No abstract

Ftr1p is the permease component of the Fet3p-Ftr1p high affinity iron-uptake complex, in the plasma membrane of Saccharomyces cerevisiae, that transports the Fe3+ produced by the Fet3p ferroxidase into the cell. In this study we show that Ftr1p probably has seven transmembrane domains with an orientation of N-terminal outside, and C-terminal inside the cell. Within the context of this topology of the Fet3p-Ftr1p complex, we have identified several sequence elements in Ftr1p that are required for wild-type uptake function. First to be identified were two REXLE (Arg-Glu-Xaa-Leu-Glu) motifs in transmembrane domains 1 and 4. Alanine substitutions at any one of these combined six arginine or glutamic acid residues inactivated Ftr1p in iron uptake, indicating that both motifs were essential to iron permeation. R-->K and E-->D substitutions in these two motifs led to a variable loss of activity, suggesting that while all six residues were essential, their contributions to uptake were quantitatively and/or mechanistically distinct. The terminal glutamate in an EDLWE89 element, associated with transmembrane domain 3, and a DASE motif, located in extracellular loop 6, were also required. The double substitution to AASA in the latter, inactivated Ftr1p in iron uptake while the Ftr1p(E89A) mutant had only 20% of wild-type activity. The two REXLE and the EDLWE and DASE motifs are strongly conserved among fungal Ftr1p homologues, suggesting that these motifs are essential to iron permeation. Finally another important residue, Ile369, was identified in the Ftr1p cytoplasmic C-terminal domain. Deletion or substitution of this residue led to a 70% loss of iron-uptake activity. Ile369 was the only residue identified in this domain that made such a major contribution to iron uptake by the Fet3p-Ftr1p complex.

The high affinity iron uptake complex in the yeast plasma membrane (PM) consists of the ferroxidase, Fet3p, and the ferric iron permease, Ftr1p. We used a combination of yeast two-hybrid analysis, confocal fluorescence microscopy, and fluorescence resonance energy transfer (FRET) quantification to delineate the motifs in the two proteins required for assembly and maturation into an uptakecompetent complex. The cytoplasmic, carboxyl-terminal domain of each protein contains a four-residue motif adjacent to the cytoplasm-PM interface that supports an interaction between the proteins. This interaction has been quantified by two-hybrid analysis and is required for assembly and trafficking of the complex to the PM and for the ϳ13% maximum FRET efficiency determined. In contrast, the Fet3p transmembrane domain (TM) can be exchanged with the TM domain from the vacuolar ferroxidase, Fet5p, with no loss of assembly and trafficking. A carboxyl-terminal interaction between the vacuolar proteins, Fet5p and Fth1p, also was quantified. As a measure of the specificity of interaction, no interaction between heterologous ferroxidase permease pairs was observed. Also, whereas FRET was quantified between fluorescent fusions of the copper permease (monomers), Ctr1p, none was observed between Fet3p and Ctr1p. The results are consistent with a (minimal) heterodimer model of the Fet3p⅐Ftr1p complex that supports the trafficking of iron from Fet3p to Ftr1p for iron permeation across the yeast PM.

Fet3p is a multicopper oxidase recently isolated from the yeast, Saccharomyces cerevisiae. Fet3p is functionally homologous to ceruloplasmin (Cp) in that both are ferroxidases. However, by sequence homology Fet3p is more similar to fungal laccase, and both contain a type 1 Cu site that lacks the axial methionine ligand present in the functional type 1 sites of Cp. To determine the contribution of the electronic structure of the type 1 Cu site of Fet3p to the ferroxidase mechanism, we have examined the absorption, circular dichroism, magnetic circular dichroism, electron paramagnetic resonance, and resonance Raman spectra of wild-type Fet3p and type 1 and type 2 Cu-depleted mutants. The spectroscopic features of the type 1 Cu site of Fet3p are nearly identical to those of fungal laccase, indicating a very similar three-coordinate geometry. We have also examined the reactivity of the type 1 Cu site by means of redox titrations and stopped-flow kinetics. From poised potential redox titrations, the E degrees of the type 1 Cu site is 427 mV, which is low for a three-coordinate type 1 Cu site. The kinetics of reduction of the type 1 Cu sites of four different multicopper oxidases with two different substrates were compared. The type 1 site of a plant laccase (Rhus vernicifera) is reduced moderately slowly by both Fe(II) and a bulky organic substrate, 1,4-hydroquinone (with 6 equiv of substrate, k(obs) = 0.029 and 0.013 s(-)(1), respectively). On the other hand, the type 1 site of a fungal laccase (Coprinus cinereus) is reduced very rapidly by both substrates (k(obs) > 23 s(-)(1)). In contrast, both Fet3p and Cp are rapidly reduced by Fe(II) (k(obs) > 23 s(-)(1)), but only very slowly by 1,4-hydroquinone (10- and 100-fold more slowly than plant laccase, respectively). Semiclassical theory is used to analyze the origin of these differences in reactivity in terms of type 1 Cu site accessibility to specific substrates.