Deletion C-terminal thioesterase abolishes melanin biosynthesis, affects metabolism and reduces the pathogenesis of Fonsecaea monophora

Dematiaceous Fonsecaea monophora is one of the major pathogens of chromoblastomycosis. It has been well established that melanization is catalyzed by the type I polyketide synthase (PKS) in F. monophora. Multidomain protein Type I PKS is encoded by six genes, in which the last enzyme thioesterase (TE) catalyzes the cyclization and releases polyketide. Two PKS genes AYO21_03016 (pks1) and AYO21_10638 have been found in F. monophora and both PKS loci have the same gene arrangement but the TE domain in AYO21_10638 is truncated at 3’- end. TE may be the key enzyme to maintain the function of pks1. To test this hypothesis, we constructed a 3’-end 500 bp deletion mutant of AYO21_03016 (Δpks1-TE-C500) and its complemented strain. We profiled metabolome of this mutant and analyzed the consequences of impaired metabolism in this mutant by fungal growth in vitro and by pathogenesis in vivo. Compared with wild-type strain, we found that the mutant repressed pks1 expression and other 5 genes expression levels were reduced by more than 50%, perhaps leading to a corresponding melanin loss. The mutant also reduced sporulation and delayed germination, became vulnerable to various environmental stresses and was less resistance to macrophage or neutrophil killings in vitro, and less virulence in mice footpad model. Metabolomic analysis indicated that many metabolites were remarkably affected in Δpks1-TE-C500, in particular, an increased nicotinamide and antioxidant glutathione. In conclusion, we confirmed the crucial role of C-terminal TE in maintaining fully function of pks1 in F. monophora. Deletion of TE negatively impacts on the synthesis of melanin and metabolites that eventually affect growth and virulence of F. monophora. Any potential inhibitor of TE then could be a novel antifungal target for drug development.


Introduction
Chromoblastomycosis (CBM) is a chronic granulomatous infection caused by dematiaceous fungi invading the skin and subcutaneous tissue, which can infect healthy people in tropical and temperate zones. The disease is endemic worldwide and is considered as a neglected tropical disease (NTD) in 2017 [1]. F. monophora is one of the most common pathogens for chromoblastomycosis. Recently, we have reanalyzed the ITS rRNA sequences of causative strains previously collected in southern China and found that more than 80% of morphologically identified Fonsecaea pedrosoi in the past are actually F. monophora [2]. A number of studies have demonstrated that F. monophora exhibits neurotropism or causes primary brain infection without skin damage [3,4].
Little is known about the pathogenic traits of F. monophora and their impact in host immune defense response. Studies have shown that melanin on the fungal cell wall is a vital virulence factor in dematiaceous fungi because of its distinct chemical and biological attributes such as negative charge, hydrophobicity, acid resistance, and heat resistance [5]. The ability of resistance to external physical and chemical stresses confers melanin a protection from the host defense response, which has been confirmed with melanin-deficient strains of Cryptococcus neoformans [6], F. pedrosoi [7] and Exophiala dermatitidis [8]. In our previous study, we also found that albino F. monophora CBS125149 was more sensitive to environmental stresses and less virulent in mice model, compared with pigmented F. monophora CBS122845 [9,10].
Since the albino strain can only be obtained by fully deletion of entire AYO21_03016 gene, but not AYO21_10638 [13], we thus speculated that AYO21_03016 (pks1) is the dominate gene for melanin synthesis in F. monophora. Subsequently, we compared the gene complex structures between AYO21_03016 and AYO21_10638, and found that the size of the TE gene was different between two gene complexes. As shown in S1 Fig, TE gene of AYO21_03016 is longer at its 3'-end than AYO21_10638, which leaves a protein with 132 amino acids truncation at Cterminus of TE in AYO21_03106. TE domain catalyzes the last step of polyketides synthesis for polyketide cyclization and release through a canonical catalytic triad consisting of Ser-His-Asp residues [14]. Sequence analysis revealed that the TE domain in AYO21_03016, likes many other PKS, contains these three residues at S72, D106 and H231 (S2 Fig), but the truncated TE in AYO21_10638 has no D106 and H231 residues, which can make TE loss of function. Therefore, TE domain may be necessary to maintain the function of PKS.
Polyketide products possess diverse architectures and biological functions. PKS biosynthesis even shares several steps with fatty acid synthesis, in which TE is an important link between PKSs and fatty acid synthases [15]. As PKS is also involved in the melanin and metabolite synthesis pathway, it is speculated that TE on pks1 may also affect the resistance of F. monophora to external stress as well as virulence, and could be a potential drug target for F. monophora infection. In this study, we aimed to investigate the function of TE by characterizing the TE gene on pks1 and exploring the mechanism of the host's immune response to F. monophora.

Ethics statement
This study protocol was approved by the Ethics committee of the Dermatology Hospital, Southern Medical University (ethics amendment dated 7 March 2018, approval number 2018002. All experiments in this study were conducted according to internationally accepted standards and regulations on the administration of experimental animals in China (8/1/2011 C-WISC).

Strains, growth conditions, targeted gene deletion, and complementation analysis
Strains and plasmids used in this study are listed in Table 1. F. monophora SUMS0310 (CBS269.37) was used as the wild-type (WT) strain in all experiments. Transformation was performed using the Agrobacterium tumefaciens-mediated transformation (ATMT) method [16]. The Δpks1-TE-C500 was generated by transforming F. monophora with plasmids to delete the terminal 500 bp region of the pks1, followed by selection with hygromycin (Fig 1A). The complemented Δpks1-Com strain was generated by transforming the Δpks1-TE-C500 with the pks1-TE-C-neo-pBHt2 plasmid and selecting for geneticin-resistant transformants (Fig 1B). The plasmids used for pks1 TE targeted deletion and complementation, pks1 TE-hph-pBHt2 and pks1-TE-C-neo-pBHt2, were constructed by In-Fusion HD Cloning Kit (Takara, Japan) according to the manufacturer's instructions. For pks1-TE-hph-pBHt2 vector, PCR was used to obtain the four fragments: the 326 bp fragment upstream (primers 1 and 2) (  Table) downstream of the F. monophora AYO21_03016 gene. The fragment containing XhoI cleavage site was inserted into SacII and AsiSI-restricted pBHt2 plasmid using In-Fusion HD Cloning Kit to obtain pBHt2-XhoI. As for pks1-TE-hph-pBHt2 vector, two AYO21_03016 gene homologous flank ranges were cloned to pBHt2-XhoI by the similar strategy. To construct TE complementation vector (pks1-TE-Cneo-pBHt2), fragment containing geneticin resistance gene was synthesized and inserted into XhoI and AflII-restricted pks1-TE-hph-pBHt2 by in-fusion method to generate pks1-TE-neo-pBHt2. PCR was used to obtain the 1300 bp fragment containing the 500 bp deletion coding sequence (primers pks1-Com arm3. FOR and pks1-Com arm3. REV), and inserted into SacI and BamHI-restricted pks1-TE-neo-pBHt2 by in-fusion method to obtain pks-TE-C-neo-pBHt2. PCR method was used to identify knockout or complemented pks1 TE transformants with the primers (pks1-TE. FOR and pks1-TE. REV) (S1 Table). Primers pks1-TE. FOR and pks1-TE. REV are located at the left and right flank regions adjacent to the pks1 TE homology respectively. The negative transformant was about 3.1 kb, the positive transformants were confirmed by 4.5 kb amplified fragment for Δpks1-TE-C500 (Fig 1A), and 4.3 kb PCR fragment for complemented transformants (Fig 1B). Sanger sequencing also confirmed the transformants. The whole gene deletion strain Δpks1 was also performed with the same strategy described above to compare the efficacy of Δpks1-TE-C500 transformation. The overall transformation efficiency of Δpks1-TE-C500 was higher than Δpks1 (S3 Fig).

RNA isolation and real time PCR
Total RNA of each sample was extracted using Trizell Reagent (Invitrogen, USA). Then, the RNA was reverse transcribed using RevertAid Master Mix (catalog no.M16325, Thermo, USA). Quantitative real-time reverse transcription-PCR (RT-qPCR) was carried out on the BioRad instrument (BioRadCFX96Touch, USA) by using PowerUp SYBR Green Master Mix (Thermo, USA), and the primers used are listed in S1 Table.

Metabolomics analysis
All strains were cultured on potato dextrose agar (PDA) at 25˚C for 14 days. Subsequently, the strains were collected, and metabolites were extracted with 50% methanol buffer. In brief, 100 mg of sample was extracted with pre-cooled 50% methanol, homogenized for 1 min, and incubated at room temperature for 10 min; the extraction mixture was then stored overnight at -20˚C. After centrifugation at 4,000 g for 20 min, the supernatants were transferred to new 96-well plates. The samples were stored at -80˚C prior to Liquid chromatography-mass Schematic diagram of gene deletion. The resistance gene (hph) replacement strategy was used to destruct the 500 bp C-terminal of TE in F. monophora. Through homologous recombination, overlapping DNA fragments of the hygromycin resistance gene cassette (hph) were used for gene replacement. PCR identification analysis of the wild-type and deletion mutant strains. (B) Schematic diagram of gene complementation. Another resistance gene (neo) replacement strategy was used to complement the 500 bp C-terminal of TE in F. monophora. Through homologous integration, overlapping DNA fragments of the neomycin resistance gene cassette (neo) were used for gene replacement. PCR identification analysis of the wild-type and complementation strain. (C) Colony morphology and microscopic morphology examination of the wild-type and mutant strain. (D) The mRNA expression of conserved functional domains of AYO21_03016 protein in F. monophora were detected, including starter ACP transacylase (SAT), β-ketoacyl synthase (KS), acetyl transferase (AT), dehydratase (DH), acyl carrier protein (ACP) and thioesterase (TE). TE gene expression was negative in Δpks1-TE-C500 and other 5 genes (SAT, KS, AT, DH, ACP) were significantly less expressed than WT. All statistical analysis were performed using twotailed t-test ( �� , P<0.01; ��� , P<0.001; ND means not detected, relative mRNA expression = 0). All the assays were performed in triplicate.
https://doi.org/10.1371/journal.pntd.0010485.g001 spectrometry (LC-MS) analysis. In addition, pooled Quality control (QC) samples were prepared by combining 10 μl of each extraction mixture. The identification and quantitative analysis of metabolites were carried out as described by Yukuo Li et al. [17].

Fungal growth, stress resistance
The growth rate and spore-producing ability of the mutants and WT were grown in PDA medium for 14 days at 25˚C and 37˚C, and radial growth was measured every 3 days over a period of 14 and 24 days. For each strain, 5 μl of conidial suspension in PBS containing 1×10 6 conidia/mL was inoculated. Conidial number was counted as the number of spores per mm 2 after 14 days. Germination was examined with initial spores (1×10 6  . Formation of melanin was tested by supplementation of PDA with 25 mg/l L-DOPA or 25 mg/l L-tyrosine in all strains. Concentrations of L-DOPA, L-tyrosine and tricyclazole were selected according to previous research [19]. The cultures were kept in the dark to withstand the autopolymerization of L-DOPA at 25˚C for 14 days.

Evaluation of macrophage and neutrophil killing ability
RAW264.7 macrophages were infected with F. monophora conidia (multiplicity of infection (MOI) = 10:1, 10 conidia per cell) for 24 h at 37˚C. The cells were incubated at 37˚C with 5% CO 2 for 24 h. Then, sterilized double-steamed water (1 mL) was used to release the spores phagocytosed by the cells. The diluted suspension was transferred to PDA plates and incubated at 25˚C, and the number of colonies on each plate was counted after 7 days. For microscopy, infected macrophages were either fixed and prepared for Transmission electron microscopy (TEM) or stained with 5 mM Dil for 10 min, washed twice with PBS, fixed in 4% paraformaldehyde, and stained with calcofluor to examine morphogenesis. For TEM, cells were fixed with 2.5% glutaraldehyde for 2 h at 4˚C. Then, they were washed in 0.1 mol/L PBS six times for 30 min each time, fixed with 1% osmic acid for 2 h, washed in 0.1 mol/L PBS three times for 10 min each time, and ethanol-dehydrated by sequential washing in 30%, 50%, 70%, 80%, 90%, 100% ethanol, 100% ethanol: 100% acetone (1:1), and 100% acetone for 10 min each time. The samples were embedded in white resin, and thin sections were examined with a JEM-1400 PLUS Transmission Electron Microscope (Japan Electron Optics Laboratory Co., Ltd., Japan). Human neutrophils were isolated using PolymorphPrep density gradient media (1114683; PROGEN BioTEchnik GmbH, Heidelberg, Germany) according to the manufacturer's protocol. Neutrophil fungicidal assay was performed with reference to a previous study [20]. In brief, purified neutrophils (1×10 6 ) were infected with F. monophora conidia (MOI = 1:1) for 1 h or 2 h at 37˚C under homogenization. Different MOIs were previously tested, which showed similar results; thus, a MOI of 1:1 was chosen. As a control, conidia were incubated under the same conditions without neutrophils. After incubation, an aliquot was taken and diluted in distilled water to induce neutrophil lysis without harming the fungi. The fungi were seeded on PDA and incubated for 7 days at 25˚C for colony-forming unit (CFU) counting. The survival rate of the control group (fungi without neutrophils) was set as 100%.
Phagocytosis was determined by flow cytometry analysis according to Voyich et al. [21]. In brief, 2×10 5 neutrophils were infected with the FITC-labeled conidia (2×10 6 ) of different strains of F. monophora in RPMI 1640 media (10% FBS and 1% penicillin/streptomycin) for 1 h and 2 h at 37˚C. Phagocytosis was stopped by the addition of ice-cold PBS. Samples were washed, taken up in cold PBS and 2% FBS, and analyzed with FACSAriaIII (Becton Dickinson). The mean fluorescence intensity per cell was calculated to estimate the amount of phagocytosed conidia.

Animal model
All animals were treated according to National Institutes of Health guidelines for the use of experimental animals. The study protocol was approved by the ethics committee of Dermatology Hospital, Southern Medical University, and conformed to the Guide for the Care and Use of Laboratory Animals. Female BALB/c mice (6-8 weeks old, 16 mice in each group). The mice were randomly divided into four groups: control group, wild-type strain, pks1 TE mutant strain, and complemented strain group. The footpad was injected with 50 μl of the fungal solution containing 1×10 6 fungal cells or PBS as the control in the plantar cushion. The thickness of the footpad was measured with the aid of a caliper. At 3, 5, 7, and 14 days after injection, the infected mice were sacrificed to obtain footpad samples for future analyses, including CFU counting, histopathological examination, RNA sequence analysis, and cytokine detection. For cytokine detection, Mouse Cytokine & Chemokine Panel 1 (eBioscience, San Diago, USA) was used according to manufacturer's instructions. The footpad samples were put in the PBS with protease inhibitor, after homogenization, centrifuged (4˚C, 16,000g, 10min) and took the supernatant for Th1 and Th17-related cytokines (IL-6, TNF-α, IL-1β, IL-17A, IL-22), and neutrophil-related chemokines (CCL2, CCL3, CXCL1, CXCL2) detection. The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of Dermatology Hospital, Southern Medical University, Protocol 2018002 (7 March 2018).

Immunohistochemistry
To verify above suggestive neutrophil activation within WT-infected mice tissues, we then analyzed the inflammatory cells by immunohistochemistry using myeloperoxidase (MPO) antibody. Immunohistochemical staining was carried out as described previously [22]. Briefly, paraffin-embedded footpad tissues were sectioned at 3 μm. Tissue sections were deparaffinized and hydrated and incubated in 10% H 2 O 2 for 10 minutes to block endogenous peroxidase. After sections were blocked with goat serum, the sections were incubated with MPO antibody (ab208670) overnight at 4˚C. Subsequently, tissue sections were stained with horseradish peroxidase-coupled secondary antibody and detected by the Envision System. The assays were performed at least in triplicate.
2.8.1 mRNA sequencing by Illumina HiSeq /Novaseq or MGI2000 and data analysis. Total RNA of each sample was extracted from infected footpads using Trizell Reagent. According to the previous method [23], total RNA of each sample was quantified and qualified by Agilent 2100/2200 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA), NanoDrop (Thermo Fisher Scientific Inc.). 1 μg total RNA was used for following library preparation. Next generation sequencing library preparations were constructed according to the manufacturer's protocol. The poly(A) mRNA isolation was performed using Poly(A) mRNA Magnetic Isolation Module. The mRNA fragmentation and priming were performed using First Strand Synthesis Reaction Buffer and Random Primers. First strand cDNA was synthesized using ProtoScript II Reverse Transcriptase and the second-strand cDNA was synthesized using Second Strand Synthesis Enzyme Mix. The purified double-stranded cDNA by beads was then treated with End Prep Enzyme Mix to repair both ends and add a dA-tailing in one reaction, followed by a T-A ligation to add adaptors to both ends. Size selection of Adaptor-ligated DNA was then performed using beads, and fragments of~400 bp (with the approximate insert size of 300 bp) were recovered. Each sample was then amplified by PCR using P5 and P7 primers, with both primers carrying sequences which can anneal with flow cell to perform bridge PCR and P5/ P7 primer carrying index allowing for multiplexing. The PCR products were cleaned up using beads, validated using a Qsep100 (Bioptic, Taiwan, China), and quantified by Qubit 3.0 Fluorometer (Invitrogen, Carlsbad, CA, USA). Then libraries with different indices were multiplexed and loaded on an Illumina HiSeq/Novaseq instrument according to manufacturer's instructions (Illumina, San Diego, CA, USA) or a MGI2000 instrument according to manufacturer's instructions (MGI, Shenzhen, China). Sequencing was carried out using a 2x150 paired-end (PE) configuration; image analysis and base calling were conducted by the HiSeq Control Software (HCS) + OLB + GAPipeline-1.6 (Illumina) on the HiSeq instrument, image analysis and base calling were conducted by the NovaSeq Control Software (NCS) + OLB + GAPipeline1.6 (Illumina) on the NovaSeq instrument, image analysis and base calling were conducted by the Zebeacall on the MGI2000 instrument. The sequences were processed and analyzed by GENEWIZ.

Quality control & mapping.
In order to remove technical sequences, including adapters, PCR primers, or fragments thereof, and quality of bases lower than 20, pass filter data of fastq format were processed by Cutadapt (V1.9.1) to be high quality clean data.
Reference genome sequences and gene model annotation files of relative species were downloaded from NCBI. Secondly, Hisat2 (v2.0.1) was used as index reference genome sequence. Finally, clean data were aligned to reference genome via software Hisat2 (v2.0.1).

Differential gene expression and Gene Ontology (GO) and KEGG enrichment analysis.
In the beginning transcripts in FASTA format are converted from known GFF annotation file and indexed properly. Then, with the file as a reference gene file, HTSeq (v0.6.1) estimated gene and isoform expression levels from the pair-end clean data.
Differential expression analysis used the DESeq2 Bioconductor package, a model based on the negative binomial distribution. he estimates of dispersion and logarithmic fold changes incorporate data-driven prior distributions, Padj of genes were settled. GOSeq (v1.34.1) was used identifying GO terms that annotate a list of enriched genes with a significant padj less than 0.05. And topGO was used to plot DAG. KEGG databases (Kyoto Encyclopedia of Genes and Genomes) is a collection of databases dealing with genomes, biological pathways, diseases, drugs, and chemical substances. We used scripts in house to enrich significant differential expression gene in KEGG pathways.

Statistical analysis
Data are presented as the mean ± SEM. Unpaired Student's t-test with two-tailed P-values and ANOVA test was used for statistical analyses unless indicated otherwise (GraphPad Prism 5 software). In all tests, P-values less than 0.05 were considered statistically significant.
The numerical data used in all figures are included in S1 Data.

C-terminal TE deletion down-regulates the pks1gene expression and causes albino phenotype in F. monophora
To determine the role of TE on pks1, particularly C-terminus of TE in metabolism and growth of F. monophora, C-terminal deletion strain Δpks1-TE-C500 and complemented strain were successfully constructed (Fig 1A and 1B) and used to test loss-of-function phenotypes. After knocking out 500 bp C-terminal of TE domain, there are only 86 amino acids on the TE of AYO21_03016 (S1 Fig), without D106 and H231 residues, which can make TE loss of function. Melanin loss was clearly shown in Δpks1-TE-C500 that exhibited small and white colony on PDA plate (Fig 1C). Under microscopy, the conidia and hyphae were also less pigmented in Δpks1-TE-C500 strain (Fig 1C). Depigmentation in Δpks1-TE-C500 was sustained despite of supplementation of melanin agonist L-DOPA or L-tyrosine (S4 Fig). Tricyclazole, a DHN melanin inhibitor, caused a de-coloration in WT but had no effect on Δpks1-TE-C500. At transcription level, TE gene expression was negative in Δpks1-TE-C500 and the expression level of other 5 genes SAT, KS, AT, DH, ACP was reduced by more than 50%, compared with WT (Fig 1D).

C-terminal TE deletion reduces in vitro growth and germination of F. monophora
To further understand the growth effects of a down-regulated pks1, growth rate, sporulation and germination of F. monophora were measured. We found that the colony diameter of Δpks1-TE-C500 was significantly reduced, especially at 25˚C when compared with the wildtype strain (Fig 2A). The biomass of 14 days growth of Δpks1-TE-C500 was only half of wildtype and complemented strains at 25˚C and 37˚C (Fig 2B). In addition, the Δpks1-TE-C500 strain exhibited an approximately 10-fold decrease in sporulation (Fig 2C) and showed a delayed germination. Compared with the germination course of wild-type and complemented strains in which germination began at 12 h and the long hyphae formed at 24 h, the mutant strain showed no germination at 12 h, started to germinate at 24 h and no hyphae were formed until 48 h (S5 Fig).

C-terminal TE deletion increases sensitivity of F. monophora to osmotic and oxidative stress
Stress adaptation of TE mutant to salt stress and or oxidative stress were tested in the PDA plates containing different concentrations of KCl, H 2 O 2 , SNAP, menadione (VitK), and sorbitol. The data revealed that the mutant was highly sensitive to oxidative stress. The colony diameters of Δpks1-TE-C500 are markedly reduced on PDA containing various concentrations of H 2 O 2 , SNAP, or VitK versus wild-type strain. The mutant strain also showed significant growth inhibition at various concentrations of KCl and sorbitol except 0.4 M KCl and 1 M sorbitol, and was easily killed by UV irradiation. When compared with mutant strain, the parental strain and complemented strain showed resistance to each of tested chemicals in a dosedependent manner (Fig 3).

Deletion of C-terminal TE F. monophora alters secondary metabolites, lipid and polyketides metabolism
To understand the roles of F. monophora PKS TE domain on metabolism, metabolomic profiles of the C-terminal TE mutant Δpks1-TE-C500 was compared with WT and complemented strain (Fig 4). In the principal component analysis (PCA) diagram, WT and the complemented strain were separated from the Δpks1-TE-C500, which implied that WT and the complemented strain were not similar to Δpks1-TE-C500. (Fig 4A). We found that deletion of C-terminal TE accounted for 23% differential metabolites between mutant and either WT or complemented strain, a total of 4585 metabolites (77%) were overlapped among three strains (Fig 4B). KEGG enrichment analysis showed that knockout pks1 TE C-terminus specifically effected terpenoid and polyketide production, as evidenced by the 90% loss of gossypol and hydroxyspheroidenone and the 85% loss of okenone (S2 Table and Fig 4C). Lipid biosynthesis and amino acids metabolism were also influenced after knocking out TE (Fig 4C). We also found that biosynthesis of other secondary metabolites such as cephalosporin_C and peni-trem_D reduced by 90% as well in mutant strain (S2 Table). Accompanying with metabolites impairment as shown above, reduced form of L-glutathione, S-lactoylglutathione and niacinamide increased in the Δpks1-TE-C500 versus those of the wild-type and complemented strain (Fig 4E), which could be a compensatory response to decreased polyketides. In DOPA-melanin pathway, glutathione is required for formation of pheomelanin (yellow or red), but not for eumelanin (black) (Fig 4D).

Deletion of C-terminal TE F. monophora increases phagocytotic killing
The consequence of melanin loss and decreased growth rate of the Δpks1-TE-C500 in immune escape is then evaluated. All strains were co-cultured with macrophages (RAW264.7) to detect phagocytotic killing at 24 and 48 h after incubation. We found that WT conidia could be rapidly phagocytosed by the macrophages at 12 h and they started to germinate at 24 h and formed hyphae at 48 h within macrophages, causing immune escape (Fig 5A). The conidia of the mutant strain were also readily phagocytosed by macrophages at 24 h, however, the amount of conidia in the cells were significantly reduced by 48 h with no hyphae as most of the spores were killed and degraded. TEM indicated wild-type conidia maintained normal cell morphology inside the macrophages with the intact cell wall and organelles. The mutant conidia were in the phagosome at 24 h (Fig 5B). After 24 h of incubation, wild-type or mutant strain-infected macrophages were lysed for detecting fungal burden. Compared with the wildtype group, the mutant group exhibited sharply decreased CFU in the infected macrophages (Fig 5C). In addition, the mitochondria of wild-type and complemented strain-stimulated  macrophages showed swelling and vacuolation (usually due to cell damage), which could not induce by the Δpks1-TE-C500 strain.

Deletion of C-terminal TE F. monophora decreases virulence
We next tested virulence of mutant on our previously-established murine paw infectious model. We found that the Δpks1-TE-C500 strain is significantly less virulent than control strains (Fig 6). The paw swelling and inflammation of mice infected with the Δpks1-TE-C500 strain are significantly less severe than those of mice infected with the wild-type strain and complemented strain (Fig 6A and 6B), especially at 7 days. The fungal burden in the infected footpads of the Δpks1-TE-C500 (136.5 CFU/g) is significantly less than those of the wild-type strain (374.7 CFU/g) and complemented strain (380 CFU/g) at 14 days (Fig 6C). The inflammatory response is overall weakly induced by Δpks1-TE-C500, when compared with abundant inflammatory foci over the injection site in the footpads of mice injected with the wild-type and complemented (Fig 6D), in which a large number of mixed inflammatory infiltrates and granulocytes are found.

TheΔpks1-TE-C500-infected mice exhibited a weak pro-inflammatory cytokine response
The cytokine profile revealed a strong neutrophil chemotaxis response in WT-infected footpads, which was consistent with granulocytes enhancement in histopathology. PCA analysis showed that the Δpks1-TE-C500 group was relatively similar to the PBS group on the 3rd and 7th day, while WT group was alike to the complemented group (Fig 7A). The gene expression between the WT and the complemented strain at day 3 and day 7 were analogous in heatmap (Fig 7B) and volcano map (S6 Fig). GO enrichment analysis showed that the differentially expressed genes were significantly enriched in inflammatory response and neutrophil chemotaxis in WT versus Δpks1-TE-C500-infected footpads at day 3 and day 7 after infection (labeled with black arrow) (S7A Fig). KEGG enrichment demonstrated that the differential genes were enriched in chemokine signaling pathway and Th17 cell differentiation (labeled with black arrow) (S7B Fig). Heatmap of neutrophil chemotaxis indicated that gene expression in wildtype strain was higher than mutant strain (Fig 7C). Proinflammatory cytokines associated with Th17/Th1 response such as IL-6, TNF-α, and IL-1β, and neutrophil-recruiting chemokines including CCL2, CCL3, CXCL1, and CXCL2 were highly produced from day 3 to day 14 after WT infection. Avirulent Δpks1-TE-C500 produced much low levels of the same panel of cytokines, especially at day 3 and day 7 after infection (Fig 7D). These results were consistence with RNA-seq data.

Δpks1-TE-C500-infected mice has less neutrophil recruitment in the footpads
There were significantly fewer neutrophils in the footpads of mice injected with the mutant versus the wild-type or complemented strain when detected by immunohistochemistry or western blotting (Fig 8A and 8B). In neutrophil killing assay, each strain was co-cultured with human blood neutrophils for 1 h and 2 h, the phagocytosis rate was determined by flow cytometry. Through counting the CFU, our data confirmed that the mutant strain was more likely to be phagocytosed and killed by neutrophils, which were consistent with the findings of in vitro macrophage killing assay (Fig 8C). . In comparison with footpads injected with the wild type and complemented strain, footpads injected with the Δpks1-TE-C500 were less swollen, and the number of CFU in the tissues was significantly reduced. In addition, the number of inflammatory cells of Δpks1-TE-C500 group in the pathological tissues was decreased. All statistical analysis were performed using two-tailed t-test, and the results were statistically significant ( � , P<0.05; ��� , P<0.001; ���� , P<0.0001).

Discussion
PKSs in many organisms contain a C-terminal TE in their multidomain modules [24] and integrity of TE domain in F. monophora PKS seemed to be critical for melanin synthesis in our previous study [13]. TE domain in F. monophora PKS can be classified into TE16 family according to a previous report [25] that has oleoyl-ACP hydrolase (EC3.1.2.14) activity. The functions of the TE domain include termination of fatty acid synthesis, polyketide biosynthesis, and non-ribosomal peptide biosynthesis, known as thioesterase I. To verify the function of TE domain of PKS on F. monophora virulence, we constructed a C-terminal TE deletion mutant by knocking out the last 500 bp of AYO21_03016 (pks1) in this study. The consequences of metabolic abnormality in fungal growth, pathogenic traits and virulence are then examined in vitro and in vivo conditions. The TE-mediated pathogenicity has been demonstrated in some bacteria which have TEs belonging to different TE families. However, the pathogenesis of the TE domain on PKS has not been determined [26][27][28]. To the best of our knowledge, it is the first study to assess pathogenic effects of TE on fungal PKS. PKS gene can regulate the development of conidia in Beauveria bassiana [29], Pestalotiopsis microspore [30] and Chaitomium globosum [31]. In Pestalotiopsis fici, PKS has been proved to be essential not only to protect spores from biological and abiotic stress, but also to the development of spore structure [32]. In this study, the Δpks1-TE-C500 strain also shows defects in growth and the development of conidia demonstrated by smaller-sized, less, and delayed germination. Our results are similar to the reports found in other fungi after deletion of PKS [30,[33][34][35]. Given the fact that pks1 expression significantly decreased in the Δpks1-TE-C500 strain, all these findings indicate that TE on pks1 act as a trigger for growth and sporulation in F. monophora.
The reproduction and germination of conidia are the core components of survival in the environment, as well as the important determinants of pathogenicity [36]. For example, fungal conidia face various stresses in the human body, such as heat, salt, oxidative stress, osmotic pressure, nutritional deficiencies, and cytokine-mediated killing. In addition, melanin in fungi is also essential for the resistance to external stresses. Studies on C. neoformans and F. pedrosoi have shown that melanin can convert ferric iron to ferrous iron, thereby forming an iron-melanin complex to eliminate oxidative free radicals [37,38], which plausibly explains that the Δpks1-TE-C500 strain is more sensitive to an increased sensitivity to H 2 O 2 and NO in vitro during mycelial growth. We speculate that the loss of TE affects the synthesis of melanin, which may not effectively eliminate oxidative free radicals, resulting in the increase sensitivity to oxidative stress.
The stress adaptation not only helps fungi survive in extreme environments, but also plays an important role in anti-killing and replication in macrophages and neutrophils. In our study, viability and morphogenesis of the Δpks1-TE-C500 within the macrophages are impaired and mutant cells were enclosed within the lysosomes of microphages. These results are consistent with the study in Aspergillus fumigatus, in which pksP (alb1) was required for defense of lysosomal fusion and intracellular killing [39]. Melanin is believed to decrease a pathogen's resistance to the host antimicrobial mechanisms and influence the host immune  Western blotting was performed to detect MPO in footpad tissues. MPO was highly expressed in footpad tissues on the 3rd, 7th, and 14th day after injection with the wild-type and complemented strains. (C) After each strain was co-cultured with human blood-derived neutrophils for 1 h and 2 h, the phagocytosis rate was determined by flow cytometry, and the survival rate of conidia was measured by counting CFU on PDA. Statistical significance was determined by two-tailed t-test ( ���� , P<0.0001). The Δpks1-TE-C500 had higher phagocytosis and survival rate.
https://doi.org/10.1371/journal.pntd.0010485.g008 response to infection [40]. Recent studies have demonstrated that calcium sequestration by DHN-melanin inhibits essential host effector pathways regulating phagosome biogenesis and prevents A. fumigatus conidia killing by phagocytes [41,42]. All these findings indicated that the loss of melanin is an important reason for the attenuated virulence of the Δpks1-TE-C500 strain besides the defects in growth and germination mentioned above.
PKS regulates the biosynthesis of various secondary metabolites through the biosynthetic intermediates besides being a key enzyme for melanin formation [43]. Our findings indicate that the TE domain is required for pks1 expression in F. monophora since down-regulated expression of each of remaining five enzymes in pks1 after knocking out TE. Metabolomic profiles showed a significant increase on anti-oxidant glutathione (reduced form) or nicotinamide in Δpks1-TE-C500 strain. The pathogenic impact of increased anti-oxidant glutathione and nicotinamide in Δpks1-TE-C500 strain is unclear, but it could be an overall result from a downregulated pks1. Villarama et al. found that glutathione inhibits the formation of melanin through mediating the switch mechanism from eumelanin (L-DOPA pathway) to pheomelanin production [44], which could be used to explain the albino phenotype of Δpks1-TE-C500 strain due to supplementation of L-tyrosine and L-DOPA did not restore melanin in Δpks1-TE-C500 strain.
In addition to melanin biosynthesis, glutathione has anti-inflammatory effect by inhibiting the production of IL-6, IL-8, and TNF-α [45] and nicotinamide also shows potent inhibitory effects on proinflammatory cytokines from the host [46]. Ciebiada-Adamiec et al. found that nicotinamide inhibited antifungal activity against Candida albicans, Trichophyton rubrum, and Trichophyton mentagrophytes by decreasing pathogenesis-associated enzymatic activities [47]. These anti-inflammatory effects of fungal glutathione and nicotinamide might explain the mild footpad swelling and decreased inflammation in mice infected with the Δpks1-TE-C500 mutant. Besides, there were some metabolites dropped sharply in mutant strain, including gossypol, cephalosporin_C, ydroxyspheroidenone, okenone and penitrem_D. Gossypol is an inhibitor of eukaryotic cells with an undetermined mode of action [48,49]. The relationship between its reduction in vivo and the pathogenicity of our mutant strain is still not clear, which deserves further investigation. An inverse relation was observed between the ability to synthesize cephalosporin C and the growth rate in Cephalosporium acremonium mutants [50]. However, the Δpks1-TE-C500 strain in our study has a significantly reduced production of cephalosporin C and slow growth at the same time, suggesting a various relationship between cephalosporin C and growth in different strains. Studies on hydroxyspheroidenone, okenone and penitrem_D in fungi were limited and their function are still unclear.
The importance of neutrophils for the prevention and clearance of invasive fungal infection is widely recognized [51]. IL-17A and Th17 cells are considered to confer protection against fungi by recruitment via CXC chemokines [52]. In footpads injected with wild-type strain, neutrophils and Th17 were the main cell lineages found in the infected area during the early stage of experimental murine CBM, which is similar to increased Th17 cells found in the earlyinfected stage of F. pedrosoi [53]. In the mutant strain infection, we noted that Th17-related cytokines (IL-17A, IL-22) and neutrophil-related chemokines (CCL2, CCL3, CXCL1, CXCL2) are significantly reduced at the early stage of infection, which might be related to the absence of melanin and defects in growth and germination. However, the pksP/alb1 mutant of A. fumigatus conidia promoted the production of pro-inflammatory cytokines [54], suggesting a variety of pks1-associated inflammatory effects among different fungi. In consistence of cytokines responses, in particular increased neutrophil-related chemokines, an abundant neutrophil infiltration is shown in the footpads of mice injected with the wild-type strain but not in mutant infected mice. These results indicate that the TE plays an important role in promoting the host's inflammatory response and is essential for the virulence of F. monophora.
As a NTD, CBM is well-known for its resistance to most treatments and prone to recurrence [55]. Recent studies have shown that impaired fungal clearance in CBM could be related to inherent virulence traits of pathogens such as thermotolerance, cellularity with thick cell walls, cell adhesion, hydrophobicity and melanin [56][57][58]. Our data indicate that the knockout of TE affected the synthesis of melanin and virulence of F. monophora, which provides new insights into the pathogenic mechanism of CBM and new therapeutic development.
In summary, an albino F. monophora strain is successfully obtained by knocking out C-terminus of TE gene in this study. Several findings and conclusions can be drawn. First, disruption of TE catalytic active site (Ser-His-Asp triad) results in the melanin loss, decreases pks1 gene expression and affects biosynthesis of secondary metabolites. Second, reduction of melanin or other secondary metabolites in mutant leads to slow growth, delays germination, and increases the sensitivities to environmental stresses such as oxidative stresses. Third, mutant strain has a compromised capacities to invade murine footpads and to induce inflammatory responses, and is easily killed by macrophages and neutrophils. These results suggest that Cterminal TE is essential in maintaining fully function of pks1 and fungal pathogenicity of F. monophora. The plasmid with knockout of the terminal 500 bp of the pks1 gene (pks1-TE) or whole gene (pks1) was transformed to A. tumefaciens strain EHA105. The pre-induced EHA105 cells and spores were mixed and co-cultured at the same time. Transformants were selected on PDA plates with hygromycin B (50 μg/ml) and cefotaxime (200 μM) and incubated at room temperature for 7 days. The number of transformants and the ratio of positive transformants (determined by phenotype) under each condition were calculated. Statistical significance was determined by two-tailed t-test ( �� P<0.01; ��� P<0.001).