Sib1, Sib2, and Sib3 proteins are required for ferrichrome-mediated cross-feeding interaction between Schizosaccharomyces pombe and Saccharomyces cerevisiae

Abstract

Although Saccharomyces cerevisiae is unable to produce siderophores, this fungal organism can assimilate iron bound to the hydroxamate-type siderophore ferrichrome (Fc) produced and secreted by other microbes. Fc can enter S. cerevisiae cells via Arn1. Unlike S. cerevisiae, Schizosaccharomyces pombe synthesizes and secretes Fc. The sib1 ⁺ and sib2 ⁺ genes encode, respectively, a Fc synthetase and an ornithine-N⁵-oxygenase, which are required for Fc production. When both genes were expressed in S. pombe, cross-feeding experiments revealed that S. cerevisiae fet3Δ arn1-4Δ cells expressing Arn1 could grow in the vicinity of S. pombe under low-iron conditions. In contrast, deletion of sib1 ⁺ and sib2 ⁺ produced a defect in the ability of S. pombe to keep S. cerevisiae cells alive when Fc is used as the sole source of iron. Further analysis identified a gene designated sib3 ⁺ that encodes an N⁵-transacetylase required for Fc production in S. pombe. The sib3Δ mutant strain exhibited a severe growth defect in iron-poor media, and it was unable to promote Fc-dependent growth of S. cerevisiae cells. Microscopic analyses of S. pombe cells expressing a functional Sib3-GFP protein revealed that Sib3 was localized throughout the cells, with a proportion of Sib3 being colocalized with Sib1 and Sib2 within the cytosol. Collectively, these results describe the first example of a one-way cross-feeding interaction, with S. pombe providing Fc that enables S. cerevisiae to grow when Fc is used as the sole source of iron.

Keywords: budding yeast; cross-feeding; ferrichrome; fission yeast; siderophore.

FIGURE 1

Expression of ARN1 in S. cerevisiae fet3Δ arn1-4Δ cells complements the Fc acquisition deficiency of cells defective in the uptake of Fc. (A) S. cerevisiae fet3Δ arn1-4Δ mutant cells expressing an empty plasmid, ARN1 or ARN1-GFP alleles were grown in SC medium to the mid-logarithmic phase. The cells were then incubated in the presence of Dip (100 μM) without Fc supplementation or with Fc supplementation (2 μM) for 5 h. Whole extracts were prepared and total Fc content was analyzed by thin-layer chromatography on silica gel sheets. Commercially purified Fc (15 μg) was run as a reference (control, far left lane). The solid arrowhead indicates the migration position of Fc, whereas the open arrowheads show the position of sample loading and the front of gel migration. (B) S. cerevisiae fet3Δ arn1-4Δ cells expressing Arn1-GFP were grown to the mid-logarithmic phase and subsequently left untreated (-) or treated with Dip (100 μM), Dip plus Fc (2 μM), or iron (Fe, 100 μM) for 5 h. Whole cell extract preparations were analyzed using immunoblot assays with anti-GFP and anti-PGK antibodies. The positions of the molecular weight standards (in kDa) are indicated on the right side. (C) Aliquots of cultures used in panel B were analyzed by fluorescence microscopy for visualizing cellular location of Arn1-GFP. Cell morphology was examined using Nomarski optics. White arrows indicate the cell periphery. The results are representative of three independent experiments. (D) Aliquots of cultures used in panel A were spotted in serial dilutions onto medium without Dip or Fc supplementation (control) or supplemented with Dip or a combination of Dip (75 μM) and Fc (2 μM). An isogenic wild-type (WT) strain was grown under the same conditions used in (A) and then spotted at different cellular densities onto the same indicated solid media as a control. All plates were incubated for 4 days at 30°C, and photographed.

FIGURE 2

S. pombe Fc promotes growth of a S. cerevisiae fet3Δ arn1-4Δ mutant in which the ARN1 gene is returned. (A) Schematic representation of a S. cerevisiae fet3Δ arn1-4Δ mutant expressing ARN1 in co-culture with a S. pombe strain that was competent to synthesize and excrete Fc-iron. (B,C) S. pombe wild-type (WT) and sib1Δ sib2Δ strains were grown to an OD₆₀₀ of 1.0 in the presence of FeCl₃ (10 μM), and half of cells (1 × 10⁷ cells/10 μl) were spotted (right side of each pair of spots) onto SD^–Cu–Fe medium. S. cerevisiae fet3Δ arn1-4Δ cells harboring an empty vector (v. alone) or expressing ARN1 or ARN1-GFP were grown in SD^–Cu–Fe to an OD₆₀₀ of 1.0. Cells were washed, diluted 10,000-fold, and spotted (3,000 cells/10 μL) (left side of each pair of spots) onto SD^–Cu–Fe medium in the vicinity of the S. pombe strains. In the representative experiment in (C) (right side), a physical barrier was installed between point-inoculated S. cerevisiae and S. pombe strains. S.c., S. cerevisiae; S.p., S. pombe. (D,E) S. pombe wild-type (WT) and sib1Δ sib2Δ strains were mixed with the indicated plasmid-transformed S. cerevisiae fet3Δ arn1-4Δ strain in a 1:1 ratio. Each pairwise co-culture was grown in liquid SD^–Cu–Fe medium in the presence of Dip (100 μM) for 18 h. The morphology of yeast cells was examined by Nomarski optics for determining the number of S. cerevisiae cells vs. the total number of fungal cells. Representative results are presented for the percentages of S. cerevisiae cells with the indicated plasmids in co-culture with S. pombe (WT or sib1Δ sib2Δ) (D). Comparative results of percentages of each population representing S. cerevisiae cells vs. S. pombe cells after they were grown together for 18 h (E). A minimum of 300 cells were examined for each pairwise co-culture. The results are representative of three independent experiments. The error bars indicate the standard deviation (± SD; error bars). The asterisks correspond to p < 0.0001 (****) (two-way ANOVA with Tuckey’s multiple comparisons test against the genotypes of S. pombe and S. cerevisiae strains), whereas ns stands for not significant.

FIGURE 3

str1⁺ gene disruption in S. pombe favors Fc-dependent growth of an S. cerevisiae fet3Δ arn1-4Δ mutant expressing ARN1. (A) S. pombe str1Δ strain containing an empty vector (v. alone) or expressing str1⁺ and str1⁺-GFP alleles were grown to an OD₆₀₀ of 1.0. Cells were spotted in serial dilutions onto YES medium containing Dip (140 μM) and Fc (2 μM) (right side). Cell viability of S. pombe wild-type (WT) and str1Δ strains was assayed on untreated medium (control, left side). (B) S. pombe wild-type, sib1Δ sib2Δ, and str1Δ strains were grown to an OD₆₀₀ of 1.0 under iron-replete conditions (10 μM FeCl₃). str1Δ cells contained an empty vector (v. alone) or expressed either the str1⁺ or str1⁺-GFP allele. At this point, the cells were spotted (1 × 10⁷ cells/10 μl) (right side of each pair of spots) onto SD^–Cu–Fe medium. S. cerevisiae fet3Δ arn1-4Δ cells expressing ARN1 were grown to an OD₆₀₀ of 1.0. At this stage, the cells were diluted 10,000-fold and spotted (3,000 cells/10 μL) in the vicinity of the S. pombe strains (left side of each pair of spots). S.c., S. cerevisiae; S.p., S. pombe. (C) The indicated S. pombe strains were mixed with the plasmid-transformed S. cerevisiae fet3Δ arn1-4Δ strain (harboring an empty plasmid or ARN1 allele) in a 1:1 ratio. Each pairwise co-culture was grown in liquid SD^–Cu–Fe medium in the presence of Dip (100 μM) for 18 h. Morphology of yeast cells was examined by Nomarski optics for determining the number of S. cerevisiae cells vs. the total number of fungal cells. Representative results are shown for the percentages of S. cerevisiae cells in co-culture with S. pombe (WT, sib1Δ sib2Δ or str1Δ containing an empty plasmid, str1⁺, or str1⁺-GFP allele). A minimum of 300 cells were examined for each pairwise co-culture. The results are representative of three independent experiments. Error bars indicate standard deviation (± SD; error bars). The asterisks correspond to p < 0.0001 (****) (two-way ANOVA with Tukey’s multiple comparisons test against the indicated S. pombe and S. cerevisiae fet3Δ arn1-4Δ ARN1 strains), whereas ns stands for not significant. (D) str1Δ cells expressing Str1-GFP were left untreated (-) or treated with FeCl₃ (Fe, 100 μM), Dip (100 μM), or Dip plus Fc (2 μM) for 90 min. Aliquots of total cell extract preparations were analyzed using immunoblot assays with anti-GFP and anti-α-tubulin antibodies. The positions of the molecular weight standards (in kDa) are indicated on the right side. (E) Fluorescence microscopy was performed on cells incubated from each group of cultures described in (D) to visualize cellular location of Str1-GFP. Cell morphology was examined using Nomarski optics. White arrowheads indicate the cell periphery. Results are representative of three independent experiments.

FIGURE 4

S. pombe requires Sib3 to promote Fc-dependent growth of S. cerevisiae fet3Δ arn1-4Δ cells expressing ARN1. (A) Wild-type, sib1Δ sib2Δ, and sib3Δ strains were assayed for their ability to grow on YES medium containing 0 μM (control) or 140 μM Dip. In the case of the sib3Δ mutant, an empty plasmid (v. alone), sib3⁺, or sib3⁺-GFP allele was returned in the strain. Once spotted on the control and iron-starved media, the strains were incubated for 4 days at 30°C, and photographed. (B) The indicated S. pombe strains described in (A) were grown in the presence of iron (10 μM) to an OD₆₀₀ of 1.0. Subsequently, these strains were inoculated (1 × 10⁷ cells/10 μl) (right side of each pair of spots) onto SD^–Cu–Fe medium. S. cerevisiae fet3Δ arn1-4Δ cells expressing ARN1 were grown to the mid-logarithmic phase, diluted, and spotted (3,000 cells/10 μL) in the vicinity of S. pombe strains (left side of each pair of spots). S.c., S. cerevisiae; S.p., S. pombe. (C) The indicated S. pombe strains were grown to an OD₆₀₀ of 0.5 in YES medium and incubated in the presence of Dip (100 μM) for 5 h. Total Fc was extracted and analyzed by TLC on silica gel sheets. Commercially purified Fc (15 μg) (control) was loaded as a reference. Solid arrowhead indicates the migration position of Fc, whereas open arrowheads show the origin of sample loading and front of gel migration. (D) Shown are expression profiles of sib1⁺, sib2⁺, and sib3⁺ genes. Wild-type (WT) and fep1Δ strains were grown in YES medium and were either left untreated or treated with Dip (250 μM) or FeCl₃ (Fe; 100 μM) for 90 min. Total RNA was prepared from culture aliquots, and steady-state mRNA levels of sib1⁺, sib2⁺, and sib3⁺ were analyzed by RT-qPCR assays. Graphic representations of quantification of three independent RT-qPCR assays. Error bars indicate the standard deviation (± SD; error bars). The asterisks correspond to p < 0.05 (*), p < 0.01 (**), and p < 0.0001 (****) (two-way ANOVA with Tukey’s multiple comparisons test against the indicated strain grown under low-iron conditions), whereas ns stands for not significant. (E) A genetic approach was used for allowing homologous integration of the GFP coding sequence at the chromosomal loci of sib2⁺ and sib3⁺, respectively, therefore creating strains containing GFP-tagged sib2⁺ and sib3⁺ alleles. Subsequently, these two strains were used for disrupting the fep1⁺ gene and generating sib2⁺-GFP fep1Δ and sib3⁺-GFP fep1Δ strains. All four S. pombe strains were left untreated (-) or treated with Dip or FeCl₃ (Fe), as described in (D). Aliquots of whole cell extract preparations were analyzed by immunoblot assays using anti-GFP and anti-α-tubulin antibodies. The positions of the molecular weight standards (in kDa) are indicated on the right side.

FIGURE 5

Localization of Sib1, Sib2 and Sib3 proteins in S. pombe. (A) Strains in which the GFP coding sequence was directly integrated at the chromosomal locus of sib1⁺, sib2⁺, or sib3⁺ were created and termed sib1⁺-GFP, sib2⁺-GFP, and sib3⁺-GFP strains, respectively. In the case of sib3⁺, an identical approach was performed using the Cherry coding sequence, thereby creating the sib3⁺-Cherry strain. Wild-type (WT), sib1Δ sib2Δ, sib1⁺-GFP, sib2⁺-GFP, sib3⁺-GFP, and sib3⁺-Cherry strains were grown to the logarithmic phase under non-selective growth conditions and spotted in serial dilutions onto YES medium without Dip (control) or with Dip (140 μM). (B) Strains expressing sib1⁺-GFP, sib2⁺-GFP, and sib3⁺-GFP alleles under the control of their own promoters were untreated (-) or treated with Dip (250 μM) or FeCl₃ (100 μM) for 90 min. Cells were analyzed by fluorescence microscopy for the presence of GFP-dependent fluorescence signals (bottom of each pair of panels). Nomarski optics (top of each pair of panels) was used for examining cell morphologies. (C) Strains co-expressing Sib3-Cherry in combination with Sib1-GFP or Sib2-GFP were incubated in the presence of Dip (250 μM) for 90 min. Fluorescence signals of Sib3-Cherry co-expressed with Sib1-GFP or Sib2-GFP were observed by fluorescence microscopy (middle panels). Merged images of co-expressed fluorescent proteins are shown in the far right panels. Cell morphology was examined by Nomarski optics in the far left panels. Results of microscopy are representative of five independent experiments. (D) sib2Δ sib3Δ cells co-expressing GFP-tagged Sib2 and TAP-tagged Sib3 or GFP alone and TAP-tagged Sib3 were grown to the mid-logarithmic phase and incubated in the presence of Dip (250 μM) for 90 min. Whole cell extracts (Total) were incubated with IgG-Sepharose beads. The immunoprecipitated fractions (IP) were analyzed by immunoblot assays using anti-mouse IgG (α-IgG), anti-GFP (α-GFP) and anti-α-tubulin antibodies. Aliquots of whole cell extracts (Total) were probed with the same antibodies to ascertain the presence of epitope-tagged proteins prior to incubation with beads. As an additional control, aliquots of total cell lysates and bound fractions were assayed using an anti-α-tubulin antibody. The positions of the molecular weight standards (in kDa) are indicated on the left and right sides of the panels.