CpOCH1 was identified in the Candida genome database (http://www.candidagenome.org/) by homology to the C. albicans ortholog (Systematic name orf19.7391). The CpOCH1 open reading frame of 1089 bp (Systematic name CPAR2_404930) is predicted to encode a type-II transmembrane protein of 362 amino acids of the glycosyl transferase family 32, which shows 67 and 78% of identity and similarity to C. albicans Och1, respectively. This open reading frame is unlikely to encode the closely related α1,6-mannosyltransferase Hoc1, as it shows 39 and 57% of identity and similarity to C. albicans Hoc1 (Systematic name orf19.3445). The CpOCH1 alleles were deleted by sequential gene replacement in the CPL2H1 strain (Holland et al., 2014) as described for C. albicans, using disruption cassettes complementing leucine and histidine auxotrophies (Figure 1SA; Noble et al., 2010). Southern blotting assays were performed to discard ectopic integrations within the C. parapsilosis genome, using specific probes for CmLEU2, CdHIS1 and CpOCH1. Results showed loss of CpOCH1 gene and a single copy of each replacing cassette within the mutant strain genome (Figure 1SB), demonstrating the production of a C. parapsilosis och1Δ null mutant. As a control, the CpOCH1 open reading frame, under the control of the C. albicans TDH3 promoter was re-integrated into the C. parapsilosis och1Δ null mutant, generating a re-integrant control strain. In order to demonstrate that CpOCH1 is the functional ortholog of the CaOCH1, we complemented a C. albicans och1Δ null mutant (Bates et al., 2006) with CpOCH1. Results indicated that the C. parapsilosis gene was able to restore the levels of phosphomannosylation (Figure 1A) and the electrophoretic mobility of Hex1 (Figure 1B), a secreted protein previously used to assess the status of the N-linked glycosylation pathway (Bates et al., 2005, 2006; Mora-Montes et al., 2007; Lopes-Bezerra et al., 2015). Therefore, C. parapsilosis OCH1 is the functional ortholog of C. albicans OCH1.

The growth rate of the Cpoch1Δ null mutant was significantly reduced, with doubling times of 2.79 ± 0.22 h for the null mutant vs. 1.54 ± 0.11 and 1.57 ± 0.15 h for the wild-type (WT) and reintegrant control strains, respectively (P < 0.05). Experiments conducted in presence of 2 units/mL chitinase to disrupt cell aggregates (Bates et al., 2006) showed similar results (data not shown). The Cpoch1Δ null cells displayed a clumpy phenotype, i.e., they tended to form cell aggregates when cultured in liquid media (Figure 2B). Colony morphology was not significantly affected when cells were grown on Sabouraud plates at 20 and 28°C, or under pH conditions ranging from 4 to 8 (Figure 2A and data not shown). However, when strains were grown at 37°C, the null mutant failed to develop pseudohyphae, even when cultured in filamentation-inducing media, such as RPMI supplemented with 10% (v/v) fetal bovine serum, Lee or Spider media (Figure 2D, and data not shown). Under such growth conditions, the Cpoch1Δ null mutant displayed well-defined colony edges on the plates and no filamentous structures were visible by light microscopy (Figure 2C). The reintegrant control strain did not show the clumpy phenotype and effectively generated pseudohyphae, confirming that the phenotypes observed in the null mutant strain are indeed related to OCH1 disruption. Therefore, loss of OCH1 affects C. parapsilosis morphogenesis.

To assess the effect of CpOCH1 mutation on cell wall integrity, we tested the susceptibility of the null mutant to a range of cell wall perturbing agents and compounds associated with glycosylation defects. The Cpoch1Δ null mutant exhibited enhanced susceptibility to Calcofluor White and Congo Red (P = 0.04055 and 0.00124, respectively), which interact with cell wall chitin and β-glucans, respectively (Figure 3). Furthermore, the null mutant had an increase in susceptibility to Tunicamycin (P = 0.0278), an inhibitor of the first steps during N-linked mannan biosynthesis; and it was highly susceptible to SDS (P = 0.0074), a detergent that affects the plasma membrane (Bates et al., 2006; Mora-Montes et al., 2007); whereas the WT and control strains were largely resistant (Figure 3). Hygromycin B, vanadate, and osmotic stressors such as NaCl and KCl were also tested, but no significant differences were observed (data not shown). Similar results were generated when the experiments were performed in presence of chitinase to disaggregate cells (not shown).

We next assessed the cell wall composition of the Cpoch1Δ null mutant. Cell walls were purified, acid hydrolyzed, washed and analyzed by High-Performance Anion-Exchange Chromatography with Pulsed Amperometric Detection (HPAEC-PAD). Results indicated similar carbohydrate content among the control strains, but the null mutant displayed a reduction of about 67% in the mannan content, along with a significant increment in β-glucan and chitin levels, when compared to the control strains (Table 1). No significant differences in the cell wall protein level or the phosphomannan content were observed (Table 1). The relative cell wall porosity to polycations, an indirect reporter of the glycan arrangement (De Nobel et al., 1990; Cheng et al., 2011), was greater in the Cpoch1Δ null mutant than in the heterozygous or control strains (Table 1). Furthermore, we assessed whether rearrangements in the organization of cell wall components occurred upon disruption of CpOCH1. For this, we quantified the ability of the fluorescein isothiocyanate-wheat germ agglutinin conjugate (WGA-FITC) and IgG Fc-Dectin-1 chimera to bind chitin and β1,3-glucan, respectively (Graham et al., 2006; Mora-Montes et al., 2011; Marakalala et al., 2013). Results shown in Figure 4 indicate that both lectins displayed a weak ability to bind either chitin or β1,3-glucan at the cell wall of live WT, reconstituted WT, heterozygous mutant and re-integrant control cells. However, the live och1Δ null mutant showed an increased ability to bind both WGA-FITC and the IgG Fc-Dectin-1 chimera (Figure 4). C. albicans inactivation by heat exposes inner cell wall components at the cell surface, such as β1,3-glucan and chitin (Gow et al., 2007; Mora-Montes et al., 2011). Thus, as expected, enhanced binding by both lectins occurred upon inactivation of yeast cells by heating (Figure 4). Altogether, these results indicate that the Cpoch1Δ null mutant has significant defects in cell wall composition and fitness.

We next assessed the relevance of proper cell wall N-linked mannosylation during C. parapsilosis interaction with human PBMCs, quantifying the levels of pro- and anti-inflammatory cytokines as a read out of this interaction. Live C. parapsilosis yeast cells from WT control strains and the heterozygous mutant stimulated low and similar TNFα, IL-1β, IL-6, and IL-10 levels (Figure 5). As mentioned, C. albicans inactivation by heat exposes β1,3-glucan at the cell surface, and heat-treated C. albicans cells are therefore more immunostimulatory compared to untreated cells (Gow et al., 2007). Here, upon heat inactivation of WT C. parapsilosis cells, the interaction with PBMCs resulted in stimulation of TNFα, IL-6, and IL-10, but not IL-1β (Figure 5). Live C. parapsilosis och1Δ null mutant stimulated significantly higher levels of both pro- and anti-inflammatory cytokines compared to WT and control cells (Figure 5), and, accordingly, WT and control cells treated with endo-H, i.e., with N-linked mannan enzymatically removed from the cell wall (Kobata, 1979), induced cytokine levels similar to that with the och1Δ null mutant cells (Figure 5). The heat-killed (HK) och1Δ null mutant and the endo-H-treated WT cells stimulated significantly less TNFα and IL-6 production than the heterozygous mutant, the reintegrant control and the WT control cells (Figure 5), suggesting that recognition of both N-linked mannans and inner cell wall components is required for maximal cytokine stimulation. Production of IL-1β and IL-10 was insensitive to heat killing in both the och1Δ null mutant and the endo-H-treated WT cells. Therefore, proper N-linked mannosylation is required for stimulation of cytokines by human PBMCs.

Next, to obtain insights into the role of O-linked mannans as PAMPs during C. parapsilosis sensing, we trimmed this oligosaccharide from the cell wall by β-elimination (Diaz-Jimenez et al., 2012), and used these cells to stimulate cytokine production by human PBMCs. After the removal of O-linked mannans, cell viability was not significantly altered (not shown), about 50% of mannan content was removed from the cell wall of WT and control cells, and no mannose was recorded in the cell wall preparations from och1Δ null mutant cells (Table 1). Furthermore, C. albicans β-eliminated cells stimulated similar cytokine levels than those stimulated by the C. albicans mnt1-mnt2Δ null mutant (Munro et al., 2005) that lacks long O-linked mannans from the cell wall (Figure 2S). Live WT control, heterozygous, re-integrant strain and och1Δ null mutant cells stimulated similar TNFα and IL-6 levels in comparison to the β-eliminated-treated cells (Figure 5). HK WT controls, re-integrant, and OCH1/och1Δ heterozygous cells, but not och1Δ null mutant cells, induced lower TNFα and IL-6 levels, suggesting that O-linked mannans significantly contribute to C. parapsilosis sensing by human PBMCs. Upon O-linked mannan trimming, HK och1Δ cells lost the ability to stimulate high IL-1β levels, suggesting a key role for this cell wall component in IL-1β stimulation. As in the case of N-linked mannans, loss of O-linked mannans did not affect production of IL-10 (Figure 5). Overall, these data indicate key roles of both N- and O-linked mannans during the interaction of C. parapsilosis with human PBMCs.

Next, we analyzed the pathogen recognition receptors involved in the stimulation of IL-1β. The production of this cytokine was dependent on proper engagement of either dectin-1 or TLR4, as fungal cells lacking N-linked mannans stimulated reduced levels of IL-1β production upon blocking with laminarin and antibodies to TLR4, respectively (Figure 6). No effect was observed in the IL-1β levels when TLR2 was blocked with an antibody to TLR2, indicating this receptor does not affect stimulation of this cytokine (Figure 3S). As a control assay for this blocking agent, we found that this antibody did not affect the levels of IL-10 (Figure 4S), but significantly reduced TNFα production by WT cells harboring N-linked mannans at the cell wall (Figure 3S). Control experiments using isotype-matched, irrelevant antibodies did not show significant variations in cytokine stimulation by C. parapsilosis cells (Figure 6, Figure 3S). Overall, these data indicate that N-linked mannans are able to inhibit IL-1β production whereas stimulation of this cytokine requires presence of O-linked mannans, dectin-1 and TLR4.

To assess the relevance of C. parapsilosis cell wall mannans during phagocytosis, primary human PBMC-derived macrophages were co-cultured with fluorescently labeled WT or C. parapsilosis och1Δ for 1.5 h and the ratio of phagocytosing macrophages was determined by flow cytometry. Interestingly, we found that C. parapsilosis och1Δ null mutant cells were as efficiently phagocytosed by the PBMC-derived macrophages as WT cells, indicating that N-linked mannans are not essential cell wall components for C. parapsilosis phagocytosis (Figure 5S). Furthermore, we similarly did not find a difference in the engulfment of C. parapsilosis WT or och1Δ null mutant cells by J774 mouse macrophages (data not shown).

To determine how the disruption of the protein N-linked mannosylation affects the virulence of C. parapsilosis, we compared the susceptibility of BALB/c mice to C. parapsilosis WT and the och1Δ null mutant in a non-lethal experimental model of disseminated candidiasis, as previously reported (Ifrim et al., 2014). We found that mice infected with the och1Δ null mutant had significantly decreased fungal burdens in the spleen, kidneys and liver at 1, 2, and 7 days post-infection, when compared to those mice infected with the WT strain (Figure 7). Animals infected with the reintegrant control had fungal burdens similar to mice challenged with WT yeast cells (Figure 7). Therefore, these results demonstrate that the loss of OCH1 significantly affects the virulence of C. parapsilosis in vivo.

We also assessed the virulence of the mutant strains generated in this work in the Galleria mellonella model of disseminated candidiasis (Gago et al., 2014; Jacobsen, 2014). Interestingly, larvae inoculated with either the WT control cells or the Cpoch1Δ null mutant displayed similar mortality rates, with most of the larvae dying after 10 days post-inoculation (Figure 5S). As a control, we included the C. albicans och1Δ null mutant that was attenuated in a mouse model of systemic candidiasis (Bates et al., 2006), and this strain also displayed virulence attenuation in the G. mellonella infection system (Figure 6S). Therefore, these results indicate that the virulence of the C. parapsilosis och1Δ null mutant significantly depends on the host milieu.

The strains utilized in the experiments are listed in Table 2. Unless otherwise indicated, cells were maintained and propagated at 30°C in Sabouraud medium [1% (w/v) mycological peptone, 4% (w/v) glucose]. Two % (w/v) agar was included when solid medium was required. In order to prepare cells for cytokine assays and cell wall analyses, 500 μL from an overnight culture were transferred to 500 mL flasks containing 100 mL of fresh medium and incubated at 30°C with constant shaking at 200 rpm, until mid-log growth phase was reached. Filamentation was induced by growing 5 × 10⁶ cells/mL in RPMI 1640 medium (Sigma) supplemented with 10% fetal bovine serum (Sigma) or Lee's medium (all reagents from Sigma; Lee et al., 1975), at 37°C for 4 h. Cells were heat inactivated by incubating at 56°C for 60 min (Mora-Montes et al., 2007). Loss of cells viability was confirmed by absence of fungal growth in Sabouraud broth at 30°C for 72 h. β-Elimination was achieved by incubating cells overnight with 100 mM NaOH at room temperature and mild shaking (Diaz-Jimenez et al., 2012). Cells maintained viability upon β-elimination, since they showed no significant differences in CFU/mL before and after this treatment (average loss of cell viability upon β-elimination was 5.0 ± 2.0%). Trimming of N-linked glycans was accomplished by treating cells with endo-H (New England BioLabs) as described (Mora-Montes et al., 2012). The release of cell wall mannans was assessed by quantifying the free mannan content upon either β-elimination or treatment with endo-H, using HPAEC-PAD, in an ion chromatograph Dionex ICS-3000 (USA) with a guard-column CarboPac PA-100 (4 × 50 mm) and analytical CarboPac-PA100 (4 × 250 mm) column as reported (Mora-Montes et al., 2012). In some experiments, the medium was added with 2 units/mL chitinase (Sigma) to disperse cell aggregates as reported (Bates et al., 2006).

The gene disruption strategy performed here consisted in the use of a double auxotrophy (His⁻ Leu⁻) system in the CLIB-214 isolate of C. parapsilosis (CPL2H1 strain) (Holland et al., 2014), as described for C. albicans (Noble et al., 2010). Briefly, disruption cassettes containing either C. dubliniensis HIS1 or C. maltosa LEU2 genes flanked by ~500 nucleotides matching sequences upstream and downstream of the C. parapsilosis OCH1 gene were generated by fusion PCR. The markers C. dubliniensis HIS1 or C. maltosa LEU2 were amplified by PCR from pSN52 or pSN40 (Noble and Johnson, 2005), respectively, using the same primers (primer pair: 5′-CCGCTGCTAGGCGCGCCGTGACCAGTGTGATGGA TATCTGC-3′and 5′-GCAGGGATGCGGCCGCTGACT CCGCTTAAACAATCGGCAAAGCTCGGATCCACTAGTAAC G-3′, underlined sequences align with the 5′ and 3′ OCH1 fragments, respectively). The C. parapsilosis OCH1 5′ and 3′ regions of homology were amplified by PCR (primer pair: 5′-CAACATTTACATTCTTTTGC-3′ and 5′-CAC GGCGCGCCTAGCAGCGGATCAACGTATAAGCACTGC C-3′; and primer pair: 5′-GTCAGCGGCCGCATCCCTGCTGCC TGAAATGCTTACATAG-3′ and 5′-AACATCTCAAAACCGC AAGA-3′; homologous sequences to C. dubliniensis HIS1 or C. maltosa LEU2 markers are underlined, respectively). The heterozygous strain lacking one OCH1 allele was constructed by chemical transformation of C. parapsilosis CPL2H1 with a LEU2-marked disruption cassette. Transformants were selected in minimal SC medium [0.67% (w/v) yeast nitrogen base, 2% (w/v) dextrose, 2% (w/v) agar] supplemented with L-histidine (1 mg/mL), and quick DNA extraction (Looke et al., 2011) was performed to screen Leu⁺ transformants by PCR, looking for the presence of expected 5′ and 3′ junctions of the integrated DNA. Homozygous och1Δ null mutant was constructed by transforming the heterozygous knockout strain with a HIS1-containing disruption cassette. Transformants were selected in minimal medium and quick DNA extractions (Looke et al., 2011) were performed to screen His⁺Leu⁺ mutants by PCR. Furthermore, to confirm the PCR results, genomic DNA samples were analyzed by means of southern blotting. Briefly, samples of genomic DNA from the null mutant, and control strains were digested with RI, run onto agarose gels, and transferred to nylon membranes. DNA samples were hybridized with specific probes targeting either CmLEU2, CdHIS1, or CpOCH1 genes. These probes were amplified using Digoxigenin-marked dNTPs (Invitrogen) whose signal was revealed by enzymatic means (Roche). To reintegrate OCH1 in the null mutant, the Invitrogen Gateway® cloning strategy was used. First the OCH1 orf was amplified from the genomic DNA of C. parapsilosis CLIB 214. The 1089bp PCR product was purified using the PEG/MgCl₂ method according to manufacturer's instruction from Invitrogen gateway cloning kit. Next, the BP clonase was used to integrate the OCH1 ORF into the pDONR221 vector, and then subcloned into a modified destination vector, which contained a constitutive promoter, pTDH3, and a permanent URA3 terminator sequence. Nourseothricin (NAT) was used as dominant selection marker. To achieve successful integration into the C. parapsilosis RP10 locus, the destination vector was further modified by replacing the C. albicans RP10 region to an artifical StuI restriction site containing C. parapsilosis RP10 locus. The expression vector was then digested with StuI, and the linearized plasmid was transformed in the och1Δ/och1Δ null mutant by chemical transformation. Integration of the cassette in the CpRP10 locus was checked by colony PCR and Southern blotting analysis. pDONR221 and Clp10-CaTDH3-GTW-URA3 plasmid vectors were kindly provided by Prof. Christophe d'Enfert.

To confirm that CpOCH1 is the functional ortholog of CaOCH1, we performed the complementation of a Caoch1Δ null mutant (Bates et al., 2006). The CpOCH1 open reading frame, along with 1 Kbp upstream and 582 bp downstream sequences, was amplified by PCR (primer pair 5′- GCGGCCGCAGGCTAATCAAGAGGTTCCTG-3′ and 5′- GCGGCCGCTTCATAGCGAGTGAGAGAC-3′, with the bases to generate a NotI site underlined), cloned into pCR®2.1-TOPO® (Invitrogen), and subcloned into the NotI site of CIp10 (Murad et al., 2000). The resulting plasmid was linearized with StuI before transformation of the Caoch1Δ null mutant.

Cells in mid-log phase were collected by centrifugation, washed twice with deionized water, and mechanically broken, under a CO₂ stream, in a Braun homogenizer during 5 min. Cell disruption was performed in cycles of 1 min, with 2-min resting periods on ice. Then, the homogenate was centrifuged for 10 min at 13 206 × g, 4°C and the supernatant was recovered. The samples were loaded onto a 4% PAGE gel and run for 11 h at 40 V under native conditions. The N-acetylhexosaminidase activity was determined by incubating with 0.4 mM 4-methylumbelliferyl N-acetyl-β-D-glucosamine (Sigma) in 0.1 M citrate-KOH buffer (pH 4.5) during 30 min at 37°C, and the results were observed by exposing the gel to UV light (Hernandez-Cervantes et al., 2012).

Cells were mechanically broken as above described, the homogenate was centrifuged, the pellet recovered and extensively washed with deionized water for debris elimination. Cell walls were lyophilized and acid-hydrolyzed as described (Mora-Montes et al., 2007). Acid-hydrolyzed samples were analyzed by HPAEC-PAD and the carbohydrate separation was achieved with a gradient of sodium acetate in 150 mM NaOH (flow 0.5 mL/min) as follows: 0–5 min = 45–75 mM NaOH, 5.1–15.0 min = 90 mM NaOH, 15.1–17.0 min = 105 mM NaOH + 75 mM sodium acetate, 17.1–20.0 min = 75 mM NaOH + 150 mM sodium acetate, and 20.1–25.0 min = 45 mM NaOH, at a column temperature of 25°C. Applied potentials, for detection by the amperometric pulse were: E1 (400 ms), E2 (20 ms), E3 (20 ms), and E4 (60 ms) of + 0.1, −2.0, + 0.6, and −0.1 V, respectively. For quantification of cell wall protein content, lyophilized cell walls were alkali-hydrolyzed as reported (Mora-Montes et al., 2007), and analyzed using the Bradford protein assay.

Cell wall porosity was determined by relative porosity to polycations as described (De Nobel et al., 1990). Briefly, overnight-grown cells were inoculated into fresh Sabouraud broth, incubated for 4 h at 30°C and 200 rpm, and washed twice with PBS. Cell pellets containing 1 × 10⁸ cells were suspended in either 10 mM Tris-HCl, pH 7.4 (buffer A), buffer A plus 30 μg/mL poly-L-lysine (MW 30–70 kDa, Sigma Cat. No. P-2636) or buffer A plus 30 μg/mL DEAE-dextran (MW 500 kDa, Sigma Cat. No. D-9885), and incubated for 30 min at 30°C with constant shaking at 200 rpm. Preparations were centrifuged to pellet cells, supernatants recovered and further centrifuged before absorbance at 260 nm was measured. The relative cell wall porosity to DEAE-dextran was calculated as described (De Nobel et al., 1990).

Cells grown at mid-log phase were pelleted, washed twice with deionized water and adjusted at an OD₆₀₀ of 0.2 in deionized water. Aliquots of 1 mL were pelleted and cells suspended in 1 mL of Alcian blue (Sigma; 30 μ g/mL, in 0.02 M HCl) and assayed as described (Hobson et al., 2004).

Strains were tested for susceptibility to cell wall perturbing agents using a microdilution method as described (Bates et al., 2005). Briefly, cells from overnight culture in medium containing 2 units/mL chitinase were washed with water, passaged through a syringe with a 32-gauge needle, and suspended at an OD₆₀₀ = 1. These cells were then inoculated into fresh medium at an OD₆₀₀ of 0.01 and 95 μL of this suspension were dispensed into 96-well plates. A 5 μL volume of the cell wall perturbing agents was added to each well by duplicate across a range of double dilutions. The final OD₆₀₀ was determined after 16 h of incubation at 30°C. The maximum concentrations tested for each agent were Calcofluor White (Sigma), Congo Red (Sigma) and tunicamycin (Sigma, 100 μ g/mL, each), SDS (BioRad, 0.1%, w/v), hygromycin B (Sigma, 500 μ g/mL), NaCl₂ and KCl (1M, each), caffeine (Sigma; 50 mM) and vanadate (Sigma, 80 mM). Cells were grown in medium with no stressor added to normalize data. Growth data were normalized as percentage of those generated with the same strains without treatment.

The assay was prepared as previously described (Mora-Montes et al., 2011). For chitin staining, cells were stained with 1 mg/mL WGA-FITC (Sigma). For β1,3-glucan staining cells were incubated with 5 μg/mL IgG Fc-Dectin-1 chimera (Graham et al., 2006) for 40 min at room temperature, followed by incubation with 1 μg/mL donkey anti Fc IgG-FITC for 40 min at room temperature (Marakalala et al., 2013). Samples were examined by fluorescence microscopy using a Zeiss Axioscope-40 microscope and an Axiocam MRc camera. From the pictures acquired, the fluorescence quantification of 50 cells was achieved using Adobe Photoshop™ CS6 with the next formula: [(total of green pixels-background green pixels) × 100]/total pixels.

The Ethics Committee from Universidad de Guanajuato approved the use of human cells in this study (permission number 17082011); while University of Szeged granted permission XII./00455/2011 to work with mice. Human cells were collected from healthy adult volunteers after information about the study was provided and written informed consent was obtained.

Human PBMCs were isolated by density centrifugation using Histopaque-1077 (Sigma) as described (Endres et al., 1988). For stimulation experiments, the interactions were performed in round-bottom 96-well microplates with 100 μL of cells adjusted to 5 × 10⁵ PBMCs in RPMI 1640 Dutch modification (added with 2 mM glutamine, 0.1 mM pyruvate and 0.05 mg/mL gentamycin; all reagents from Sigma) and 100 μL with 1 × 10⁵ fungal cells freshly harvested or treated. The interactions were incubated for 24 h at 37°C with 5% (v/v) CO₂. In some experiments, PBMCs were preincubated for 1 h at 37°C with either laminarin (200 μg/mL), anti-TLR2 (10 μg/mL, eBioscience, Cat. No. 16-9922) or antibodies to TLR4 (10 μg/mL, Santa Cruz Biotechnology, Cat. No. sc-293072) prior to stimulation with Candida cells. Isotype matched, irrelevant antibodies, IgG₂aκ (10 μg/mL, eBioscience, Cat. No. 14-4724-85) and IgG₁(10 μg/mL, Santa Cruz Biotechnology, Cat. No. sc-52003) were used as controls for experiments assessing TLR2 and TLR4, respectively. All reagents used for the pre-incubation experiments were negative to contamination with LPS (tested with the Limulus amebocyte lysate from Sigma), and all reactions were performed in presence of 5 μ g/mL polymyxin B (Sigma) (Schwartz et al., 1972). Plates were centrifuged for 10 min at 3000 × g at 4°C, the supernatant saved and kept at −20°C until used.

The concentration of TNFα, IL-6, and IL-10 was quantified by ELISA (Peprotech), according to the manufacturer's instructions. The IL-1β levels were measured using a commercial ELISA kit from R&D Systems.

After isolation, aliquots of 1 mL containing 2 × 10⁷ PBMCs in RPMI supplemented with 1% (v/v) penicillin-streptomycin solution (PS, Sigma-Aldrich) were placed in flat bottom 12-well plates, and incubated 1.5 h at 37°C, 5% (v/v) CO₂. Non-adherent cells were removed gently with the medium and adherent cells were washed with PBS at 37°C. One mL of X-VIVO 15 serum-free medium (Lonza) supplemented with 1% (v/v) PS and 10 ng/mL recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF, Sigma) were added into each well and incubated for 6–7 days at 37°C, 5% (v/v) CO₂, with fresh medium exchanged every 2–3 days (Netea et al., 2009).

Phagocytosis of C. parapsilsosis cells was performed as previously described (Nemeth et al., 2013). Briefly, C. parapsilosis cells were labeled with AlexaFlour488 Succinimidyl ester (Life Technologies) and subsequently co-incubated with human PBMC-derived macrophages at an effector:target ratio of 1:5 for 1.5 h. Non-phagocytosed yeast cells were removed by gentle washing and macrophages were detached from cell culture plates by using TrypLE™ Express solution (Gibco). To inhibit phagocytosis, macrophages were pre-incubated for 30 min with 2.5 μM cytochalasin D (R&D Systems) before infection. Samples were measured on a FACSCalibur instrument and analyzed by FlowJo vX.0.7 software.

A non-lethal experimental model of disseminated candidiasis was used as reported previously (Ifrim et al., 2014). Briefly, groups containing 8–12-weeks-old male Balb/c WT mice (22–27 g. of weight) were injected via the lateral tail vein with 2 × 10⁷ C. parapsilosis cells, previously passaged through a syringe with a 32-gauge needle, in 100 μL of sterile PBS. A group of control mice was injected with 100 μL of sterile PBS. Animals were mantained with sterile water and pet aliment ad libitum. After 1, 2, or 7 days post infection, animals were humanitarianly euthanatized, and liver, kidneys and spleen were aseptically removed, weighed, and homogenized in sterile PBS in a tissue grinder. The fungal burden in these tissues was determined by plating serial dilutions on three YPD agar plates per tissue. The CFU were counted after 48 h of incubation at 30°C and expressed as CFU/g tissue.

Wax moth larvae killing assays were performed as described (Mylonakis et al., 2005). Briefly, 10 μL of a cell suspension prepared in PBS and containing 2 × 10⁷ yeast cells, passaged through a syringe with a 32-gauge needle, were injected directly into the haemocele, through last left pro-leg of the larva, using a 26-gauge needle and a Hamilton syringe. Larvae were incubated at 25°C after injection and survival monitored daily. Groups of 10 larvae were used for each strain analyzed. A group of untreated larvae and a PBS-injected group were included in each experiment as controls.

Statistical analysis was performed using GraphPad Prism 6 software. Results obtained upon incubation with the cell wall perturbing agents were analyzed by two-way ANOVA. Cytokine stimulation using human PMBCs was performed in duplicate with six healthy donors, whereas the rest of the experiments were performed at least thrice in duplicate. Data represent cumulative results of all experiments performed. The Mann-Whitney U test or unpaired t-test was used to establish statistical significance (see figure legends for details), with a significance level set at P < 0.05.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.