Copper-dependent trafficking of the Ctr4-Ctr5 copper transporting complex

Abstract

Background: In Schizosaccharomyces pombe, copper uptake is carried out by a heteromeric complex formed by the Ctr4 and Ctr5 proteins. Copper-induced differential subcellular localization may play a critical role with respect to fine tuning the number of Ctr4 and Ctr5 molecules at the cell surface.

Methodology/principal findings: We have developed a bimolecular fluorescence complementation (BiFC) assay to analyze protein-protein interactions in vivo in S. pombe. The assay is based on the observation that N- and C-terminal subfragments of the Venus fluorescent protein can reconstitute a functional fluorophore only when they are brought into tight contact. Wild-type copies of the ctr4(+) and ctr5(+) genes were inserted downstream of and in-frame with the nonfluorescent C-terminal (VC) and N-terminal (VN) coding fragments of Venus, respectively. Co-expression of Ctr4-VC and Ctr5-VN fusion proteins allowed their detection at the plasma membrane of copper-limited cells. Similarly, cells co-expressing Ctr4-VN and Ctr4-VC in the presence of Ctr5-Myc(12) displayed a fluorescence signal at the plasma membrane. In contrast, Ctr5-VN and Ctr5-VC co-expressed in the presence of Ctr4-Flag(2) failed to be visualized at the plasma membrane, suggesting a requirement for a combination of two Ctr4 molecules with one Ctr5 molecule. We found that plasma membrane-located Ctr4-VC-Ctr5-VN fluorescent complexes were internalized when the cells were exposed to high levels of copper. The copper-induced internalization of Ctr4-VC-Ctr5-VN complexes was not dependent on de novo protein synthesis. When cells were transferred back from high to low copper levels, there was reappearance of the BiFC fluorescent signal at the plasma membrane.

Significance: These findings reveal a copper-dependent internalization and recycling of the heteromeric Ctr4-Ctr5 complex as a function of copper availability.

Figure 1. Design of the BiFC system used in S. pombe.

A, Schematic representation of the ctr4⁺-VC fusion gene including a linker region (hatched box) that was inserted in-frame of and between the ctr4⁺ allele and the nonfluorescent C-terminal (VC) coding fragment of Venus. The ctr4⁺-VC fusion allele is under the control of its native promoter. The numbers refer to the position of the nucleotides relative to the translational initiator codon of ctr4⁺. B, The VN DNA fragment was fused downstream of and in-frame with the ctr5⁺ allele, which itself is under the control of its endogenous promoter. The hatched box represents a DNA sequence derived from a linker region. The amino acid sequence of each linker (57 base pairs (bps) for ctr4⁺-VC and 30 bps for ctr5⁺-VN) is depicted using the single-letter amino acid code. C, Schematic representation of Ctr4-VC and Ctr5-VN fusion proteins in a membrane. Transmembrane proteins Ctr4-VC and Ctr5-VN are represented by round-shaped red and yellow cylinders, respectively. Each cylinder represents a transmembrane domain. A third putative transmembrane protein partner (represented in dark) corresponds to a putative monomer of Ctr4 or Ctr5. Ctr4-VC and Ctr5-VN close association reconstitutes a functional fluorophore.

Figure 2. Co-expression of Ctr4-VC and Ctr5-VN functionally complements the respiratory deficiency of a ctr4Δ ctr5Δ mutant and produces a BiFC signal at the plasma membrane.

A, S. pombe cells harboring a ctr4Δ ctr5Δ double deletion were transformed with empty vectors (v. alone), a vector alone (v. alone) and ctr4⁺-VC, a vector alone and ctr5⁺-VN, ctr4⁺-VC and ctr5⁺-VN, or ctr4⁺-GFP and ctr5⁺-MYC₁₂. Cultures were spotted onto YES media containing glucose or glycerol/ethanol (EtOH) and BCS (0, 5, 10, 20 and 25 µM). WT, isogenic wild-type (WT) strain FY435 (ctr4⁺ ctr5⁺). B, BiFC signal of Ctr4-VC and Ctr5-VN fusion proteins at the plasma membrane. Yeast cells disrupted for ctr4⁺ and ctr5⁺ that expressed individually, or in combination, the indicated tagged genes were grown in EMM medium containing BCS (100 µM) and analyzed by fluorescence microscopy. As a positive control, co-expression of the Ctr4-GFP and Ctr5-Myc12 fusion proteins allowed Ctr4-GFP detection in the plasma membrane. As an additional proof of the specificity of the BiFC, no Venus-associated fluorescence was detected when the unrelated ctr4⁺-VC and VN-php4⁺ fusion alleles were co-expressed together. Cell morphology was examined through Nomarski optics.

Figure 3. Cells co-expressing Ctr4-VN and Ctr4-VC in the presence of Ctr5-Myc₁₂ display a fluorescent signal at the plasma membrane, whereas Ctr5-VN and Ctr5-VC co-expressed in the presence of Ctr4-Flag₂ fail to interact at the plasma membrane.

A, The strain JSY22 (ctr4Δ ctr5Δ) was co-transformed with vectors alone (v. alone), integrative plasmids co-expressing only ctr4⁺-VN and ctr4⁺-VC, or ctr4⁺-VN and ctr4⁺-VC in the presence of ctr5⁺-MYC₁₂. Cells were grown either for 5 days on YES medium in the presence of glucose or, for 9 days, on YES medium in the presence of glycerol/ethanol (EtOH) containing or not BCS (25 µM). WT, parental wild-type strain FY435. B, ctr4Δ ctr5Δ cells co-expressing the indicated fusion alleles were grown to an A₆₀₀ of 0.5. At this optical density, BCS (100 µM) was added and the treated cultures were incubated for 3 h at 30°C, and then visualized for BiFC by fluorescence microscopy. The cells were also examined by Nomarski microscopy for cell morphology. C, JSY22 (ctr4Δ ctr5Δ) cells were co-transformed with vectors alone, or with ctr5⁺-VN and ctr5⁺-VC in the absence (v. alone) or the presence of ctr4⁺-FLAG₂. Growth was tested on both fermentable (glucose) and non-fermentable (glycerol/ethanol) media that were either supplemented, or not with BCS (25 µM). D, BiFC analysis of Ctr5-VN + Ctr5-VC and Ctr5-VN + Ctr5-VC + Ctr4-FLAG₂ in ctr4Δ ctr5Δ cells. Representative fluorescence images of BiFC are shown. Nomarski microscopy was used to determine cell morphology.

Figure 4. Copper-dependent internalization of the Ctr4-Ctr5 complex.

A, Cells harboring a ctr4Δ ctr5Δ double deletion were co-transformed with the ctr4⁺-VC and ctr5⁺-VN alleles. Co-transformed cells were grown in EMM containing 160 nM of copper (low Cu) to an A₆₀₀ of 0.5 (T₀), and then were left untreated (low Cu), or were treated with BCS (100 µM), CuSO₄ (1, 25, and 100 µM) (Cu) or FeCl₃ (100 µM) (Fe). After being incubated for 3 h, the cells were visualized by fluorescence microscopy (BiFC). The cells were also examined by Nomarski microscopy for cell morphology. B, ctr4Δ ctr5Δ cells co-expressing the Ctr4-GFP and Ctr5-MYC₁₂ fusion proteins were grown to mid-logarithmic phase in EMM copper-poor (160 nM) media (low Cu). Cells were then incubated in the absence (low Cu) or the presence of CuSO₄ (100 µM) or BCS (100 µM). After a 3 h treatment, the full-length Ctr4-GFP protein was viewed by direct fluorescence microscopy (GFP). The corresponding Nomarski images are also shown for each GFP panel.

Figure 5. Preexisting Ctr4-Ctr5 complexes undergo internalization in response to high concentrations of copper.

A, A double-disruption strain (ctr4Δ ctr5Δ) was co-transformed with the ctr4⁺-VC and ctr5⁺-VN alleles. A mid-logarithmic phase culture (A₆₀₀ of 0.5) was treated with cycloheximide (100 µg/ml) for 30 min so as to inhibit protein synthesis, followed by treatment with CuSO₄ (25 µM) for the indicated times. To ensure the complete inhibition of protein synthesis, the cycloheximide treatment was repeated after 3 h of exposure of the cells to copper. The cells were then analyzed by fluorescence microscopy (BiFC). Cell morphology was examined using Nomarski optics. B, Cells were cultured as described for panel A, except that they were treated with cycloheximide only, without exposure of the cells to CuSO₄. C, Growth of cells was assessed in liquid cultures in the absence (untreated) or the presence of CuSO₄ (25 µM), or a combination of CuSO₄ and cycloheximide (100 µg/ml). All cultures were restarted at an A₆₀₀ of ∼0.5. Data are the average of triplicate samples from three independent cultures. Error bars indicate the average ± standard deviation.

Figure 6. Ctr4-Ctr5 complexes recycle during a shift of high to low copper concentrations.

ctr4Δ ctr5Δ cells containing the tagged ctr4⁺-VC and ctr5⁺-VN coding sequences were grown in EMM containing CuSO₄ (160 nM) for 16 h. The culture was then diluted (1×10⁷ cells/ml) and incubated in the presence of CuSO₄ (25 µM) for 4 h, at which time (T₀) the cells were washed twice and incubated in the presence of cycloheximide (100 µg/ml) for 30 min. At this point, the culture was divided into three treatment groups. A first group was left untreated (−), while the two other were treated with 25 µM CuSO₄ (Cu) or 250 µM BCS, respectively. Nomarski microscopy was used to determine cell morphology.