PI3K signaling of autophagy is required for starvation tolerance and virulenceof Cryptococcus neoformans

Autophagy is a process by which cells recycle cytoplasm and defective organelles during stress situations such as nutrient starvation. It can also be used by host cells as an immune defense mechanism to eliminate infectious pathogens. Here we describe the use of autophagy as a survival mechanism and virulence-associated trait by the human fungal pathogen Cryptococcus neoformans. We report that a mutant form of C. neoformans lacking the Vps34 PI3K (vps34Δ), which is known to be involved in autophagy in ascomycete yeast, was defective in the formation of autophagy-related 8–labeled (Atg8-labeled) vesicles and showed a dramatic attenuation in virulence in mouse models of infection. In addition, autophagic vesicles were observed in WT but not vps34Δ cells after phagocytosis by a murine macrophage cell line, and Atg8 expression was exhibited in WT C. neoformans during human infection of brain. To dissect the contribution of defective autophagy in vps34Δ C. neoformans during pathogenesis, a strain of C. neoformans in which Atg8 expression was knocked down by RNA interference was constructed and these fungi also demonstrated markedly attenuated virulence in a mouse model of infection. These results demonstrated PI3K signaling and autophagy as a virulence-associated trait and survival mechanism during infection with a fungal pathogen. Moreover, the data show that molecular dissection of such pathogen stress-response pathways may identify new approaches for chemotherapeutic interventions.

Cryptococcus neoformans (Cn) is a major fungal pathogen primarily infecting immunocompromised hosts, especially patients with impaired cellular immunity related to HIV infection (1). It is generally believed that inhalation of desiccated yeasts or basidiospores from the environment is the route for primary pulmonary Cn infection. The alveolar macrophages (AMs) therefore serve as one of the first lines of anticryptococcal defense that must be overcome by a successful fungal intruder (2). While much is known about specific immunological killing mechanisms of macrophages that result in containment of intracellular yeast cells (3), only recently has study been devoted to understanding the host’s ability to deprive the pathogen of essential elements necessary for survival and virulence. For example, in cryptococcosis, deprivation of oxygen and glucose results in a requirement for the oxygen sensor, Sre1 (4), and the gluconeogenesis enzyme Pck1 (5) in the pathogen. Encountering such a hostile nutrient environment thus requires a specific programmatic response by the pathogen, dissection of which may yield new insights into unique attributes of the host-pathogen interaction. Such study may also contribute to our understanding of how organisms adapt during their transition from environmental saprophytes to pathogens and may provide genetic predictors of strains that have evolved to become the more successful pathogens. For example, clinical isolates showing more effective induction of high-affinity copper transport were recently found to be associated with a propensity to cause more severe neurological disease in a cohort of transplant patients (6).

The class III PI3K in ascomycete yeast, Vps34, has been shown to be involved in vesicular transport of vacuolar hydrolases to the vacuole lumen via endosomes (the so-called “CPY pathway”) and autophagy under nutrient-deprived conditions (7). CPY requires the formation of PI3K complex II composed of Vps34, Vps15, Vps30 (Atg6), and Vps38 (8). Autophagy, involved in nutrient conservation during starvation, requires formation of a type I kinase subcomplex composed of Vps34, Vps15, Atg14, and Atg6 (8, 9), leading to transport vesicle formation initiated by lipid binding (lipidation) of Atg8 (10). The Tor signaling pathway in yeast responds to nutrient levels, and inhibition by rapamycin leads to an artificial induction of autophagy under nutrient-replete conditions (11). In mammalian systems, an additional class I PI3K activity is associated with activation of Tor signaling through a TSC2/TSC1-dependent pathway and is inhibitable by nanomolar concentrations of wortmannin (12). Indeed, while there is no TSC2/TSC1 homolog in Saccharomyces cerevisiae (Sc), a homolog of TSC2 has been described in Schizosaccharomyces pombe (13), and a homolog in Cn is apparent in the UniProt genome database ( http://beta.uniprot.org/; accession no. Q5KJR4; 51% identical to S. pombe), suggesting higher yeast species may share a relationship between PI3K activity and TSC-mediated Tor activity.

In the present study, we report the targeted deletion of a cryptococcal VPS34 PI3K gene linked to intact laccase expression and use this to identify genetically linked attributes of virulence. The results show that although VPS34 was largely dispensable for growth of Cn in nutrient-replete conditions at 37°C and was not required for expression of several important virulence factors such as capsule and urease, it played a large role in virulence and survival in macrophages. In addition, the VPS34 mutation was associated with a defect in tolerance to nutrient starvation, related to an inability to undergo autophagy. Furthermore, molecular dissection of VPS34-related phenotypes was undertaken to determine the contribution of autophagy in cryptococcal pathogenesis, with the conclusion that RNAi suppression of the autophagy-related ATG8 gene also resulted in loss of virulence in 2 mouse models of cryptococcosis. These studies thus provide genetic support for a role for PI3K and autophagy in the adaptation to nutrient starvation and virulence for a pathogenic fungus.

Identification and mutation of a VPS34 homolog in Cn. Analysis of a laccase-associated cryptococcal annotated gene sequence (TIGR Database locus name 186.m03746; http://www.tigr.org/tdb/e2k1/cna1) showed strong homology to PI3Ks, including VPS34 from Sc (35% identity; E value, e-130), and contained 2 conserved PI3K sequences, a catalytic domain (PI3Kc) at aa 554–918 (E value, 2e-124), and an accessory domain (PI3Ka) at aa 344–527 (E value, 6e-48). BLAST analysis of the NCBI database showed homology with PI3Ks only within the first 200 matches. In addition, Clustal-W analysis (EMBL-EBI database; http://www.ebi.ac.uk/Tools/clustalw/index.html) comparing closely related kinase members, as described by Shelton et al. (14), showed highest similarity to the class III Vps34 class of PI3Ks exemplified by Vps34 of Sc, less homology with the class I PI3Ks such as those from mice or humans, and even less similarity to other related kinases (Figure 1A). A Cn knockout strain, vps34Δ, was constructed (Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI32053DS1) that demonstrated production of a number of phosphorylated phosphatidylinositol phosphates, including phosphatidylinositol 4 phosphate (PI4P), as described in ascomycete yeast (15), but also showed defective production of PI3P (Figure 1B) and reduced expression of laccase. However, the mutant retained production of the virulence factors capsule and urease (Figure 1C). Growth rate of the cryptococcal vps34Δ mutant was similar to WT at 30°C on agar (data not shown) and in liquid culture (doubling times: WT, 2.6 ± 0.3 h; vps34Δ, 2.8 ± 0.3 h; vps34Δ + VPS34, 2.4 ± 0.3 h), but there was only a barely discernible difference at 37°C on agar (Figure 1C) or in liquid culture (doubling times: WT, 2.8 ± 0.3 h; vps34Δ, 3.5 ± 0.4 h; vps34Δ + VPS34, 2.6 ± 0.3 h; P < 0.05 vps34Δ vs. WT or vps34Δ + VPS34). Retention of near WT growth at 37°C is in contrast to a more severe temperature-dependent growth defect of the homolog mutant strain from Sc (16). This difference could be due to redundancy in growth pathways in the basidiomycetous fungi and has also been found with other vps-class mutants from this fungus (17). To further characterize the cryptococcal Vps34 activity, a Sc vps34Δ mutant was complemented with a cDNA fragment of the cryptococcal gene expressed under a Saccharomyces constitutive promoter. As shown in Figure 1D, complementation of the Sc vps34Δ mutant with Cn VPS34 resulted in restoration of its temperature-sensitive defect at 37°C as well as production of autophagic bodies under starvation conditions (WT, 3.6 ± 0.3; Sc vps34Δ, 0.1 ± 0.1; Sc vps34Δ + Cn VPS34, 2.3 ± 0.4 vacuolar vesicles per cell; n = 15, P < 0.0001, vps34Δ vs. either WT or Sc vps34Δ + Cn VPS34) that were freely mobile within the Sc vacuole (data not shown). Furthermore, WT cryptococcal cell PI3P production was resistant to nanomolar concentrations of wortmannin (Figure 1D), consistent with a class III PI3K activity (15).

Figure 1

Phenotypes of a cryptococcal VPS34 homolog in vitro. (A) Clustal-W analysis comparing closely related kinases. Mus musculus class I PI3K (GenBank NP 032865); human class I PI3K, Vps34, DNA-PK, and PI4K (GenBank AAX41023, NP 002638, AAA79184, and AAA56839, respectively); Sc Tel1, Esr1, PI4KI, Vps34 (GenBank CAA84909, CAA85094, NP 013408, and CAA37610, respectively); S. pombe PI4K (GenBank NP 595304); and Cn Tor1 (GenBank AAD16273). (B) Indicated mid–log phase fungal cells were starved for 4 h in YNB without amino acids or ammonium sulfate and cell lysates assayed for PI kinase activity by autoradiography of thin-layer chromatograms, as described in Methods. (C) Indicated Cn cells were assayed for laccase by melanin formation, capsule by India Ink microscopy, urease activity, and growth at 37°C on YPD agar as described in Methods. (D) Indicated Sc mid–log phase cells were assayed for growth at 37°C in YPD (upper panel) or incubated in YNB without amino acids or ammonium sulfate for 4 h in the presence of 1 mM PMSF and observed using DIC microscopy or as described in Methods (middle panel). Arrows point to autophagic bodies. Lower panel: Cn cells were subjected to starvation conditions and assayed for PI kinase activity as in B with or without addition of 50 nM wortmannin.

Role of PI3K in cryptococcal pathogenesis. Because of retention of growth at 37°C in the cryptococcal vps34Δ mutant strain, we tested for virulence in mice. While some attenuation in virulence was expected because of the minimally reduced growth rate and reduced laccase expression, the vps34Δ mutant strain was found to be avirulent in an intravenous (Figure 2A) as well as an intranasal (Figure 2B) mouse model and organ culture of surviving mice failed to identify live fungal cells in brain or lung. Parallel experiments comparing virulence of a lac1Δ mutant with WT in the same genetic background and using the same models and fungal inoculum found a smaller reduction in virulence in the lac1Δ mutant; (intravenous, Swiss Albino: median survival, WT, 7 d, lac1Δ, 11 d, P < 0.05; intranasal, CBA/J: median survival, WT, 25 d, lac1Δ, 29 d, P < 0.05), suggesting that the avirulence of the vps34Δ mutant was not due solely to its reduction in laccase activity. Remarkably, after i.n. inoculation of CBA/J mice (Figure 2C), the vps34Δ strain was rapidly cleared from the lungs, in contrast to WT or 2 previously described hypovirulent knockout strains of the virulence factors laccase (lac1Δ) or capsule (cap60Δ) (18, 19). Since there was a slight reduction in fungal CFU at the first time point 1 hour after inoculation (WT, 5.8 ± 0.6 × 10⁴ CFU/lung vs. vps34Δ, 4.4 ± 0.5 × 10⁴ CFU/lung), we conducted a second set of experiments focusing on the vps34Δ mutant strain to see whether a more prompt recovery from mice and addition of a glucose nutrient (1% glucose in PBS) could prevent lung clearance of the mutant strain. As shown in Figure 2D, initial recovery of fungal strains at 15 minutes after using an inoculum containing glucose resulted in vps34Δ fungal burden equal to that of WT (WT, 5.0 × 10⁴ ± 0.3 × 10⁴ vs. vpsΔ, 5.0 × 10⁴ ± 0.4 × 10⁴ CFU/lung) but still showed a rapid clearance compared with WT over the 24-hour time period. In addition, simultaneous inoculation of mice with the vps34Δ mutant strain containing an equivalent admixture of WT cells did not protect mutant fungal cells from rapid clearance, suggesting that the protection of WT cells compared with mutants, as shown in Figure 2C, was not due primarily to suppression of clearance mechanisms by the WT cells. (Differentiation of cryptococcal WT from vps34Δ mutant cells in mixtures recovered from lung was made by laccase assay of 100 cells and confirmed by PCR using specific primers).

Figure 2

Role of VPS34 in cryptococcosis. Indicated strains were inoculated intravenously into 10 Swiss Albino mice (10⁶ CFU; A) or intranasally into 10 CBA/J mice (10⁵ CFU; B) and sacrificed when moribund. (C) Same as in B, but animals were sacrificed at the indicated times, lungs harvested, and fungal cells measured by CFU on YPD (values represented as ± SEM; n = 3 for each strain at each time point) (D) Same as in C except that mice were inoculated with either 10⁵ CFU of the indicated strains or a mixture of 10⁵ WT and 10⁵vps34Δ mutant strains in PBS containing 1.0% glucose, and cells were recovered from lungs and fungal strains identified in mixtures, as described in Methods.

In addition, the rapid clearance of the vps34Δ mutant cells was accompanied by a vigorous host response as measured by the early procytokine marker TNF-α. TNF-α is induced during human and murine cryptococcosis (20) and is normally present 3–7 days after pulmonary inoculation with intact fungal cells (21). In as little as 5 hours, TNF-α production was significantly induced after inoculation with the vps34Δ mutant (Supplemental Figure 2), most likely a result of host exposure to liberated fungal proteins from rapidly dying mutant cells, which has been shown previously to induce this cytokine (22). These data show the importance of the VPS34 locus during murine cryptococcosis.

Role of VPS34 and autophagy during macrophage residence of Cn. Since Cn is found predominantly within macrophages during lung infection, we examined the survival of the vps34Δ mutant strain after phagocytosis by a J774.16 macrophage cell line as previously described (23). As shown in Figure 3A, while WT strains exhibited good survival after macrophage phagocytosis, the vps34Δ mutant strain was rapidly killed over a similar time frame as that exhibited in mouse lungs. Control experiments showed that the vps34Δ mutant strain showed greater than 95% survival in tissue culture medium over the time period tested (data not shown). The clearance of the vps34Δ strain was also more rapid than mutants of the 2 best-known virulence factors, laccase (lac1Δ) and capsule (cap60Δ) (data not shown). However, the degree of phagocytosis was similar between WT and vps34Δ cells (WT: 32% ± 8%; vps34Δ: 40% ± 8%, mean ± SEM; P > 0.10). In addition, we conducted mixing experiments to determine if the relative survival of the WT mutant fungal strains was due to differential effects on host macrophages. The macrophage cell line was again induced as in experiments shown in Figure 3A but was inoculated with a 50:50 mixture of WT and vps34Δ cells, followed by removal of non-adherent cells and recovery of fungal cells at the indicated times by lysis of macrophages and quantitation on YPD agar. As shown in Figure 3B, addition of WT cells to the vps34Δ cells did not result in protection of mutant cells, suggesting that differences in survival between the strains was not due to suppression of macrophage killing mechanisms by WT cells. In addition, to confirm the mixing studies result, an equivalent macrophage-killing experiment was conducted using a 50:50 mixture of WT and vps34Δ cells, the latter labeled by a trace amount of calcofluor that was shown not to affect cell viability. After removal of non-adherent cells, macrophage cultures were followed by microscopy, and at the indicated times, macrophages containing both WT and mutant cells were assayed for fungal cell viability by FUN-1 staining, used previously to assess cryptococcal viability in macrophages (24). As shown in Figure 3C, at 30 minutes, both WT and mutant cells showed good viability evident by intact FUN-1. FUN-1 staining was also evident within the cytoplasm of the metabolically active macrophage as previously described during fungal phagocytosis (24). However, over the 24-hour observation period, the vps34Δ mutant cells selectively lost viability, whereas the WT cells were unaffected in the same macrophage, similar to that expected based on the CFU mixing studies (P = NS at 30 min, n = 45; P < 0.001 at 3 h, n = 54; P < 0.001 at 6 h, n = 42; P < 0.001 at 24 h, n = 53 fungal cells counted; see Figure 3D). In addition, infection of J774.16 macrophages with either WT or vps34Δ mutant cells showed equivalent expression and physiologic cleavage of the mammalian Atg8 homolog LC3-I into LC3-II in the macrophage cell line (Figure 3E), indicative of equivalent induction of macrophage autophagy, which occurs after IF-G–stimulated macrophage killing (25).

Figure 3

Survival of VPS34 strains in a J774. 16 macrophage cell line. (A) Fungal cells were opsonized with anti-capsular antibody, and 10⁵ cells were incubated with J774.16 macrophages for 30 min, washed extensively to remove non-phagocytosed cells, and at the indicated times, wells were washed with water containing 0.01% SDS and cultured for growth of fungal cells on YPD (n = 3 for each strain at each time point.) (B) Same as in A, but macrophages were incubated with 10⁵ of a 50:50 mixture of WT and vps34Δ mutant cells, and strains were identified from recovered mixtures as described in Methods. (C) Same as in B, but vps34Δ mutant strains were prestained with calcofluor (blue) prior to macrophage inoculation and fungal viability determined by FUN-1 epifluorescence (red) at the indicated times. Blue arrows indicated vps34Δ cells, yellow arrows indicate WT cells. Scale bars: 5 microns. (D) Analysis of viable fungal cells in macrophage by FUN-1 staining. Macrophages containing mixtures of WT (unlabeled, blue bars) and vps34Δ mutant (calcofluor-labeled, yellow bars) cells were identified and fungal cells scored by the presence of FUN-1 staining at the indicated times (*P < 0.001). (E) Macrophages were infected with either WT or vps34Δ mutant fungal cells as in A, and at 3 h, macrophage protein was harvested and subjected to 15% SDS-PAGE, followed by western blot using 1:1000 anti-LC3 antibody and developed using a horseradish peroxidase anti-rabbit antibody as described in Methods. Lane 1 corresponds to macrophages infected with WT cells and lane 2 to macrophages infected with vps34Δ mutant fungal cells.

Since VPS34 has been implicated in resistance to starvation due to a role in autophagy (8), we next performed survival studies in vitro under starvation conditions. Indeed, in whole-genome microarray experiments, the autophagy-related genes ATG3 and ATG9 were shown to be upregulated in macrophages, although fungal autophagy was not demonstrated (26). As shown in Figure 4A, the Cn vps34Δ cells showed poor tolerance to starvation and the rates of fungal cell death were comparable with that exhibited after phagocytosis by J774.16 cells (comparison of slopes of log survival of vps34Δ between data in Figure 2D and Figure 3A; P > 0.10). In contrast, vps34Δ mutant cells showed vigorous growth under other conditions expected in macrophages such as oxidative stress (Supplemental Figure 3). Examination of vps34Δ cells under starvation conditions revealed defects in properties attributable to the process of autophagy. For example, while WT cells displayed evidence of autophagy-related (Atg8) association into vesicular structures in response to starvation conditions, the vps34Δ mutant cells showed a diffuse cytoplasmic localization (Figure 4B). Atg8 is a microtubule-associated protein that is induced and associated with vesicles during autophagy, and this vesicle-associated phenomenon has been used extensively as a marker of autophagy (27). Interestingly, as shown in Figure 4B, Atg8-labeled autophagic bodies were distributed throughout each cell, as opposed to the typical appearance of autophagic bodies principally within large vacuoles in yeast (28), most likely due to formation of smaller multiple vacuolar compartments, which is a characteristic of Cn (29). However, addition of the vacuolar-modifying microtubular inhibitor nocodazole (30) resulted in the formation of larger vacuoles in Cn during starvation conditions, which allowed better visualization of motile autophagic bodies using DIC microscopy under starvation conditions. Using this protocol, autophagic bodies were observed within WT cells but not in identically treated vps34Δ mutant cells (Figure 4C and Supplemental Movie). Analysis of multiple cryptococcal cells revealed a number of autophagic bodies in WT cells (range, 0–5; mean ± SEM, 2.0 ± 0.4; n = 15), whereas none were observed in vps34 cells (n = 15; P < 0.001). Furthermore, because of the conserved nature of a number of autophagic proteins including Vps34 described above, Atg6 (22% similarity between Sc and Cn) and Atg8 (89% similarity between Sc and Cn) rabbit antibodies against Sc Vps34, Sc Atg6, and Sc Atg8 cells were tested and showed immunoreactive bands equal to their predicted sizes (101, 63, and 14.5 kDa, respectively), with cryptococcal extracts on western blot (Figure 4D and Supplemental Figure 4). These immuno-crossreactive antibodies were used to show formation of autophagic complexes during starvation, demonstrated by coimmunoprecipitation of Vps34 by the autophagic protein Atg6, as reported previously in yeast (9, 31). Recruitment of autophagy-related proteins by Vps34 is an important step in the formation of the autophagic complex in yeast, and these data support a similar mechanism in Cn.

Figure 4

Tolerance of VPS34 strains to nutrient starvation is associated with autophagic processes. (A) Starvation survival: Indicated log phase cells (grown in YPD) were incubated in YNB media without amino acids or ammonium sulfate at 37°C for the indicated times, followed by removal of aliquots, and cultured on YPD agar. (B) Indicated log phase cells were harvested (+gluc) or inoculated in YNB without amino acids or ammonium sulfate for 4 h at 37°C (no gluc) and labeled with anti-Atg8 antibody (+anti-Atg8) or rabbit polyclonal IgG (IgG alone), followed by anti-rabbit Alexa Fluor 594, and observed by for epifluorescence (Epi) or bright field (BF). (C) Indicated log phase (+gluc) or log phase cells incubated in YNB without amino acids or ammonium sulfate containing 1 mM PMSF and 10 μg/ml nocodazole for 4 h (no gluc) were harvested and subjected to DIC microscopy. Arrows point to autophagic bodies within vacuoles. (D) Indicated cells were subjected to starvation conditions as in C, and whole-cell lysates (left panel) were subjected to western blot using anti-Vps34 antibody or anti-Atg6 antibody or immunoprecipitated with anti-Atg6 antibody or rabbit IgG, washed and eluted, and subjected to western blot using the indicated antibodies.

To determine whether autophagic processes occurred during macrophage residence of cryptococcal cells, vesicularization of Atg8 was investigated after phagocytosis by the J774.16 macrophage cell line. Cryptococcal cells were identified by cell wall staining by the dye calcofluor, which also demonstrated some fluorescence with macrophages, as previously described (32). As shown in Figure 5A, J774.16 phagocytosis of WT but not vps34Δ mutant cells resulted in formation of Atg8-immunoreactive vesicles. This was visualized by epifluorescence after 3 hours of phagocytosis by the FM4/80-stained macrophage cell line and by the formation of Atg8-stained fungal autophagic bodies visualized by immunoelectron microscopy (Figure 5B). Most likely, induction of autophagy in phagocytosed fungal cells is the result of exposure to a nutrient-poor phagolysosome in the macrophage, suggested recently by a requirement for virulence of phosphoenolpyruvate carboxykinase, which is a gluconeogenic enzyme required to synthesize glucose under carbon starvation conditions (5).

Figure 5

Formation of autophagic bodies in Cn after macrophage phagocytosis. Indicated cells were subjected to phagocytosis by J774.16 cells as in Figure 3A, followed by fixation and embedding, and stained with anti-Atg8 antibody (+anti-Atg8) or rabbit IgG (IgG alone) and followed by an anti-rabbit Alexa Fluor 488 fluorescent antibody and anti-FM4/80 Alexa Fluor 594 antibody and observed for epifluorescence (A) or an anti-rabbit 5-nm gold-labeled antibody and observed by electron microscopy (B). AB, autophagic bodies; Mac, mouse macrophage; Vac, vacuole. Arrows point to intracellular fungal cells in A and gold particle labeling of Atg8 in B. Scale bars: 500 nm.

Reduced tolerance to starvation and virulence of an iATG8 knockdown strain. To molecularly dissect the contribution of autophagy to the VPS34 phenotype during cryptococcal infection, a reduction in autophagic capacity was produced by RNA interference of the autophagy-related gene ATG8. Control strains were produced by transformation of fungal cells with an equivalent plasmid in equivalent copy number but not containing the RNAi construct. The Cn iATG8 strain retained WT growth under nutrient-replete conditions at 37°C (see Methods) and retained the virulence factors laccase, urease, and capsule (Figure 6A). However, the iATG8 strain showed sensitivity to starvation (Figure 6B), although not quite as severe as the vps34Δ strain, which could be due to residual ATG8 expression in the iATG8 knockdown strain (see Supplemental Figure 4B). The iATG8 knockdown strain also showed attenuated virulence in the intravenous mouse model (Figure 6C) as well as in an intranasal model (Figure 6D) compared with the control strains, with only small amounts of residual iATG8 organisms noted in brains of sacrificed animals (<10³ CFU/brain). To determine whether cryptococcal Atg8 is expressed during human CNS disease, Atg8 immunoreactivity was demonstrated in fungal cells during infection of human brain (Figure 6, E and F), which showed vigorous Atg8 expression in fungal cells. In contrast, we did not see significant Atg8 reactive signal throughout brain tissue or on western blots of mouse brain (Supplemental Figure 4), suggesting that the Atg8 antibody was yeast specific. In summary, these studies demonstrate the role of VPS34 signaling and autophagy in starvation response and in the pathogenesis of cryptococcosis.

Figure 6

Role of ATG8 in murine and human cryptococcosis. (A) Cn cells transformed with the ATG8 RNAi construct (iATG8) or control plasmid (controls 1 and 2) assayed for laccase, urease, and capsule, as described in Methods. (B) Indicated mid–log phase strains incubated in YNB without amino acids or ammonium sulfate for the indicated times and cultured for growth of fungal cells on YPD. (C) Indicated strains (10⁶ CFU) were inoculated intravenously into Swiss Albino mice and followed until moribund. (D) Indicated strains (10⁵ CFU) were inoculated intranasally into CBA/J mice and followed until moribund. (E) Antibody staining of Cn-infected human brain tissue incubated with an anti-Atg8 antibody (+anti-Atg8) or rabbit IgG (IgG control) as well as an anti-rabbit horseradish peroxidase secondary antibody and developed with diaminobenzidine and counterstained with H&E. Lower panels show mucicarmine and H&E stain of equivalent tissue. (F) Higher magnification of intracellular fungal cells stained as in E. Arrows point to fungal cells. Scale bars: 10 μm.

In the present study, identification and deletion of a VPS34 gene associated with the virulence factor laccase allowed for the analysis of the role of inositol phosphate signaling, autophagy, and starvation response in the pathogen Cn. This expands further the genetic relationships between an immune modulatory virulence factor, laccase (33, 34), and starvation response programs such as autophagy that play a central role during virulence. Previously, regulation of laccase activity has been genetically linked to processes important under nutrient-limiting conditions such as gluconeogenesis (5) and high-affinity copper acquisition (6). Such a genetic association between laccase and attributes such as autophagy, important to survival during the search for nutrients, recalls the organism’s environmental niche within the hollows of trees where laccase is involved in the generation of energy by the use of alternative substrates including degradation of wood lignins in the absence of glucose (35). The existence of such strong genetic relationships between virulence determinants and environmental genetic programs are a testament to the success of opportunistic pathogens such as Cn that appear to effortlessly switch between an environmental saprophyte and a virulent pathogen. In addition, laccase’s extensive relationships to virulence determinants provides additional insights into why laccase and its product melanin has proven successful as a strong marker of pathogenic strains of Cryptococcus (36).

Autophagy is an evolutionarily conserved response to stress and has been linked to important protective roles in mammalian cells such as lifespan extension and protection from diverse human diseases including cancer, muscular disorders, neurodegeneration (e.g., Huntington, Alzheimer, and Parkinson diseases) (37–39). In the mammalian host-pathogen interaction, the role of autophagy has mostly favored the host, as induction of autophagy in macrophages has been found to be key in the control of replication of diverse prokaryotes such as Mycobacterium tuberculosis and Legionella pneumophila within autophagic vesicles (25, 40). However, the present studies indicate that PI3K-dependent induction of autophagy is also a tool used by the eukaryotic pathogen to ensure survival and effect host cell damage. Interestingly, the fungal pathogen Candida albicans has not been found to be dependent on autophagy for virulence (41), although a reduction in virulence was found by deletion of a vps34Δ homolog (42), which may suggest that the 2 pathogens encounter different host-cell environments leading to varied requirements for an autophagic response. Alternatively, this could be a reflection of pathogen-specific alterations in the host cell environment. For example, Cn liberates an extensive polysaccharide capsule within macrophages which has macrophage modulating properties that could alter the environment of the phagolysosome (43). In addition, Candida species may elicit specific protective properties that improve nutrient conditions similar to Mycobacterium within the host environment (44), making it less dependent on its own autophagosomal machinery. Regardless of mechanisms, these different requirements for autophagy between pathogens offer insights into the unique aspects of interactions between fungi of different species and the mammalian host.

Induction of autophagy in Cn was found to be dependent on the PI3K locus VPS34, as previously described in ascomycetous yeast (8). Interestingly, the almost WT growth of the cryptococcal vps34Δ mutant at 37°C distinguishes it from the temperature sensitivity of the homologous mutation in ascomycetous yeast and may reflect the evolution of higher eukaryotes. However, in spite of robust growth at 37°C, the cryptococcal vps34Δ mutant showed a remarkably fast clearance from lungs of mice, which was associated with a defect in tolerance of starvation and formation of autophagic bodies. Such extremely different survival patterns of the vps34Δ mutant in tissue culture media containing glucose compared with survival patterns in the host environment suggest that the former is not representative of host conditions. This discrepancy in pathogen response between these 2 conditions may have implications for the selection of conditions used to mimic the host environment, such as the determination of antifungal minimum inhibitory concentrations (MICs). These analyses, used to predict antifungal resistance during clinical infections, were conducted under nutrient-replete conditions as recommended by the National Committee for Clinical Laboratory Standards (45, 46). Unfortunately, cryptococcal fungal MICs have shown a remarkable inability to predict clinical endpoints (47), which could be partly due to differences in antifungal resistance within differing nutrient environments.

In summary, while these data suggest that an effective response, such as PI3K signaling and autophagy, by the pathogen to starvation encountered during infection may be detrimental to the host, they also suggest how attention to fungal stress response pathways may lead to more effective preventative or treatment modalities tailored to the unique environment of the host.

View Supplemental data

View Supplemental movie 1

View Supplemental movie 2

We would like to thank J. Moscat, P. Dennis, and G. Thomas for review of the manuscript. This work was supported in part by US Public Health Service Grant NIH-AI45995 and -AI49371. We also acknowledge the use of the C. neoformans Genome Project, Stanford Genome Technology Center ( http://www-sequence.stanford.edu), funded by the NIAID/NIH under cooperative agreement AI47087.

Address correspondence to: Peter R. Williamson, University of Illinois at Chicago, Room 888, Building 910, m/c 735, 808 S. Wood St., Chicago, Illinois 60612, USA. Phone: (312) 996-6070; Fax: (312) 413-1657; E-mail: [email protected].

Nonstandard abbreviations used: AM, alveolar macrophage; Cn, Cryptococcus neoformans; PI3P, phosphatidylinositol 3 phosphate; Sc, Saccharomyces cerevisiae.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J. Clin. Invest.118:1186–1197 (2008). doi:10.1172/JCI32053

1. Perfect, J.R., Wong, B., Chang, Y.C., Kwon-Chung, K.J., Williamson, P.R. 1998. Cryptococcus neoformans: virulence and host defences. Med. Mycol. 36(Suppl. 1):79-86.
2. Huffnagle, G.B., et al. 2000. Leukocyte recruitment during pulmonary Cryptococcus neoformans infection. Immunopharmacology. 48:231-236.
   View this article via: CrossRef PubMed Google Scholar
3. Feldmesser, M., Tucker, S., Casadevall, A. 2001. Intracellular parasitism of macrophages by Cryptococcus neoformans. Trends Microbiol. 9:273-278.
   View this article via: CrossRef PubMed Google Scholar
4. Chun, C., Liu, O., Madhani, H. 2007. A Link between virulence and homeostatic responses to hypoxia during infection by the human fungal pathogen Cryptococcus neoformans. PLOS Pathogens. 3:e22.
   View this article via: PubMed Google Scholar
5. Panepinto, J., et al. 2005. The DEAD-box RNA helicase Vad1 regulates multiple virulence-associated genes in Cryptococcus neoformans. J. Clin. Invest. 115:632-641.
   View this article via: JCI PubMed Google Scholar
6. Waterman, S., et al. 2007. Role of a CUF1-CTR4 copper regulatory axis in the virulence of Cryptococcus neoformans. J. Clin. Invest. 117:794-802.
   View this article via: JCI CrossRef PubMed Google Scholar
7. Burda, P., Padilla, S., Sarkear, S., Emr, S. 2002. Retromer function in endosome-to-Golgi retrograde transport is regulated by the yeast Vps35 PtdIns 3-kinase. J. Cell Sci. 115:3889-3900.
   View this article via: CrossRef PubMed Google Scholar
8. Kihara, A., Noda, T., Ishihara, N., Ohsumi, Y. 2001. Two distinct vps34 phohatidylinositol 3-kinase complexes function in autophagy and carboxyeptidase Y sorting in Sacharomyces cerevisiae. J. Cell Biol. 152:519-530.
   View this article via: CrossRef PubMed Google Scholar
9. Kametaka, S., Okano, T., Ohsumi, M., Ohsumi, Y. 1998. Apg14p and Apg6/Vps30p form a protein complex essential for autophagy in the yeast, Saccharomyces cerevisiae. J. Biol. Chem. 273:22284-22291.
   View this article via: CrossRef PubMed Google Scholar
10. Ichimura, Y., et al. 2000. A ubiquitin-like system mediates protein lipidation. Nature. 408:488-492.
    View this article via: CrossRef PubMed Google Scholar
11. Noda, T., Ohsumi, Y. 1998. Tor, a phosphatidylinositol kinase homologue, controls autophagy in yeast. J. Biol. Chem. 273:3963-3966.
    View this article via: CrossRef PubMed Google Scholar
12. Ma, L., Chen, Z., Erdjument-Bromage, H., Tempst, P., Pandolfi, P. 2005. Phosphorylation and functional inactivation of TSC2 by Erk: Implications for tuberous sclerosis and cancer pathogenesis. Cell. 121:179-193.
    View this article via: CrossRef PubMed Google Scholar
13. van Slegtenhorst, M., Carr, E., Stoyanova, R., Kruger, W.D., Henske, E.P. 2004. Tsc1 and tsc2 regulate arginine uptake and metabolism in Schizosaccharomyces pombe. J. Biol. Chem. 279:12706-12713.
    View this article via: CrossRef PubMed Google Scholar
14. Shelton, S., et al. 2003. Saccharomyces cerevisiae contains a type II phosphoinositide 4-kinase. Biochem. J. 371:533-540.
    View this article via: CrossRef PubMed Google Scholar
15. Chen, H., Salopek, T., Jimbow, K. 2001. The role of phosphoinositide 3-kinase in the sorting and transport of newly synthesized tyrosinase-related protein-1 (TRP-1). J. Investig. Dermatol. Symp. Proc. 6:105-114.
    View this article via: CrossRef PubMed Google Scholar
16. Herman, P., Emr, S. 1990. Characterization of VPS34, a gene required for vacuolar protein sorting and vacuole segregation in Saccharomyces cerevisiae. Mol. Cell. Biol. 10:6742-6754.
    View this article via: PubMed Google Scholar
17. Liu, X., Hu, G., Panepinto, J., Williamson, P. 2006. Role of a VPS41 homolog in starvation response and virulence of Cryptococcus neoformans. Mol. Microbiol. 61:1132-1146.
    View this article via: CrossRef PubMed Google Scholar
18. Pukkila-Worley, R., et al. 2005. Transcriptional network of multiple capsule and melanin genes governed by the Cryptococcus neoformans cyclic AMP cascade. Eukaryot. Cell. 4:190-201.
    View this article via: CrossRef PubMed Google Scholar
19. Moyrand, F., Klaproth, B., Himmelreich, U., Dromer, F., Janbon, G. 2002. Isolation and characterization of capsule structure mutant strains of Cryptococcus neoformans. Mol. Microbiol. 45:837-849.
    View this article via: CrossRef PubMed Google Scholar
20. Lortholary, O., et al. 1999. Cytokine profiles of AIDS patients are similar to those of mice with disseminated Cryptococcus neoformans infection. Infect. Immun. 67:6314-6320.
    View this article via: PubMed Google Scholar
21. Herring, A.C., Lee, J., McDonald, R.A., Toews, G.B., Huffnagle, G.B. 2002. Induction of interleukin-12 and gamma interferon requires tumor necrosis factor alpha for protective T1-cell-mediated immunity to pulmonary Cryptococcus neoformans infection. Infect. Immun. 70:2959-2964.
    View this article via: CrossRef PubMed Google Scholar
22. Lindell, D., Ballinger, M., McDonald, R., Toews, G., Huffnagle, G. 2006. Diversity of the T-cell response to pulmonary Cryptococcus neoformans infection. Infect. Immun. 74:4538-4548.
    View this article via: CrossRef PubMed Google Scholar
23. Feldmesser, M., Kress, Y., Novikoff, P., Casadevall, A. 2000. Cryptococcus neoformans is a facultative intracellular pathogen in murine pulmonary infection. Infect. Immun. 68:4225-4237.
    View this article via: CrossRef PubMed Google Scholar
24. Henry-Stanley, M., Garni, R., Wells, C. 2004. Adaptation of FUN-1 and calcofluor white stains to assess the ability of viable and nonviable yeast to adhere to and be internalized by cultured mammalian cells. J. Microbiol. Methods. 59:289-292.
    View this article via: CrossRef PubMed Google Scholar
25. Gutierrez, M., et al. 2004. Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages. Cell. 119:753-766.
    View this article via: CrossRef PubMed Google Scholar
26. Fan, W., Kraus, P., Boily, M., Heitman, J. 2005. Cryptococcus neoformans gene expression during murine macrophage infection. Eukaryot. Cell. 4:1420-1433.
    View this article via: CrossRef PubMed Google Scholar
27. Lang, T., et al. 1998. Aut2p and Aut7p, two novel microtubule-associated proteins are essential for delivery of autophagic vesicles to the vacuole. EMBO J. 17:3597-3607.
    View this article via: CrossRef PubMed Google Scholar
28. Kirisako, T., et al. 1999. Formation process of autophagosome is traced with Apg8/Aut7p in yeast. J. Cell Biol. 147:435-446.
    View this article via: CrossRef PubMed Google Scholar
29. Erickson, T., et al. 2001. Multiple virulence factors of Cryptococcus neoformans are dependent on VPH1. Mol. Microbiol. 42:1121-1131.
    View this article via: CrossRef PubMed Google Scholar
30. Guthrie, B., Wickner, W. 1988. Yeast vacuoles fragment when microtubules are disrupted. J. Cell Biol. 107:115-120.
    View this article via: CrossRef PubMed Google Scholar
31. Kim, J., Huang, W., Stromhaug, P., Klionsky, D. 2002. Convergence of multiple autophagy and cytoplasm to vacuole targeting components to a perivacuolar membrane compartment prior to de novo vesicle formation. J. Biol. Chem. 277:763-773.
    View this article via: CrossRef PubMed Google Scholar
32. Shin, Y., Kim, K., Paik, Y. 2005. Alterations of protein expression in macrophages in response to Candida albicans infection. Mol. Cells. 20:271-279.
    View this article via: PubMed Google Scholar
33. Liu, L., Tewari, R.P., Williamson, P.R. 1999. Laccase protects Cryptococcus neoformans from antifungal activity of alveolar macrophages. Infect. Immun. 67:6034-6039.
    View this article via: PubMed Google Scholar
34. Mednick, A.J., Nosanchuk, J.D., Casadevall, A. 2005. Melanization of Cryptococcus neoformans affects lung inflammatory responses during cryptococcal infection. Infect. Immun. 73:2012-2019.
    View this article via: CrossRef PubMed Google Scholar
35. Randhawa, H.S., Mussa, A.Y., Khan, Z.U. 2001. Decaying wood in tree trunk hollows as a natural substrate for Cryptococcus neoformans and other yeast-like fungi of clinical interest. Mycopathologia. 151:63-69.
    View this article via: CrossRef PubMed Google Scholar
36. Racicot, T.A., Bulmer, G.S. 1985. Comparison of Media for the Isolation of Cryptococcus neoformans. Appl. Environ. Microbiol. 50:548-549.
    View this article via: PubMed Google Scholar
37. Levine, B., Klionsky, D.J. 2004. Development by self-digestion: molecular mechanisms and biological functions of autophagy. Dev. Cell. 6:463-477.
    View this article via: CrossRef PubMed Google Scholar
38. Cuervo, A. 2004. Autophagy: in sickness and in health. Trends Cell Biol. 14:70-77.
    View this article via: CrossRef PubMed Google Scholar
39. Shintani, T., Klionsky, D.J. 2004. Autophagy in health and disease: a double-edged sword. Science. 306:990-995.
    View this article via: CrossRef PubMed Google Scholar
40. Amer, A., Swanson, M. 2005. Autophagy is an immediate macrophage response to Legionella pneumophila. Cell. Microbiol. 7:765-778.
    View this article via: CrossRef PubMed Google Scholar
41. Palmer, G., Kelly, M., Sturtevant, J. 2007. Autophagy in the pathogen Candida albicans. Microbiology. 153:51-58.
    View this article via: CrossRef PubMed Google Scholar
42. Kitanovic, A., et al. 2005. Phosphatidylinositol 3-kinase VPS34 of Candida albicans is involved in filamentous growth, secretion of aspartic proteases, and intracellular detoxification. FEMS Yeast Res. 5:431-439.
    View this article via: CrossRef PubMed Google Scholar
43. Monari, C., et al. 2005. Cryptococcus neoformans capsular glucuronoxylomannan induces expression of fas ligand in macrophages. J. Immunol. 174:3461-3468.
    View this article via: PubMed Google Scholar
44. Deretic, V., et al. 2006. Mycobacterium tuberculosis inhibition of phagolysosome biogenesis and autophagy as a host defence mechanism. Cell. Microbiol. 8:719-727.
    View this article via: CrossRef PubMed Google Scholar
45. Barchiesi, F., Colombo, A.L., McGough, D.A., Rinaldi, M.G. 1994. Comparative study of broth macrodilution and microdilution techniques for in vitro antifungal susceptibility testing of yeasts by using the National Committee for Clinical Laboratory Standards’ proposed standard. J. Clin. Microbiol. 32:2494-2500.
    View this article via: PubMed Google Scholar
46. National Committee on Clinical and Laboratory Standards. 2002. Reference method for broth dilution antifungal susceptibility testing of yeasts; approved standard — second edition. NCCLS document M27-A2. http://www.nccls.org/source/orders/free/m27-a2.pdf .
47. Dannaoui, E., et al. 2007. Results obtained with various antifungal susceptibility testing methods do not predict early clinical outcome in patients with cryptococcosis. Antimicrob. Agents Chemother. 50:2464-2470.
    View this article via: PubMed Google Scholar
48. Zhu, X., Williamson, P.R. 2003. A CLC-type chloride channel gene is required for laccase activity and virulence in Cryptococcus neoformans. Mol. Microbiol. 50:1271-1281.
    View this article via: CrossRef PubMed Google Scholar
49. Varma, A., Edman, J.C., Kwon-Chung, K.J. 1992. Molecular and genetic analysis of URA5 transformants of Cryptococcus neoformans. Infect. Immun. 60:1101-1108.
    View this article via: PubMed Google Scholar
50. Cox, G.M., Toffaletti, D.L., Perfect, J.R. 1996. Dominant selection system for use in Cryptococcus neoformans. J. Med. Vet. Mycol. 34:385-391.
    View this article via: CrossRef PubMed Google Scholar
51. Storz, P., Doppler, H., Wernig, A., Pfizenmaier, K., Muller, G. 1999. Cross-talk mechanisms in the development of insulin resistance of skeletal muscle cells palmitate rather than tumour necrosis factor inhibits insulin-dependent protein kinase B (PKB)/Akt stimulation and glucose uptake. Eur. J. Biochem. 266:17-25.
    View this article via: CrossRef PubMed Google Scholar
52. Bain, J., Stubberfied, C., Gow, N. 2001. Ura-status-dependent adhesion of Candida albicans mutants. FEMS Microbiol. Lett. 204:323-328.
    View this article via: CrossRef PubMed Google Scholar
53. Panepinto, J., Williamson, P. 2006. Intersection of fungal fitness and virulence in Cryptococcus neoformans. FEMS Yeast Res. 6:489-498.
    View this article via: CrossRef PubMed Google Scholar
54. Tucker, S.C., Casadevall, A. 2002. Replication of Cryptococcus neoformans in macrophages is accompanied by phagosomal permeabilization and accumulation of vesicles containing polysaccharide in the cytoplasm. Proc. Natl. Acad. Sci. U. S. A. 99:3165-3170.
    View this article via: CrossRef PubMed Google Scholar
55. Levitz, S., DiBenedetto, D., Diamond, R. 1987. A rapid fluorescent assay to distinguish attached from phagocytized yeast particles. J. Immunol. Methods. 101:37.
    View this article via: PubMed Google Scholar
56. Kozel, T.R., Pfrommer, G.S., Guerlain, A.S., Highison, B.A., Highison, G.J. 1988. Strain variation in phagocytosis of Cryptococcus neoformans: dissociation of susceptibility to phagocytosis from activation and binding of opsonic fragments of C3. Infect. Immun. 56:2794-2800.
    View this article via: PubMed Google Scholar
57. Zhu, X., Gibbons, J., Garcia-Rivera, J., Casadevall, A., Williamson, P.R. 2001. Laccase of Cryptococcus neoformans is a cell wall-associated virulence factor. Infect. Immun. 69:5589-5596.
    View this article via: CrossRef PubMed Google Scholar
58. Varma, A., Kwon-Chung, K.J. 1991. Rapid method to extract DNA from Cryptococcus neoformans. J. Clin. Microbiol. 29:810-812.
    View this article via: PubMed Google Scholar
59. Mizushima, N. 2004. Methods for monitoring autophagy. Int. J. Biochem. Cell Biol. 36:2491-2502.
    View this article via: CrossRef PubMed Google Scholar
60. Zhang, S., et al. 2006. The Hsp70 member, Ssa1 acts as a DNA-binding transcriptional co-activator in Cryptococcus neoformans. Mol. Microbiol. 62:1090-1101.
    View this article via: CrossRef PubMed Google Scholar
61. Noverr, M.C., Williamson, P.R., Fajardo, R.S., Huffnagle, G.B. 2004. CNLAC1 is required for extrapulmonary dissemination of Cryptococcus neoformans but not pulmonary persistence. Infect. Immun. 72:1693-1699.
    View this article via: CrossRef PubMed Google Scholar
62. Cox, G., Mukherjee, J., Cole, G., Casadevall, A., Perfect, J. 2000. Urease as a virulence factor in experimental cryptococcosis. Infect. Immun. 68:443-448.
    View this article via: CrossRef PubMed Google Scholar
63. Liu, L., Wakamatsu, K., Ito, S., Williamson, P.R. 1999. Catecholamine oxidative products, but not melanin, are produced by Cryptococcus neoformans during neuropathogenesis in mice. Infect. Immun. 67:108-112.
    View this article via: PubMed Google Scholar
64. Salas, S.D., Bennett, J.E., Kwon-Chung, K.J., Perfect, J.R., Williamson, P.R. 1996. Effect of the laccase gene CNLAC1, on virulence of Cryptococcus neoformans. J. Exp. Med. 184:377-386.
    View this article via: CrossRef PubMed Google Scholar
65. Cox, G., et al. 2003. Superoxide dismutase influences the virulence of Cryptococcus neoformans by affecting growth within macrophages. Infect. Immun. 71:173-180.
    View this article via: PubMed CrossRef Google Scholar