| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Molecular Genetics and Microbiology, Durham, NC 27710, USA
2 Department of Pharmacology and Cancer Biology, Durham, NC 27710, USA
3 Department of Medicine, Durham, NC 27710, USA
4 Duke University Medical Center, Durham, NC 27710, USA
Correspondence
Joseph Heitman
heitm001{at}duke.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
mutations conferred fludioxonil resistance. Fludioxonil treatment caused cell growth inhibition following cell swelling and cytokinesis defects in the sensitive wild-type but not in a hog1 mutant strain, suggesting that Hog1 activation results in morphological cellular defects. Fludioxonil exerted a fungistatic effect on the wild-type strain H99, but exhibited fungicidal activity against calcineurin mutant strains, indicating that the calcineurin pathway contributes to drug resistance in this fungus. Combination of fludioxonil and the calcineurin inhibitor FK506 synergistically inhibited C. neoformans growth. mpk1 MAPK mutant strains exhibited fludioxonil hypersensitivity, indicating that this pathway also contributes to drug resistance. These studies provide evidence that the broad-spectrum antifungal drug fludioxonil exerts its action via activation of the Hog1 MAPK pathway and provide insight into novel targets for synergistic antifungal drug combinations.
A table of primer sequences is available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Cryptococcosis is one of the most common fungal infections diagnosed in AIDS patients, particularly in regions where antifungal drugs such as amphotericin B and fluconazole are not readily available. However, amphotericin B has a number of adverse side effects and fluconazole exhibits only fungistatic activity. Furthermore, mutants resistant to these drugs are emerging in Candida species and Cryptococcus neoformans (Cameron et al., 1993; Paugam et al., 1994; White et al., 1998). Therefore, it has become an important issue to develop new antifungal agents that are fungicidal, less toxic, and employ different mechanisms of action for use in combination drug therapies.
Fludioxonil [4-(2,2-difluoro-1,3-benzodioxol-4-yl)pyrrole-3-carbonitrile] is a phenylpyrrole antifungal drug derived from the antibiotic pyrrolnitrin (Gehmann et al., 1990); it has a broad antifungal spectrum and is now used to control a variety of important plant-pathogenic fungi. It is a unique antifungal drug that targets signal transduction. Several lines of evidence implicate the Hog1 MAP kinase (MAPK) pathway in fludioxonil-mediated antifungal effects. In the filamentous fungus Neurospora crassa, the OS-2 gene, a homologue of the Saccharomyces cerevisiae HOG1 gene, is responsible for adapting to hyperosmotic conditions by accumulating intracellular glycerol, and its deletion causes hyperosmosensitivity and resistance to fludioxonil (Zhang et al., 2002). Exposure of several plant-pathogenic fungi to fludioxonil, including Colletotrichum lagenarium, Botrytis cinerea and Cochliobolus heterostrophus, activates the Hog1 MAPK homologue and inhibits fungal growth (Kojima et al., 2004). In Candida albicans and N. crassa, fludioxonil treatment seems to induce excessive intracellular glycerol accumulation (Fujimura et al., 2000a; Ochiai et al., 2002; Zhang et al., 2002).
Tacrolimus (FK506) and cyclosporin A (CsA) were both originally isolated as immunosuppressive agents (Borel, 1976; Kino et al., 1987) and also exhibit potential antifungal activity (Odom et al., 1997a; Perfect & Durack, 1985). These drugs exert their toxic effect by binding to target immunophilins and subsequently inhibiting calcineurin activity (Cruz et al., 2000; Odom et al., 1997a). In Cryptococcus neoformans, inhibition of calcineurin by FK506 causes a growth defect at high temperature. Similar to drug inhibition, mutation of the calcineurin A (CNA1) or calcineurin B (CNB1) genes renders cells inviable at high temperature (Fox et al., 2001; Odom et al., 1997b) and severely attenuates virulence of C. neoformans. Furthermore, the calcineurin pathway is involved in mating and cell wall integrity of C. neoformans (Cruz et al., 2001; Kraus et al., 2003).
Combination treatment with multiple drugs can exert a more potent antifungal activity than treatment with a single drug. For instance, FK506 exhibits synergistic antifungal action against C. neoformans with several other drugs in both calcineurin-dependent and -independent manners. Inhibition of calcineurin by FK506 is required for its synergy with the echinocandins, but is not required for FK506-mediated synergy with fluconazole (Del Poeta et al., 2000), indicating another effect of FK506 on cells in addition to calcineurin inhibition. Similarly, the toxic action of fludioxonil may be exerted via other signal transduction pathways in addition to the Hog1 MAPK pathway. In N. crassa, a cAMP- and calcium-independent protein kinase is inhibited by phenylpyrroles, including fludioxonil (Pillonel & Mayer, 1997). On the other hand, in Ustilago maydis, mutants lacking cAMP-dependent protein kinase A (PKA) exhibit increased resistance to dicarboximide fungicides, although these fungicides do not directly inhibit PKA (Orth et al., 1995; Ramesh et al., 2001). Interestingly, a Colletotrichum lagenarium mutant lacking the PKA regulatory subunit shows significantly increased sensitivity to fludioxonil (Kojima et al., 2004). These results indicate that these signalling pathways could also be useful targets for designing novel antifungal agents. Therefore, elucidating the relationships between drugs and their target signal transduction pathways may provide the foundation for new combination drug therapies and less toxic therapies.
In this study, we demonstrate that the Hog1 MAPK pathway promotes sensitivity to fludioxonil in C. neoformans, whereas the calcineurin and Mpk1 MAPK pathways mediate resistance to fludioxonil. Furthermore, we provide evidence that simultaneous perturbation of the Hog1 and calcineurin pathways by combined treatment with fludioxonil and FK506 inhibits the growth of the pathogen even more effectively. Biochemical and genetic data demonstrate that fludioxonil exerts its cytotoxic effects through hyperactivation of the Hog1 pathway and that cell wall integrity plays an important role in maintaining cell viability in the presence of fludioxonil.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Among clinical serotype A strains, IN-38 is from India; UG-20020 is from Uganda. Other serotype A and D clinical or environmental strains were described previously (Bahn et al., 2005). AD hybrid strains were previously generated by crossing the serotype D strain JEC171 (MATa ade2 lys2) and a 5-fluoroorotic acid (5-FOA)-resistant mutant of the serotype A strain H99 (MAT
ura5) (Lengeler et al., 2001). The S. cerevisiae hog1/hog1 mutant (from the diploid homozygous deletion mutant collection) and its parental strain BY4743 (diploid of strains BY4741/BY4742; Giaever et al., 2002) were also used. YPD medium contained 2 % (v/v) glucose, 2 % (w/v) Bacto Peptone (Difco), and 1 % (w/v) yeast extract (Difco). YPD agar plates also contained 2 % (w/v) Bacto Agar (Difco). The top agar used in these assays was 0·7 % Bacto Agar (Difco) in water.
|
Disruption of the CNA1, CNB1, MPK1, MKK1 and BCK1 genes.
The CNA1, CNB1, MPK1, MKK1 and BCK1 genes were disrupted by biolistic transformation in C. neoformans serotype A strain H99 with alleles generated by PCR overlap as previously described (Davidson et al., 2002). The 5' and 3' regions of these genes were PCR-amplified with the following primers: 12339/12409 and 12408/12336 for the 5' and 3' regions of the CNA1 gene respectively, 12347/12412 and 12410/12344 for the CNB1 gene, 13460/13461 and 13462/13463 for the MPK1 gene, 14419/14420 and 14421/14422 for the MKK1 gene, and 13979/13980 and 13981/13982 for the BCK1 gene (see supplementary Table S1, available with the online version of this paper, for the primer sequences). M13 forward (M13F) and reverse (M13R) primers were used to generate the Natr (nourseothricin acetyltransferase) dominant selectable marker with plasmids pNATSTM with a unique sequence tag (#117 for the CNA1 gene, #122 for the CNB1 gene, #150 for the MPK1 gene, #224 for the MKK1 gene, #43 for the BCK1 gene, kindly provided by Dr Jennifer K. Lodge, Saint Louis University School of Medicine; tag sequence is available upon request) as templates (Hensel et al., 1995). The disruption alleles of these genes were generated by PCR overlap using primers 12339/12336 for the CNA1 gene, 12347/12344 for the CNB1 gene, 13460/13463 for the MPK1 gene, 14419/14422 for the MKK1 gene, and 13979/13982 for the BCK1 gene.
The gel-extracted disruption cassettes were precipitated onto 600 µg of gold microcarrier beads (0·8 µm, bioWORLD) and biolistically transformed into the serotype A strain H99. Stable transformants were selected on YPD medium containing nourseothricin (100 mg l–1). The cna1, cnb1, mpk1, mkk1 and bck1 mutant strains were screened by diagnostic PCR for the 5' and 3' junctions and by Southern blot analysis using specific probes generated by PCR with primers 12343/12342 for the CNA1 gene, 12351/12350 for the CNB1 gene, 13466/13467 for the MPK1 gene, 14424/14425 for the MKK1 gene and 13979/13980 for the BCK1 gene.
To construct the serotype A cna1 hog1 strain, M13F and M13R primers were used to generate the Neor (neomycin phosphotransferase II) dominant selectable markers with template pJAF1 (Fraser et al., 2003). The cna1 : : NEO disruption allele was generated by PCR overlap with primers 12339/12336 and biolistically transformed into the hog1 mutant strain (YSB64). Stable transformants were selected on YPD agar containing both nourseothricin (100 mg l–1) and G418 (200 mg l–1). Positive cna1 hog1 strains were screened by diagnostic PCR for the 5' and 3' junctions and further confirmed by Southern blot analysis (not shown).
To construct the serotype A cna1+CNA1 complemented strain, H99 genomic DNA containing the full-length CNA1 gene was isolated from a C. neoformans H99 bacterial artificial chromosome (BAC) library by colony hybridization with a CNA1 gene-specific probe amplified by PCR. The 5·7 kb BglII–BamHI fragment containing the full-length CNA1 gene was subcloned into pJAF12 (Neor), generating plasmid pNEOCNA1A. NotI-digested linearized pNEOCNA1A DNA was introduced by biolistic transformation into strain KK1 (Table 1
) and integrated at the genomic CNA1 locus. Integration of plasmid pNEOCNA1A was confirmed by Southern blotting.
Antifungal drug activity tests.
For fludioxonil sensitivity testing on agar media, each strain was incubated overnight in liquid YPD medium at 30 °C and subcultured in fresh liquid YPD medium to OD600 0·7–0·9. Then cells were serially diluted (101–105) in dH2O, and spotted (2 µl) onto YPD agar containing 1 or 10 µg fludioxonil ml–1, dissolved in DMSO.
The MICs of antifungal drugs for C. neoformans strains were assessed in accordance with the National Committee for Clinical Laboratory Standards (NCCLS) criteria for antifungal drug activity as previously described (Cruz et al., 2002). Briefly, In vitro testing was performed in RPMI 1640 medium containing L-glutamine without sodium bicarbonate (Gibco), buffered to pH 7·0 with 0·165 M MOPS using 96-well plates (96-well cell culture cluster, flat-bottom; Costar). The fludioxonil concentrations tested were 10, 5, 1, 0·5, 0·1, 0·02, 0·01 and 0·005 µg ml–1. Cells were incubated at 30 °C without shaking, and after 72 h, the OD600 was determined with a microtitre plate reader (Molecular Devices Thermomax). The MIC80 of fludioxonil was defined as the drug concentration at which 80 % reduction in optical density is observed compared with the no-drug control. Minimum fungicidal concentrations (MFCs) were determined by sampling 100 µl from each well that showed growth inhibition in the MIC assay and incubation on YPD medium at 30 °C for 72 h. The lowest concentration that yielded three or fewer colonies was recorded as the MFC.
The fractional inhibitory concentrations (FICs) and indexes were calculated as previously described (Cruz et al., 2002). The FIC index is the sum of the FICs for each drug, which in turn is defined as the MIC80 of each drug when used in combination divided by the MIC80 of the drug when used alone. Based on the FIC index, drug interactions were classified as synergistic (FIC<1·0), additive (FIC=1·0), autonomous (FIC between 1 and 2) or antagonistic (FIC>2). In the MFC and FIC experiments, FK506 was tested at 2, 0·5, 0·1 and 0·04 µg ml–1 as the final drug concentrations. For calculation purposes, MICs of <5·0, >2·0, 
0·5 and 0·04 were assumed to be 5·0, 2·0, 0·5 and 0·04 respectively.
Western blot analysis of Hog1 phosphorylation.
Yeast cells were grown to mid-exponential phase as described above, added to an equal volume of YPD medium containing 2 M NaCl, 2 µg or 20 µg fludioxonil ml–1 (final 1 M NaCl, 1 µg or 10 µg fludioxonil ml–1), and further incubated for the indicated amount of time. Protein extraction and measurement of concentration were performed as described previously (Bahn et al., 2005). An equal amount of protein (20 µg) was loaded into a 10 % Tris/glycine gel (Novex). Separated proteins were further transferred to Immunoblot PVDF membrane (Bio-Rad) and incubated at 4 °C overnight with a rabbit anti-dually phosphorylated p38 antibody (Cell Signalling) and for 1 h with a secondary anti-rabbit IgG horseradish-peroxidase-conjugated antibody. The blot was developed using the ECL Western Blotting Detection System (Amersham Bioscience). Subsequently, the blot was stripped and further used for detection of Hog1 with a rabbit polyclonal anti-Hog1 antibody (Santa Cruz Biotechnology) as loading control.
Microscopy.
Strains were inoculated into liquid YPD medium and grown overnight at 30 °C. The cells were washed with sterile dH2O and reincubated in RPMI 1640 medium containing fludioxonil for 48 h. The cells were resuspended in dH2O, and observed with a Zeiss Axioskop 2 equipped with an AxioCam MRM digital camera.
Intracellular glycerol measurement assay.
To measure the intracellular glycerol content, each strain was grown to mid-exponential phase, added to an equal volume of YPD medium containing 20 µg fludioxonil ml–1 (final 10 µg fludioxonil ml–1), and further incubated for the indicated amount of time. After incubation, cells were collected by centrifugation, washed with dH2O and then extracted by boiling at 100 °C for 10 min. After centrifugation, the glycerol concentration in the supernatant was measured using a UV-glycerol assay kit according to the manufacturer's instructions (R-Biopharm). The same amount of each sample was lyophilized to measure cellular dry weight for normalization.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). To elucidate the role of the HOG pathway in fludioxonil sensitivity, we tested the sensitivity of hog1 and pbs2 mutants that had been constructed before (Bahn et al., 2005). Both mutants exhibited complete resistance to fludioxonil, indicating that the Hog1 pathway is involved in fludioxonil sensitivity of C. neoformans (Fig. 1A). To examine whether phosphorylation and kinase activity of Hog1 MAPK are required to confer fludioxonil sensitivity, we tested the sensitivity of cells expressing site-directed mutants of Hog1 at the phosphorylation sites (hog1+HOG1T171A+Y173A) or the catalytic site (hog1+HOG1K49S+K50N) (Fig. 1A). These Hog1 mutants were as resistant to fludioxonil as the hog1 mutant, indicating that Pbs2-dependent phosphorylation and catalytic activation of the Hog1 MAPK are prerequisites for fludioxonil sensitivity (Fig. 1A).
|
). Given the role of the HOG pathway in fludioxonil sensitivity of C. neoformans, we tested whether calcineurin is also involved in resistance to fludioxonil. For this purpose, we deleted the genes encoding the calcineurin catalytic (CNA1) or regulatory subunit (CNB1) in the H99 background with dominant selectable markers. The cna1 and cnb1 mutants exhibited hypersensitivity to fludioxonil, indicating that calcineurin promotes resistance to fludioxonil in C. neoformans (Fig. 1A). To further examine this hypothesis, we tested whether a synergistic fungicidal effect would be observed with concomitant exposure to fludioxonil and the calcineurin inhibitor FK506. In fact, FK506 greatly enhanced growth inhibition when combined with fludioxonil, but had no effect on cell growth by itself (Fig. 1A), strongly suggesting a synergistic fungicidal effect between the two drugs. Interestingly, a cna1 hog1 double mutant still exhibited complete resistance to fludioxonil (Fig. 1B), indicating that calcineurin-dependent fludioxonil resistance is also mediated directly or indirectly by the Hog1 MAPK.
To quantitatively measure fludioxonil sensitivity, we performed drug susceptibility assays according to NCCLS criteria using a range of fludioxonil concentrations (5 ng ml–1 to 10 µg ml–1). In this assay, the MIC80 of fludioxonil for the WT strain was <5 µg ml–1 whereas the MIC80 for the cna1 mutant was <100 ng ml–1 (Fig. 1B, Table 2). In contrast, hog1 and cna1 hog1 mutants exhibited a modest reduction of growth, but still showed robust resistance, even with 10 µg fludioxonil ml–1 (Fig. 1B). Taken together, these findings indicate that sensitivity of C. neoformans to fludioxonil is oppositely regulated by the HOG and calcineurin pathways.
|
). To confirm that calcineurin was the target of the observed drug synergy with FK506, a cna1 mutant strain was also tested. When disks containing 10, 50 or 100 µg fludioxonil were placed over the cna1 strain, we observed large haloes similar to those of the wild-type strain exposed to fludioxonil in combination with FK506 (Fig. 1C). Fludioxonil and FK506 did not produce any haloes on the hog1 strain, which is consistent with the result that the hog1 mutant was resistant to medium containing fludioxonil and FK506 (Fig. 1A, C). Additionally, the fractional inhibitory concentration (FIC) was calculated to determine the FIC index, of which a value <1·0 denotes a synergistic interaction. The calculated FIC index of fludioxonil is 0·12 with FK506, denoting a synergistic relationship between fludioxonil and FK506 (Table 2). These results indicate that FK506 participates in drug synergy with fludioxonil by inhibiting the calcineurin pathway.
To determine whether fludioxonil is fungicidal or fungistatic to C. neoformans, minimal fungicidal concentrations (MFCs) were investigated in accordance with the NCCLS criteria (Table 3). Although fludioxonil dramatically inhibited growth of the WT strain in liquid medium (Fig. 1B), 10 µg fludioxonil ml–1 did not produce an MFC against the WT strain, indicating that fludioxonil at <10 µg ml–1 is not fungicidal against C. neoformans (Table 3). On the other hand, when fludioxonil was tested in combination with FK506, the MFC of fludioxonil was 0·5 µg ml–1, indicating that the combination treatment of fludioxonil with FK506 has a fungicidal effect on the WT strain (Table 3). The MFC of fludioxonil for the cna1 mutant was 0·5 µg ml–1, which is consistent with the MFC of fludioxonil in combination with FK506 against the WT strain.
|
). Therefore, we hypothesized that fludioxonil sensitivity could also be differentially regulated between the two strains if it is controlled by the HOG pathway. In fact, the serotype D strain JEC21 exhibited complete resistance to fludioxonil, similar to the serotype A hog1 mutant; this is in stark contrast to the sensitivity of the WT serotype A strain H99 (Fig. 2). S. cerevisiae exhibits a similar Hog1 regulatory pattern to the JEC21 C. neoformans strain (Bahn et al., 2005) and was also completely resistant to fludioxonil (Fig. 2).
|
). However, we have recently shown that Hog1 is constitutively phosphorylated under normal conditions and dephosphorylated in response to osmotic shock in the C. neoformans serotype A strain H99 (Bahn et al., 2005). We therefore examined how Hog1 is regulated in response to fludioxonil in C. neoformans. For this purpose, Hog1 phosphorylation patterns were monitored in response to fludioxonil by Western blot analysis. When the serotype A strain H99 was exposed to 1 or 10 µg fludioxonil ml–1, Hog1 was dephosphorylated within 15 min and its dephosphorylation status was maintained for 3 h (Fig. 3). This regulatory pattern is quite similar to that of Hog1 in the H99 strain under osmotic stress (Fig. 3), indicating that fludioxonil activates the Hog1 pathway in WT H99 cells. On the other hand, in the serotype D strain JEC21 Hog1 was only slightly phosphorylated under normal conditions, only minimally further phosphorylated if at all after 15 min of exposure to fludioxonil, and subsequently maintained in an unphosphorylated state for up to 3 h (Fig. 3). In contrast, Hog1 was rapidly phosphorylated in response to osmotic shock in strain JEC21, as reported previously (Bahn et al., 2005). Therefore, we conclude that Hog1 is rapidly activated by dephosphorylation in response to fludioxonil in the drug-sensitive H99 strain, whereas Hog1 is only minimally activated, if at all, in the presence of fludioxonil in the resistant serotype D strain JEC21. Hog1 was also found to be completely inactive during exposure to fludioxonil in S. cerevisiae, which is also resistant to fludioxonil (Fig. 3).
|
mutant that is hypersensitive to fludioxonil (Fig. 1A–C). To address this, we monitored the Hog1 phosphorylation pattern of the cna1 mutant in response to fludioxonil. The Hog1 phosphorylation pattern in the cna1 mutant exposed to 1 µg or 10 µg fludioxonil ml–1 was almost identical to that observed in the wild-type strain in response to fludioxonil (Fig. 3). The data indicate that the calcineurin pathway promotes fludioxonil resistance, but does not directly regulate Hog1 phosphorylation or activation.
A question remains whether the differential fludioxonil sensitivity observed between the serotype A strain H99 and the serotype D strain JEC21 results from serotype- or strain-specific differences. Previously, we have shown that the H99-type Hog1 phosphorylation pattern predominates in a majority of C. neoformans strains regardless of serotype (Bahn et al., 2005). Therefore, we investigated fludioxonil sensitivity in multiple serotype A and D clinical and environmental strains, and monitored the Hog1 phosphorylation pattern after a 1 h exposure to fludioxonil. A majority of C. neoformans strains (8 of 10 serotype A and 6 of 9 serotype D strains) were found to be sensitive to fludioxonil, and in these strains Hog1 was regulated in a manner similar to that of the H99 strain (Fig. 4A, B). Two serotype A strains (IN-38 and UG-20020), and three serotype D strains (NIH433, JEC21, MMRL757) exhibited clear resistance to fludioxonil (Fig. 4A, B). In the resistant strains, the Hog1 phosphorylation signal was almost undetectable under normal conditions, and this dephosphorylated state persisted after 1 h incubation with fludioxonil, indicating that Hog1 is not activated in response to fludioxonil. Taken together, these data demonstrate that fludioxonil exerts a fungicidal effect via activation of the Hog1 pathway in a majority of C. neoformans strains.
|
). As expected, the parental control serotype A H99 (ura5) and serotype D JEC171 (ade2 lys2) strains exhibited sensitivity and resistance to fludioxonil, respectively (Fig. 5). All of 12 independently derived AD hybrid strains exhibited resistance similar to the parental strain JEC171, indicating that fludioxonil sensitivity is a recessive phenotype (Fig. 5 and data not shown).
|
A-b). Although the cna1 mutant strain exhibited a cytokinesis defect without fludioxonil, when this mutant strain was treated with fludioxonil, the cells exhibited an even more severe cytokinesis defect (Fig. 6A-c). On the other hand, a majority of hog1 mutant cells exhibited no swollen morphology or defects in cell division (Fig. 6A-d). As expected, the morphology of JEC21 cells was not affected by fludioxonil (data not shown). These results demonstrate that fludioxonil-mediated cell growth inhibition is accompanied by defects in cell morphology and cell cycle that are dependent on integrity of the Hog1 pathway.
|
). Therefore, we hypothesized that fludioxonil treatment hyperactivates the HOG pathway to stimulate excessive glycerol accumulation, which may cause morphological defects, including cell swelling. To test this hypothesis, we measured the glycerol content in C. neoformans after fludioxonil treatment for 1 and 3 h (Fig. 6B). In the WT strain H99, the glycerol content increased after 1 h exposure, and to an even greater extent after 3 h incubation. In contrast, no increase in glycerol was observed in the hog1 mutant compared to the WT. These results indicate that fludioxonil treatment hyperactivates the Hog1 osmotic response pathway, which results in overaccumulation of intracellular glycerol. Increased intracellular glycerol levels may trigger non-physiological levels of water influx into the cell, resulting in cell swelling and growth inhibition. In the cna1 mutant, intracellular glycerol content increased following 1 h treatment with fludioxonil but accumulation levels at 3 h were lower than those of the WT (Fig. 6B). Thus, the cna1 mutant does not maintain intracellular glycerol levels similar to the WT strain, and may release glycerol to the extracellular environment, possibly due to impaired cell wall integrity. Alternatively, the cna1 mutant cells could be rapidly killed by Hog1 activation prior to accumulating glycerol, because fludioxonil has a fungicidal effect on the cna1 mutant (Table 2). As expected, the resistant strain JEC21 accumulated little or no glycerol after treatment with fludioxonil compared to the H99 WT or cna1 mutant strains (Fig. 6B), further supporting our finding that Hog1 is not activated in strain JEC21 in response to fludioxonil (Fig. 3).
Cell wall integrity is required for resistance to fludioxonil
To test whether general defects in cell wall integrity result in hypersensitivity to fludioxonil, we examined the fludioxonil sensitivity of a mutant lacking the MPK1 MAPK gene, which is also known to regulate cell wall integrity in C. neoformans (Kraus et al., 2003) In accord with this hypothesis, the mpk1 mutant exhibited a growth defect at 37 °C (data not shown) and hypersensitivity to fludioxonil similar to that of the cna1 mutant (Fig. 7). In addition, C. neoformans mutants lacking the highly conserved MKK1 and BCK1 genes, which encode a MAPK kinase (MAPKK) and a MAPKK kinase (MAPKKK), respectively, which function upstream of the Mpk1 MAPK, also showed hypersensitivity to fludioxonil (Fig. 7). Interestingly, supplementation with 1 M sorbitol as an osmotic stabilizer partially rescued the growth defect of the mpk1, mkk1 and bck1 mutants in response to fludioxonil treatment. These results further support models in which cell wall integrity promotes cell viability in the presence of fludioxonil.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Here, we propose a model in which C. neoformans sensitivity to fludioxonil is not only positively controlled by the HOG pathway, but also negatively controlled by the calcineurin and Mpk1 MAPK pathways, which are involved in maintaining cell wall integrity (Fig. 8). In this model, the Hog1 MAPK in a majority of serotype A and D C. neoformans strains is activated by dephosphorylation in response to fludioxonil treatment, which in turn causes a rapid accumulation of intracellular glycerol, cell swelling, cytokinesis defects and cell growth inhibition. Several lines of evidence show that the Hog1 MAPK pathway plays a central role in promoting fludioxonil sensitivity in C. neoformans. First, hog1 and pbs2 mutants exhibited strong resistance to fludioxonil. Second, strain-specific variation in fludioxonil sensitivity in different C. neoformans strains is completely correlated with their Hog1 MAPK phosphorylation patterns. Independently, the calcineurin and the Mpk1 MAPK pathways promote resistance to fludioxonil via the cell wall integrity pathway (Fig. 8).
|
). However, we speculate that glycerol accumulation is not the only factor contributing to fungal toxicity exerted by fludioxonil. In N. crassa, a cut mutant, which was deficient in glycerol synthesis, remained sensitive to fludioxonil (Fujimura et al., 2000b). We found that some WT cells, but not hog1 mutant cells, of C. neoformans, were attached to each other following fludioxonil exposure, indicating that normal cytokinesis is also affected by fludioxonil treatment via the Hog1 pathway. This cell–cell attachment was not attributable to cell aggregation after cell division because even sonication could not separate these avidly attached cells (data not shown).
Another novel finding in this study is that the cell wall integrity pathways, including the calcineurin and Mpk1 MAPK pathways, promote resistance to fludioxonil treatment. These pathways appear to contribute to fludioxonil resistance independent of Hog1 MAPK regulation because the cna1 mutation did not significantly change Hog1 phosphorylation patterns. In the model yeast S. cerevisiae, however, perturbation of the HOG pathway partially suppresses cell wall integrity defects via the SVG (STE vegetative growth) pathway that works in parallel with the Mpk1 pathway (Lee & Elion, 1999). In this process, the HOG pathway seems to negatively regulate cell wall integrity through the SVG pathway, antagonizing the Mpk1 MAPK pathway. Therefore, it is possible that overactivation of the HOG1 pathway by fludioxonil treatment impairs cell wall integrity in C. neoformans, resulting in a cytokinesis defect (Fig. 8). Furthermore, fludioxonil hypersensitivity of the mpk1 mutant was partially suppressed by osmostabilization or incubation at low temperature (Fig. 7 and data not shown), suggesting that fludioxonil affects cell wall integrity in fungi. On the other hand, we speculate that fludioxonil treatment increases internal glycerol accumulation by Hog1 activation, which generates osmotic pressure to decrease cell wall integrity, which can be further affected by defects in the calcineurin or Mpk1 pathway (Fig. 8). Furthermore, decreased cell-wall integrity by fludioxonil might be involved in the cytokinesis defects. In S. cerevisiae, cytokinesis defects are observed in mutants deleted for genes encoding PIR (protein with internal repeat) family members, which are covalently attached to 1,3-
-glucan in the cell wall (Mrsa et al., 1997; Mrsa & Tanner, 1999). These mutants also exhibit decreased cell wall integrity and a defect in mother–daughter cell separation (Teparic et al., 2004).
The data from this study lead us to propose a unique way of treating cryptococcosis by simultaneously controlling two independent signalling pathways, the Hog1 MAPK and calcineurin pathways. Potential application of a combination drug therapy for treatment of human fungal pathogens has been proposed by others. The immunophilin-targeting drugs, FK506 and cyclosporin A, which are inhibitors of calcineurin, exhibit antifungal activity against several human fungal pathogens, such as Candida albicans, Cryptococcus neoformans and Aspergillus fumigatus (Cruz et al., 2001; High, 1994; Odom et al., 1997a). Combinations of a calcineurin inhibitor with an inhibitor of 1,3--glucan synthase or an inhibitor of ergosterol exhibit dramatic synergistic antifungal activity against these pathogens in vitro (Cruz et al., 2002; Del Poeta et al., 2000; Onyewu et al., 2003; Steinbach et al., 2004). In this study, we discovered a novel drug combination of fludioxonil and a calcineurin inhibitor that exhibited synergistic fungicidal activity against C. neoformans, in contrast to the fungistatic activity by fludioxonil alone. In plant-pathogenic fungi such as Magnaporthe grisea and Botrytis cinerea, a calcineurin inhibitor prevents development of infection-specific structures (Viaud et al., 2003, 2002) and fludioxonil inhibits the growth of these fungi (Gehmann et al., 1990), suggesting the possibility of a synergistic lethal impact on plant-pathogenic fungi by combining these drugs for agricultural use.
The protein target of fludioxonil remains to be elucidated. Notably, C. neoformans seems to contain an evolutionarily conserved phosphorelay system upstream of the Pbs2-Hog1 MAPK pathway. In contrast to S. cerevisiae, which contains only a single sensor kinase (Sln1), C. neoformans has seven putative two-component histidine-kinase-like proteins. The functions of these sensor proteins have not as yet been studied. Heterologous expression of M. grisea Hik1, a member of the Os-1 family of the two-component histidine kinases, can confer fludioxonil sensitivity to S. cerevisiae (Motoyama et al., 2005). These observations indicate that Hik1 could be a direct fludioxonil target or a signalling mediator for its fungicidal action that activates the Hog1 pathway. In N. crassa, however, an intracellular cAMP and calcium-independent protein kinase binds fenpicolonil, which is a related phenylpyrrole fungicide (Pillonel & Mayer, 1997), raising the possibility that the phenylpyrrole class of drug could directly bind to targets in pathways other than the HOG pathway. Further characterization of upstream factors in the C. neoformans Hog1 MAPK, and of other signalling pathways that are related to fludioxonil sensitivity, may lead to the identification of the direct target of fludioxonil. At this point, we cannot exclude the possibility that signalling cross-talk may exist between the Hog1 pathway and other signalling pathways, and may play an important role in sensing and mediating antifungal action in C. neoformans. Furthermore, our data demonstrate that fludioxonil sensitivity is a recessive phenotype although the majority of C. neoformans strains are drug-sensitive. We speculate that during evolution C. neoformans might have lost either precise upstream regulation or downstream feedback control of Hog1, resulting in constitutive phosphorylation of the MAPK and contributing to fludioxonil sensitivity. Such evolutionary events may have provided selective benefits to the pathogen for host infection, such as high stress resistance. In conclusion, our study proposes new target pathways for the development of therapies for human fungal infections and expands options for the utility of existing antifungal drug classes, such as calcineurin inhibitors, by combination therapy with fludioxonil to exert synergistic antifungal effects.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bahn, Y. S., Kojima, K., Cox, G. M. & Heitman, J. (2005). Specialization of the HOG pathway and its impact on differentiation and virulence of Cryptococcus neoformans. Mol Biol Cell 16, 2285–2300.
Borel, J. F. (1976). Comparative study of in vitro and in vivo drug effects on cell-mediated cytotoxicity. Immunology 31, 631–641.[Medline]
Cameron, M. L., Schell, W. A., Bruch, S., Bartlett, J. A., Waskin, H. A. & Perfect, J. R. (1993). Correlation of in vitro fluconazole resistance of Candida isolates in relation to therapy and symptoms of individuals seropositive for human immunodeficiency virus type 1. Antimicrob Agents Chemother 37, 2449–2453.
Cruz, M. C., Del Poeta, M., Wang, P. & 7 other authors (2000). Immunosuppressive and nonimmunosuppressive cyclosporine analogs are toxic to the opportunistic fungal pathogen Cryptococcus neoformans via cyclophilin-dependent inhibition of calcineurin. Antimicrob Agents Chemother 44, 143–149.
Cruz, M. C., Fox, D. S. & Heitman, J. (2001). Calcineurin is required for hyphal elongation during mating and haploid fruiting in Cryptococcus neoformans. EMBO J 20, 1020–1032.[CrossRef][Medline]
Cruz, M. C., Goldstein, A. L., Blankenship, J. R., Del Poeta, M., Davis, D., Cardenas, M. E., Perfect, J. R., McCusker, J. H. & Heitman, J. (2002). Calcineurin is essential for survival during membrane stress in Candida albicans. EMBO J 21, 546–559.[CrossRef][Medline]
Davidson, R. C., Blankenship, J. R., Kraus, P. R., de Jesus Berrios, M., Hull, C. M., D'Souza, C., Wang, P. & Heitman, J. (2002). A PCR-based strategy to generate integrative targeting alleles with large regions of homology. Microbiology 148, 2607–2615.
Del Poeta, M., Cruz, M. C., Cardenas, M. E., Perfect, J. R. & Heitman, J. (2000). Synergistic antifungal activities of bafilomycin A(1), fluconazole, and the pneumocandin MK-0991/caspofungin acetate (L-743,873) with calcineurin inhibitors FK506 and L-685,818 against Cryptococcus neoformans. Antimicrob Agents Chemother 44, 739–746.
Fox, D. S., Cruz, M. C., Sia, R. A., Ke, H., Cox, G. M., Cardenas, M. E. & Heitman, J. (2001). Calcineurin regulatory subunit is essential for virulence and mediates interactions with FKBP12-FK506 in Cryptococcus neoformans. Mol Microbiol 39, 835–849.[CrossRef][Medline]
Fraser, J. A., Subaran, R. L., Nichols, C. B. & Heitman, J. (2003). Recapitulation of the sexual cycle of the primary fungal pathogen Cryptococcus neoformans var. gattii: implications for an outbreak on Vancouver Island, Canada. Eukaryot Cell 2, 1036–1045.
Fujimura, M., Ochiai, N., Ichiishi, A., Usami, R., Horikoshi, K. & Yamaguchi, I. (2000a). Fungicide resistance and osmotic stress sensitivity in os mutants of Neurospora crassa. Pestic Biochem Physiol 67, 125–133.[CrossRef]
Fujimura, M., Ochiai, N., Ichiishi, A., Usami, R., Horikoshi, K. & Yamaguchi, I. (2000b). Sensitivity to phenylpyrrole fungicides and abnormal glycerol accumulation in os and cut mutant strains of Neurospora crassa. J Pestic Sci 25, 31–36.
Gehmann, K., Nyfeler, R., Leadbeater, A. J., Nevill, D. & Sozzi, D. (1990). CGA 173506: a new phenylpyrrole fungicide for broad-spectrum disease control. Brighton Crop Prot Conf Pests Dis 2, 399–406.
Giaever, G., Chu, A. M., Ni, L. & 71 other authors (2002). Functional profiling of the Saccharomyces cerevisiae genome. Nature 418, 387–391.[CrossRef][Medline]
Heitman, J., Allen, B., Alspaugh, J. A. & Kwon-Chung, K. J. (1999). On the origins of congenic MAT and MATa strains of the pathogenic yeast Cryptococcus neoformans. Fungal Genet Biol 28, 1–5.[CrossRef][Medline]
Hensel, M., Shea, J. E., Gleeson, C., Jones, M. D., Dalton, E. & Holden, D. W. (1995). Simultaneous identification of bacterial virulence genes by negative selection. Science 269, 400–403.
High, K. P. (1994). The antimicrobial activities of cyclosporine, FK506, and rapamycin. Transplantation 57, 1689–1700.[Medline]
Hull, C. M. & Heitman, J. (2002). Genetics of Cryptococcus neoformans. Annu Rev Genet 36, 557–615.[CrossRef][Medline]
Kino, T., Hatanaka, H., Hashimoto, M., Nishiyama, M., Goto, T., Okuhara, M., Kohsaka, M., Aoki, H. & Imanaka, H. (1987). FK-506, a novel immunosuppressant isolated from a Streptomyces. I. Fermentation, isolation, and physico-chemical and biological characteristics. J Antibiot 40, 1249–1255.[Medline]
Kojima, K., Takano, Y., Yoshimi, A., Tanaka, C., Kikuchi, T. & Okuno, T. (2004). Fungicide activity through activation of a fungal signalling pathway. Mol Microbiol 53, 1785–1796.[CrossRef][Medline]
Kraus, P. R., Fox, D. S., Cox, G. M. & Heitman, J. (2003). The Cryptococcus neoformans MAP kinase Mpk1 regulates cell integrity in response to antifungal drugs and loss of calcineurin function. Mol Microbiol 48, 1377–1387.[CrossRef][Medline]
Kwon-Chung, K. J., Edman, J. C. & Wickes, B. L. (1992). Genetic association of mating types and virulence in Cryptococcus neoformans. Infect Immun 60, 602–605.
Lee, B. N. & Elion, E. A. (1999). The MAPKKK Ste11 regulates vegetative growth through a kinase cascade of shared signaling components. Proc Natl Acad Sci U S A 96, 12679–12684.
Lengeler, K. B., Cox, G. M. & Heitman, J. (2001). Serotype AD strains of Cryptococcus neoformans are diploid or aneuploid and are heterozygous at the mating-type locus. Infect Immun 69, 115–122.
Moore, T. D. & Edman, J. C. (1993). The alpha-mating type locus of Cryptococcus neoformans contains a peptide pheromone gene. Mol Cell Biol 13, 1962–1970.
Motoyama, T., Ohira, T., Kadokura, K., Ichiishi, A., Fujimura, M., Yamaguchi, I. & Kudo, T. (2005). An Os-1 family histidine kinase from a filamentous fungus confers fungicide-sensitivity to yeast. Curr Genet 47, 298–306.[CrossRef][Medline]
Mrsa, V. & Tanner, W. (1999). Role of NaOH-extractable cell wall proteins Ccw5p, Ccw6p, Ccw7p and Ccw8p (members of the Pir protein family) in stability of the Saccharomyces cerevisiae cell wall. Yeast 15, 813–820.[CrossRef][Medline]
Mrsa, V., Seidl, T., Gentzsch, M. & Tanner, W. (1997). Specific labelling of cell wall proteins by biotinylation. Identification of four covalently linked O-mannosylated proteins of Saccharomyces cerevisiae. Yeast 13, 1145–1154.[CrossRef][Medline]
Ochiai, N., Fujimura, M., Oshima, M., Motoyama, T., Ichiishi, A., Yamada-Okabe, H. & Yamaguchi, I. (2002). Effects of iprodione and fludioxonil on glycerol synthesis and hyphal development in Candida albicans. Biosci Biotechnol Biochem 66, 2209–2215.[CrossRef][Medline]
Odom, A., Del Poeta, M., Perfect, J. & Heitman, J. (1997a). The immunosuppressant FK506 and its nonimmunosuppressive analog L-685,818 are toxic to Cryptococcus neoformans by inhibition of a common target protein. Antimicrob Agents Chemother 41, 156–161.[Abstract]
Odom, A., Muir, S., Lim, E., Toffaletti, D. L., Perfect, J. & Heitman, J. (1997b). Calcineurin is required for virulence of Cryptococcus neoformans. EMBO J 16, 2576–2589.[CrossRef][Medline]
Onyewu, C., Blankenship, J. R., Del Poeta, M. & Heitman, J. (2003). Ergosterol biosynthesis inhibitors become fungicidal when combined with calcineurin inhibitors against Candida albicans, Candida glabrata, and Candida krusei. Antimicrob Agents Chemother 47, 956–964.
Orth, A. B., Rzhetskaya, M., Pell, E. J. & Tien, M. (1995). A serine (threonine) protein kinase confers fungicide resistance in the phytopathogenic fungus Ustilago maydis. Appl Environ Microbiol 61, 2341–2345.[Abstract]
), and cna1
), hog1
) and cna1
) mutant strains. (C) WT (H99), hog1
