Wor1‐regulated ferroxidases contribute to pigment formation in opaque cells of Candida albicans

Candida albicans is a harmless commensal resident in the human gut and a prevalent opportunistic pathogen. A key part of its commensalism and pathogenesis is its ability to switch between different morphological forms, including white‐to‐opaque switching. The Wor1 protein was previously identified as a master regulator of white‐to‐opaque switching in mating type locus (MTL) homozygous cells. The mechanisms by which the dark color of the opaque colonies is controlled and the pimpled surface of opaque cells is formed remain unknown. Candida albicans produces melanin pigment in vitro and during infection. However, the molecular mechanism underlying the regulation of melanin production is unclear. In this study, we demonstrated that ferroxidases (Fets) function as pigment multicopper oxidases and regulate the production of dark‐pigmented melanin in opaque cells. The FET genes presented distinct regulation patterns in response to different extracellular stimuli. In YPD (1% yeast extract, 2% peptone and 2% dextrose)‐rich medium, four of the five FET genes were up‐regulated by Wor1, especially at the human body temperature of 37 °C. In minimal medium with low ammonium concentrations, all five FET genes were up‐regulated by Wor1. However, at high ammonium concentrations, some FET genes were down‐regulated by Wor1. Wor1‐up‐regulated Fets contributed to dark pigment formation in opaque colonies, but not to the elongated shape of these opaque cells. Increased melanin externalization was associated with the pimpled surface of the opaque cells. Melanized C. albicans cells were more resistant to fungal clearance. Deletion of the five FET genes completely blocked melanin production in opaque cells and resulted in the generation of white elongated ‘opaque’ cells. In addition, the up‐regulated Fets are important for defense against oxidant attacks. The functional diversity of Fets may reflect the multiple strategies of C. albicans to rapidly adapt to diverse host niches.

Candida albicans is a harmless commensal resident in the human gut and a prevalent opportunistic pathogen. A key part of its commensalism and pathogenesis is its ability to switch between different morphological forms, including white-to-opaque switching. The Wor1 protein was previously identified as a master regulator of white-to-opaque switching in mating type locus (MTL) homozygous cells. The mechanisms by which the dark color of the opaque colonies is controlled and the pimpled surface of opaque cells is formed remain unknown. Candida albicans produces melanin pigment in vitro and during infection. However, the molecular mechanism underlying the regulation of melanin production is unclear. In this study, we demonstrated that ferroxidases (Fets) function as pigment multicopper oxidases and regulate the production of dark-pigmented melanin in opaque cells. The FET genes presented distinct regulation patterns in response to different extracellular stimuli. In YPD (1% yeast extract, 2% peptone and 2% dextrose)-rich medium, four of the five FET genes were up-regulated by Wor1, especially at the human body temperature of 37°C. In minimal medium with low ammonium concentrations, all five FET genes were upregulated by Wor1. However, at high ammonium concentrations, some FET genes were down-regulated by Wor1. Wor1-up-regulated Fets contributed to dark pigment formation in opaque colonies, but not to the elongated shape of these opaque cells. Increased melanin externalization was associated with the pimpled surface of the opaque cells. Melanized C. albicans cells were more resistant to fungal clearance. Deletion of the five FET genes completely blocked melanin production in opaque cells and resulted in the generation of white elongated 'opaque' cells. In addition, the up-regulated Fets are important for defense against oxidant attacks. The functional diversity of Fets may reflect the multiple strategies of C. albicans to rapidly adapt to diverse host niches.
Candida albicans is a harmless commensal resident in the human gut and a prevalent opportunistic pathogen [1][2][3]. Important to its commensalism and pathogenesis is its ability to switch between different morphological forms, including white-to-opaque switch and white-togastrointestinally induced transition (GUT) [3][4][5][6][7]. The white and opaque phenotypes have different colonial morphologies in terms of colony size, shape and color. White cells form hemispherical and white-colored colonies, whereas opaque cells form much larger, flatter and opaque or dark-colored colonies. Opaque cells are elongated with a pimpled surface, but the mechanisms that control the dark color of opaque colonies and the pimpled surface of opaque cells remain unknown. White cells are more virulent than opaque cells in systemic infections, whereas the virulence of opaque cells seems to increase during cutaneous infection [8,9]. Opaque cells are significantly less susceptible to phagocytosis by Drosophila and mouse phagocytes than white cells [10]. GUT cells are elongated in shape with a smooth surface. GUT cells showed enhanced commensal fitness when passing through the mammalian gastrointestinal (GI) tract; however, they attenuated virulence in the bloodstream [3,11].
Wor1 was previously identified as a master regulator of white-to-opaque switching in mating type locus (MTL) homozygous cells [12][13][14]. Chromatin immunoprecipitation assay showed that it binds to more than 100 target promoters and its own upstream region [15]. Wor1 is expressed when passing through the mammalian gut, thus triggering white-to-GUT switching in MTLa /a heterozygous cells and promoting C. albicans commensalism [3]. Wor1 overexpression also increases the adhesion of C. albicans cells to the mouse gut mucosa [16].
Melanin is usually described as one of the most common natural pigments and is broadly produced by various organisms [17,18]. Melanins are negatively charged hydrophobic macromolecules, which are composed of polymerized phenolic and/or indolic compounds, and referred to as 'fungal armor' because of the ability of the polymer to protect microorganisms against a broad spectrum of toxic insults. Several types of melanin are known to exist in the fungal kingdom, most of which are derived from 1, 8-dihydroxynaphthalene (DHN), known as DHN-melanin, and the others are derived from L-3,4-dihydroxyphenylalanine (L-DOPA), known as DOPA-melanin [19]. DHN-melanin is usually synthesized by filamentous fungi, such as Aspergillus spp., through a polyketide synthase pathway, by using endogenous substrates, including acetyl-CoA and malonyl-CoA. After a series of hydrolysis, reduction, dehydration and polymerization, DHN-melanin is formed. In contrast, DOPA-melanin is produced via a polyphenoloxidase (a laccase), which catalyzes a one-step oxidation of dihydroxyphenols to quinone intermediates, which subsequently auto-oxidize to form melanin [17,18]. Cryptococcus neoformans produces pigments from many aminophenol and diaminobenzene compounds in vitro [20]. During central nervous system infection in vivo, C. neoformans synthesizes melanin from neurotransmitters, such as dopamine, norepinephrine and epinephrine as substrates [21]. C. neoformans cannot generate melanin without an exogenous substrate, and the type of pigment synthesized varies depending on the chemical structure of the substrate added to the medium [18]. C. neoformans melanin is catalyzed by laccases and encoded by two tandem localized genes, CNLAC1 and CNLAC2 [22,23], and CNLAC1 plays a dominant role in melanin production [24,25].
Laccases constitute the largest subfamily of multicopper oxidases (MCOs) and are widely distributed in fungi, higher plants, bacteria and insects. MCOs consist of four enzyme families: laccases, ascorbate oxidases, ferroxidases (Fets) and ceruloplasmin [26]. Most MCOs can use a wide variety of aromatic phenols and amines as reducing substrates. Laccases and Fets share some MCO-specific patterns and signature sequences (https://lcced.biocatnet.de/) [27]. Fets can oxidize laccase substrates in addition to ferrous iron at a lower catalytic efficiency. Laccases also can oxidize ferrous iron other than phenolic substrates at a lower rate [28]. In the C. albicans genome (http://www.candidage nome.org), there are five predicted MCOs, which were previously reported as Fets [29]. Interestingly, the five Fets showed distinct iron-dependent regulation patterns, and two of them (Fet34 and Fet99) were significantly up-regulated by low iron [29]. Whether the five Fets can oxidize laccase substrates and produce pigments in C. albicans remains to be investigated.
C. albicans can produce melanin pigment in vitro and during infection [30]. Melanin externalization also has been observed in C. albicans [31]. In this study, we proved that the previously reported Fets could function as pigment MCOs necessary for melanin pigment production. We found that all the FET genes presented distinct regulation patterns in response to different extracellular stimuli. The Wor1-up-regulated Fets contributed to dark pigment formation in opaque cells, but not to the elongated shape of opaque cells. To the best of our knowledge, this is the first study to link the dark color of opaque cells to melanin pigment production and the pimpled surface of opaque cells to melanin externalization, thus revealing the functional diversity of Fets in C. albicans.

Strains and growth conditions
The C. albicans strains used in this study are listed in Table 1. C. albicans strains were routinely grown at 22°C in YPD (1% yeast extract, 2% peptone and 2% dextrose). YPS (YP + 2% sorbose) was used to isolate MTLa/a or MTLa/a strains. Transformants were selected on synthetic complete medium with glucose (SCD; 2% dextrose, 0.17% yeast nitrogen base without amino acids and ammonium sulfate, 0.5% ammonium sulfate and auxotrophic supplements). Cells were cultured in minimal medium (15.0 mM glucose, 10.0 mM MgSO 4 , 29.4 mM KH 2 PO 4 , 13.0 mM glycine, 3.0 M vitamin B 1 , pH 5.5) with L-DOPA and N-acetylglucosamine (GlcNAc) for melanin production. Glucose minimal medium (GMM; 2% w/v glucose, 0.17% w/v yeast nitrogen base without amino acids and ammonium sulfate, and 0.5% w/v ammonium sulfate) and GMM with low ammonium (GMM-LA; 2% w/v glucose, 0.17% w/v yeast nitrogen base without amino acids and ammonium sulfate, and 0.025 mgÁL À1 ammonium sulfate) were used to test the effects of iron on FET expression and the impact of Wor1 on FET expression. To prepare the minimal medium containing different concentrations of iron, we first depleted GMM or GMM-LA of iron by adding 1 mM ferrozine and 10 mM ascorbic acid, followed by the addition of different amounts of FeCl 3 . carrying CaURA3, CmLEU2, CdHIS1 and CdARG4 markers were used as templates to delete target genes by PCRbased homologous recombination, as previously described [32]. LoxP-CmLEU2-LoxP (PLP), LoxP-CdHIS1-LoxP (PHP) and LoxP-CdARG4-LoxP (PAP) were assembled with 5 0 and 3 0 fragments of target genes by the fusion PCR method. The BWP17 strain [33] was used to construct a single gene mutant fet31/fet31D/D, fet99/fet99D/D or fet3/ fet3D/D. The fet33/fet33D/D and fet34/fet34D/D mutants were gifts from M. Li [34]. The MTLa /a 4fetsD/D (FET31-FET99-FET3-FET34) mutant was generated from the SN148 strain [35] by PCR-based homologous recombination with CmLEU2 and CdHIS1 markers and then streaked onto YPS for isolation of MTLa/a 4fetsD/D strains. The two copies of FET33 were disrupted by the 'URA-BLAST' method [36]. PstI-digested pFET33-KO was transformed into the MTLa/a 4fetsD/D (CBD9) mutant to generate the 5fetsD/D mutant (CBD12). All disruptions were confirmed by Southern blotting and PCR analysis (Fig. S1). The primers used for PCR amplification are listed in Table 3.

Measurement of melanin production in C. albicans
Melanin production was measured according to previously described methods [30,31] with minor modifications. GlcNAc (5 mM) was added to the minimal medium to stimulate melanin production. Melanin production was approximated by measuring the optical density at 310 nm (OD 310 ) or 470 nm (OD 470 ) of the whole culture versus a matched blank. Cells grown without DOPA were used as a blank. Melanin production was also approximated by OD 310 or OD 470 in resuspended cell pellets. To monitor the visible pigment of melanized cells, we resuspended cells grown with DOPA in PBS and transferred to 96-well plates for photography.

Enzyme assay
The oxidative enzyme associated with melanin synthesis in C. albicans was determined by a previously reported method [23,37], and the enzyme solution was prepared as previously described [38,39]. In brief, overnight-cultured C. albicans cells were reinoculated at 2.0 9 10 6 cellsÁmL À1 in YPD and cultured at 22 or 37°C. Then, the cells were washed with cold PBS, and all further procedures were carried out at 4°C. Cells were resuspended in a lysis buffer ( was determined with DOPA as a substrate. Enzyme solutions (10 lL) were incubated in a 200-lL minimal medium with 1 mM DOPA at 22 or 37°C for 30 min, and the OD 310 was recorded. One unit was defined as 0.001 absorbance unit in 30 min of the assay. Enzyme activity was expressed as mU AbsÁmin À1 Álg protein À1 .

Bone marrow-derived macrophage induction
Bone marrow-derived macrophages (BMDMs) were prepared as previously described [40]. In brief, bone marrow cells were obtained by flushing the femur and tibia of Institute of Cancer Research (ICR) mice aged 6-8 weeks with BMDM culture medium (RPMI-1640 medium containing 10% FBS, 30% L929 conditioned medium and 1% penicillin-streptomycin). After removal of nonadherent cells on day 4, fresh BMDM culture medium was added, and on day 5, BMDMs were seeded into 12-well plates at 5 9 10 5 cells per well and used on day 7.

Phagocytosis of C. albicans cells
BMDMs were plated at 5 9 10 5 cells per well in 12-well plates and infected with C. albicans cells (multiplicity of infection = 3). At 2 h postinfection, after washing away free fungal cells with 1% PBS, C. albicans uptake by macrophages was counted by light microscopy after fixation and staining using a modified Wright-Giemsa stain (Sigma) as previously described [41]. Phagocytosis was defined as the percentage of macrophages taking up at least one fungal cell, and the phagocytic index was defined as the number of fungal cells taken up per 100 macrophages. Data are shown as means AE standard deviation (SD) from three independent experiments by analyzing at least 200 macrophages per well. We also used the colony-forming units (CFUs) method to analyze macrophage phagocytosis [40]. BMDMs were plated at 5 9 10 5 cells per well in 12well plates and infected with C. albicans cells (multiplicity of infection = 0.5-3). After 0.5-12 h of coincubation, the wells were washed twice with 1% PBS to remove free fungal cells. BMDMs were scraped, lysed in a lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA), resuspended, serially diluted and plated onto YPD agar. Fungal CFUs were counted after 48 h of incubation at 30°C.
Analysis of melanized C. albicans resistance to UV radiation, H 2 O 2 oxidation and antifungal drugs C. albicans cells were cultured in a minimal medium containing 5 mM GlcNAc with 1 mM DOPA at 22°C for 4 days and then suspended in 1% PBS. Cells (1.0 9 10 7 ) were exposed to UV at 30 000 lJÁcm À2 for 30 s or treated with 1.5 mM H 2 O 2 for 2 h, and then spread onto the YPD plate for survival analysis. Similarly, DOPA-incubated C. albicans cells were suspended in PBS, and 1.0 9 10 7 cells were treated with 32 lgÁmL À1 caspofungin or 1 lgÁmL À1 amphotericin B (AMB) for 2 h. Fungal CFUs were counted after 48-h incubation at 30°C.

Scanning electron microscopy and transmission electron microscopy
Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were performed according to the methods described by Walker et al. [31]. C. albicans cells were collected, washed twice with cold PBS, fixed in 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer (pH 7.3) for 24 h at 4°C, postfixed with 1% OsO 4 for 1 h and serially dehydrated in ethanol (30%, 50%, 70%, 80%, 95% and 100%) for 10 min. Then, the cells were critical point dried in CO 2 with a Leica EM CPD300 dryer (FEI Ltd., Hillsboro, USA), sputter-coated with gold using a Leica EM SCD050 coater (FEI Ltd.) and viewed in a FEI Quanta 250 scanning electron microscope (FEI Ltd.). C. albicans cell pellets were fixed in 2.5% (v/v) glutaraldehyde in 0.1 M sodium phosphate buffer (pH 7.3) for 24 h at 4°C [31]. After three washes with PBS, the cells were treated with 1% OsO 4 for 1.5 h, serially dehydrated in ethanol (30%, 50%, 70%, 80%, 95% and 100%) and infiltrated with acetone/Spurr resin. Additional infiltration was provided under vacuum in a mixture of acetone and Epon 812 for 1.5 h. The samples were embedded in Epon 812 and polymerized at 60°C for 48 h, and further treated with a Leica EM TRIM2 machine (Leica Mikrosysteme GmbH A, Wien, Austria). Semithin survey sections were prepared with a diamond knife on a Leica EM UC7 ultramicrotome and stained with 1% toluidine blue to detect the area with the best cell density. Ultrathin sections (70 nm) were prepared with a diamond knife, stained with uranyl acetate and lead citrate, and observed under a FEI Tecnai G2 Spirit transmission electron microscope (FEI Ltd.).

Quantitative PCR
RNA extraction, cDNA synthesis and quantitative PCR (qPCR) amplification were performed as described previously [42]. Exponentially growing C. albicans cells were collected for RNA extraction. The first-strand cDNA was synthesized from 3 lg total RNA in a 60-lL reaction volume using the cDNA synthesis kit (TaKaRa Biotechnology, Dalian, China). qPCRs were performed using the Chromo 4 Real-Time PCR System (Bio-Rad, Indianapolis, IN, USA). SYBR Green I (TaKaRa) was used to visualize and monitor the amplified product in real time. Transcription levels of genes in different samples were normalized against the levels of ACT1. Primers used are shown in Table 3. The data were measured with three independent experiments.
Detection of laccase-like activity in C. albicans Nondenaturing gel electrophoresis was used to detect the laccase-like activity in C. albicans as previously described [30]. C. albicans cells were cultured in YPD overnight at 22°C. After the cells were harvested, the total proteins were extracted with acid-washed glass beads (Sigma) using a Fast-Prep homogenizer (FP120; Thermo Electron). The homogenate was centrifuged at 16 200 g for 30 min. The protein concentration of the cell lysate was measured by the Bradford method, and 300 lg total protein in each 40 lL sample was loaded onto gels. Proteins were also estimated using the Coomassie brilliant blue method (Fig. S2). Commercially, laccase (from Rhus vernicifera; Sigma) was used as a positive control, and an equal protein sample was boiled for 5 min as a negative control. Rhus vernicifera laccase (40 U equivalent) and 300 lg total protein of C. albicans cells were loaded onto gels. After electrophoresis, the gels were immersed in 1 mM L-DOPA in 0.1 M citric acid/ 0.2 M Na 2 HPO 4 (pH 6.0) buffer overnight. Positive laccase activity was revealed by C. albicans protein extracts as shown by dark bands, which confirmed that L-DOPA had polymerized to form melanin.

Quantitative mating assays in C. albicans
Quantitative mating analysis was performed as previously described [43]. YPD or SCD medium was used for mating between the C. albicans MTLa and MTLa strains. The strains were mixed at 1 : 1 and incubated in YPD plates at 22°C for 6 days, then spread onto SCD plates without amino acids. The mating efficiency = conjugants/(limiting parent + conjugants) = the greater of (-Ade-Ura-His-Arg)/ (-Ade) or (-Ade-Ura-His-Arg)/(-Ura-His-Arg).

Results
Opaque cells produce more dark-pigmented melanin than white cells C. albicans was featured as a white yeast and was later found to exist in an opaque form [4]. C. albicans can switch between white and opaque cellular phenotypes to survive at different host niches. The white colony appears white and hemispherical, whereas the opaque colony appears dark and flat (Fig. 1A). The white cells appear round, whereas the opaque cells are elongated (Fig. 1B). To investigate whether the dark color of the opaque colony was correlated with melanin pigment production in C. albicans, we measured melanin pigment in white or opaque cells with L-DOPA as a substrate using previously reported methods [30,31] and found that opaque cells produced more melanin pigment than white cells. After 4-day incubation at 22°C in a defined liquid minimal medium containing 1 mM DOPA and 5 mM GlcNAc, the resuspended opaque cell pellets showed a black color, while the white cell pellets showed a white color (Fig. 1C), suggesting that opaque cells produced more melanin pigment than white cells. To quantify the melanin production, we measured the OD of opaque cells in whole cultures ( Fig. 2A) and cell pellets (Fig. 2B) at OD 310 nm and OD 470 nm ; opaque cells grown with DOPA showed a higher absorption value than white cells.

Wor1 promotes melanin production in both MTL homozygous and heterozygous cells
Wor1 is a master regulator for white-to-opaque switching in C. albicans [12][13][14], and this white-to-opaque switching is coupled with the sexual mating process under a1-a2 repression [44]. The ability to switch to the opaque state depends on whether the cells are homozygous for the MTL that controls the cell type [43]. In MTLa/a or MTLa/a homozygous cells, ectopically expressed Wor1 induces the transcription of endogenous WOR1 in a positive feedback manner, promotes white-to-opaque switching and locks cells in the opaque phase [12]. In MTLa /a heterozygous cells, overexpression of Wor1 had lower efficiency in whiteto-opaque switching [12]. We first tested the impact of Wor1 on melanin production in MTLa/a homozygous cells. Compared with white cells, the wild-type (WT) opaque cells produced more melanin in the DOPAcontaining medium. At both OD 310 nm ( Fig. 2A,B, left) and OD 470 nm ( Fig. 2A,B, right), the opaque cells grown with DOPA showed a higher absorption value than the white cells during the 4-day incubation. The opaque cells produced more melanin than the white cells in whole cultures ( Fig. 2A) or cell pellets (Fig. 2B). Interestingly, when we used a WOR1 overexpression strain (WT + WOR1) in which the endogenous WOR1 was positively induced by ectopic Wor1 and the opaque phase was stabilized, the WOR1-stabilized opaque cells showed even higher absorption values than the WT opaque cells (WT + vector; Fig. 2A, B). This phenomenon is probably due to the unstable opaque state of the WT opaque cells in the minimal medium used for melanin production. Therefore, we used WOR1 overexpression-stabilized opaque cells (WT + WOR1) for later analyses. We tested the impact of Wor1 on melanin production in MTLa /a heterozygous cells. Because it is not possible to maintain the WT MTLa /a heterozygous cells in an opaque state in air conditions in vitro, we used ectopic Wor1 formed MTLa /a opaque cells to measure the melanin production. As expected, WOR1 overexpression-stabilized MTLa /a opaque cells (WT + WOR1) produced more melanin than the WT MTLa /a white cells (WT + vector) when they were incubated in the DOPA-containing medium and showed a higher absorption value at OD 310 nm or OD 470 nm in whole cultures (Fig. 2C) and cell pellets (Fig. 2D), indicating that Wor1 promoted melanin production in both MTL homozygous and heterozygous cells. However, WOR1 overexpression promoted less melanin production in MTLa /a heterozygous cells than in MTL homozygous cells, reflecting the impact of a1-a2 repression on melanin production.

The pimpled surface of opaque cells is correlated with externalization of melanin
White and opaque cells have distinct features [4]. Opaque cells are elongated with pimples on the surface, forming flat and opaque colonies. The mechanisms that control the dark color of opaque colonies and the pimpled surface of opaque cells are unknown. We speculated that the dark color and pimpled surface may be associated with melanin synthesis and externalization. To analyze the secretion of melanin, we collected homozygous WOR1-overexpressing opaque cells (MTLa/a WT + WOR1) and incubated them with DOPA for a 10-day growth period to observe melanin externalization by SEM and TEM using the method described previously [31,45]. SEM showed several types of opaque cell surfaces during the process of melanin externalization. Opaque cells exhibited a smooth surface during the early growth stage in the log phase (Fig. 3A, left), a bubbly form during the middle growth stage after 4-day incubation (Fig. 3A, middle) and a pimpled surface during the late growth stage after 10-day incubation (Fig. 3A, right). TEM also showed different patterns of melanin externalization corresponding to the three opaque growth stages.
Most of the dark flecks remained inside the opaque cells during the early stage (Fig. 3B, left); dark melanin particles were secreted constitutively outside the opaque cells during the middle stage (Fig. 3B, middle), and fewer dark particles were observed both inside and outside the opaque cells in the late stage (Fig. 3B, right). Walker et al. [31] reported that the presence of exogenous substrate DOPA is required for the externalization of melanin in C. albicans. Melanin particles were not observed in DOPA-free cultures [31]. To verify that the small dark flecks from opaque cells are indeed melanin, we compared the TEM images of opaque cells grown with or without the addition of DOPA for 4 days (Fig. S3). As expected, the cells cultured with DOPA were surrounded by melanin clumps, whereas melanin clumps were not observed in cells cultured in DOPA-free media. All these SEM and TEM findings suggest that the cell surface appearance was associated with melanin synthesis and externalization. During the early growth period, melanin was synthesized but barely secreted, and the opaque cell surface appeared smooth. With the increase in melanin secretion during the middle stage, the opaque cells presented a bubbly form. With the exhaustion of melanin externalization during the late stage, opaque cells exhibited a pimpled surface. Therefore, the pimpled surface of opaque cells described previously could be correlated with complete melanin externalization.
Fets function as pigment MCOs for melanin production in C. albicans Knowing that melanin production is usually catalyzed by a polyphenol oxidase (laccase) with an exogenous diphenolic as a substrate [30,31], we searched the C. albicans genome database (http://www.candidage nome.org) and found five predicted MCOs, which have been previously reported as Fets annotated as Fet3, Fet31, Fet33, Fet34 and Fet99 [29]. All five Fets contain three multicopper binding domains that are conserved in Fets such as Saccharomyces cerevisiae Fet3 [46] and laccases such as C. neoformans Lac1 (CnLac1) [23]. The conserved multicopper domains also exist in Homo sapiens ceruloplasmin (HsCp; Fig. 4A). As expected, the conserved histidine and cystatin residues required for copper binding were found in the multicopper domains, and all potential coordination sites for the three different types of copper ions were found in these regions (Fig. 4B). Interestingly, laccase-specific signature sequences, namely, L1 and L3, were present in the five Fets and CnLac1 (Fig. 4B). However, the L1 pattern was absent in HsCp. The proposed patterns L1 (H-W-H-G-x(9)-D-G-x(5)-Q-C-P-I) and L3 (H-Px-H-L-H-G-H) were suggested to be specific for laccases and proposed to distinguish laccases from other MCOs [26].
To determine whether the five Fets could function as laccase-like CnLac1, we constructed a series of MTL homozygous Fet deletion mutants by PCR-based homologous recombination [32]. Because four of the five Fet-encoding genes were clustered in a tandem order of FET31-FET99-FET3-FET34 (orf19.4211, orf19.4212, orf19.4213 and orf19.4215) on chromosome 6, and FET33 (orf19.943) was located on chromosome 5 alone (Fig. 4C), we constructed five single FET gene deletion mutant strains, including fet3D/D, fet31D/D, fet33D/D, fet34D/D and fet99D/D, four FET genes deleting mutant strain 4fetsD/D (FET31-FET99-FET3-FET34) and five FET genes deleting mutant strain 5fetsD/D for functional analysis and examined the ability of these mutant strains in melanin production using DOPA as a substrate. Knowing that white cells can also produce melanin, we cultured white phase cells in minimal medium with 1 mM DOPA and 5 mM GlcNAc and used the whole culture for measurement to simplify the procedures. Then, we measured the OD 310 and found that pigment production was reduced in each fet single gene with the mutant depleted, especially when the strains were cultured at 37°C (Fig. 4D), which reflects the contribution of each Fet to pigment oxidase activities. When the four clustered FET genes were deleted (4fetsD/D), pigment production was significantly decreased. The OD 310 was reduced 10-fold in the 4fetsD/D mutant compared with the WT when the cells were cultured at 22°C and 13fold when the cells were cultured at 37°C (Fig. 4D). After depletion of all five FET genes (5fetsD/D), pigment production was completely blocked (Fig. 4D). These results suggest that each Fet contributed to the activity of the pigment MCO. We further measured the global melanin-associated oxidase activity of these mutant cells using DOPA as a substrate. Specific oxidase activity was indicated as mU AbsÁmin À1 Álg protein À1 [23,37]. We found that melanin-associated oxidase activity was completely abolished in the 5fetsD/D mutant when the cells were cultured at 22 or 37°C (Fig. 4E). These data suggest that all five Fets function as pigment MCOs associated with melanin synthesis in C. albicans using exogenous phenolic substrates, and the four clustered FET gene products play major roles in melanin pigment production.
In this study, we used CAI4 background strain for Fet-mediated melanin analysis, because a gene, IRO1, which is involved in iron metabolism, is mutated in CAI4, although the CAI4 strain is derived from a WT strain SC5314 [36]. To avoid misunderstanding the functional role of Fets in melanin production, we compared the melanin production in SC5314 and CAI4 background strains. We screened and selected the MTLa/a homozygous strain of SC5314, the parent of CAI4, containing a WT IRO1 gene. In white cells, the melanin levels produced in SC5314 MTLa/a strain are similar to that in the CAI4 MTLa/a strain (Fig. S4A,  B). In opaque cells, the melanin productions are also similar in the two MTLa/a homozygous strains (Fig. S4C,D). We further measured the melanin productions in MTLa /a heterozygous strains of SC5314 and CAI4; the melanin productions are almost the same in these two heterozygous strains (Fig. S4E,F). Furthermore, we examined the MCO activities in the two MTLa/a homozygous strains. The MCO activities from the SC5314 MTLa/a strain are almost equal to that of the CAI4 MTLa/a strain at both 22 and 37°C (Fig. S5). Thus, absence of the IRO1 gene seems to have no obvious effect on the melanin production in the conditions we used. Considering the technical difficulty, and consistent with our previous results, we choose CAI4 or BWP17 and their homozygous descendants for the melanin production analysis.

Wor1 up-regulates FET expression in rich medium at physiological temperature of 37°C
Fet homologs were found in all Candida species. However, the four-gene FET clusters were present in only two Candida species, C. albicans and C. dubliniensis, and were absent in other Candida species or closely related fungal species, such as S. cerevisiae (Fig. 5A).
The existence of the four-gene Fet clusters in C. albicans and C. dubliniensis was probably correlated with melanin pigment production. FET33 homologs, however, existed in all Candida species and S. cerevisiae. Conversely, the four permeases encoding genes FTR1, FTR2, FTH1 and FTH2 were broadly present in Candida species, unlike that in S. cerevisiae, which had only two iron permeases encoding genes, FTR1 and FTH1. Therefore, the existence of Fet clusters and multiple permease homologs in opportunistic pathogenic fungi is of evolutionary importance. Knowing that Wor1 promoted melanin production in opaque cells (Fig. 2), we then tested the impact of the master regulator Wor1 on the expression of MCO encoding genes. Interestingly, all five FET promoters contained predicted Wor1 binding sites (Fig. S6). To determine whether Wor1 regulated FET gene transcription, we used MTL homozygous WOR1-stabilized opaque cells (WT + WOR1) that highly expressed WOR1 and WT white cells (WT + vector) that did not express WOR1 (Fig. 5B), to detect the mRNA levels of all five FET genes in C. albicans cells cultured in a YPD-rich medium. We found that WOR1 overexpression up-regulated FET3 and FET31 transcription at 22°C but had a subtle effect on the other FET genes (Fig. 5B). At the human physiological temperature (37°C), WOR1 overexpression up-regulated the expression of the FETs more obviously and increased the transcription of FET3, FET31, FET33 and FET34 by 6-, 5.7-, 7.5-and 11-fold, respectively (Fig. 5B, upper panels). Except for FET99, all four other FET genes were up-regulated by Wor1 at 37°C. The promoters of four permease-encoding genes, FTR1, FTR2, FTH1 and FTH2 [29], also contained the predicted Wor1 binding sites (Fig. S6). We subsequently examined the transcription level of the permease-encoding genes in WOR1-stabilized opaque cells and WT white cells. It was found that WOR1 overexpression up-regulated FTR1 and FTH1 transcription at 37°C and down-regulated FTR2 transcription but had a subtle effect on FTH2 transcription (Fig. 5B, lower panels). Our data suggest that Wor1 enhances the expression of FET genes in rich medium, especially at the human physiological temperature of 37°C.

Wor1 up-regulates FET expression in GMM-LA
Previous reports showed that the five Fets had distinct iron-dependent regulation patterns in response to iron concentration [29]. Unlike the growth condition of opaque cells, we used GMM to test the effect of iron on FET expression and analyzed the impact of Wor1 on FET expression in the minimal medium with a low (20 lM FeCl 3 ) or high (300 lM FeCl 3 ) iron concentration. Overnight-cultured WT white cells or WOR1-stabilized opaque cells from YPD medium were released into the iron-chelated minimal medium supplemented with FeCl 3 and grown at 22°C for 6 h or at 37°C for 3 h. It was found that the FET expression pattern in the GMM was different from that in YPD-rich medium (Fig. 6). In the GMM with regular ammonium The fet33D/D and fet34D/D mutants were introduced with a vector pPR671, and the other strains were transformed with a vector pBA1. The melanin production of each strain was detected by OD 310 nm and corrected the value by its own negative control cultured without DOPA. (E) Melanin-associated oxidase activity in WT (JYC5 + pBA1) and 5fetsD/D mutant (CDB11 + pBA1) cells. The white cells were cultured in YPD at 22 or 37°C and collected for extraction of total proteins. Specific oxidase activity was determined using DOPA as a substrate and indicated as mU AbsÁmin À1 Álg protein À1 . Bars, mean AE SD. Data are representative of at least three independent experiments, each with similar results. Data relative to OD 310 in WT cells. *P < 0.05, **P < 0.01, by Student's t-test. aa, amino acids.
sulfate (38 mM), Wor1 down-regulated FET99 expression in an iron-independent manner but had a subtle effect on the expression of other FETs (Fig. 6A,B). were up-regulated by Wor1 in GMM-LA with 20 lM FeCl 3 , and this up-regulation was independent of growth temperature (Fig. 6C). In GMM-LA with 300 lM FeCl 3 , three FETs (FET3, FET34 and FET99) were found to be up-regulated by Wor1 (Fig. 6D), indicating that FET genes were up-regulated by Wor1 in minimal medium with LA sulfate. Among the FETs, the FET33 expression level was relatively high in the GMM with high ammonium or GMM-LA, independent of iron concentrations. Expression of the other FETs varied depending on the growth conditions. Among the permease-encoding genes, Wor1 had a repressive effect on FTR1 in high-ammonium medium (Fig. 6A,B), but a repressive effect on FTR2 in LA medium (Fig. 6C,D). Based on the earlier-mentioned data, we concluded that the expressions of the FET and permease-encoding genes were controlled in multiple layers in response to different environmental stimuli.

CO 2 and GlcNAc induce FET expression and melanin production
Knowing that the physiological level of CO 2 induced white-opaque switching and stabilized the opaque phenotype at 37°C [47], we analyzed the impact of CO 2 on the expression of MCO encoding genes and melanin production and found that the transcription levels of FET31 and FET34 increased significantly when the MTL homozygous WT white cells were exposed to a high CO 2 concentration (20% CO 2 ) on YPD plates at 37°C for 1 day, when compared with the air condition (0.03% CO 2 ). However, the FET3, FET33 and FET99 expression levels remained unchanged significantly (Fig. 7A). Consistently, higher melanin production was detected at a high CO 2 concentration (Fig. 7B). GlcNAc was reported to be another white-opaque switching inducer [48] that promoted C. albicans melanin production [31]. The combination of GlcNAc and CO 2 even induced white-opaque switching in MTLa /a heterozygous cells [7]. Thus, we tested the effect of GlcNAc on FET gene expression. After 1 day of treatment with GlcNAc on YPG plates in 20% CO 2 , the FET31 and FET34 mRNA levels were dramatically increased to a level higher than that with 20% CO 2 alone. FET3 and FET33 expression were also increased; however, FET99 remained unchanged significantly (Fig. 7A). More melanin production was detected under high CO 2 condition and GlcNAc (Fig. 7B). Our data showed that a high CO 2 concentration together with GlcNAc promoted melanin production by increasing the expression level of the FET genes at the host physiological temperature of 37°C.
Fets contribute to the dark color, but not the elongated shape of opaque cells C. albicans opaque cells exhibit an elongated shape, and opaque colonies exhibit a dark color [4]. We found that the dark color of the opaque colonies was associated with increased melanin pigment formation and Wor1-induced up-regulation of FET expression. We then investigated whether the elongated form of the opaque cells was also correlated with Fet expression. To promote and maintain the 'opaque' phase, we introduced ectopically expressed Wor1 into homozygous WT cells and 4fetsD/D and 5fetsD/D mutant cells to induce their endogenous WOR1. Similar to WT cells, Wor1-induced 4fetsD/D or 5fetsD/D mutant cells exhibited an elongated shape, which is different from the non-Wor1-induced round-shaped white form (Fig. 8A). We also observed colonial color. Unlike the dark and flat WT opaque colonies, WOR1-promoted 4fetsD/D and 5fetsD/D mutants exhibited smooth, white and small hemispherical colonies (Fig. 8B). The colonies of the WOR1-overexpressing 4fetsD/D and 5fetsD/D retained the white-looking shape, but smaller. Thus, Wor1-induced up-regulation of the Fets contributed to the dark color, but not to the elongated shape of opaque cells.
To confirm that the dark color formation was due to Fet-mediated melanin pigment production, we cultured the WOR1-overexpressed WT cells and 4fetsD/D and 5fetsD/D mutant cells in liquid minimal medium at 22°C and detected pigment production using DOPA as the substrate. Compared with the WT cell pellets that turned black, the cell pellets of 4fetsD/D and 5fetsD/D mutants overexpressing WOR1 remained white (Fig. 8C), resembling the non-Wor1-induced WT white cells. We further measured the global melaninassociated oxidase activities using DOPA as the substrate. In minimal medium with 1 mM DOPA and 5 mM GlcNAc at 22°C, WOR1 overexpression WT opaque cells induced a higher melanin-associated oxidase activity, reaching 0.732 mU AbsÁmin À1 Álg protein À1 , which was approximately 2.7-fold higher than that from the non-Wor1-induced WT white cells (Fig. 8D). In the 5fetsD/D mutant, the melanin-associated oxidase activity was completely depleted, and WOR1 overexpression could not restore the enzyme activity (Fig. 8D). The depleted melanin-associated oxidase activity of the 5fetsD/D mutant could be restored by reintroducing FET34 back to the mutant (Fig. 8E). To determine the function of Fets as melanin-associated oxidases, we performed in vitro biochemical assays. The total protein extracts of C. albicans opaque phase cells were loaded into nonreducing SDS-polyacrylamide gels and incubated with L-DOPA. In comparison with a commercial laccase, positive laccase activity was detected by the protein extracts of WT cells, as shown by a dark band (Fig. 8F), demonstrating that L-DOPA polymerized to form melanin. The extracts from the 5fetsD/D mutant cells did not show a dark band, and the extracts from the FET34 complementary strain showed a dark band (Fig. 8F), demonstrating the laccase activity of the reintroduced Fet34. Among the four plasma membrane (PM) Fets, Fet34 is the main PM Fet for iron acquisition under iron-limiting conditions [29]. Considering the high similarities but distinct expression patterns among the four PM Fets, we chose Fet34 under the control of a constitutive promoter (ADH1p) for complementation assays. Consistent with its Fet activity, Fet34 showed melanin-associated oxidase activity (Fig. 8E,F). Using SEM, we further compared the cell surfaces of WT opaque and 5fetsD/D mutant 'opaque' cells from the late growth stage. As expected, the WT opaque cells showed a pimpled surface, whereas the 5fetsD/D mutant 'opaque' cells showed a smooth surface (Fig. 8G), indicating that the pimpled surface of opaque cells was associated with melanin synthesis and externalization. We also tested the effects of Fets on mating efficiency. The quantitative mating assay was performed as described previously [43]; MTLa 5fets mutant cells showed no significant difference with MTLa WT when mating with MTLa WT white or opaque cells (Fig. S7), reflecting no impact of Fets on mating processes. These data suggest that the effects of Wor1 on the opaque cell shape and opaque colony color were regulated by two different mechanisms. Wor1-up-regulated Fets contributed to the dark color, but not to the elongated shape of the opaque cells.

Melanization protects C. albicans against BMDM phagocytosis and oxidative killing
Melanin is an efficient free radical scavenger [49,50] and confers resistance to UV light by absorbing a broad spectrum of electromagnetic waves, thus preventing photo-induced damage. To investigate the functional role of melanin products in protecting C. albicans against damage, we first performed a BMDM macrophage phagocytosis assay. As shown in Fig. 9A, the DOPA preincubated C. albicans cells survived better than the non-DOPA-treated cells and showed a lower phagocytosis ratio and phagocytosis index. These results suggest that melanized C. albicans  cells are resistant to BMDM phagocytosis and protect themselves against host immune killing. Next, we tested the melanization of C. albicans on other antistress responses and found that after 30 s of treatment with 30 000 lJÁcm À2 UV, approximately 10% of the melanized cells survived, whereas those nonmelanized cells almost all died (Fig. 9B, left). After 2 h of incubation with 1.5 mM H 2 O 2 , approximately 60% of the DOPA-pretreated cells survived, but only 4% of the non-DOPA-treated cells survived (Fig. 9B, right). The DOPA-pretreated cells were also more resistant to the antifungal drugs. Approximately 60% of the DOPApretreated cells survived when they were incubated with caspofungin (32 lgÁmL À1 ) or AMB (1 lgÁmL À1 ) for 2 h, whereas few non-DOPA-treated cells survived (Fig. 9C). To rule out the possibility that the DOPApretreatment itself, but not melanin production, was responsible for the protective effects observed, we included the 5fetsD/D mutant as controls. Compared with the WT cells, the melanin production-defective mutant cells were equally sensitive to all the damage with or without DOPA pretreatment (Fig. 9). Therefore, melanin production with exogenous DOPA substrate protected C. albicans cells against BMDM macrophage phagocytosis, UV radiation, H 2 O 2 oxidation and antifungal drugs.   (Fig. 10, upper panels). At an opaque maintaining temperature of 22°C in vitro, the growth of opaque cells overexpressing Wor1 was similar to that of WT white cells (Fig. 10, lower panels). To eliminate the effect of ferric ions on cell growth, all plates were supplied with 20 lM FeCl 3 [29]. All the single fet gene deletion mutants had no growth defect on the H 2 O 2 -containing plates at 22°C, but 4fetsD/D and 5fetsD/D mutant cells were sensitive to H 2 O 2 (Fig. 10, lower panel). Given the slow growth rate of the 4fetsD/D and 5fetsD/D mutant cells, we incubated them on the plate for a longer time. Wor1 overexpression in the 4fetsD/D or 5fetsD/D mutant could not recover cell growth in the medium containing H 2 O 2 , but reintroducing FET34 from the ADH1 promoter into the 4fetsD/D or 5fetsD/D mutant restored cell growth. All these findings suggest that the C. albicans MCO Fets have antioxidant capacity.

Discussion
Previous studies [30,31] have shown that C. albicans can produce dark-pigmented melanin in vitro and during infection. However, no gene encoding laccase has been identified thus far. Here, we demonstrated that the MCO Fets, which were previously reported as Fets [29], function as pigment MCOs and promote melanin production under the control of Wor1 in opaque cells (Fig. 11). C. albicans can undergo white-opaque switching, and the white and opaque cells have known distinct features. We found that the dark color of opaque colonies was associated with increased FETs expression and melanin production, and the pimpled surface of opaque cells was associated with melanin externalization.
We also found evidence that Wor1 had additional regulatory impacts on the expression of FETs, along with iron. The five FET genes had distinct expression patterns in response to iron concentrations, among which FET34 and FET99 were down-regulated by high iron concentrations and up-regulated by low iron concentrations, FET31 was slightly up-regulated by low iron concentrations, and FET3 and FET33 exhibited iron-independent expression [29,34]. To examine the impact of Wor1 on the FET gene transcription, we cultured MTL homozygous WOR1-stabilized opaque cells (WT + WOR1) and WT white cells (WT + vector) in rich or poor medium under different growth conditions. In YPD-rich medium, Wor1 overexpression greatly up-regulated four (FET3, FET31, FET33 and FET34) of the five FET genes, especially at the host physiological temperature of 37°C, but had a subtle effect on FET99 expression. In GMM, all five FET genes were up-regulated by Wor1 with LA. However, FET99 was down-regulated by Wor1 in minimal medium with high ammonium. In addition, high CO 2 combined with high concentrations of GlcNAc at 37°C mimicking the host GI tract environment also up-regulated FET31 and FET34 expression. As investigated previously, genome-wide gene expression comparisons showed different FET gene expression patterns in white and opaque cells [29,[52][53][54][55]. Because different media and growth conditions were used in each report, we speculated on whether the components in the media and growth conditions had any impact on the regulation of FET genes. Therefore, the expression of FET genes is controlled in multiple layers in response to different environmental stimuli, which reflect the multiple enzymatic activities of Fets in different host niches.
The host GI tract is a complex and changeable environment where large numbers of microorganisms live inside. Knowing that trillions of microorganisms reside in the human GI tract, the gut microbiota is now regarded as a virtual endocrine organ that produces abundant chemicals of hormonal nature, such as the precursor to neuroactive compounds L-DOPA [56]. Therefore, it is especially important for commensal pathogen C. albicans to rapidly adapt to diverse host niches, furnishing iron and other chemicals in different forms and levels. Under iron-limited conditions, the Fets Fet34 and Fet99 were up-regulated and partnered with membrane permeases to form a high-affinity iron transporter to acquire iron. In the GI tract, which is thought to be replete in GlcNAc, and with high CO 2 [7,48], the up-regulated Fets function as pigment MCOs to catalyze the exogenous L-DOPA or other phenolic substrates to form melanin pigment. Given that C. albicans is a commensal resident in the human gut, C. albicans can undergo white-to-opaque switching in host gut conditions [7,47,48], and Wor1 overexpression triggers white-to-GUT switching and promotes C. albicans commensalism in the mouse gut [3]. Our finding that Wor1 up-regulated Fets and subsequently catalyzed melanin production may explain the different appearances between the opaque and GUT forms. Wor1 can promote white-to-opaque switching in MTL homozygous cells and trigger whiteto-GUT switching in MTLa /a heterozygous cells. According to our data in Fig. 2, the melanin production in MTL homozygous cells overexpressing Wor1 was higher than that in MTLa /a heterozygous cells overexpressing Wor1. We speculate that the pimpled surface of the opaque form in MTL homozygous cells may be associated with the exhaustion of melanin externalization, and that the smooth surface of GUT form in MTLa /a heterozygous cells is probably associated with less melanin secretion.
Melanins are used by microbes to protect pathogens against host immune responses [57], and conversely used by the host to defend against microbes [58]. Fungal melanin also can protect microbes against oxidants, UV, heat, enzymatic degradation, and antimicrobial compound stress [50]. DOPA-melanin plays an immunomodulatory role in C. neoformans infection [59,60] by inhibiting host cell phagocytosis and cytokine production and reducing the toxicity of microbicidal peptides. DHN-melanin from Aspergillus fumigatus facilitates the microbes to inhibit host cell apoptosis [61] and spread within the host niche [62]. DHN-melanin can also be sensed by the host through a melanin-sensing C-type lectin receptor (MelLec), which plays a crucial role in the control of systemic A. fumigatus infection in both mice and humans [63]. The C. albicans-produced melanin not only contributes to the color formation of opaque cells but also plays a protective role in C. albicans infection. Melanized C. albicans cells are more resistant to BMDM phagocytosis, radiation and oxidative damage and are tolerant to antifungal drugs.
In addition, we found that Fet MCOs had antioxidant capacity and functionally resembled the mammalian ceruloplasmin. Ceruloplasmin is a blue-colored plasma protein with multiple physiological functions, including copper transport, oxidation of organic amines, radical scavenging, Fet activity and antioxidant activity [28,51]. Ceruloplasmin inhibits the ferrous iondependent formation of hydroxyl radicals in the Fenton reaction, scavenges ROS and protects hepatocytes from oxidative damage in orthotopic liver transplantation [51]. C. albicans acquires iron from specific host molecules in regions where iron is scarce, while also defending against iron overdose toxicity in regions where iron occurs in excess [11,64]. Therefore, Fig. 11. Wor1-up-regulated Fets function as pigment MCOs required for pigment formation in opaque cells and contribute to antioxidant reaction. C. albicans presents as a commensal resident in the human gut, which is thought to be replete in GlcNAc, and with high CO 2 . WOR1 overexpression promotes white-to-opaque or white-to-GUT switching. Wor1 up-regulates the FETs expression in rich media at the physiological temperature of 37°C or in glucose minimal media with LA. The increased Fets function as pigment MCOs to promote melanin production using exogenous DOPA or other phenolic substrates. At low iron host niches, the up-regulated Fets function as Fets to facilitate iron acquisition. The Fets also have antioxidant capacity to protect the C. albicans against oxidative damage. The distinct expression patterns of the FET genes and the diverse functions of the Fets reflect the ability of C. albicans to rapidly adapt to diverse host niches and survive in the host.
C. albicans possesses five Fets representing the functional diversity of the enzymes, as Fets required for iron acquisition and laccases/pigment MCOs required for melanin production and antioxidant ability, befitting its commensal-pathogenic lifestyle.
Biotin-labeled (North2South TM Biotin Random Prime DNA Labeling Kit, Thermo). Fig. S2. Total proteins were estimated by the Coomassie brilliant blue method.   Fig. S5. Melanin-associated oxidase activity in MTLa/ a SC5314 and CAI4 + V (JYC5 + V) white cells. The white cells were cultured in YPD at 22 or 37°C and collected for extraction of total proteins. Specific oxidase activity was determined using DOPA as a substrate and indicated as mU AbsÁmin À1 Álg protein À1 . Bars, mean AE SD. Data are representative of at least three independent experiments, each with similar results. Data are relative to OD 310 in SC5314 cells. ns, no significance, by Student's t-test. Fig. S6. Prediction of Wor1 binding sites on the promoters according to previous reports. Fig. S7. Mating efficiency of the MTLa 5fets mutant. The tester WT a strain used is CHY477 (mtla1/MTLa URA3 HIS1 ade2). JYC5 is WT MTLa ura3 ADE2 strain. CBD11 is 5fets MTLa ura3 ADE2 mutant. Mating efficiency = mean AE SD. Data relative to mating efficiency in the WT cell. P White = 0.85, P Opaque = 0.52, by Student's t-test. ns, no significance.