CgHog1 is required for Cg survival in the gastrointestinal candidiasis model

Cg resides in the iron-rich gastrointestinal tract, and CgHog1 is essential for survival of the high-iron stress in vitro and for Cg virulence in the murine model of systemic candidiasis [4,18,20,26]. However, how iron abundance modulates Cg survival in systemic and gastrointestinal candidiasis models is unknown. To address this, we examined wild-type (wt) and Cghog1Δ survival in both models wherein the mice were fed either a regular-iron (RI; 190 mg/kg diet) or a high-iron (HI; 500 mg/kg diet) diet for 10 days, prior to infection (Fig 1A). In the systemic candidiasis model, we recovered similar wt CFUs (colony-forming units) from the kidneys, liver, spleen and brain of RI diet- and HI diet-fed mice at day 1 (d1) and day 4 (d4) post-infection (Fig 1B). These results suggest that the surplus-iron does not impact wt survival appreciably during the early infection stages in the systemic model. Notably, wt d4-CFUs in the brain were 7-fold higher than d1-CFUs (Fig 1B), indicative of Cg proliferation in the brain. Cghog1Δ d4- brain CFUs were 2.5-fold lower in HI diet-fed mice, compared to RI diet-fed mice (Fig 1B), indicating that surplus iron may impair Cghog1Δ growth in vivo (Fig 1B). Importantly, compared to wt 4d-CFUs, diminished 4d-CFUs of Cghog1Δ in all organs, kidneys, liver, spleen and brain (Fig 1B), corroborate the earlier finding of CgHog1 being required for Cg survival in the murine systemic candidiasis model [4]. Altogether, these results suggest that increased iron abundance does not promote Cg survival in the systemic model of candidiasis.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Iron abundance impacts Cg colonization of the gastrointestinal tract.

A. Schematic diagram of the experimental design for systemic candidiasis (SC) and gastrointestinal candidiasis (GIC). B and C. Groups of regular-iron (190 mg/kg diet) or high-iron (500 mg/kg diet) diet-fed BALB/c (B) and C57BL/6 (C) mice were infected with C. glabrata (Cg) via tail-vein injection and oral gavage, respectively. At 1^st and 4^th day post-infection, mice were sacrificed and fungal load in indicated organs was determined. wild-type (wt; YRK20) and Cghog1Δ (YRK964) survival in SC (B; n = 11) and GIC (C; n = 17–20) models. Circles and triangles represent CFUs in individual mouse organs at 1^st and 4^th day post-infection, respectively. Bars indicate the CFU geometric mean. Black asterisks denote differences in organ CFUs between regular and high iron-fed mice infected with the same C. glabrata strain and sacrificed on the same day. Green asterisks denote organ CFU differences between regular iron-fed and same day-sacrificed wt- and Cghog1Δ-infected mice. Olive asterisks denote organ CFU differences between regular iron-fed, same C. glabrata strain-infected 1^st and 4^th day-sacrificed mice. *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001; ****, p ≤ 0.0001, Mann-Whitney U-test.

https://doi.org/10.1371/journal.pgen.1011281.g001

Next, we examined Cg burden in the gastrointestinal candidiasis model. D1-CFUs for wt and Cghog1Δ were 2- and 4-fold higher and lower, respectively, in the caecum of HI diet-fed mice, compared to respective RI diet-fed mice (Fig 1C), suggesting that while high iron aids wt colonization of the mouse caecum, it is detrimental to Cghog1Δ survival. Notably, the caecal-d4- wt CFUs were lower in HI diet-fed mice, compared to RI diet-fed mice (Fig 1C), indicating that high-iron negatively impacts wt survival, and that, d1- and d4-CFUs may represent initial colonization and survival in tissues, respectively. Since Cghog1Δ d1-CFUs in RI-fed mice ceacum were 2.5-fold higher than wt d1-CFUs, it is conceivable that Cghog1Δ colonizes caecum better at day 1 (Fig 1C). Importantly, compared to wt, Cghog1Δ d4-CFUs were lower in all organs, stomach, ileum, caecum and colon (Fig 1C), thereby reinforcing CgHog1 functions in promoting Cg survival in the host.

The Cghog1Δ mutant grows slowly, is sensitive to high-iron stress in vitro (4), and exhibited 1.3- and 3.0-fold lower cell density, compared to wt, after 24 to 48 h growth in YPD and YNB medium, respectively (S1A Fig). Thus, we infected RI diet-fed mice with a 3-fold higher Cghog1Δ inoculum to investigate the effect of mutant’s reduced growth on organ colonization and its survival in the gastrointestinal candidiasis model. We found similar results with the increased inoculum, with Cghog1Δ displaying reduced survival in the stomach at d1 and d4-post infection, and in the caecum, colon and ileum at d4-post-infection, compared to wt (S1B Fig). Notably, Cghog1Δ d1-CFUs in the caecum were higher than wt d1-CFUs (S1B Fig), indicating that CgHog1 adversely affects the initial Cg colonization of the caecum. Importantly, Cghog1Δ, at both 1X and 3X inoculum, failed to survive as well as the wt strain at 4^th day post-infection (S1B Fig), implicating CgHog1 in Cg survival in the mouse gastrointestinal candidiasis model. Altogether, while underscoring CgHog1 essentiality for Cg survival in both systemic and gastrointestinal models, these results suggest that the increased iron abundance may confer a transient advantage to Cg for colonization in the caecum, but it has no long-term positive impact on Cg survival in the gastrointestinal candidiasis model. Notably, Hog1 and Sfu1 (a negative regulator of iron-uptake genes in the gut) in C. albicans are required for mouse gut colonization and persistence, respectively [27,28], with surplus iron and HOG1 deletion in C. albicans resulting in cell flocculation and perturbed iron homeostasis, respectively [29]. Further, intestinal expansion and translocation of C. albicans and C. parapsilosis into the bloodstream has been reported in transplant patients [30], thereby linking intestinal survival with Candida bloodstream infections. Collectively, our findings highlight CgHog1 requirement for Cg survival in the gastrointestinal candidiasis model.

Identification of CgHog1 interactome

To examine how CgHog1 regulates iron homoeostasis and Cg survival in the host, we next identified CgHog1 interactors, using the N-terminally SFB-tagged CgHog1, whose functionality was verified by rescue of osmotic stress and surplus-iron sensitivity of Cghog1Δ (S2A Fig), and by activation upon growth in the iron-rich medium (S2B Fig). We found 16, 21 and 21 proteins to interact with CgHog1 under growth in low [CAA containing extracellular iron chelator BPS (bathophenanthroline disulphonate)], regular (CAA), and high (CAA containing sodium ascorbate and ferrous ammonium sulfate)-iron medium, respectively (Fig 2A and S1 Table). These three datasets shared 8 common proteins (Fig 2A and 2B), which belonged to ‘Heme transport’, ‘ATP metabolic process’, ‘Cellular response to hypoxia’, ‘Ion transmembrane transport’ and ‘Regulation of mRNA stability’ gene ontology (GO) terms for Biological Process (BP), as revealed by FungiFun analysis (S2 Table). ‘Translation’ and ‘mRNA export from nucleus’ GO terms were uniquely enriched in the CgHog1 interactome under high- and regular-iron conditions, respectively (S2 Table). The core histone H2A was present in the low-iron interactome, while the high-affinity iron uptake system component, CgFet3 ferroxidase, was identified in both low- and regular-iron interactomes of CgHog1 (Fig 2B and S1 Table). Notably, three common CgHog1 interactors between regular- and high-iron grown cells were CgHsp60 (mitochondrial chaperonin), CgKar2 (ATPase involved in protein import into the Endoplasmic Reticulum) and CgPor1 (mitochondrial voltage-dependent anion channel) that are involved in ATP-dependent protein folding, unfolded protein response and protein folding in the ER, and mitochondrial osmotic stability and membrane permeability, respectively (S1 Table). Altogether, these results suggest that CgHog1 interacts with more functionally-diverse proteins than previously anticipated.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. CgHog1 interactome analysis.

A. Venn diagram illustrating the overlap among CgHog1 interactors that were identified under low (LI), regular (RI), and high-iron (HI) conditions. B. Cytoscape network illustrating unique and common interactors of CgHog1 under LI, RI and HI conditions. Blue, black and green-colored circles represent unique interactors of CgHog1 under LI, RI and HI conditions, respectively. Cyan, violet and pink color-filled circles represent common interactors of CgHog1 under LI and HI, RI and HI, and RI and LI, respectively. Dark green color-filled circles represent the common interactors of CgHog1 under all three-iron conditions.

https://doi.org/10.1371/journal.pgen.1011281.g002

CgSub2, a CgHog1 interactor, is essential for cell viability

From identified CgHog1 interactors, we selected CgSub2 for further analysis for five reasons. First, it is neither characterized in Cg nor implicated in Hog1 signalling in any other fungal pathogen. Second, CgSub2 interacted with CgHog1 under all three (low, regular and high)-iron conditions (S1 Table). Third, Sub2 in Saccharomyces cerevisiae is a transcription export complex subunit, and a DEAD-box RNA helicase, that is involved in nuclear mRNA export and spliceosome assembly [31–35]. Fourth, SUB2 overexpression upregulates iron homeostasis genes and suppresses heterochromatic gene silencing at telomeres in S. cerevisiae [36]. Finally, both CgHOG1 deletion and surplus environmental iron in Cg activates expression of the subtelomerically-encoded EPA1 gene [4].

To verify and characterize CgHog1-CgSub2 interaction, we first generated antibodies against Escherichia coli-purified CgSub2 and CgHog1 proteins in mice, and checked their specificities (S3A and S3B Fig). Next, we verified the interaction between endogenous CgSub2 and CgHog1 proteins, and found increased CgSub2-CgHog1 interaction under high-iron condition (Fig 3A). Notably, high-iron had no appreciable effect on CgSub2 and CgHog1 protein levels, but it activated CgHog1, as evidenced in elevated CgHog1 phosphorylation (Fig 3A). CgSub2 also interacted with the phosphorylated form of CgHog1 (S3C Fig), further corroborating CgHog1-CgSub2 interaction.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Surplus-iron increases CgHog1-mediated phosphorylation of CgSub2.

A. A representative Western blot showing elevated CgHog1-CgSub2 interaction in wt (YRK20) grown in ferric chloride (1 mM)-containing YNB medium for 2 h. Cell lysates were immunoprecipitated (IP) using anti-CgHog1 antibody, and resolved on 10% SDS-PAGE. Cghog1Δ (YRK964) was used as negative control. For input samples, 100, 100, 120 and 60 μg protein were loaded for CgSub2, CgHog1, phohphorylated-CgHog1 (Hog1-P) and CgGapdh detection, respectively. CgGapdh was used as loading control. The signal intensity in each lane was quantified using the ImageJ software, and CgSub2 intensity values in input samples were normalized against the corresponding CgGapdh signal values, followed by normalization of the immunoprecipitated CgSub2 against the pulled-down, CgGapdh-normalized CgHog1 signal. Fold-increase (mean ± SEM, n = 3) in CgHog1-CgSub2 interaction in high-iron-grown wt cells, compared to regular-iron (YNB)-grown wt cells (considered as 1.0), is shown underneath the blot. *, p ≤ 0.05, paired two-tailed Student’s t-test. The green asterisk marks non-specific band. IP, Immunoprecipitated samples. B. Intracellular iron levels in indicated strains. wt (YRK20), Cghog1Δ (YRK964), and wt/CgSUB2 (YRK2803) strains were grown to log-phase in YNB medium, while Cgsub2Δ* (YRK3294) and Cghog1Δsub2Δ* (YRK3372) strains were grown in methionine (2 mM) and cysteine (2 mM)-containing YNB medium. Data represent mean ± SEM (n = 4–8). Black and red asterisks denote fold-differences in iron content between indicated strains and wt (taken as 1.0), and between Cghog1Δ and Cghog1Δsub2Δ, respectively. ****, p ≤ 0.0001; two-tailed Student’s t test. C. A representative Western blot showing CgHog1 phosphorylation (CgHog1-P) in CgHog1 variants-expressing Cghog1Δ that was grown in YNB medium lacking or containing 1 mM ferric chloride for 2 h. CgGapdh was used as loading control. The signal intensity in each lane was quantified using the ImageJ software, and CgHog1 and CgHog1-phosphorylation signal intensity values in input samples were normalized against the corresponding CgGapdh signal values, followed by normalization of the phosphorylated-CgHog1 against the total CgHog1 protein signal. Fold-difference in CgHog1 and CgHog1-phosphorylation levels (mean ± SEM, n = 3) in indicated condition/strains is shown, compared to regular-iron (YNB)-grown Cghog1Δ/CgHOG1 cells (considered as 1.0). *, p ≤ 0.05; ***, p ≤ 0.001, paired two-tailed Student’s t-test. Cghog1Δ+CgHOG1, Cghog1Δ+CgHOG1^K52A, Cghog1Δ+CgHOG1^T174A/Y176F and Cghog1Δ/V strains correspond to YRK1047, YRK2127, YRK2129 and YRK1166 strains, respectively. D. Serial dilution spotting assay. Overnight YPD medium-grown cultures were normalized to OD₆₀₀ of 1.0, followed by spotting of 3 μl of three 10-fold serial dilutions on YNB medium lacking or containing 3 mM FeCl₃. V, empty vector. E. A representative autoradiograph showing recombinant CgSub2 (rCgSub2) phosphorylation. Log-phase wt (YRK20) cultures were either treated with 1 mM FeCl₃ for 2 h or left untreated in YNB medium, while Cghog1Δ (YRK964) cultures were kept untreated. After cell lysis, whole cell lysates (200 μg) were incubated with 100 μg E. coli-purified rCgSub2, 0.5 mM ATP and 10 μCi γ-³²P-ATP at 30°C for 1 h. As a negative control, rCgSub2 was incubated with 0.5 mM ATP and 10 μCi γ-³²P-ATP without Cg cell lysate. Samples were resolved on 10% SDS-PAGE and analyzed by autoradiography. The last lane contained all reagents but for the cell lysate. The signal intensity in lanes was quantified using the ImageJ software, and fold-increase in CgSub2 phosphorylation (mean ± SD, n = 2) is shown in high-iron-grown wt, compared to regular-iron-grown wt cells (considered as 1.0). *, p ≤ 0.05, paired two-tailed Student’s t-test.

https://doi.org/10.1371/journal.pgen.1011281.g003

To study CgSub2 functions in Cg, we sought to generate a CgSUB2 deletion strain. However, our attempts to delete CgSUB2 gene were unsuccessful, indicating that it may be essential for Cg growth. Therefore, we generated a conditional CgSUB2-deletion strain, that expressed CgSUB2 from methionine (met)-repressible CgMET3 promoter. This Cgsub2Δ/MET3Pro-CgSUB2 strain will be referred to as Cgsub2Δ* from hereon, and reflects the conditional deletion of CgSUB2. Cgsub2Δ* grew poorly on both methionine and cysteine (cys)-containing solid (S3D Fig) and liquid medium (S3E Fig), with cellular CgSub2 levels decreasing gradually over time in the medium containing methionine and cysteine (S3F Fig).

Next, we checked the effect of CgSub2 loss on iron homeostasis, as CgHog1 is pivotal to maintain intracellular iron content, with Cghog1Δ containing elevated iron levels [4]. We found that while CgSUB2 conditional deletion had no effect on the intracellular iron levels in wt, it led to a reversal of the high intracellular iron content of Cghog1Δ (Fig 3B), thereby implicating CgSub2 in perturbed iron homeostasis of Cghog1Δ. As a control, we also checked if methionine and cysteine addition to the medium alters intracellular iron levels. We found similar iron levels in wt cells that were grown in the medium lacking or containing methionine and cysteine (S3G Fig). Similarly, methionine and cysteine presence in the medium had no effect on elevated iron levels in the Cghog1Δ mutant (S3G Fig).

Further, to check the effect of CgSUB2 overexpression, we expressed CgSUB2 in wt cells from the strong PDC1 promoter, and found substantially higher CgSub2 protein levels in CgSUB2-overexpressing strain (wt/CgSUB2), compared to the wt strain containing endogenous CgSub2 (S3H Fig). Notably, we found that CgSUB2 overexpression led to increased intracellular iron levels in wt (Fig 3B), indicating that CgSub2 may govern cellular iron content maintenance in Cg. Altogether, these data suggest that like CgHog1, its interactor CgSub2 also regulates iron homeostasis, and that, elevated iron in Cghog1Δ could be due to altered CgSub2 functions.

CgHog1 phosphorylates CgSub2

Next, to investigate if CgSub2 is a substrate of CgHog1, we performed two experiments. First, we identified catalytic and phosphorylated amino acid residues in CgHog1 through multiple sequence alignment of Hog1 of human and yeast species (S4A Fig). Of note, Hog1 is a conserved MAPK and p38 is the human homolog of CgHog1, with MAP kinase kinases activating p38 MAPKs by phosphorylating Threonine and Tyrosine in the dual phosphorylation motif, Thr-X-Tyr [1,2]. Our in silico analysis predicted Lysine-52, and Threonine-174 and Tyrosine-176, to be required for CgHog1 kinase activity and HOG pathway activation, respectively (S4A Fig). Thus, we simultaneously mutated the Thr-174 and Tyr-176 residues in CgHog1 to alanine and phenylalanine, respectively, to generate the non-phosphorylatable CgHog1. Additionally, Lys-52 was mutated to alanine to generate the catalytically-inactive CgHog1. We found that Lys-52 mutation led to higher protein levels, and diminished high iron-induced increase in CgHog1 phosphorylation (Fig 3C). Contrarily, both basal and high-iron-responsive CgHog1 phosphorylation were absent in Cghog1Δ-expressing CgHog1^T174A/Y176F (Fig 3C), indicating that Thr-174 and Tyr-176 are phosphorylatable residues, that undergo phosphorylation upon HOG pathway activation. Notably, neither CgHog1^K52A nor CgHog1^T174A/Y176F could rescue the high-iron sensitivity of Cghog1Δ (Fig 3D), suggesting that the activated HOG pathway and CgHog1 activity are pivotal to high-iron stress survival.

Second, we purified CgSub2 from E. coli for in vitro phosphorylation assay. Co-incubation of 6XHis-tagged CgSub2 and γ-³²P-ATP with whole cell lysates of wt, grown in regular- and high-iron medium, revealed the phosphorylated CgSub2 in both lysates, with CgSub2 phosphorylation signal being significantly higher in high-iron samples (Fig 3E). Notably, CgSub2 was not autophosphorylated, and CgSub2 phosphorylation was significantly reduced in the absence of CgHog1 (Fig 3E). A faint signal in Cghog1Δ (Fig 3E) may indicate that another kinase is capable of CgSub2 phosphorylation. Further, we also verified high-iron induced increase in CgSub2 phosphorylation by performing in vitro phosphorylation assay using cold ATP, and running samples on a Phos-tag gel. A clear mobility shift of CgSub2 was observed with cell lysates of high-iron grown wt, compared to regular-iron-grown wt (S4B Fig), indicating that CgSub2 is phosphorylated under excess-iron conditions. Importantly, both HOG pathway and CgHog1 kinase activity are required for CgSub2 phosphorylation, as Cghog1Δ-expressing catalytically-dead CgHog1^Lys52A or non-activatable CgHog1^T174A/Y176F could not phosphorylate CgSub2 (S4C Fig). Collectively, these results suggest that CgHog1 phosphorylates CgSub2, with surplus-iron-inducible HOG pathway activation elevating the CgHog1 kinase activity.

CgSub2 modulates Cg adherence to host epithelial cells

Cg, upon growth in the high-iron medium, activates EPA1 [codes for the major adhesin contributing to host adherence in vitro; [37]] expression and displays elevated adherence to Lec2 ovary epithelial cells [4]. Further, Cghog1Δ contained increased EPA1 levels, and exhibited higher adherence to Lec2cells [4]. Since elevated adhesin expression may confer an advantage in the iron-rich host gut by promoting host adhesion, and impeding clearance by the host [24,38], we next asked if CgSub2 regulates EPA1 expression. For this, we performed three experiments. First, we showed that CgSUB2 deletion and overexpression led to decreased and increased adherence, respectively, to A-498 kidney epithelial cells, compared to wt (Fig 4A). The epa1Δ mutant (used as control) showed 1.5-fold lower adherence than wt (Fig 4A). Importantly, the increased adherence of Cghog1Δ is predominantly mediated by Epa1, as Cghog1Δepa1Δ was hypo-adherent (S4D Fig). We also verified that the presence of methionine and cysteine in the medium had no effect on Cg adherence to A-498 cells, as wt grown in the medium lacking or containing methionine and cysteine exhibited similar adherence (S4E Fig). Further, Cghog1Δsub2Δ* and CgSUB2 overexpressing-Cghog1Δ strains exhibited 2.5-fold less and similar adherence, respectively, to A-498 cells, compared to Cghog1Δ (Fig 4A), thereby linking increased adherence of Cghog1Δ to CgSub2. These results, along with diminished intracellular iron levels of Cghog1Δ, upon CgSUB2 conditional deletion (Fig 3B), indicate that CgHog1 and CgSub2 interact genetically. Importantly, CgSUB2 conditional deletion, CgSUB2 overexpression and CgHOG1 deletion also rendered Cg cells hypoadherent, hyperadherent and hyperadherent to human stomach epithelial (AGS) cells, respectively (S4F Fig), indicating that CgSub2 is a positive regulator of adhesion to both stomach and kidney epithelial cell types.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. CgSub2 is required for Cg adherence to host epithelial cells.

A. Adherence analysis of indicated S³⁵-labelled Cg cells to A-498 kidney epithelial cells. For Cgsub2Δ* (YRK3294) mutants, the sub-culturing was done in YNB medium containing S³⁵ mix, and cold methionine (2 mM) and cysteine (2 mM). Data represent mean ± SEM (n = 3). Black and red asterisks denote statistically significant differences in the percentage adherence between wt (YRK20) and indicated strains, and between Cghog1Δ (YRK964) and Cghog1Δsub2Δ* (YRK3372), respectively. *p ≤ 0.05; **p ≤ 0.01; ***, p ≤ 0.001, unpaired two-tailed Student’s t-test. B. qRT-PCR-based measurement of EPA1 and CgSUB2 transcript levels. Total RNA (500 ng), extracted from log-phase cultures of wt (YRK20), Cgsub2Δ* (YRK3294) and wt/CgSUB2 (YRK2803) strains, using the acid phenol method, was used for cDNA synthesis, followed by real-time quantitative PCR amplification. Cgsub2Δ* cells were grown in methionine and cysteine-containing YNB medium. Transcript levels were quantified using the 2^-ΔΔC_t method. Data (mean ± SEM, n = 3) were normalized against CgACT1 mRNA control, and represent fold change in EPA1 and CgSUB2 expression in indicated strains compared to wt (taken as 1.0). *p ≤ 0.05; **p ≤ 0.01, paired two-tailed Student’s t-test. C. Anti-CgSub2 antibody-based chromatin immunoprecipitation (ChIP) analysis showing CgSub2 enrichment on EPA1 promoter. Log-phase wt (YRK20) and Cghog1Δ (YRK964) cells were grown in YNB or YNB medium containing 1 mM FeCl₃ for 2 h, whereas Cgsub2Δ* (YRK3294; used as control) cells were grown in methionine and cysteine-containing YNB medium. Amplification values from ChIP samples were normalized with their respective input samples. Data represent mean ± SD (n = 2–3). Black and red asterisks denote statistically significant differences in CgSub2 binding at EPA1 promoter in Cgsub2Δ* and indicated strains, and between wt and indicated strains, respectively. *, p ≤ 0.05; **, p < 0.01, unpaired two-tailed Student’s t test. D. Venn diagram illustrating overlap in CgSub2 interactors between wt/CgSUB2 (YRK2803) and Cghog1Δ/CgSUB2 (YRK3013). E. Fungifun-based functional analysis of 55 CgSub2-interactors, that were uniquely present in Cghog1Δ. Enriched gene ontology (GO) terms for Biological Process (BP) and Cellular Component (CC) are shown.

https://doi.org/10.1371/journal.pgen.1011281.g004

Second, to determine if adhesion changes co-relate with EPA1 transcription, we performed qRT-PCR. The 2-fold lower and 2-fold higher EPA1 transcript levels in Cgsub2Δ* (CgSUB2-conditional knockout) and wt/CgSUB2, (CgSUB2-overexpressing strain), respectively (Fig 4B), highlight the opposite effects of CgSUB2 deletion and overexpression on EPA1 expression, and attribute the Cg adherence potential predominantly to EPA1 levels. Notably, EPA1 expression has been reported to be lower, higher and higher in iron-limited, iron-surplus and CgHOG1-deleted conditions, respectively [4], with Epa1 being the major adhesin for Cg adherence in vitro [37].

Further, similar to Cghog1Δ, CgSUB2 overexpression also led to increased caecum colonization at day 1-post infection in the gastrointestinal candidiasis model (S5 Fig). Contrarily, CgSUB2 loss attenuated the Cg ability to colonize and survive in the stomach, ileum, caecum and colon, as reduced fungal CFUs were recovered from Cgsub2Δ*-infected mice, compared to wt-infected mice at both day 1 and day 4 post-infection (S5 Fig). Since wt/CgSUB2 CFUs in all organs were also lower than wt CFUs at day 4 post-infection (S5 Fig), similar to Cghog1Δ mutant (Fig 1C), it is unlikely that CgSUB2 overexpression confers any significant survival advantage to Cg, despite higher initial caecum colonization in the gastrointestinal candidiasis model. Besides implicating CgSub2 in organ colonization and Cg survival in the host gut, these results also underscore that the increased adherence in vitro does not necessarily translate into elevated colonization and increased survival in vivo. This could in part be due to Cg colonization of the gut involving at least two steps; adherence and persistence, with an ability to withstand stress and survive in the host tissue being a prerequisite for the latter step. It is possible that Cghog1Δ and wt/CgSUB2, due to diverse stress susceptibility [4] and ectopic CgSUB2 overexpression, respectively, are attenuated for survival post-adherence, thereby showing impaired gut colonization, compared to the wt strain (Figs 1C, S1B and S5). In this context, it is worth noting that increased EPA1 expression has recently been implicated in macrophage activation, elevated secretion of the pro-inflammatory cytokine IL-1β and poor Cg survival in human macrophages [39]. Therefore, it is possible that while higher EPA1 expression is beneficial for the initial Cg adherence to host tissues, it is detrimental for long-term persistence and survival, as Epa1 cell surface exposure may activate the host defense system against Cg. Consistent with this, Cg clearance in the murine systemic candidiasis model is dependent upon the recognition of three adhesins including Epa1, by the natural cytotoxic receptor NCR1 on Natural Killer cells [40].

Third, we performed ChIP-qPCR to examine if CgSub2 directly regulates EPA1 expression. A 3.5- and 28-fold increase in CgSub2 occupancy on EPA1 promoter in Cghog1Δ and high-iron-grown wt, respectively, compared to regular-iron-grown wt (Fig 4C) indicated that CgSub2 enrichment on EPA1 promoter may contribute to EPA1 activation, upon both CgHOG1 deletion and growth in the surplus-iron environment [4]. Notably, Sub2 in S. cerevisiae binds RNA and unwinds RNA:DNA duplexes [41]. Therefore, it is possible that CgSub2 recruitment on EPA1 promoter may be mediated by another protein/complex. Nevertheless, these results altogether underscore that CgHog1 and CgSub2 modulate EPA1 expression and host adhesion negatively and positively, respectively, and that, excess iron stimulates CgSub2 phosphorylation, CgSub2 enrichment on EPA1 promoter and EPA1 transcription.

Identification of CgSub2 interactome

To unveil the molecular basis underlying CgHog1- and CgSub2-controlled EPA1 expression, we profiled CgSub2 interactome in the presence and absence of CgHog1. We identified 29 and 76 CgSub2-interacting proteins in wt and Cghog1Δ, respectively, including 21 common interactors (Fig 4D and S3 Table). Notably, of 29 CgSub2-interactors identified in wt, 7 were earlier found as CgHog1 interactors (Fig 2B and S1 and S3 Tables), reinforcing CgHog1-CgSub2 interaction.

Functional analysis of identified proteins revealed CgSub2 interactors to primarily belong to ‘Translation’, ‘rRNA export from Nucleus’, and ‘Maturation of SSU-rRNA from tricistronic rRNA transcript (SSU-rRNA, 5.8S rRNA, LSU-rRNA) processes (Fig 4E and S4 Table), underscoring CgSub2 functions in RNA metabolism. CgSub2 interactome in wt was specifically enriched for GO-BP terms ‘Intracellular protein transport’ and Glycolytic process’ (S4 Table). Importantly, the unique enrichment of ‘Chaperonin-containing T-complex’, ‘COPI vesicle coat’ and ‘Cdc48p-Npl4p-Ufd1p AAA ATPase complex’ GO terms for Cellular Component (S4 Table), and the 2.6-fold larger interactome of CgSub2 in the absence of CgHog1 (Fig 4D and 4E) may reflect several new interactions of CgSub2, including those with constituents of the protein-folding machinery. Consistently, the ‘Protein refolding’ BP-term was uniquely enriched in CgSub2 interactome in Cghog1Δ (S4 Table). In this context, it is worth noting that Hog1 in S. cerevisiae promotes survival of the stress caused by accumulation of the unfolded/misfolded proteins [42,43]. It thus is possible that CgHog1 deletion may result in ER stress and activation of the unfolded protein response pathway in Cg, and that, CgSub2 may facilitate protein refolding via direct or indirect interaction with protein chaperones. Of note, three unique chaperone interactors of CgSub2 viz., Hsp60, Hsp70 and Hsp90, identified in the Cghog1Δ mutant background (S3 Table) are likely to aid this process. However, the validation of CgSub2-CgHsp protein interaction as well as misfolded protein accumulation in Cghog1Δ need to be demonstrated to support this hypothesis.

CgHog1 negatively regulates CgSub2-CgHtz1 interaction

From identified interactors of CgSub2, we selected CgHtz1, a variant of the core histone H2A, for further analysis for four reasons: (i) CgHtz1 interacted with CgSub2 uniquely in Cghog1Δ; (ii) Htz1 in S. cerevisiae regulates subtelomeric and telomeric gene expression [44]; (iii) Htz1 occupancy facilitates optimal transcription activation for a subset of genes with inactive promoters [45]; and (iv) high environmental iron relieves the subtelomeric adhesin gene silencing in Cg [4]. Of note, CgHtz1 displayed 64% and 63% identity to the canonical histone H2A protein encoded by two 396 nt-long ORFs, CAGL0K11440g (CgHTA1), and CAGL0C04411g (CgHTA2), respectively, in Cg (http://www.candidagenome.org/;S6 Fig). CgHta1 and CgHta2 differ from each other in one amino acid at the 9^th position (S6 Fig).

Since CgHtz1 functions were unknown, we first performed three experiments to examine its role in Cg pathobiology. First, we deleted CgHTZ1, and found that Cghtz1Δ grew slowly in minimum YNB medium, and displayed sensitivity to surplus-iron (Fig 5A). Cghtz1Δ also exhibited attenuated growth under DNA damage and oxidative stress conditions (S7A Fig). Second, we checked proliferation in human THP-1 macrophages via CFU-based assay, and found the Cghtz1Δ mutant to display 2-fold reduced replication, compared to wt cells (S7B Fig). Third, CgHtz1 was required for survival in mice, as 2- to 10-fold lower yeast colony-forming units were recovered from kidneys, liver, spleen and brain of Cghtz1Δ-infected mice, compared to wt-infected mice (Fig 5B). Importantly, CgHTZ1 expression complemented all defects associated with CgHTZ1 deletion (Figs 5A, 5B, S7A and S7B). Altogether, these data underscore CgHtz1 requirement for stress survival and pathogenesis of Cg.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. CgHog1 negatively regulates CgSub2-CgHtz1 interaction.

A. Serial dilution spotting assay showing sensitivity of indicated strains to ferric chloride (3 mM) and ferrous sulfate (20 mM). B. Cghtz1Δ (YRK4392) survival analysis in systemic candidiasis model. Overnight YPD medium-grown strains were injected intravenously into female BALB/c mice. At 7^th day post-infection, mice were sacrificed, indicated organs were collected and fungal burden was determined. Circles and bars denote colony-forming units recovered from an individual mouse, and the CFU geometric mean (n = 3–8), respectively, for each organ. *, p ≤ 0.05; **, p < 0.01; ***, p ≤ 0.001, Mann-Whitney U-test. C. A representative Western blot showing CgSub2-CgHtz1 interaction. Whole cell lysates (3 mg) of wt and Cghog1Δ strains expressing CgSub2-SFB, CgHtz1-GFP or Vector (V)-GFP were incubated with S-protein beads, followed by sample resolution on 15% SDS-PAGE and immunoblotting. 100 μg, 100 μg and 60 μg protein were loaded for CgSub2, CgHtz1 and CgGapdh detection, respectively, in input samples. Fold-increase in CgSub2-CgHtz1 interaction (mean ± SEM; n = 3) in Cghog1Δ, compared to wt (taken as 1.0), is shown underneath the blot. *, p ≤ 0.05, paired two-tailed Student’s t test. wt/CgSUB2&CgHTZ1, Cghog1Δ/CgSUB2&CgHTZ1, wt/CgSUB2&V and Cghog1Δ/CgSUB2&V strains correspond to YRK3872, YRK3979, YRK3870 and YRK3871 strains, respectively. IP, Immunoprecipitated samples. D. A representative Western blot showing diminished CgSub2-CgHtz1 interaction under high-iron conditions. wt cells expressing CgSUB2-SFB (YRK2803) were grown in YNB and ferric chloride (1 mM)-containing YNB medium for 2 h. Cells were lysed and whole cell lysates (3 mg) were incubated with streptavidin beads, followed by sample resolution on 15% SDS-PAGE and immunoblotting. wt cells expressing SFB-tagged GFP were used as negative control. Hog1-P indicates phosphorylated CgHog1 under iron-rich conditions (ferric chloride-containing medium). The arrow denotes CgSub2 protein band (65 kDa). Fold-decrease in CgSub2-CgHtz1 interaction (mean ± SEM; n = 3) in ferric chloride-grown wt, compared to YNB-grown wt cells (taken as 1.0), is shown underneath the blot. *, p ≤ 0.05, paired two-tailed Student’s t test. IP, Immunoprecipitated samples. E. Cg adherence analysis to A-498 cells. Data represent mean ± SEM (n = 3). Black, red and blue asterisks denote percentage adherence differences between wt and indicated strains, Cghog1Δ and Cghog1Δhtz1Δ, and Cghtz1Δ and Cghtz1Δ-complemented strain, respectively. **p ≤ 0.01; ***, p ≤ 0.001, unpaired two-tailed Student’s t-test. wt, Cghog1Δ, Cghtz1Δ, Cghog1Δhtz1Δ, Cghtz1Δ+CgHTZ1 and wt/CgHTZ1 strains, that correspond to YRK20, YRK964, YRK4392, YRK4390, YRK4737 and YRK4729, respectively, were grown in YNB medium. F. ChIP analysis showing CgSub2 occupancy on EPA1 promoter in Cghtz1Δ cells. Log-phase wt (YRK20) and Cghtz1Δ (YRK4392) cells were grown in YNB medium. Amplification values from ChIP samples were normalized with their respective input samples. Data represent mean ± SD (n = 2). Black and red asterisks denote statistically significant differences in CgSub2 binding at EPA1 promoter in indicated strains, compared to IgG control, and between wt and indicated strains, respectively. *, p ≤ 0.05; **, p < 0.01, unpaired two-tailed Student’s t test. G. Adherence analysis of YNB-grown CgSUB2-overexpressing cells to A-498. Data represent mean ± SEM (n = 3). Black and red asterisks denote percentage adherence differences between wt and indicated strains, and CgSub2-overexpressing wt and indicated strains, respectively. **p ≤ 0.01; ***, p ≤ 0.001; ****p ≤ 0.0001, unpaired two-tailed Student’s t-test. wt, wt/CgSUB2, Cghtz1Δ, Cghtz1Δ/CgSUB2, Cghog1Δhtz1Δ and Cghog1Δhtz1Δ/CgSUB2 strains correspond to YRK20, YRK2803, YRK4392, YRK4909, YRK4390 and YRK4911 strains, respectively.

https://doi.org/10.1371/journal.pgen.1011281.g005

Next, to examine CgHtz1 functions in CgHog1-mediated iron homeostasis, we first deleted CgHTZ1 in Cghog1Δ background, and found that the double mutant Cghog1Δhtz1Δ was also growth-attenuated in the high-iron medium (Fig 5A). Cghog1Δhtz1Δ also exhibited increased susceptibility to DNA damage and oxidative stress (S7A Fig). Mutant complementation analysis revealed that CgHTZ1 and CgHOG1 expression in the Cghog1Δhtz1Δ mutant resulted in Cghog1Δ and Cghtz1Δ phenotypes, respectively (Figs 5A and S7A).

Next, we verified CgHtz1-CgSub2 interaction, and found higher interaction in Cghog1Δ, compared to wt (Fig 5C), thereby corroborating that CgHog1 negatively regulates CgSub2-CgHtz1 interaction. Notably, we found no interaction between CgHtz1 and CgHog1 (S7C Fig). Importantly, neither CgHtz1 levels were altered in Cghog1Δ (Fig 5C) nor CgHog1 phosphorylation was impaired in Cghtz1Δ (S7D Fig). However, high-iron led to diminished CgHtz1 levels (S7E Fig) and a decrease in CgSub2-CgHtz1 interaction (Figs 5D and S7F), thereby underscoring an inhibitory effect of surplus-iron on CgSub2-CgHtz1 interaction. Altogether, we draw three inferences from these data. First, CgHog1 and CgHtz1 do not interact physically. Second, CgHtz1 acts downstream of CgHog1 activation. Third, surplus-iron and CgHog1 have negative effect on CgSub2-CgHtz1 interaction, indicating that CgSub2 may activate EPA1 transcription, via distinct mechanisms, upon CgHOG1-deleted and high-iron conditions.

CgHtz1 is essential for CgSUB2 overexpression-associated host adhesion

A strong and a weak CgSub2-CgHtz1 interaction under CgHOG1-deleted and high-iron conditions, respectively, prompted us to examine CgHtz1 functions in host adhesion. Unlike CgSUB2, CgHTZ1 overexpression did not increase the wt Cg adherence (Fig 5E), despite containing 6-fold higher CgHtz1 levels, compared to endogenous CgHtz1 levels (S8A Fig). However, Cghtz1Δ displayed 2-fold lower adherence to A-498, compared to wt, while CgHTZ1 deletion in Cghog1Δ background reduced the elevated adherence of Cghog1Δ to Cghtz1Δ levels (Fig 5E), suggesting that CgHog1-regulated host adherence is dependent on CgHtz1. The reduced adhesion of Cghtz1Δ was associated with diminished EPA1 transcription (S8B Fig), and 2.5-fold reduced CgSub2 occupancy on EPA1 promoter in the Cghtz1Δ mutant (Fig 5F). Further, CgSUB2 overexpression, intriguingly, did not increase the adhesion of Cghtz1Δ and Cghog1Δhtz1Δ (Fig 5G), thereby highlighting CgHtz1 requirement for the elevated host adhesion associated with CgSub2 overexpression and CgHOG1 deletion. Collectively, these results suggest that CgHtz1 is required for CgSub2 functions in host adhesion, and that, CgHOG1-deletion induced EPA1 expression may be due to elevated occupancy of the CgSub2-CgHtz1 complex at the EPA1 promoter.

CgSub2 is essential for surplus-iron-induced host adherence

Heat shock stress in S. cerevisiae disrupts histone-DNA interactions, and leads to nucleosome reassembly [46]. Therefore, in light of the high-iron-triggered CgSub2 enrichment on EPA1 promoter, and decrease in CgSub2-CgHtz1 interaction, we hypothesized that EPA1 transcriptional activation under high-iron conditions may arise from the canonical histone H2A replacing CgHtz1 as the CgSub2 interacting partner, with CgSub2 being indispensable for the high-iron-induced EPA1 transcription. To test this, we performed three experiments. First, we measured adhesion of the surplus-iron-grown wt, Cgsub2Δ* (CgSUB2 conditional knockout) and Cghtz1Δ strains to A-498 cells. Notably, like Cghog1Δ [4], Cgsub2Δ* did not respond to iron-excess by displaying elevated adhesion (Fig 6A), while growth in the high-iron medium led to an increase in the adherence of Cghtz1Δ (Fig 6A). Next, we checked EPA1 expression, and found a substantial and a modest increase in EPA1 transcript levels in high-iron-grown wt and Cgsub2Δ* cells, respectively, compared to respective regular iron-grown cells (Fig 6B). These results suggest that CgSub2 is pivotal to high-iron induced EPA1 activation and elevated host adhesion.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. CgSub2 is required for surplus-iron-induced Cg adherence.

A. Adherence analysis. wt (YRK20), Cghtz1Δ (YRK4392) and Cgsub2Δ* (YRK3294) cells were grown in YNB (regular-iron) and ferric chloride (500 μM)-containing YNB (high-iron) medium for 16 h, and labelled with S³⁵. YNB medium used for subculturing Cgsub2Δ* also contained cold methionine (2 mM) and cysteine (2 mM). Data represent mean ± SEM (n = 3–6). Black asterisks denote percentage adherence differences between regular and high iron-grown Cg strains. Red and blue asterisks denote percentage adherence differences between regular iron-grown wt and indicated strains, and high iron-grown wt and indicated strains, respectively. *, p ≤ 0.05; **p ≤ 0.01; ***, p ≤ 0.001; ****p ≤ 0.0001, unpaired two-tailed Student’s t-test. B. qRT-PCR-based EPA1 expression analysis in indicated strains grown in YNB lacking or containing 1 mM and 2 mM ferric chloride. Data (mean ± SEM, n = 3) were normalized against CgACT1 mRNA control, and represent fold change in EPA1 transcript levels in ferric chloride-grown cells, compared to respective YNB-medium grown cells (taken as 1.0). Black and red asterisks denote fold change in EPA1 expression between YNB-grown Cg strains and indicated conditions, and similar-condition-grown wt (YRK20) and Cgsub2Δ* (YRK3294), respectively. *p ≤ 0.05; **p ≤ 0.01, two-tailed Student’s t-test. C. A representative Western blot illustrating CgSub2-CgH2A interaction. Lysates (15 mg) of wt cells, that were grown in YNB medium lacking or containing ferric chloride (1 mM) for 2 h, were incubated with anti-CgSub2 antibody, followed by sample resolution on 15% SDS-PAGE and immunoblotting. wt cell lysates incubated with 2.5 μg IgG-coated rProtein A-Sepharose beads were used as control. Fold-increase in CgSub2-CgH2A interaction (mean ± SEM; n = 4) in ferric chloride-grown wt, compared to YNB-grown wt cells (taken as 1.0), is shown underneath the blot. **, p < 0.01; ***, p ≤ 0.001, paired two-tailed Student’s t test. The green asterisk marks a non-specific band. IP, Immunoprecipitated samples.

https://doi.org/10.1371/journal.pgen.1011281.g006

Third, to check CgSub2 association with histone H2A under high-iron conditions, we performed co-immunoprecipitation assay. We found that CgSub2 interacted with H2A, with surplus-iron leading to a higher CgSub2-CgH2A interaction, despite a reduction in H2A levels (Fig 6C). Of note, decreased histone H2A levels under surplus-iron conditions may partly be due to general stress-responsive transcriptional downregulation of core histone genes in Cg, as the genotoxic stress is known to result in transcriptional downregulation of the histone H3 and H4 genes in Cg [47], and a reduction in histone mRNAs in S. cerevisiae [48]. Further, our co- immunoprecipitation results point towards the environment cue (iron abundance)-responsive interaction of CgSub2 with the canonical histone H2A or the histone H2A variant CgHtz1, which may potentially modulate chromatin dynamics. In this context, it is worth noting that Htz1-containing nucleosomes, due to their susceptibility to ejection, poise genes for rapid and full transcriptional activation, with heat shock-activated gene promoters displaying reduced Htz1 occupancy in S. cerevisiae, and oxidative stress causing Htz1 eviction from cat-3 locus in Neurospora crassa [49,50]. Although the role of CgSub2-CgHtz1 and CgSub2-CgH2A interaction in EPA1 regulation remains to be determined via CgHtz1 and CgH2A occupancy measurement on EPA1 promoter under regular- and high-iron conditions, it is possible that the surplus extracellular iron in Cg leads to CgHtz1-CgH2A exchange at the EPA1 promoter to activate EPA1 transcription.

Identification of phosphorylated amino acids in CgSub2

CgSub2 undergoes CgHog1-dependent phosphorylation (Fig 3E), exhibits decreased (Fig 5D) and increased (Fig 6C) interaction with CgHtz1 and CgH2A, respectively, under surplus-iron conditions, and is pivotal to high-iron induced increase in adherence to epithelial cells (Fig 6A). Therefore, we sought to identify all amino acid residues in CgSub2 that are phosphorylated. For this, we purified CgSub2-SFB from wt, Cghog1Δ, Cgslt2Δ and Cghog1Δslt2Δ strains, and mapped phosphorylation sites in CgSub2 by mass spectrometry analysis. The rationale for using Cgslt2Δ and Cghog1Δslt2Δ mutants was twofold. First, basal CgSub2 phosphorylation was observed in Cghog1Δ (Fig 3E). Second, CgSlt2, terminal MAPK of the cell wall integrity pathway, is required for iron homeostasis, and is activated in response to high iron [4]. Thus, a double-deletion strain Cghog1Δslt2Δ, that lacked both kinases, was used to investigate the cross talk between the iron-responsive CgHog1 and CgSlt2 MAPKs.

We identified three, four, three and two phosphorylated residues in CgSub2 in wt, Cghog1Δ, Cgslt2Δ and Cghog1Δslt2Δ, respectively (Fig 7A). Threonine-176, Serine-200 and Threonine-230 were phosphorylated in three strains, wt, Cghog1Δ and Cgslt2Δ (Fig 7A). CgSub2 was also phosphorylated at an additional Threnonine-395 residue in Cghog1Δ (Fig 7A). Contrarily, Thr-230 phosphorylation was absent in Cghog1Δslt2Δ (Fig 7A), thereby highlighting CgHog1 and CgSlt2 indispensability for CgSub2 phosphorylation at Thr-230 residue. Notably, the NetPhos 3.1 tool predicted all four identified residues to be phosphorylated (https://services.healthtech.dtu.dk/services/NetPhos-3.1/). The identified residues, Thr-176, Ser-200 and Thr-230, and Thr-395, reside in the ATP-binding helicase domain and the Helicase conserved C-terminal domain of CgSub2, respectively (S9A Fig), with Thr-230 phosphorylation lacking in Cghog1Δslt2Δ, and Thr-395 phosphorylation being uniquely present in Cghog1Δ (Fig 7A). Importantly, while all four identified residues were conserved between Cg and S. cerevisiae Sub2, Thr-230 and Thr-395 were also conserved across their Drosophila and human counterparts [31] (S9B Fig).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. CgSub2 phosphorylation at Threonine-395 is important for CgSub2 functions.

A. Phosphosite-mapping of CgSub2 protein. A list of identified phosphorylation sites in CgSub2 protein in indicated strains. a, the phosphorylated amino acid is marked with the hashtag symbol, #. b, ModScore represents confidence in the assigned phosphorylation sites. B. Serial dilution spotting analysis showing growth of CgSUB2 knockout cells, that were overexpressing CgSub2 or CgSub2 variants with indicated serine and threonine residues mutated to alanine. Cg strains were grown overnight in CAA medium, collected, suspended in PBS and spotted on YNB and YNB medium containing methionine and cysteine. The Cgsub2Δ* strain (used as control) was grown in YNB medium containing 2 mM methionine and 2 mM cysteine. Plates were imaged after 2 days of growth at 30°C. wt, Cgsub2Δ*, Cgsub2Δ+CgSUB2, Cgsub2Δ+CgSUB2^T176A, Cgsub2Δ+CgSUB2^S200A, Cgsub2Δ+CgSUB2^T230A and Cgsub2Δ+CgSUB2^T395A strains correspond to YRK20, YRK3294, YRK6158, YRK6154, YRK6150, YRK6152 and YRK6156 strains, respectively. C. Adherence analysis of CgSUB2 knockout cells, that were overexpressing CgSub2 or CgSub2 variants, to A-498 cells via CFU-based assay. For regular-iron condition, Cg strains were grown in YNB medium for 16 h, but for the Cgsub2Δ* strain (used as control) which was grown for 16 h in the YNB medium containing methionine (2 mM) and cysteine (2 mM). The high-iron medium contained 500 μM ferric chloride. Cg cells were collected and incubated with p-formaldehyde-fixed A-498 cells at 2:1 MoI. After 2 h incubation at room temperature, A-498 epithelial cells were washed with PBS and lysed in 1 ml PBS. The lysate was plated on YNB medium, and Cg colonies (Output) appeared were counted. To determine % adherence, the output CFUs were divided by input CFUs (Number of Cg cells infected to A-498 cells), and the number was multiplied by 100. Data represent mean ± SEM (n = 3–5). *p ≤ 0.05, unpaired two-tailed Student’s t-test. wt, Cgsub2Δ*, Cgsub2Δ+CgSUB2, Cgsub2Δ+CgSUB2^T176A, Cgsub2Δ+CgSUB2^S200A, Cgsub2Δ+CgSUB2^T230A and Cgsub2Δ+CgSUB2^T395A strains, correspond to YRK20, YRK3294, YRK6158, YRK6154, YRK6150, YRK6152 and YRK6156 strains, respectively.

https://doi.org/10.1371/journal.pgen.1011281.g007

To investigate the significance of CgSub2 phosphorylation in Cg biology, we mutated Thr-176, Ser-200, Thr-230 and Thr-395 to alanine, and checked if CgSub2 mutant proteins could perform the essential function of CgSub2 in Cg growth. For this, we replaced CgSub2 in the Cgsub2Δ/CgSUB2 strain (expresses CgSUB2 from PDC1 promoter) with CgSub2^T176A, CgSub2^S200A, CgSub2^T230A and CgSub2^T395A, and monitored cell growth in the YNB medium. We found that alanine replacements of three phosphorylated residues viz., Thr-176, Ser-200 and Thr-230, in CgSub2 had no effect on cell growth, as similar growth was observed for Cgsub2Δ carrying CgSUB2^T176A, CgSUB2^S200A, CgSUB2^T230A or CgSUB2 (Fig 7B). Contrarily, compared to Cgsub2Δ/CgSUB2, Cgsub2Δ/CgSUB2^T395A grew slowly on YNB medium and YNB medium containing methionine and cysteine (Fig 7B), suggesting that Thr-395 in CgSub2 may be important for CgSub2 functions in cell growth. Expectedly, Cgsub2Δ* was unable to grow in methionine and cysteine-containing YNB medium (Fig 7B), reflecting CgSub2 essentiality. These data suggest that Thr-176, Ser-200 and Thr-230 phosphorylation in CgSub2 is dispensable for CgSub2 essentiality in Cg.

Next, we checked the effects of CgSub2 mutations on host adhesion. For this, we first examined if overexpression of CgSub2 mutant proteins results in hyperadherence in wt cells. We found that wt-expressing CgSub2^T176A, CgSub2^S200A, CgSub2^T230A and CgSub2^T395A displayed elevated adherence similar to wt-expressing CgSub2 (S9C Fig), suggesting that mutating these residues individually had no effect on host adhesion of wt cells that contain endogenous CgSub2. Secondly, we asked if CgSub2 phosphorylation at Thr-176, Ser-200, Thr-230 and Thr-395 residues is important to rescue the diminished adherence associated with CgSUB2 conditional deletion. For this, we checked the adhesion of Cgsub2Δ expressing CgSub2^T176A, CgSub2^S200A, CgSub2^T230A, CgSub2^T395A or CgSub2, to A-498 cells. We found that while all CgSub2 mutant proteins could complement diminished adherence of the CgSUB2-conditional knockout, the elevated adherence, upon growth in surplus-iron condition, was observed only upon expression of CgSub2^T176A, CgSub2^S230A and CgSub2 proteins (Fig 7C). The similar adherence of Cgsub2Δ/CgSUB2^S200A and Cgsub2Δ/CgSUB2^T395A under both regular- and high-iron conditions (Fig 7C) implicate Ser-200 and Thr-395 amino acids in iron abundance-responsive CgSub2 phosphorylation and EPA1 activation. Notably, Thr-395, that lies in the C-terminal helicase domain of CgSub2, was phosphorylated uniquely in Cghog1Δ (S9A Fig), with Thr-395 mutation to alanine impairing CgSub2 functions in cell growth (Fig 7A). Besides highlighting the importance of Thr-395 residue in proper functioning of CgSub2, these data point towards two distinct, basal and iron-induced phosphorylation-dependent functions, of CgSub2 in host adherence. Further studies are warranted to identify phosphorylated residues in CgSub2 under surplus-iron conditions, and determine identified residues’ phosphorylation significance for the helicase activity, EPA1 transcriptional regulatory function and CgHtz1/CgH2A interaction of CgSub2. However, detection of the phosphorylated amino acids in CgSub2 in the Cghog1Δ mutant suggests that CgHog1 is unlikely to be the sole kinase that phosphorylates CgSub2. In this regard, it is worth noting that since our in vitro CgSub2 phosphorylation assays were carried out with whole-cell lysates of wt and Cghog1Δ strains (Fig 3E), it is possible that CgSub2 is phosphorylated by other CgHog1-interacting/activating kinases.

In summary, our findings report a nexus among a MAPK, a putative RNA helicase and a histone variant, that links adhesin gene expression with iron availability in Cg.

Ethics statement

Mice infection experiments were performed at the Experimental Animal Facility of Centre for DNA Fingerprinting and Diagnostics (CDFD), Hyderabad, India in accordance with guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals, Government of India. All procedures were designed to minimize mice suffering, and were approved by the Institutional Animal Ethics Committee [EAF/RK/21/2023 (for gastrointestinal candidiasis model) and EAF/RK/23/2023 (for systemic candidiasis model)].

Cg growth and viability analysis

The vaginal isolate BG2 was used as the Cg wild-type (wt) strain, and all mutants were derived from this wt strain. Cg strains were routinely cultured in rich YPD medium (1% Yeast extract, 2% Peptone and 2% Dextrose) at 30°C. The effect of iron on cell growth was assessed in synthetic YNB or CAA medium. YNB medium consisted of 0.67% YNB (Yeast Nitrogen Base; with ammonium sulfate and without amino acids, and contains 1.23 μM ferric chloride) and 2% dextrose [4]. To prepare the CAA medium, 0.6% casamino acids was added to the YNB medium. To achieve logarithmic-phase Cg cells, overnight-grown Cg strains were grown for 3–5 h. Strains used are listed in the S5 Table.

For stress susceptibility analysis, serial dilution spotting assay was performed. Briefly, overnight-grown Cg cells were collected, washed, suspended in PBS and OD₆₀₀ was normalized to 1.0. Next, cultures were serially diluted 10-fold in PBS and 3 μl of each dilution was spotted on medium lacking or containing different compounds. After 1–3 days of incubation at 30°C, growth was recorded using the Gel Doc Imaging System.

For time-course analysis, overnight YNB-grown cultures were re-inoculated in fresh YNB lacking or containing methionine (2 mM) and cysteine (2 mM) medium at an initial OD₆₀₀ of 0.2. Culture absorbance was measured at 600 nm at regular intervals till 48 h, and OD₆₀₀ values were plotted against time.

Gene cloning and disruption

Cg ORFs were replaced at the genomic locus with nourseothricin resistance-conferring nat1 gene (encodes nourseothricin acetyltransferase enzyme) using homologous recombination, as described previously [4]. Cg deletion strains were confirmed by PCR. Cghog1Δ mutant was used to generate Cghog1Δslt2Δ, Cghog1Δhtz1Δ, Cghog1Δsub2Δ* (Conditional deletion of CgSUB2 in Cghog1Δ mutant background) and Cghog1Δepa1Δ double-deletion strains.

To generate CgSUB2-conditional knockout, CgSUB2 ORF (CAGL0L06908g; 1.32 kb)- was tagged with the SFB (S protein-Flag-Streptavidin-binding peptide)-epitope-encoding sequence at the C-terminus via cloning in XbaI-XmaI sites in the pRK1349 plasmid. Next, CgSub2-SFB was sub-cloned in XbaI-SalI sites in the pCU-MET3 plasmid (pRK1011) under the methionine-repressible CgMET3pro promotor and transformed into wt cells. Transformants were selected for uracil prototrophy and colony purified. CgSUB2 at the genomic locus of one purified transformant was replaced with the nat1 gene using homologous recombination-based strategy. To achieve this replacement, 5’ and 3’ UTRs (about 650 bp) of the CgSUB2 gene were PCR amplified using wt genomic DNA as a template. Next, both 5’ and 3’ gene UTRs were fused to one half each of the nat1 gene, which were PCR amplified from the nat1 gene-containing plasmid pRK625. The two nat1 amplified fragments shared about 300–350 bp of complementary region. Fused PCR products were co-transformed into the uracil prototrophic wt transformant expressing CgSUB2 ectopically from CgMET3Pro. Transformants were plated on YNB medium and incubated at 30°C for 12–16 h, followed by replica plating on nourseothricin (200 μg/ml)-containing YNB medium, containing monosodium glutamate as nitrogen source, in place of ammonium sulphate. After 24 h incubation at 30°C, nourseothricin-resistant colonies were purified and checked for CgSUB2 replacement at the genomic locus with nat1 by PCR. For growth analysis, CgSUB2-SFB expression was repressed by growing CgSUB2-conditional knockout in the presence of 2 mM methionine and 2 mM cysteine. CgSUB2-SFB, cloned under CgPDC1 promoter in the pRK1349 plasmid, was used for overexpression and protein interaction analyses.

For complementation studies, CgHTZ1 (CAGL0E02315g; 0.40 kb) was cloned in SpeI-XmaI sites in the pRK1349 plasmid. For CgHOG1 complementation, CgHOG1 (CAGL0M11748g; 1.34 kb) cloned in BamHI-XmaI sites in the pRK74 plasmid, was used, as described previously [4]. CgHTZ1 were tagged with GFP-encoding sequence at the C-terminus via cloning in XbaI-SpeI sites in the pRK1965 plasmid. CgHOG1 was cloned in BamHI-XmaI sites in the pRK1131 plasmid for N-terminal SFB tagging. Specific residues in CgSUB2 or CgHOG1 were mutated using mutagenic primers. All clones were verified by PCR, restriction digestion, sequencing and complementation analysis. The plasmids and primers used are listed in S6 and S7 Tables, respectively.

Protein extraction and immunoblotting

Log-phase Cg cells were grown in YNB or YNB medium containing 2 mM FeCl₃ for 2 h, collected, washed with chilled PBS and were snap-frozen. Frozen cells were lysed with 0.5 mm glass beads in protein extraction buffer [Tris-HCl (50 mM; pH 7.5), EDTA (2 mM), sodium fluoride (10 mM), sodium orthovanadate (1 mM) and 1 X protease inhibitor cocktail]. Cell lysates were collected after centrifugation at 13,000 rpm for 10 min at 4°C, and protein concentration was estimated using BCA kit. For protein analysis, 120, 100, 100, 100, 80 and 60 μg protein were loaded to detect phosphorylated CgHog1, total CgHog1, CgSub2, CgHtz1, CgSub2-SFB and CgGapdh, respectively.

For immunoprecipitation, Cg cell lysates were prepared in lysis buffer [Na₂HPO₄ (15 mM), NaCl (150 mM), Triton X- 100 (2%), SDS (0.1%), sodium deoxylcholate (0.5%), EDTA (10 mM), NaN₃ (0.02%)] using glass beads. rProtein A-Sepharose Fast Flow beads (GE Healthcare) or rProtein G-Sepharose Fast Flow beads were coated with appropriate antibodies for 6 h in lysis buffer at 4°C, followed by addition of pre-cleared cell lysates (5 to 15 mg). After 12 h incubation at 4°C, beads were washed 5 times with lysis buffer to remove non-specific interactions, and bead-bound proteins were resolved on a SDS-PAGE gel after boiling in SDS-PAGE sample buffer at 95°C for 10 min. Gels were transferred to PVDF membrane and probed with appropriate antibodies. CgHog1 and phosphorylated CgHog1 were detected using anti-CgHog1 (raised in mice) and phospho-p38 MAPK (Thr180/Tyr182) antibody (Cell Signaling Technology #4511L) antibodies, respectively. All antibodies used are listed in the S8 Table.

Mice maintenance and infections

The BALB/c and C57BL/6 mice are routinely maintained on iron-adequate diet containing 192 mg iron/kg for systemic and gastrointestinal candidiasis, respectively. For iron abundance experiments, the mice were fed the formulated identical diet but for the iron content. The pellet diet contained two iron concentrations, regular-iron (190 mg/kg) or excess-iron (500 mg/kg). Mice were fed experimental diets for 10 days, prior to Cg infection, and were maintained on the same diet during infection studies. Mice had free access to water and food during the entire study. No difference in weight or diet consumption was observed for the experimental mice. Mice were euthanized by CO₂ inhalation before organ collection.

For the systemic candidiasis model, overnight YPD-grown Cg (4 X 10⁷ cells; 100 μl cell suspension in PBS) were injected into the tail vein of 6–8 week-old female BALB/c mice. At 1 and 4^th day post-infection (dpi), mice were sacrificed and four organs, kidneys, liver, spleen and brain, were collected. Organs were homogenized in PBS and plated on penicillin and streptomycin-containing YPD medium. Cg colonies were counted after 2 days of growth at 30°C.

For gastrointestinal candidiasis model, 6–8 week-old female C57BL/6 mice were given three antibiotics, tetracycline (1 mg/ml), streptomycin (2 mg/ml) and gentamicin (0.1 mg/ml), in drinking water for 7 days, prior to infection. Antibiotic-treated C57BL/6 mice were orally inoculated by a single gavage with overnight YPD-grown Cg (200 μl PBS cell suspension; 2.5 X 10⁸ cells) using a 24-gauge feeding needle. At 1 or 4 dpi, mice were sacrificed, dissected to separate individual gastrointestinal organs, and four organs, stomach, ileum, caecum and colon were collected. Organs were homogenized in NETN tissue lysis buffer [Tris HCl (20 mM; pH 8.0), NaCl (100 mM), EDTA (1 mM) and Nonidet P-40 (0.5%)] by three rounds of gentle mechanical shearing on Fastprep-24 (MP biochemicals). Fungal bioburden in organ lysates was calculated by plating appropriate dilutions of organ homogenates on YPD medium containing penicillin and streptomycin. Cg colonies were counted after 2 days of growth at 30°C.

Mass spectrometry (MS) analysis

For CgHog1 interactome analysis, the Cghog1Δ mutant expressing either the N-terminally SFB (S protein-Flag-Streptavidin-binding peptide)-tagged CgHog1 (Cghog1Δ/CgSFB-HOG1) or the SFB tag (Cghog1Δ/V) were grown in CAA medium (contains 1.23 μM ferric chloride; regular- iron medium) for 12 h. Next, the Cghog1Δ/CgSFB-HOG1 culture was starved for iron for 12 h via growth in 50 μM BPS (bathophenanthroline disulfonate; an extracellular iron chelator)-containing CAA medium. After washes, the iron-starved culture was further incubated in the CAA (regular-iron) medium and CAA medium supplemented with either 50 μM BPS (low-iron), or sodium ascorbate (1 mM) plus ferrous ammonium sulfate (500 μM) (high-iron) media for 2 h at 30°C. The Cghog1Δ/V strain, grown in CAA medium, was used as a control. After 2 h growth, cells were washed and lysed in NETN buffer [Tris-HCl (20 mM; pH 8.0), NaCl (100 mM), EDTA (1 mM) and Nonidet P-40 (0.5%)] using glass beads. Whole cell lysates were subjected to tandem affinity purification with streptavadin-agarose and S-protein-agarose beads. Briefly, 8 mg protein per lysate sample was first incubated with streptavidin beads for 2 h at 4°C, followed by elution in the biotin (2 mg/ml)-containing buffer. The resulting supernatant was incubated with S-protein agarose beads for 2 h at 4°C. After washes, beads were boiled, and proteins were resolved on a 10% SDS-PAGE gel till bromophenol blue dye in the sample buffer entered about 3 mm into the gel. The gel was stained with Coomassie Brilliant Blue and destained. A portion of the gel displaying protein bands was excised, placed in water and sent to the Taplin Biological Mass Spectrometry (MS) facility at Harvard Medical School, Boston, USA for protein identification via LC-MS/MS analysis. Samples were prepared and sent from two independent biological replicates.

For CgSub2 interactome analysis, overnight cultures of wt or Cghog1Δ expressing C-terminally SFB tagged CgSub2 (CgSub2-SFB) were grown in CAA medium for 5 h at 30°C in CAA medium. wt-expressing GFP (GFP-SFB) was used as control. Cells were lysed using glass beads. CgSub2 was purified from cell lysates (8 mg), using streptavadin-agarose and S-protein-agarose beads, as described above for CgHog1 interactor analysis, and samples were sent to the Taplin Biological Mass Spectrometry (MS) facility for LC-MS/MS analysis.

For MS data analysis, two criteria were applied. First, proteins, that were found in control lysate samples, were not considered, as these represented proteins bound non-specifically to the affinity matrices. Second, proteins that were represented by a minimum of two total peptides in each replicate sample, were selected.

For phosphorylation site identification in CgSub2, CgSub2-SFB was pulled down from whole cell lysates (15 mg) of CAA medium-grown wt, Cghog1Δ, Cgslt2Δ and Cghog1Δslt2Δ strains expressing CgSUB2-SFB using the two-step pull down assay involving lysate binding with streptavidin beads, followed by elution in biotin solution, and incubation of the first-step-eluate with S-protein agarose beads. Pulled-down protein samples were resolved on 12% SDS-PAGE. The 65 kDa band corresponding to CgSub2-SFB was cut from the gel, and sent to Taplin MS facility, Harvard Medical School, Boston for identification of phosphorylated residues in CgSub2 protein.

Intracellular iron estimation

The intracellular iron content was measured by flame atomic absorption spectrometry (AAS). YNB-grown log phase Cg (50 OD₆₀₀) were collected and washed twice with HPLC-grade water. Cells were lysed overnight in 30% HNO₃ at 95°C, and supernatants were collected after centrifugation at 13000 rpm for 10 min. Iron content in supernatant samples was determined, via AAS, using the standard curve generated for different iron concentrations in 5% HNO₃.

CgSub2 and CgHog1 purification

CgHOG1 was cloned in the pET28a⁺ plasmid between BamHI and SacI Sites, and transformed into the E. coli BL21 (DE3) strain. A purified transformant was induced with 1 mM IPTG for CgHOG1 expression at 37°C for 3 h. Cells were collected, suspended in lysis buffer [Tris-HCl (50 mM; pH 7.5), NaCl (500 mM), β-ME (5 mM), Triton X-100 (1%) and Glycerol (10%)] and sonicated using Diagenode bioruptor plus. After sample centrifugation at 12,000 rpm for 5 min, the pellet was suspended in 6 M urea and resolved on 12% SDS-PAGE. The recombinant CgHog1 protein band was cut from the gel, gel slices were crushed in the gel elution buffer [Tris-HCl (25 mM; pH 8.8), glycerol (5%), SDS (1%) and glycine (0.24 mM)] using motor and pestle, and were incubated overnight at 37°C. Samples were centrifuged at 13,000 rpm for 15 min, pellet was suspended in gel elution buffer and the extraction process was repeated 3 times. Next, five volumes of chilled acetone were added to eluates and samples were spun down after overnight storage stored at -20°C. The pellet was washed with acetone, dried and solubilized with 10 mM Tris-HCl (pH 8.8) and 0.1% SDS. The soluble protein was collected after centrifugation at 12,000 rpm for 5 min and stored at -80°C.

CgSUB2 was cloned in BamHI and SacI sites in pET-28a⁺ plasmid, and expressed in the E. coli BL21 (DE3) CodonPlus-RIL strain. Bacterial cells were grown for 3 h at 37°C, and CgSUB2 expression was induced with 1 mM IPTG. After 3 h growth at 37°C, cells were collected, suspended in lysis buffer [Tris-HCl (50 mM; pH 8.0), NaCl (200 mM), imidazole (10 mM) and sarkosyl (1%)], and sonicated using Diagenode bioruptor plus. The recombinant CgSub2 protein carrying hexa-His epitope at the N-terminus was purified using Ni-NTA beads and polypropylene column (Thermo Fisher Scientific #R64050). After washing protein-bound beads with wash buffer [Tris-HCl (50 mM; pH 8.0), NaCl (150 mM) and imidazole (50 mM)], protein was eluted with the elution buffer [Tris-HCl (50 mM; pH 8.0), NaCl (150 mM) and imidazole (150 mM)]. Amicon Ultra centrifugal filter units (3 kDa cut off) were used to remove imidazole, and purified protein was stored at -80°C, after verifying protein purity on 12% SDS-PAGE.

For antibody generation in mice, purified CgHog1 or CgSub2 protein (100 μg), emulsified with complete Freud’s adjuvant were subcutaneously injected into 6–8 week-old BALB/c mice. After two booster doses at 15^th and 22^nd days, mice sera were collected and used to check the specificity of raised polyclonal antibodies by Western analysis. For anti-CgHog1 antibody, the mouse serum was pre-cleared with Cghog1Δ cell lysates (100 μg) for 12 h, prior to use. Anti-CgHog1 and anti-CgSub2 antibodies detected 51 kDa CgHog1 and 51 kDa CgSub2 protein bands in wt whole cell extracts, respectively.

In vitro phosphorylation assay

CgSub2 phosphorylation was checked by in vitro kinase assay. Briefly, YNB-grown log phase Cg cells were left either untreated or treated with 1 mM FeCl₃ for 2 h at 30°C. Cells were collected and lysed using glass bead in lysis buffer [Tris-HCl (50 mM; pH 7.5), NaCl (50 mM), EGTA (5 mM), MgCl_2, (5 mM), DTT (1 mM), triton X-100 (0.1%), and sodium orthovanadate (1 mM)]. Cell lysates were collected after centrifugation at 13000 rpm for 15 min at 4°C. For in vitro phosphorylation assay, the reaction mixture consisting of 200 μg cell lysate, 100 μg purified rCgSub2, and kinase buffer [Tris-HCl (50 mM; pH 7.5), NaCl (50 mM), EGTA (5 mM), MgCl_2, (20 mM), DTT (1 mM), and sodium orthovanadate (0.1 mM)] containing 10 μCi γ-³²P ATP and 0.5 mM ATP, was prepared, and incubated for 1 h at 30°C. After adding SDS-PAGE sample buffer, samples were heated at 95°C for 5 min, and resolved on 10% SDS-PAGE gel. The gel was dried and exposed to the phosphor screen for 12–16 h, followed by screen scanning on Typhoon FLA 9500 biomolecular imager.

To detect the mobility shift of phosphorylated CgSub2 using Phos-tag gels, 100 μg purified rCgSub2 was incubated with 200 μg whole cell lysates and kinase buffer [Tris-HCl (50 mM; pH 7.5), NaCl (50 mM), EGTA (5 mM), MgCl_2, (20 mM), DTT (1 mM), and sodium orthovanadate (0.1 mM)] containing 0.5 mM ATP. After 1 h incubation at 30°C, SDS-PAGE sample buffer was added, and samples were heated at 95°C for 5 min, followed by sample resolution in a Phos-tag (50 μM; Phos-tag Acrylamide AAL-107, NARD)-containing 10% SDS-PAGE, and probing with anti-His and anti-CgSub2 antibodies.

Human cell lines and culture conditions

The human kidney epithelial [A-498 (ATCC #HTB-44)], human stomach epithelial [AGS (ATCC #CRL-1739)] and human moncytic cell line [ATCC #TIB-202] were maintained in Minimum Essential Medium (MEM), Ham’s F-12 Medium (F-12K Medium), and RPMI medium, respectively, at 37°C and 5% CO₂.

Adherence measurement

A-498 and AGS cells were seeded at a seeding density of 2 × 10⁵ and 5 × 10⁵ cells, respectively, per well in a 24-well tissue culture plate and incubated at 37°C and 5% CO₂ for 12–14 h, followed by cell fixation with 3.7% p-formaldehyde for 15 min. For adherence potential measurement, Cg strains were grown in YNB medium for 24 h at 30°C, followed by re-inoculation of 100 μl yeast culture in YNB medium containing 200 μCi S35(Met : Cys-65:25) [INVIVO PROTWIN label mix]. After 16 h growth at 30°C, radio-labelled Cg were harvested, and culture OD₆₀₀ was normalized to 0.1. 200 μl S³⁵-labelled Cg cell suspension (4 X 10⁵ Cg cells) was added to seeded epithelial cells, and incubated for 90 min at room temperature. After washing off non-adherent Cg, epithelial cells were lysed in 5% SDS, and radioactive counts (output values) in lysates were measured. The amount of radioactivity in labelled-Cg cell suspension was considered as ‘input value’. The percentage adherence was determined by dividing the output counts by input counts, and multiplying the obtained number by 100. To study the effect of high iron on adherence, Cg strains were grown in YNB medium containing ³⁵S and 500 μM FeCl₃ for 16 h, prior to incubation with fixed A-498 cells.

Chromatin immunoprecipitation

Log-phase Cg wt cells were grown in YNB medium lacking or containing 1 mM FeCl₃ for 2 h. Cghog1Δ and the conditional Cgsub2Δ* (Cgsub2Δ/MET3Pro-CgSUB2) mutants were grown in YNB and YNB medium containing 2 mM methionine and 2 mM cysteine, respectively. Both FeCl₃-treated and untreated Cg cells were cross-linked with 1% formaldehyde for 20 min at 30°C, followed by reaction quenching with 125 mM glycine for 10 min at room temperature. Cells were harvested, and suspended in FA lysis buffer [EDTA (1 mM; pH 8.0), HEPES (50 mM; pH 7.5), NaCl (140 mM), sodium deoxycholate (0.1%), 1X protease inhibitor, and triton X-100 (1%)]. After cell lysis by bead-beating, lysates were spun down, and the supernatant was sonicated in cycles of 40 pulses (30 sec ON and 30 sec OFF) at the highest amplitude. The soluble fraction was collected after centrifugation at 15,000 rpm for 10 min, and 1/10th volume was stored as ‘Input fraction’. 3 mg protein lysate was immunoprecipitated with anti-CgSub2 antibody-conjugated protein-A Sepharose beads for 3 h at 4°C. Beads were sequentially washed with FA lysis buffer, FA lysis buffer containing 0.5 M NaCl, Wash buffer [Tris-HCl (100 mM; pH 8.0), LiCl (0.25 M), NP-40 (0.5%), deoxycholic acid (0.5%), and EDTA (1 mM)], and TE buffer. Finally, beads were suspended in elution buffer [Tris-HCl (50 mM; pH 8.0), EDTA (10 mM), and SDS (1%)] and subjected to de-crosslinking, along with the Input fraction, for overnight at 65°C. After proteinase K treatment for 1 h at 60°C, DNA was extracted, suspended in TE, treated with RNase for 1 h at 37°C, and used as a template for qRT-PCR to examine amplification of the EPA1 promoter region.

Other procedures

qRT-PCR and THP-1 macrophage infection were performed as described previously [4,23]. The freely available web tool cytoscape (https://cytoscape.org/) was used to depict the overlap among identified interacting partners of CgHog1.

Statistical analysis

GraphPad Prism was used for statistical analysis with significance level p≤0.05. Data are presented as mean and standard error of the mean. The appropriate statistical tests used are indicated in figure legends.