Secretion of serine proteases by planktonic- and biofilm-growing cells of Candida parapsilosis

Abstract

Candida parapsilosis is a common cause of candidiasis worldwide, with biofilm formation and secretion of aspartic proteases (Saps) as key virulence factors. Conversely, serine protease secretion by this fungus is poorly understood. In this study, we investigated the secretion of serine-type proteases by planktonic- and biofilm-forming cells of C. parapsilosis cultured in brain heart infusion (BHI) medium. Cell-free supernatant from the reference strain (ATCC 22019) was screened against various serine protease substrates, revealing pronounced activity toward N-benzoyl-Phe-Val-Arg-pNa (0.74 nmol pNA.mg^− 1.min^− 1), with optimal activity at pH 9.0 and temperatures between 32 °C and 40 °C. Proteolytic activity was significantly reduced by serine protease inhibitors PMSF (32.8%), TLCK (40.2%) and benzamidine (50.7%), while inhibitors of other protease classes had no effect, confirming its serine-type specificity. Notably, serine protease activity was detected in supernatants from cells grown in BHI but absent in those cultured in albumin-supplemented yeast carbon base medium, a known inducer of Saps, suggesting culture-dependent regulation of protease expression. Serine protease activity also increased over time, rising from 0.36 pNA.mg^− 1.min^− 1 at 24-hour to 1.14 pNA.mg^− 1.min^− 1 at 72-hour. Clinical isolates of C. parapsilosis exhibited significantly higher serine protease activity than the reference strain under optimal conditions. Serine-type protease activity was also detected in the supernatant of mature biofilms, showing a correlation with metabolic activity and biomass. Infection of Galleria mellonella larvae with C. parapsilosis isolates revealed no correlation between larval mortality and serine protease production. These findings suggest that C. parapsilosis serine proteases contribute to fungal growth and biofilm development, representing potential targets for antifungal intervention.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-025-01765-z.

Introduction

Fungal infections caused by emergent species such as Candida parapsilosis sensu stricto (C. parapsilosis) have significantly increased over the last decades [1, 2]. C. parapsilosis has a unique capacity to grow in hyperalimentation solutions, adhere to intravascular medical devices and form biofilms, making its infections particularly associated with the presence of catheters, central venous lines and parenteral nutrition systems [3–6]. Due to these characteristics, C. parapsilosis is frequently linked to healthcare-associated infections. Although some reports indicate that C. parapsilosis can surpass C. albicans in the number of infections [7–9], it generally ranks as the second or third most common species isolated in cases of invasive candidiasis across regions such as Latin America, North America, Europe and Asia [6]. To establish an infection in the host, C. parapsilosis deploys a variety of virulence attributes, including adhesins, biofilm formation and secretion of hydrolytic enzymes, which collectively enhance its pathogenicity and enable it to evade host immune defenses [10, 11].

Secreted aspartic proteases (Saps) are considered among the primary key virulence factors produced by C. parapsilosis, playing a pivotal role in its interactions with host structures and cells. These enzymes are crucial for processes such as adhesion, colonization, tissue invasion, dissemination and immune evasion [12]. However, other classes of proteases may also contribute significantly to these pathogenic mechanisms. Notably, serine-type proteases have been identified in C. albicans, C. glabrata, C. tropicalis, C. dubliniensis, C. guilliermondii and C. haemulonii species complex, and they are implicated in facilitating adhesion to and degradation of host tissues and extracellular matrix components, degrading complement system factors and immunoglobulins, breaching epidermal and epithelial barriers, and targeting immune cells such as lymphocytes and macrophages [13–15].

Although a serine protease with keratinolytic activity has previously been identified in C. parapsilosis [13], the regulation, dynamics of secretion and functional roles of serine-type proteases in this species remain poorly understood. To bridge this gap, the present study aimed to investigate the presence of secreted serine-type proteases in C. parapsilosis, with particular emphasis on their production by both planktonic and biofilm-associated cells. In parallel, an in vivo infection assay was performed using Galleria mellonella larvae as a model to investigate the potential correlation between serine protease production and fungal virulence in a living host. This model system offers a valuable platform to assess pathogenicity and host-pathogen interactions in a physiologically relevant context.

Materials and methods

Fungi and growth conditions

The reference American Type Culture Collection (ATCC 22019) strain and nine clinical isolates of C. parapsilosis [16] (kindly provided by Dr. Marcos Dornelas-Ribeiro, Instituto de Biologia do Exército, RJ, Brazil), were employed in the present study. The clinical isolates were biochemically/molecularly identified in a previous study by our group as C. parapsilosis sensu stricto (C. parapsilosis) [16] and designated as 222 (isolate recovered from toe), 225 (from toe), 229 (from a finger), 234 (unknown origin), 235 (from finger), 241 (from ear), 251 (from gluteus), 275 (form finger) and 454 (unknown origin). The fungal cells were cultured into brain heart infusion (BHI) medium for 48 h at 37 °C under constant agitation (200 rpm), a condition that induces the secretion of serine-type proteases in Candida spp [13].

Obtainment of cell-free culture supernatant

The 48-h fungal cultures were harvested by centrifugation at 4000×g for 10 min at 4 °C, and the spun culture medium was filtered through a 0.22-µm membrane (Millipore, SP, Brazil). The cell-free culture supernatants were concentrated 100-fold using a 10 kDa molecular weight cut-off AMICON micropartition system (AMICON^®, Beverly, MA, USA) and stored at − 20 °C. Protein levels were quantified using the method proposed by Lowry and colleagues [17].

Cleavage of peptide substrates by cell-free supernatant

Protease activity was measured by using different chromogenic substrates of the peptidyl-p-nitroaniline (pNA) type, namely: N-Suc-Ala-Ala-Ala-pNA, a substrate for elastase; N-Suc-Gly-Gly-Phe-pNA and N-Suc-Ala-Ala-Pro-Phe-pNA, substrates for chymotrypsin; Nα-benzoyl-DL-4Arg and N-benzoyl-Phe-Val-Arg-pNA, substrates for trypsin. The substrates were diluted in dimethylsulfoxide (DMSO) at 40 mM and stored at – 20 °C. The reaction was conducted in a total volume of 200 µL, in a 96-well-plate containing the reaction solution (50 mM glycine-NaOH buffer, pH 9.0), 200 µM of the substrate, and cell-free culture supernatant (500 µg protein) from C. parapsilosis (ATCC 22019). The reaction mixture was incubated at 37 °C for 60 min and hydrolysis was measured at 405 nm using a spectrophotometer. Characterization of cell-free supernatant proteolytic activity was carried out using the substrate N-benzoyl-Phe-Val-Arg-pNA as described, under different conditions of pH using the following buffer solutions: 50 mM sodium citrate (pH 3.0–5.0), 50 mM sodium phosphate (pH 6.0–8.0) or 50 mM glycine-NaOH (pH 9.0–12.0), as well as under different temperatures (28, 32, 37 and 40 °C) using 50 mM glycine-NaOH (pH 9.0) as standard buffer. Control wells containing only the respective buffers at each pH value were included to monitor abiotic substrate degradation, and the background signal from these controls was subtracted from the experimental readings. Finally, cell-free supernatant and N-benzoyl-Phe-Val-Arg-pNA were incubated in the presence of phenylmethylsulfonyl fluoride (PMSF, 10 mM), benzamidine (10 mM), Nα-tosil-l-lysine chloromethyl ketone hydrochloride (TLCK, 10 mM), pepstatin A (PepA, 10 µM), trans-epoxysuccinyl-l-leucyl-amido(4-guanidino butane) (E-64, 10 µM) and 1,10-phenanthroline (phen, 10 mM) in 50 mM glycine-NaOH buffer, pH 9.0, as previously described.

Modulation of protease production by distinct culture media

Cell-free supernatants were obtained from cultures of C. parapsilosis (ATCC 22019) grown in BHI or yeast carbon base (YCB) medium supplemented with 0.1% bovine serum albumin (BSA), a condition known to induce Sap production, and incubated (500 µg of protein) with the chromogenic substrate N-Benzoyl-Phe-Val-Arg-pNA in 50 mM glycine-NaOH buffer, pH 9.0, as previously described, to measure serine protease activity. The supernatants were also incubated with the fluorogenic peptide substrate for cathepsin D (7-methoxycoumarin-4-acetyl-Gly-Lys-Pro-Ile-Leu-Phe-Phe-Arg-Leu-Lys (DNP)-D-Arg-amide) (Sigma Chemical Co., St. Louis, USA) to measure aspartic protease activity. For this assay, the reaction was started by adding the substrate (2 µM) to each well of a 96-well plate containing the supernatant and citrate buffer, pH 4.0, in a total volume of 60 µL. The reaction mixture was incubated at 37 °C for 60 min and substrate cleavage was evaluated with a spectrofluorometer (SpectraMax Gemini XPS, Molecular Devices, CA, USA) using an excitation wavelength of 340 nm and an emission wavelength of 490 nm. The assays were controlled for self-liberation of the fluorophore over the same time interval.

Enzymatic activity of planktonic growth supernatant

C. parapsilosis (ATCC 22019) was inoculated (10⁶ cells/mL) into a Falcon tube containing BHI broth and incubated at 37 °C under agitation. A sample of 1 mL of the culture was collected at different time points (24 h, 48 h and 72 h), and after centrifugation, the supernatant was collected. In comparison, the cells were resuspended in 1 mL of phosphate-buffered saline (PBS, 7.2), diluted, and counted using a Neubauer chamber to establish a growth curve. Protein levels in the supernatant were measured and a sample of 500 µg of protein was incubated with the substrate N-Benzoyl-Phe-Val-Arg-pNA in 50 mM glycine-NaOH buffer, pH 9.0, using the conditions previously described.

Enzymatic activity of different clinical isolates

Nine clinical isolates of C. parapsilosis and the ATCC 22019 strain were grown in BHI broth for 48 h at 37 °C under constant agitation (200 rpm), the cell-free supernatant was obtained and the protein concentration was measured. The non-concentrated supernatant containing 100 µg of proteins adjusted to 50 µL was incubated with the substrate N-Benzoyl-Phe-Val-Arg-pNA (200 µM) in 50 mM glycine-NaOH buffer, pH 9.0, as previously described, and proteolysis was determined by spectrophotometry at 405 nm after 60 min.

Enzymatic activity of mature biofilm supernatant

C. parapsilosis (ATCC 22019) (10⁶ cells) was inoculated in 24-well polystyrene plates containing 1 mL of BHI broth in each well and incubated for 48 h at 37 °C without shaking to allow adherence and biofilm formation [16]. The supernatant was collected, centrifuged, and the protein concentration measured. The presence of enzymatic activity and its optimum pH was determined using 500 µg of proteins and the substrate N-Benzoyl-Phe-Val-Arg-pNA (200 µM), as previously described. Reactions were performed at 37 °C across a range of pH values using distinct 50 mM buffers: sodium citrate (pH 3.0–5.0), sodium phosphate (pH 6.0–8.0), and glycine–NaOH (pH 9.0–12.0).

Enzymatic activity of the supernatant during biofilm formation

C. parapsilosis (ATCC 22019) cells (10⁶) were inoculated in different 96-well plates containing 200 µL of BHI medium in each well, and the plates were incubated at 37 °C without agitation. The assays were performed at different timepoints (24, 48, 72 and 96 h), and three parameters were analyzed: supernatant enzymatic activity, biofilm cellular viability and biofilm biomass. Enzymatic activity was determined using the collected supernatant (500 µg of protein) and the substrate N-Benzoyl-Phe-Val-Arg-pNA (200 µM) in 50 mM glycine–NaOH, pH 9.0, as previously described. The biofilm cellular viability was measured through the metabolic activity of the cells by the reduction of XTT (2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide), as previously described [18]. Shortly, the wells of the plate were washed three times with PBS, and 158 µL of PBS, 40 µL of XTT (1 mg/mL), and 2 µL menadione (0.4 mM) were added to each well. The plates were incubated for 3 h at 37 °C, away from light sources, and the reduction of XTT was measured at 492 nm in an ELISA plate reader. For biomass quantification, the wells were washed three times with PBS and 200 µL of crystal violet dye 0.4% was added to each well, and the plate was incubated at room temperature for 5 min. The wells were washed with distilled water to remove excess dye and 200 µL of absolute ethanol was added to the wells for 5 min. The bleach solution was then transferred to another 96-well plate, and the absorbance was measured at 570 nm in an ELISA reader.

Galleria mellonella assay

In this set of experiments, G. mellonella larvae weighing approximately 0.2–0.3 g and exhibiting uniform, healthy coloration were selected for use. For survival assays, fungal cells were cultured for 48 h at 37 °C in liquid BHI medium, then harvested, washed with sterile PBS, and resuspended in the same buffer. Larvae (n = 10 per group) were injected with 1 × 10⁶ fungal cells per larva through the last right proleg using a 10-µL Hamilton syringe (Hamilton Company, Reno, NV, USA), and incubated at 37 °C for up to five days. Survival was assessed daily by evaluating the larvae’s responsiveness to gentle tactile stimulation on the head and body; lack of movement was recorded as death. In addition, visual inspection of larval coloration was performed as an auxiliary indicator of physiological stress or disease progression.

Statistical analysis

All experiments were performed in triplicate, in three independent experimental sets. Statistical analyses were performed using either Student’s t-test or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test, using GraphPad Prism 8 for Windows (GraphPad Software, La Jolla, CA, USA). In the case of the survival curve, the log-rank (Mantel-Cox) test was used to indicate statistical significance. The correlation test was determined by Pearson’s correlation coefficient (r). In all the analyses, P values of 0.05 or less were considered statistically significant.

Results

C. parapsilosis secretions cleave serine protease substrates: effects of protease inhibitors, pHs and temperatures

Initially, five chromogenic peptide substrates, with specific affinities for elastase, trypsin and chymotrypsin families of serine-type proteases, were used to assess the ability of cell-free conditioned supernatants from C. parapsilosis planktonic cells to cleave these substrates. The degradation rates of N-Suc-Gly-Gly-Phe-pNA (0.38 nmol pNA mg⁻¹ min⁻¹), N-Suc-Ala-Ala-Pro-Phe-pNA (0.41 nmol pNA mg⁻¹ min⁻¹), N-Suc-Ala-Ala-Ala-pNA (0.44 nmol pNA mg⁻¹ min⁻¹) and Nα-benzoyl-DL-4Arg (0.45 nmol pNA mg⁻¹ min⁻¹) were observed, confirming the presence of all three families of serine proteases in C. parapsilosis secretions. Notably, the cleavage rate of the N-benzoyl-Phe-Val-Arg-pNA substrate was significantly higher (0.74 nmol pNA mg⁻¹ min⁻¹) than that of the other substrates, suggesting that trypsin-like serine proteases are predominantly secreted by C. parapsilosis (Fig. 1A). As a result, this substrate was selected for further assays.

Fig. 1.

Cell-free supernatants of C. parapsilosis degrade various serine protease substrates under different pH and temperature conditions. (A) Concentrated supernatant from C. parapsilosis (ATCC 22019), containing 500 µg of protein, was incubated with 200 µM chromogenic substrates specific for different serine protease classes (subtilisin-, chymotrypsin- and trypsin-like) at 37 °C in 50 mM glycine-NaOH buffer, pH 9.0. Enzymatic activity was measured spectrophotometrically at 405 nm after 60 min of incubation. The asterisk denotes a statistically significant difference in the hydrolysis of trypsin-specific substrate N-benzoyl-Phe-Val-Arg-pNA compared to the other serine-type substrates (P < 0.05, Tukey’s multiple comparisons test). (B) Hydrolysis of N-Benzoyl-Phe-Val-Arg-pNA was confirmed to be catalyzed by serine-type proteases through inhibition assays using specific inhibitors: E-64 (cysteine), pepstatin A (aspartic), 1,10-phenanthroline (metallo), and PMSF, TLCK and benzamidine (serine proteases). Asterisks denote statistically significant differences compared to the control group and the group treated with the protease inhibitor (P < 0.05, Student’s t-test). (C) A total of 500 µg of protein was incubated with 200 µM N-Benzoyl-Phe-Val-Arg-pNA across a pH range using different buffers at 50 mM: sodium citrate buffer (pH 3.0–5.0), sodium phosphate buffer (pH 6.0–8.0) and glycine-NaOH buffer (pH 9.0–12.0). (D) Substrate degradation (N-Benzoyl-Phe-Val-Arg-pNA) was also evaluated at different temperatures using 50 mM glycine-NaOH buffer, pH 9.0. The asterisk denotes a statistically significant difference in substrate hydrolysis at 28 °C compared to the other tested temperatures (P < 0.05, Tukey’s multiple comparisons test). Bars represent the mean ± standard deviation of three independent experiments

In the presence of inhibitors targeting different protease classes, a significant reduction in substrate cleavage was observed with PMSF (32.8%), TLCK (40.2%) and benzamidine (50.7%), all classical inhibitors of serine-type proteases (Fig. 1B). In contrast, no inhibition was detected with E-64 (a cysteine protease inhibitor), pepstatin A (an aspartic protease inhibitor) or 1,10-phenanthroline (a metalloprotease inhibitor), further confirming the specificity of serine-type proteases in the observed activity (Fig. 1B).

Serine-type proteolytic activity was observed across a broad pH range, from 6.0 (0.05 nmol pNA mg⁻¹ min⁻¹) to 12.0 (0.27 nmol pNA mg⁻¹ min⁻¹), with optimal activity at pH 9.0 (0.74 nmol pNA mg⁻¹ min⁻¹) (Fig. 1C). Similarly, proteolytic activity was detected across a wide temperature range (28 °C to 40 °C), peaking at 32 °C (0.75 nmol pNA mg⁻¹ min⁻¹) (Fig. 1D).

Culture media influence the secretion of proteases by C. parapsilosis cells

Distinct proteolytic activities were observed in the supernatants of C. parapsilosis cultured in BHI and YCB-BSA media. Minimal substrate cleavage of the cathepsin D (aspartic protease) substrate was detected in the BHI-derived supernatant (0.05 fluorescent arbitrary units), while significantly higher proteolysis was observed with the YCB-BSA-derived supernatant (0.68 fluorescent arbitrary units) (Fig. 2A). In contrast, when using the serine protease substrate N-benzoyl-Phe-Val-Arg-pNA, the BHI-derived supernatant exhibited high levels of substrate degradation (0.74 nmol pNA mg⁻¹ min⁻¹). In comparison, the YCB-BSA-derived supernatant displayed considerably lower proteolytic activity (0.08 nmol pNA mg⁻¹ min⁻¹) (Fig. 2B).

Fig. 2.

Culture medium composition modulates protease secretion in C. parapsilosis. Fungal cells (ATCC 22019) were cultured into BHI and YCB supplemented with 0.1% BSA, and incubated for 48 h at 37 °C under constant agitation. (A) Cell-free supernatants (500 µg of protein) were incubated with 2 µM of the fluorogenic substrate 7-methoxycoumarin-4-acetyl-Gly-Lys-Pro-Ile-Leu-Phe-Phe-Arg-Leu-Lys (DNP)-D-Arg-amide in 50 mM sodium citrate buffer (pH 4.0), and (B) with 200 µM of the chromogenic substrate N-Benzoyl-Phe-Val-Arg-pNA in 50 mM glycine-NaOH buffer (pH 9.0), both for 60 min at 37 °C. Enzymatic activities were measured by spectrophotometry at 405 nm for the serine protease substrate, and by spectrofluorometry using an excitation wavelength of 340 nm and an emission wavelength of 490 nm for the aspartic protease substrate. Bars represent the mean ± standard deviation of three independent experiments. Asterisks denote statistically significant differences between the two systems (P < 0.05, Student’s t-test). FAU = Fluorescent arbitrary units

C. parapsilosis growth impacts protease activity level

The cellular proliferation of C. parapsilosis was measured at various time points, revealing a typical growth curve in which cell density increased steadily from 24 h (9.5 × 10⁶ cells/mL) to 72 h (39.5 × 10⁶ cells/mL) of incubation in BHI medium, followed by a slight decline at 96 h (36.5 × 10⁶ cells/mL) (Fig. 3A). In parallel, the degradation of N-benzoyl-Phe-Val-Arg-pNA by C. parapsilosis secretions was observed at all time points. Consistent with the growth curve, enzymatic activity peaked at 72 h (1.14 nmol pNA mg⁻¹ min⁻¹) before decreasing at 96 h (0.97 nmol pNA mg⁻¹ min⁻¹) (Fig. 3B).

Fig. 3.

Evaluation of serine protease activity during planktonic growth of C. parapsilosis. Fungal cells (ATCC 22019) were inoculated into 20 mL of BHI medium (1 × 10⁶ cells/mL), and incubated at 37 °C under constant agitation. At 24, 48, 72 and 96 h, 1 mL aliquots were collected, centrifuged, and the supernatant harvested for analysis. (A) Cell density at each time point was determined using a Neubauer chamber to construct a growth curve. (B) The supernatant from each time point was concentrated and then incubated (500 µg of protein) with 200 µM N-Benzoyl-Phe-Val-Arg-PNA in 50 mM glycine-NaOH buffer, pH 9.0, at 37 °C for 60 min, followed by spectrophotometric measurement of enzymatic activity. Bars represent the mean ± standard deviation of three independent experiments. Symbols indicate statistically significant differences according to Tukey’s multiple comparisons test, as follows: (*, P < 0.05 or **, P < 0.001) denote significant differences between 24 h and 48, 72–96 h; (♦, P < 0.05) indicates significant differences between 48 h and 72–96 h; and (◼) indicates a significant difference between 72 h and 96 h

Serine proteolytic activity in various C. parapsilosis clinical isolates

Conditioned supernatants (100 µg of protein) from nine different clinical isolates of C. parapsilosis were tested for their ability to cleave the substrate N-Benzoyl-Phe-Val-Arg-pNA without prior concentration. All isolates demonstrated substrate degradation, albeit at varying intensities (Fig. 4). The levels of substrate degradation observed with the ATCC 22019 strain using 100 µg of protein were comparable to those obtained in previous assays with 500 µg, suggesting that substrate saturation had already been reached at the lower protein concentration. Remarkably, all clinical isolates of C. parapsilosis exhibited significantly higher degradation efficiency against the serine-type protease substrate compared to the reference strain ATCC 22019, with the following results: 83.7% (strain 251), 114.8% (strain 241), 243.2% (strains 454 and 225), 291.9% (strain 235), 297.3% (strain H234), 328.3% (strain 276), 331% (strain 222) and 423% (strain 275). Notably, strain 275 displayed the highest serine protease activity, significantly surpassing the other strains.

Fig. 4.

Comparison of serine-type protease activity in cell-free supernatants from different clinical isolates of C. parapsilosis. Cell-free supernatants were collected from nine clinical isolates and the reference strain ATCC 22019 after cultivation in 10 mL of BHI medium at 37 °C for 48 h under constant agitation. Following centrifugation, protein concentrations were determined, and 100 µg of protein from each sample was incubated with 200 µM N-benzoyl-Phe-Val-Arg-PNA in 50 mM glycine-NaOH buffer, pH 9.0, at 37 °C for 60 min. Enzymatic activity was assessed spectrophotometrically at 405 nm. Bars represent the mean ± standard deviation of three independent experiments. Symbols (*, P < 0.05 or **, P < 0.001) indicate statistically significant differences in serine-type protease production between the reference strain ATCC 22019 and clinical isolates of C. parapsilosis, as determined by Tukey’s multiple comparisons test

Serine-type protease activity during biofilm formation

The supernatant obtained from C. parapsilosis biofilm also demonstrated the ability to degrade N-Benzoyl-Phe-Val-Arg-pNA across a range of pH values, from pH 6.0 (0.094 nmol pNA mg⁻¹ min⁻¹) to pH 12.0 (0.255 nmol pNA mg⁻¹ min⁻¹) (Fig. 5). Over time, proteolytic activity showed a progressive increase, from 0.20 to 0.36, 0.43 and 0.54 nmol pNA mg⁻¹ min⁻¹ at 24, 48, 72 and 96 h, respectively (Fig. 6A). Similarly, biofilm viability increased over time, rising by 31% at 48 h (absorbance = 1.71) and by 79.5% at 72 h (absorbance = 2.35) compared to the initial 24-h measurement (absorbance = 1.30), reaching a significant peak. However, after 96 h, viability declined by 21.77% (absorbance = 1.02) relative to the 24-h point (Fig. 6B). Additionally, biofilm biomass also increased over time, becoming 46.5% denser after 96 h (absorbance = 0.142) compared to 24 h (absorbance = 0.076) (Fig. 6C).

Fig. 5.

Influence of pH on serine protease activity in cell-free supernatants from biofilm-forming C. parapsilosis cells. Fungal cells (ATCC 22019) at 1 × 10⁶ cells per well were inoculated into 24-well polystyrene plates containing 1 mL of BHI medium and incubated for 48 h at 37 °C without agitation to promote adherence and biofilm formation. After incubation, supernatants were collected and protein concentrations determined. Supernatants (500 µg protein) were incubated with 200 µM N-benzoyl-Phe-Val-Arg-pNA in 50 mM glycine-NaOH buffer, pH 9.0, at 37 °C for 60 min in buffers spanning different pH ranges: 50 mM sodium citrate (pH 3.0–5.0), 50 mM sodium phosphate (pH 6.0–8.0) and 50 mM glycine–NaOH (pH 9.0–12.0). Enzymatic activity was measured spectrophotometrically at 405 nm. Bars represent the mean ± standard deviation of three independent experiments

Fig. 6.

Comparison of serine protease activity, cellular viability and biomass levels at different time points during biofilm formation of C. parapsilosis cells on polystyrene. C. parapsilosis (ATCC 22019) was inoculated (1 × 10⁶ cells) into each well of a 96-well plates containing 200 µL of BHI medium and incubated at 37 °C. At 24, 48, 72 and 96 h, supernatants were collected, centrifuged, and used for assays. (A) Serine protease activity was assessed by incubating 500 µg of protein from supernatants with 200 µM N-benzoyl-Phe-Val-Arg-pNA in 50 mM glycine-NaOH buffer, pH 9.0, at 37 °C for 60 min, followed by spectrophotometric measurement at 405 nm. (B) Biofilm cellular viability was determined by metabolic activity using the XTT reduction assay. (C) Biofilm biomass was quantified via staining with 0.4% crystal violet dye. Bars represent the mean ± standard deviation of three independent experiments. Symbols indicate statistically significant differences (P < 0.05) according to Tukey’s multiple comparisons test, as follows: (*) denotes significant differences between 24 h and 48, 72–96 h; (♦) indicates significant differences between 48 h and 72–96 h; and (◼) indicates a significant difference between 72 h and 96 h

In vivo infection in G. mellonella

Overall, the in vivo infection assay demonstrated that all C. parapsilosis isolates induced high levels of larval mortality by 96 h post-infection (Fig. 7). However, no significant correlation was found between the rate of larval death and the level of serine protease production by the fungal isolates (P > 0.05, Pearson correlation test). Despite this, clear isolate-dependent differences in virulence were observed. For example, at 24 h post-infection, isolate 225 was the most virulent, causing 50% larval mortality. In contrast, isolates 234 and 241 each killed 30% of larvae, isolates 275 and 454 killed 20%, and isolate 276 caused 10% mortality. At this same time point, isolates 222, 235, 251 and the ATCC reference strain did not kill any larvae, with mortality rates significantly lower than that of isolate 225 (P < 0.05, Mantel-Cox test) (Fig. 7). By 48 h post-infection, isolates 234, 241 and 251 had killed all larvae, while isolates 225, 275 and 454 reached 100% lethality by 72 h. The remaining fungal isolates exhibited larval mortality ranging from 80 to 90% by the end of the experimental period (Fig. 7).

Fig. 7.

Comparison of the virulence of C. parapsilosis isolates in the Galleria mellonella infection model. Larvae were inoculated with 1 × 10⁶ fungal cells per larva and incubated at 37 °C for up to 4 days. Survival rates are expressed as the percentage of surviving larvae relative to the uninfected control group (PBS-treated)

Discussion

The incidence of fungal infections has significantly increased, leading to higher rates of morbidity and mortality, particularly among immunocompromised individuals [1]. Notably, non-albicans Candida species, such as C. parapsilosis, are becoming more commonly isolated from patients with systemic fungal infections [12]. Several virulence factors contribute to the establishment of infection, with hydrolytic enzymes playing a key role in this process. While the contributions of secreted aspartic proteases (Saps) to the physiology and virulence of Candida have been well-documented [19, 20], very little is known about the role of serine proteases. It has been suggested that serine proteases can degrade serum proteins and components of the extracellular matrix, aiding in breaching host barriers and facilitating the attack on macrophages and lymphocytes [13–16, 21]. This suggests that serine proteases could serve as potential targets for novel therapeutic interventions. However, research on the production of serine proteases by C. parapsilosis remains limited. In this study, we aimed to evaluate the secretion and activity of these proteases in both planktonic and biofilm-growing C. parapsilosis cells.

The cell-free supernatant of C. parapsilosis exhibited the ability to cleave various peptidyl-pNA chromogenic substrates specific to different serine protease classes, including subtilisin, trypsin and chymotrypsin. However, significantly higher levels of degradation of the N-Benzoyl-Phe-Val-Arg-pNA substrate suggest that trypsin-like serine proteases are predominantly secreted by C. parapsilosis. This is further supported by the lack of inhibition from metallo, cysteine or aspartic protease inhibitors, and a marked reduction in proteolytic activity when exposed to classical serine protease inhibitors like PMSF, as well as specific trypsin-like protease inhibitors, such as TLCK and benzamidine. The production of trypsin-like serine proteases has been documented in various fungal species [22–26], with the expression of these proteases often linked to fungal pathogenesis [27–29]. It has been previously reported that species within the emerging pathogenic C. haemulonii complex secrete serine-type proteases that exhibit activity across a broad range of pH levels and temperatures, as well as against various substrates. Interestingly, high levels of substrate degradation of N-Benzoyl-Phe-Val-Arg-pNA at alkaline pH, which is significantly inhibited by trypsin inhibitors such as PMSF and AEBSF, suggest that, similar to our observations with C. parapsilosis, the C. haemulonii complex primarily secretes trypsin-like serine proteases [21, 30]. Notably, trypsin inhibitors have been shown to induce significant morphological changes, alter plasma membrane integrity, reduce intracellular ATP levels, promote lipid peroxidation and inhibit the proliferation of medically relevant Candida species, such as C. albicans, C. krusei and C. tropicalis [31, 32]. Previous studies from our group have demonstrated that the serine protease inhibitor TPCK can trigger a range of morphophysiological changes and interfere with virulence traits in C. parapsilosis and C. haemulonii [33, 34]. Collectively, these findings highlight the crucial role of trypsin-like serine proteases in C. parapsilosis biology and virulence, underscoring the need for further investigation into the dynamics of their production and activity.

During its life cycle, C. parapsilosis is exposed to various stressors, including fluctuations in environmental pH and temperature. Its ability to adapt to these changing conditions is a critical factor for successful host colonization and establishing infection. This adaptability enables the organism to thrive in diverse environments, enhancing its virulence and persistence in the host [35, 36]. This evolutionary adaptation likely explains the thermal stability and the broad pH range over which we observed substrate degradation. The adaptability of serine proteases to diverse environmental conditions is a widespread phenomenon, and has been documented across a variety of organisms, including bacteria, fungi and mammalian tissues. This versatility is essential for survival in fluctuating environments and plays a significant role in the pathogenesis of many microbial species [37–40]. The alkaline pH (pH 9.0), at which optimal substrate degradation was observed, also appears to be a common characteristic of many pathogenic fungal species, including C. albicans, Scedosporium apiospermum, Cryptococcus neoformans, Trichophyton rubrum, Coccidioides immitis, Microsporum canis, Histoplasma capsulatum, Aspergillus niger, Aspergillus flavus and Aspergillus fumigatus. This feature is not limited to pathogenic fungi, as it has also been observed in non-pathogenic species such as Aspergillus clavatus, Conidiobolus coronatus, Rhizopus oligosporus, Trichoderma harzianum and Cephalosporium sp. This shared trait underscores the evolutionary significance of alkaline tolerance in fungi, particularly for those involved in pathogenic interactions [41–48]. It is noteworthy that a previous study identified a serine protease secreted by C. parapsilosis with an optimal activity pH of 5.5 [49]. This difference in pH preference could be attributed to variations in culture conditions and the nutrient composition of the culture medium, both of which are known to influence protease expression [50, 51]. Our data corroborates this, demonstrating a predominance of serine proteases secreted into BHI culture medium, while their expression was repressed when YCB supplemented with 0.1% BSA was used, favoring the secretion of aspartic proteases. This observation aligns with previous research showing that when C. albicans is cultivated in a defined medium with BSA or hemoglobin as a nitrogen source, Saps’ expression is predominant, whereas in BHI broth or RPMI, serine and metalloproteases become more prevalent [52, 53]. It has been suggested that this phenomenon may be modulated by factors such as the molecular mass of proteins in the culture medium, pH, metal ions and incubation time, although the precise mechanisms remain unclear [52, 54, 55].

When comparing the ability to cleave the substrate N-Benzoyl-Phe-Val-Arg-pNA, we found that the supernatant from the reference strain of C. parapsilosis (ATCC 22019) exhibited significantly lower proteolytic activity than several clinical isolates. Remarkably, all of the more recently isolated clinical samples displayed much higher proteolytic activity. This suggests that these proteases could play a crucial role in the interaction between the pathogen and the host during the infection process. Similar trends have been observed in other human fungal pathogens, such as C. neoformans, C. albicans, Paracoccidiodes brasiliensis and Blastomyces dermatitidis, where more recently isolated strains tend to be more virulent than older strains. Prolonged in vitro cultivation of these pathogens often leads to a reduction in virulence, which can be restored after passage through an in vivo model. This further highlights the potential link between proteolytic activity and the pathogenicity of these clinical isolates [56–60]. Additionally, studies have shown that more recently isolated samples of Fonsecaea pedrosoi exhibit significantly higher Sap activity compared to older strains. Similarly, the production of Saps in C. albicans is notably increased after passage through a murine model. These findings further underscore the potential role of protease activity in the virulence and pathogenicity of fungal pathogens, suggesting that protease expression may be modulated in response to environmental factors, such as host interactions, which could enhance the fungal ability to cause infection [50, 61, 62]. Our data indicate that serine protease levels are significantly higher in more recently isolated samples, suggesting that these proteases may play a crucial role in the colonization process and in the establishment of infection within the host.

We observed a peak in secreted proteolytic activity in the supernatant that coincided with the highest cellular density during planktonic growth, suggesting a role for these serine proteases in C. parapsilosis proliferation. This finding is consistent with the established role of fungal proteases in various essential biological processes, including nutrient acquisition, cellular growth and proliferation. Proteolytic activity likely facilitates the breakdown of host proteins and extracellular matrix components, contributing to the pathogen’s ability to proliferate and thrive in diverse environments [63, 64]. It is noteworthy that serine protease activity increased over time during biofilm formation, following a trend that paralleled the increase in cellular viability up to 72 h. The peak of serine protease activity coincided with the highest biomass levels. This is significant, as C. parapsilosis is known for its ability to adhere to medical devices and form biofilms, structured communities of cells encased in an extracellular matrix, which significantly enhances the resistance of fungal cells to antifungal treatments and the host immune response [4, 65]. The production of Saps during Candida biofilm formation is dynamic and variable. For example, upregulation of the SAP1, SAP2, SAP4, SAP5, SAP6, SAP9 and SAP10 genes during C. albicans biofilm formation has been reported [65–72]. Disruption of SAP5, SAP6, SAP9 and SAP10 genes reduces its biofilm, while deletion of SAP9 alters the biofilm architecture [66, 73]. In contrast, C. tropicalis does not exhibit increased expression of SAP1, SAP2 and SAP4 during biofilm formation, and even shows decreased expression of SAP3 [69]. In C. parapsilosis, reduced levels of Sap1 and Sap2 have been reported in the supernatant of cells growing in a biofilm [19], while Candida auris mutants deficient in SAP2 and SAP6 exhibit impaired biofilm formation [74]. These data demonstrate differential expression of Saps in biofilm formation, suggesting a probable role for the enzymes in biofilm development. However, while the involvement of serine proteases in bacterial biofilm formation has been well documented [75, 76], the specific role of secreted serine proteases in C. parapsilosis biofilm formation remains unclear. In a previous work, we demonstrated that the serine protease inhibitor TPCK can reduce cellular viability and partially disrupt mature biofilms of C. parapsilosis [33]. Similarly, TPCK has been shown to inhibit biofilm formation in C. albicans [77]. These findings, combined with the data from this study, strongly suggest that serine proteases play a crucial role in biofilm formation, a key virulence factor for Candida species [77]. More importantly, biofilm formation is essential for C. parapsilosis’s ability to establish infections [78, 79], particularly in medical devices and parenteral nutrition systems, areas where this pathogen is particularly problematic [78, 79]. Consequently, targeting serine proteases could be a promising strategy to prevent biofilm formation and mitigate the risk of C. parapsilosis infections in clinical settings.

In conclusion, the cell-free culture supernatant of C. parapsilosis demonstrates the ability to cleave various substrates, with a significantly higher affinity for the serine-type protease substrate N-Benzoyl-Phe-Val-Arg-PNA. The observed serine-type proteolytic activity appears to be primarily mediated by an alkaline, trypsin-like serine protease that demonstrates remarkable tolerance to variations in pH and temperature. Additionally, the production of this protease is modulated by the composition of the culture medium. The serine proteases secreted by C. parapsilosis are implicated in both cellular proliferation and biofilm formation, processes critical for the pathogen’s ability to interact with the host. Notably, higher levels of serine protease production were observed in the more recently isolated clinical specimens, suggesting a potential role for these enzymes in promoting infection and enhancing virulence. However, this hypothesis was not supported by the in vivo G. mellonella infection model, as all C. parapsilosis isolates tested in this study exhibited comparable levels of virulence. In this context, high larval mortality rates were observed across isolates, with significant lethality occurring after 96 h post-infection. These findings indicate that, although protease production may contribute to pathogenicity, it is not the sole determinant of virulence in this model system. Nevertheless, our findings, in conjunction with previous data demonstrating reduced virulence upon treatment with the serine protease inhibitor TPCK [33], support the notion that secreted serine-type proteases represent promising molecular targets for the development of novel antifungal strategies aimed at mitigating infections caused by this opportunistic pathogen. Although we did not directly identify or characterize the specific protease responsible for the observed activity, which can be considered an acknowledged limitation of our study, these findings open new avenues for future research aimed at the identification, biochemical characterization, and functional analysis of serine proteases in this opportunistic fungal pathogen.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Rafael Gandra (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing - original draft), Lucas Giovanini (Formal analysis, Investigation, Methodology, Writing - original draft), Marta Branquinha (Conceptualization, Data curation, Formal analysis, Funding acquisition, Supervision, Writing - review & editing), André Santos (Conceptualization, Data curation, Formal analysis, Funding acquisition, Project administration, Supervision, Writing - review & editing).

Declarations

Competing interests

Authors have no competing interests.