The genotypes of S. pombe strains are listed in Table 1. Alleles were inactivated using a resistance gene cassette engineered for multiple use in fission yeast (Iwaki and Takegawa, 2004). This cassette contains the kanamycin/G418 resistance gene (kanMX) flanked by loxP sequences on either side. Each disruption cassette is flanked by short DNA segments homologous to the chromosomal sequences lying upstream and downstream of the gene to be deleted. After gene disruption, the cassette can be recycled by excision from the yeast genome using a Cre recombinase/loxP-mediated removal process (Gueldener et al., 2002). Yeast extract with supplements (YES) medium was used to grow yeast strains under standard conditions (Sabatinos and Forsburg, 2010). Edinburgh minimal medium (EMM) lacking specific nutrients was used to select transformed yeast strains carrying integrative or episomal plasmids.

For transcript and protein steady-state level assessments, yeast liquid cultures were seeded to an OD₆₀₀ of 0.5. When the cultures reached an OD₆₀₀ of 1.0, cells were either left untreated or treated with the iron chelator 2,2′-dipyridyl (Dip, 250 μM) or FeCl₃ (100 μM) for 1.5 h or 3 h, unless otherwise stated. For vacuole purification, the indicated cultures were treated with Dip for 3 h, with holo-Fc (1 μM) added during the final hour of treatment. Growth assays on solid media were performed by growing cells in YES medium to an OD₆₀₀ of 1.0, then spotting serial dilutions (6,000 cells/10 μL; 600 cells/10 μL; and, 60 cells/10 μL) onto media without Dip or Fc supplementation (control) or supplemented with Dip (140 μM) or a combination of Dip and Fc (0.1 or 1 μM).

To create the integrative plasmid pJK-1000str2⁺-GFP, a 2,791-bp EcoRI-BamHI PCR-amplified DNA fragment containing the str2⁺ allele and its promoter region (starting from 1,000 bp upstream the initiator codon to the penultimate codon) was cloned into the EcoRI and BamHI sites of pJK148 (Keeney and Boeke, 1994). This plasmid was named pJK-1000str2⁺nostop. The GFP coding sequence was then amplified by PCR using primers designed to introduce BamHI and NotI restriction sites at the 5′ and 3′ ends, respectively. The resulting BamHI-NotI PCR-amplified DNA fragment was inserted in-frame with str2⁺ at the corresponding sites in pJK-1000str2⁺nostop, generating the plasmid pJK-1000str2⁺-GFP. This plasmid was subsequently used to introduce mutations into the str2⁺ coding sequence. In the str2Y539A/Y553A mutant allele, the codons for tyrosine (Tyr) at positions 539 and 553 were replaced by nucleotide triplets encoding alanine residues. For the str2Y539A/R546A/Y553A mutant allele, the same substitutions as in str2Y539A/Y553A were made, with an additional mutation replacing the arginine (Arg) codon at position 546 with a codon for alanine. These site-specific substitutions were introduced using a PCR overlap extension method (Ho et al., 1989). The construction of the plasmids pJK-1478fep1⁺, pJK-1478TAPfep1⁺, and pSP-808abc3⁺-GFP has been described previously (Pelletier et al., 2005; Pouliot et al., 2010).

Total RNA was extracted from cell cultures using the hot phenol method, as previously described (Chen et al., 2003). To analyze the transcript levels of str2⁺ and str1⁺, real-time quantitative reverse transcription PCR (RT-qPCR) assays were performed, as previously described (Brault et al., 2022). The procedures for RT reactions, cDNA synthesis, and qPCR reactions were carried out as previously reported (Brault et al., 2022). Each target transcript was analyzed in experiments with a minimum of three biological replicates, with each sample reaction performed in triplicate. Results were considered valid if the target-specific fluorescent signal produced a C_t value ≤37 cycles, and all positive and negative control reactions yielded successful amplification and no amplification, respectively. Fold changes in str2⁺ or str1⁺ transcript levels between wild-type and fep1Δ mutant samples were calculated using the ΔΔCt method, normalized to the internal control act1⁺ (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008; Protacio et al., 2022). The following equation was used for calculations: ΔΔCt = [(Ct gene–Ct ref) in wild-type] versus [(Ct gene–Ct ref) in fep1Δ]; or, ΔΔCt = [(Ct gene–Ct ref) in sib1Δ sib2Δ] versus [(Ct gene–Ct ref) in sib1Δ sib2Δ fep1Δ], under the indicated experimental conditions related to iron availability. In the case of str2⁺, the primer pair allowed the detection of an amplicon corresponding to the coding region between positions +502 and + 601 down to the first nucleotide of the initiator codon. For str1⁺, the amplicon covered the coding region from positions +376 to +475, down to the A of the start codon. To detect act1⁺ expression, a primer pair was used to amplify the coding sequence between positions +173 and + 280 down to the first base of the ATG codon of act1⁺.

Early logarithmic fep1Δ php4Δ cells expressing untagged or TAP-tagged fep1⁺ alleles were grown in the presence of FeCl₃ (Fe, 75 μM). When cultures reached an OD₆₀₀ of 0.5, they were washed and incubated with either Dip (250 μM) or FeCl₃ (Fe, 100 μM) for 3 h. Following the treatments, in vivo chemical cross-linking of proteins was performed by incubating the cell cultures in the presence of 1% formaldehyde for 20 min. The crosslinking reaction was neutralized by adding glycine (0.45 M), and cell lysates were prepared by glass bead disruption, as previously described (Larochelle et al., 2012; Brault et al., 2016). The samples were subsequently sonicated to shear the chromatin DNA into fragments of 500 to 1,000 bp. Immunoprecipitation of TAP-tagged Fep1 bound to chromatin was conducted using immunoglobulin G (IgG)-Sepharose beads, following the procedures as previously described (Jacques et al., 2014). Bead manipulation, which includes washings, elution, reversal cross-linking, and DNA precipitation, was performed in accordance with the protocols previously described (Adam et al., 2001; Jbel et al., 2009). Quantification of the immunoprecipitated DNA was conducted by quantitative real-time PCR (qPCR) using different sets of primers targeting the promoter regions of str2⁺ and str1⁺. The occupancy of TAP-Fep1 at the str2⁺ or str1⁺ loci was determined by calculating the enrichment of the specific str2⁺ and str1⁺ promoter regions relative to a GATA-free 18S ribosomal DNA coding region, which served as an internal background control. The primer pairs used for amplifying the str2⁺ and str1⁺ promoter regions were: str2-105 (5’-CCAACTTCATTAAACATCTCGGTTAG-3′)/str2-2 (5’-CAGAGTGTATGGTAAATGGCAGTA-3′), and str1-921 (5’-GACAGTCCCGTACAAGGAAGAA-3′)/str1-812 (5’-AAGATGGAGGTGAAGGCAACTT-3′), respectively. The primers used for amplifying the 18S ribosomal DNA coding region were described previously (Brault et al., 2016). Each qPCR reaction was performed in triplicate using the Perfecta SYBR Green Fast mix (Quanta) on a CFX96 Touch Real-Time PCR instrument (BioRad). All ChIP experiments were repeated at least three times with independent chromatin preparations.

For all growth conditions, phenylmethylsulfonylfluoride (PMSF) (1 mM) was added directly to the cell cultures 15 min before harvesting to protect proteins from proteolysis. Whole cell extracts were prepared with glass beads using a FastPrep-24 instrument (MP Biomedicals, Solon, OH). Cells were lysed in HEGN₁₀₀ buffer containing 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pH 7.9, 100 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 10% glycerol, 0.1 mM Na₃VO₄, 1 mM PMSF, 1 mM dithiothreitol (DTT), and a complete protease inhibitor mixture (Sigma-Aldrich; P8340). Cell lysates were incubated in the presence of Triton X-100 (1%) for 30 min at 4°C. Equal concentrations of each sample were resuspended in loading buffer (100 mM Tris–HCl, pH 7.5, 1.4 M β-mercaptoethanol, 140 mM sodium dodecyl sulfate (SDS), 5 mM EDTA, 4 M urea, 1 M thiourea, and 0.72 mM bromophenol blue), and proteins were resolved by electrophoresis on 6% SDS-polyacrylamide gels. Proteins were then transferred to nitrocellulose membranes, and the following antibodies were used for immunodetection of Str2-GFP, Abc3-GFP, and α-tubulin: monoclonal anti-GFP antibody B-2 (Santa Cruz Biotechnology) and monoclonal anti-α-tubulin antibody B-5-1-2. After incubation, the membranes were washed and incubated with appropriate horseradish peroxidase-conjugated secondary antibodies, developed with enhanced chemiluminescence (ECL) reagents (Amersham Biosciences), and visualized using an ImageQuant LAS 4000 instrument (GE Healthcare).

Cells were subjected to microscopic analysis using 1,000× magnification using the following filters: 340–380 nm (bimane-GS), 465–495 nm (GFP-tagged proteins), and 510–560 nm (FM4-64). Both fluorescence and differential interference contrast (DIC) images (Nomarski) were captured using a Nikon Eclipse E800 epifluorescence microscope equipped with a Hamamatsu ORCA-ER digital cooled camera. The representative fields shown in the images were obtained from a minimum of three independent experiments. Furthermore, the cell fields shown represent protein localization in 200 cells tested per condition.

The indicated strains were subjected to cell wall digestion, cell-surface membrane disruption, and then cell fractionation using a differential centrifugation method as previously described (Sooksa-Nguan et al., 2009). After the second Percoll step gradient (50%, v/v), the integrity of the isolated vacuoles was assessed by monitoring their ability to actively retain the fluorescent compound bimane-GS. This fluorescent compound is derived from the nonfluorescent, membrane-permeant monochlorobimane. Upon entering the cells, monochlorobimane is glutathionylated by cellular glutathione S-transferases, resulting in the production of bimane-GS, which is actively transported into the vacuoles, where it becomes fluorescent (Sarry et al., 2007).

Vacuole preparations were used either to quantify iron content or to detect the presence of GFP-tagged Str2 and Abc3 proteins. For protein detection, purified vacuoles were disrupted using glass beads in Thorner buffer (40 mM Tris–HCl, pH 6.8, 0.1 mM EDTA, 5% SDS and 8 M urea) supplemented with 1% Triton X-100. Vacuole lysates were incubated in a thermomixer at 37°C for 15 min with periodic rotation at 600 rpm, followed by an additional round of glass bead disruption using a MP-24 FastPrep instrument. The extracted vacuolar proteins were then subjected to immunoblot assays to detect Str2-GFP and Abc3-GFP.

For measuring vacuole iron content, the vacuole preparations were washed and lysed in a Tris–HCl solution (50 mM, pH 7.5) supplemented with Triton X-100 (1%), using glass beads disruption. The protein concentrations of the vacuolar lysates were determined using the Bradford assay, and equal amounts of these extracts were treated with citric acid (100 mM, pH 2.0) and incubated at 60°C for 4 h, as previously described (Rad et al., 2007). After centrifugation, the supernatant was mixed with an equivalent volume of citric acid and transferred to a fresh tube. BPS (5 mM) and freshly prepared ascorbic acid (100 mM) were then added. After 45 min of incubation at 25°C in the dark, the absorbance of the samples containing iron was measured, and the total iron content was calculated using a separate calibration curve, as previously described (Rad et al., 2007).

Although previous studies have identified target genes of Fep1 (Pelletier et al., 2003; Rustici et al., 2007), several remain poorly characterized in terms of their roles in iron homeostasis, including the str2⁺-encoded putative siderophore transporter. To validate that str2⁺ expression is repressed by iron repletion, we monitored str2⁺ transcript levels in wild-type or sib1Δ sib2Δ strains either left untreated or treated with the iron chelator 2,2′-dipyridyl (Dip, 250 μM) or FeCl₃ (Fe, 100 μM) for 90 min. As a control, we concurrently analyzed str1⁺ transcripts, which are known to be down-regulated under iron-replete conditions (Pelletier et al., 2003; Plante and Labbé, 2019). For both strains, results showed that str2⁺ and str1⁺ transcript levels were highly expressed in the presence of Dip (Figures 1A–D). In contrast, their expression markedly decreased under basal and iron-replete conditions. In the wild-type strain, str2⁺ and str1⁺ mRNA levels were repressed 2.4- and 37.0-fold, respectively, in response to iron compared to levels under iron-starved conditions (Figures 1A,B). Similarly, in the sib1Δ sib2Δ mutant strain, str2⁺ and str1⁺ transcript levels were repressed 2.0- and 27.0-fold, respectively, by iron repletion compared to levels under low-iron conditions (Figures 1C,D). To examine the role of Fep1 in str2⁺ gene regulation, similar experiments were performed using isogenic fep1Δ and sib1Δ sib2Δ fep1Δ strains. In both strains lacking Fep1, str2⁺ and str1⁺ transcripts exhibited high and constitutive expression levels that were unresponsive to iron for repression (Figures 1A–D). However, when the wild-type fep1⁺ allele or a functional TAP-tagged fep1⁺ allele was reintroduced by integration into fep1Δ and sib1Δ sib2Δ fep1Δ strains, the ability to repress str2⁺ and str1⁺ gene expression under basal and iron-replete conditions was restored (Figures 1C,D). Collectively, these results demonstrated that str2⁺ is an iron-regulated gene, with its expression moderately repressed by Fep1 in response to iron.

To further investigate whether Fep1 physically occupies the str2⁺ and str1⁺ promoters in an iron-dependent manner, we used a fep1Δ php4Δ mutant strain expressing either an untagged or TAP-tagged fep1⁺ allele, which had been re-integrated. Since this mutant strain lacks Php4, it ensures that the transcription of the re-integrated fep1⁺ or TAP-fep1⁺ alleles occurs independently of Php4, allowing their constitutive expression regardless of cellular iron levels (Mercier et al., 2008; Jbel et al., 2009; Mercier and Labbé, 2009). Thus, this biological system enabled us to separate the iron-dependent DNA binding activity of Fep1 and TAP-Fep1 from potential changes in their gene expression. Under this setup context, we used a ChIP approach to test whether the presence of TAP-Fep1 could be detected at the str2⁺ and str1⁺ promoters in vivo (Figure 2A). We probed for promoter occupancy using primers specific to the str2⁺ and str1⁺ promoter regions, which are known to contain GATA elements that are bound by Fep1 in vitro (Pelletier et al., 2003). ChIP analysis revealed that TAP-Fep1 occupied the str2⁺ and str1⁺ promoters at maximum levels when cells were treated with FeCl₃ (100 μM). Under these conditions, TAP-Fep1 occupancy at the str2⁺ and str1⁺ promoters increased 7.9- and 10.7-fold, respectively, compared to a control region encoding the 18S ribosomal RNA (Figure 2B). In contrast, when fep1Δ php4Δ cells expressing TAP-Fep1 were treated with Dip (250 μM), only low levels of str2⁺ and str1⁺ promoter fragments were immunoprecipitated (Figure 2B). In iron-replete cells, TAP-Fep1 exhibited 8.8- and 13.4-fold higher binding to the str2⁺ and str1⁺ promoter regions, respectively, compared to iron-starved cells expressing TAP-Fep1 (Figure 2B). As negative controls, untagged Fep1 immunoprecipitated only background levels of the str2⁺ and str1⁺ promoter regions (Figure 2B). Collectively, these results showed that TAP-Fep1 is recruited to str2⁺ and str1⁺ promoters under conditions of high levels of iron.

The iron- and Fep1-dependent regulation of str2⁺ expression led us to investigate whether the steady-state protein levels of Str2 mirrored the changes in str2⁺ transcript levels as a function of iron availability. A functional str2⁺-GFP allele, expressed under the control of its own promoter, was integrated into the genomes of str2Δ, str2Δ fep1Δ, str2Δ sib1Δ sib2Δ, and str2Δ fep1Δ sib1Δ sib2Δ mutant strains. The first two strains were grown to an OD₆₀₀ of 1.0 and either left untreated or treated with Dip (250 μM) or FeCl₃ (100 μM) for 3 h. The last two strains, which are deficient in Fc biosynthesis, were grown under same conditions as the first two. However, when they reached an OD₆₀₀ of 1.0, Dip-treated cells were either incubated without Fc supplementation or supplemented with holo-Fc (1 μM) during the final hour of treatment. Whole cell extracts were then prepared and analyzed by immunoblotting. The results showed that Str2-GFP steady-state levels correlated with str2⁺ transcript levels, increasing in the presence of Dip but remaining low in untreated or iron-treated str2Δ and str2Δ sib1Δ sib2Δ cells harboring a str2⁺-GFP allele (Figures 3A,B). In contrast, in str2Δ fep1Δ and str2Δ fep1Δ sib1Δ sib2Δ cells expressing str2⁺-GFP, the steady-state levels of Str2-GFP remained higher in both untreated and iron-treated conditions compared to strains with a functional fep1⁺ allele (Figures 3A,B). Furthermore, steady-state levels of Str2-GFP were detected when str2Δ sib1Δ sib2Δ and str2Δ fep1Δ sib1Δ sib2Δ strains harboring an integrated str2⁺-GFP allele were incubated in the presence of Dip with Fc supplementation (1 μM) (Figure 3B).

Next, we aimed to determine the subcellular localization of Str2-GFP when expressed in untreated, iron-replete, and iron-starved str2Δ, str2Δ fep1Δ, str2Δ sib1Δ sib2Δ, and str2Δ fep1Δ sib1Δ sib2Δ strains. Moreover, we examined the Str2-GFP fluorescent signal in str2Δ sib1Δ sib2Δ and str2Δ fep1Δ sib1Δ sib2Δ strains that had been incubated with exogenous Fc (1 μM) under low-iron conditions. Fluorescence microscopy analysis of iron-starved str2Δ and str2Δ sib1Δ sib2Δ cells expressing Str2-GFP revealed that Str2-GFP-mediated fluorescence was localized to the vacuole membranes, regardless of the presence of exogenous Fc in the case of str2Δ sib1Δ sib2Δ cells (Figures 3C,D). The Str2-GFP signal colocalized with the vacuole-staining dye FM4-64, which served as a marker for the vacuolar membrane (Figures 3C,D). Consistent with iron-dependent repression of str2⁺ expression, Str2-GFP fluorescence levels were markedly reduced in Str2-GFP-expressing str2Δ and str2Δ sib1Δ sib2Δ cells grown under basal or high-iron conditions (Figures 3C,D). In contrast, the fluorescence signal at the vacuole membrane persisted when GFP-tagged str2⁺ was expressed in str2Δ fep1Δ and str2Δ fep1Δ sib1Δ sib2Δ strains under all tested conditions (Figures 3C,D). Taken together, these results led us to conclude that Str2 functions at the vacuole membrane in iron-starved cells, regardless of whether they are Fc prototrophic or auxotrophic. Furthermore, the vacuolar localization of Str2 remains unchanged in the presence of exogenous Fc.

When the fep1⁺ gene is disrupted, fep1 null strains exhibit substantially increased expression of several genes encoding proteins involved in cellular iron homeostasis, such as iron transporters, iron–sulfur proteins, and iron-consuming proteins (Rustici et al., 2007; Brault et al., 2015). Based on these previous findings, we used str2Δ fep1Δ and abc3Δ fep1Δ mutant strains expressing re-integrated str2⁺-GFP and abc3⁺-GFP alleles, respectively. Moreover, for the str2Δ fep1Δ strain, an integrative empty vector was transformed as a negative control. Cells were grown to an OD₆₀₀ of 1.0 and then incubated with Dip (250 μM) for 3 h. During the final hour of treatment, holo-Fc (1 μM) was added, and aliquots of the cells were subsequently visualized using fluorescence microscopy. The results showed that Str2-GFP-associated fluorescence was detected at the vacuolar membrane (Figure 4A). To further validate the vacuolar localization of Str2-GFP, the vacuolar-sequestered fluorescent compound bimane-GS was used as a marker (Sarry et al., 2007). Merged images showed that Str2-GFP and bimane-GS shared a similar subcellular localization pattern within the cells (Figure 4A). As an additional control, we examined the vacuolar localization of Abc3-GFP, a known vacuolar membrane transporter (Pouliot et al., 2010). Under identical growth conditions, Abc3-GFP fluorescence was observed at the vacuoles and co-localized with the bimane-GS marker (Figure 4A). In contrast, no green fluorescence signal was detected in str2Δ fep1Δ cells containing an empty vector (Figure 4A). To distinguish protein localization at the vacuole from other organelles, we used cox4Δ fep1Δ cells expressing Cox4-Cherry, a protein known to be a mitochondrial resident marker. Results showed that the fluorescence associated with Cox4-Cherry exhibited a distinct localization pattern compared to the subcellular localization of Str2-GFP and showed no overlap with the bimane-GS signal (Figure 4A).

Vacuoles were purified from the above-mentioned cultures, and sample aliquots were examined by fluorescence microscopy. The results indicated that the vacuoles maintained their integrity throughout the purification process, as the bimane-GS-associated blue fluorescence was retained in most of the isolated vacuoles (Figure 4B). Furthermore, some of these isolated vacuoles exhibited a green fluorescence signal when purified from str2Δ fep1Δ and abc3Δ fep1Δ cells expressing str2⁺-GFP and abc3⁺-GFP, respectively (Figure 4B). In contrast, vacuoles isolated from str2Δ fep1Δ cells containing an empty vector or cox4Δ fep1Δ cells expressing Cox4-Cherry lacked any green and red fluorescence signal, respectively.

Proteins were extracted from vacuole preparations and analyzed by immunoblot assays. Str2-GFP was detected as vacuolar protein when yeast vacuoles were isolated from str2Δ fep1Δ cells expressing str2⁺-GFP (Figure 4C). Similarly, Abc3-GFP, a known vacuolar membrane protein, was detected when vacuolar proteins were analyzed by immunoblotting from Abc3GFP-expressing abc3Δ fep1Δ cells (Figure 4C). In contrast, when proteins were extracted from vacuole preparations, immunoblot experiments consistently showed no signal corresponding to Cox4-Cherry.

Vacuoles isolated from str2Δ fep1Δ cells expressing str2⁺-GFP or containing an empty plasmid were analyzed for their iron content by a BPS-based spectrophotometric method (Rad et al., 2007; Pouliot et al., 2010). In str2Δ fep1Δ cells treated with Fc and expressing str2⁺-GFP, purified vacuoles exhibited a total iron concentration of 0.079 μg per μg of vacuolar proteins (Figure 4D). In contrast, vacuoles from str2Δ fep1Δ cells harboring an empty plasmid showed 31.7% less iron (0.054 μg iron/μg of vacuolar proteins) compared to those isolated from Str2GFP-expressing str2Δ fep1Δ cells (Figure 4D). As a control, vacuoles isolated from str2Δ fep1Δ cells expressing str2⁺-GFP or carrying an empty plasmid, and incubated with Dip without Fc supplementation, displayed very low total iron concentrations (0.012 and 0.010 μg iron/μg of vacuolar proteins, respectively) (Figure 4D). Interestingly, these results showed that adding Fc in the absence of Str2-GFP led to an increase in vacuolar iron concentration, although not to the same extent as when Str2-GFP was expressed. This observation suggests the possible existence of an alternative transport system that facilitates iron transport to the vacuole, independent of Str2. Nonetheless, these findings strongly suggested that Str2 plays a role in mobilizing iron within the vacuole when cells are grown with holo-Fc as the sole source of iron.

The Fc biosynthetic pathway is essential for S. pombe survival under iron-limiting conditions (Mercier and Labbé, 2010; Brault et al., 2022). The sib1⁺ and sib2⁺ genes are necessary for Fc production in S. pombe (Schrettl et al., 2004b; Mercier and Labbé, 2010; Brault et al., 2022). Yeast strains with deletions of these two genes (sib1Δ sib2Δ) are unable to grow on iron-poor media supplemented with Dip (Figure 5A; Mercier and Labbé, 2010; Brault et al., 2022). Notably, this growth defect due to iron deficiency was rescued by adding exogenous Fc (0.1 and 1 μM) to the medium (Figure 5A). To determine whether Str2 is required for the utilization of exogenous Fc, the str2⁺ gene was disrupted in the sib1Δ sib2Δ strain and tested on Dip-supplemented media containing 0.1 and 1 μM Fc. As shown in Figure 5A, sib1Δ sib2Δ str2Δ mutant cells exhibited a severe growth defect on this Fc-supplemented medium compared to wild-type and sib1Δ sib2Δ cells expressing str2⁺ (Figure 5A). Given the established role of Str1 in the uptake of exogenous Fc (Plante and Labbé, 2019; Brault et al., 2022), we validated that its inactivation (str1Δ) in the sib1Δ sib2Δ strain led to an inability to grow on Dip-supplemented media containing 0.1 and 1 μM Fc (Figure 5A). Moreover, the results consistently showed that the sib1Δ sib2Δ str1Δ str2Δ quadruple mutant strain was unable to grow in the presence of exogenous Fc under low-iron conditions (Figure 5A).

Amino acid sequence analysis of Str2 suggests that the protein belongs to the MFS-type transporter family (Rutherford et al., 2024). Topological models of Str2 predict the presence of 14 transmembrane spans connected by hydrophilic loops (Figure 5B). The final loop is predicted to contain a putative siderophore transporter domain (SITD) with highly conserved amino acid residues (Nevitt and Thiele, 2011). Among these conserved residues, Tyr⁵³⁹, Arg⁵⁴⁶, and Tyr⁵⁵³ in Str2 are found within the SITD of other predicted or known hydroxamate-type siderophore transporters, including Str1 (S. pombe), MirB (Aspergillus fumigatus), MirC (Aspergillus nidulans), Sit1 (Candida glabrata), Arn1, and Arn3 (S. cerevisiae) (Figure 5C; Yun et al., 2000; Kim et al., 2002; Kim et al., 2005; Philpott, 2006; Haas et al., 2008; Nevitt and Thiele, 2011; Raymond-Bouchard et al., 2012; Plante and Labbé, 2019). Based on previous studies that had demonstrated the functional importance of conserved Tyr residues in the SITD domain of S. pombe Str1 (Tyr⁵⁵³ and Tyr⁵⁶⁷) and C. glabrata Sit1 (Tyr⁵⁷⁵) (Nevitt and Thiele, 2011; Plante and Labbé, 2019), we generated two mutant derivatives of Str2. In the first mutant, Tyr⁵³⁹ and Tyr⁵⁵³ were substituted with Ala residues, whereas in the second mutant, Tyr539, Arg546, and Tyr553 were replaced by Ala residues. To assess the role of Str2 in Fc-dependent growth under iron-deficient conditions, spot assays were performed using sib1Δ sib2Δ str2Δ cells expressing either str2⁺-GFP, str2-Y539A/Y553A-GFP, or str2-Y539A/R546A/Y553A-GFP allele. As shown in Figure 5A, sib1Δ sib2Δ str2Δ cells expressing str2⁺-GFP exhibited growth on Dip-supplemented media containing 0.1 and 1 μM Fc. In contrast, sib1Δ sib2Δ str2Δ cells expressing str2-Y539A/Y553A-GFP or str2-Y539A/R546A/Y553A-GFP allele displayed a severe growth defect when spotted on iron-depleted medium supplemented with Fc (0.1 and 1 μM) compared to those expressing str2⁺-GFP (Figure 5A).

To confirm that all GFP-tagged str2 alleles were expressed in sib1Δ sib2Δ str2Δ cells, steady-state protein levels of Str2-GFP and its mutant derivatives were analyzed by immunoblot assays. The results showed that all these proteins were expressed in the sib1Δ sib2Δ str2Δ strain (Figure 5D). As a negative control, whole cell extracts from sib1Δ sib2Δ str2Δ cells transformed with an empty plasmid were also analyzed by immunoblotting.

To ensure that the mutated forms, Str2-Y539A/Y553A-GFP and Str2-Y539A/R546A/Y553A-GFP, exhibited the same subcellular localization as the wild-type Str2-GFP protein, microscopic analyses were performed on the two GFP-tagged mutants alongside Str2-GFP. The results showed that Str2-GFP and its mutant derivatives displayed similar fluorescence patterns, localizing to the vacuoles and colocalizing with the vacuole-staining dye FM4-64 (Figure 5E). Taken together, the results showed that Str2 is required for sustaining cell growth in the presence of exogenous holo-Fc under iron-starvation conditions. Furthermore, the conserved amino acid residues Tyr⁵³⁹, Arg⁵⁴⁶, and Tyr⁵⁵³, located within the predicted final loop of Str2, are critical for its Fc-related function.

We next assessed the impact of the Str2-Y539A/Y553A-GFP and Str2-Y539A/R546A/Y553A-GFP mutants on the ability of str2Δ fep1Δ cells to mobilize iron within vacuoles. Logarithmic phase str2Δ fep1Δ cells expressing the str2⁺-GFP, str2-Y539A/Y553A-GFP, or str2-Y539A/R546A/Y553A-GFP alleles were treated with Dip (250 μM) for 3 h. During the final hour of treatment, holo-Fc (1 μM) was added, followed by vacuole isolation from each culture. Purified vacuoles were analyzed using a BPS-based spectrophotometric method to quantify iron concentrations. In str2Δ fep1Δ cells expressing the str2-Y539A/Y553A-GFP and str2-Y539A/R546A/Y553A-GFP alleles, purified vacuoles contained 0.062 and 0.054 μg iron/μg of vacuolar proteins, respectively (Figure 6). These values represent 40.4 and 48.1% less iron compared to str2Δ fep1Δ cells expressing the str2⁺-GFP allele, which had 0.104 μg iron/μg of vacuolar proteins (Figure 6). Taken together, these results indicated that substituting the Tyr⁵³⁹, Arg⁵⁴⁶, and Tyr⁵⁵³ residues with alanines in Str2 lead to a reduction in vacuolar iron levels when iron-starved cells are grown with holo-Fc as the sole source of iron.