Structure of the Plant Alternative Oxidase
2002; Elsevier BV; Volume: 277; Issue: 2 Linguagem: Inglês
10.1074/jbc.m109853200
ISSN1083-351X
AutoresMary S. Albury, Charles Affourtit, Paul G. Crichton, Anthony L. Moore,
Tópico(s)Metal-Catalyzed Oxygenation Mechanisms
ResumoAll higher plants and many fungi contain an alternative oxidase (AOX), which branches from the cytochrome pathway at the level of the quinone pool. In an attempt, first, to distinguish between two proposed structural models of this di-iron protein, and, second, to examine the roles of two highly conserved tyrosine residues, we have expressed an array of site-specific mutants inSchizosaccharomyces pombe. Mitochondrial respiratory analysis reveals that S. pombe cells expressing AOX proteins in which Glu-217 or Glu-270 were mutated, no longer exhibit antimycin-resistant oxygen uptake, indicating that these residues are essential for AOX activity. Although such data corroborate a model that describes the AOX as an interfacial membrane protein, they are not in full agreement with the most recently proposed ligation sphere of its di-iron center. We furthermore show that upon mutation of Tyr-253 and Tyr-275 to phenylalanines, AOX activity is fully maintained or abolished, respectively. These data are discussed in reference to the importance of both residues in the catalytic cycle of the AOX. All higher plants and many fungi contain an alternative oxidase (AOX), which branches from the cytochrome pathway at the level of the quinone pool. In an attempt, first, to distinguish between two proposed structural models of this di-iron protein, and, second, to examine the roles of two highly conserved tyrosine residues, we have expressed an array of site-specific mutants inSchizosaccharomyces pombe. Mitochondrial respiratory analysis reveals that S. pombe cells expressing AOX proteins in which Glu-217 or Glu-270 were mutated, no longer exhibit antimycin-resistant oxygen uptake, indicating that these residues are essential for AOX activity. Although such data corroborate a model that describes the AOX as an interfacial membrane protein, they are not in full agreement with the most recently proposed ligation sphere of its di-iron center. We furthermore show that upon mutation of Tyr-253 and Tyr-275 to phenylalanines, AOX activity is fully maintained or abolished, respectively. These data are discussed in reference to the importance of both residues in the catalytic cycle of the AOX. alternative oxidase plastid terminal oxidase 3-(N-morpholino)propanesulfonic acid alternative oxidase, all antibodies The alternative oxidase (AOX)1 is a ubiquinol:oxygen oxidoreductase that catalyzes the four electron reduction of oxygen to water (for recent reviews, see Refs. 1Vanlerberghe G.C. McIntosh L. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1997; 48: 703-734Google Scholar, 2Wagner A.M. Moore A.L. Biosci. Rep. 1997; 17: 319-333Google Scholar, 3Siedow J.N. Umbach A.L. Biochim. Biophys. Acta. 2000; 1459: 432-439Google Scholar, 4Affourtit C. Krab K. Moore A.L. Biochim. Biophys. Acta. 2001; 1504: 58-69Google Scholar). This terminal oxidase branches from the cytochrome pathway at the level of the ubiquinone pool and is non-protonmotive. The AOX is resistant to inhibitors of the cytochrome pathway such as cyanide and antimycin A but inhibited by a number of compounds including hydroxamic acids (e.g.salicylhydroxamic acid) and n-alkyl-gallates.Two structural models of the AOX currently exist. The first model proposed by Siedow et al. (5Siedow J.N. Umbach A.L. Moore A.L. FEBS Lett. 1995; 362: 10-14Google Scholar, 6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) was based on relatively few AOX sequences and classified the AOX as a member of the di-iron family of proteins that also includes the R2 subunit of ribonucleotide reductase and the hydroxylase component of methane monooxygenase. Based on hydropathy analysis, the AOX was predicted to contain two transmembrane helices that are connected by a helix located in the intermembrane space (6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) (Fig. 1A). Since this model was proposed, further AOX sequences were identified, resulting in the proposal by Andersson and Nordlund (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar) of an alternative structural model (Fig. 1B). Although this second model also classifies the AOX as a di-iron protein, it differs in the precise ligation sphere of the di-iron center (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar). For instance, one of the C-terminal Glu-X-X-His motifs identified by Siedow et al. (5Siedow J.N. Umbach A.L. Moore A.L. FEBS Lett. 1995; 362: 10-14Google Scholar, 6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar), containing Glu-270, appeared not to be fully conserved in the newly identified sequences and consequently seemed unlikely to play a role in ligating iron. Instead, Andersson and Nordlund used a third Glu-X-X-His motif (that contains Glu-217, which is located in the intermembrane space according to the Siedow et al. model) to coordinate the iron atoms. Since such a choice implies that the transmembrane helices can no longer be retained, Andersson and Nordlund (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar) proposed that the AOX is an interfacial rather than a transmembrane protein.Recently, the IMMUTANS (Im) gene fromArabidopsis thaliana has been sequenced and, interestingly, was found to encode a plastid terminal oxidase (PTOX) that appears to be distantly related to the AOX (8Wu D. Wright D.A. Wetzel C. Voytas D.F. Rodermel S. Plant Cell. 1999; 11: 43-55Google Scholar, 9Carol P. Stevenson D. Bisanz C. Breitenbach J. Sandmann G. Mache R. Coupland G. Kuntz M. Plant Cell. 1999; 11: 57-68Google Scholar). The limited but significant homology of the Im gene to the AOX includes several glutamate and histidine residues located in positions that could contribute to iron binding. In the Im sequence, all but one (Glu-269) of the amino acid residues that were proposed by Andersson and Nordlund (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar) to coordinate the di-iron centers are also present, which these authors considered strong support for their model (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). Importantly, however, the model was subtly adapted in so much that Glu-269 was replaced by Glu-268, a residue that indeed is fully conserved throughout all AOX and PTOX sequences.To probe structure-function relationships, we have previously developed a heterologous expression system in which the Sauromatum guttatum AOX is expressed in the fission yeastSchizosaccharomyces pombe. In this system, the AOX is functionally targeted to the mitochondria and as such confers antimycin-resistant mitochondrial respiration upon S. pombecells (11Albury M.S. Dudley P. Watts F.Z. Moore A.L. J. Biol. Chem. 1996; 271: 17062-17066Google Scholar). Using the system, we have previously shown that mutating Glu-270 to alanine abolishes this antimycin-resistant respiratory activity, suggesting that this residue is important for AOX activity (12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar). In agreement with this finding, mutation of a corresponding glutamate residue in the Trypanosoma brucei sequence that was expressed in Escherichia coli also resulted in an inactive AOX protein (13Chaudhuri M. Ajayi W. Hill G.C. Mol. Biochem. Parasitol. 1998; 95: 53-68Google Scholar).In addition to a catalytic site for the reduction of oxygen, the AOX must also contain a binding site for its reducing substrate, ubiquinol. Recognition of several structural features generally considered important for quinone binding (14Deisenhofer J. Epp O. Miki K. Huber R. Michel H. Nature. 1985; 318: 618-624Google Scholar, 15Allen J.P. Feher G. Yeates T.O. Komiya H. Rees D.C. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8487-8491Google Scholar) led Moore et al. (6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) to the suggestion that several residues situated at the matrix end of the two membrane-spanning helices, including Tyr-253, could potentially be involved in binding quinone.In this study, we have used our S. pombe expression system to investigate the role of Glu-217, a residue that may distinguish between the two existing models, and to study the potential role of Tyr-253 in quinone binding. In addition, we have examined another tyrosine residue, Tyr-275 (equivalent to Tyr-280 in Berthold's nomenclature; see Ref. 10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar), which is also highly conserved and which has been suggested to function in electron transport or to play a role in quinone binding (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). Respiratory analysis reveals that mutation of Tyr-253 to phenylalanine does not affect the activity of the AOX but, importantly, that the oxidase is inactive upon mutation of Tyr-275 to phenylalanine. Of particular significance is the finding that mutation of Glu-217 to alanine results in an inactive AOX, demonstrating that this residue plays an important role in the catalytic activity of the AOX. Although our results suggest that the AOX is interfacial (as proposed in Ref. 7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar) rather than a transmembranous membrane protein, they are not in full agreement with the precise ligation sphere as outlined by Berthold et al. (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar).RESULTSFig. 2 shows typical respiratory traces and a Western blot analysis of Percoll-purified S. pombe mitochondria containing the wild-type (Fig. 2A), E270N (Fig. 2B), or E217A (Fig. 2C) AOX. It can be seen that mitochondria from the three types of cells readily oxidize NADH in a manner that can be stimulated between 1.7- and 2-fold by the addition of carbonyl cyanide m-chlorophenylhydrazone, demonstrating that electron transfer in these mitochondria is well coupled to the generation of a protonmotive force. Fig. 2Areveals that the respiratory activity exhibited by mitochondria containing the wild-type AOX is partially resistant to antimycin A (∼18% of the NADH-dependent rate). This antimycin-resistant respiratory rate is inhibited by the addition of octyl-gallate, demonstrating that it is due to AOX activity, confirming earlier observations made in nonpurified mitochondria (11Albury M.S. Dudley P. Watts F.Z. Moore A.L. J. Biol. Chem. 1996; 271: 17062-17066Google Scholar, 12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar). Fig. 2also shows that mitochondria isolated from cells containing the mutant oxidases exhibit relatively little (E270N; Fig. 2B) and almost no (E217A; Fig. 2C) antimycin-resistant respiratory activity (∼6 and 2% of the respective NADH-dependent rates). Western blot analysis indicates that the AOX protein is expressed in all three types of cells (Fig. 2D). The intensity of the immunoreactive bands indicates that wild-type, E270N, and E217A forms of the AOX are expressed in amounts that are not significantly different and therefore that lack of AOX activity observed with the mutants is not due to lack of protein expression. These results demonstrate that the mutation of either Glu-217 or Glu-270 results in the loss of AOX activity, suggesting that both residues are equally essential for AOX activity.Figure 2Respiratory and Western analyses of Percoll-purified S. pombe mitochondria.Mitochondria were prepared from cells containing the wild-type AOX (A), the E270N mutant (B), or the E217A mutant (C). Respiratory activity was determined with additions as indicated: 1.8 mm NADH, 2 μm carbonyl cyanidem-chlorophenylhydrazone (CCCP), 3.5 μm antimycin A (AA), and 5 μmoctyl-gallate (OG). Numbers on thetrace refer to oxygen consumption rates (expressed in nmol of O2/min/mg of protein). D, immunoblot of mitochondria from the same samples as in A–C probed with AOA antibodies. The numbers on the right of theimmunoblot refer to the molecular masses of the standards (in kDa).View Large Image Figure ViewerDownload (PPT)Table I summarizes data obtained from extensive mitochondrial respiratory analyses and indicates the rates of coupled, uncoupled, and antimycin-resistant oxidation of NADH exhibited by the wild type and a variety of AOX mutants. From these results, it would seem that both the coupled and uncoupled NADH-dependent activities of the E217A mutant are slightly lower than those of the wild type and indeed of all the mutants studied. Although close scrutiny of the data would suggest that this observation is statistically not significant, a possible reason for the potential variations is, at present, unclear.Table IRespiratory activity of Percoll-purified S. pombe mitochondria containing a wild-type or mutated AOX proteinNADHCCCPAARateS.D.RateS.D.RateS.D.nmol O2/min/mg proteinWild type22381 (28, 6)373137 (15, 6)4115 (13, 3)E270N16227 (5, 2)33277 (5, 2)6.71.4 (5, 2)E217A12831 (5, 2)21645 (5, 2)1.71.5 (5, 2)C172A17915 (3, 1)39925 (3, 1)339 (2, 1)Y253F17041 (21, 6)35299 (7, 6)4913 (14, 3)Y275F18269 (9, 3)29119 (9, 3)2.41.7 (10, 3)Activity was determined upon the subsequent addition of 1.8 mm NADH, 2 μm carbonyl cyanidem-chlorophenylhydrazone (CCCP), and 3.5 μmantimycin A (AA). The rates are average values of oxygen consumption (expressed in nmol of O2/min/mg of protein) and were corrected by subtracting the residual respiratory rate observed after the addition of 5 μm octyl-gallate. The numbers in parentheses indicate the number of respiratory traces and mitochondrial preparations performed, respectively, on which the average rates are based. Open table in a new tab In an attempt to determine if mutagenesis per se results in an inactive enzyme, which would invalidate the data shown in Fig. 2 and Table I, Cys-172, a highly conserved residue located in the N-terminal region of the protein, was mutated to alanine. Previous mutation of the corresponding cysteine in plant AOX sequences has been shown not to have any deleterious effect on AOX activity (24Vanlerberghe G.C. McIntosh L. Yip J.Y.H. Plant Cell. 1998; 10: 1551-1560Google Scholar, 25Rhoads D.M. Umbach A.L. Sweet C.R. Lennon A.M. Rauch G.S. Siedow J.N. J. Biol. Chem. 1998; 273: 30750-30756Google Scholar), and we considered this a suitable residue to act as a control. It is clear from Table Ithat upon mutation of Cys-172, the AOX is indeed still fully active as reflected by a mitochondrial antimycin-resistant respiratory activity of 33 nmol of O2/min/mg of protein, which is not significantly different from wild-type activity. Such an observation confirms that site-directed mutagenesis per se does not result in an inactive enzyme, and we therefore consider the data presented in Table I a valid indication of the importance of the studied residues for the structure and mechanism of the AOX.In the Moore et al. (6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) model, a conserved tyrosine residue, Tyr-253, that is situated at the base of one of the transmembrane helices was proposed to be involved in quinone binding. In this respect, an additional tyrosine residue, Tyr-275, has also been identified as of potential importance by Berthold et al.(10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). Table I shows that cells expressing the AOX protein in which Tyr-253 was mutated to phenylalanine exhibit a mitochondrial antimycin-resistant respiratory activity that is comparable with that of the wild type. However, in cells expressing a protein in which Tyr-275 was mutated to phenylalanine, such activity is barely detectable (Table I). Western blot analysis indicates that the Y253F and Y275F AOX proteins are expressed in equal amounts (data not shown).DISCUSSIONDespite significant advances in our understanding of the AOX, the enzyme has proved difficult to purify to homogeneity and is elusive to spectroscopic analyses. However, progress in the characterization of the AOX has included isolation of cDNA sequences from a number of species, which has led to the development of two hypothetical models of the AOX (5Siedow J.N. Umbach A.L. Moore A.L. FEBS Lett. 1995; 362: 10-14Google Scholar, 6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar, 7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar). Although both models classify the AOX as a member of the di-iron carboxylate family of proteins, the major difference between them concerns the topology of the enzyme within the inner mitochondrial membrane. In this paper, we have presented data obtained from studies in which an array of site-directed AOX mutants was expressed in S. pombe, a functional expression system that is highly suited to study AOX activity in situ (11Albury M.S. Dudley P. Watts F.Z. Moore A.L. J. Biol. Chem. 1996; 271: 17062-17066Google Scholar, 12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar, 26Affourtit C. Albury M.S. Krab K. Moore A.L. J. Biol. Chem. 1999; 274: 6212-6218Google Scholar). Importantly, our results provide the first experimental information to distinguish between the two structural models and, furthermore, provide insights into the possible roles of two highly conserved tyrosine residues.From Fig. 2C it is evident that mutation of Glu-217 to alanine produces an inactive AOX, strongly suggesting that Glu-217 is essential for activity. This result has major implications as to the plausibility of the two structural models. Siedow et al. (5Siedow J.N. Umbach A.L. Moore A.L. FEBS Lett. 1995; 362: 10-14Google Scholar,6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) implicitly suggested that Glu-217 would be located at the cytoplasmic side of the inner mitochondrial membrane. Since both reduction of oxygen and oxidation of ubiquinol were proposed to occur toward the matrix side of the membrane, it is hard to envisage how Glu-217 would play a functional role in such a model. In contrast, Glu-217 plays a key role in the coordination of the di-iron center in the Andersson and Nordlund model (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar) (it bridges the iron atoms). Our result, therefore, supports the latter model, which would imply that the AOX is indeed an interfacial membrane protein associated with the matrix side of the membrane. Analysis of AOX sequences, using the membrane protein topology prediction method that was recently reported by Krogh et al. (27Krogh A. Larsson B. von Heijne G. Sonnhammer E.L.L. J. Mol. Biol. 2001; 305: 567-580Google Scholar), corroborates this notion. Although two relatively hydrophobic regions were identified in the primary structure of the AOX, these were considered, based on the algorithm's stringent criteria (27Krogh A. Larsson B. von Heijne G. Sonnhammer E.L.L. J. Mol. Biol. 2001; 305: 567-580Google Scholar), not to be membrane-spanning (data not shown).Following the identification of the Im gene, 19 amino acid residues appeared totally conserved between the AOX and PTOX sequences (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). Berthold et al. (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar) focused upon these totally conserved residues as additional evidence in support of the Andersson and Nordlund model. However, certain regions of significant homology in AOX sequences are absent in the PTOX sequences. For example, three glutamate residues (Glu-268, Glu-269, and Glu-270) are completely conserved in all AOX sequences examined to date, but only one glutamate (Glu-268) is present in the PTOX. In the Andersson and Nordlund model, Glu-270 plays no essential role, since it was suggested not to be part of the ligation sphere. In contrast, the results presented in this paper, in addition to our previous publication (12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar) and along with the mutation of the Trypanosome AOX (13Chaudhuri M. Ajayi W. Hill G.C. Mol. Biochem. Parasitol. 1998; 95: 53-68Google Scholar), would suggest that Glu-270 is an important residue. If it is the case that Glu-270 does not ligate iron, then the only likely way that mutation of Glu-270 could affect AOX activity is by disrupting the structural stability of the active site because of its proximity to the di-iron center. However, we have previously shown that, like the wild-type AOX, the E270N mutant is correctly targeted and integrated within the inner mitochondrial membrane (12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar). As argued previously (12Albury M.S. Affourtit C. Moore A.L. J. Biol. Chem. 1998; 273: 30301-30305Google Scholar), it is highly unlikely that an incorrectly folded protein would be targeted and integrated in such a successful manner. Based on Fig. 2, it can therefore be concluded that Glu-217 is required for activity of the AOX, supporting the proposal by Andersson and Nordlund that the protein is interfacial. However, the lack of activity of the E270N mutant AOX would suggest that the detailed ligation sphere proposed by these authors is not necessarily the correct one for all species.With the identification of the Im gene and its homology to the AOX, it may be tempting to focus totally on the conserved residues between these two distantly related proteins. Equally, focusing on theIm gene may have the effect of grouping all AOX and PTOX proteins into a single category, which may not be advisable. Umbach and Siedow (28Umbach A.L. Siedow J.N. Arch. Biochem. Biophys. 2000; 378: 234-245Google Scholar) suggest that there appear to be two general types of AOX sequences in plants and fungi. From sequence comparisons, they have demonstrated that Im contains neither a domain present in plants, surrounding the regulatory cysteine, nor a fungus-specific sequence (28Umbach A.L. Siedow J.N. Arch. Biochem. Biophys. 2000; 378: 234-245Google Scholar). Hence, there may be subtle differences between AOX sequences from different species with roles not yet identified, and it is therefore interesting to speculate that there could be two or more classes of AOX and that possibly Im belongs to a third group. We feel that one should be cautious in the classification of all of the AOX proteins (plants, algae, protist, and fungi) together with the PTOX as one group. Whereas the overall structure of the enzyme, including its active site, might be conserved, subtle species-specific differences as to the exact ligation sphere of the di-iron center may exist.The structure of several enzymes that interact with quinone/quinol have been determined to atomic resolution (14Deisenhofer J. Epp O. Miki K. Huber R. Michel H. Nature. 1985; 318: 618-624Google Scholar, 15Allen J.P. Feher G. Yeates T.O. Komiya H. Rees D.C. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8487-8491Google Scholar, 29Iwata S. Lee J.W. Okada K. Lee J.K. Iwata M. Rasmussen B. Link T.A. Ramaswamy S. Jap B.K. Science. 1998; 281: 64-71Google Scholar). However, this has not yet led to the elucidation of a universal structure of a quinone-binding site. Several features that might generally be important in quinone binding have nevertheless been suggested. For example, mapping of the quinone-binding site in complex III has resulted in a model analogous to quinone sites in photosynthetic reaction centers in which the quinone-binding pocket is formed by the ends of two transmembrane helices (30Zhang Z. Huang L. Shulmeister V.M. Chi Y.I. Kim K.K. Hung L.W. Crofts A.R. Berry E.A. Kim S.H. Nature. 1998; 392: 677-684Google Scholar). Aromatic residues have also been identified near quinone-binding regions where the aromatic ring may interact with the quinone head group in a parallel manner (15Allen J.P. Feher G. Yeates T.O. Komiya H. Rees D.C. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8487-8491Google Scholar, 31Stowell M.H.B. McPhillips T.M. Rees D.C. Soltis S.M. Abresch E. Feher G. Science. 1997; 276: 812-816Google Scholar). Tyr-253, originally identified in the model proposed by Moore et al. (6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar) as situated at the base of the transmembrane helices, has been proposed to be involved in quinone binding. In the Andersson and Nordlund model, Tyr-253 is located in a hydrophobic crevice that could potentially form a quinone-binding site (7Andersson M.E. Nordlund P. FEBS Lett. 1999; 449: 17-22Google Scholar, 10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). However, it is important to recognize that this residue could have other functions. Reactive tyrosines are important in a number of enzyme activities including photosystem II (32Hoganson C.W. Babcock G.T. Science. 1997; 277: 1953-1956Google Scholar) and cytochrome oxidase (33Proshlyakov D.A. Pressler M.A. DeMaso C. Leykam J.F. DeWitt D.L. Babcock G.T. Science. 2000; 290: 1588-1591Google Scholar, 34Ostermeier C. Harrenga A. Ermler U. Michel H. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 10547-10553Google Scholar). Modification of bacterial reaction centers to mimic photosystem II indicates that, after mutation to tyrosines, these substituted residues could participate in radical formation (35Kalman L. LoBrutto R. Allen J.P. Williams J.C. Nature. 1999; 402: 696-699Google Scholar). Therefore, it has been suggested that, in addition to a role in quinone binding, Tyr-253 could provide a tyrosyl radical involved in electron transport or, alternatively, may play a structural role (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). Our results appear to rule out one of the possible roles for Tyr-253, namely a role in electron transport, since its mutation to phenylalanine does not result in any loss of catalytic activity. However, based on the data in this report, it is not possible yet to say whether Tyr-253 has a role in quinone binding or some structural role. Preliminary results from our laboratory suggest that both the wild-type and the Y253F mutated AOX exhibit similar sensitivities to the phenolic inhibitors salicylhydroxamic acid and octyl-gallate. 2P. G. Crichton, C. Affourtit, M. S. Albury, and A. L. Moore, unpublished observations. This may indicate that Tyr-253 is not involved in quinone binding. It should be emphasized, however, that our steady-state oxygen consumption measurements lack the time resolution to conclusively exclude a role for Tyr-253 in the binding and/or oxidation of ubiquinol.The proposal that Tyr-253 could play a role in quinone binding was originally based on a limited amount of sequence data (6Moore A.L. Umbach A.L. Siedow J.N. J. Bioenerg. Biomembr. 1995; 27: 367-377Google Scholar). However, examination of the currently available sequences reveals that this residue is totally conserved among all AOX and even the PTOX sequences. Recently, a computer search of respiratory and photosynthetic complexes that react with quinones has resulted in the identification of a putative quinone-binding motif, consisting of a triad element and a His-Arg pair, which may be present, although weakly conserved, in the AOX (36Fisher N. Rich P.R. J. Mol. Biol. 2000; 296: 1153-1162Google Scholar). The proposed quinone-binding role of Tyr-253 is all the more intriguing, since this residue is situated close to the His-Arg pair. Furthermore, Tyr-253 is in close proximity to residues that have been identified through inhibitor (salicylhydroxamic acid) resistance screens, as of importance in quinone binding (37Berthold D.A. Biochim. Biophys. Acta. 1998; 1364: 73-83Google Scholar).As discussed previously, tyrosine residues could fulfill a number of roles, and, interestingly, several conserved tyrosine residues are present in the AOX. These include Tyr-266, Tyr-275, and Tyr-299, all of which are highly conserved, although Tyr-266 and Tyr-299 not fully across all species. Tyr-266 has been replaced by a phenylalanine inChlamydomonas reinhardtii and by a cysteine in all of the PTOX sequences. Tyr-299 is conserved in the PTOX and in all AOX sequences except the Trypanosome sequence. Apart from Tyr-253, Tyr-275 is the only tyrosine residue that is present in all available AOX and PTOX sequences. Interestingly, the Andersson and Nordlund model places Tyr-275 in close spacial proximity to the di-iron center (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar). The results presented in Table I show that mutation of Tyr-275 results in an inactive AOX, suggesting that Tyr-275, unlike Tyr-253, is essential for AOX activity and is potentially involved in electron transport.Based on the proposed reaction mechanisms of MMOH (the hydroxylase component of methane monooxygenase) and the R2 subunit of ribonucleotide reductase (cf. Ref. 38Wallar B.J. Lipscomb J.D. Chem. Rev. 1996; 96: 2625-2657Google Scholar), two mechanisms have been put forward for the catalytic cycle of the AOX (10Berthold D.A. Andersson M.E. Nordlund P. Biochim. Biophys. Acta. 2000; 1460: 241-254Google Scholar, 39Affourtit C. Albury M.S. Whitehouse D.G. Moore A.L. Pest Manag. Sci. 2000; 56: 31-38Google Scholar), neither of which predicts involvement of a reactive tyrosine residue. Our results are the first to suggest such a role, which is not without precedence in oxidase activity (40Babcock G.T. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 12971-12973Goo
Referência(s)