Bst1 is required for Candida albicans infecting host via facilitating cell wall anchorage of Glycosylphosphatidyl inositol anchored proteins

Glycosylphosphatidyl inositol anchored proteins (GPI-APs) on fungal cell wall are essential for invasive infections. While the function of inositol deacylation of GPI-APs in mammalian cells has been previously characterized the impact of inositol deacylation in fungi and implications to host infection remains largely unexplored. Herein we describe our identification of BST1, an inositol deacylase of GPI-Aps in Candida albicans, was critical for GPI-APs cell wall attachment and host infection. BST1-deficient C. albicans (bst1Δ/Δ) was associated with severely impaired cell wall anchorage of GPI-APs and subsequen unmasked β-(1,3)-glucan. Consistent with the aberrant cell wall structures, bst1Δ/Δ strain did not display an invasive ability and could be recognized more efficiently by host immune systems. Moreover, BST1 null mutants or those expressing Bst1 variants did not display inositol deacylation activity and exhibited severely attenuated virulence and reduced organic colonization in a murine systemic candidiasis model. Thus, Bst1 can facilitate cell wall anchorage of GPI-APs in C. albicans by inositol deacylation, and is critical for host invasion and immune escape.


Orf19.1053 (BST1) is responsible for inositol deacylase of GPI-APs in C. albicans. The sensitivity
of GPI-APs to phosphoinositide-phospholipase C (PI-PLC) was used to evaluate the level of inositol deacylation in our study. After attaching to GPI anchors, GPI-APs are resistant to PI-PLC. Following inositol deacylation in the ER, GPI-APs become sensitive to PI-PLC and are transferred from the detergent phase to aqueous phase when separated by Triton X-114 18 .
C.albicans expresses on its surface Als (Agglutinin like sequence) proteins, which play an important role in the development of candidiasis. Als1p, a well characterized GPI-anchored protein, is known to mediate the adhesion of C. albicans to host cells 20 . We extracted the hemagglutinin (HA)-tag fused Als1p from bst1Δ/Δ (bst1Δ/Δ als1Δ/ALS1-HA) and parent strain (als1Δ/ALS1-HA), and found that the majority of the purified Als1p from both strains were partitioned into the detergent phase. However, a subset of Als1p from the parental strain was partitioned into the aqueous phase following treatment with PI-PLC. By contrast, the majority of Als1p from bst1Δ/Δ strain remained partitioned into the detergent phase following PI-PLC treatment, suggesting that GPI-APs from bst1Δ/Δ strain were resistant to PI-PLC (Fig. 1B). This result indicated that inositol deacylation of GPI-APs was defective in bst1Δ/Δ mutant, and suggested that BST1 played an important role in this process.
To further confirm the role of BST1 in inositol deacylation of GPI-APs, we extracted cytoplasm proteins from C. albicans and demonstrated that GPI-APs with peroxidase labeled concanavalin A (ConA) can bind to mannose residues of GPI-anchor. ConA-stained GPI-APs from bst1Δ/Δ strain were found to be resistant to PI-PLC, as compared to the parental strain.
According to the sequence information from Saccharomyces genome database, the catalytic site of BST1 in S. cerevisiae is serine-236. Meanwhile, the catalytic site of PGAP1 (homology gene of BST1) in human has been reported to be serine-174. The corresponding serine of BST1 in C. albicans was found located in its 202 site by BLAST (basic local alignment search tool). Therefore, we mutated the serine-202 into alanine to ensure that the observed phenotypes are in fact due to the genetic defects at the BST1 locus rather than artifacts from mutations accumulated during strain construction. We constructed a C. albicans strain [BST1(S202A)] in which the putative catalytic serine 202 of BST1 was substituted for alanine, and found that its ConA-stained GPI-APs exhibited similar resistance to PI-PLC with bst1 null mutant (Fig. 1C).
The above results indicate that BST1 is responsible for inositol deacylation of GPI-APs in C. albicans.

Impaired cell wall anchorage of GPI-APs in BST1-deficient C. albicans strains. GPI-APs link to
the cell wall β -(1,6)-glucan through GPI remnant in C. albicans via phosphodiester bonds which can be cleaved using hydrogen fluoride (HF)-pyridine 5,21 . The highly mannosylated cell wall GPI-APs were extracted from C. albicans cells using HF-pyridine, and stained by peroxidase-linked ConA.
Immunoblotting of cell wall GPI-APs extraction from parent strain with peroxidase labeled ConA revealed marked protein bands (Fig. 2B). while a remarkable reduction of ConA-stained material was observed in the bst1Δ/Δ strain, suggesting fewer GPI-APs linkage to the cell wall. In addition, we also found mannoproteins significantly accumulated in the cytoplasm of bst1Δ/Δ strain which was supported by an increase of ConA-stained materials ( Fig. 2A).
To further determine whether inositol deacylation was required for the cell anchorage of GPI-APs, we analyzed the cell wall HF-pyridine extraction by LC-MS/MS. Twenty-one GPI-APs were identified to be significantly reduced in the cell wall of bst1Δ/Δ strain as compared to the parent and BST1-complemented strains (see Supplementary Table S3). These proteins were thought to play specific roles, including participating in cell wall biogenesis and assembly (Pga4p, Cht2p, Utr2p, Dfg5p, Ecm33p, Dcw1p, Ccw14p, Crh11p and Ihd1p), adhesins and morphogenesis switch (Als1p, Pga24p and Pga29p), enzymes participating in host permitting their invasion and nutrient uptake (Plb5p, Carp9p, Csap and Rbt5p), degradation of damaging superoxide radicals (Sod4p and Sod5p), as well as other unknown functions (Pga30p, Pga45p, Pga52p) 6 . Seven of these proteins (Rbt5p, Utr2p, Pga29p, Als1p, Sod5p, Ecm33p and Plb5p) are known to impact the virulence of C. albicans (Fig. 2E) 7,20,22-26 . Inositol deacylation of GPI-APs is required for host invasion ability of C. albicans. GPI-APs of C. albicans participate in several stages of host invasive including morphogenesis (the reversible transition between unicellular yeast cells and hyphae), adhesion and invasion of host cells. Our results indicate a remarkable decreased in the attachment of GPI-APs to the cell wall of bst1Δ/Δ. Therefore, we assessed the invasion ability of the bst1Δ/Δ mutant strain in vitro.
Bst1 mutants (bst1Δ/Δ and S202A mutant of BST1) displayed defective filamentation in either liquid (RPMI1640, 10% serum liquid medium) or solid media (Spider and Lee's agar media) favoring hyphal growth ( Fig. 3A and Supplementary Fig. S4). Our results indicate that the cell wall anchorage of phospholipase Plb5, which functions as an enzyme to damage the host cell membranes 26 , was decreased in bst1Δ/Δ mutant strain. So we assayed the total extracellular phospholipase activity by growing various strains on egg yolk agar and noted that bst1Δ/Δ exhibited significantly reduced secretion of phospholipases compared to the parent and BST1-complemented strains (Fig. 3B).
Adherence of C. albicans to epithelial cells, such as those in intestinal and oral mucosa, is essential for commensal growth and also an initial step in the pathophysiology of candidiasis. In the adherence assay, we found a significantly lower adherence, following a 60 minutes interaction period, of bst1Δ/Δ to KB (derived from a buccal carcinoma) and Caco-2 (derived from a colorectal carcinoma) monolayers as compared to the parent and BST1-complemented strains (Fig. 3C).
Following adhesion, the entry of C. albicans into host tissues is the next most crucial step in the pathophysiology of superficial and invasive candidiasis. We used a scanning electron microscopy (SEM) to visualize the  Fig. 1A. After phase separation, both aqueous (A) and detergent (D) phases of proteins underwent immunoblotting using 2.5 μ g/ml peroxidase labeled-Con.
invasion of C. albicans into KB and Caco-2 cells after 2 hours of co-culture. We observed the penetration of hyphal forms of the parent and BST1-complemented strains into KB and Caco-2 cells at the apical face, and microvillis were attached to their hyphae at the point of penetration (Fig. 3D). However, the bst1Δ/Δ mutant strain was not only defective in hyphal formation, but also hardly able to penetrate KB and Caco-2 cells (Fig. 3D).
These data suggest that diminished cell wall GPI-APs by BST1 gene deletion in C. albicans significantly impairs its capacity to invade host cells.
BST1 is required for cell wall polysaccharides in C. albicans. The cell wall of C. albicans is composed of an inner skeletal layer of chitin and β -glucans (β -1,3-and β -1,6-glucan) and an outer layer of highly glycosylated proteins containing GPI-APs. Via transmission electron microscopy (TEM), we found that the inner cell  The rates of adherence of parent (SN152), BST1-complemented (bst1Δ/Δ::BST1), bst1Δ/Δ null mutant (bst1Δ/Δ) strains to human KB oral epithelial cells or intestinal Caco-2 epithelial cells were evaluated by coincubating them for 60 minutes in six-well tissue culture plates, after which the adherent clones were counted. Data represent mean (± SD) of triplicates from one representative experiment of three. **P < 0.01 (Error bars indicate SD. One-way ANOVA with Bonferroni post-test). (D) Representative micrographs of scanning electron microscope (SEM) of KB and Caco-2 cells invaded or penetrated by parent strain (SN152), bst1Δ/Δ::BST1 and bst1Δ/Δ after 2 hours co-incubation.
wall layer of the parental strain is surrounded by an external layer of mannosylated proteins, while the bst1Δ/Δ strain was nearly devoid of an outer cell wall layer (Fig. 4A).
To further examine the effect of BST1 deficiency on cell wall polysaccharides, we stained polysaccharides (mannan, β -glucan or chitin) with ConA alexa fluor 488 conjugate, anti-β -glucan primary antibody and calcofluor white (CFW). ConA alexa fluor 488 conjugate staining indicated that bst1 mutants (bst1Δ/Δ and S202A mutant of BST1), either in yeast or hyphae form, had a markedly decreased mannan content within their cell walls (Fig. 4B,E Consistent with the aberrant cell wall structures, surface properties of the bst1Δ/Δ strain, such as hydrophobicity and flocculation, were also markedly increased (see Supplementary Fig. S6A,B ). The change of cell wall structures and properties also impaired the resistance of bst1Δ/Δ to different stresses including oxidative stress, congo red, CFW, and antifungal agents (fluconazole and miconazole) (see Supplementary Fig. S6C).

C. albicans with defective inositol deacylation of GPI-APs can induce stronger inflammatory response.
Macrophages are essential to initiate antifungal innate immune responses and also prime an adaptive immune response. We tested whether the bst1Δ/Δ mutant strain, could stimulate enhanced macrophage response. We found that bst1Δ/Δ mutant stimulation induced more nuclear translocation of nuclear factor-κ B (NF-κ B) (p65), spleen tyrosine kinase (Syk) phosphorylation, and inhibitor of kappa B α (Iκ Bα ) phosphorylation and degradation in thioglycollate-elicited peritoneal macrophages, suggesting increased NF-κ B signaling. Consistent with this hypothesis, we observed enhanced extracellular-signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK) and p38 phosphorylations induced in macrophages by bst1Δ/Δ mutant strain stimulation as compared to the parental strain. These results confirm the activation of mitogen-activated protein kinase (MAPK) signaling (Fig. 5A). NF-κ B and MAPK activation can modulate the expression of various inflammatory cytokines. We found that bst1Δ/Δ mutant strain induced higher levels of inflammatory cytokines Respiratory burst of neutrophils stimulated by the indicated C. albicans were assayed by detecting the cellular reactive oxygen species production in neutrophils after incubation with C. albicans for 1 hour (right panel). Data represent mean (± SD) of triplicates from one representative experiment of three. *P < 0.05 and **P < 0.01 (One-way ANOVA with Bonferroni post-test).
Scientific RepoRts | 6:34854 | DOI: 10.1038/srep34854 including tumor necrosis factor α (TNF-α ), interleukin 6 (IL-6), interleukin-12 p40 (IL-12p40), and interleukin 10 (IL-10) in macrophages, while the production of these cytokines were significantly lower in the parent and BST1-complemented strain, respectively (Fig. 5B). In addition, BST1 (S202A) mutant strain also induced higher levels of inflammatory cytokines in macrophages (see Supplementary Fig. S7). We also noted that a higher percentage of bst1Δ/Δ strains were killed by neutrophils than those levels observed in the parent or BST1-complemented strains in co-culture model system. These results are consistent with an augmented respiratory burst response to bst1Δ/Δ strain (Fig. 5C).

Inositol deacylation of GPI-APs is required for invasive infection in a mouse model of hematogenously disseminated candidiasis.
The dimorphism of C. albicans and its capabilities of adhesion and invasion of host epithelial cells are required for hematogenously disseminated candidiasis. The host immune response also significantly affects the outcome of invasive infection. Therefore, we investigated the contribution of BST1 to the development of hematogenously disseminated candidiasis in a mouse model. Over a 30-day observation period, none of the mice infected intravenously with the bst1Δ/Δ mutant strains died (Fig. 7A). By contrast, the median survival of mice infected with parent and bst1Δ/Δ::BST1-complemented strains was 8 and 17 days, respectively (Fig. 7A). The kidney and liver fungal burdens of mice inoculated with bst1 mutants (bst1Δ/Δ and S202A mutant of BST1) were significantly lower than those infected with either the parent or BST1-complemented strains (Fig. 7B,D).
The lower fungal burden in the kidneys of mice infected with the bst1Δ/Δ mutant strain was confirmed by histopathological review. Following two days of infection, no C. albicans filaments, except for a few inflammatory cells, were observed in the kidneys, of mice infected with the bst1Δ/Δ mutant strain (Fig. 7C). By contrast, in the parent or BST1-complemented strains, microabscesses were observed in the kidneys, which consisted of numerous neutrophils surrounding multiple C. albicans filaments. These results suggest that inositol deacylation of GPI-APs by BST1 is required maximal virulence of C. albicans in the mouse model of hematogenously disseminated candidiasis.

Discussion
The biosynthesis of GPI-APs can be separated into two phases. In the first phase, the GPI-anchor is synthesized and attached to the proteins. The inositol moiety of GPI is acylated at an early step in GPI biosynthesis, which is essential for the generation of mature GPI capable of proteins attachment in fungi. In the second phase, both the GPI and protein portions of the precursor GPI-APs are modified during transport from the ER to the cell surface 8 . The acyl group is usually absent from GPI-APs on the cell surface due to inositol deacylation that occurs in the ER shortly after GPI-anchor attachment in the second phase of GPI-APs biosynthesis. Mammalian inositol deacylase (Pgap1p) has been cloned and demonstrated to be important for efficient transport of GPI-APs from the ER to the Golgi, but not for the normal cell surface expression 18,19 . In the present study, we found that Bst1p could work as inositol deacylase in C. albicans, which was involved in the start of the second phase of GPI-APs biosynthesis. We also demonstrated that inositol deacylation of GPI-APs played a critical role in the pathogenicity of C. albicans.
Moreover, we demonstrated that GPI-APs from bst1 mutant strain were more resistant to PI-PLC compared to those of parent strain, highlighting the function of Bst1p as inositol deacylase in C. albicans (Fig. 1B,C). Nevertheless, the remove of palmitate residue by inositol deacylase in many mammalian GPI-APs is not obligatory 28 . By example, the mature GPI-APs on human erythrocyte membrane remain an acylated status. Blocking inositol deacylase of GPI-APs via PGAP1 deficiency in mammalian cells failed to affect the cell surface expression of GPI-APs, and only led to the expression of structurally abnormal GPI-APs with the acylated inositol 19 . However, our results suggested that inositol deacylation of GPI-APs was critical for their efficient anchorage to the cell wall of C. albicans (Fig. 2B-E). Consistent with the diminished cell wall GPI-APs, bst1 mutant strains exhibited an accumulation of GPI-APs within the cytoplasm ( Fig. 2A,C), indicating the key role of inositol deacylation in targeting GPI-AP to the cell wall. The different roles of inositol deacylation of GPI-APs in C. albicans and mammalian cells may be associated with the evolution of Bst1p. The evolutionary conservation of this protein is relatively low between C. albicans and mammalian cells, with a sequence similarity of only 28% (see Supplementary Fig. S8).
GPI-APs covalently link to the skeletal polysaccharides of fungal cell wall by GPI remnant. Many cell wall GPI-APs, such as Als1p, Als3p, Ecm33p and Hwp1p, act as virulence factors in the development of an invasive infection. In this work, we demonstrated that inositol deacylation is required for GPI-AP anchorage to the cell wall. Proteomic analysis indicated that GPI-APs including several known virulence factors diminished in the cell wall of bst1Δ/Δ strain. These virulence factors play roles in cell wall biogenesis and assembly (Utr2p, Ecm33p), adhesion, morphological switch (Als1p), invasion and damage to the host cells (Plb5p, Rbt5p), as well as degradation superoxide radicals to protect themselves (Sod5p) (see Supplementary Table S3 invasive infection including filaments transition, phospholipases secretion, adherence and invasion of host epithelial cells, which emphasized the importance of inositol deacylase (Fig. 3).
C. albicans cell wall is a complex and dynamic architecture based on a core structure of β -(1,3)-glucan covalently linked to β -(1,6)-glucan and chitin, as well as an outer layer or matrix composed mainly of mannose-glycosylated proteins 29 . These polysaccharides serve as PAMPs that can be recognized by PRRs which may subsequently trigger innate immune cell response and render antigen-presenting cells competent to prime T cells, thereby initiate the adaptive immunity. The recognition of β -(1,3)-glucan by Dectin-1 plays a crucial role in host antifungal defense. Previous studies have indicated that the Y238X polymorphism in Dectin-1 was associated with human mucosal C. albicans infection 30 . However, β -(1,3)-glucan is usually buried beneath the outer layer of mannoproteins except for the region between the parent cell and the mature bud, in which Dectin-1 is bound in a more restricted pattern. It has also been reported that exposure of cell wall β -(1,3)-glucan in C. albicans could increase its recognition by phagocytic cells 31 . Moreover, deficient inositol deacylase of GPI-APs in C. albicans could also induce β -(1,3)-glucan exposure (Fig. 4C,F). This exposure may be a result of the inadequate concealment of the inner layer by cell surface mannose-glycosylated proteins of the bst1Δ/Δ mutant strain.
Our study demonstrated that bst1Δ/Δ mutant strain could be recognized by the host innate immune cells and thus induced stronger immune responses. Neutrophils contribute to the initial response against fungi, and play a critically important role in individuals who are neutropenic or immunosuppressed [32][33][34] . We demonstrated that neutrophils could kill bst1Δ/Δ mutant strains more efficiently, and produced more ROS when challenged (Fig. 5C). After encountering microbial pathogens, host macrophages secrete cytokines and chemokines and engulf the pathogens in phagosomes 35 . Not only did bst1Δ/Δ mutant strains elicited an enhanced recognition by macrophages, but an increased cellular responses of macrophages was observed. This response included Syk phosphorylation, activation of NF-κ B and MAPK signaling, and triggering cytokine production (TNF-α , IL-6, IL-12p40, and IL-10) (Fig. 5A,B). IL-6 and IL-23 (consists of IL-12 p40 and p19) are key cytokines leading to the induction of Th17 cell differentiation 36 . TNF-α and IL-6 participate in the innate immune response against Candida infection by promoting neutrophils production and activation. Increased levels of these inflammatory cytokines may result in a recruitment of innate immune cells such as neutrophils to the infected organs to clear fungi. Using a peritoneal infection model, we demonstrated that the marked elevation of chemokines in peritoneal cavity stimulated by bst1Δ/Δ mutant strain resulted in an obvious recruitment of neutrophils and monocytes (Fig. 6). Besides β -(1,3)-glucan recognition by Dectin-1, mannan recognition by Dectin-2/Dectin-3 is important in resistance to fungal infection 27,37,38 . The present study indicated that deletion of BST1 in C. albicans resulted in decreased mannose content and a more remarkable exposure of β -(1,3)-glucan on the cell surface. Based on our present result, it could not determine that the change of mannan or exposure of β -(1,3)-glucan contribute to the altered host immune responses. The role of mannan recognition and glucan recognition in the altered immune responses induced by bst1 mutant is worth to be further investigated.
The development of invasive fungal infection is determined both by the invasive ability of the infective strains and the host antifungal defense. Our results suggested that defective inositol deacylation of GPI-APs not only inhibited the invasive ability of C. albicans, but also enhanced its recognition by the host immune system. As expected, our study further demonstrated that bst1Δ/Δ mutant strain displays a markedly diminished virulence in hematogenously disseminated candidiasis (Fig. 7). Both impaired invasive ability and enhanced immunogenicity of bst1Δ/Δ mutant strain contribute to its diminished virulence. We demonstrated the effects on virulence are not a result of a reduce growth rate in the bst1 null mutant (see Supplementary Fig. S9).
In conclusion, our study describes the novel efffets of Bst1 acting as inositol deacylase of GPI-APs in C. albicans, and facilitating cell wall anchorage of GPI-APs. Our study suggests that abolishing inositol deacylation of GPI-APs in C. albicans not only impaired its invasion ability, but also exposed PAMPs recognized by host immune system. Furthermore, inositol deacylation of GPI-APs defection rendered C. albicans more sensitive to azole antifungal agents. Therefore, our findings may provide a potential target for the development of new drugs with a new modality of treatment as compared to current antifungal agents. Given the high mortality assoiated with systemic candidiasis infection, future studies should urgently evaluate the crystal structure of Bst1p in an effort to identify potential drug candidates by high-throughput screening. The entire coding sequence of BST1 was deleted from the parent strain SN152 by two-step homologous recombination using a fusion-PCR-based strategy as described previously 39 (see Supplementary Fig. S1). The BST1-complemented (bst1Δ/Δ:: BST1) and BST1 S202A mutant strain were generated using the SAT1 flipping method 40,41 (see Supplementary Figs S1 and S2). We got 8 independently created homozygous bst1 mutants of C. albicans totally. These homozygous mutants (Mutant 1#, 2#, 3#, 4#, 5#, 6#, 7#, 8#) showed similar phenotype (Data not shown), and we showed the phenotype of mutant 1# in our manuscript.

Methods
Strains expressing HA-tag fused Als1p (bst1Δ/Δ als1Δ/ALS1-HA and als1Δ/ALS1-HA) were generated in strain bst1Δ/Δ and SN152. One allele of ALS1 was deleted according to the fusion-PCR-based strategy, and a 102-bp fragment containing three repeated sequence of HA epitope-tag was inserted into the ALS1 gene allele using the SAT1 flipping method (see Supplementary Fig. S3). And we validated the tagging by sequencing of the mutant.
All of the strains and primers used for strain construction are listed in Supplementary Tables S1 and S2. PI-PLC sensitivity assay. The sensitivity of GPI-APs to PI-PLC was used to evaluate the level of inositol deacylation. PI-PLC sensitivity assay was performed as described previously with some modifications 42 . HA-tag fused Als1p was purified by anti-HA-tag mouse mAb (agarose conjugated) from C. albicans expressing HA-tag fused Als1p. The purified Als1p was incubated with or without 0.5 unit of PI-PLC for 1h at 30 °C and then separated into aqueous and detergent phase by 1.4% Triton X-114. The aqueous and detergent phase were subjected to SDS-PAGE, and then blotted using anti-HA monoclonal antibody. Cytoplasm proteins were partitioned into aqueous and detergent phase using Triton X-114. The detergent phase were collected and incubated with 1 unit of PI-PLC for 1h before phase separation. Both phases were analyzed by western blotting analysis using 2.5 μ g/ml peroxidase labeled-ConA.
Cell wall integrity assays. C. albicans strains were grown in YPD broth in a rotary shaker at 30 °C, and exponentially growing C. albicans yeast were harvest. For hyphae growth in our study, 1 × 10 6 C. albicans were cultured in RPMI 1640 medium [RPMI 1640 (Gibco BRL) 10 g, NaHCO 3 2.0 g and morpholinepropanesulfonic acid (Sigma) 34.5 g, pH 7.0, in 1 L ddH 2 O, sterilization by filtration] at 37 °C for 3 hours. And then the hyphae cells were harvest. C. albicans yeast or hyphae cells were then stained with specific anti-β -glucan primary antibody overnight at 4 °C on a rotator and Cy3-labeled goat-anti mouse secondary antibody for 1 h at 30 °C to visualize β -glucan exposure 45 . To stain the cell wall for chitin and mannan, cells were incubated with 50 μ g/ml CFW or 50 μ g/ml ConA alexa fluor 488 conjugate for 45 minutes away from light 46 . The stained yeast or hyphae cells described above were scanned with a confocal laser scanning microscope (TCS SP5; Leica). The fluorescence intensity of confocal laser scanning microscope photograph for Fig. 4B-G were acquired and analyzed by LAS AF Lite program (Version 2.1.1 build 4443, Leica Microsystems) (n = 100 for yeast cells and n = 20 for hyphae).
The ultrastructure of cell wall was observed by TEM. Exponentially growing C. albicans were washed in PBS and fixed in 3ml fixative solution (sodium cacodylate buffer, pH 7.2, containing 4% polyoxymethylene) for 24 h at 4 °C. The samples were then washed with saline and postfixed with 1% phosphotungstic acid for 90 minutes. The fixed cells were dehydrated increasing concentrations of ethanol. Ultrathin sections were prepared, double stained with uranium and lead, and observed under a transmission electron microscope (Hitachi H-800, Japan). Murine systemic candidiasis model. C57BL/6 female mice (6-8weeks) were anaesthetized intraperitoneally before infection. A total of 5 × 10 5 live C. albicans strains were administrated intravenously in a final volume of 200 μ l of saline. Survival rates were monitored for 30 days. Fungal burden was assessed by plating a series of diluted solutions of homogenized kidneys and livers from mice at 2 days or 5 days after infection. For histopathological analysis, kidneys were fixed in 10% neutral buffered formalin and stained with hematoxylin and eosin (HE) and periodic acid-schiff (PAS) to reveal inflammatory infiltration and fungal hyphae.
Phospholipase activity assays of C. albicans. The phospholipase activity was assayed according to the method previously described 47 . A total of 10 5 exponentially growing C. albicans in PBS was plated on egg yolk agar (consisting of Sabouraud's dextrose agar, NaCl, CaCl 2 and egg yolk emulsion) and incubated at room temperature for 2 days. The extracellular phospholipase of each strain was determined by measuring the width of zone of precipitation around the colony.
Epithelial cell adherence and invasion assay. Adherence assays were performed, according to a previously described protocol, using two cell lines 48 . Caco-2 cells and KB cells were grown to 95% confluency in a 6-well tissue culture plate. Each well was infected with 10 2 organisms. After 60 minutes of infection, the wells were then overlaid with 2 ml of YPD agar (about 45 °C) and incubated at 30 °C for 48 hours and the percentage of adherence was analyzed.
Caco-2 and KB cells were grown to 95% confluency on 8 mm diameter glass coverslips. Each coverslip was infected with 1 × 10 6 exponentially growing C. albicans. After 2 hours of infection, the epithelial cells were examined using a XL-30 SEM (Philips, Holland) 49 .
Macrophage-yeast interaction. C. albicans yeast cells were harvested, and exposed to four doses of 100,000 μ joules/cm 2 in a CL-1000 Ultraviolet Crosslinker (UVP), with agitation between each dose to treat cells evenly 50 . The thioglycollate-elicited macrophages were stimulated with the UV-inactivated C. albicans [multiplicity of infection (MOI) = 5] for the indicated time.
For nuclear extracts, 1 × 10 7 murine macrophages were incubated in 6 cm plates and stimulated with UV-inactivated C. albicans for 30 minutes and 1 hour, respectively; whereas for total cell lysates, 3 × 10 6 cells were incubated in 12-well plates and stimulated with UV-inactivated C. albicans for 15, 30, and 45 minutes, respectively. For cytokine production assay, 1 × 10 7 cells were incubated in 6-well plates and stimulated with the UV-inactivated C. albicans for 6 hours, the cell supernatants were collected for cytokine production assay.
Scientific RepoRts | 6:34854 | DOI: 10.1038/srep34854 Murine peritoneal infection model. C57BL/6 female mice were injected intraperitoneally with 5 × 10 5 UV-inactivated C. albicans and killed by CO 2 asphyxiation after 4h. The inflammatory infiltrate was collected by lavage with ice-cold PBS containing 5 mM EDTA. Inflammatory cell populations were counted and then were analyzed by flow cytometry (BD FACS Verse TM ) to determine the leukocyte composition as previously described 27 . Cytokines in peritoneal lavage fluid were determined as described above.
Measurement of neutrophil killing of C. albicans. Thioglycollate-elicited peritoneal neutrophils (0.6 × 10 6 ) were infected with live C. albicans (0.3 × 10 4 to 1 × 10 4 ) in the wells of a 24-well plate and were kept for 60 minutes at 4 °C to allow the cells to 'settle' , before being transferred to an incubator at 37 °C for an additional 60 minutes. Cells were then mixed with scraping, and plated on YPD agar for counting viable C. albicans after incubation for 48h at 30 °C 27 .
Respiration burst assay. Freshly isolated thioglycollate-elicited peritoneal neutrophils pooled from at least two mice, were loaded with 10 μ M dihydrorhodamine 123, and then stimulated with C. albicans (MOI = 1) for 60 minutes at 37 °C. Reactive oxygen species production was determined by measuring the conversion of dihydrorhodamine 123 to rhodamine using a SpectraMax M5 multi-mode microplate reader with excitation and emission filters of 485 and 538 nm 51 .
Statistical analysis. At least three biological replicates were performed for all experiments unless otherwise indicated. One-way analysis of variance (ANOVA) with Bonferroni or Dunns post-tests was used when multiple groups were analyzed. For analysis of nonparametrically distributed data, Kruskal-Wallis test was used. Statistical significance was set at P value of 0.05 or 0.01.