The 14-3-3 Gene Function of Cryptococcus neoformans Is Required for its Growth and Virulence

Abstract

Cryptococcus neoformans is a life-threatening pathogenic yeast that causes devastating meningoencephalitis. The mechanism of cryptococcal brain invasion is largely unknown, and recent studies suggest that its extracellular microvesicles may be involved in the invasion process. The 14-3-3 protein is abundant in the extracellular microvesicles of C. neoformans, and the 14-3-3-GFP fusion has been used as the microvesicle's marker. However, the physiological role of 14-3-3 has not been explored. In this report, we have found that C. neoformans contains a single 14-3-3 gene that apparently is an essential gene. To explore the functions of 14-3-3, we substituted the promoter region of the 14-3-3 with the copper-controllable promoter CTR4. The CTR4 regulatory strain showed an enlarged cell size, drastic changes in morphology, and a decrease in the thickness of the capsule under copper-enriched conditions. Furthermore, the mutant cells produced a lower amount of total proteins in their extracellular microvesicles and reduced adhesion to human brain microvascular endothelial cells in vitro. Proteomic analyses of the protein components under 14-3-3-overexpressed and -suppressed conditions revealed that the 14-3-3 function(s) might be associated with the microvesicle biogenesis. Our results support that 14-3-3 has diverse pertinent roles in both physiology and pathogenesis in C. neoformans. Its gene functions are closely relevant to the pathogenesis of this fungus.

Introduction

Cryptococcus neoformans is the most common yeast infection of the central nervous system and the most frequent neurological complication in AIDS patients. C. neoformans primarily affects immunocompromised patients, resulting in the devastating meningoencephalitis. Annually, close to one million cases of cryptococcal meningoencephalitis occur worldwide, resulting in about 600,000 deaths per year [30]. Recent surveys have also revealed that C. neoformans also infacts large numbers of HIV-negative individuals in regions such as China, Korea, and Vietnam [6,7]. These outbreak cases may be due to novel hypervirulent strains or because of other reasons that have not yet been identified. C. neoformans, the etiologic agent of cryptococcosis, has become one of the most severe fungal pathogens in the world today.

C. neoformans has a predilection for infecting the brain. However, the mechanism by which C. neoformans invades the brain is largely unknown. In order to cause meningitis or meningoencephalitis, C. neoformans must transverse the boundary between the brain parenchyma and blood circulation-the blood-brain barrier (BBB), which consists of a single lining of a specialized capillary endothelium called the brain microvascular endothelial cells [12,35]. As such, human brain microvascular endothelial cells (HBMEC) have been used as the in vitro model for its brain infection. These studies have shown that C. neoformans is able to adhere to, penetrate, and hijack host systems during its invasion process [3,4,13,17,18,20,22,23,37]. Likewise, C. neoformans induces the alteration of gene profiling [19] and protein expression [41] in HBMEC. Together, C. neoformans is able to elicit complicated host responses to facilitate its traversal of the BBB.

It is well accepted that the C. neoformans capsule polysaccharide is a major virulence factor for this pathogen, and the secretion of microvesicles plays a role in the building block transport for the biosynthesis of its extracellular capsule, and microvesicles contain a few proteins associated with the cryptococcal virulence trait [32]. Indeed, recent studies have shown that C. neoformans-derived microvesicles (CnMVs) are closely associated with its brain invasion capability, relevant to the pathogenesis of cryptococcal meningoencephalitis [14]. Purified CnMVs can enhance C. neoformans traversal of the brain endothelial cell monolayer in vitro. This activity is presumably due to the CnMVs’ ability to activate HBMEC membrane rafts and induce cell fusogenic activity. The 14-3-3 protein is an abundant protein in CnMVs; therefore, GFP to 14-3-3-fusion has been used as the CnMV marker for both in vitro and in vivo studies [14]. In infected mice brains, CnMVs have been observed to distribute themselves both inside and around C. neoformans-induced cystic lesions. Moreover, host glial cells (reactive astrocytes) migrate to these cystic lesions. Substantial damage could be observed in areas that have a high density of 14-3-3-GFP markers. Thus, it has been determined that CnMVs are pertinent in both the physiology and pathogenesis (brain invasion) of this pathogen.

The 14-3-3 proteins are highly conserved but ubiquitously expressed proteins in all eukaryotes. They assemble as stable dimers that dock onto phosphorylated serine and threonine residues in hundreds of intracellular target proteins [16]. Biological regulation by 14-3-3 is mainly mediated through phosphorylation-dependent proteinprotein interactions [27]. Several molecular mechanisms of 14-3-3 have been proposed. The 14-3-3 binding sequesters specific sequences of the target protein, and modulates target protein stability, phosphorylation state, localization, and molecular interactions [8,24,28]. The 14-3-3 proteins may also induce target protein structure changes and thus modify target protein functions [8]. Distinct temporal and spatial expression patterns of 14-3-3 isoforms have been observed in development and in acute response to extracellular signals. These observations suggest that 14-3-3 isoforms may perform different functions despite their sequence similarities [8]. Over 200 targets have been identified for 14-3-3 proteins; they fall into four main categories: signaling components, primary metabolic enzymes, vesicular transport/trafficking, and chromatin functions. Based on the multitude of their binding partners and their involvement in numerous cellular functions, 14-3-3 dimers are thought to act as adaptor proteins that participate in spatial organization of signaling complexes and subcellular localization of targeted proteins [2]. In humans, there are at least seven 14-3-3 isoforms, which are differentially regulated in specific cell types, cancers, and neurological syndromes [10]. In Saccharomyces cerevisiae, two 14-3-3 genes, BMH1 and BMH2 [39], are required for multiple functions, including several downstream events of the RAS-mediated pathway [40], exocytosis, and vesicular transport steps [11]. In Candida albicans, its 14-3-3 protein is required for vegetative growth and optimal filamentation, suggesting its role in the regulatory pathways required for colonization and invasion [21]. As the 14-3-3 protein has multiple roles in different organisms and because of its abundance in the CnMVs, we are interested to know whether the C. neoformans 14-3-3 protein plays any pathological role during infection. In this report, we replaced the 5’-flanking promoter region of the 14-3-3 gene with the copper-controllable promoter CTR4 [29]. The CTR4 regulatory strain showed a reduction in growth rate, some morphological changes, an impediment in cell division, and a decrease in total CnMV proteins. Furthermore, TEM images showed a delay of cell separation and the reduction in the thickness of the capsule. The adhesion of the mutant cells to HBMEC in vitro was reduced. Alterations of protein components were observed upon induction and suppression of 14-3-3 gene expression. These results support our hypothesis that 14-3-3 has various pertinent roles in the physiology and pathology of C. neoformans.

 

Materials and Methods

Reagents and Materials

C. neoformans parental strain B-3501A was used in this study. To construct the copper-regulatory 14-3-3 strain, we performed PCR fusion using a primer set (Ctr1A1: 5’-CGAGGATCCCGGGGTGATACACCATTTTC-3’; Ctr1B1: 5’-GACAGAGTCTTCTCGGTTAGACATGATTGGTGAAGTCGTTGTCGTAG-3’) to amplify the CTR4 promoter and another (Bmh1C1: 5’-ATGTCTAACCGAGAAGACTCTG-3’; Bmh1D1: 5’-ATGATGAGGGTAGAGTCCCTGA-3’) to amplify the coding region of 14-3-3 (Fig. 1A). The resulting fusion PCR clone, pYCC1016, was sequenced. The replacement cassette was constructed by ligating the 5’ flanking region of 14-3-3, NEO resistance marker, and pYCC1016 to yield pYCC1018. Parental strain B-3501A was transformed by an ApaI/XbaI linearized DNA fragment of pYCC1018 using biolistic transformation, and transformants were selected on YPD plates containing 100 μM of bathocuproine-disulfonic acid (BCS) and 120 μg/ml of G418 sulfate. The desired replacement strain, C1617, was obtained by PCR screening and Southern blot analysis. To reconstitute the 14-3-3 strain, the entire 14-3-3 gene was cloned by PCR using primers (Bmh1F2: 5’-CGGGCTTCTCTTCCTTCTCT-3’; Bmh1E3: 5’-GCTAGTTTCTACATCTCTTCCGTGAATGC CGTTGCAAATAAAGC-3’) and the resulting clone was sequenced and cloned into the HYG vector to yield pYCC1039. C1617 was transformed with ApaI-linearized pYCC1039 and the hygromycin-resistant and G418-sensitive transformants were selected. Southern blot analysis was used to identify the correct reconstituted strain, C1617. Yeast cells were grown aerobically at 30℃ in 1% yeast extracts, 2% peptone, and 2% dextrose (YPD). Cells were harvested at the early log phase and washed with YPD, and the cell concentration was adjusted to an optical density of 1 (~108 cells/ml), or as indicated, prior to the experiments. For protein blot studies, the anti-14-3-3 (pan) antibody (#8312) was purchased from Cell Signaling Technology.

Fig. 1.The protein level of 14-3-3 mutants is modulated by the presence or absence of copper. The copper control strain was constructed as described in details in Materials and Methods. (A) After transforming the disruption vector pYCC1018, single colonies were randomly selected and analyzed by PCR screening using Bhm1E2 and Bhm1F2 as the primers. The candidate clones were identified with an up-shifted band, as the larger size of the CTR4 promoter. Then chromosomal DNA was prepared from the selected clones, digested with ScaI or PvuII for the chromosomal restriction mapping. A Dig-labeled NEO coding region fragment was used as a probe. Among the four PCR-positive clones, two clones contained a single NEO gene (clones 70 and 84). Clone #70 was selected and designed as C1617. (B) Protein blots were used to detect the 14-3-3 levels from the identified strain. Yeast cells were grown in YPD and harvested at the early log phase. Equal amounts of protein extracts from each strain were separated by SDS-PAGE and the blot was probed with anti-14-3-3 antibody. β-Actin was used as the loading control. B-3501A is the parental strain, C1617 is the copper control mutant strain, and C1637 is the reconstituted control strain. (C) Protein extracts of YPD-grown B-3501A (1st row) and mutant strain C1617 (2nd row) were used as the controls. Protein extracts were also obtained after incubation of C1617 in 100 μM BCS (3rd row) or 25 μM copper sulfate plus 1 mM ascorbic acid (4th row) for 0, 2, 4, and 6 h. The blot was probe with anti-14-3-3 antibody.

Transmission Electron Microscopy (TEM)

TEM was used to examine the effect of 14-3-3 mutation regulated by the copper-regulatory promoter. Cells were incubated in YPD in the presence of 25 μM copper sulfate and 1 mM ascorbic acid for 8 h. The cells were washed four times with PBS, and fixed with 4% paraformaldehyde and 2.5% glutaraldehyde in PBS for 1 h. The cells were then centrifuged for 10 min at 7,000 ×g and postfixed with buffered 2% OsO4 for 1 h, dehydrated through graded ethanol solutions and propylene oxide, and embedded in Epon. Ultrathin sections were cut, stained with uranyl acetate and lead citrate, and examined with a Philips CM transmission electron microscope.

Isolation of CnMVs

CnMVs were isolated as described previously [33], with slight modification [14]. The yeast cultures were centrifuged at 8,000 ×g for 10 min. The culture supernatant was filtered through a Millipore filter unit with 0.8 μm pore size to avoid any contamination of cryptococcal cells in the preparation. The crude CnMVs were collected from the supernatant by ultracentrifugation at 100,000 ×g for 1 h. The pellet was resuspended in PBS containing gentamicin and placed on a gradient of 0.5 ml of 0.5 M, 1 ml of 1.0 M, and 1 ml of 2.0 M sucrose solutions in an SW51 tube and centrifuged at 100,000 ×g for 2 h. After centrifugation, the purified CnMVs were harvested from the interface between the 1.0 and 2.0 M sucrose solutions.

Proteomic Studies

MV proteins (10 μg) from each sample were separated on a 1D gel and in-gel digested with trypsin as previously described [38]. The resulting peptides were cleaned up using C18 OmixTips (Varian) and analyzed at the CHLA Proteomics Facility using an Eksigent nanoLC-2D coupled to a Thermo Orbitrap XL mass spectrometer, as previously described [1]. Proteins were identified on the basis of their tandem mass spectra using the Bioworks (Thermo, MA, USA) and Scaffold (Proteome Software Inc., OR, USA) [36] protein identification software packages. All proteins were identified with at least two peptides, with the peptide and protein probabilities of at least 95%, and with C carboxymethylation, M oxidation, and STY phosphorylation as variable modifications. Relative protein abundance across the three samples was determined in Scaffold using the number of assigned spectra.

Measurement of Laccase and Acid Phosphatase

The enzymatic activity of laccase and acid phosphatase was determined as described previously [32]. Briefly, 10 μl of CnMV resuspension from fresh cultures was incubated with either L-DOPA for the laccase assay, or nitrophenyl phosphate in acetate buffer (pH 5) for the acid phosphatase assay. The reactions (100 μl) were carried out at room temperature for 12 h and the absorbance was analyzed at 450 nm (laccase) or 405 nm (acid phosphatase) spectrophotometrically.

In Vitro C. neoformans Adhesion Assay

The C. neoformans in vitro adhesion assay was performed as previously described [5]. HBMEC were grown in collagen-coated 24-well culture plates (Costar) until confluence. An inoculum of 106 Cryptococcus cells in 1 ml of experimental medium was added (moi of ~10) at 37℃ in the presence or absence of suppressor (50 μM CuSO4, 1 mM ascorbic acid). After 3 h, free yeast cells were removed from the HBMEC by washing four times with the experimental medium. Some adherent yeast cells were retained by the HBMEC. Subsequently, the HBMEC were lysed with 0.5% Triton, diluted, and plated onto YPD agar plates, and colonies were counted after incubation at 30℃ for 2 days.

Statistical Analysis

Analysis of variance (ANOVA) was used for statistical analysis of the data obtained in this study. The dependent variable was the percent of cells or the CFU, whereas the independent fixed factors were the treatments (different strains, different culture media in the presence or absence of CuSO4, BSC, etc.). Raw data were entered into Microsoft Excel files and converted automatically into statistical packages. ANOVA and co-variates were followed by a multiple comparison test, such as the Newmann-Keuls test, to calculate the statistical significance between the control and treatment groups. p < 0.05 was considered significant.

 

Results

Construction of Copper-Repressible 14-3-3 Mutant Strain

In order to study the function of C. neoformans 14-3-3, we attempted to delete the gene using the neomycin cassette to replace the chromosomal 14-3-3 gene in serotype D strain B-3501A. However, no 14-3-3 deletant was obtained after screening more than 200 transformants. The inability to disrupt the gene is most likely due to the essentiality of 14-3-3 in C. neoformans. As the 14-3-3 gene was resistant to disruption, we then constructed a 14-3-3 mutant strain by replacing the endogenous promoter of 14-3-3 with the CTR4 promoter. The resulting strain, confirmed through PCR screening and chromosome restriction mapping with ScaI and PvuII, was designated as the C1617 mutant (Fig. 1A). Since CTR4 expression can be regulated by availability of copper in the culture media [29], thereby the expression of the 14-3-3 gene in our mutant strain could be controlled by the presence or absence of copper. We also constructed a control strain, C1637, by reconstituting the CTR4 promoter in the 14-3-3 locus of C1617 with the native 14-3-3 promoter.

We first used an antibody raised against conserved residues surrounding Met223 of human 14-3-3γ protein, anti-14-3-3(pan)Ab, to monitor the 14-3-3 protein levels of C. neoformans (Fig. 1B). In YPD medium, the 14-3-3 protein level was much lower in the mutant compared with the wild-type strain B-3501A whereas the reconstituted strain C1637 produced similar levels of the 14-3-3 proteins as the wild type. Since YPD is a rich medium and contains a trace amount of copper, the lower expression of 14-3-3 in the mutant strain might be due to the effect of copper suppression. The time-point study was performed by growing C. neoformans in YPD and then the cultures were shifted to different media. We found that the 14-3-3 protein levels were not altered in the wild-type strain B-3501A grown in YPD (Fig. 1C, row 1) or YPD with the presence of copper chelator bathocuproine-disulfonic acid (BCS) or 50 μM CuSO4 and 1 mM ascorbic acid (data not shown). In contrast, the 14-3-3 protein levels in the mutant strain C1617 were gradually increased in the presence of BCS, in which the available copper was chelated (Fig. 1C, row 3), but gradually reduced in the presence of copper (Fig. 1C, row 4). Thus, the 14-3-3 protein levels can be regulated by the presence or absence of copper through the CTR4 promoter in the mutant strain C1617.

Phenotypes of the 14-3-3 Mutants

The morphology of the strains was examined under the microscope. The mutant cells showed similar morphology as the wild type in YPD medium in the presence of inducer BCS (Fig. 2Aa, 2Ab, and 2Ac). However, in the presence of copper/ascorbic acid, the size of mutant cells was enlarged compared with the wild type and the separation between the daughter and mother cells was affected and produced a chain-like structure in the mutant (Fig. 2Ad).

Fig. 2.Phenotype of 14-3-3 mutant strain. (A) Strains B-3501A, C1617, and C1637 were grown on YPD medium at 30℃. Cell morphology was examined under a light microscope. (a) B-3501A grown in YPD; (b) C1617 grown in YPD; (c) C1617 grown in YPD with 100 μM BCS, and (d) C1617 grown in YPD with 25 μM copper sulfate and 1 mM ascorbic acid. (B) Immunofluorescence microscopy was used to probe B-3501A cells (left), C1617 (middle), and C1637 (right) with anti-GXM monoclonal antibody (18b7) and FITC-conjugated second antibody. Cells were grown on YPD in the presence of 25 μM copper sulfate and 1 mM ascorbic acid for 8h. Capsules are shown in green signals. Bar = 10 μm.

Immunofluorescence microscopic images of capsular component glucouronxylomannan (GXM) stains was further performed to compare the images between wild-type B-3501A and mutant C1617 cells in the presence of copper (Fig. 2B). The wild-type cells showed dense, bright GXM signals surrounding the cells (Fig. 2B, first panel). The size of mutant cells was slightly larger, as shown previously (Fig. 2Ad). However the GXM stains were much thinner, showing only the contour of the attached cells (Fig. 2B, middle panel).

Expression Levels of C. neoformans 14-3-3 Are Relevant to the Capsule Biosynthesis

To explore the development of the mutant morphological features, TEM was performed in the presence of copper/ascorbic acid for 8 h. The size of the mutant cells was generally larger than the wild type (Fig. 3A vs. 3C). Multiple cells attached to each other were often observed in the mutant strain, consistent with the observation by light microscopy (Fig. 2). Nonetheless, the presence of cell wall was observed between the attached mother and daughter cells (Fig. 3B). Presumably, lack of sufficient 14-3-3 expression led to a defect in cytokinesis and resulted in difficulty in cell separation. Taken together, adequate expression of 14-3-3 seems necessary for normal cell growth and morphology.

Fig. 3.Morphological analysis of 14-3-3 mutant cells. Cells were cultured for 8 h in YPD in the presence of 25 μM copper sulfate and 1 mM ascorbic acid. Samples were fixed and processed for TEM. (A) C1617 cells. The square boxes in (A) are amplified and shown in (B). CW: cell wall; Caspule, indicated by arrows. (C) Wild-type B-3501A cells. (D) The lengths of capsule filaments were measured from strains B-3501A and C1617 under the copper treatment. Bar = either 1 μm or 5 μm, as indicated in each image (**p < 0.01).

In addition to the morphological alteration, another obvious differentiating feature was the capsule production. In general, the wild-type B-3501A showed a thick, dense capsule, and displayed ray-like stains in the TEM image (Fig. 3C). However, compared with the wild type, the mutant cells displayed lighter and shorter capsule stains (Fig. 3A vs. 3C). The extension of the fiber-like capsule stains was quantified to examine the distinctions between these two strains. Ten sections were randomly chosen from the TEM images of the wild-type and mutant cells and the extensions of those capsule fiber-like structures were measured. The relative length of the hair-like structure extensions was significantly shorter in the mutant cells than in the wild type (Fig. 3D, p < 0.01). Overall, these findings are consistent with the notion that 14-3-3 plays an important role for C. neoformans capsule biosynthesis.

14-3-3 Expression Levels Affect the Production of Microvesicles

14-3-3 is an abundant protein in C. neoformans-derived microvesicles and has been used as the CnMV marker for both in vitro and in vivo studies [18]. We examined the effect of down-regulated 14-3-3 on the CnMV secretion. An equivalent cell number of C. neoformans cells grown in YPD at 30℃ was harvested for the CnMV preparation. The yield of total MV proteins was lower in the mutant strain compared with the wild type (Fig. 4A). When the same amount of proteins was used from each preparation, the enzymatic activities of two known MV proteins, laccase (Fig. 4B) and acid phosphatase (Fig. 4C), were also reduced in the mutant strain. These results suggest a dependency of MV secretion on 14-3-3 expression.

Fig. 4.Expression levels of 14-3-3 affect the MV protein content. CnMVs were isolated from equal amounts of cells of indicated strains. (A) Total protein concentration of the isolated CnMV was determined by the Bradford assay. The enzymatic activities of laccase (B) and acid phosphatase (C) associated with the isolate CnMVs were measured in a reaction volume of 100 μl (see Materials and Methods). An equal volume of MV preparations from fresh culture (10 μl) was used in parallel for comparison. After incubation for 1 h, the difference in readings between the initial time and 60 min indicated their enzymatic reactivity as indicated on the ordinate.

Different Expression Levels of 14-3-3 Proteins Alter C. neoformans Adherence to HBMEC in Vitro

It has been suggested that CnMV production may play a role in cryptococcal pathogenicity [18]. Experiments were performed to investigate whether different levels of 14-3-3 affect the adhesion between C. neoformans and HBMEC. Parental strain B-3501A, mutant strain C1617, and reconstituted strain C1637 were used in parallel. To modulate the levels of 14-3-3 protein, one experimental condition was to grow the cells in YPD, and the other condition was to grow the cells in the presence of 25 μM copper sulfate and 1 mM ascorbic acid. The adherence of C. neoformans to HBMEC was determined. As shown in Fig. 5, in YPD medium, the adhesion ability of the mutant cells to HBMEC was slightly reduced compared with the wild type, which most likely was due to the slower growth rate of the mutant strain than the wild type. In the presence of copper sulfate, the binding ability to HBMEC showed some effects in all strains; but a significantly reduction was observed in the mutant strain C1617 (Fig. 5, p < 0.05). These results suggest that 14-3-3 is relevant to C. neoformans brain invasion.

Fig. 5.Different expression levels of the 14-3-3 protein alter C. neoformans adherence to HBMEC in vitro. In vitro adhesion assays were performed. HBMEC were grown in 24-well tissue plates with 106 yeast cells in the indicated conditions. The reaction was carried out for 3 h in the absence (left three bars) or presence of 25 μM copper sulfate and 1 mM ascorbic acid (three bars on the right). Data showed that copper-sulfate-treated mutant cells had a significant decrease in HBMEC adhesion (*p < 0.05).

Analysis of Protein Components of Secreted Microvesicles under Different 14-3-3 Expression Conditions

It is known that 14-3-3 gene functions are involved in vesicular transport/trafficking and chromatin functions in different species [31]. Proteomic analyses were performed using equal amounts of CnMV proteins from induced and suppressed conditions, relative to the untreated control. A Venn diagram of the proteomic studies shows the number of unique identified protein, not protein abundance, in each group (Fig. 6A). The control group detected 163 polypeptides, and both overexpressed (+BCS) and suppressed (+copper) groups detected 110 polypeptides. Despite the same number, the latter two contained slightly different protein compositions. The majority of protein components were located within the control sample (65 in the overlapping regions), whereas some unique components were identified in the overexpressed and suppressed conditions, respectively. Quantitatively, abundance ratios for the 38 proteins were found to be more copious only in the 14-3-3 overexpression condition (Fig. 6B). One example, the 14-3-3 protein was increased >7-folds in the presence of inducer, and was essentially undetectable in the presence of suppressor. On the other hand, abundance ratios for the 52 proteins were found to be more plentiful only in the 14-3-3 suppression condition (Fig. 6C). The original data that showed the relative abundance of protein components from isolated CnMVs under induction, control, and suppression conditions are listed in the Supplementary file (Columns, A: Identified proteins; B: Gene accession number; C: Molecular weight; D: Protein abundance under control condition; E: Protein abundance under suppression condition; F: Protein abundance under overexpression condition). Our results showed that fluctuation of protein components could be readily observed at different 14-3-3 levels, suggesting that alterations of physiological status might occur under different 14-3-3 expression conditions.

Fig. 6.Protein components of secreted microvesicles in different 14-3-3 expression conditions. (A) Venn diagram showing the number of different proteins identified in one or more of the three expression conditions. (B) Abundance ratios for each of the 38 CnMV proteins found to have at least 2-fold greater abundance only in 14-3-3 overexpression versus the control. The line showing the 14-3-3 protein level is indicated. (C) Abundance ratios for each of the 52 CnMV proteins found to have at least 2-fold greater abundance only in 14-3-3 suppression versus the control. The original data are provided in the Supplementary file.

 

Discussion

We are interested in this protein because 14-3-3 is an abundant protein in secreted microvesicles [14] and its biological role has never been explored in C. neoformans. We attempted to knock out the 14-3-3 gene to examine its phenotypic consequence. As the 14-3-3 gene was recalcitrant to disruption, we then replaced the endogenous promoter of 14-3-3 with the CTR4 promoter. The CTR4 promoter is suppressed in the presence of CuSO4, and in opposite, the downstream gene can be turned on in the presence of a copper chelator, such as BCS [29]. In our studies, BCS did not affect the expression of 14-3-3 and morphology of the wild-type strain B-3501A, but BCS could boost 14-3-3 expression in the CTR4 promoter-controlled 14-3-3 mutant strain (Fig. 1). Time-point studies indicated that 14-3-3 protein (>90%) was dramatically reduced after 6 h in the presence of copper in the mutant. Correspondingly, in the presence of copper, the mutant strain also altered its morphology drastically, and the morphological change could be observed in the cell population as short as half an hour, suggesting a threshold of 14-3-3 molecules is required for normal cell morphology. In Schizosaccharomyces pombe, two 14-3-3 genes, rad24 and rad25, share overlapping functions that are essential for cell growth and cell wall integrity [15]. The rad24 null cells are defective in cytokinesis, displaying cone-shaped cell morphology, but this is not the case with the rad25 null mutant [9]. Consistent with these observations, our results also suggest that a low level of 14-3-3 expression may alter the cytoskeleton and also jeopardize separation during cell division (Figs. 2 and 3). As such, the growth rate of mutant cells was slower in YPD medium and was further reduced in the presence of copper sulfate. Overall, C. neoformans 14-3-3 protein, as in other organisms, plays several important roles relevant to its growth, morphology, cell division, and other functions (see below).

In S. cerevisiae, overexpression of the C-terminal region of 14-3-3 shows an effect on the disruption of the actin cytoskeleton, resulting in vesicle targeting defects [34]. Proteomic studies show that yeast 14-3-3 interacts with more than 200 human protein components, including a group of cellular trafficking proteins, such as Sec23, Sec24, vps33, importin, etc. [31]. Indeed, the 14-3-3 protein was first identified in bovine adrenal chromaffin cells as a cytosolic protein Exo1, required for exocytosis [25]. Based on the above observations, one would speculate that the C. neoformans 14-3-3 protein may also play a role in secretory/trafficking pathways. Interestingly, in our 14-3-3 mutant, both total CnMV proteins and the activities of two enzymes associated with CnMV, laccase and acid phosphatase, were reduced (Fig. 4). It is also known that the secretion of microvesicles plays a role in the building block transport for its extracellular capsule biosynthesis [32]. It is possible that reduction of 14-3-3 protein levels in strain C1617 results in the decline of MV secretion and subsequently drops the supplies for capsule biosynthesis, which results in the reduction of capsule size. Furthermore, the exocytosis pathway is required for PKC functions, and the C. neoformans 14-3-3 has a conserved PKC phosphorylation site [26]. Whether PKC exerts its gene function through the 14-3-3 protein to regulate the exocytosis or secretion pathways is currently unknown.

Lack of a good way to control body copper concentrations prevents from doing animal studies. However, the in vitro adhesion studies show that the decline in 14-3-3 protein levels in the mutant strain causes a decrease in its ability to bind to HBMEC (Fig. 5). This decrease may be attributed to the less amount of CnMV secreted in the mutant, since CnMV can affect the binding of cryptococcal cells to HBMEC [18], although we cannot rule out other possible reasons. Given the fact that the 14-3-3 functions are relevant to its growth, morphology, possible cytokinesis, and capsule and microvesicle biogenesis, it is perceivable that the C. neoformans 14-3-3 functions are closely associated with C. neoformans physiology and pathogenesis.

It is widely accepted that 14-3-3 is a chaperone protein, primarily for phosphorylation regulation [24,27,42]. The interacting component list consists of more than 200 proteins [31], and the number of interacting partners is ever increasing. However, the physiological roles of most these interactions have yet to be rigorously tested. Unicellular yeasts have intractable genetic systems, which may provide a solution for exploring the biological significance of a large amount of interacting partners. Nevertheless, some commonly studied yeasts such as S. cerevisiae contain two 14-3-3 genes (BHM1 and BHM2) [39], and S. pombe has RAD24 and RAD25 [15]. On account of C. neoformans consisting of one 14-3-3 gene, it would be an excellent model system to understand the physiological and pathological roles of this conspicuous protein.

In conclusion, our results support the notion that 14-3-3 has various pertinent roles in the physiology of C. neoformans, including cell growth, temperature sensitivity, cell division, morphology, microvesicle secretion, and gene expression. Its gene functions are closely relevant to the pathogenesis of this fungus. Our constructed strains will also be a useful model to explore its gene functions in eukaryotes.