Candida albicans Chitin Increases Arginase-1 Activity in Human Macrophages, with an Impact on Macrophage Antimicrobial Functions

ABSTRACT   The opportunistic human fungal pathogen Candida albicans can cause a variety of diseases, ranging from superficial mucosal infections to life-threatening systemic infections. Phagocytic cells of the innate immune response, such as neutrophils and macrophages, are important first-line responders to an infection and generate reactive oxygen and nitrogen species as part of their protective antimicrobial response. During an infection, host cells generate nitric oxide through the enzyme inducible nitric oxide synthase (iNOS) to kill the invading pathogen. Inside the phagocyte, iNOS competes with the enzyme arginase-1 for a common substrate, the amino acid l-arginine. Several pathogenic species, including bacteria and parasitic protozoans, actively modulate the production of nitric oxide by inducing their own arginases or the host’s arginase activity to prevent the conversion of l-arginine to nitric oxide. We report here that C. albicans blocks nitric oxide production in human-monocyte-derived macrophages by induction of host arginase activity. We further determined that purified chitin (a fungal cell wall polysaccharide) and increased chitin exposure at the fungal cell wall surface induces this host arginase activity. Blocking the C. albicans-induced arginase activity with the arginase-specific substrate inhibitor Nω-hydroxy-nor-arginine (nor-NOHA) or the chitinase inhibitor bisdionin F restored nitric oxide production and increased the efficiency of fungal killing. Moreover, we determined that C. albicans influences macrophage polarization from a classically activated phenotype toward an alternatively activated phenotype, thereby reducing antimicrobial functions and mediating fungal survival. Therefore, C. albicans modulates l-arginine metabolism in macrophages during an infection, potentiating its own survival.

L-arginine. Several pathogenic species, including bacteria and parasitic protozoans, actively modulate the production of nitric oxide by inducing their own arginases or the host's arginase activity to prevent the conversion of L-arginine to nitric oxide. We report here that C. albicans blocks nitric oxide production in human-monocyte-derived macrophages by induction of host arginase activity. We further determined that purified chitin (a fungal cell wall polysaccharide) and increased chitin exposure at the fungal cell wall surface induces this host arginase activity. Blocking the C. albicans-induced arginase activity with the arginase-specific substrate inhibitor N-hydroxy-nor-arginine (nor-NOHA) or the chitinase inhibitor bisdionin F restored nitric oxide production and increased the efficiency of fungal killing. Moreover, we determined that C. albicans influences macrophage polarization from a classically activated phenotype toward an alternatively activated phenotype, thereby reducing antimicrobial functions and mediating fungal survival. Therefore, C. albicans modulates L-arginine metabolism in macrophages during an infection, potentiating its own survival. IMPORTANCE The availability and metabolism of amino acids are increasingly recognized as crucial regulators of immune functions. In acute infections, the conversion of the "conditionally essential" amino acid L-arginine by the inducible nitric oxide synthase to nitric oxide is a resistance factor that is produced by the host to fight pathogens. Manipulation of these host defense mechanisms by the pathogen can be key to successful host invasion. We show here that the human opportunistic fungal pathogen Candida albicans influences L-arginine availability for nitric oxide production by induction of the substrate-competing host enzyme arginase-1. This led to a reduced production of nitric oxide and, moreover, reduced eradication of the fungus by human macrophages. We demonstrate that blocking of host arginase-1 activity restored nitric oxide production and increased the killing potential of macrophages. These results highlight the therapeutic potential of L-arginine metabolism in fungal diseases. C andida albicans is an opportunistic fungal pathogen that frequently colonizes the mucosal surfaces of healthy individuals without causing infection (1). However, in individuals in whom the normally suppressive endogenous bacterial microflora is

RESULTS
C. albicans influences host arginase activity. We hypothesized that C. albicans promotes its own survival in phagocytes by manipulating the availability of L-arginine for NO synthesis. We therefore analyzed the induction of iNOS and arginase-1 protein expression in IFN-␥and LPS-activated human monocyte-derived macrophages, cocultured with C. albicans (multiplicity of infection [MOI] ϭ 1) for 3 h by immunoblotting (Fig. 1A). C. albicans significantly increased both iNOS and arginase-1 (Arg-1) protein expression, although the total increase in protein level was higher for Arg-1 (7.43fold Ϯ 1.61-fold) than for iNOS (1.21-fold Ϯ 0.05-fold). We next examined NO production by measuring the amount of nitrite in the supernatants of the same cocultures (Fig. 1B). Interestingly, although a slightly increased iNOS protein level was detected, we observed a 50% reduction of NO released from the activated macrophages in the presence of C. albicans (Fig. 1B). However, the observed increased arginase-1 protein expression led to a 40% increase in arginase activity in the analyzed cell lysates (Fig. 1C).
To prove that the reduced NO production was due to the observed host arginase-1 induction and not due to upregulation of the fungal NO detoxification mechanism, we added the selective arginase-1 inhibitor N-hydroxy-nor-arginine (nor-NOHA) to block arginase activity (11). This restored C. albicans-suppressed NO levels in cocultured macrophages to levels observed in macrophages that were not exposed to C. albicans (Fig. 1D). The increased arginase activity in macrophages cocultured with C. albicans was also reduced by nor-NOHA to levels observed in nor-NOHA-treated activated macrophages in the absence of C. albicans (Fig. 1E). Therefore, we conclude that the presence of C. albicans increased arginase-1 activity in IFN-␥and LPS-activated human monocyte-derived macrophages, limiting the availability of L-arginine for NO production by iNOS.
Increased arginase activity is chitin mediated and leads to reduced macrophage antimicrobial function. Next, we determined the impact of increased arginase activity and blocked NO production on antimicrobial macrophage function and C. albicans germ tube formation. Blocking arginase activity with nor-NOHA significantly increased fungal uptake ( Fig. 2A) and killing (Fig. 2B) by IFN-␥-and LPS-activated human monocyte-derived macrophages. Moreover, reduced arginase activity led to decreased C. albicans hyphal extension inside macrophage phagosomes (Fig. 2C). Germ tube formation of nonphagocytosed C. albicans was not affected by nor-NOHA treatment (Fig. 2D). Incubation of C. albicans with nor-NOHA in the absence of macrophages did not impact Candida viability (Fig. 2B) or germ tube formation ( Fig. 2C and D).
Arginase-1 induction in macrophages can be induced by type 2 cytokines, such as IL-4, IL-10, and IL-13 (7). Chitin is a cell wall polysaccharide which is a potent stimulator of allergic responses, and we showed previously that purified fungal chitin induces IL-10 secretion in human myeloid cells (12). Therefore, we tested the ability of purified fungal chitin to induce arginase protein expression and arginase activity in IFN-␥-and LPS-activated human monocyte-derived macrophages (Fig. 3). C. albicans-derived chitin increased arginase-1 protein levels but did not induce iNOS expression (Fig. 3A). Arginase activity was, however, significantly increased in chitin-stimulated macrophages (Fig. 3B), whereas NO synthesis was reduced (Fig. 3C). Next, we determined whether chitin on the surface of C. albicans was accessible for immune recognition by macrophages. We stained C. albicans cells that had been cocultured with IFN-␥-and LPS-activated human monocyte-derived macrophages for 3 h, both phagocytosed and nonphagocytosed cells, for ␤-glucan (red) and chitin (green) exposure (Fig. 4A). C. albicans cells recovered from macrophage phagosomes showed significantly increased chitin surface exposure, but no major changes in ␤-glucan exposure were observed (Fig. 4B). To test whether chitin released from phagocytosed C. albicans by the activity   Therefore, exposed chitin in the cell wall of C. albicans led to increased activity of arginase-1 in IFN-␥/LPS-activated human monocyte-derived macrophages, thereby reducing C. albicans killing.
A chitin-mediated increase in arginase activity requires active fungal cell wall remodeling and correlates with C. albicans virulence. Increased chitin surface exposure may be driven by C. albicans cell wall remodeling in response to the phagosome environment, similar to the increased chitin deposition in the cell wall described after an attack by neutrophils (13) and/or unmasking by the actions of host enzymes or exposure to echinocandins (e.g., caspofungin) (14). We therefore coincubated IFN-␥/LPS-activated human monocyte-derived macrophages with nonviable yeast cells (thimerosal-treated yeast cells [THY]) for 3 h and determined macrophage NO production and arginase-1 activity, as well as the surface exposure of ␤-glucan and chitin of control samples, for both nonphagocytosed and phagocytosed THY (Fig. 5). Nonviable C. albicans cells did not significantly reduce NO production ( Fig. 5A) or induce arginase activity (Fig. 5B). Cell wall staining revealed that phagocytosis by macrophages did not significantly alter the cell surface presentation of ␤-glucan or chitin in nonviable yeast cells during the experimental time course (3 h) (Fig. 5C).
C. albicans cell wall remodeling during infection is predicted to have a significant impact on signaling pathways important for mounting a protective immune response Chitin Modulates Macrophage Arginase-1 Activity ® (15,16). In vivo infection with different C. albicans strains from different multilocus sequence type (MLST) clades revealed that mice deficient in the major ␤-glucan receptor dectin-1 can exhibit a protective immune response in a strain-dependent manner. Interestingly, the pathogenicity of the strains used in this study correlated with increased chitin levels observed in vivo (16). We therefore tested four additional C. albicans strains with different virulence attributes for their ability to increase macrophage arginase activity and to reduce NO production (Table 1; Fig. 5). Levels of induction of arginase activity and reduction of NO synthesis varied between the strains and correlated with their reported in vivo pathogenicity (Table 1)  We therefore conclude that increased chitin exposure is actively driven by fungal cell wall remodeling and that the levels of arginase activity induction and reduction of  NO synthesis are C. albicans strain specific and correlate with the virulence attributes of the tested strains.
Purified fungal chitin and C. albicans surface-exposed chitin influence the macrophage phenotype. Macrophage activation depends on stimuli present in their immediate environment. Macrophages adapt to environmental signals, such as cytokines produced by surrounding cells (7). This leads to flexibility in macrophage programing, enabling macrophages to switch readily from one functional phenotype to another in response to changing microenvironmental signals. Pathogen-and damageassociated molecular patterns, together with IFN-␥, polarize macrophages toward an M1 phenotype, characterized by the production of ROS and RNS, which facilitate the killing of microbial pathogens. In contrast to M1 macrophages, alternatively activated M2 macrophages suppress inflammation and facilitate wound repair. High arginase activity has been described for the M2 macrophage phenotype, which is important in facilitating the production of prolines required for wound healing (18). To determine whether the chitin-induced increase in arginase activity is also impacted the macrophage phenotype, we analyzed the expression of macrophage phenotype-specific markers by flow cytometry to distinguish macrophage phenotypes (Fig. 6).
Up to 70% of the in vitro-differentiated IFN-␥/LPS-activated macrophages expressed markers characteristic of the M1 phenotype (Fig. 6). Stimulation with purified fungal chitin significantly reduced this M1 population while simultaneously significantly increasing the proportion of cells expressing M2 markers (Fig. 6B). To overcome technical restrictions on using live C. albicans cells in flow cytometry due to hyphal formation, we mimicked the increased chitin surface expression observed in live C. albicans cells (Fig. 4) by pretreating yeast cells with caspofungin, a cell wall-targeting antifungal drug which increases the exposure of both chitin and ␤-glucan (19,20), before inactivating the yeast cells chemically with thimerosal to preserve the cell wall changes. Macrophages stimulated with untreated C. albicans cells showed no change in polarization markers, whereas macrophages stimulated with caspofungin-treated yeast cells showed a decrease in the M1 population and a significant increase in the M2 macrophage population (Fig. 6C), which was comparable to levels induced by purified fungal chitin alone (Fig. 6B). Therefore, the recognition of purified fungal chitin and exposed cell wall chitin shifted the equilibrium from M1 to M2 macrophages.
Macrophage phenotype is an important determinant of the antifungal response to C. albicans. Immune surveillance and tissue homeostasis is an important task of tissue-resident macrophages. Tissue-resident macrophages, alongside other tissue-resident cells, such as stromal cells, mast cells, and dendritic cells, respond to pathogens by initiating inflammation. The initial pathogen encounter drives the influx of neutrophils and inflammatory monocytes, the source of inflammatory macrophages  (21). Tissue-resident macrophages are classified as M2-like due to their functional similarity to M2 macrophages. We therefore tested the influence of C. albicans on arginase activity and NO production in tissueresident macrophages. We repeated the coincubation experiments with nonactivated (resting) and IFN-␥/LPS-activated human monocyte-derived macrophages ( Fig. 7A and  B) and, in addition, challenged nonactivated and IFN-␥/LPS-activated murine peritoneal (tissue-resident) macrophages ( Fig. 7C and D) with C. albicans. We found that resting and tissue-resident macrophages secreted very low levels of nitric oxide ( Fig. 7A and C) and that NO production increased upon activation with IFN-␥/LPS (Fig. 7B and D), which supports the results of prior reports (5). Coincubation with C. albicans did not impact NO levels in nonactivated macrophages ( Fig. 7A and C). However, coincubation with C. albicans significantly increased the arginase activity in both nonactivated human macrophages and mouse peritoneal macrophages ( Fig. 7B and D).
Alternative activation of macrophages is complex, and several macrophage subtypes have been described in the literature (7,18). In general, M2 macrophages are marked by increased anti-inflammatory cytokine production (e.g., IL-10, TGF-␤, IL-1RA) and increased expression of C-type lectins (CTLs) (e.g., dectin-1, dendritic cell-specific intercellular adhesion molecule-3-grabbing nonintegrin [DC-SIGN], and mannose receptor) and scavenging receptors for the uptake of cellular debris (7). Alternative activation of macrophages due to Th2 responses (e.g., IL-4 and IL-13) has been shown to promote fungal persistence, allergy, and disease relapse, and neutralization of endogenous IL-4 has beneficial effects on the survival of mice systemically infected with C. albicans (22). When we challenged alternatively (IL-4) activated macrophages with live C. albicans, we found that these macrophages were more potent in their ability to phagocytose C. albicans than IFN-␥/LPS-activated (M1) macrophages (Fig. 8A). Although hypha formation of nonphagocytosed (Fig. 8B) and phagocytosed C. albicans (Fig. 8C) cells was significantly suppressed compared to that in M1 macrophages, alternatively activated macrophages showed a markedly reduced ability to kill C. albicans (Fig. 8B).
In summary, classical macrophage activation is important for pathogen eradication, and an increased alternative activation of macrophages may result in increased C. albicans persistence following tissue invasion.

DISCUSSION
Amino acid metabolism has been increasingly recognized as an important modulator of immune responses (23). The amino acid L-arginine is classified as "conditionally essential," depending on the developmental state and health status of an individual (6). During an infection, L-arginine availability has been shown to be a crucial regulator of immune functions, and several pathogens, such as Salmonella and Leishmania spp., can impact host L-arginine metabolism in various ways (4). In this study, we show that the human-pathogenic fungus C. albicans alters macrophage nitric oxide production by affecting host arginase-1 expression and activity. Furthermore, we identified fungal cell wall chitin as an important component that mediates arginase-1 induction in human macrophages.
Previous studies have reported that C. albicans can suppress nitric oxide production in IFN-␥/LPS-activated primary mouse macrophages and in activated cell lines (RAW264.7 and J774.A1) (8-10). A common observation in these studies was that NO inhibition was highest when live C. albicans cells were in direct contact with macrophages. A decrease in iNOS expression (mRNA/protein) was observed with C. albicans (8, 10) and coculture (9) supernatants. In contrast, our results with human IFN-␥/LPSactivated macrophages showed a slight increase in iNOS protein induced by C. albicans, but this did not result in higher enzyme activity as measured by nitric oxide production. Therefore, the reduced NO response of human macrophages to C. albicans was not likely to be due to transcriptional downregulation of the iNOS gene. Inhibition of NO production was also not mediated by Candida-generated competitive inhibitors (9). We determined that reduced NO production was most likely to be due to competition for the L-arginine substrate by the host arginase-1 enzyme. Blocking arginase-1-mediated consumption of its substrate L-arginine by the substrate inhibitor nor-NOHA restored NO production in C. albicans-challenged human macrophages and decreased C. albicans survival. Using a transwell system, Collette et al. (9) showed that direct cell-cell contact was not necessary for reduced nitric oxide production and that it also occurred in response to conditioned cell-free supernatants from macrophage-Candida cocultures. They described a "secreted" mediator as a small, aqueous, and heat-stable compound. Chitin is a heat-stable carbohydrate that is released from the fungal cell wall by human chitinases during macrophage-Candida interactions (12) and is, therefore, a credible candidate for this class of mediator. Chitin is also known to induce alternatively activated macrophages in mouse lungs, which are marked by a high expression of the enzyme arginase-1 (24). Although chitin in the native cell wall is covered by a layer of mannoproteins, we showed that chitin can be exposed during C. albicans-macrophage interactions and that C. albicans chitin is a strong inducer of arginase-1 activity in human macrophages. Exposure of underlying cell wall layers has also been shown to occur when C. albicans is attacked by neutrophils (13). Using a specific inhibitor (bisdionin F) for the human acidic mammalian chitinase (AMCase) to prevent chitin release from the cell walls of phagocytosed C. albicans cells, we demonstrated that chitin release negatively influenced macrophage antimicrobial functions. Moreover, chitin recognition influenced macrophage polarization, pushing macrophages from a classically activated, antimicrobial phenotype toward an alternatively activated phenotype, thereby decreasing antimicrobial activity. The ability of C. albicans to influence macrophage activation and phenotype has been show previously (25,26), although the impact on antimicrobial functions and C. albicans survival was not determined.
Alternative activation of macrophages by C. albicans in the context of the local immune response may have either beneficial or detrimental effects for the host. The activation of a protective Th1/Th17 response is important for the host to clear an infection, but uncontrolled inflammation leads to overinduction and immunopathology. Maintaining the correct balance of the host's immune response is therefore crucial for the clearance of an infection and the resolution of inflammation. In this context, the predominant, but not exclusive, induction of classical macrophages has been shown to be protective against several fungal species, including Cryptococcus neoformans, Aspergillus fumigatus, and C. albicans, whereas a predominant induction of alternative macrophages is implicated in the protection against Pneumocystis spp. (18). Interestingly, the recognition of chitin and the activity of the mammalian chitinase chitotriosidase have recently been shown to mediate pathological Th2-associated inflammation during C. neoformans lung infection in vivo (27). Further, the abrogation of IL-4 receptor ␣-dependent alternative activation of macrophages can mediate resistance to pulmonary C. neoformans infection marked by reduced arginase-1 expression and increased nitric oxide production (28). The importance of developing a protective Th1/Th17 response over a nonprotective Th2 response in clearing fungal infections has been described previously (29), and the negative impact of IL-4 in macrophage activation during fungal infection is linked to fungal persistence (22).
We show here that alternatively (IL-4) activated human macrophages have increased phagocytic capabilities and an enhanced ability to suppress C. albicans hypha formation, but they fail to eradicate the fungus. C. albicans can utilize amino acids, such as L-arginine, as an alternative carbon source under conditions where other carbon sources, such as glucose, are limited (30). Such conditions are predicted to exist inside phagosomes (31,32), and gene transcription analysis confirmed that expression of C. albicans genes coding for arginine import and degradation are increased after phagocytosis by macrophages (30). Moreover, increased arginine catabolism was shown to be important for phagosome alkalinization, hyphal induction, and macrophage escape (30,33). The restricted availability of host L-arginine as an energy source, due to high arginase-1 activity in alternatively activated macrophages, can explain the reduced hypha formation (intra-and extracellular) observed in these experiments.
In humans, arginase-1 is constitutively expressed in granulocytes and has been shown to participate in neutrophil antifungal activity, possibly through arginine depletion (34,35). On the other hand, arginase-1 release after granulocyte cell death at the site of inflammation and the exocytosis of arginase-1-containing granules due to "frustrated phagocytosis" has been shown to suppress T-cell proliferation and cytokine synthesis (34,36,37). Sustained arginase-1 activity can also deplete extracellular L-arginine levels, thereby impairing not only T-cell function (38, 39) but also NK cell activation (40). Increased arginase activity occurs as a result of tissue injury and subsequent wound healing, cancer, cystic fibrosis, or sepsis, thereby increasing susceptibility to secondary infections, including mycoses (41)(42)(43)(44). Alternatively, increased macrophage arginase-1 expression has been shown to be important in the suppression of T-cell-mediated immunopathology in a mouse model of Mycobacterium tuberculosis infection (45). These observations highlight the importance of L-arginine metabolism mediated by arginase-1 in the resolution of inflammation.
In summary, our findings demonstrate that changes in the exposure of C. albicans cell wall chitin can influence macrophage functions by impacting host L-arginine metabolism. Our data reveal an increased fungal killing potential of macrophages if either host chitinase (bisdionin F) or arginase (nor-NOHA) activity is reduced. Therapeutic inhibition of arginase-1 has been shown to have beneficial effects for other diseases, such as asthma, cancer, and parasitic infections, and shows potential as a therapeutic target to explore in fungal diseases.

MATERIALS AND METHODS
Monocyte isolation, purification, and macrophage differentiation. Blood from healthy volunteers was collected according to the local guidelines and regulations, approved by the College Ethics Review Board of the University of Aberdeen (CERB/2012/11/676). Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Paque Plus (GE Healthcare) density centrifugation according to the manufacturer's instructions. Further, highly pure unlabeled monocytes were obtained from PBMCs by depletion of nonmonocytes using a magnetically activated cell sorting (MACS) system together with a human pan-monocyte isolation kit (Miltenyi Biotec, Inc.). Purified monocytes were washed twice with phosphatebuffered saline (PBS), and cells were suspended in RPMI 1640 (Dutch modification) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2 mM L-glutamine, 1 mM sodium pyruvate, 1% minimum essential medium (MEM), nonessential amino acids (NEAA [100ϫ]; Gibco), 100 U/ml penicillin, and 100 mg/ml streptomycin. An aliquot of 50 ng/ml recombinant human granulocyte-macrophage colonystimulating factor (hGM-CSF) (Gibco) was added to differentiate monocytes into an M1-like polarized phenotype, and recombinant hM-CSF (Gibco) was added to generate an M2-like polarized phenotype. Cells were seeded into cell culture dishes at a density of 1 ϫ 10 6 cells/cm 2 and incubated at 37°C in a humidified atmosphere containing 5% CO 2 , and fresh medium containing recombinant cytokines was added every other day. After 5 to 7 days, cells were collected, seeded into 12-or 24-well plates at a density of 5 ϫ 10 5 cells or 2 ϫ 10 5 cells/well, respectively, and left overnight to adhere. M1-polarized macrophages were fully activated by adding 100 ng/ml IFN-␥ and LPS, and M2-polarized macrophages were activated by adding 20 ng/ml IL-4 for 24 h.
Isolation of mouse peritoneal macrophages. C57BL/6 (wild-type) mice were bred and housed under pathogen-free conditions in the registered animal facility at the University of Aberdeen. All animal work at the University of Aberdeen is regulated under the UK Home Office's Animals (Scientific Procedures) Act of 1986 (ASPA) (46) and European Directive 2010/63/EU (47). All work is approved by the University of Aberdeen Animal Welfare and Ethical Review Body (AWERB). A total of four mice were used in this study.
Peritoneal macrophages were isolated from 12-to 16-week-old female mice. Mice were euthanized by cervical dislocation and injected intraperitoneally with PBS containing 5 mM EDTA, and the resident peritoneal cells were harvested. Peritoneal cells were counted and seeded into 12-well plates at a density of 5 ϫ 10 5 cells/well in RPMI 1640 supplemented with 10% heat-inactivated FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin. Cells were incubated at 37°C in a humidified atmosphere containing 5% CO 2 overnight. The next day, nonadherent cells were removed by gently washing them with prewarmed PBS. Peritoneal macrophages were either left untreated or activated with 100 ng/ml IFN-␥/LPS for 4 h prior to coincubation with C. albicans.  Table 1 and were maintained as glycerol stocks at Ϫ80°C. When required, strains were plated onto YPD agar (1% [wt/vol] yeast extract, 2% [wt/vol] peptone, 2% dextrose, and 2% agar) and incubated at 30°C for 48 h. Fungal chitin from C. albicans yeast cells was extracted and purified as described previously (12). For Candida-macrophage cocultures, yeast cells were precultured twice in 10  caspofungin-treated yeast cells that presented chitin on their cell surfaces, yeast cells were inoculated to stationary phase in YPD broth at 30°C at 200 rpm and then diluted in fresh YPD broth (starting optical density at 600 nm [OD 600 ] ϭ 0.2) and grown to mid-exponential phase in the presence or absence of caspofungin (3.2 g/ml). Cells were washed three times with PBS and incubated with 50 mM thimerosal overnight in the dark before being washed repeatedly with PBS and used for stimulation assays or staining for cell wall chitin exposure. Inactivation of C. albicans cells by thimerosal was confirmed by plating cells onto YPD agar and incubating them at 30°C for 5 days. Phagocytosis assay. Following coincubation with C. albicans for 3 h at 37°C in a humidified atmosphere containing 5% CO 2 , samples were fixed with 4% (wt/vol) paraformaldehyde overnight at 4°C. Cells were washed three times with PBS and blocked with goat serum (1/20) in PBS for 30 min at room temperature. Nonphagocytosed Candida cells were stained with a fluorochrome-conjugated rabbit polyclonal anti-C. albicans antibody (Acris Antibodies) diluted 1/250 in 1% (wt/vol) BSA in PBS for 1 h at room temperature, followed by three washes with 0.05% (wt/vol) BSA in PBS containing 0.05% (vol/vol) Tween 20 (BSA/PBS-T). Macrophages were permeabilized with 0.5% (vol/vol) Triton X-100 in PBS for 10 min at room temperature, and then samples were washed three times with BSA/PBS-T and incubated with fluorochrome-conjugated phalloidin (1/1,000; Sigma-Aldrich) and 10 g/ml calcofluor white (CFW) to distinguish nonphagocytosed (FITC-and CFW-positive) from phagocytosed (CFW-positive) fungal cells. Fluorescence was observed with a Zeiss Axio Observer Z1 microscope equipped with a Zeiss MRm digital camera and Zeiss AxioVision software. Images were analyzed with ImageJ. Phagocytic activity was calculated as a phagocytic index (PI) as described before (48), using the following formula: PI ϭ (percentage of phagocytic cells containing Ն1 Candida cell) ϫ (mean number of Candida cells/ phagocytic cell containing Candida).
Candida killing assay. Activated macrophages were coincubated with C. albicans for 3 h at 37°C in a humidified atmosphere containing 5% CO 2 . To determine the killing of C. albicans by macrophages, the supernatant was removed and macrophages were lysed in 200 l 0.01% SDS in water to release fungal cells. A serial dilution was performed and plated onto YPD agar. After an incubation period of 24 h at 30°C, numbers of CFU were counted and Candida survival/killing was determined by comparing cocultures to C. albicans control cultures without macrophages.
Immunoblots. Protein expression was analyzed by Western blot analysis. After coincubation, macrophages were washed once with ice-cold PBS before proteins were extracted by lysing cells with 2ϫ Laemmli buffer (49). Samples were heated for 10 min at 95°C and centrifuged for 5 min at 13,000 rpm to pellet cellular debris. Protein samples (20 l) were separated by SDS-PAGE using the XCell SureLock minicell system (Invitrogen) with NuPAGE Novex Bis-Tris 4 to 12% precast gels (Invitrogen) in NuPAGE MOPS (morpholinepropanesulfonic acid)-SDS running buffer (Invitrogen) as per the manufacturer's instructions. Proteins were then transferred to Invitrolon polyvinyl difluoride (PVDF) membranes (Invitrogen) in NuPAGE transfer buffer using the XCell II blot module (Invitrogen). Following transfer, the membranes were blocked in 10% (wt/vol) nonfat dry milk in TBS-T (0.1% Tween 20 in TBS) for 1 h at room temperature. Membranes were washed with TBS-T and incubated with primary antibodies in 5% BSA in TBS-T overnight at 4°C, followed by incubation with appropriate horseradish peroxidase (HRP)conjugated secondary antibodies diluted in 5% BSA-TBS-T for 1 h at room temperature. Arginase-1 was detected using a polyclonal sheep anti-arginase 1 antibody (1/200; R&D Systems), followed by rabbit anti-sheep IgG-HRP (1/1,000; Abcam, Inc.); iNOS was detected using a rabbit polyclonal anti-NOS (pan)antibody (1/500; Cell Signaling) followed by goat anti-rabbit IgG-HRP (1/1,000; Cell Signaling), and ␤-actin was detected using an HRP-conjugated rabbit monoclonal anti-␤-actin antibody (1/1,000; Cell Signaling). Membranes were washed in TBS-T, and signals were detected using the LumiGLO chemiluminescent substrate (Cell Signaling). Signals were quantified using a Fusion FX7/SL chemiluminescence and fluorescence combination imaging system.
Arginase activity and NO production assays. Arginase activity was determined using the arginase activity assay kit (Sigma-Aldrich) by following the manufacturer's instructions. In short, macrophages were washed once with ice-cold PBS before being lysed for 10 min by the addition of 100 l of 0.4% Triton X-100 in 10 mM Tris-HCl, pH 7.4, supplemented fresh with 1 M pepstatin A and 1 M leupeptin. Samples were centrifuged for 10 min at 4°C and 13,000 rpm to remove cellular debris. Lysates were transferred into fresh tubes and stored at Ϫ20°C until analyzed. Nitric oxide production was determined using the Griess reagent kit for nitrite determination (Molecular Probes).
Cell wall staining. C. albicans cells were harvested by washing macrophages with PBS to collect nonphagocytosed fungal cells. Macrophages were then lysed with 0.01% SDS in water to release phagocytosed fungal cells. Fungal cells were washed three times with FACS buffer (1% [wt/vol] BSA-0.5 mM EDTA in PBS) and blocked with 5% goat serum for 30 min at room temperature before being stained with 5 g/ml purified human Fc-dectin-1 for 1 h on ice. Samples were repeatedly washed before being incubated with fluorochrome-conjugated anti-human IgG Fc antibody (1/250; Life Technologies, Inc.) and fluorochrome-conjugated wheat germ agglutinin (WGA-FITC) (100 g/ml; Sigma-Aldrich) for 45 min on ice. Cells were washed and stained with CFW (25 g/ml; Sigma-Aldrich) for 10 min before being intensively washed and finally fixed with 4% (wt/vol) paraformaldehyde. Fluorescence was observed with a Zeiss Imager M2 equipped with a Zeiss MRm digital camera and Zeiss AxioVision software. Images were analyzed with ImageJ.
Statistical analyses. All experiments included two biological replicates per donor and were performed at least 3 times, with the exception of the experiments whose results are shown in Fig. 5 and S1 (performed twice), and revealed comparable results. Values are presented as means Ϯ standard deviations (SD) (n ϭ the number of blood donors/mouse). Statistical significance was determined using GraphPad Prism 5 (as applicable) using a two-tailed paired Student t test and one-way analysis of variance (ANOVA) followed by Tukey post hoc analysis. A P value of 0.05 or less was considered significant.