RNA (51) were separated in a 1% agarose gel and transferred to nylon membranes (Amersham Biosciences, Little Chalfont, Buckinghamshire, England). Northern blots were hybridized with PCR-amplified probes, which were 32P-labeled dCTP radiolabeled by using a MegaPrime labeling kit (Amersham) under the conditions recommended by the manufacturer. Methylene blue staining of rRNA on nylon membranes was used to control for equal RNA loading (29).
Nylon membranes were prehybridized in 10 ml of 10× Denhardt buffer (1 g of Ficoll 400, 1 g of polyvinylpyrrolidone, 1% [wt/vol] bovine serum albumin fraction V), 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 1% sodium dodecyl sulfate (SDS), and 20 μg of salmon sperm DNA/ml for 3 to 6 h at 65°C. The radiolabeled probes were directly added to the prehybridization solution after purification on a NICK column (Amersham) and subsequent heat denaturation. After an overnight incubation at 65°C, membranes were washed at 65°C three times in 2× SSC-1% SDS and three times in 1× SSC-1% SDS and then exposed to X-ray films at −70°C or analyzed with a PhosphorImager (Storm 1840; Molecular Dynamics, Sunnyvale, Calif.).
Fragments of yeast genes used as probes for Northern blots were amplified by PCR with genomic DNA as a template (29). The following PCR program was used for fragment amplification: denaturation at 94°C for 2 min, followed by 35 cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 2 min, with a final extension at 72°C for 10 min. The primers used in the present study included SLT2s (5′-CAT GGA GCA TAC GGC ATA GT-3′), SLT2as (5′-GCT TGT GAA TTG GCA TTT GG-3′), HSP150s (5′-CTA AGA CTA CCG CTG CTG CT-3′), HSP150as (5′-AGC TGG TGC CAT CTA CGC TG-3′), CWP1s (5′-GTC TGT CGC TTT ATT CGC CT-3′), and CWP1as (5′-GGG CCA TTT CAT ATT ACA TTA CGC-3′).
Yeast cells were grown overnight in YPD medium to the mid-logarithmic growth phase at 30°C. The cultures were next diluted to an OD600 of 0.2 and then grown to an OD600 of 1, at which point caspofungin was added to the cultures. Aliquots were harvested after the time intervals indicated in the figures, and cells were collected by centrifugation. Cells were lysed in 250 μl of cold YEX lysis buffer (1.85 M NaOH, 7.5% β-mercaptoethanol), and proteins were precipitated with cold 50% (wt/wt) trichloroacetic acid. Cell extracts corresponding to ca. 5 × 106 cells were separated in SDS-10% polyacrylamide gels and then transferred to nitrocellulose membranes (Protran; Schleicher & Schuell, Dassel, Germany). Phosphorylated Slt2p was detected by using an anti-phospho-p44/42 MAP kinase (Thr202/Tyr204) antibody (Cell Signaling, Beverley, Mass.) at a 1:2,000 dilution to detect dually phosphorylated Slt2p. Total Slt2p was detected with anti-GST-Slt2p antibodies (a generous gift from Humberto Martín, Madrid, Spain) (40) at a 1:1,000 dilution. Polyclonal antibodies to Swi6p (a generous gift from Kim Nasmyth, Vienna, Austria) were used as a loading control. Immunoblots were developed with the Amersham enhanced chemiluminescence detection system under the conditions recommended by the manufacturer.
A 20-ml culture of cells in YPD medium was inoculated from a colony and grown overnight at 30°C and 200 rpm. The culture was diluted until it reached an OD600 of 0.1, and then it was allowed to recover from the stationary phase for 4 h. The culture was split into two halves; one remained untreated, and the other was treated with caspofungin by adding the drug at a concentration of 10 ng/ml. Samples were taken at different time points after drug treatment, followed by preparation of RNA for fluorescence labeling as described below. We used different combinations of cells, such as wild-type untreated and treated cells, at an OD600 of 1 (Fig. (Fig.1B,1B, arrows A2 and B4) or after one (arrows A1 and B1), two (arrows A2 and B2), and 3 h (arrows A3 and B3) caspofungin challenges. Cells were pelleted at 3,000 rpm for 5 min and washed with 1 ml of water, followed by RNA preparation by a routine procedure (53). RNA was quantified by spectrophotometry at 260 nm in Tris-EDTA buffer.
Aliquots containing about 20 μg of total RNA from treated and untreated cells were used for cDNA synthesis with 200 U of Superscript II reverse transcriptase (Invitrogen, Carlsbad, Calif.). Reactions included either the Cy3-dCTP or the Cy5-dCTP kit (Amersham). Labeled cDNAs were pooled, and RNA was destroyed by hydrolysis the samples for 20 min in 50 mM NaOH at 65°C, followed by neutralization with acetic acid, and then cDNA was precipitated with isopropanol. Hybridization to whole-genome cDNA microarrays (Ontario Cancer Institute, Toronto, Ontario, Canada) was done in DigEasyHyb (Roche, Mannheim, Germany) solution at 37°C overnight with 70 μg of salmon sperm DNA/ml as a carrier. Microarrays were washed three times in 1× SSC-0.1% SDS at 50°C, followed by a 1-min wash in 1× SSC at room temperature. Glass slides were spun for 5 min at 500 rpm in a tabletop centrifuge to remove liquid, scanned on an Axon 4000B scanner (Axon Instruments, Inc., Union City, N.J.), and analyzed by using Gene Pix Pro4.1 software (Axon). DNA microarrays and protocols were obtained from the Ontario Cancer Institute. All microarray experiments were carried out with independent RNA preparations. Microarray data were collected and normalized with GenePix Pro4.1, and data filtering was done by using conventional spreadsheets. The Saccharomyces Genome Database at Stanford University (http://www.yeastgenome.org/) was used to retrieve functional annotations of yeast genes. The complete microarray data set is also available as supplementary material.
Caspofungin is a new antifungal drug, which acts fungicidal, since it blocks cell wall synthesis through inhibition of fungal β-1,3-glucan synthases. During our attempts to characterize fungal caspofungin resistance mechanisms (54), we noticed that baker's yeast displayed a much higher caspofungin tolerance than the fungal pathogen C. albicans (54). At sublethal concentrations, yeast cells showed 100% viability on agar plates when treated with 10 ng of caspofungin/ml (Fig. (Fig.1A),1A), whereas viability of C. albicans cells was drastically reduced to ca. 12% of untreated control cells under the same conditions (data not shown). Hence, we pursued a global microarray analysis to identify yeast genes implicated in caspofungin tolerance. To pinpoint the optimal conditions for RNA isolation, wild-type yeast cells were treated with increasing drug concentrations until reduced growth was obvious (Fig. (Fig.1B).1B). Notably, cell growth was severely impaired in liquid culture at 10 ng of caspofungin/ml (Fig. (Fig.1B),1B), whereas the viability appeared to be unaffected, as judged from the ability of cells to form colonies (Fig. (Fig.1A1A).
Since 10 ng of caspofungin/ml caused a strongly reduced growth of baker's yeast, we chose this concentration to extract RNA for global microarray analysis. RNA was then isolated from treated and untreated controls cells at a similar OD600 or from cells in the same growth phases (Fig. (Fig.1B).1B). Several independent experiments were carried out to compare transcriptomes of cells after drug treatment for 1 h (A1 and B1), 2 h (A2 and B2), and 3 h (A3 and B3), as well as at a similar OD600 of 1 (A2 and B4). Total RNA was purified and labeled by incorporation of Cy3- and Cy5-dCTP and then hybridized to whole-genome cDNA microarrays of S. cerevisiae. The representative results from these microarray profiling experiments are represented in Table Table11 and Table Table22 and in the supplementary material. Based on functional annotation (http://www.yeastgenome.org/), the genes whose expression was modulated by caspofungin were classified into several categories.
Caspofungin enhanced transcription of genes required for cell wall biogenesis or maintenance of its architecture. At least one of the three chitin synthase genes, CHS1 (4), as well as five genes encoding GPI-anchored proteins (CWP1, SED1, CRH1, YLR194c, and PST1) were induced. Furthermore, PIR2/HSP150, one out of a total of five PIR family members (Pir1p, Pir2p/Hsp150p, Pir3p, Cis3p, and YJ160p) was also induced (23). A similar induction was observed for YGR189c encoding a homolog of bacterial β-glucanases and eukaryotic endotransglycosidases. Most strikingly, the MAP kinase SLT2/MPK1 and the related gene MLP1 (YKL161c), both of which appear to be induced by active cell integrity signaling (18), were also upregulated after caspofungin challenge. Remarkably, however, none of the genes involved in general stress response through Msn2p was induced (Table (Table11).
Further, caspofungin also caused repression of certain genes involved in cell wall organization and biogenesis (Table (Table2;2; see also Supplementary Material), including the putative WSC4 stress sensor, the SIT4 phosphatase, and calcineurin CNA1, as well as the α-1,6-mannosyltransferase OCH1 and the mannosyltransferase KTR4 genes. Taken together, the results of the mRNA expression profiling demonstrate the activation of target genes of PKC cell integrity pathway (18). Despite the rapid deleterious effect of caspofungin on cell proliferation and cell wall synthesis, other stress-related genes remained unaffected in their expression profile.
The microarray data indicated that members of the PKC pathway increased expression upon caspofungin treatment. Thus, we used Northern blotting to test whether SLT2 expression is controlled by caspofungin in order to verify the microarray data. Indeed, induction of SLT2 mRNA was observed in cells treated with 10 or 30 ng of caspofungin/ml after 30 or 60 min (Fig. (Fig.2A2A).
The activation of the PKC pathway also leads to phosphorylation of Slt2p on Thr190/Tyr192 residues (32). Therefore, we also tested whether caspofungin increased the amount of dually phosphorylated Slt2p, which would indicate active cell integrity signaling. We used a phospho-specific antibody raised against dually phosphorylated p44/42 MAP kinase and a specific anti-Slt2p antibody as described in Material and Methods. Caspofungin triggered a rapid and transient Slt2p phosphorylation and thus activation of Slt2p within 5 min after drug treatment (Fig. (Fig.2B).2B). The amount of activated phospho-Slt2p kinase decreased after about 30 min, which is consistent with stress adaptation. We used CR, a cell wall-damaging drug known to induce Slt2p phosphorylation (10, 24, 41), as a positive control. As expected, no Slt2p was detectable in cells lacking SLT2 (Fig. (Fig.2B).2B). Interestingly, although the microarray data indicated slightly upregulated mRNA levels, corresponding total Slt2p protein levels did not significantly change under activating conditions (19), a result perhaps due to autoregulatory mechanisms (18).
The data indicate that caspofungin specifically triggers PKC pathway signaling, since it induced Slt2p phosphorylation and thus its activation. We therefore tested whether Slt2p is also required for caspofungin tolerance. Wild-type and isogenic slt2Δ cells were diluted and spotted onto plates containing various amounts of caspofungin. Indeed, cells lacking Slt2p displayed a severe growth retardation at low caspofungin doses and were unable to grow at higher concentrations compared to the isogenic wild-type control (Fig. (Fig.3A).3A). CR and caffeine, which are drugs known to inhibit the growth of cells lacking a functional PKC pathway (6, 7), were used as positive controls. Finally, we also tested the growth of slt2Δ cells in liquid culture in the presence of caspofungin (Fig. (Fig.3B).3B). Within the first 6 h after caspofungin addition, no growth differences were detectable between the wild-type and the slt2Δ cells. However, after prolonged treatment, slt2Δ cells showed dramatic growth defects in liquid culture compared to wild-type cells (Fig. (Fig.3B).3B). Taken together, these data demonstrate that the MAP kinase Slt2p is necessary for caspofungin tolerance.
Because Slt2p mediates caspofungin tolerance, we investigated whether other upstream and downstream components of the PKC pathway (Fig. (Fig.4)4) are implicated in caspofungin tolerance. We also asked the question which cell surface sensor mediates sensing caspofungin-induced cell wall damage. Several gene products are known to feed into the PKC pathway by transducing surface signals. Mid2p is required for the induction of Slt2p tyrosine phosphorylation after exposure to high temperature, mating pheromones or Calcofluor White (24, 41). In contrast, Wsc1p/Slg1p/Hcs77p has only been implicated in the sensing of thermal stress (12, 59). Hence, we analyzed levels of phospho-Slt2p in wild-type cells and in isogenic wsc1Δ, wsc2Δ, wsc3Δ, wsc4Δ, and mid2Δ strains, as well as in an lre1Δ strain (data not shown), and we also tested their caspofungin sensitivities. Interestingly, a caspofungin-induced increase in Slt2p phosphorylation, and thus activation was detected in all mutants tested except for the wsc1Δ mutant (Fig. (Fig.5A).5A). All other putative sensor deletion strains tested showed normal caspofungin-induced activation of Slt2p (Fig. (Fig.5A).5A). Notably, levels of phospho-Slt2p were generally lower in wsc4Δ cells, but the caspofungin-mediated activation was still detectable. Conversely, wsc2Δ, wsc3Δ, and rom2Δ mutants, as well as the lre1Δ mutant (data not shown), displayed higher levels of dually phosphorylated Slt2p than did the wild-type strain. Immunoblotting with anti-Slt2p antibodies showed that the Slt2p and Swi6p levels did not change under these conditions, the latter serving as a loading control (Fig. (Fig.5A5A).
Next, we tested caspofungin susceptibilities of wild-type cells and isogenic strains carrying deletions of the PKC pathway genes, including the deletions mid2Δ, rho2Δ, mid2Δ, mkk2Δ, bck1Δ, wsc1Δ, wsc2Δ, wsc3Δ, wsc4Δ, lre1Δ, and slt2Δ. Strains were grown in YPD medium, and identical volumes, as well as 10-fold serial dilutions, were spotted onto agar plates containing the indicated caspofungin concentrations. The growth assays demonstrated that all mutants lacking genes of the PKC pathway were also caspofungin hypersensitive (Fig. (Fig.5B).5B). Among the most sensitive mutants were bck1Δ and slt2Δ cells, as well as pkc1Δ cells (Fig. (Fig.5B).5B). Interestingly, the GDP/GTP exchange factor Rom2p showed higher levels of phospho-Slt2p but was only slightly sensitive to caspofungin in the plate assay (Fig. (Fig.5B).5B). These results indicate that an active PKC pathway is necessary but not sufficient for full caspofungin tolerance. These data establish the importance of the PKC pathway and suggest that Wsc1p senses caspofungin-induced cell wall damage (Fig. (Fig.5B5B).
Caspofungin is a noncompetitive inhibitor of 1,3-β-glucan synthases encoded by FKS1 and FKS2 (30). The FKS1 gene is expressed during vegetative growth, whereas FKS2 is mainly expressed during sporulation (30). We therefore investigated whether cells lacking Fks1p display altered caspofungin sensing. fks1Δ mutant cells showed high levels of phospho-Slt2p even in uninduced cells, suggesting compensatory activity of the PKC pathway. However, in the presence of caspofungin, the level of phospho-Slt2p still increased, suggesting FKS1-independent sensing. Likewise, caspofungin-induced Slt2p phosphorylation was similar to that of the wild type in fks2Δ cells (Fig. (Fig.6A).6A). As expected, mutants lacking FKS1 also showed pronounced caspofungin hypersensitivity in agar plate assays because expression levels of Fks2p under these conditions apparently cannot fully compensate for the loss of Fks1p (Fig. (Fig.6B).6B). Notably, chs3Δ cells were slightly hypersensitive. These results demonstrate that an activated PKC pathway, as well as a functional 1,3-β-glucan synthase, is required for caspofungin tolerance (Fig. (Fig.6B6B).
To identify regulators acting downstream of the PKC pathway, we investigated the Rlm1p and Swi4p transcription factors implicated in the PKC pathway. Rlm1p and Swi4p (SBF) act as downstream effectors of Slt2p/Mpk1p signaling (14, 16, 38, 61). Most Rlm1p-regulated genes encode cell wall proteins or enzymes involved in cell wall biosynthesis (18). The Swi4p and Swi6p proteins form a heterodimeric complex known as SBF, which regulates gene expression during the G1/S transition (27). SBF-activated genes are involved in budding, as well as in membrane and cell wall biosynthesis. Previous studies have shown that the Rlm1p transcription factor, which is activated by Slt2p-dependent phosphorylation, is also mediating SLT2 mRNA induction in response to heat shock (18, 61, 62). To examine the role of Rlm1p and Swi4p in the SLT2 activation upon caspofungin stress, we determined the mRNA levels of SLT2, HSP150, and CWP1 by Northern analysis (Fig. (Fig.7).7). SLT2 and CWP1 expression was induced in wild-type and swi4Δ cells but strongly impaired in the rlm1Δ mutant strain (Fig. (Fig.7).7). Notably, HSP150 mRNA levels were slightly induced by caspofungin after 1 h in wild-type cells but increased significantly in the swi4Δ mutant. However, similar to SLT2 and CWP1, no induction was observed for HSP150 in the rlm1Δ mutant. ACT1 served as a control for RNA loading. Finally, we also tested whether the absence of Rlm1p causes caspofungin hypersensitivity. However, although Slt2p induction requires Rlm1p, rlm1Δ cells failed to display dramatic caspofungin hypersusceptibilities, implying that Slt2p might require the function of other, as-yet-unknown transcription factors to mediate candin tolerance through target genes such as FKS1 or FKS2, since candin tolerance is apparently also controlled by other factors (31). Likewise, agar plate assays demonstrated that cells lacking Swi4p displayed normal caspofungin susceptibility compared to the wild-type control (Fig. (Fig.7B).7B). Taken together, our data establish caspofungin and perhaps other echinocandins as specific activators of cell integrity signaling through the Wsc1p surface sensor and downstream Slt2p kinase.
The cell wall is an essential component of the fungal cell and the prime barrier to the surrounding environment. Modulation of cell wall architecture is required for cell growth, mating, and adaptation to changing environmental conditions. Hence, cell wall composition perhaps undergoes constant dynamic changes controlled by a network of sensors regulating cell wall-modifying enzymes, most of which appear to be controlled by the Slt2p/Mpk1 MAP kinase pathway (26). We show in this study that blocking β-1,3-glucan synthesis by caspofungin rapidly and selectively activates the yeast PKC pathway and that the integrity of this pathway is required for tolerance to caspofungin.
Caspofungin, an echinocandin family member, kills Candida and Cryptococcus spp., as well as Aspergillus spp., by inhibition of the enzymes synthesizing β-1,3-glucan (28, 30, 39, 56). We used microarray profiling to analyze the global response of S. cerevisiae to sublethal but growth-inhibitory doses of caspofungin (Fig. (Fig.1B).1B). Although transcript profiles were generated from distinct growth phases and at different time points, the datasets of all experiments are remarkably similar (Table (Table11 and supplementary material). Among the genes induced by caspofungin, we found a group of genes recently identified to be under the control of the PKC pathway (18). These genes also overlap with those identified to be upregulated by lack of Fks1p, the main yeast 1,3-β-glucan synthase (31, 57). Further, about 20 genes are upregulated in response to overexpression of an activated allele of the MAP kinase kinase Mkk1p (S386P), which is acting upstream of Slt2p (18). Caspofungin also induces five GPI-anchored proteins (PST1, CRH1, SED1, YLR194c, and CWP1) important for cell wall function, as well as chitin synthase Chs1p. This is in good agreement with the genes encoding genes upregulated in fks1Δ mutants (YLR194c, CRH1, PST1 and CWP1) reported earlier (57). Strikingly, caspofungin specifically induces and activates the MAP kinase Slt2p and its related serine/threonine protein kinase Mlp1p, both of which are also induced in strains lacking FKS1 (18, 57). Among genes downregulated by caspofungin, we find RHO4 (Rho small monomeric GTPase), the sensor WSC4, the α-1,6-mannosyltransferase MNN10, and the serine/threonine phosphatase SIT4, the last of which is required for the downregulation of the PKC pathway, as well as the calcium-dependent protein phosphatase calcineurin CNA1.
An exhaustive analysis of transcript profiles of several cell wall biosynthesis mutants also indicated the involvement of stress response pathways regulated by the transcription factors Msn2/4p, Crz1p, and Hsf1p (31). Interestingly, and in contrast to these studies, our results do not indicate expression modulation of any key genes from these stress pathways. Hence, the experimental conditions we used in terms of sublethal drug doses and exponentially growing cells is sufficient to elicit the most specific response, namely, PKC-mediated signaling.
Activation of the PKC pathway by caspofungin is apparent from the phosphorylation status of the MAP kinase Slt2p. We used a combination of an antibody recognizing dually phosphorylated p44/42 MAP kinase (on Thr202/Tyr204) and a specific anti-Slt2 antibody to detect the phosphorylated and active form of the MAP kinase Slt2p. The intracellular activation of MAP kinases requires phosphorylation on conserved Thr and Tyr residues in subdomain VIII (5). As for Slt2p, these residues correspond to Thr190 and Tyr192. Phosphorylation of both sites is essential for MAP kinase activation. Therefore, the amount of dually phosphorylated Slt2p is a direct indicator of PKC/Slt2p pathway activation (32). Our results suggest a rapid response of the PKC pathway to caspofungin-induced damage, since phospho-Slt2p levels peak after only 10 min of caspofungin challenge. The levels of phospho-Slt2p reached a level comparable to a 2-h treatment with CR (Fig. (Fig.2B).2B). The time frame of Slt2p activation is quite similar to the induction by hypo-osmotic shock that occurs within 1 min after stress exposure (8). These fast kinetics suggests that the status of the PKC pathway is tightly and dynamically linked to the activity of the glucan synthase and cell wall integrity or biosynthesis. Furthermore, we also demonstrate that the integrity of the PKC pathway is important for sensing cell wall damage, as well as for the downstream response to caspofungin. Mutants lacking SLT2 are hypersensitive to caspofungin both in plate assays and in liquid culture. In addition, we found that bck1Δ cells lacking the MAP kinase kinase kinase and rom2Δ cells lacking the GDP/GTP exchange factor for the PKC activator Rho1p, as well as pkc1Δ mutants, display marked caspofungin hypersensitivity.
The PKC pathway is also important for other fungal species, such as C. albicans or Cryptococcus neoformans. The C. albicans MAP kinase MKC1 complements the lytic phenotype of the S. cerevisiae slt2 mutants, thus representing a functional homologe of the yeast SLT2 gene. This is further supported by the fact that homozygous C. albicans mkc1Δ/mkc1Δ mutant strains display cell surface alterations and increased sensitivity toward antifungals that inhibit β-1,3-glucan and chitin synthesis (11, 43, 44). Likewise, PKC signaling is involved in response to caspofungin-imposed perturbations of cell wall biosynthesis in Cryptococcus neoformans (28). The Cryptococcus neoformans Mpk1 homologue of yeast Slt2p controls the cellular response to high temperature, as well as challenge by cell wall synthesis inhibitors such as caspofungin (28). A C. albicans mkc1Δ/mkc1Δ strain displays caspofungin hypersensitivity, even though C. albicans cells are already at least 10 times more sensitive to caspofungin than yeast (C. Reinoso-Martin et al., unpublished data). Thus, it appears likely that the response to caspofungin is conserved between S. cerevisiae, C. albicans, and Cryptococcus neoformans, suggesting that the PKC signaling pathway could harbor several potential drug targets for novel antifungals interfering with cell wall function or architecture.
We also analyzed the role of the transcription factors Rlm1p and Swi4p, which are acting downstream of Slt2p (2, 60). The activation of Slt2p by caspofungin also results in Rlm1p-dependent transcription activation, whereas a lack of Swi4p does not reduce the caspofungin-mediated gene regulation, at least for the genes analyzed in the present study. Nevertheless, and somewhat unexpectedly, cells lacking Rlm1p are not caspofungin sensitive (Fig. (Fig.7B).7B). However, this is consistent with the fact that the rlm1Δ mutant cells do not show a cell integrity defect and, more importantly, FKS1 levels do not change upon loss of Rlm1p (18). These findings support our data demonstrating the activation of the PKC pathway by caspofungin but also indicate that certain Rlm1p-dependent target genes are not essential for caspofungin tolerance.
The rapid response of the PKC pathway to caspofungin-induced damage requires a highly active sensing machinery. Among several cell surface factors that could potentially signal through the PKC pathway, we show that only wsc1Δ mutants, but not mutants in other cell wall damage sensor proteins, are hypersensitive to caspofungin. Wsc1p localizes to the plasma membrane and is a member of the WSC family composed of four genes (WSC1 to WSC4) (17, 59). Notably, deletion of either ROM2 or WSC1 leads to a defect of β-1,3-glucan synthesis (55), which is consistent with our data. Furthermore, Wsc1p colocalizes with Fks1p (9), the actual target of caspofungin. The C-terminal cytoplasmic domains of Wsc1p and Mid2p interact with Rom2p, a guanine nucleotide exchange factor for Rho1p (48). However, Mid2p activates Pkc1p without affecting β-1,3-glucan synthesis. Given the role of Wsc1p in the regulation of Fks1p, the caspofungin sensitivity of the wsc1Δ mutant comes not entirely unexpected. However, we find that absence of Wsc1p also prevents phosphorylation and activation of Slt2p. This result supports the notion of a signaling feedback from the cell wall via Wsc1p to monitor the activity of Fks1p. The situation is probably similar to fks1Δ cells, which have a reduced glucan content of the cell wall and compensate for this by increasing the chitin content as well as by the higher levels of the second glucan synthase gene, FKS2 (42, 65). We and others (10) have shown that fks1Δ cells demonstrate increased basal levels of phospho-Slt2p, a finding that further supports the existence of a feedback loop. Our results suggest a highly dynamic connection between Fks1p and Wsc1p and offer exciting possibilities to address the tantalizing question as to how the actual sensing is accomplished by Wsc1p. To address this possible interplay, we constructed a wsc1Δ fks1Δ strain. Interestingly, a wsc1Δ fks1Δ double mutant already exhibits a drastic slow-growth phenotype on YPD medium lacking any drugs (data not shown). These data not only show a genetic interaction between WSC1 and FKS1 but also strongly support the role of Wsc1p in sensing cell wall damage that arises from a loss of glucan synthase. Our data suggest that baker's yeast can sense the presence of caspofungin through rapid activation of the PKC pathway, leading to the induction of a salvage response to maintain cell wall integrity. The results may also explain the inherently higher tolerance of baker's yeast against this drug. Recent experimental data demonstrate that caspofungin also activates the Mkc1p orthologue of Slt2p in C. albicans (data not shown). However, we do not know at this point whether and how C. albicans cells can counteract caspofungin-induced cell wall perturbations.
We thank all laboratory members, and especially Yasmine M. Mamnun, for critical manuscript reading and helpful discussions. Michael Schuster is acknowledged for help with computational analysis. We thank Jeremy Thorner, Kim Nasmyth, Maria Molina, Humberto Martín, David Levin, Javier Arroyo, and Frans Klis for providing strains, plasmids, and antibodies or for stimulating discussions about unpublished data.
This study was supported by grants from the Austrian Science Foundation (P-15934-B08) and the Austrian National Bank (OeNB-9985) to K.K. C.R.-M. is a recipient of an FP5 Marie Curie postdoctoral fellowship from the European Commission.
†The supplemental material for this article may be found at http://ec.asm.org/.