Deciphering the Role of PIG1 and DHN-Melanin in Scedosporium apiospermum Conidia

Scedosporium apiospermum is a saprophytic filamentous fungus involved in human infections, of which the virulence factors that contribute to pathogenesis are still poorly characterized. In particular, little is known about the specific role of dihydroxynaphtalene (DHN)-melanin, located on the external layer of the conidia cell wall. We previously identified a transcription factor, PIG1, which may be involved in DHN-melanin biosynthesis. To elucidate the role of PIG1 and DHN-melanin in S. apiospermum, a CRISPR-Cas9-mediated PIG1 deletion was carried out from two parental strains to evaluate its impact on melanin biosynthesis, conidia cell-wall assembly, and resistance to stress, including the ability to survive macrophage engulfment. ΔPIG1 mutants did not produce melanin and showed a disorganized and thinner cell wall, resulting in a lower survival rate when exposed to oxidizing conditions, or high temperature. The absence of melanin increased the exposure of antigenic patterns on the conidia surface. PIG1 regulates the melanization of S. apiospermum conidia, and is involved in the survival to environmental injuries and to the host immune response, that might participate in virulence. Moreover, a transcriptomic analysis was performed to explain the observed aberrant septate conidia morphology and found differentially expressed genes, underlining the pleiotropic function of PIG1.


Introduction
Scedosporium species are filamentous fungi widely distributed throughout the environment, especially in soil, where they live as saprophytes. However, they can also behave as opportunistic pathogens. Scedosporium sp. are involved in a large range of human diseases, from chronic localized infections to immuno-allergic diseases in immunocompetent hosts and severe invasive infections in immunocompromised patients [1]. Scedosporium is also the second most prevalent mold after Aspergillus involved in airway colonization of cystic fibrosis (CF) patients, with a prevalence ranging from 3 to 25% depending on the geographical area and technical procedures used for mycological culture [2][3][4]. In addition, there has been an increasing number of reports on the involvement of Scedosporium/Lomentospora in invasive life-threatening infections in individuals with underlying immune-system impairment, including those with hematological malignancies and bone marrow or solid organ transplants. [5,6]. The observed poor outcome is also related to the frequent resistance to systemic mold-active antifungals available to treat such infections [7].  (36.1 kb), from the publishe apiospermum genome [30]. The scytalone deshydratase encoding gene is located out of the cluste In this context, we aimed to assess the functions of PIG1 and DHN-melanin in apiospermum conidia using three complementary approaches applied to PIG1-deleted bino strains vs. parental S. apiospermum strains: (i) phenotypic characterization, (ii) tr scriptomic analysis, and (iii) the evaluation of the macrophage response.

Strains and Culture Conditions
Two different parental strains were used in this study: - The Scedosporium apiospermum wild-type strain deposited at the BCCM/IHEM cult collection (Brussels, Belgium) under the accession number IHEM 14462 (WT 144 was isolated in 1998 from a sputum sample of a CF patient. The genome of this str has been sequenced and annotated [30]; -A non-homologous end joining-deficient strain (ΔKU70) previously obtained fr the WT 14462 strain [33].
All strains were conserved frozen at −80 °C and cultured at 37 °C. Fungi were r tinely maintained by passage on supplemented potato dextrose agar plus 0.5% chlora phenicol supplemented with phleomycin (20 µ g/mL) for the culture of the ΔKU70 stra Hygromycin B (50 µ g/mL) was also added to the culture medium of the PIG1-disrup strains to maintain the selection pressure. Moreover, the WT 14462 strain was cultured Potato dextrose agar (PDA) supplemented with 50 µ g/mL tricyclazole (TRC, PEST NAL ® , Sigma-Aldrich) to obtain a melanin-deficient control (WT-TRC).
All experiments were performed with conidia harvested from nine-day cultures scraping. The conidia were then filtered through sterilized Miracloth (pore size 22-25 µ for hyphae removal and washed once in sterile water. Conidia were counted usin LUNA™ automated particle counter (Logos Biosystems, Villeneuve d'Ascq, France).

Disruption of the PIG1 Gene
The alignment of S. apiospermum PIG1 protein sequence with homologous sequen of phytopathogen fungi [31] let to adjust the nucleotide sequence of PIG1-encoding g (SAPIO_CDS8655), previously annotated [30].The 2793 bp gene encoding PIG1 was rupted using CRISPR-Cas9 technology to induce a double-strand break at the 5′ end of PIG1 gene using a specific crRNA protospacer (sequence in Table S1). After preparat  [30]. The scytalone deshydratase encoding gene is located out of the cluster.
In this context, we aimed to assess the functions of PIG1 and DHN-melanin in S. apiospermum conidia using three complementary approaches applied to PIG1-deleted albino strains vs. parental S. apiospermum strains: (i) phenotypic characterization, (ii) transcriptomic analysis, and (iii) the evaluation of the macrophage response.

Strains and Culture Conditions
Two different parental strains were used in this study: - The Scedosporium apiospermum wild-type strain deposited at the BCCM/IHEM culture collection (Brussels, Belgium) under the accession number IHEM 14462 (WT 14462) was isolated in 1998 from a sputum sample of a CF patient. The genome of this strain has been sequenced and annotated [30]; -A non-homologous end joining-deficient strain (∆KU70) previously obtained from the WT 14462 strain [33].
All strains were conserved frozen at −80 • C and cultured at 37 • C. Fungi were routinely maintained by passage on supplemented potato dextrose agar plus 0.5% chloramphenicol supplemented with phleomycin (20 µg/mL) for the culture of the ∆KU70 strains. Hygromycin B (50 µg/mL) was also added to the culture medium of the PIG1-disrupted strains to maintain the selection pressure. Moreover, the WT 14462 strain was cultured in Potato dextrose agar (PDA) supplemented with 50 µg/mL tricyclazole (TRC, PESTANAL ® , Sigma-Aldrich) to obtain a melanin-deficient control (WT-TRC).
All experiments were performed with conidia harvested from nine-day cultures by scraping. The conidia were then filtered through sterilized Miracloth (pore size 22-25 µm) for hyphae removal and washed once in sterile water. Conidia were counted using a LUNA™ automated particle counter (Logos Biosystems, Villeneuve d'Ascq, France).

Disruption of the PIG1 Gene
The alignment of S. apiospermum PIG1 protein sequence with homologous sequences of phytopathogen fungi [31] let to adjust the nucleotide sequence of PIG1-encoding gene (SAPIO_CDS8655), previously annotated [30]. The 2793 bp gene encoding PIG1 was disrupted using CRISPR-Cas9 technology to induce a double-strand break at the 5 end of the PIG1 gene using a specific crRNA protospacer (sequence in Table S1). After preparation of the gRNA by mixing crRNA and the universal tracrRNA (Table S1) in equimolar amounts at a final concentration of 33 µM (95 • C for 5 min and then RT for 15 min), RNP complex formation was performed according to the method developed by Abdallah et al. for A. fumigatus [34].
The repair cassette was obtained by amplification of the hygromycin-resistance gene (hygromycin B transferase-HYGR) from the plasmid pCB1636 using M13 primers flanked by an 80-bp sequence homologous to that of the extremities of the target gene. PCR amplification was performed using Q5 high-fidelity DNA polymerase. Transformation was performed as described previously [35] on 10 7 protoplasts from the two strains of S. apiospermum (WT 14462 and ∆KU70) with 5 µg of repair cassette and 13.25 µL of the RNP complex (manuscript in preparation). Cultures incubated at 37 • C were checked daily for disrupted upward growth in the selection marker layer. Monosporic isolates of disruptants were then subcultured on YPDA (yeast potato dextrose agar) supplemented with 20 µg/mL phleomycin and/or 50 µg/mL hygromycin B. Genomic DNA was extracted using phenol/chloroform.
The genotype was analyzed and validated using two different methods: (i) PCR amplification with primers designed against the upstream and downstream regions of the insertion and sequencing of the amplicons and (ii) Southern blotting, as previously described [35].

Melanin Detection by Electronic Paramagnetic Resonance (EPR)
A 5 × 10 7 /mL conidial suspension was prepared in sterile water and 300 µL transferred to a quartz EPR tube. EPR examination was performed at 100 K using an Elexsys EPR spectrometer (Bruker, Wissembourg, France) with a 3330 G center field, 2 mW microwave power, 9.377 GHz microwave frequency, 10 G modulation amplitude, and 100 kHz modulation frequency. Synthetic melanin (ref M8631, Sigma Aldrich, Saint-Quentin-Fallavier, France) was used as a standard for melanin quantification. Experiments were performed at least three times.
For transmission electron microscopy (TEM), ethanol was replaced by propylene oxide (3 × 20 min) and samples were impregnated overnight in a propylene oxide-Epon mixture (1:1 v/v) and then in pure Epon for 16 h and 8 h. After polymerization (24 h at 37 • C, 24 h at 45 • C, and then 48 h at 60 • C), thin sections were directly examined on a JEM-1400 transmission electron microscope (Jeol, Paris, France; 120 kV).

Quantification of Polysaccharide Exposure of the Conidia Cell Wall
The exposure of cell-wall polysaccharides was assessed using wheat germ agglutinin Alexa Fluor TM 633 (WGA) and concanavalin A Alexa Fluor TM 488 (ConA) conjugates (Ther-moFisher Scientific, Illkirch-Graffenstaden, France), which bind to N-acetylglucosamine oligomers, and α-d-mannosyl and α-d-glucosyl residues, respectively. Briefly, 5 × 10 6 conidia in HBSS were incubated with WGA or ConA at 100 µg/mL under orbital shaking, protected from light, for 30 min at 37 • C. Conidia were washed twice in HBSS and fixed with 1% PFA in HBSS before storage at 4 • C until microscopic observation. Stained conidia (20 µL) were observed under a widefield inverted fluorescence microscope (Olympus IX71) using a 20× magnification lens. Images were digitally recorded using SoftWorx software version 6.5.2. Analysis and fluorescence quantification were performed using ImageJ (Fiji) software.
2.6. Germination and Biofilm Assays 2.6.1. Conidial Germination Assay A standardized inoculum of 2 × 10 3 conidia in YPD was added to each well of a 96-well microplate, which was incubated at 37 • C in 5% CO 2 in an IncuCyte ® Live Cell Imaging System (Sartorius, Dourdan, France). Each well was automatically imaged every hour for 7 h. Acquisitions were then manually analyzed by counting at least 200 cells per strain at each time point. 2.6.2. Biofilm Formation 2 × 10 6 conidia (200 µl) were transferred to YPD medium in flat-bottom 96-well polystyrene microtiter plates, centrifugated at 200 g for 2 min, and then incubated at 37 • C in 5% CO 2 for 2 h for adhesion [36]. Supernatants were discarded to remove non-adherent conidia and 200 µL RPMI supplemented with 2% glucose and 20% fetal calf serum (FCS) was added; plates were then incubated without agitation at 37 • C in 5% CO 2 for 24 h, 48 h, or 72 h. In parallel, blanks were prepared containing medium only. Subsequently, the supernatants were carefully discarded and the wells washed twice with PBS to remove nonadherent cells. The biofilm biomass was measured according to Mello et al. [37]. Biofilms were fixed with 200 µL 99% methanol for 15 min. After removal of the supernatant, plates were air-dried for 5 min. Then, 200 µL of 0.3% crystal violet in PBS (Sigma Aldrich) were added to each well before an incubation at RT for 20 min. The wells were washed twice with PBS to remove excess stain and the biomass then decolorized with 200 µL glacial acetic acid for 5 min. One hundred microliters of the acetic acid solution were transferred to a new 96-well plate and the absorbance measured at 590 nm using a microplate reader (SPECTROstar Nano, BMG Labtech, Champigny sur Marne, France).

Resistance to Ultraviolet Light
Conidia were stained with 0.1 mg/mL fluorescein isothiocyanate (FITC, Sigma-Aldrich) in carbonate buffer (0.1 M, pH = 9) at RT for 30 min, washed three times with carbonate buffer, and suspended in phosphate buffered saline (PBS). Suspensions of FITC-labelled conidia were adjusted to 5 × 10 5 /mL in uncovered sterile Petri dishes (53 mm in diameter, 6 mL) and exposed to UV-C radiation (λ = 254 nm) using a UV Stratalinker 2400 (Stratagene, La Jolla, CA, USA) at a dose of 30 mJ/cm 2 , according to the results of preliminary assays. Pellets from exposed and unexposed conidia were stained with 25 µg/mL propidium iodide (PI, Sigma-Aldrich) for 30 min at RT and protected from light. Conidia were washed with sterile water, fixed with 4% PFA, and analyzed using a LSRFortessa TM cytometer (BD Biosciences, Le-Pont-de-Claix, France). Each replicate included a negative control, reflecting live conidia (unexposed FITC + conidia), and a positive control consisting of heatkilled conidia (85 • C, 20 min). Both controls were used to set the gates for the positivity of each marker. Live conidia were FITC + PI − and the dead conidia were double-positive (FITC + PI + ). The percentage of live conidia was determined as the percentage of PI − conidia among FITC + conidia. The reduction in conidia survival was calculated relative to control conidia not exposed to UV (viability control) as: 100*((% of live conidia among unexposed conidia-% of live conidia among exposed conidia)/% of live conidia among unexposed conidia).

Resistance to Heat Shock
Conidial suspensions adjusted to 10 3 cells/mL were incubated at 40 • C or 50 • C for 15 min. After homogenization, 200 µL conidia suspension were inoculated in triplicate onto PDA plates. The survival rate was calculated by counting the number of colonies after incubation at 37 • C for 72 h and defined as the ratio of the colony number between exposed and control (unexposed) conditions.

Growth under Oxidative Conditions
The ability to grow in the presence of oxidizing compounds, i.e., menadione (Sigma Aldrich, stock solution prepared in DMSO) or cumene hydroperoxide (CHP, Sigma Aldrich), was studied by nephelometry [38]. To study hyphal growth in liquid media, 10 5 conidia were inoculated into microplate wells containing 40 µM menadione or 200 µM cumene hydroperoxide in potato dextrose broth (PDB). Fungal growth was monitored automatically for 40 h at 37 • C using a laser-based microplate nephelometer (NEPHELOstar, BMG Labtech). For each condition, the area under the growth curve representative of the lag phase was used to calculate the percentage of growth under the oxidative condition relative to the standard growth condition (PDB medium). At least three replicates were conducted per treatment.

Cell Wall Integrity Assay
The growth phenotypes of the parental strains (WT and ∆KU70) and their disruptants were assessed on PDA supplemented with 1 mg/mL Congo red or 0.0064% sodium dodecyl sulfate (SDS, purchased from Sigma Aldrich). Concentrations were chosen according to the results of preliminary assays. A growth control on PDA was added in each experiment. Plates were incubated at 30 • C for seven days. The colony diameter, reflecting the growth rate, was expressed relative to the growth on control PDA as follows: (100-100*(diameter on PDA − diameter on PDA + stressor)/diameter on PDA]). Each test was repeated three times, with four replicates each time.

Antifungal Susceptibility Testing
In vitro susceptibility testing was performed using broth microdilution for filamentous fungi according to EUCAST guidelines with concentrations ranging from 0.016 µg/mL to 8 µg/mL. The antifungals used were amphotericin B (AmB), voriconazole (VRC), itraconazole (ITC), posaconazole (POS), and anidulafungin (ANI), all purchased from Sigma-Aldrich, reconstituted in DMSO and stored frozen at −80 • C until use. The MICs were determined in flat-bottomed 96-well plates with conidial suspensions prepared in RPMI 1640 supplemented with 2% glucose, buffered with MOPS, and adjusted to a final concentration of 5 × 10 5 CFU/mL, as previously described [39]. Inoculated plates were incubated for 72 h at 35 • C. MICs were spectroscopically determined at 405 nm using a spectrophotometer (NEPHELOstar, BMG Labtech) as the antifungal concentration that resulted in 90% (AmB) or 50% growth inhibition (VRC, ITC, POS, ISA).

Transcriptome Analysis
For transcriptome sequencing, 10 8 S. apiospermum conidia from the ∆KU70 strain and its M1 mutant were incubated in YPD medium under continuous agitation (300 rpm) at 37 • C for 30 h. Mycelia were harvested, immediately frozen in liquid nitrogen, and stored at −80 • until RNA extraction. Four biological replicates were performed. Fifty milligrams of mycelium sprayed with liquid nitrogen were then subjected to total RNA extraction using the Plant/Fungi Total RNA Purification kit (Norgen Biotek, Thorold, Canada), with additional DNase treatment with TURBO DNase (Thermo Fisher Scientific) to remove residual genomic DNA. Total RNA was checked for integrity using a Bioanalyzer system (Agilent) and quantified using the fluorescent-based quantification Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific).
Directional libraries were prepared from rRNA-depleted RNA using the TruSeq Stranded mRNA sample Preparation Kit following the manufacturer's instructions (Illumina, Evry-Courcouronnes, France). Libraries were checked for quality on Bioanalyser DNA chips (Agilent, Les Ulis, France) and quantified using the fluorescent-based quantification Qubit dsDNA HS Assay Kit. RNA sequencing was performed on the Illumina NextSeq 500 platform using a High Output Kit (75 cycles).
Gene annotation was performed to analyze the functions of the DEGs using Blast2GO for gene ontology (GO) annotation and GO for functional classification.
Additionally, Interproscan (https://www.ebi.ac.uk/, accessed on 1 April 2022) was used for the functional analysis of proteins, classifying them into families and predicting domains based on several databases, including the Pfam, CDD, and Panther databases.
For quantitative real-time PCR analysis, RNA was reverse transcribed to cDNA using a high-capacity cDNA reverse transcription kit (Applied Biosystems, Illkirch-Graffenstaden, France) according to the manufacturer's instructions. Quantitative PCR amplifications were carried out in duplicate using PowerUp SYBR ® green PCR master mix (Applied Biosystems), 0.3 µM primers, and 2 µL of cDNA in a final volume of 10 µL, in 384-well optical plates, using a CFX384 OPUS TM real-time PCR detection system (Bio-Rad, Marnesla-Coquette, France). Gene-specific primers (Table S1) were synthesized by Eurogentec. Expression levels of target genes were normalized by comparison to the expression of the S. apiospermum UBC6 and FIS1 housekeeping genes. Results are expressed as 2 −∆∆Cq , referring to the fold induction in the T, M1, and M2 strains relative to the mean quantification cycle obtained with the parental strains.

Phagocytosis and Killing Assays
After differentiation, macrophages were plated at 5 × 10 5 cells per well in 24-well plates and infected with FITC-labelled conidia with a MOI of 5:1 for the ingestion assay and 1:1 for killing experiments. Plates were centrifugated at 500 g for 5 min at RT to synchronize phagocytosis and incubated at 37 • C in 5% CO 2 , protected from light, for 2 h (phagocytosis) or 6 h (phagocytosis and killing). Plates were then put on ice to block macrophage activity, the supernatants removed, and the cells washed three times with cold DPBS to remove extracellular non-adherent conidia.
-Phagocytosis assay: Cells were removed from the wells by gentle mixing and transferred to Eppendorf tubes, washed once with DPBS, and centrifuged at 500 g for 5 min. Macrophages were then transferred into 96-well U-bottom plates for labelling. Cells were stained for viability using the LIVE/DEAD™ Fixable Yellow Dead Cell Stain Kit (ThermoFisher Scientific) for 30 min at RT, protected from light. After washing, Fc receptors were blocked with Fc-R Blocking Reagent (Miltenyi) for 10 min at 4 • C and marked with anti-CD11b/Mac1 (Mouse antihuman, BD horizon) for 15 min at 4 • C, washed twice with PBS-2% SVF, fixed with 4% PFA, and analyzed on a LSRFortessa TM cytometer (BD Biosciences) to quantify the percentage of macrophages with ingested conidia. Data were analyzed using Flowlogic software according to the gating process previously designed to remove dead macrophages and distinguish between the CD11b + /FITC − population (non-infected macrophages) and the double-positive CD11b + /FITC + population (macrophages with adherent and/or ingested conidia) [47].
-Killing assay After 6 h of incubation, cells were immediately lysed using cold sterile water and incubated at 4 • C for at least 1 h or overnight. The well contents were then transferred into Eppendorf tubes and centrifuged at 10,000× g for 5 min. Pellets were then stained with PI and analyzed as described above (see Resistance to Ultraviolet Light).

Cytokine Assays
Primary macrophages were plated in 24-well plates at 10 6 cells/mL and incubated with 10 7 (MOI 10:1) live conidia in complete RPMI for 12 h at 37 • C in 5% CO 2 . Supernatants were then collected and immediately stored at −80 • C. Pro-inflammatory cytokines were measured using a 13 plex-bead-based immunoassay LEGENDplex ® multi-analyte flow assay (BioLegend, Paris, France) according to the manufacturer's instructions. Cytokine concentrations were determined using standard curves obtained using recombinant cytokine standards provided in the kit. Samples were analyzed on a LSRFortessa TM cytometer (BD Biosciences) and the data processed using the online platform provided by the kit manufacturer.

Statistical Analysis
All experiments were performed in triplicate or quadruplicate at least three times independently. Data are expressed as the mean ± standard error of the mean (SEM). Statistical analysis was performed using GraphPad Prism software (GraphPad Software, Inc., La Jolla, CA, USA). A one-way analysis of variance Kruskall-Wallis test or a Mann-Whitney test were used to evaluate differences between strains. In all analyses, p-values ≤ 0.05 were considered statistically significant.

Disruption of the PIG1 Gene in S. apiospermum
The CRISPR/Cas9 technology recently adapted to S. apiospermum was used to disrupt the PIG1 gene. Because of the low frequency of the homologous recombination, we performed the experiment in the wild-type strain (WT 14462), as well as in the previously developed non-homologous end joining-deficient strain ∆KU70 [33]. Three hygromycin-resistant strains corresponding to three putative PIG1 disruptants were obtained, one derived from the WT strain (T strain) and two derived from the ∆KU70 strain (∆KU70/∆PIG1-1 (M1) and ∆KU70/∆PIG1-2 (M2)). The genotype of the monosporic isolates was verified by Southern blotting. A band of expected size (1123 bp) was visualized for M1 and M2 after probe hybridization, instead of the 1700-bp band observed for the parental strain (Figure 2a,b). Replacement of the PIG1 gene by the repair cassette was confirmed by DNA sequencing (data not shown) and PIG1 mRNA expression was completely repressed (Figure 2c). By contrast, two bands (2411 bp and 1700 bp) were observed in the T strain. This result, associated with DNA sequencing, showed integration of the repair cassette 782 bp upstream of the PIG1 gene. No mutation was found in the PIG1 coding sequence, which could have possibly resulted in a reading-frame shift. Moreover, PIG1 mRNA was overexpressed in the T strain relative to the parental WT strain (Figure 2c).

Melanin Content and Cell-Wall Structure
In contrast to the dark brown pigment observed in the parental strains (Figure 3a), colonies from the three transformants showed a similar non-melanized white phenotype. The color defect did not result from an inability to sporulate, as supported by the presence of conidia by microscopic observation. Moreover, based on the paramagnetic structure of DHN-melanin, the quantification of conidia melanin could be assessed by electronic paramagnetic resonance (EPR), and confirmed the absence of a melanin-related EPR signal in conidia from the T, M1, and M2 strains ( Figure 3b). These results indirectly show that the transcription factor PIG1 regulates the expression of (a) gene(s) of the DHN-melanin cluster. Thus, the PIG1 mRNA overexpression observed in the T strain did not correlate with a functional transcription factor. This mutant was included in the subsequent experiments to decipher the role of DHN-melanin in S. apiospermum as the Southern blot confirmed that there was no ectopic integration elsewhere in the genome and as EPR confirmed the absence of DHN-melanin in the T strain.
J. Fungi 2023, 9, x FOR PEER REVIEW 9 of 28 repressed ( Figure 2c). By contrast, two bands (2411 bp and 1700 bp) were observed in the T strain. This result, associated with DNA sequencing, showed integration of the repair cassette 782 bp upstream of the PIG1 gene. No mutation was found in the PIG1 coding sequence, which could have possibly resulted in a reading-frame shift. Moreover, PIG1 mRNA was overexpressed in the T strain relative to the parental WT strain (Figure 2c).

Melanin Content and Cell-Wall Structure
In contrast to the dark brown pigment observed in the parental strains (Figure 3a), colonies from the three transformants showed a similar non-melanized white phenotype. The color defect did not result from an inability to sporulate, as supported by the presence of conidia by microscopic observation. Moreover, based on the paramagnetic structure of Of note, tricyclazole (TRC) was previously shown to be a fungicide that specifically inhibits both 1,3,6-trihydroxynaphtalene and 1,3,6,8-tetrahydroxynaphtalene reductases at nontoxic concentrations [48,49]. TRC is thus commonly used as a melanin biosynthesis inhibitor in studies focusing on DHN-melanin function. Here, the light reddish-brown pigment of the WT strain grown in the presence of TRC, similar to previous descriptions, reflects the accumulation of melanin shunt intermediates, confirming the inhibitory effect of TRC on DHN-melanin biosynthesis in S. apiospermum (Figure 3a).
Of note, tricyclazole (TRC) was previously shown to be a fungicide that specifically inhibits both 1,3,6-trihydroxynaphtalene and 1,3,6,8-tetrahydroxynaphtalene reductases at nontoxic concentrations [48,49]. TRC is thus commonly used as a melanin biosynthesis inhibitor in studies focusing on DHN-melanin function. Here, the light reddish-brown pigment of the WT strain grown in the presence of TRC, similar to previous descriptions, reflects the accumulation of melanin shunt intermediates, confirming the inhibitory effect of TRC on DHN-melanin biosynthesis in S. apiospermum (Figure 3a). In the absence of the external melanin layer in the T, M1, and M2 conidia, the cell wall content was expected to be different, or at least disorganized. We thus investigated the accessibility of the cell wall carbohydrates to mannose-binding and N-acetylglucosamine-binding lectins (ConA and WGA, respectively). Most of the parental cells were not labeled, whereas almost all conidia from the mutant strains were intensely labeled, together with the WT-TRC conidia, suggesting polysaccharide unmasking by the removal of the melanized outer layer (Figures 4 and S1). In the absence of the external melanin layer in the T, M1, and M2 conidia, the cell wall content was expected to be different, or at least disorganized. We thus investigated the accessibility of the cell wall carbohydrates to mannose-binding and N-acetylglucosaminebinding lectins (ConA and WGA, respectively). Most of the parental cells were not labeled, whereas almost all conidia from the mutant strains were intensely labeled, together with the WT-TRC conidia, suggesting polysaccharide unmasking by the removal of the melanized outer layer (Figures 4 and S1).  In addition, TEM examination showed the mean cell wall thickness to be significantly less for the three disruptants (T, M1, and M2) than that of the parental strains by a factor of 1.5 (p < 0.01), 2.7 (p < 0.001), and 2.0 (p < 0.001), respectively (Figure 5a,b). We examined the cell wall integrity of transformants by assessing the effect of cell-wall perturbing In addition, TEM examination showed the mean cell wall thickness to be significantly less for the three disruptants (T, M1, and M2) than that of the parental strains by a factor of 1.5 (p < 0.01), 2.7 (p < 0.001), and 2.0 (p < 0.001), respectively (Figure 5a,b). We examined the cell wall integrity of transformants by assessing the effect of cell-wall perturbing agents through the measurement of mycelial growth on PDA plates [50]. Growth of the T, M1, and M2 strains in contact with Congo red and SDS was significantly less than that of the WT and ∆KU70 strains (Figure 5c), supporting the role of PIG1 in cell-wall structure and integrity.

Cell Morphology, Germination and Biofilm Formation
In addition to the loss of the external melanin coating, conidia from the T and M1 and M2 recombinants showed significant changes in terms of shape and structure. Unexpectedly, the recombinant strains produced multicellular cross-sectioned spores with one or two septa, accounting for 35 to 43% of the total conidia (Figure 6a,c). The remaining spores from recombinants were similar in shape, although a portion appeared to be more elon-

Cell Morphology, Germination and Biofilm Formation
In addition to the loss of the external melanin coating, conidia from the T and M1 and M2 recombinants showed significant changes in terms of shape and structure. Unexpectedly, the recombinant strains produced multicellular cross-sectioned spores with one or two septa, accounting for 35 to 43% of the total conidia (Figure 6a,c). The remaining spores from recombinants were similar in shape, although a portion appeared to be more elongated than those of the parental strain, corresponding to the fraction of 1-cell conidia displaying a size comprised between 80 and 100µm (Figure 6b). By contrast, the shape of WT-TRC conidia remained unchanged (Figure 6a). The results suggest that the transcription factor PIG1 not only regulates genes of the DHN-melanin cluster but may also indirectly regulate the expression of other genes. WT-TRC conidia remained unchanged (Figure 6a). The results suggest that the transcription factor PIG1 not only regulates genes of the DHN-melanin cluster but may also indirectly regulate the expression of other genes. We also investigated germination and biofilm formation. Although the kinetics of germ-tube formation were slightly slower in the T strain, it was not hampered in the two ΔKU70-derived transformants relative to their parental strain (data not shown). Similarly, the ability of both M1 and M2 to adhere and grow on inert material appeared to be significantly enhanced with respect to the ΔKU70 strain (+40%, p < 0.05), as shown by the amount of biomass after 24 h of incubation on a polystyrene plate (Figure 7). By contrast, the growth of the T strain appeared to be less than that of the WT strain. The metabolic activity within the biofilms remained unchanged (data not shown). Of note, however, the mutant strains exhibited a similar ability to form biofilms and the differences observed We also investigated germination and biofilm formation. Although the kinetics of germ-tube formation were slightly slower in the T strain, it was not hampered in the two ∆KU70-derived transformants relative to their parental strain (data not shown). Similarly, the ability of both M1 and M2 to adhere and grow on inert material appeared to be significantly enhanced with respect to the ∆KU70 strain (+40%, p < 0.05), as shown by the amount of biomass after 24 h of incubation on a polystyrene plate (Figure 7). By contrast, the growth of the T strain appeared to be less than that of the WT strain. The metabolic activity within the biofilms remained unchanged (data not shown). Of note, however, the mutant strains exhibited a similar ability to form biofilms and the differences observed upon comparing each mutant to its parental strain appeared to be mainly based on different characteristics of the two parental strains.

Resistance to Exogenous Stress
Many studies have shown that melanized fungi are more resistant to stressful conditions than non-melanized species, enabling them to survive under extreme environmental conditions. Thus, we exposed the mutant and parental strains to deleterious conditions to determine whether unpigmented mutant conidia were more vulnerable to environmental stress. First, exposure of the strains to UV-C light showed the survival rate of M1 and M2 conidia to be only 35.6% and 39.4% that of unexposed conidia, respectively, whereas survival of the parental ΔKU70 strain was 73.6% relative to the unexposed controls ( Figure  8a). By contrast, there was no significant difference in the survival of the WT-derived T strain nor the TRC-treated WT strain relative to that of the parental WT strain. This may be related to the difference in susceptibility of the parental strains, as suggested by the three-fold lower viability of the WT strain relative to that of ΔKU70.
Moreover, although incubation at 40 °C for 15 min had no significant effect on conidia viability, the survival rate of the three mutants markedly dropped after exposure to 50 °C to 4.1% (p < 0.05), 9.1% (p < 0.01), and 5.9% (p < 0.001), respectively, whereas 47.5% to 62.8% of the parental strain conidia survived (Figure 8b). The survival of WT-TRC conidia was similar to the WT strain. Similarly, we also evaluated the ability of the T, M1, and M2 strains to grow in oxidizing medium by nephelometry. The three mutant strains showed a large reduction in growth in cumene hydroperoxide relative to growth in PDB (control medium): a reduction of 88.0% for the T (p < 0.001), 72.5% for the M1 (p < 0.01), and 56.8% for the T (p < 0.01) strains. By contrast, the parental strains were able to maintain more than 70% of their growth in the reference medium (Figure 8c). When exposed to menadione, the relative growth of the T strain was lower than that of the WT strain, which contrasted with the growth of both the M1 and M2 strains, which was slightly enhanced relative to that of the ΔKU70 strain (p < 0.05) (Figure 8d).

Resistance to Exogenous Stress
Many studies have shown that melanized fungi are more resistant to stressful conditions than non-melanized species, enabling them to survive under extreme environmental conditions. Thus, we exposed the mutant and parental strains to deleterious conditions to determine whether unpigmented mutant conidia were more vulnerable to environmental stress. First, exposure of the strains to UV-C light showed the survival rate of M1 and M2 conidia to be only 35.6% and 39.4% that of unexposed conidia, respectively, whereas survival of the parental ∆KU70 strain was 73.6% relative to the unexposed controls ( Figure 8a). By contrast, there was no significant difference in the survival of the WTderived T strain nor the TRC-treated WT strain relative to that of the parental WT strain. This may be related to the difference in susceptibility of the parental strains, as suggested by the three-fold lower viability of the WT strain relative to that of ∆KU70.
Moreover, although incubation at 40 • C for 15 min had no significant effect on conidia viability, the survival rate of the three mutants markedly dropped after exposure to 50 • C to 4.1% (p < 0.05), 9.1% (p < 0.01), and 5.9% (p < 0.001), respectively, whereas 47.5% to 62.8% of the parental strain conidia survived (Figure 8b). The survival of WT-TRC conidia was similar to the WT strain. Similarly, we also evaluated the ability of the T, M1, and M2 strains to grow in oxidizing medium by nephelometry. The three mutant strains showed a large reduction in growth in cumene hydroperoxide relative to growth in PDB (control medium): a reduction of 88.0% for the T (p < 0.001), 72.5% for the M1 (p < 0.01), and 56.8% for the T (p < 0.01) strains. By contrast, the parental strains were able to maintain more than 70% of their growth in the reference medium (Figure 8c). When exposed to menadione, the relative growth of the T strain was lower than that of the WT strain, which contrasted with the growth of both the M1 and M2 strains, which was slightly enhanced relative to that of the ∆KU70 strain (p < 0.05) (Figure 8d). J. Fungi 2023, 9, x FOR PEER REVIEW 15 of 28 Figure 8. Susceptibility of the T and the M1 and M2 deleted strains exposed to exogenous stresses. (a) Conidia survival after exposure to UV-C light. After exposure to a dose of 30 mJ/cm 2 at 254 nm, conidia were stained with 25 µ M propidium iodide (PI) before being analyzed on a LSRFortessa TM cytometer. The graph presents the percentage survival among exposed FITC + conidia expressed relative to that of unexposed FITC + conidia (live control), as follows: 100× ((% PI -FITC + conidia among unexposed control conidia-% PI -FITC + conidia among exposed conidia)/(% PI -FITC + conidia among unexposed control conidia)).

Antifungal Susceptibility
The fungal cell wall, composed of molecules not present in humans, constitute an ideal target for the development of clinical antifungal compounds. Therefore, modifications of the cell wall may influence responses to antifungal drugs. The susceptibility of mutants to various antifungals was assessed by the broth microdilution method according to EUCAST recommendations (Table 1). Amphotericin B had a deleterious impact on the growth of both the M1 and M2 recombinants, shown by the lower mean MICs of 0.58 ± 0.38 and 0.83 ± 0.29 µg/mL, respectively, relative to that of for ∆KU70, which was resistant to AmB (MIC > 8 µg/mL). By contrast, both the T and WT strains showed similar MICs of approximately 0.5 µg/mL. These results were further supported by E-test susceptibility testing (not shown). The susceptibility to triazole compounds and anidulafungin were unchanged between the parental strains and their transformants ( Table 1).

Transcriptomic Analysis of M1 Transformant
We next conducted a comparative transcriptomic analysis between the MI strain and its parental ∆KU70 strain to understand the involvement of the transcription factor PIG1 in melanin biosynthesis and the phenotypic characteristics described above, especially relative to the morphogenesis of the S. apiospermum cell wall. Mycelia were harvested from YPDA medium after 30 h of incubation and RNA extracted and purified for transcriptome sequencing. In total, eight biological samples (four biological replicates) were thus sequenced by next-generation sequencing. After a stringent quality check and data cleaning, 9.4 to 15 million assigned reads were obtained for each sample (Table S2), which were further spliced into 10,919 assembled unigenes.
In addition to biostatistical quality controls, RNA-seq based differential gene expression data were further validated by RT-qPCR for eight selected representative genes, resulting in an excellent agreement between the fold change values determined by global sequencing and specific qPCR ( Table 2).
In total, 416 differentially expressed genes (DEGs) were obtained for M1 relative to ∆KU70, regardless of the fold change value (FC). Considering only DEGs with a |log 2 FC| > 1 (adjusted p-value < 0.05) resulted in 278 DEGs, including 78 pseudogenes, of which 115 were upregulated and 163 downregulated. Only 110 genes (39.6%) were assigned with GO terms, including biological processes (n = 67), molecular functions (n = 107), and cellular components (n = 41). In addition, we performed a functional analysis of the resulting proteins using Interproscan, making it possible to associate a function or metabolic pathway with 198 DEGs (71.2%), of which 83 were upregulated and 115 downregulated (Figure 9a). Among DEGs with a known function found in M1, the five most represented functions were associated with oxidoreduction processes (N = 43), transmembrane transport (N = 23), carbohydrate metabolic processes (N = 16), transcription factors (N = 16), and methyltransferases (N = 16) (Figure 9a). The significant representation of oxidoreduction enzymes among DEGs in the M1 transformants could at least partially explain some of the phenotypic observations with the recombinant. For example, a number of key proteins involved in conidial protection against oxidative injuries, such as peroxidases, catalases, or glutathione enzymes, appeared to be differentially expressed (Figure 9b). Dysregulated enzymes were also involved in cell-wall metabolism included, for example, various glycosidases, a polysaccharide synthase (log2FC = −1.2, p = 9.6 × 10 −5 ) and a neutral/alkaline non-lysosomal ceramidase (log2FC =-1.45, p = 0.038). Other DEGs potentially related to conidiogenesis were also identified in M1, such as the genes encoding the membrane protein SUR7/RIM9-like (log2FC = 2.09, p = 0.0024) and conidiation-specific protein-6 (log2FC = −3.18, p = 0.030) (data not shown).   (Figure 9a). The significant representation of oxidoreduction enzymes among DEGs in the M1 transformants could at least partially explain some of the phenotypic observations with the recombinant. For example, a number of key proteins involved in conidial protection against oxidative injuries, such as peroxidases, catalases, or glutathione enzymes, appeared to be differentially expressed (Figure 9b). Dysregulated enzymes were also involved in cell-wall metabolism included, for example, various glycosidases, a polysaccharide synthase (log 2 FC = −1.2, p = 9.6 × 10 −5 ) and a neutral/alkaline non-lysosomal ceramidase (log 2 FC =-1.45, p = 0.038). Other DEGs potentially related to conidiogenesis were also identified in M1, such as the genes encoding the membrane protein SUR7/RIM9-like (log 2 FC = 2.09, p = 0.0024) and conidiation-specific protein-6 (log 2 FC = −3.18, p = 0.030) (data not shown).
The expression of genes selected for the qPCRs performed to validate the M1 RNA-seq data was also explored in the M2 and T strains ( Figure 10). Interestingly, the scytalone dehydratase-encoding gene was the only melanin biosynthesis-related gene found to be downregulated in all three transformants. The repression of this key-enzyme from the melanin biosynthesis cascade supports the melanin-deficient phenotype observed for the three recombinants. The expression of genes selected for the qPCRs performed to validate the M1 RNAseq data was also explored in the M2 and T strains ( Figure 10). Interestingly, the scytalone dehydratase-encoding gene was the only melanin biosynthesis-related gene found to be downregulated in all three transformants. The repression of this key-enzyme from the melanin biosynthesis cascade supports the melanin-deficient phenotype observed for the three recombinants.

In Vitro Macrophage Infection Assays
The first step of the host immune response to fungal conidia has been shown to be highly dependent on the early recognition of cell-wall specific patterns by phagocytic cells. Thus, the objective of these experiments was to assess the response of human macrophages exposed to conidia from the three melanin-deficient transformants. In particular, a phagocytosis assay was conducted to determine whether cell-wall sugar unmasking was associated with easier recognition, binding, and subsequent conidia internalization by macrophages. Both M1 and M2 recombinants showed a higher ability to adhere to and Figure 10. Fold-induction mRNA values for a selected panel of genes in the T, M1, and M2 strains relative to their parental ∆KU70 strain. Expression levels of target genes were normalized against the expression of the S. apiospermum UBC6 and FIS1 housekeeping genes. Results are expressed as 2 −∆∆Cq , referring to the fold induction in the T, M1, and M2 strains relative to the mean quantification cycle obtained with the parental strains. Data are from four biological replicates: * p ≤ 0.05 (Mann-Whitney test). T: transformant derived from the WT strain; M1 and M2: transformants derived from the ∆KU70 strain.

In Vitro Macrophage Infection Assays
The first step of the host immune response to fungal conidia has been shown to be highly dependent on the early recognition of cell-wall specific patterns by phagocytic cells. Thus, the objective of these experiments was to assess the response of human macrophages exposed to conidia from the three melanin-deficient transformants. In particular, a phagocytosis assay was conducted to determine whether cell-wall sugar unmasking was associated with easier recognition, binding, and subsequent conidia internalization by macrophages. Both M1 and M2 recombinants showed a higher ability to adhere to and be internalized by macrophages, as at 6 h post-infection, the percentage of FITC + macrophages infected with M1 and M2 reached 60.5% (p < 0.001) and 55.1% (ns), respectively, vs. only 46.5% of ∆KU70 infected cells (Figure 11a). Different results were obtained with the WT-derived strains, as the percentages of infected macrophages were significantly lower at 2 and 6 h when challenged with the T transformant and the TRC-treated WT strain than with the WT strain: 34.3% (p < 0.05) and 25.4% (p < 0.001) vs. 49.3% 2 h post-infection. Although the phagocytosis assay resulted in contrasting conclusions between the WT and ∆KU70-derived strains, the ability of recombinant conidia to survive macrophage killing was equally impeded, as shown by the decrease in conidia viability of approximately 50% for the three transformants (48.2-56%, p < 0.001 vs. 35.7% and 15.2% for WT and ∆KU70) (Figure 11b). By contrast, conidia survival of the TRC-grown WT strain was unchanged.
J. Fungi 2023, 9, x FOR PEER REVIEW 20 of 28 Figure 11. Phagocytosis, killing, and cytokine release from human macrophages infected with T, M1, and M2 conidia. (a) Percentage of macrophages with adherent or ingested conidia after 2 or 6 h of incubation with FITC-stained conidia (MOI 5:1) at 37 °C in 5% CO2. Graphs present the percentage of FITC + CD11b + events among CD11b + cells (macrophages). (b) Conidia killing after ingestion by macrophages. After 6 h of incubation (MOI 1:1), extracellular conidia were removed by three PBS washes and internalized conidia were released by macrophage lysis (cold water). The graph presents the decrease in conidia viability among ingested conidia relative to the viability of conidia not exposed to macrophages (live control). Conidia viability was determined as the percentage of FITC + PIevents among FITC + events (total conidia). The reduction in conidia survival was calculated relative to control conidia as: 100*((% of live conidia among control conidia-% of live conidia among cell-exposed conidia)/% of live conidia among control conidia). (c) Cytokine release in culture supernatants from macrophages incubated with conidia for 12 h (MOI 10:1). Concentrations were determined using a 13 plex-bead-based LEGENDplex ® multianalyte flow assay. A non-infected (NI) condition was included in each experiment. Data are from three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001 (Kruskall-Wallis test) T: transformant derived from the WT strain; M1 and M2: transformants derived from the ΔKU70 strain.

Discussion
The relevance of DHN-melanin in fungal virulence has been increasingly assessed in recent decades for both plant and human pathogens. In addition to protecting fungal Figure 11. Phagocytosis, killing, and cytokine release from human macrophages infected with T, M1, and M2 conidia. (a) Percentage of macrophages with adherent or ingested conidia after 2 or 6 h of incubation with FITC-stained conidia (MOI 5:1) at 37 • C in 5% CO 2 . Graphs present the percentage of FITC + CD11b + events among CD11b + cells (macrophages). (b) Conidia killing after ingestion by macrophages. After 6 h of incubation (MOI 1:1), extracellular conidia were removed by three PBS washes and internalized conidia were released by macrophage lysis (cold water). The graph presents the decrease in conidia viability among ingested conidia relative to the viability of conidia not exposed to macrophages (live control). Conidia viability was determined as the percentage of FITC + PIevents among FITC + events (total conidia). The reduction in conidia survival was calculated relative to control conidia as: 100 × ((% of live conidia among control conidia − % of live conidia among cell-exposed conidia)/% of live conidia among control conidia). (c) Cytokine release in culture supernatants from macrophages incubated with conidia for 12 h (MOI 10:1). Concentrations were determined using a 13 plex-bead-based LEGENDplex ® multianalyte flow assay. A non-infected (NI) condition was included in each experiment. Data are from three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001 (Kruskall-Wallis test) T: transformant derived from the WT strain; M1 and M2: transformants derived from the ∆KU70 strain.
The inflammatory macrophage response to the PIG1-deficient strains was also indirectly evaluated by measuring the level of various pro-inflammatory cytokines released within 12 h post-infection. Although there was a trend toward increasing IL-1β release by M1-and M2-infected macrophages, the IL-18 concentration increased by a factor of 1.9 (p < 0.05), 5.0 (p < 0.001), and 6.1 (p < 0.001) for macrophages infected with the T, M1, and M2 transformants, respectively, relative to the parental strain, (Figure 11c). However, there was no significant change in the level of IL-6, TNFα, IL-12p70, IL-17A, IL-23, or IL-10 released by macrophages infected by any of the mutants (Figure 11c).

Discussion
The relevance of DHN-melanin in fungal virulence has been increasingly assessed in recent decades for both plant and human pathogens. In addition to protecting fungal structures from environmental stresses, the melanin layer of the conidia cell wall has been recognized in certain genera to be an effective camouflage against the host immune system. In certain phytopathogens in which DHN-melanin has been extensively explored as a key-component of invasion organs, called appressoria [48], melanin biosynthesis involves several enzymes under the regulation of conserved and mostly clustered genes. In particular, in Magnaporthe oryzae [31], Colletotrichum gloeosporioides [51], Cochliobolus heterostrophus [32], and Alternaria brassicicola [14], a transcription factor has been identified within the cluster and numerous studies have shown its activating role in melanization, which is subsequently related to virulence.
As there are only limited data on the intrinsic factors related to S. apiospermum pathogenicity, we aimed to assess the role of the transcription factor PIG1 in the biosynthesis of DHN-melanin in S. apiospermum conidia. As an ortholog gene encoding the PIG1 transcription factor was identified in the S. apiospermum genome, a CRISPR-Cas9-mediated PIG1 deletion was conducted on two melanized S. apiospermum strains, the WT reference 14462 strain and a ∆KU70 deleted strain derived from the reference strain. Three melanindeficient recombinants displaying white colonies were obtained, although the complete PIG1 deletion was obtained only in the two ∆KU70-derived mutants. Unexpectedly, the 14462 WT-derived T strain integrated the repair template upstream of PIG1. This may be explained by (i) an ectopic insertion in this area or (ii) by a the HPH insertion by a micro-homology-end joining mechanism as recently described in Magnaporthe oryzae [52]. This mutant T strain was indeed able to grow similarly in the presence of hygromycin as the ∆KU70-derived mutants, and the function of PIG1 was apparently hampered, as shown by the absence of melanin in the conidia. However, the increased production of PIG1 mRNA observed suggests that a gene-silencing mechanism may occurred by transcriptional/posttranscriptional interference [53].
To assess the role of DHN-melanin in S. apiospermum, some experiments included a control tricyclazole (TRC)-treated WT strain acting on specific reductases from the DHNmelanin biosynthesis. Indeed, TRC-treated colonies appeared lightly pigmented due to the accumulation of melanin shunt intermediates. Thus, even if there were some slight differences with albino strains, this condition generated melanin-lacking spores. This comparison allowed us to confirm the involvement of the melanin in some observed phenotypes, and to reveal additional melanin-independent phenotypes, possibly related to PIG loss. Besides, we analyzed the global transcriptome of M1 relative to its parental strain ∆KU70 to identify differentially expressed genes (DEGs) that could explain the melanin-independent phenotypes observed.
For the first time, the phenotype of melanin-lacking spores of S. apiospermum was deeply explored, emphasizing the major structural role of DHN-melanin in the conidia cell wall. It is commonly thought that DHN-melanin plays a role of "fungal armor" considering its external location in the cell wall. This has been supported by the increase of polysaccharide exposure in conidia from the three transformants together with TRCtreated strain. The structural disorganization may be associated with the reduced cell wall thickness highlighted in T, M1 and M2 conidia. These results are in line with a previous report revealing the drastic decrease in the cell wall thickness of Alternaria alternata deficient conidia, following pyroquilon-mediated DHN-melanin inhibition [54]. Similarly, the cell wall of TRC-treated hyphae from another phytopathogen fungus Gaeumannomyces graminis (Magnaporthaceae family) was thinner than the untreated strain [55].
Moreover, several unexpected phenotypes were observed in the PIG1 recombinants that cannot definitely be the consequence of melanin-deficiency but result from the absence of functional PIG1.
The conidia morphology of the three recombinants appeared to be equally affected, showing aberrant septate and large conidia, contrasting with the regular and ovoid unicellular spores released by the parental strains and the tricyclazole-treated strain. No change in conidia shape has previously been associated with melanin deficiency due to the specific disruption of genes encoding biosynthetic enzymes, such as polyketide synthase or scytalone dehydratase. This observation has also never been reported in mutants with the deletion of PIG1 orthologs from fungi producing monocellular conidia [51,52]. By contrast, in some genera that normally produce pluricellular conidia, a reduced number of conidia septations has previously been reported, such as for conidia from CMRA-deleted Alternaria alternata [56] or tricyclazole-exposed Bipolaris [57].
The complexity of the cell-wall architecture, which is still partially unresolved, did not allow us to definitively identify the differentially expressed gene(s) responsible for the presence of compartmentalized conidia. Although the process or functions could not be determined for all DEGs, several important enzymes dedicated to processes associated with cell-wall metabolism were overexpressed in M1 relative to WT, such as a neutral/alkaline ceramidase. The fungal cell wall is a dynamic and complex structure in which polysaccharides, oligosaccharides, lipids, and proteins are intertwined [24]. For the first time, the transcriptomic analysis of a ∆PIG1 fungal strain was performed. Interestingly, a significant number of genes encoding hydrolases, glycosyltransferases, and other enzymes involved in carbohydrate metabolic processes were differentially regulated in the M1 recombinant. In Scedosporium, glycosphingolipids are components of the plasma membrane, the cell wall, and extracellular vesicles involved in fungal growth and virulence [26,58]. In particular, glucosylceramides (GlcCer) are major neutral sphingolipids bound to the fungal cell surface, characterized by a fatty-acid chain bound to a sphingosine and a sugar unit [59] and highly conserved among pathogenic fungi, such as Candida albicans, Cryptococcus neoformans, and Aspergillus nidulans [60][61][62]. The crucial role of GlcCer in fungal physiology, i.e., in cell division, growth, alkaline tolerance, the promotion of virulence, and the modulation of the host immune response [63], and their unique structure compared to the components of mammalian cells, make them interesting targets for new antifungal strategies [25,64,65]. Although there are some data on the essential role of GlcCer in fungal homeostasis, there is no data on the specific role of the neutral and alkaline ceramidases that regulate ceramide and sphingosine levels, a fortiori in S. apiospermum. Thus, the overexpression of a neutral/alkaline ceramidase resulting from PIG1 inactivation in the ∆PIG1 recombinant may hinder normal physiology, but this hypothesis is yet to be specifically addressed. Finally, little is known about the function of conidiation specific protein 6 (CON6), which was downregulated in the two ∆PIG1 recombinants. Initially described in Neurospora crassa, a filamentous fungus used as a model for studies on developmental processes, the transcription of this regulatory gene increases during conidiation and following exposure of vegetative mycelia to light [66,67]. However, its exact function and association with the phenotype of the PIG1 mutants are still unknown.
In addition, PIG1 is involved in the protection of conidia heat stress as revealed by the higher susceptibility of ∆PIG1 disruptants to heat relative to their parental strains. Similar observations were previously made in Verticillium dahliae with a deletion of CMR1, an ortholog of PIG1, when exposed to heat stress [23,68].
In recent years, our understanding of host-fungi interactions has led to a great interest in deciphering the role of surface cell-wall components in fungal virulence, with specific attention on DHN-melanin. In Scedosporium species, polysaccharides and peptidopolysac-charides are especially relevant for the architecture of the conidia cell wall and several are immunologically active, with great potential as regulators of pathogenesis and the host immune response. For example, a significant reduction in both phagocytosis and killing of S. apiospermum conidia was demonstrated using anti-peptidoramnomannan mAbs in murine macrophage-like cells relative to control conidia [69]. Here, the removal of DHNmelanin resulted in equally higher exposure of underlying cell-wall polysaccharides in the three mutants and the TRC-treated WT strain, potentially increasing the availability of surface antigenic patterns and conidia interaction with phagocytic cells. Thus, we assessed the in vitro response of human macrophages to infection by PIG1 recombinants relative to their parental strains. However, the uptake (adherence) of T, WT-TRC, M1 and M2 conidia by macrophages relative to their parental strain let to heterogeneous findings between WT and ∆KU70-derived strains, consistent with the contradictory results of previous studies conducted on melanin-deficient fungi [70][71][72]. Then, we showed that the conidial killing of the three melanin-deficient recombinants was equally enhanced, which unexpectedly contrasted with the unchanged survival of WT-TRC conidia. Together with the enhanced macrophage release of IL-18 following infection with T, M1 M2 but not with WT-TRC conidia, these results suggest the role of PIG1 in conidia survival upon macrophage interaction, rather than the direct involvement of melanin.
Other phenotypes were described, without establishing if they were related to the loss of melanin or to the absence of functional PIG1.
We showed a reduced growth of PIG1 recombinants incubated with cell wall stressors (Congo Red, SDS) and oxidative chemicals, in particular cumene. Such increased vulnerability may result from a combination of several mechanisms. First, the radical structure of DHN-melanin provides free radical scavenging properties, which could explain why, in addition to the loss of cell-wall integrity, the three melanin-deficient mutants were less resistant to oxidizing molecules. Another important point is that fungi, Scedosporium in particular, possess a considerable enzymatic arsenal capable of detoxifying numerous toxic compounds, especially those that induce oxidative stress [73,74]. In the M1 recombinant we highlighted the dysregulation of various enzymes that play a role in oxidoreduction processes, such as the protection against oxidative stress, for example, through glutathione metabolism. It is therefore likely that certain overexpressed enzymes, such as thioredoxin reductase, shown in the M1 disruptant, play a role in the detoxification of menadionegenerated superoxides.
DHN-melanin has been commonly associated with fungal protection against stress that triggers cell damage, such as electromagnetic radiation and oxidants [12,[75][76][77]. This is what we observed in the ∆KU70 strain exposed to UV-C light relative to both the ∆KU70derived M1 and M2 recombinants. By contrast, the WT strain showed much lower UV resistance than its derived ∆KU70 strain and paradoxically, its melanin-deficient derived recombinant tended to be more resistant to UV light. These results prevent consistent conclusions to be drawn regarding the role of PIG1 orthologs and DHN-melanin in S. apiospermum protection against UV exposure, in line with previous contradictory findings varying according to fungal models [14,29,51].
The melanization of the fungal cell wall has also previously been linked to antifungal resistance, especially for C. neoformans, Histoplasma capsulatum [21], and Exophiala dermatitidis against AmB [23] or Madurella mycetomatis against azoles [78]. Here, the susceptibility to triazoles and anidulafungin was unchanged by PIG1 deletion in S. apiospermum, and the M1 and M2 disruptants sporadically recovered AmB susceptibility, relative to the parental ∆KU70. Thus, we were not able to draw any conclusions about the role of melanin in S. apiospermum antifungal susceptibility. Of note, the mechanism explaining the lower susceptibility of the ∆KU70 strain to AmB than the 14462 WT strain is still unclear. Nonetheless, independently of DHN-melanin, the restored susceptibility of the ∆KU70-derived recombinants to AmB may be linked to the downregulation of certain key-antioxidative enzymes, such as catalases, as shown by the transcriptomic analysis. Indeed, the AmB resistance of Scedosporium/Lomentospora has been recently shown to be associated with the ability to thwart the cellular redox imbalance triggered by AmB through potent activity of antioxidative enzymes [79].
Finally, the heterogeneous phenotype between the two melanized parental strains observed for some assays, in particular those dedicated to evaluating UV resistance, biofilm formation and antifungal susceptibility, could be possibly related to the higher melanin content in ∆KU70 conidia, as well as the absence of the KU70 protein itself [80,81]. Indeed, although the benefit of KU70 deletion is widely recognized to facilitate genetic transformation in fungi, the loss of this component of the nonhomologous end-joining pathway involved in DNA double-strand break repair could also have an impact on other physiological processes, which has never been specifically explored.

Conclusions
Overall, we demonstrate that the transcription factor PIG1 regulates the DHN-melanin biosynthetic pathway in S. apiospermum through the construction of PIG1-deleted strains, which displayed phenotypic features similar to previously described PKS-disrupted fungi. Supported by the transcriptomic analysis, this study highlights the pleiotropic action of PIG1 in S. apiospermum beyond melanin regulation, including on the cell-wall architecture and enzymatic pathways. Although, we were unable to determine whether certain phenotypic observations were attributed directly to the loss of the melanin layer or due to a collateral effect of PIG1 loss of function, we provide evidence for the involvement of PIG1/melanin in the protection of conidia against environmental damages and macrophage killing, and possibly in the pathogenicity of Scedosporium.
This study provides new insights on the melanization process of S. apiospermum and paves the way for further research on DHN-melanin as a potential therapeutic target in a fungus with low susceptibility to most current antifungal options.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/jof9020134/s1, Figure S1: Fluorescence labelling of conidia surface carbohydrates with FITC-conjugated lectins; Table S1: Nucleotide sequence of constructs and primers used in the study; Table S2: Summary of RNA-Seq quality data of M1 and its ∆KU70 parental strain. Four biological repeats were performed for each isolate.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.