Identification and Characterization of FET3

Using a selection system that was enriched for mutants unable to grow on low-iron media, Askwith et al. reported the identification of a mutant, fet3, that was unable to grow on low-iron media (14). This mutant had a normal surface reductase activity but was unable to accumulate 59Fe. A gene that could complement both the low-iron growth defect and the inability to accumulate radioactive iron was identified by complementation of the mutant strain with a genomic library. Genetic studies determined that the complementing gene was the normal allele of the mutated gene. The sequence of this gene, termed FET3, showed homology to members of the multicopper oxidase family. The closest homology was to ascorbate oxidase, but FET3 showed significant homology to all multicopper oxidases. The homology was seen in a signature sequence that defines multicopper oxidases and in the sequences thought to ligate the coppers. A feature that distinguished Fet3p from the other known multicopper oxidases was the presence of a single transmembrane domain. Fet3p is made as a higher-molecular-weight precursor with an amino terminal extension. This extension is a leader sequence and gains entry of Fet3p into the secretory apparatus where Fet3p is glycosylated. The carboxyl terminal transmembrane domain is a membrane anchor tethering Fet3p to the membrane. Studies using epitope-tagged Fet3p demonstrated that Fet3p is a plasma membrane protein. At the time of its identification, most multicopper oxidases were found as secreted products in both plants and animals.

The sequence of Fet3p indicated that it was a multicopper oxidase. This indication was strengthened by the discovery that copper depletion results in an inability to grow on low-iron media. Copper depletion was accomplished either genetically through the identification of mutants unable to grow on copper (15) or through nutritional studies in which media copper was made limiting (14). In either case, copper-deprived yeast does not have a functional high-affinity iron transport system. This work provided strong evidence that copper was required for high-affinity iron transport and supported the role of a multicopper oxidase in iron transport.

Studies on oxygen consumption provided further evidence for a putative ferroxidase in iron transport. Measurement of oxygen consumption by yeast showed both a FET3-dependent and FET3-inde-pendent oxygen consumption (16). Most of the oxygen consumed by cells grown on iron-rich media was the result of mitochondrial consumption. In cells that were iron starved, however, there was an increased rate of oxygen consumption. If the FET3 gene was deleted, then the enhanced rate of oxygen consumption was reduced and could be accounted for by mitochondrial respiration. Inhibi tion of respiration, either by pharmacological agents or by introduction of mutations into genes encoding respiratory proteins, resulted in a marked decrease in mitochondrial oxygen consumption, but not in FET3-induced oxygen consumption. Further studies demonstrated that the FET3-depen-dent rate of oxygen consumption was reduced in copper-starved yeast and could also be reduced by the addition to spheroplasts of an antibody directed against the extracellular domain of Fet3p but not by antibodies directed against the intracellular domain. Calculation of the ratio of the rates of FET3-dependent oxygen consumption and iron accumulation revealed a value of 4/1. This ratio conforms to the stoichiometry expected of a multicopper oxidase. The final piece of physiological data showing a role for a multicopper oxidase in iron consumption is that cells grown under anaerobic conditions are unable to accumulate iron by the FET3-dependent iron-transport system, rather, they employ the low-affinity iron-transport system that relies on the product of the FET4 gene. Under anaerobic conditions, cells can only grow in media containing high concentrations of iron. A /et3-deletion strain that express the FET4 gene product can grow, but a FET4-deletion strain cannot grow even though it has a good FET3 gene. Further, the transcription of the FET3 gene is repressed under anaerobic conditions (17). These results show that FET3 effects iron-dependent oxygen consumption, and without oxygen, cells are unable to take up iron by the Fet3p transport system.

Although the above studies showed that FET3 was a multicopper oxidase, the specific role of the oxidase remained unclear. It is possible that Fet3p oxidizes a protein, rather than iron. The suggestion that Fet3p acted as a ferroxidase required the isolation of Fet3p and a direct measurement of ferroxidase activity. This was first accomplished by deSilva et al., who demonstrated that purified Fet3p could oxidize iron and that it could also load iron onto transferrin (18). Examination of a number of potential substrates for multicopper oxidases showed iron to be the preferred substrate. The Km for iron oxidation, 0.5 uM was very close to the Km for iron transport. Purified Fet3p could iron load transferrin, although the rate of iron loading was low relative to that of ceruloplasmin. Reasons for the low rate may be that detergent was use to isolate Fet3p and as isolated Fet3p is heavily glycosylated. The presence of both detergent and carbohydrate may reduce the rate of transferrin iron loading through steric hindrance. Hasset et al. using molecular engineering produced a Fet3p that lacked a transmembrane domain (19). This protein is secreted into the media, permitting its isolation without the use of detergents. The soluble Fet3p showed a similar rate of iron oxidation as the transmembrane containing Fet3p, both of which then had a lower rate of iron turnover than Cp. A highly active fragment of Fet3p can be released by proteases from the yeast Pichia pastoris (20). This fragment lacks the transmembrane domain and can be isolated without the use of detergents. The isolated Fet3p oxidizes iron at the same rate as ceruloplasmin. It is unclear why the Fet3p from Pichia would have a faster rate of iron oxidation than the Fet3p from S. cerevisiae.

That purified Fet3p can oxidize iron suggests that it functions as an ferroxidase in iron transport. Mutagenesis studies designed to determine the molecular basis of iron oxidation support the view that the ferroxidase activity of Fet3p is essential for iron transport. Although there are many multicopper oxidases, all of which can oxidize organic substrates, only Fet3p and Cp (and its homologs) can also oxidize iron. A comparison of the sequence of Fet3p to the sequences of other multicopper oxidases led to the suggestion of several structural features of Fet3p that were thought to be responsible for iron oxidation. Askwith et al. made a series of site-specific mutants in an attempt to determine what features were responsible for iron oxidation (21). Among the residues mutagenized were those thought to ligate the type-1 copper and thus were responsible for setting the oxidation potential. In most multicopper oxidases, a methionine is a ligand for the type-1 copper. In Fet3p, that methionine is replaced by leucine, suggesting a geometry typical of a high-redox-potential copper site. When that leucine was replaced by a methionine, the protein retained ferroxidase activity and the rate of iron uptake was reduced by less than twofold. Replacement of the leucine with other amino acids severely reduced both multicopper oxidase activity and cellular iron transport. Other residues suggested to be involved in iron oxidation (glutamic acid 227, aspartic acid 228, and glutamic acid 330) when changed to alanine did not affect either iron transport or ferroxidase activity. In a more recent study, structural determinants for ferroxidase activity were based on modeling studies (22). The amino acids Glu185 and Tyr354 were thought to be important for iron binding. Alteration of these amino acids by site-specific mutagenesis had an effect on ferroxidase activity (23). Most impressively, alteration of Glu185 to Ala reduced ferroxidase activity by 95% but only reduced oxidase activity by 60%. This is the first result that suggests that the two enzymatic activities could be separated. Unfortunately, iron transport activity was not assayed. A measurement of iron transport could have provided compelling evidence that the ferroxidase activity of Fet3p is the critical aspect of its role in iron transport. If the ferroxidase activity of Fet3p is required for iron transport, then a prediction is that the Glu185Ala mutant would show a 95% decrease in iron transport. Even without this experiment, the consensus view is that Fet3p plays a role in iron transport as a result of its ferroxidase activity.

Although Fet3p is a ferroxidase converting Fe(II) into Fe(III), it only has a single transmembrane domain and is unlikely to transport iron across the bilayer. A genetic screen performed by Stearman et al. revealed a second gene required for high-affinity iron transport (24). This gene, termed FTR1, encoded a protein with six transmembrane domains. Stearman et al. identified a sequence within Ftrlp comprising the amino acids REGLE, which is similar to a motif that implicated in the iron-binding region of mammalian L-chain ferritin. Mutation of these residues in Ftrlp abrogates its ability to transport iron. These authors also provided genetic evidence that Fet3p and Ftrlp form a complex. In the absence of Ftrlp, Fet3p did not localize to the cell surface. Similarly, in the absence of Fet3p, Ftrlp did not localize to the cell surface. These results suggest that both molecules must be synthesized simultaneously so that they can be assembled to form a complex that permits the correct targeting of either to the cell surface.

Confirmation that Fet3p and Ftrlp formed a functional complex came from studies of a homologous complex in different yeast Schizosaccharomycespombe. The fission yeast, S. pombe, also has a ferroxidase-based high-affinity iron-transport system (25). This system consists of an oxidase termed fio1+ and a transmembrane transporter fip1+. These genes are highly homologous to the S. cerevisiae genes, with fiolp + sharing 38% identity and 60% similarity on the amino acid level with Fet3p and frplp sharing 46% identity and 70% homology on the amino acid level percentage with Ftrlp. Expression of fiolp in a S. cerevisiae strain lacking FET3 does not permit high-affinity iron uptake (Fig. 3). Thus, fiolp cannot partner with Ftrlp. If, however, fiplp is expressed at the same time as fiolp in the S. cerevisiaefet3-deletion strain, then the strain is capable of high-affinity iron transport. These results suggest that the oxidase and permease function as a unit and imply that Fet3p and Ftrlp are the only two plasma membrane proteins required to carry out iron transport.

Other yeasts, aside from S. pombe and S. cerevisiae, show an oxidase/permease-based iron-transport system. A ferroxidase was shown to be essential for iron transport in Candida albicans as deletion of the C. albicans FET3 prevents high-affinity iron transport (26,27). In one study, it was reported that the C. albicansfet3-deletion strain grew poorly in infected mice, suggesting that the FET3 gene may be a virulence factor (27). It is curious that the C. albicans FET3 when expressed in a S. cerevisiaefet3-deletion strain complements the low-iron growth defect (26). This result suggests that C. abicans Fet3p can form a complex with S. cerevisiae Ftrlp, whereas the FET3 homolog of S. pombe, fio1+, cannot. Biochemical studies have shown that P. pastoris also expresses a ferroxidase similar Fet3p (19). It may well be that a ferroxidase/permease iron-transport system is common among yeasts.

The role of a ferroxidase in yeast iron transport, however, presents a conundrum. The reaction of Fe(II) with Fet3p generates Fe(III). Yet, the substrate for Fet3p is Fe(II) produced by the ferrireductases. The overall reaction is therefore:

Fe(III) + e- ^ Fe(II) (ferriductase) Fe(II) + 4H+ +O2 ^ Fe(III) + 2H2O (oxidase)

Fe Fe Fe

Fig. 3. Genetic evidence for complex formation between the multicopper oxidase and the transmembrane permease. See text for details.

Within the same domain, the cell surface, there are two diametrically opposed enzymatic reactions: a reduction and then an oxidation. Why do yeasts require iron to be first reduced and then oxidized? The explanation for this requirement is twofold. First, ferric iron is essentially insoluble and is not readily bioavailable. Iron is most often found as insoluble ferric hydroxide. Most bioavailable iron is found chelated to small organic molecules. There is little structural similarity in the organic chelates and, most often, iron is shielded from the media by organic molecules. Therefore, there is little basis for a chemical recognition of ferric chelates. Reduction of ferric iron by electron addition generates ferrous iron, which is much more soluble and can exist in aqueous fluids at measurable concentrations.

Ferric oxidation may be required to provide specificity to the transport process. To date, all known ferrous transport systems show a low degree of specificity, as they can mediate the accumulation of other transition metals. In yeast, for example, the FET4-transport system can also accumulate copper and cobalt in addition to iron (13). In mammalian cells, DMT1 (also referred to as Nramp2 or DCT1) transports to iron, zinc, manganese, and cobalt (28). In contrast, the two known Fe(III)-transport systems, FET3/FTR1 in yeast and Cp and transferrin in vertebrates, are exquisitely specific for iron. There is no other relevant transition metal transported by these systems. (Aluminum and gallium can bind to transferrin.) The basis for specificity is the binding of iron to the transporter (FET3) or carrier, transferrin. As iron is insoluble, it needs specific ligands to generate high-affinity binding. The ligation of Fe(III) to amino acids imparts specificity on the binding process. Further, the enzyme reaction carried out by the multicopper oxidases generates Fe(III) without producing oxygen radicals.

Cure Your Yeast Infection For Good

Cure Your Yeast Infection For Good

The term vaginitis is one that is applied to any inflammation or infection of the vagina, and there are many different conditions that are categorized together under this ‘broad’ heading, including bacterial vaginosis, trichomoniasis and non-infectious vaginitis.

Get My Free Ebook

Post a comment