The transcription factor Ndt80 is a repressor of Candida parapsilosis virulence attributes

ABSTRACT Candida parapsilosis is an emergent opportunistic yeast among hospital settings that affects mainly neonates and immunocompromised patients. Its most remarkable virulence traits are the ability to adhere to prosthetic materials, as well as the formation of biofilm on abiotic surfaces. The Ndt80 transcription factor was identified as one of the regulators of biofilm formation by C. parapsilosis; however, its function in this process was not yet clarified. By knocking out NDT80 (CPAR2-213640) gene, or even just one single copy of the gene, we observed substantial alterations of virulence attributes, including morphogenetic changes, adhesion and biofilm growth profiles. Both ndt80Δ and ndt80ΔΔ mutants changed colony and cell morphologies from smooth, yeast-shaped to crepe and pseudohyphal elongated forms, exhibiting promoted adherence to polystyrene microspheres and notably, forming a higher amount of biofilm compared to wild-type strain. Interestingly, we identified transcription factors Ume6, Cph2, Cwh41, Ace2, Bcr1, protein kinase Mkc1 and adhesin Als7 to be under Ndt80 negative regulation, partially explaining the phenotypes displayed by the ndt80ΔΔ mutant. Furthermore, ndt80ΔΔ pseudohyphae adhered more rapidly and were more resistant to murine macrophage attack, becoming deleterious to such cells after phagocytosis. Unexpectedly, our findings provide the first evidence for a direct role of Ndt80 as a repressor of C. parapsilosis virulence attributes. This finding shows that C. parapsilosis Ndt80 functionally diverges from its homolog in the close related fungal pathogen C. albicans.


Introduction
Candida parapsilosis is a ubiquitous yeast, often recovered from domestic animals, soil, and marine environments, but is also a commensal of the human skin. Among hospital settings, this species is considered a major opportunistic pathogen involved in invasive fungal infections [1,2]. Its incidence has dramatically increased, being the second most common Candida species isolated from blood cultures in Latin America, Asia, and Southern Europe countries [3][4][5][6][7][8]. C. parapsilosis is of particular concern among susceptible populations, comprising low birth weight neonates, immunocompromised individuals, and patients requiring prolonged use of indwelling devices such as central venous catheters [1,9]. Besides its ability to grow and persist in the hospital environment surfaces, C. parapsilosis stands out for its capacity to adhere to the abiotic surface of implanted devices, later involving biofilm formation [1,10,11]. In fact, adhesion and formation of biofilm are intimately related with C. parapsilosis virulence and are critical for its involvement in hospital outbreaks [2].
Biofilm is an organized community comprised of a dense network of microbial cells embedded in an extracellular matrix of polymers, which clinically restricts drug access and the immune response [12,13]. The ability of fungal cells to adhere to host tissues or medical indwelling devices, as well as cellcell binding are required for biofilm development and for infection proliferation [14][15][16]. In contrast to C. albicans, C. parapsilosis does not form true hyphae and, therefore, its biofilm only involves yeast and pseudohyphal forms [17,18]. To identify putative C. parapsilosis biofilm regulators, more than 100 transcription factors were knocked-out and mutants were assessed for biofilm formation ability [19]. Previously identified as biofilm regulators in C. albicans, Bcr1, Efg1 and Ace2 were also directly implicated in biofilm development in C. parapsilosis [16,[19][20][21][22], together with the transcription factor Gzf3, whose involvement in biofilm formation seems to be restricted to C. parapsilosis [19]. In this largescale screen of C. parapsilosis biofilm defective mutants, NDT80 was firstly pointed as a putative biofilm regulator, in analogy with C. albicans biofilm regulation network. However, in the case of C. parapsilosis, NDT80 role was undisclosed due to marked growth defects exhibited by ndt80 mutant [19]. In C. albicans, Ndt80 was first described as a key modulator of azole drug sensitivity, being involved in the control of ergosterol biosynthesis [23] and activation of the efflux pump Cdr1 [24]. We firstly identified C. parapsilosis Ndt80 ortholog to be a transcription factor upregulated following azole resistance acquisition [25]. Later, we showed that ndt80 mutant exhibits increased susceptibility to azoles and that, together with Upc2 transcription factor, also regulates the expression of various genes of ergosterol biosynthetic pathway, namely ERG25, ERG6, ERG2, ERG3 and ERG4 [26].
In this study, we address the role of Ndt80 in C. parapsilosis as a repressor of virulence attribute expression, namely morphogenesis, adhesion, and biofilm formation. Additionally, we explore the morphological phenotypes, its constitutive filamentous growth, and the adhesion profile resulting from NDT80 knockout, as well as its interaction with host immune system by assessing macrophage-mediated response.
RAW 264.7 murine macrophages were obtained from the European Collection of Cell Cultures and maintained in DMEM (Sigma-Aldrich) with 10% noninactivated Fetal Calf Serum (FCS), 10 mM HEPES, 12 mM sodium bicarbonate and 11 mg ml −1 sodium pyruvate at 37°C in a humidified atmosphere with 5% CO 2 . The culture medium was changed every 2 days, until ~70% of cell confluence was reached. RAW 264.7 cells were resuspended in RPMI 1640 medium (Sigma-Aldrich) supplemented with 10% inactivated FCS, 23.8 mM sodium bicarbonate and 50 mM glucose for the experimental assays (initiated until the cells 15 th generation).

Plasmid construction
To knockout NDT80 gene in C. parapsilosis BC014S (wild-type strain) [25], the pNG4 disruption cassette described by Branco et al. [26] was used. Briefly, a 478 bp upstream and 460 bp downstream sequences of NDT80 gene were amplified using CpNDT80up_F and CpNDT80up_R primers (containing recognition sites for KpnI and ApaI) and CpNDT80down_F and CpNDT80down_R primers (containing recognition sites for SacII and SacI), respectively, and cloned into the flanking sites of pCD8 plasmid [18]. After restriction with KpnI and SacI, pNG4 disruption cassette was introduced into the native locus of NDT80 gene of C. parapsilosis BC014S. All primer sequences are listed in Table 1.

C. parapsilosis transformation
Transformation of wild-type strain was performed by electroporation as described by Ding et al. [18]. Briefly, an overnight cell culture was diluted in 50 ml of YPD broth medium for an initial OD 600 of 0.2 and incubated at 30°C until reaching approximately OD 600 of 2.0. After being pelleted, yeast cells were resuspended in 10 ml of Tris-EDTA buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.5) containing 10 mM dithiothreitol and incubated at 30°C for 1 h with agitation (100 rpm). Yeast cells were washed twice with 40 ml of cold water plus once with 10 ml 1 M Sorbitol and, subsequently resuspended in 125 μl of this solution. Approximately 1 µg of purified KpnI-SacI fragment of pNG4 was added to 50 µl of competent cells. The cell mixture was then transferred to a 1 mm electroporation cuvette. Electroporation shock was performed at 1.25 kV, using a Gene Pulser X-cell Electroporator (Bio-Rad). Afterward, 950 μl of YPD containing 1 M sorbitol was immediately added; the mixture was incubated at 30°C for 4 h with agitation; afterward 100 μl were plated on YPD agar supplemented with nourseothricin at final concentration of 200 μg ml −1 . Transformants were obtained after 24 h of incubation at 30°C.

Adhesion assay
Yeast adhesion was quantified by flow cytometry, as described by Silva-Dias et al. [27]. Briefly, yeasts were grown overnight at 30°C in Sabouraud broth medium, with agitation (180 rpm); the culture was centrifuged at 10,000 g for 5 min and washed twice with phosphate buffer saline (PBS) (Sigma-Aldrich). A yeast suspension was standardized to 2 × 10 6 cells ml −1 in the same buffer and mixed with 2 × 10 7 microspheres ml −1 of 1 μm uncoated carboxylated highly green fluorescent polystyrene microspheres (Molecular Probes). This mixture was incubated at room temperature for 30 min at 150 rpm. The suspensions were vortexed, and 50,000 events were analyzed using a FACS Calibur flow cytometer (BD Biosciences). Cell adhesion results are expressed as the percentage of cells with microspheres attached, representative of at least three independent experiments, performed in triplicate.

Biofilm formation assays
After overnight growth at 37°C with agitation (180 rpm) in Sabouraud broth medium, yeast cells were collected by centrifugation at 10,000 g for 5 min, washed once with PBS and standardized to obtain a suspension of 1 × 10 6 yeast cells ml −1 in RPMI-1640 medium supplemented with L-glutamine and buffered with MOPS acid (Sigma-Aldrich). One ml of such cell suspension was placed in each of a 12-well polystyrene microplate and incubated for 24 and 48 h at 37°C. Following incubation, total biomass was quantified by Crystal Violet (CV) assay, as previously described by Silva-Dias et al. [28]. Biofilm mass was calculated from at least three independent experiments, performed in triplicate. For dry mass assessment, C. parapsilosis strains were set up as previously described, except the standardization of the cell suspension, which was diluted to an OD 600 of 1; afterward, 5 ml were distributed in each well of a 6-well polystyrene plate. After 24 and 48 h of incubation at 37°C, adherent biofilms were washed with PBS, scrapped from the bottom of the wells, and vacuum filtered, as described by Holland, et al. [19]. The average of the total biomass was calculated by subtracting the initial weight of the filter to the final weight, determined from three independent experiments, performed in triplicate.

Microscopic imaging
Colony phenotypes were observed and photographed under 20× magnification using a Stereo zoom S9i (Leica Microsystems) dissection microscope, after growth on YPD agar at 30°C, for 72 h. Images of yeast cell morphology were taken with a Zeiss Axioplan microscope, coupled with an AxioVision image acquisition system (Zeiss), after staining with Calcofluor White (Sigma-Aldrich) and mounting on glass slides. Yeast cells were photographed under 1000× magnification, oil immersion.

RNA extraction, cDNA synthesis and RT-qPCR
RNA was extracted as described by Kohrer and Domdey [29]. Concentration and quality of RNA samples were measured using a Nanodrop equipment (Eppendorf). Only samples yielding A 280 /A 260 ratios ranging from 1.6 to 2.2 and showing no signs of degradation, after electrophoresis, were used in subsequent analyses.
From 100 ng of total RNA, the first-strand cDNA was synthesized using the SensiFAST cDNA Synthesis Kit (Bioline) according to the manufacturer´s instructions. The resulting cDNA was stored at −20°C prior to use for real-time quantitative polymerase chain reaction (RT-qPCR). The genes analyzed were the followed: For each real-time quantitative PCR, five replicates per strain were analyzed. All primers used are detailed in Table 1. PCRs were performed using the SensiFAST SYBR Hi-ROX Kit (Bioline) 3-step cycling, according to the manufacturer's instructions, in a PikoReal Real-Time PCR System instrument (Thermo Scientific). ACT1 gene expression was used to normalize the signal obtained for each gene. Data obtained were analyzed with REST software.

Bioinformatic analysis
Sequences from C. parapsilosis CDC317 open reading frames (ORFs) plus 1000 bp upstream and downstream (version s01-m03-r14, from 7 February 2016) were downloaded from the Candida Genome Database (CGD, http://candidagenome.org/). To identify putative Ndt80-regulated genes, a search for the MSE consensus motif (gNCRCAAAY) was performed in the promoter regions (1000 bp upstream the start codon). The resulting ORFs containing MSE sequences were grouped according to Gene Ontology (GO) terms using the CGD Gene Ontology Slim Mapper with the default parameters.

Macrophage-yeast interaction assays
Macrophage-yeast interaction assays were carried out as previously described [30]. Briefly, RAW 264.7 macrophage cells were platted in 96-, in 12-well (with 16 mm glass coverslips) or in µ-slide 8 well plates, and incubated for 18 h at 37°C, under a 5% CO2 atmosphere. After this incubation period, yeast cells were added to the macrophages at an MOI (Multiplicity of Infection) of 1:1.

Immunofluorescence and microscopic analysis
Macrophages grown in coverslips were incubated with C. parapsilosis as described below. At the end of each incubation period (10 min, 30 min, 1 h 30 min, 3 h), coverslips were washed twice with ice-cold PBS and fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. After 3 washing steps with PBS, cell membranes were stained with WGA, for 10 min, protected from light. Macrophages were treated with a blocking solution of 10% bovine serum albumin in PBS for 30 min at 37°C. Cells were then incubated overnight, at room temperature, with the primary rabbit polyclonal antibody against Candida (GTX40096; GeneTex), diluted (1:200) in blocking solution. Coverslips were washed and incubated for 2 h at room temperature with the AlexaFluor 488 donkey anti-rabbit IgG secondary antibody (A21206; Invitrogen). Finally, after a washing step, macrophage cells were incubated with DAPI 0.02% for 10 min at room temperature. Cells were subsequently washed and the coverslips were mounted in glass slides with DAKO mounting medium and kept at −20°C until observation under confocal or fluorescence microscopy. Digital images were captured using a Carl Zeiss LSM 710 Confocal Microscope, using Plan-ApoChromat 40x/ 63x/1.4 oil objectives; Zen Blue and FiJi software's were used to analyze the images.

Yeast and macrophage viability assays
The yeast cell viability following interaction with RAW 264.7 macrophage cells was assessed by a colonyforming unit (CFU) assay. After 30 min and 3 h of coincubation, supernatants were collected and plated on YPD agar, to count non-internalized or non-adhered yeast cells. The remaining adhered RAW 264.7 macrophages were scraped and lysed with 0.5% Triton X-100. This cell suspension, representing the amount of yeast cells internalized was plated on YPD agar, using serial dilutions. Following 3 days of incubation, at 30°C, the number of yeast colonies per ml was calculated.
For macrophage viability assay, after 30 min and 3 h of co-incubation, viable, and death macrophage cells were calculated using a hemocytometer, after staining with Trypan Blue (T8154; Sigma-Aldrich).

Live cell imaging assays
For live cell imaging assays, culture media without phenol-red was used and macrophage cell membranes were stained with Wheat Germ Agglutinin, Tetramethylrhodamine conjugate (WGA, W849; Molecular Probes). Image acquisitions were conducted during at least 45 min, using a confocal Cell Observer Spinning Disk microscope (Zeiss), equipped with an LCI PlanNeofluar 63x/1.3 glycerol objective; Zen Blue software was used to analyze the time-lapse videos obtained.

Statistical analysis
Statistical analysis of results of adhesion, biofilm and infection assays was performed using one-way ANOVA followed by a Dunnett post hoc test. Differences were considered statistically significant for a p-value <0.05. Significant differences were marked with an asterisk character (*), in which *p < 0.05, **p < 0.01, ***p < 0.001. All results are presented as mean ± standard deviation, of at least three independent experiments.

Deleting NDT80 transcription factor gene triggers morphogenesis
To gain insight into the role of Ndt80 in C. parapsilosis virulence attribute expression, two independent lineages lacking one (ndt80Δ -NG2 strain) or both (ndt80ΔΔ -EF16 strain) copies of NDT80 were generated from C. parapsilosis strain BC014S (wild-type strain) [25]. Deletion was carried out using a genespecific disruption cassette (pNG4) based on the recyclable nourseothricin-resistant marker as previously described [18]. The introduction of pNG4 into theNDT80 locus of the wild-type strain, generated NG1 clone, which after cassette recycling, resulted in the NG2 strain. To delete the second copy of NDT80 gene, a second round of integration/recycling were performed, generating EF15 and EF16 clones, respectively. Gene knockout was confirmed by PCR (Figure 1 (a) and (b)).
Deletion of NDT80 had a major effect upon colony and yeast cell morphology (Figure 2(a) and (b)). The parental strain and the ndt80Δ haploid mutant grow as smooth-white and creaky-opaque colonies, respectively, whereas colonies from ndt80ΔΔ diploid mutant display a crepe phenotype. Wild-type and haploid cells are yeast-shaped cells; in contrast, the ndt80ΔΔ cell population is mostly composed of elongated cells and pseudohyphae.

Deleting NDT80 increases adhesion and biofilm formation ability
The yeast to pseudohyphae transition was observed along with the formation of fungal cell aggregates, typical of enhanced cell to cell adhesion. The ndt80Δ and ndt80ΔΔ mutants flocculate in liquid medium, suggesting that Ndt80 negatively affects the cell-cell adhesion process (Figure 3(a)). The ability of C. parapsilosis to adhere to polystyrene microspheres, representative of abiotic surfaces, was quantified using a flow cytometric adhesion assay, as described previously [27]. Compared to wild-type, manipulated strains displayed a significant increase of about 2-fold in adhesion ability (Figure 3(b)).
Filamentous growth and adhesion displayed by ndt80ΔΔ mutant are two known enhancers of biofilm formation. We assessed wild-type and mutants strains regarding the ability to form biofilm, using two independent methods, Cristal Violet (CV) staining [28] and dry weight [19]. C. parapsilosis lacking one or both copies of NDT80 gene exhibits enhanced capacity to form biofilm compared to wild-type strain (Figure 3(c) and (d)). Differences were statistically significant when using both methodologies. Nevertheless, comparatively to ndt80Δ mutant, ndt80ΔΔ mutant had lower biofilm biomass, a result statistically significant when using CV staining for biofilm quantification.
Relatively to adhesin-like genes, the expression of ALS7 in ndt80Δ and ndt80ΔΔ mutants was upregulated 210-and 180-fold, respectively, compared to wild-type. In contrast, ALS3 gene expression was not changed significantly among the studied mutant strains. The expression of UME6 was upregulated, approximately, 5-fold in the ndt80Δ haploid mutant and a 13-fold in the ndt80ΔΔ diploid mutant, compared to the wildtype. MKC1 expression was also upregulated 2.8-fold and 36-fold in haploid and diploid mutants, respectively, comparatively to the wild-type. CPH2 gene exhibited a 1.2-fold upregulation in ndt80Δ mutant and of approximately 4-fold increased expression in ndt80ΔΔ mutant, in comparison to the wild-type.
ACE2, CWH41 and OCH1 genes displayed similar expression values of approximately 3-fold, 2-fold, and 1.2-fold, respectively, in the haploid and diploid mutants. BCR1 gene was 1.5 and 1.7-fold upregulated in ndt80Δ and ndt80ΔΔ mutants in comparison to wild-type. The expression of STP3 was increased approximately 1.8-fold in ndt80Δ mutant but remained unchanged in ndt80ΔΔ mutant. In contrast, EFG1, GZF3 and RHR2 were downregulated in ndt80ΔΔ mutant comparatively to the wild-type; ndt80Δ mutant exhibited a slight upregulation of expression of such genes (of about 1.1-, 1.4-, and 2.6-fold, respectively). CZF1 gene was progressively downregulated following sequential NDT80 gene copy deletion, by approximately 30% and 70%, respectively.
As expected, no NDT80 transcript was observed in the null strain. Interestingly, the expression of NDT80 in ndt80Δ mutant was 1.6-fold up-regulated. Since NDT80 gene has in its promoter region the MSE binding sequence, we could hypothesize that to cope with one copy gene deletion, Ndt80 up-regulates itself expression, as described in S. cerevisiae and A. nidulans [33,34].

Identification of putative NDT80-regulated genes
Ndt80 was found to bind to the middle sporulation element (MSE) (5ʹ-CACAAA-3ʹ) in the target gene promoter region [35] of C. albicans and S. cerevisiae ORFeomes [22,23]. The putative colony transition, adhesion-and biofilm-related genes mentioned above were analyzed for the presence of MSE motifs using the NCBI blast tool. As some of the promoter regions bound by biofilm regulators are larger than the normal [22,35], the considered sequence was approximately 1 kb upstream of the start codon. All genes assessed for their expression (Figure 4) contain putative MSE recognition sites, being identified in promoter regions.
Attaching to such results, we further expanded the search for MSE consensus sequences in the complete C. parapsilosis ORFeome. This analysis allowed the Percentage of yeast cells with adherent beads. ndt80Δ and ndt8ΔΔ mutants exhibited significantly higher adhesion ability than wild-type. The ability to form biofilm was quantified by (c) Cristal Violet (CV) staining and (d) dry weight, following 24 and 48 h of growth; in both assays, a significant increase of biofilm formation by ndt80Δ and ndt8ΔΔ mutants compared to the parental strain was observed. CV staining revealed a statistical decrease in biofilm formation between ndt80Δ and ndt8ΔΔ mutants, at both time points. *p < 0.05, **p < 0.01 and ***p < 0.001 wild-type vs ndt80Δ and ndt8ΔΔ mutants, or both groups.
retrieval of 417 ORFs containing MSE motifs in their promoters. These were mapped to GO terms and grouped according to Biological Process, Molecular Function or Cellular Component ( Figure 5). Results showed that most ORFs with MSE elements (with over 10% and excluding the unknowns) belong to cell transport regulation, organelle organization, response to stress/chemical and RNA metabolic processes. Also, these ORFs are mostly related with enzymes with hydrolase or transferase activity which in addition to the cytoplasm and nucleus, many are located in cell membranes and mitochondria ( Figure 5).

C. parapsilosis strains lacking NDT80 are more resistant to macrophage attack and impair macrophage viability
The capacity of fungal cells to resist to macrophagemediated killing contributes to its pathogenicity [- [36][37][38]. We conducted a phagocytic assay using the murine macrophage cell line RAW264.7 in order to determine the impact resulting from NDT80 deletion upon phagocytic cells response. The interaction between macrophages and C. parapsilosis cells begins as early as 10 min (Figure 6(a)). However, while C. parapsilosis wild-type cells hardly interact, at the same time point a higher number of ndt80ΔΔ cells are attached to macrophages with clear signs of internalization, as indicated by the tridimensional green staining fading (Figure 6(a)); the ndt80Δ cells showed a intermediate behavior. Clearly, mutant strains exhibited a more effective adherence and internalization profile soon after 27 min of coculturing (Movie S1), while this process is more lagging for the wild-type macrophage interaction; after 30 min of interaction, most of the C. parapsilosis cells were outside of the macrophages, adherent or not ( Figure 6(b), i). Following 3 h of interaction, wild-type and both mutant strains were mostly internalized; notably, the number of ndt80ΔΔ mutant cells inside macrophages was statistically higher versus the two other cell types ( Figure 6(b), ii). Macrophage viability decreased along the assay (Figure 6(c), i and ii). Macrophage challenge with ndt80ΔΔ mutant cells, caused a significant reduction of the number of viable macrophages soon after 30 min (Figure 6(c), i). Following 3 h of coculture, an increase of lysed macrophages was observed with all the strains assessed; however, this result was statistically significant in the case of ndt80ΔΔ strain (Figure 6(c), ii).

Discussion
While molecular mechanisms are well characterized in C. albicans, several studies addressing the regulatory networks of non-albicans species, like C. glabrata and C. parapsilosis, demonstrate a significant difference in the evolutionary adaptation of such yeasts to the human host [19,39]. Although the available knowledge regarding the expression of C. parapsilosis virulence attributes is still somewhat limited, this species displays many biological features that are presumed to be directly related to its environmental colonization and pathogenicity, such as enhanced adherence and biofilm development on abiotic surfaces. Adhesion, morphogenetic variations and biofilm formation are virulence attributes clearly depicted for C. albicans [40,41] and are intimately related to each other. Filamentous growth is closely related to the expression of surface proteins, such as Als1, Als3 and the hyphal-specific protein, Hwp1. In turn, these proteins play relevant roles in cell-cell and cell-surface adhesion and are required for biofilm formation as contact mediators that promote further biomass accumulation and enhance biofilm resilience [14,15]. Ndt80 was identified as one of the many regulators of filamentous growth by binding to promoters of genes encoding cell wall components (e.g. ALS3 and HWP1), being required for their normal expression [42]. Thus, deletion of NDT80 reduces C. albicans virulence in vivo, by blocking yeast to hyphal transition, as well as the expression of genes involved in the filamentous transcriptional program [42].
Surprisingly, and opposing to what was described for C. albicans, the disruption of C. parapsilosis NDT80 gene triggers two noticeable phenotypic changes: morphogenesis in a spontaneous and constitutive manner (Figure 2), and prompted adhesion, both cell to cell and to abiotic surfaces, but also to murine macrophages ( Figure 3 and Figure 6, respectively). Despite the scarce knowledge on C. parapsilosis adhesion mediators, we demonstrate that ndt80 mutants adhesion is conferred by ALS7 (CPAR2_404800), whose expression is extraordinary increased. This adhesin was previously identified as a mediator of C. parapsilosis adhesion to human buccal epithelial cells [31]. Although only 0.5% of the ORFs related with cell adhesion contain putative recognition sites for Ndt80, ALS7 and ALS3 are included in this group.
According to our findings Ndt80 can have a dual role in yeast to pseudohyphae transition: on one hand, by impairing the expression of UME6 and CPH2, described as inducers of yeast to pseudohyphae transition [19]; on the other hand, by acting as an  activator of Czf1 and Efg1 [19,35], two known transcription factors regulating phenotypic switching and filamentous growth in C. albicans. Other genes like OCH1, the orthologs of C. albicans CWH41 and STP3 are also involved in C. parapsilosis phenotypic switching, as positive and negative regulators, respectively [32,36]. We found that Ndt80 has no impact upon the expression of OCH1 and the ortholog of C. albicans STP3; interestingly, the ortholog C. albicans CWH41 expression doubles in ndt80ΔΔ mutant, suggesting that this gene could be a target for Ndt80, which putatively represses the expression of this pseudohyphae formation factor.
Ndt80 is also part of a network of six transcription factors (Bcr1, Efg1, Tec1, Rob1, Bgr1, and Ndt80) responsible for the regulation of C. albicans biofilm development [22]. In this species, NDT80 deletion significantly compromises biofilm formation either in vitro or in vivo models [22]. Conversely, we found that deletion of C. parapsilosis NDT80 gene promotes biofilm growth in vitro, suggesting that this transcription factor is acting as a repressor of genes involved in such process. Other biofilm regulators, acting as repressors and activators in a circuit system were already previously identified in C. albicans and C. parapsilosis [19]. Efg1, Bcr1, and Ace2 play similar roles regarding biofilm development in both species, while Cph2, Czf1, Gzf3, and Ume6 have major roles just in C. parapsilosis [19]. In C. parapsilosis, deletion of CZF1, GZF3, UME6, and CPH2 was associated with a reduced biofilm formation ability. Although Ndt80 was not identified as a component of C. parapsilosis regulatory network due to the inherent growth defects [19], we analyzed the promoter sequences of all the biofilm transcription factors described by Holland et al. [19] for the presence of Ndt80 MSE motifs and identified putative recognition sites in all of the genes tested. The gene expression profile analysis of ndt80ΔΔ mutant revealed an approximately 36-fold, 13-fold, 4-fold and 3-fold upregulation of MKC1, UME6, CPH2 and ACE2, respectively, while other genes also described to be required for biofilm formation, such as GZF3 and CZF1, were demonstrated to be downregulated. These findings strongly suggest the role of Ndt80 as a negative regulator of MKC1, UME6, CPH2 and ACE2 expression and as an activator of GZF3 and CZF1 expression. Thus, in Ndt80 absence, and despite GZF3 and CZF1 genes exhibiting a reduced expression, the upregulation of MKC1, UME6, CPH2 and ACE2 genes occurs and biofilm development is promoted (Figure 4). RHR2 was also considered to be involved in biofilm development by C. parapsilosis, as its expression was increased during biofilm formation [19]. Nevertheless, in ndt80ΔΔ mutant characterized by enhanced biofilm production, RHR2 gene is downregulated probably denoting the lack of Ndt80 regulation as an activator.
The virulence-related phenotypes exhibited by ndt80ΔΔ mutant led us to explore its interaction with immune system cells. The ability to switch from yeast to a filamentous form is a key factor that allows successful phagocytosis evasion of C. albicans [43]. In the case of C. parapsilosis, several studies have elucidated distinct virulence traits of this species that could modulate the mechanism by which phagocytosis and the immune response proceed [44][45][46]. We found, in our in vitro infection assays a prompter interaction of both mutants with the macrophage cells in comparison to the wild-type strain. This finding is also in accordance with results obtained with the adhesion assays to abiotic surfaces and to other yeast cells.
Toth et al. [37] using other host cell models (J774.1 murine macrophage cell line and human peripheral blood mononuclear cells) described that the length of C. parapsilosis pseudohyphae did not correlate with the engulfment time. In our assays, after 3 h of coculturing, only the ndt80ΔΔ mutant induced a significantly increase of macrophage killing with concomitant higher yeast viability, while neither the wildtype nor the ndt80Δ mutant promoted significant damage of the macrophage cells. These results show that the phenotype prompted by NDT80 knockout results in a more virulent C. parapsilosis strains, more resistant to macrophage attack, associated with a decrease of macrophage cytoplasmic membrane integrity and a concomitant increase of macrophage cell death. Virulence features are not exclusively related to the constitutive pseudohyphal form; notably, the promoted expression of ALS7 and MKC1 transcripts (factors essential to cell wall integrity and remodeling) [47,48] provides a strong evidence of alterations of cell wall concerning composition and architecture in the ndt80ΔΔ mutant, with impact upon adhesion and recognition by immune system cells [49].
In fungi, NDT80-like genes recognize the conserved DNA-binding domain motif, MSE, through an Ig fold. As other members of the Ig-fold family of transcription factors, such as p53 or NFAR from mammals, NDT80like genes share a similar regulation mechanism [50]. However, the number and attributable functions of NDT80-like genes are divergent among fungal species and even within species [34]. These disparities range from NDT80 absence, as seen in Schizosaccharomyces pombe, to a family of six members, as seen in Fusarium oxysporum. While in Saccharomyces cerevisiae, NDT80 single gene functions as a master regulator of meiosis process and sporulation [51], in other fungal species possessing several paralogous of NDT80-like genes the unraveling of its function and regulation mechanism is laborious and far from being obtained. NdtA and XprG are two of the Ndt80-like proteins in the filamentous fungal species Aspergillus nidulans. The former has a high homology with Ndt80 and like in S. cerevisiae, it is crucial for sexual reproduction. The later, under carbon starvation, regulates positively fungal response by controlling its extracellular proteases, mycotoxin, and penicillin expression, which could result in autolysis, hyphal fragmentation and ultimately in cell death [52]. Neurospora crassa possesses three Ndt80-like proteins, Vib-1, Ncu04729 and Fsd-1. Vib1, closely related to XprG, is an activator of extracellular protease production and is also associated with apoptosis [53]; Fsd1 (more similar to NdtA) together with Vib-1, is involved in the female sexual structure formation, but no one is required for meiosis. So far, NCU04729 gene deletion has no effect upon phenotype, which impairs the understanding of its function. In the CTG clade, C. albicans has three NDT80like DNA-binding domain genes, NDT80, RON1 and REP1 [54]. These Ndt80-like transcription factors seem to be functionally independent from each other. Rep1 was found to be a regulator of the drug efflux pump MDR1 and is required for yeast growth on presence of N-acetylglucosamine (GlcNAc) and galactose. Ron1 is associated with GlcNAc regulation signaling.
Notably, Ndt80 was identified as a morphogenesis and biofilm regulator, in C. albicans and C. parapsilosis, although it diverged to opposite functional roles. Our study highlights the importance of Ndt80 on the complex regulation of C. parapsilosis virulence attributes, as a major repressor.