Filament Negative Regulator CDC4 Suppresses Glycogen Phosphorylase Encoded GPH1 That Impacts the Cell Wall-Associated Features in Candida albicans

We have previously identified Candida albicans GPH1 (orf19.7021) whose protein product was associated with C. albicans Cdc4. The GPH1 gene is a putative glycogen phosphorylase because its Saccharomyces cerevisiae homolog participates in glycogen catabolism, which involves the synthesis of β-glucan of the fungal cell wall. We made a strain whose CaCDC4 expression is repressed, and GPH1 is constitutively expressed. We established a GPH1 null mutant strain and used it to conduct the in vitro virulence assays that detect cell wall function. The in vitro virulence assay is centered on biofilm formation in which analytic procedures are implemented to evaluate cell surface hydrophobicity; competence, either in stress resistance, germ tube formation, or fibronection association; and the XTT-based adhesion and biofilm formation. We showed that the constitutively expressed GPH1 partially suppresses filamentation when the CaCDC4 expression is repressed. The C. albicans Gph1 protein is reduced in the presence of CaCdc4 in comparison with the absence of CaCdc4. Compared with the wild-type strain, the gph1Δ/gph1Δ mutant displayed a reduction in the capability to form germ tubes and the cell surface hydrophobicity but an increase in binding with fibronectin. Compared with the wild-type strain, the gph1Δ/gph1Δ mutant showed a rise in adhesion, the initial stage of biofilm formation, but displayed a similar capacity to form a mature biofilm. There was no major impact on the gph1Δ/gph1Δ mutant regarding the conditions of cell wall damaging and TOR pathway-associated nutrient depletion. We conclude that GPH1, adversely regulated by the filament suppressor CDC4, contributes to cell wall function in C. albicans.


Introduction
The opportunistic human fungal pathogen Candida albicans [1] is a member of the normal microflora on mucosal surfaces in healthy persons [2] but can cause vulvovaginal candidiasis in women [3,4] and oral [5,6] and systemic candidiasis in debilitated and immunocompromised patients [7][8][9][10]. C. albicans can grow in a wide variety of morphological forms, from the ellipsoid blastospore to various filamentous types [11][12][13][14]. A great effort has been made to reveal the underlying mechanism of C. albicans morphogenesis because it is proven to be coupled with virulence and pathogenesis [15][16][17][18]. However, research

General Manipulation, Media, and Growth Conditions
The E. coli strain DH5α was used for regular manipulation of the plasmids. All C. albicans strains (Table 1) were derived from either the clinically isolated wild-type strain SC5314 [64] or the auxotrophic strain BWP17 (arg4/arg4 his1/his1 ura3/ura3) [65]. The routine usage of media and growth conditions for the strains of E. coli and C. albicans were performed as described previously [66]. The pH of the SD medium with or without agar was adjusted to 7 by 100 mM HEPES after autoclaving because the default SD medium is acidic, which can suppress the filamentation of C. albicans. The E. coli strain DH5α was transformed with plasmid DNA by CaCl 2 as described [67] or by electroporation [68]. C. albicans strains were transformed using the LiAc/PEG/ssDNA method [69] or electroporation [70].

Nucleic Acid Extraction and PCR Analysis
Candida albicans cells were grown to the mid-log phase, and genomic DNA was isolated using a MasterPure™ Yeast DNA Purification Kit (Epicentre, Madison, WI, USA), following the manufacturer's instructions, as described previously [35]. The total RNA derived from cells cultured to the mid-log phase was extracted using a MasterPure TM Yeast RNA Purification Kit (Epicentre, Madison, WI, USA), following the manufacturer's instructions, as described previously [51]. Subsequently, 5 µg of total RNA was used to generate cDNA using a SuperScript III Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions. The cDNA was then subjected to PCR with a pair of GPH1-specific primers, the front of GPH1-F_SpeI, and the front of GPH1-R_SpeI (Table 2), annealing the downstream of the GPH1 coding sequence with a predictive product of 623 bp. The primers CaACT1-F and CaACT1-R were used to generate a C. albicans ACT1-specific product that was used as a loading control. To confirm the correctness of the GPH1 deletion strain, Southern blotting analyses with DIG-labelled probe amplified by a pair of primers, GPH1-probe-F and GPH1-probe-R (Table 2), were performed as previously described [74].

Protein Extraction and Western Blotting
The total protein was extracted from the C. albicans cells, as described previously (74). The protein was partially purified from cells containing the p6HF-ACT1p plasmid with the open reading frame of the gene integrated at RPS1 capable of generating a tagged (6×His and FLAG) protein using Ni 2+ -NTA-agarose beads (Qiagen, Germantown, MD, USA), as previously described [75]. Precipitated proteins were resolved using 10% SDS-PAGE and transferred electrophoretically to PVDF membranes (Pall Corporation, Port Washington, NY, USA). They were then probed with a polyclonal antibody to FLAG (Sigma) in a 1:2000 dilution and visualized using a SuperSignal West Pico Chemiluminescent Substrate Kit (Pierce). The proteins detected were recorded with a Luminescent Image Analyzer (FUJIFILM LAS-1000) and analyzed by ImageGauge 3.46 and L Process v 1.96 (FUJIFILM). ImageJ (National Institutes of Health, Bethesda, MD, USA) was used to quantify the levels of proteins.

Germ Tube Formation Assay
The morphological plasticity of C. albicans plays a vital role in biofilm maturation, as previously discussed. Germ tube formation is a prerequisite of the development of hyphal and pseudohyphal forms, and hence the length of the germ tube under hyphal induction condition is used for the assessment of yeast-to-hypha transition. To promote germ tube formation, 1 × 10 6 C. albicans cells/mL are transferred into the cell culture medium RPMI 1640 supplemented with 10% (v/v) fetal calf serum (FCS), 2 mM L-glutamine, penicillin (100 U/mL), and streptomycin (100 µg/mL) and seeded into a 24-well plate. After 1 h incubation at 37 • C, cells were visualized and recorded, and the germ tube length was determined with the Photoshop 6 software.

Cell Surface Hydrophobicity Assay
Cell surface hydrophobicity (CSH) was measured using the microbial adhesion assay to hydrocarbons (MATH) [76]. The assay was conducted as previously described [77]. Briefly, C. albicans cells grown to the mid-log phase at 30 • C were collected and washed twice with PBS. The cell suspension with an OD 600 between 0.4 and 0.5 was set up in PBS (A0); 3 mL of the cell suspension was overlaid with 0.4 mL of the hydropho-bichydrocarbon, n-hexadecane (SIGMA, H6703). Following robust vortexing, the phases were left to separate for 10 min at 30 • C, and the OD 600 of the aqueous phase was quantified (A1). The percentage of hydrophobicity is calculated as follows: hydrophobicity (%) = [1 − (A1/A0)] × 100.

Fibronectin (FN)-C. albicans Association Assay
To specifically assess the binding of fibronectin with C. albicans cells, C. albicans cells (1 × 10 6 ) from the mid-log phase were sub-cultured in 2 mL RPMI 1640 medium supplemented with 0.0001% human fibronectin (Sigma-Aldrich, St. Louis, MO, USA) for 1 h at 30 • C at 200 rpm. Subsequently, cells were examined microscopically or washed three times with 2 mL PBS before harvesting. Cells were resuspended in 80 µL PBS, with the addition of 20 µL 5× sample loading buffer (with β-ME), boiled for 10 min, and rested on ice for 10 min. The cells were spun down for 10 min, and 90 µL of the supernatant was transferred to a fresh tube prior to immunoblot analysis with specific FN-specific antibody [78].

Adhesion Assay
The adhesion assay was conducted as previously described [77]. In brief, C. albicans cells were grown overnight in YPD at 37 • C with 180 rpm agitation. Cells were collected by centrifugation for 5 min at 10,000× g, washed with PBS, and standardized to 5 × 10 7 C. albicans cells/mL in RPMI-1640 medium supplemented with L-glutamine and buffered with MOPS acid. Next, 100 µL (5 × 10 6 cells) aliquots of the cell suspension were placed in each well of a nonpyrogenic polystyrene flat-bottom 96-well microtiter plate and incubated for 1 h at 37 • C. The wells were washed three times with 10 mM PBS before being quantified by the XTT (2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5carboxanilide) reduction assay [79]. Briefly, the adherent cells were incubated with XTT (0.5 mg/mL + 1 µM menadione in Ringer' solution) in the dark. The absorbance of reduced XTT was measured in a microtiter plate reader at 490 nm, as described previously [47].

Biofilm Formation Assay
To assess the ability of C. albicans cells to form biofilm, cells of the strains were prepared as described in the adhesion assay, except that the cells were standardized to 1 × 10 6 C. albicans cells/mL after washing. Then, 200 µL aliquots (2 × 10 5 cells) of the C. albicans cell suspension was placed in the wells of a 96-well microtiter plate and incubated for 48 h at 37 • C before XTT reduction assay.

Spotting Assay
The spotting assays were carried out as previously described [74]. Concisely, cells of the C. albicans strains were grown in YPD to the mid-log phase. The cultured strains were diluted to an optical density of 1.0 at OD 600 (approximately 2 × 10 7 cells/mL) and then serially diluted from 10 7 to 10 2 cells/mL. The diluted cultures were spotted on agar plates at a volume of 5 µL and left to grow.

Cellular Image Observation and Recording
The images of the cultured cells were recorded with a Nikon 50i microscope at 400× magnification. Colonies were photographed with a MEIJI stereoscopic microscope EMZ5 at 40× magnification. The monographs were digitized and processed using Adobe Photoshop software.

Statistical Analysis
Unless stated otherwise, three independent assays were conducted, and each sample was assayed in triplicate. Statistical analyses were performed using GraphPad Prism software, v.8.0 (GraphPad Software, Inc., La Jolla, CA, USA), by one-way analysis of variance (ANOVA), followed by Tukey's post hoc analysis. The results are expressed as the mean ± standard deviation (SD). The p < 0.05 indicate a statistically significant difference. The asterisks used to indicate statistically significant difference are as follows: * p < 0.05; ** p < 0.01; *** p < 0.005; **** p < 0.001.

The Filamentous Growth Caused by the Repressed CaCDC4 Expression Could Be Partially Suppressed by the Constitutive GHP1 Expression in C. albicans
We previously identified the Gph1 protein as a C. albicans Cdc4-interactor [48]. To understand the functional link between CaCDC4 and GPH1, a C. albicans strain capable of repressing CaCDC4 expression with methionine and cysteine (Met/Cys) and constitutively expressing GPH1, together with those expressing CaCDC4 and none, were created ( Figure 1A). To evaluate the outcome of GPH1 expression on the filamentous growth of cells with CaCDC4 expression being repressed, the cells of the above strains, together with their parental strain, were plated onto YPD rich media ( Figure 1B) or were grown in the minimum media ( Figure 1C) with or without 2.5 mM Met/Cys. After obtaining the constructed strains, we unexpectedly found that the expression of CaCDC4 of CaCDC4 M3/− on plates of YPD-rich media appeared to be repressed; hence, they grew as filamentous forms but could mostly be suppressed when constitutively expressing CaCDC4 from the p6HF-based plasmid ( Figure 1B). As expected, the constitutive expression of CaCDC4 but not the empty plasmid completely suppressed the filamentous mode of growth when the CaCDC4 expression was repressed ( Figure 1C). It appeared that the filaments as a result of the repression of CaCDC4 expression were mixed with the hyphal and pseudohyphal cells, similar to previous observations by us [40] and others [28]. Significantly, the constitutive GPH1 expression could somewhat suppress the filamentous development when the CaCDC4 expression was repressed ( Figure 1C). Curiously, the suppression of filamentous growth by GPH1 on SD plates appeared to be indistinct ( Figure 1D). Overall, these results suggest that GPH1 is functionally related to CaCDC4 regarding the control morphogenesis and that GPH1 negatively modulates hyphal formation.
repressing CaCDC4 expression with methionine and cysteine (Met/Cys) and constitutively expressing GPH1, together with those expressing CaCDC4 and none, were created ( Figure 1A). To evaluate the outcome of GPH1 expression on the filamentous growth of cells with CaCDC4 expression being repressed, the cells of the above strains, together with their parental strain, were plated onto YPD rich media ( Figure 1B) or were grown in the minimum media ( Figure 1C) with or without 2.5 mM Met/Cys. After obtaining the constructed strains, we unexpectedly found that the expression of CaCDC4 of CaCDC4 M3/ ̶ on plates of YPD-rich media appeared to be repressed; hence, they grew as filamentous forms but could mostly be suppressed when constitutively expressing CaCDC4 from the p6HF-based plasmid ( Figure 1B). As expected, the constitutive expression of CaCDC4 but not the empty plasmid completely suppressed the filamentous mode of growth when the CaCDC4 expression was repressed ( Figure 1C). It appeared that the filaments as a result of the repression of CaCDC4 expression were mixed with the hyphal and pseudohyphal cells, similar to previous observations by us [40] and others [28]. Significantly, the constitutive GPH1 expression could somewhat suppress the filamentous development when the CaCDC4 expression was repressed ( Figure 1C). Curiously, the suppression of filamentous growth by GPH1 on SD plates appeared to be indistinct ( Figure 1D). Overall, these results suggest that GPH1 is functionally related to CaCDC4 regarding the control morphogenesis and that GPH1 negatively modulates hyphal formation.  The diagram illustrates the strains used. The cells were (B) plated on YPD plate or were grown in the SD media (C) or plate (D) with or without 2.5 mM Met/Cys. The "φ" represents empty plasmid p6HF-ACT1p. Bars represent 10 µm. "18" and "26" represent different isolates of strains with p6HF-ACT1p-GPH1.

C. albicans Gph1 Protein Being Reduced in the Presence of CaCdc4 May Be the Result of Polyubiquitin-Proteasome-Dependent Degradation
Because GPH1 negatively modulates hyphal development ( Figure 1B,C), we presumed that Gph1, like Sol1 [28] and our recently characterized Thr1 [51], is the target of CaCdc4 and is governed by ubiquitination for degradation. To assess the possible regulation of CaCdc4 and Gph1, cells of the same strains as in Figure 1 were grown in the minimum media with or without Met/Cys, and the proteins were extracted and subjected to Western blotting analysis. The protein level of Gph1, with or without expression of CaCDC4, showed no apparent difference ( Figure 2A). However, the de-repressed CaCDC4 expression led to a reduction of the level of Gph1 protein with translation inhibitor cycloheximide ( Figure 2B,C), suggesting that Gph1 protein is targeted for degradation by CaCdc4. The results indicate that the CaCdc4 negatively regulates the level of Gph1 protein and that the Gph1 adversely controls filamentation.

C. albicans Gph1 Protein Being Reduced in the Presence of CaCdc4 May Be the Result of Polyubiquitin-Proteasome-Dependent Degradation
Because GPH1 negatively modulates hyphal development ( Figure 1B,C), we presumed that Gph1, like Sol1 [28] and our recently characterized Thr1 [51], is the target of CaCdc4 and is governed by ubiquitination for degradation. To assess the possible regulation of CaCdc4 and Gph1, cells of the same strains as in Figure 1 were grown in the minimum media with or without Met/Cys, and the proteins were extracted and subjected to Western blotting analysis. The protein level of Gph1, with or without expression of CaCDC4, showed no apparent difference (Figure 2A). However, the de-repressed CaCDC4 expression led to a reduction of the level of Gph1 protein with translation inhibitor cycloheximide ( Figure 2B,C), suggesting that Gph1 protein is targeted for degradation by CaCdc4. The results indicate that the CaCdc4 negatively regulates the level of Gph1 protein and that the Gph1 adversely controls filamentation.

Cells Overexpressing or Lacking GPH1 Bear No Morphological Changes but Those without GPH1were Apt to Aggregate for a More Extended Period in Normal Growth Condition
To evaluate the role of GPH1 in morphogenesis, we made a strain Tet-on-GPH1 (Supplementary Figure S1A), where the expression of GPH1 is induced by doxycycline (Dox). The massive induction of GPH1 expression, both transcriptionally (Supplementary Figure  S1B) and translationally (Supplementary Figure S1C), was confirmed. However, no morphological alteration could be observed, suggesting that cells overexpressing GPH1 were unable to interfere with cellular morphology in C. albicans (Supplementary Figure S1D). To further determine whether GPH1 has a role in morphogenesis, the CaSAT1-flipper method (73) was used to create the GPH1 homozygous null mutant (gph1Δ/gph1Δ). Southern blotting analyses were used to validate the mutants. As shown in Supplementary Figure S2B, the PvuII-digested genomic DNAs extracted from each of the strains could be detected with a probe specific to the GPH1 locus flanked with PvuII sites generating the expected sizes (Supplementary Figure S2A). Therefore, we proved that the created mutants were correct. By RT-PCR analyses, as expected, the GPH1 expression was only observed in the wild-type SC5314 (GPH1/GPH1), the GPH1 heterozygous null mutant

Cells Overexpressing or Lacking GPH1 Bear No Morphological Changes but Those without GPH1were Apt to Aggregate for a More Extended Period in Normal Growth Condition
To evaluate the role of GPH1 in morphogenesis, we made a strain Tet-on-GPH1 (Supplementary Figure S1A), where the expression of GPH1 is induced by doxycycline (Dox). The massive induction of GPH1 expression, both transcriptionally (Supplementary Figure S1B) and translationally (Supplementary Figure S1C), was confirmed. However, no morphological alteration could be observed, suggesting that cells overexpressing GPH1 were unable to interfere with cellular morphology in C. albicans (Supplementary Figure S1D). To further determine whether GPH1 has a role in morphogenesis, the CaSAT1-flipper method (73) was used to create the GPH1 homozygous null mutant (gph1∆/gph1∆). Southern blotting analyses were used to validate the mutants. As shown in Supplementary Figure S2B, the PvuII-digested genomic DNAs extracted from each of the strains could be detected with a probe specific to the GPH1 locus flanked with PvuII sites generating the expected sizes (Supplementary Figure S2A). Therefore, we proved that the created mutants were correct. By RT-PCR analyses, as expected, the GPH1 expression was only observed in the wild-type SC5314 (GPH1/GPH1), the GPH1 heterozygous null mutant (GPH1/gph1∆), and the GPH1 complementation strain gph1∆/gph1∆+GPH1, but not in the homozygous null mutant (gph1∆/gph1∆) (Supplementary Figure S2C). The expression level of GPH1 was approximately two-fold less in GPH1/gph1∆) and gph1∆/gph1∆+GPH1 than that of SC5314 (Supplementary Figure S2C). No apparent morphological alteration between the GPH1 null mutant and the wild-type in the normal growth condition could be found (Supplementary Figure S3A). However, after prolonged incubation, compared with cells of the wild-type, those of the GPH1 null mutants exhibited an increase in aggregation (Supplementary Figure S3B), suggesting that cells lacking GPH1 may alter the properties of the cell wall, consequently promoting the cell-cell interaction.

The GPH1 Null Mutant Shows No Growth Defect in Normal Growth Condition and Various Stressful Conditions
To determine if GPH1 can have a general effect on growth, cells of the gph1∆/gph1∆, together with the gph1 heterozygous null mutant (GPH1/gph1∆), the GPH1 complement strain gph1∆/gph1∆+GPH1, and the wild-type SC5314 (GPH1/GPH1), were grown in either liquid or semi-solid YPD. Cells lacking GPH1 showed no growth defect both in YPD liquid medium (Supplementary Figure S4A,B) and YPD plate (Supplementary Figure S4C), suggesting that GPH1 bears no role in the maintenance of growth. GPH1 is involved in the synthesis of β-glucan and energy conversion in S. cerevisiae; hence, we presumed that this is common in fungi, including C. albicans. We set up the spotting assays with conditions including various cell wall damaging agents, different carbon sources, and distinct nutrientdepleted states at either 30 or 37 • C. However, with cells of the wild-type, those of the GPH1 null mutant showed no consequence in growth ability (Supplementary Figure S5), suggesting that either GPH1 plays no role or the presence of GPH1 redundant genes in the cell wall structure and energy conversion. As a result, we have sought alternative assays that may reveal the GPH1 function.

C. albicans Cells Lacking GPH1 Reduce the Ability to Form Germ Tube in Response to the Hypha-Inducing Condition
While the GPH1 homozygous null mutant of C. albicans did not directly contribute to the yeast-to-hypha transition, we sought to determine the influence of the hyphal induction condition on the mutant. Cells of the GPH1 homozygous null mutant (gph1∆/gph1∆), together with the GPH1 heterozygous null mutant (GPH1/gph1∆), the GPH1 complement strain gph1∆/gph1∆+GPH1, and the wild-type SC5314 (GPH1/GPH1), were grown exponentially in YPD and transferred to RPMI 1640 supplemented with 10% fetal calf serum at 37 • C and were subjected to analysis of the length of the germ tube. The ability of the GPH1 homozygous null mutant to form a germ tube was reduced compared to that of the GCN4 homozygous null mutant, which is known to be impaired in filamentation under the hyphal induction condition [80] (Figure 3), suggesting that C. albicans GPH1 is indirectly involved in the yeast-to-hypha transition.

C. albicans Cells Lacking GPH1 Reduce Their Cell Surface Hydrophobicity (CSH)
C. albicans is capable of forming biofilms on abiotic or biotic surfaces. Catheters, dentures, prosthesis (abiotic), and mucosal cell surfaces (biotic) are the most common substrates [81]. Notably, hydrophobic attachment to abiotic and biotic surfaces is also critical in the initial step of biofilm formation, demonstrated by the fact that the adhesion of C. albicans to polymeric materials correlates with the cell surface hydrophobicity (CSH) phenotype [82]. Cells of the strains were grown exponentially in YPD and subjected to incubation with hydrophobic molecule n-hexadecane; the percentage of hydrophobicity (CSH) of the cells was determined as the percentage of cells not bound with the n-hexadecane. It appeared that the GPH1 homozygous null mutant decreased further in CSH compared to that of the GCN4 homozygous null mutant (Figure 4), which is known to decrease in biofilm formation [83], suggesting that C. albicans GPH1 is required to maintain CSH in C. alibicans.

C. albicans Cells Lacking GPH1 Increase Their Ability to Bind Fibronectin
Adhesion in C. albicans refers to the adherence of candida cells to host tissues. It is a phenomenon employing several adhesion proteins (called adhesins, flocculins, or agglutinins) expressed on morphologically changing cell surfaces. Adhesins are agglutinin-like sequences (ALS) that are members of a family of glycosylphosphatidylinositol (GPI)-linked cell surface glycoproteins, capable of binding several extracellular matrix proteins (ECM) of mammalian cells, such as fibronectin (FN), lamin, fibrinogen, and collagen type I and IV [84][85][86][87][88]. To assess the ability of C. albicans cells associating with FN, cells of the strains were incubated with FN for 1 h, before washing, harvesting, boiling in the presence of loading dye, and subjecting to Western blotting analysis with a specific anti-FN antibody. It appeared that cells lacking GPH1 increase the ability to bind FN compared with the wildtype strain, suggesting that C. albicans GPH1 suppresses the binding with FN ( Figure 5).

C. albicans Cells Lacking GPH1 Increase Their Ability to Bind Fibronectin
Adhesion in C. albicans refers to the adherence of candida cells to host tissues. It is a phenomenon employing several adhesion proteins (called adhesins, flocculins, or agglutinins) expressed on morphologically changing cell surfaces. Adhesins are agglutinin-like sequences (ALS) that are members of a family of glycosylphosphatidylinositol (GPI)linked cell surface glycoproteins, capable of binding several extracellular matrix proteins (ECM) of mammalian cells, such as fibronectin (FN), lamin, fibrinogen, and collagen type I and IV [84][85][86][87][88]. To assess the ability of C. albicans cells associating with FN, cells of the strains were incubated with FN for 1 h, before washing, harvesting, boiling in the presence of loading dye, and subjecting to Western blotting analysis with a specific anti-FN antibody. It appeared that cells lacking GPH1 increase the ability to bind FN compared with the wild-type strain, suggesting that C. albicans GPH1 suppresses the binding with FN ( Figure 5). The exponentially cultured strains were subjected to 10% fetal calf serum for 1 h to induce germ tube formation. The length of the germ tube was determined from 20 randomly picked germ tube-cells. The representative monographs were from the differential interference contrast (DIC) (also known as Nomarski) microscopy. The GCN4 null mutant gcn4Δ/gcn4Δ, known to reduce germ tube formation under the hyphal induction condition, was used as a control. Statistical analyses were performed by one-way ANOVA, with ** p < 0.01 and **** p < 0.001 as indicated. Representative data from one of the three independent experiments are shown. The exponentially cultured strains were subjected to 10% fetal calf serum for 1 h to induce germ tube formation. The length of the germ tube was determined from 20 randomly picked germ tube-cells. The representative monographs were from the differential interference contrast (DIC) (also known as Nomarski) microscopy. The GCN4 null mutant gcn4∆/gcn4∆, known to reduce germ tube formation under the hyphal induction condition, was used as a control. Statistical analyses were performed by one-way ANOVA, with ** p < 0.01 and **** p < 0.001 as indicated. Representative data from one of the three independent experiments are shown.

C. albicans Cells Lacking GPH1 Improve Adhesion Ability but Remain Unchanged in Biofilm Formation
Because CaCDC4 negatively regulates biofilm formation, we attempted to ascertain whether GPH1 plays a similar role. Cells (5 × 10 6 cells in 100 µL) of the strains were subjected to biofilm induction condition for 1 h at 37 • C in each well of the nonpyrogenic polystyrene flat-bottom 96-well microtiter plate to determine the ability in adhesion assessed by the XTT reduction assay. Cells (5 × 10 6 cells in 100 µL) of the strains were subjected to biofilm induction condition for 1 h at 37 • C in each well of the nonpyrogenic polystyrene flat-bottom 96-well microtiter plate to assess the ability in adhesion before the XTT reduction assay. As shown in Figure 6A, cells the GPH1 mull mutants, both the homozygote and heterozygote, and the GPH1 complement strain increased in adhesion to polystyrene as those of the GCN4 homozygous null mutant, compared with the wildtype, suggesting that GPH1 has a role in inhibiting adhesion to the polystyrene. Cells (2 × 10 5 cells in 200 µL) of the strains were subjected to biofilm induction condition for 48 h at 37 • C in each well of the nonpyrogenic polystyrene flat-bottom 96-well microtiter plate to assess the ability in biofilm formation before the XTT reduction assay. Unlike cells of the GCN4 homozygous null mutant exhibiting increased ability in biofilm formation, those of all GPH1-related strains showed similar ability in biofilm formation ( Figure 6B), as compared with those of wild-type, suggesting that GPH1 does not affect biofilm formation.

C. albicans Cells Lacking GPH1 Improve Adhesion Ability but Remain Unchanged in Biofilm Formation
Because CaCDC4 negatively regulates biofilm formation, we attempted to ascertain whether GPH1 plays a similar role. Cells (5 × 10 6 cells in 100 μL) of the strains were subjected to biofilm induction condition for 1 h at 37 °C in each well of the nonpyrogenic

C. albicans Cells Lacking GPH1 Improve Adhesion Ability but Remain Unchanged in Biofilm Formation
Because CaCDC4 negatively regulates biofilm formation, we attempted to ascertain whether GPH1 plays a similar role. Cells (5 × 10 6 cells in 100 μL) of the strains were subjected to biofilm induction condition for 1 h at 37 °C in each well of the nonpyrogenic cells in 200 μL) of the strains were subjected to biofilm induction condition for 48 h at 37 °C in each well of the nonpyrogenic polystyrene flat-bottom 96-well microtiter plate to assess the ability in biofilm formation before the XTT reduction assay. Unlike cells of the GCN4 homozygous null mutant exhibiting increased ability in biofilm formation, those of all GPH1-related strains showed similar ability in biofilm formation ( Figure 6B), as compared with those of wild-type, suggesting that GPH1 does not affect biofilm formation. Figure 6. GPH1 has a role in adhesion but not in biofilm formation in C. albicans. Cells of the strains were induced for adhesion in RPMI-1640 medium supplemented with L-glutamine and buffered with MOPS acid as described in the Materials and Methods and were subjected to in vitro XTT reduction assay for adhesion assay (A) or biofilm formation assay (B). Statistical analyses were performed by one-way ANOVA, with * p < 0.05, *** p < 0.005, **** p < 0.001 as indicated. The GCN4 null mutant gcn4∆/gcn4∆, known to reduce biofilm formation, was included in the assay.

Discussion
In this study, we characterized a CaCdc4-associated protein Gph1 that had been identified previously [48]. The functional interaction between the Gph1-encoded gene GPH1 and CaCDC4 was assessed. Because the F-box and WD40-repeat domains are present in CaCdc4, we presumed that CaCDC4 encodes a standard F-box protein of SCF ubiquitin ligase [39] named as SCF CaCdc4 . We verified that those domains are indispensable for filamentation [40] to control its targets via SCF CaCdc4 ubiquitin ligase-dependent degradation. We revealed that the filamentous development due to the repressed CaCDC4 expression in C. albicans was moderately suppressed by the constitutive expression ofGPH1 ( Figure 1B,C), which is a positive regulator of filamentous growth. The reason for this can be justified by the hindrance of Gph1 being entirely degraded by the SCF CaCdc4 ubiquitin ligase, resembling the degradation of Sol1 blocked in the CaCdc4-depleted C. albicans cells [28]. Indeed, we were able to observe the decreased level of Gph1 when the CaCDC4 was de-repressed with the translation inhibitor cycloheximide in C. albicans ( Figure 2B,C). Hence, Gph1 denotes a typical SCF CaCdc4 target, which is negatively regulated by CaCdc4 in a ubiquitin-proteasome-dependent manner.
However, the fact that the GPH1 expression-induced strain showed no enhancement of filamentous development (Supplementary Figure S1) and the GPH1 homozygous null mutant could still form filaments under the hypha-induced conditions but with the reduced ability in germ tube formation (Figure 3), suggests that GPH1 serves no direct role to control yeast-to-hypha transition. The likely reason is that the property of the cell wall of the GPH1 null mutant has altered, which is evidenced by the improved ability to aggregate in cells lacking GPH1 (Supplementary Figure S3B); as a consequence, the ability in hyphal formation is affected. The change in cell wall property may alter the structural organization and cell wall layers, which affects the ability in flocculation. We tested if GPH1 has a role in calcium-dependent self-recognition mediated by adhesins or flocculins [45,89], which has been demonstrated to be required by the C. albicans CaCDC4 [40]. Cells lacking GPH1 did not appear to affect the ability in flocculation (Supplementary Figure S6). Hence, the altered cell wall property mediated by GPH1 has no role in the function of flocculins, including expression and activity, specifically those of Ca 2+ -dependency. The change in cell wall property in C. albicans cells lacking GPH1 was revealed in their decreased cell surface hydrophobicity (CSH) (Figure 4) and increased ability to bind fibronectin ( Figure 5), both of which are related to cell adhesion. Of note, while increased CSH [90] and binding ability to fibronectin [91] of C. albicans is known to accompany enhanced biofilm formation, C. albicans lacking GPH1 appeared to affect biofilm formation oppositely with regards to CSH and binding to fibronection. Nevertheless, our data indeed demonstrated that the homozygous gph1 null mutant improves cell adhesion on polystyrene surfaces ( Figure 6A). Interestingly, no improvement of biofilm formation was found in the homozygous gph1 null mutant ( Figure 6B). Biofilms form in a sequential process involving adherence of yeast cells to the substrate, proliferation of the yeast cells, development of hyphal or pseudohyphal cells in the upper part of the biofilm, encircled in an accumulated extracellular polymer matrix consisting of proteins and polysaccharides that form a three-dimensional structure with water channels, and finally, dispersion of yeast cells from the biofilm to seed new sites [92,93]. The loss of GPH1 in C. albicans may improve only the initial step of adherence but not the final stage of maturation in biofilm formation.
How the loss of GPH1 in C. albicans cells influences the cell wall property and related functions may be far more complicated than predicted. Of note, the homozygous gph1 null mutant did not show accumulation of glycogen in C. albicans cells (unpublished data), which is inconsistent with the homozygous gph1 null mutant affecting the no growth defect by the cell wall damaging agents, diverse carbon sources, or the nutrient-depleted conditions (Supplementary Figure S5). We presume that the loss of GPH1 in C. albicans results in the compensation of GPH1 function related to the character of the cell wall architecture in which the function of many genes in a diverse aspect has interfered. As a consequence, the cell wall is reorganized such that CSH, the ability to bind fibronectin, and adhering to the abiotic surface are altered.

Conclusions
Our findings indicates that C. albicans CaCdc4 controls the polyubiquitin-proteasomedependent degradation of Gph1. While GPH1 is a positive regulator of filamentation, GPH1 is negatively controlled by the CaCDC4, which suppresses filamentation. C. albicans cells lacking GPH1 affect several features associated with the cell wall structure. Hence, the alterations of these features impact on the adhesion of the early stage of biofilm formation and other related virulent attributes, but not on mature biofilm formation.