|Home | About | Journals | Submit | Contact Us | Français|
The opportunistic intracellular fungal pathogen Cryptococcus neoformans depends on many antioxidant and denitrosylating proteins and pathways for virulence in the immunocompromised host. These include the glutathione and thioredoxin pathways, thiol peroxidase, cytochrome c peroxidase, and flavohemoglobin denitrosylase. All of these ultimately depend on NADPH for either catalytic activity or maintenance of a reduced, functional form. The need for NADPH during oxidative stress is well established in many systems, but a role in resistance to nitrosative stress has not been as well characterized. In this study we investigated the roles of two sources of NADPH, glucose-6-phosphate dehydrogenase (Zwf1) and NADP+-dependent isocitrate dehydrogenase (Idp1), in production of NADPH and resistance to oxidative and nitrosative stress. Deletion of ZWF1 in C. neoformans did not result in an oxidative stress sensitivity phenotype or changes in the amount of NADPH produced during oxidative stress compared to those for the wild type. Deletion of IDP1 resulted in greater sensitivity to nitrosative stress than to oxidative stress. The amount of NADPH increased 2-fold over that in the wild type during nitrosative stress, and yet the idp1Δ strain accumulated more mitochondrial damage than the wild type during nitrosative stress. This is the first report of the importance of Idp1 and NADPH for nitrosative stress resistance.
The alveolar macrophage can produce microbicidal amounts of toxic reactive oxygen species (ROS) and reactive nitrogen species (RNS) following phagocytosis (27, 53). Despite this, the opportunistic fungal pathogen Cryptococcus neoformans is able to inhabit and replicate within phagocytes of the mammalian host and to exit these cells unharmed (1, 2, 40). The intracellular pathogenicity of C. neoformans is most likely facilitated by stress resistance mechanisms, including a number of antioxidant proteins and pathways involved in the detoxification of ROS and RNS. Specifically, these include the synthesis of mannitol, a free radical scavenger (9, 20); the small protein flavohemoglobin denitrosylase (Fhb1), which is essential for resistance of C. neoformans to nitrosative stress (10, 14, 32); and the glutathione and thioredoxin antioxidant systems, which are both important for stress resistance and virulence (42, 43, 45).
Even with different mechanisms of catalysis and/or cellular localization, one thing that these stress resistance proteins and pathways have in common is the requirement for NADPH as a cofactor. NADPH is used as an electron donor either in recycling of oxidized, inactive enzymes to reduced, active forms or directly in catalytic activity. For example, Fhb1 binds NADPH during its catalytic activity and uses it directly as an electron donor for the reduction of NO· to NO3 (21). Catalases, which are highly conserved antioxidants that dismute H2O2 to molecular oxygen and water, consist of four units each with a molecule of NADPH bound in the core (18, 36, 59). The tripeptide glutathione (GSH) is oxidized to glutathione disulfide (GSSG), a homodimer held together by a disulfide bridge, during its oxidative state. GSSG can be reduced back to GSH by glutathione reductase, an enzyme that requires NADPH for electrons used in reduction. Similarly, glutathione peroxidase and thiol peroxidase ultimately depend on NADPH for recycling from an oxidized, inactive form back to a reduced, active form (57).
NADPH is classically recognized as being produced by the highly conserved, cytosolic pentose phosphate pathway. This pathway has been shown to be important for reductive biochemistry during oxidative stress in many organisms. The pentose phosphate pathway is an essential factor in maintaining health of erythrocytes, cells that, due to their biological function, have considerable risk for oxidative damage. Humans deficient in the pathway have hemolytic anemia, as their erythrocytes are unable to maintain sufficient pools of reduced glutathione (68). Also, the pressure of oxidative stress can stimulate the pentose phosphate pathway. This has been shown in human lymphocytes (56); in the rat adrenal gland, liver, and pancreas (15, 16); and in bacteria (63).
In fungi, the pentose phosphate pathway has been implicated in both oxidative stress resistance and adaptation to oxidative stress. In the model yeast Saccharomyces cerevisiae, NADPH-generating systems, including the pentose phosphate pathway, are critical for the ability of this organism to resist and adapt to high levels of oxidative stress (35, 47). It has also been shown that the cytosolic copper/zinc superoxide dismutase and the pentose phosphate pathway have overlapping roles in protecting S. cerevisiae from oxidative stress and that both systems are critical for maintaining the intracellular redox state (62). Furthermore, fungi may rely on the pentose phosphate pathway for more than reducing oxidative stress. Aspergillus nidulans requires a functional pentose phosphate pathway for nitrogen metabolism. Four A. nidulans mutants with independent defects in the pentose phosphate pathway were unable to grow on nitrite, nitrate, or various carbon sources, including 1% glucose, d-xylose, or d-glucoronate (28).
The pathway has two phases, the oxidative phase and the nonoxidative phase. The oxidative phase consists of two successive oxidations and results in the production of NADPH. The first enzyme in the oxidative phase of the pentose phosphate pathway is glucose-6-phosphate dehydrogenase (Zwf). Zwf catalyzes the oxidation of glucose-6-phosphate to 6-phosphogluconate and is highly specific for NADP+ as a cofactor (49, 67). There is abundant evidence supporting the role of Zwf in oxidative stress resistance. In addition to Zwf deficiency causing hemolytic anemia, Zwf has been also been implicated in maintenance of DNA repair systems during oxidative stress, as some cancers and aging disorders have also been linked to Zwf deficiency (30). For instance, Chinese hamster ovarian cells that are Zwf null have enhanced radiation sensitivity and a reduced ability to repair double-strand breaks due to the inactivation of Ku, a heterodimer DNA repair protein. In this case, the inactivation of Ku is the result of overoxidation of key cysteine residues on the protein due to the lack of sufficient reduced GSH (3). In the model yeast Saccharomyces cerevisiae, deletion of ZWF1 results in sensitivity to oxidative stress. ZWF1 is also important for the adaptive response to oxidative stress in S. cerevisiae. ZWF1-null mutants and wild-type cells were pretreated with 0.2 mM H2O2 and then challenged with 2 mM H2O2. While a large increase in tolerance to the high level of H2O2 was observed in the wild-type cells pretreated with 0.2 mM H2O2, the zwf1Δ strain was unable to tolerate the higher concentration (33). In Candida albicans, another pathogenic fungus, ZWF1 is upregulated during oxidative stress (38).
Another source of NADPH is NADP+-dependent isocitrate dehydrogenase (Idp) (55), a ubiquitous enzyme that in systems ranging from humans to yeasts to plants has been found in the cytosol, peroxisomes, or mitochondria (12, 19, 70). Although this enzyme can be targeted to mitochondria, it is distinct from the NAD+-dependent isocitrate dehydrogenase (Idh) that functions in the mitochondria as part of the Krebs cycle. However, similarly to Idh, Idp catalyzes the decarboxylation of isocitrate to α-ketoglutarate (29). This reaction can be performed in the mitochondria, in the cytosol, or in peroxisomes using isocitrate formed from citrate exported across the mitochondrial membrane. This allows for the production of NADPH in cellular compartments without reliance of active transport of NADPH across membranes (11). It is important to have reductive power produced directly within organelles for protection from exogenous as well as endogenous stressor. For example, NADPH is consumed in peroxisomes by enzymes such as catalase and uric acid oxidase, that counteract the ROS produced during breakdown of lipids (4, 5, 31). Mitochondria particularly require reductive capability, as these organelles are susceptible to endogenous ROS produced during cellular respiration and also to exogenous RNS (52). The proteins that make up the electron transport chain are prone to damage by nitric oxide, peroxynitrite, and S-nitrosothiols (6). Nitric oxide and peroxynitrite have been shown to cause irreversible damage to cytochrome c reductase, NADH dehydrogenase, and the succinate-ubiquinone complex; the common mechanism of damage is sequestration of iron/sulfur centers of the proteins (54, 69). Thus, without a means of detoxification, the mitochondrial membrane loses potential and the ability to continue respiration, leading to death of the stressed cell. In C. neoformans, some antioxidant enzymes that are located at the mitochondria and dependent on NADPH for function include catalases, superoxide dismutases, cytochrome c peroxidase, and flavohemoglobin denitrosylase (7, 24, 25, 26). These enzymes are important for stress resistance or virulence of C. neoformans due to their role in high-temperature growth (24, 25) or nitrosative stress resistance (10, 14, 26).
In humans, there is one IDP gene that results in mitochondrial and peroxisomal products (22). In S. cerevisiae, there are three IDP genes, which encode mitochondrial (IDP1), cytosolic (IDP2), and peroxisomal (IDP3) forms of the protein. Deletion of both ZWF1 and any one of the IDP genes in S. cerevisiae results in sensitivity to oxidative stress, likely due to a substantial decrease in NADPH produced in these double deletion mutants (41). In C. neoformans there is one predicted IDP gene (IDP1). Microarray data have indicated that this gene is upregulated 2.5-fold during nitrosative stress and thus may have a role in resistance to this stressor (44).
Since so many factors essential for stress resistance in C. neoformans utilize NADPH, we hypothesize that the sources of this cofactor are likewise critical for stress resistance. Although Zwf1 is important for adaptation to oxidative stress in the fungi S. cerevisiae and C. albicans, we had previously found that C. neoformans is unable to adapt to oxidative stress (S. M. Brown and J. K. Lodge, unpublished data), and thus we had reason to suspect that the role of Zwf1 in C. neoformans may be different than that in other organisms. The role of Idp1 in stress resistance, especially in resistance to nitrosative stress, is relatively unknown. In this study we used biochemical and genetic approaches to compare the roles of Zwf1 and Idp1 in resistance to oxidative and nitrosative stress in C. neoformans. We found that the Zwf1 is dispensable for viability, for resistance to oxidative and nitrosative stress, and for NADPH production. In contrast, we found that Idp1 is important for resistance to nitrosative stress, specifically for maintaining healthy mitochondria during exposure to nitrosative stress.
KN99α (50), C. neoformans serotype A, was used for all experiments. All mutant strains were made in the KN99α background. Reagents, including 6-phosphogluconate potassium salt and rhodamine 123, were from Sigma Chemical Company (St. Louis, MO). MitoTracker Green FM was from Invitrogen (Invitrogen Corporation, Carlsbad, CA).
Gene-specific deletion cassettes containing specific drug resistance markers were created using overlap PCR technology (13). The primer pair idp1-1/idp1-4 was used to amplify a 690-bp genomic region corresponding to the 5′ region of IDP1. Primer pair idp1-2/idp1-5 was used for the 3′ region of the gene (Table 1). An entire IDP1 gene open reading frame (1.65-kb genomic fragment) was deleted. Primer pair zwf1-1/zwf1-6 (Table 1) was used to amplify a 3.48-kb region of the genome encompassing the ZFW locus; the entire ZWF1 gene open reading frame was deleted. Drug marker cassettes used were Geneticin (zwf1Δ) and hygromycin (idp1Δ). KN99α was transformed using biolistic techniques (64). Cells were grown in rich medium (yeast extract-peptone-dextrose [YPD]) to late log phase, concentrated, and plated onto YPD agar for transformation. The cells were bombarded with 0.6-μm gold beads (Bio-Rad, Richmond, CA) which were coated with DNA of the target construct according to the manufacturer's recommendations. Following the transformation, the cells were incubated at 25°C for 4 h on nonselective medium to allow for recovery and then transferred with 0.8 ml sterile phosphate-buffered saline (PBS) to the appropriate selective medium. Transformants were observed in 3 to 5 days.
To isolate stable transformants, all transformants were passaged three times on nonselective YPD agar and then tested for resistance to the appropriate selective marker. Only those transformants that grew equally well on the selective medium as on nonselective medium were considered stable transformants. A three-primer PCR screen was used to prove homologous integration on both the 5′ and 3′ ends of the deletion construct (46). In this manner, homologous recombinants can be distinguished from the wild type. Genomic DNA for the screen was prepared by a modification of the glass bead DNA extraction protocol described by Fujimura and Sakuma (17). A PCR screen using the primers outside the deletion construct will amplify the entire gene region, demonstrating that a single copy of the transforming DNA had been inserted at the desired locus. Southern blots were performed to screen for single integration in the genome. A single band of characteristic size (after in silico analysis) was observed for all the deletion colonies used for the experiments. The colonies in which Southern blot analysis revealed multiple insertion of the drug cassette were excluded from further analysis.
The IDP1 gene for complementation was amplified from strain K99 using primers idp1-3/idp1-6 (Table 1). A 1.65-kb IDP1 gene flanked by 5′ and 3′ regions (700 bp and 600 bp, respectively) was cotransformed into idp1Δ cells with supercoiled G418-containing plasmid. After biolistic transformation, cells initially were incubated at 30°C for 4 h and later were incubated in the presence of G418 at 37°C. The colonies growing at 37°C were passaged immediately on YPD in the absence of drugs and were used for colony PCR. The colonies in which the IDP1 gene was reinserted at its genomic locus replacing the hygromycin cassette were confirmed by diagnostic PCR and also tested sensitive for hygromycin. The two complemented strains designated idp1Δ::IDP1-A8 and idp1Δ::IDP1-C4 were used for further analysis.
Cells were plated in 10-fold serial dilutions on solid minimal medium (yeast nitrogen base [YNB], pH 4.0) with or without a stressor (1 mM H2O2 for oxidative stress and 0.75 mM NaNO2 for nitrosative stress). Plates were incubated and observed at 2, 3, 5, and 7 days.
For the measurement of NADP+ and NADPH, cells were grown overnight at 25°C in YNB, pH 4.0. A stressor (1 mM H2O2 or 1 mM NaNO2) was added to the cell cultures at an optical density at 650 nm (OD650) of 1.0 and incubated for 1 h. Cells were pelleted by centrifugation and frozen at −80°C prior to measurements. The measurements were made using the EnzyChrom NADP+/NADPH assay kit from BioAssay Systems (ECNP-100; BioAssay Systems, Hayward, CA). Briefly, cells were homogenized by bead beating in either acidic extraction buffer for extraction of NADP+ or alkaline extraction buffer for extraction of NADPH. Homogenates were heated at 60°C for 5 min and then neutralized by addition of the opposite extraction buffer. For the measurement, a working reagent mix including glucose-6-phosphate dehydrogenase was added, and an initial reading at 565 nm was taken immediately. Another reading was taken after 30 min of incubation at room temperature. A standard curve of NADP+ was used for the calculation of total pyridine nucleotide.
For the measurement of NAD+ and NADH, cells were prepared as for the NADP+ measurements, using the EnzyChrom NAD+/NADH assay kit from BioAssay Systems (ECND-100; BioAssay Systems). The working reagent for this assay contained alcohol dehydrogenase. The first reading at 565 nm was immediately following addition of the working reagent, and the final reading was taken after 15 min of incubation at room temperature. A standard curve of NAD+ was used for the calculation of total pyridine nucleotide.
6-Phosphogluconate (6-PG) was measured by gas chromatography-mass spectrometry (GC/MS) (61). The standard used for comparison was 6-phosphogluconate potassium salt. This standard was run at multiple concentrations to confirm the retention time of silylated 6-PG and the m/z of the corresponding fragments. The derivatized standard eluted at 30.984 min and gave distinct fragments of m/z 299, 315, 333, 357, and 387. The ion of m/z 387 was followed for quantification. Spiked samples were also run to determine matrix effects. For measurement of 6-PG produced in wild-type and zwf1Δ strains under normal conditions, oxidative stress, and nitrosative stress, 50-ml cultures of each strain were incubated in YNB (pH 4.0) in a 30°C shaking incubator overnight. One millimolar H2O2 (oxidative stress), 1 mM NaNO2 (nitrosative stress), or 1 ml PBS as a control for no stress was added to the cultures, and the cultures were incubated for an additional 2 h at 30°C. Following treatment, cells were pelleted by centrifugation at 3,220 × g for 8 min. The cells were washed three times in sterile PBS. The washed cells were disrupted by vortexing with 0.5 mM glass beads (BioSpec Products, Inc.) for 15 min in sterile, deionized water. The disrupted cells and beads were spun at 16,000 × g for 10 min and the supernatants collected and passed through a 45-μm filter.
A specified volume of filtered lysate was transferred to a Reactivial containing the following 11 internal standards: 500 nmol d3-creatine; 10 nmol d3-methylmalonic acid; and 100 nmol each of [13C3]lactate, [13C3]pyruvate, d3-serine, d5-phenylalanine, [15N2]orotate, d4-sebacic acid, [13C6]glucose, d6-inositol, and d5-tryptophan. Forty microliters of triethylammonium trifluoroacetate was added to each vial, and acetonitrile was added to the tops of the vials. The samples were then placed under a nitrogen stream and held at 70°C in a heating block. Acetonitrile was added as the volume was reduced until a precipitate appeared. Methylene chloride was added, allowed to blow off, and then added again until the sample was anhydrous and reduced to constant volume. Up to 500 μl N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA) was added, depending upon the appearance of the original sample, and the suspension was capped and heated at 70°C for 1 to 2 h. A splitless glass insert was used with the valve program, which turns off the purge for 1 min. The injection temperature was 200°C. A 25-m DB5 capillary column (film thickness, 0.5 μm; inner diameter, 0.32 mm) was used with a temperature program of 80 to 130°C at 4°C/min, 130 to 200°C at 6 C/min, and 200 to 285°C at 12°C/min, with holds of 1 min at 80°C and 10 min at 285°C for a total run time of 42.5 min. The mass spectrometer, in EI+ mode, scanned from 50 to 650 amu every 2.46 s after a 3.5-min delay.
For staining with fluorescent dyes, 1 mM H2O2 or 1 mM NaNO2 was added to 50-ml cultures of KN99α and the zwf1Δ, psi11Δ, and idp1Δ mutants that had been incubated in YNB (pH 4.0) at 25°C overnight. The cultures were incubated for an additional 2 h with the stressors, and then the cells were collected by centrifugation at 1,800 rpm for 8 min. The cells were washed in fresh YNB and then resuspended in 30 ml of YNB prewarmed to 37°C. To 1 ml of each culture, 10 μg rhodamine 123 (1-mg/ml stock solution, diluted to 10 μg/ml), 100 μg MitoTracker Green FM (lyophilized MitoTracker resuspended in fresh dimethyl sulfoxide [DMSO] to 1 mM, diluted to 100 μg/ml), or both were added and incubated with shaking at 37°C for 30 min. The cells were collected by centrifugation and washed in fresh YNB three times or until the supernatant was clear. The stained cells were resuspended in 1 ml YNB, and 10 μl of the suspension was placed on a microscope slide. The cells were visualized by confocal microscopy at a magnification of ×100.
The Zwf1 enzyme in C. neoformans (CNAG_03245.1) has been previously purified and characterized as a glucose-6-phosphate dehydrogenase by Niehaus and Mallett (49). We compared the sequence of this gene to the current C. neoformans genome (Broad Assembly 1, May 2006; www.broad.mit.edu/annotation/genome/cryptococcus_neoformans/) and also to the current S. cerevisiae genome database (SGD). There are two genes in S. cerevisiae (51), and one in C. neoformans. This gene was disrupted by overlap PCR, and deletion of the gene was confirmed by PCR and Southern blotting.
While there are three IDP genes in S. cerevisiae, BLAST homology searches resulted in only one homologue in the C. neoformans genomic sequence. This gene, IDP1 (CNAG_03920.1), was homologous to all three of the S. cerevisiae IDP genes, with highest homology to the mitochondrial IDP1. Mitoprot, a mitochondrial targeting sequence prediction server (http://ihg2.helmholtz-muenchen.de/ihg/mitoprot.html), was used to identify a putative mitochondrial signal sequence in C. neoformans IDP1. A full-length cDNA of IDP1 was isolated, and the sequencing results confirmed the presence of putative mitochondrial signal sequence. IDP1 is only 17% similar to the C. neoformans NAD+-dependent isocitrate dehydrogenase gene IDH.
Deletion of the IDP1 and ZWF1 genes and confirmation of the deletions were carried out as described in Materials and Methods. A total of 12 idp1 deletion mutants, generated by four independent biolistic transformations, were selected for phenotypic analysis. Individual colonies grown in YPD were plated in 10-fold serial dilutions on solid YPD plates. Plates were incubated at 25, 30, and 37°C; a representative result is shown in Fig. 1a. Nine of the individual deletion mutants exhibited a consistent temperature-sensitive growth phenotype. The idp1Δ cells exhibited growth similar to that of the wild type after 2 and 4 days of incubation at 25°C. However, they showed a significant impairment of growth at 30°C even after 4 days of growth. Loss of Idp1 completely abolished the growth at 37°C. The temperature-sensitive growth phenotype of idp1Δ was completely restored in the two independent idp1Δ::IDP1-A8 and idp1Δ::IDP1-C4 complemented strains. Figure 1b confirms the presence of the IDP gene-specific transcript in the complemented strains. Temperature-sensitive phenotypes were not observed in the zwf1Δ strain (data not shown).
We measured absolute values of NADP+ and NADPH produced in zwf1Δ, idp1Δ, and wild-type cells during normal growth (YNB, pH 4.0, 25°C) and growth during oxidative and nitrosative stress (YNB, pH 4.0, with 1 mM H2O2 or 1 mM NaNO2, 25°C). NADP+ and NADPH were extracted from whole-cell lysate and measured by absorbance at 565 nm. Amounts were quantified by comparison to a standard curve of NADP+, and the data were processed as a ratio of NADPH to NADP+. The ratio of NADPH to NADP+ is indicative of the demand for NADPH by the cell. If there was no demand for NADPH, then the ratio of the reduced pyrimidine to the oxidized form would be 1. The requirement for NADPH, whether for stress resistance or for biosynthesis, decreases the ratio, as NADPH is used in cellular processes (15, 48, 58). Each measurement was taken in triplicate from six biological replicates.
In wild-type cells, the same NADPH/NADP+ ratio was maintained regardless of growth condition, whether without a stressor or with an oxidative or nitrosative stressor (Fig. 2A, B, and C). We expected the ratio to decrease in this strain during oxidative stress, due to decreased production of NADPH by the oxidative phase of the pentose phosphate pathway in the absence of the rate-limiting enzyme. However, the same ratio found in the wild type was maintained in the zwf1Δ strain under all growth conditions (Fig. 2A, B, and C). In contrast, the ratio was significantly increased in the idp1Δ strain during oxidative and nitrosative stress compared to in both the wild-type and zwf1Δ strains (Fig. 2B and C). The increase was most dramatic during nitrosative stress (Fig. 2C).
Despite the increase in the NADPH/NADP+ ratio in idp1Δ cells during nitrosative stress, the idp1Δ strain was sensitive to nitrosative stress in an in vitro stress assay. Serial dilutions were plated onto solid YNB (pH 4.0) with 1 mM NaNO2 or 1 mM H2O2 or without a stressor. Plates were incubated at 25°C and observed at 3, 5, and 7 days. Compared to the wild type, the idp1Δ strain exhibited a stress phenotype on nitrosative stress plates. The idp1Δ strain was not sensitive to oxidative stress in vitro (Fig. 3).
To test whether Zwf1 is important for resistance to oxidative or nitrosative stress, serial dilutions of zwf1Δ and wild-type cells were plated on solid YNB (pH 4.0) with 1 mM H2O2 or 0.75 mM NaNO2 or without a stressor. Plates were incubated at 30°C and observed at 3, 5, and 7 days. Unlike S. cerevisiae, the C. neoformans zwf1Δ strain did not exhibit sensitivity to either nitrosative or oxidative stress (Fig. 4). Previous reports on S. cerevisiae have indicated that the zwf1Δ strain is a methionine auxotroph (62), but the C. neoformans zwf1Δ strain was able to grow in the absence of methionine (data not shown). The zwf1Δ and idp1Δ strains were both tested with lower (0.5 mM) and higher (2.0 mM) concentrations of H2O2, and there was no difference in the phenotypes seen (data not shown).
Deletion of ZWF1 did not cause a stress-sensitive phenotype, as seen in other systems such as S. cerevisiae, or a change in NADPH during stress compared to the wild type. Thus, we investigated whether the second reaction of the oxidative phase of the pentose phosphate pathway could be a source of NADPH in the absence of Zwf1.
The two reactions of the oxidative phase of the pentose phosphate pathway (see Fig. 8) produce NADPH. The first reaction is catalyzed by Zwf1 and converts glucose-6-phosphate to 6-phosphogluconate. 6-PG is then converted to ribulose-5-phosphate by 6-phosphogluconate dehydrogenase (Gnd), with the concurrent reduction of NADP+ to NADPH. Gnd is highly specific for 6-PG; so far there is no reported evidence that Gnd utilizes any other phosphosugar in vivo (8, 39). Based on current evidence, this reaction should proceed, thereby producing NADPH, only if 6-PG is provided. In eukaryotes, with the exception of Schizosaccharomyces pombe (65, 66), the source of 6-PG for the second reaction of the pentose phosphate pathway is the Zwf-catalyzed conversion of glucose-6-phosphate to 6-PG. Therefore, we reasoned that deletion of Zwf1 would eliminate production of 6-PG and thus NADPH by the 6-phosphogluconate reaction.
To test this hypothesis, trimethylsilyl derivatives of 6-PG were measured by gas chromatography-mass spectrometry of wild-type and zwf1Δ strains grown with or without oxidative or nitrosative stressors. The results are shown in Fig. 5. The amounts of 6-PG in wild-type cells under normal growth conditions (no stressor) and during nitrosative stress were the same. There was a statistically significant increase in 6-PG produced by wild-type cells during oxidative stress, indicating that the pressure of oxidative stress does increase flux through the first reaction of the oxidative phase of the pentose phosphate pathway in wild-type cells. In contrast to the case for wild-type cells, there was no 6-PG found in the zwf1Δ cells under normal growth conditions. However, some 6-PG was found in zwf1Δ cells during oxidative and nitrosative stress. This amount was 4-fold less than that in the wild type under the same conditions. Thus, in C. neoformans, deletion of ZWF1 attenuates production of 6-PG but does not completely abolish it.
IDP1 is distinct from IDH, the NAD+-dependent isoform of isocitrate dehydrogenase that functions in the tricarboxylic acid (TCA) cycle. We predict that Idp1 does not participate in this energy-producing cycle; however, we needed to confirm that the slow-growth phenotype of the idp1Δ strain was not due to disruption of the TCA cycle. NADH is a product of the cycle, and so to confirm that deletion of IDP1 does not interfere with the cycle, we measured NADH in wild-type and idp1Δ cells during normal growth and nitrosative stress. We found no difference in the amounts of NAD+ and NADH in the idp1Δ strain compared to the wild type, indicating that deletion of IDP1 does not interfere with normal production of NADH (data not shown).
If Idp1 indeed is localized to mitochondria, it may have a critical role in supplying mitochondrial antioxidants with NADPH during nitrosative stress. The sensitive phenotype despite the increase in NADPH in the idp1Δ strain during nitrosative stress could be caused by a lack of sufficient reducing equivalents within the mitochondria due to the absence of a direct source and insufficient membrane transport. Thus, NADPH would pool within the cytosol and yet not be accessible to the enzymes essential for nitrosative stress resistance. We decided to use the mitochondrial stains rhodamine 123 (Sigma) and MitoTracker Green FM (Invitrogen) to test loss of mitochondrial membrane integrity in idp1Δ cells during nitrosative stress. Rhodamine 123 passively diffuses across mitochondrial membranes and will be sequestered within the organelles if the membrane is potentiated, as it is in the healthy state. However, if the membrane has lost potential, this dye will efflux out of the organelle (34). Conversely, MitoTracker mitochondrion-selective probes passively diffuse across mitochondrial membranes and will remain within mitochondria regardless of membrane potential (Invitrogen). As a result, MitoTracker can be used to identify mitochondria by fluorescence microscopy, and rhodamine 123 can be used to determine whether mitochondria have intact, potentiated membranes.
For this experiment, we used the wild-type, idp1Δ, and zwf1Δ strains. For a control, we used the psi11Δ strain, which is sensitive to nitrosative stress but in which this sensitivity is likely because of poor cell wall integrity and/or cell signaling (K. J. Gerik and J. K. Lodge, unpublished data) and not primarily mitochondrial damage. We looked at the retention of both MitoTracker and rhodamine 123 in strains exposed to oxidative (1 mM H2O2), and nitrosative (0.75 mM NaNO2) stressors for 1 h compared to that in cells not exposed to stress (Fig. 6). To rule out the nonspecific effect of incubating cells at 37°C (employed in this experiment for labeling), we investigated the effect of brief exposure of cells (90 min) to 37°C on their viability. By counting the CFU after the exposure of either the wild-type or mutant cells to a temperature of 37°C for up to 90 min, we observed no significant difference in cell viability (data not shown). We found that 100% of wild-type cells not exposed to a stressor that stained with MitoTracker also retained rhodamine 123, indicating healthy mitochondria. Conversely, after exposure to nitrosative stress, only 30% of wild-type cells that stained with MitoTracker also retained rhodamine 123, indicating that nitrosative stress does induce mitochondrial damage (Fig. 7). This is in contrast to oxidative stress, which did not result in a decrease in the number of cells that retained both dyes. Similar results were seen with the zwf1Δ and psi11Δ strains. However, while 100% of the idp1Δ cells that took up MitoTracker also retained rhodamine 123 after exposure to no stress or oxidative stress, only 9.0% of cells that retained MitoTracker retained rhodamine 123 as well (Fig. 7). This indicates that the idp1Δ strain is more susceptible to mitochondrial damage during nitrosative stress than any of the other strains tested, supporting the hypothesis that Idp1 is localized to the mitochondria and that its presence there is essential for production of the NADPH used by the antioxidant proteins located in the mitochondria.
The ability of the fungal pathogen C. neoformans to survive within macrophages of the mammalian host is presumably due to the ability of the fungal cell to protect itself from toxic reactive oxygen and nitrogen species that can be produced by macrophages following phagocytosis. It is well established that several antioxidant proteins are essential for stress resistance and virulence. Regardless of mechanism or localization, a common requirement of many essential protective enzymes is the reductive cofactor NADPH. Since NADPH crosses membrane bilayers only by facilitated transport, it is important that NADPH be produced within the various cellular compartments so that there is a pool of reducing equivalents readily available for stress resistance. Mitochondria particularly are susceptible to damage from NO·, as several of the components of the electron transport chain contain metal centers that can be sequestered by NO·, leading to disruption of the membrane, impaired function, and ultimately cell death.
Two potential sources of NADPH for stress resistance are the cytosolic pentose phosphate pathway and NADP+-dependent isocitrate dehydrogenase. In many other systems, disruption of the pentose phosphate pathway by deletion of the first and rate-limiting enzyme, Zwf1, leads to poor growth and insufficient resistance to oxidative stress, presumably due to a decrease in NADPH. Likewise, in S. cerevisiae the three Idp proteins are important for protection from ROS under various conditions, including peroxisomal fatty acid oxidation (Idp3). Given the need for antioxidant function and the number of NADPH-dependent enzymes essential for nitrosative stress resistance and virulence in C. neoformans, we investigated the roles of Zwf1 and Idp1 as sources of NADPH for resistance to nitrosative stress as well as oxidative stress.
We first investigated the role of Zwf1 in production of NADPH and resistance to stress by creating a zwf1Δ strain, measuring NADPH in this strain, and testing in vitro stress sensitivity. Our results showed that the role of Zwf1 in production of NADPH and oxidative stress resistance in C. neoformans is unlike that in other fungi. In S. cerevisiae, deletion of Zwf1 results in a slow-growth phenotype and methionine auxotrophy (35, 47, 62). In contrast, our C. neoformans zwf1Δ strain had no growth phenotype on rich, minimal, or methionine-deficient media, nor was it sensitive to changes in temperature. In S. cerevisiae, ZWF1 has been shown to be upregulated during oxidative stress at both the transcriptional and translational levels (33, 60), and in both S. cerevisiae and C. albicans, it is essential for adaptation to oxidative stress (37, 38). In contrast, the C. neoformans zwf1Δ strain was not sensitive to oxidative stress, and further, we found no changes in mRNA expression during oxidative stress (data not shown). We also measured the amount of NADPH produced by the zwf1Δ strain under normal conditions (no stress), oxidative stress, and nitrosative stress. Finding no difference compared to results for the wild type under any condition, we considered the possibility that NADPH is still made by the pentose phosphate pathway even in the absence of Zwf1.
While the first two decarboxylation reactions of the pentose phosphate pathway can produce NADPH, the second reaction, driven by Gnd1, is dependent on 6-PG, the substrate produced by the first reaction, and therefore would proceed only if provided this substrate. Thus, without an alternative source of 6-PG, Gnd1 would not be able to reduce NADP+ in the zwf1Δ strain. We found that deletion of ZWF1 in C. neoformans did not completely abolish production of 6-PG, and therefore we propose that there is an alternative source of this phosphosugar in C. neoformans. To date, the only reported source of 6-PG other than the decarboxylation of glucose-6-phosphate by Zwf1 is gluconate kinase (Fig. 8, arrow 3) in the “gluconate route,” a pathway that, while described mainly for prokaryotes, has been reported for Schizosaccharomyces pombe. While the bacterial gluconate kinase is NAD+ dependent, the S. pombe ortholog (CAA93562.1) is NADP+ dependent. Despite the authors' suggestion that the active gluconate route in S. pombe is a distinguishing factor between fission and budding yeasts (65, 66), bioinformatic searching of the C. neoformans genome (www.broad.mit.edu/annotation/genome/cryptococcus_neoformans/) resulted in identification of a possible homologue (CNAG_03048; E value = 9e−27) to the S. pombe gluconate kinase, annotated only as “cytoplasm protein.” We suggest that this enzyme may an alternative source of 6-PG in the absence of Zwf1 as well as another source of cytosolic NADPH (Fig. 8) and that while the amount of 6-PG produced in the absence of Zwf1 is still 4-fold less than that in the wild type, this is enough to support antioxidant proteins and biosynthetic pathways. This hypothesis is supported by the lack of an oxidative stress phenotype in the zwf1Δ strain.
Second, we looked at the role of Idp1 in stress resistance. Our idp1Δ strain was not susceptible to oxidative stress but was sensitive to both high temperature and nitrosative stress. Despite the phenotype, the idp1Δ cells accumulated NADPH during nitrosative stress, as indicated by the increase in NADPH/NADP+. In an effort to explain the phenotype, we predicted that the sensitivity to nitrosative stress is due to an inability to reduce mitochondrion-damaging RNS in the absence of Idp1. Containing a mitochondrial signal sequence, C. neoformans' Idp1 is likely to be localized to the mitochondria, and as such it could produce intracompartmental NADPH for use by mitochondrial antioxidant and denitrosylating proteins. We found that the idp1Δ strain accumulated significantly more inviable mitochondria during exposure to nitrosative stress than either the wild-type or zwf1Δ strain. Thus, we propose the following working model. Wild-type cells, with both intracompartmental and cytosolic sources of NADPH, have normal growth and stress resistance. In the absence of Zwf1, C. neoformans maintains normal stress resistance through cytosolic NADPH produced by an alternative protein or pathway and maintaining intracompartmental levels with Idp1. However, in the absence of Idp1, the cells sustain considerable mitochondrial damage during nitrosative stress, leading to cell death, while NADPH produced by cytosolic sources pools in the cytosol. The presence of antioxidant enzymes in the cytosol prevents the idp1Δ strain from being completely inviable when exposed to nitrosative stress. Of note, we were not able to obtain a zwf1Δ idp1Δ double deletion strain, suggesting that these enzymes have some redundant, essential role.
Our study shows that despite overwhelming evidence for a role of Zwf1 in oxidative stress resistance in many, various systems, its role in C. neoformans is exceptional. One indication from the results is that recycling of oxidized antioxidants, which is purported to be a significant use of NADPH, is not the main source of active enzymes in C. neoformans and that instead, increased protein turnover may be the prominent source of reduced enzymes in this species. This hypothesis is supported by proteomic data showing that there are very few changes in steady-state levels of proteins in C. neoformans during oxidative stress (S. M. Brown and J. K. Lodge, unpublished data). We also provide evidence of a functional gluconate kinase in C. neoformans. Finally, our results suggest that compartmentalization of NADPH production is important for stress resistance in C. neoformans and reveal a key role for Idp1 in nitrosative stress.
We acknowledge Jack Kennel for helpful discussions and guidance with the mitochondrial staining experiments, John Corbett for many helpful conversations regarding nitrosative stress, Wandy Beatty for microscope images, and Anthony Thomas for excellent technical support.
This work was supported by NIH grant HL088905 to J.K.L.
Published ahead of print on 16 April 2010.