Siroheme- and Fe4-S4-dependent NirA from Mycobacterium tuberculosis Is a Sulfite Reductase with a Covalent Cys-Tyr Bond in the Active Site

The nirA gene of Mycobacterium tuberculosis is up-regulated in the persistent state of the bacteria, suggesting that it is a potential target for the development of antituberculosis agents particularly active against the pathogen in its dormant phase. This gene encodes a ferredoxin-dependent sulfite reductase, and the structure of the enzyme has been determined using x-ray crystallography. The enzyme is a monomer comprising 555 amino acids and contains a Fe4-S4 cluster and a siroheme cofactor. The molecule is built up of three domains with an α/β fold. The first domain consists of two ferredoxin-like subdomains, related by a pseudo-2-fold symmetry axis passing through the whole molecule. The other two domains, which provide much of the binding interactions with the cofactors, have a common fold that is unique to the sulfite/nitrite reductase family. The domains form a trilobal structure, with the cofactors and the active site located at the interface of all three domains in the center of the molecule. NirA contains an unusual covalent bond between the side chains of Tyr69 and Cys161 in the active site, in close proximity to the siroheme cofactor. Removal of this covalent bond by site-directed mutagenesis impairs catalytic activity, suggesting that it is important for the enzymatic reaction. These residues are part of a sequence fingerprint, able to distinguish between ferredoxin-dependent sulfite and nitrite reductases. Comparison of NirA with the structure of the truncated NADPH-dependent sulfite reductase from Escherichia coli suggests a binding site for the external electron donor ferredoxin close to the Fe4-S4 cluster. The nirA gene of Mycobacterium tuberculosis is up-regulated in the persistent state of the bacteria, suggesting that it is a potential target for the development of antituberculosis agents particularly active against the pathogen in its dormant phase. This gene encodes a ferredoxin-dependent sulfite reductase, and the structure of the enzyme has been determined using x-ray crystallography. The enzyme is a monomer comprising 555 amino acids and contains a Fe4-S4 cluster and a siroheme cofactor. The molecule is built up of three domains with an α/β fold. The first domain consists of two ferredoxin-like subdomains, related by a pseudo-2-fold symmetry axis passing through the whole molecule. The other two domains, which provide much of the binding interactions with the cofactors, have a common fold that is unique to the sulfite/nitrite reductase family. The domains form a trilobal structure, with the cofactors and the active site located at the interface of all three domains in the center of the molecule. NirA contains an unusual covalent bond between the side chains of Tyr69 and Cys161 in the active site, in close proximity to the siroheme cofactor. Removal of this covalent bond by site-directed mutagenesis impairs catalytic activity, suggesting that it is important for the enzymatic reaction. These residues are part of a sequence fingerprint, able to distinguish between ferredoxin-dependent sulfite and nitrite reductases. Comparison of NirA with the structure of the truncated NADPH-dependent sulfite reductase from Escherichia coli suggests a binding site for the external electron donor ferredoxin close to the Fe4-S4 cluster. Mycobacterium tuberculosis, the causative agent of tuberculosis, poses a major threat to human health. Over the last decade, the number of registered cases has been progressively increasing, resulting presently in approximately 2 million deaths per year (statistics available on the World Wide Web at www.who.int). The emergence of multidrug-resistant strains of M. tuberculosis raises serious concerns about future capabilities to control this pathogen. Chemotherapy is further complicated by the ability of M. tuberculosis to persist in the lungs of infected individuals for decades by switching to a dormant or latent phase (1Bloom B.R. McKinney J.D. Nat. Med. 1999; 5: 872-874Crossref PubMed Scopus (41) Google Scholar), which also induces tolerance to current antibiotics (2Wayne L.G. Sramek H.A. Antimicrob. Agents Chemother. 1994; 38: 807-811Crossref Scopus (275) Google Scholar, 3Wallis R.S. Patil S. Cheon S.H. Edmonds K. Phillips M. Perkins M.D. Joloba M. Namale A. Johnson J.L. Teixeira L. Dietze R. Siddiqi S. Mugerwa R.D. Eisenach K. Ellner J.J. Antimicrob. Agents Chemother. 1999; 43: 2600-2666Crossref PubMed Google Scholar). Estimates by the WHO suggest that about one-third of the world's population is infected with persistent mycobacteria. Reactivation of these dormant bacteria can occur either spontaneously or as the consequence of an immunocompromised state (e.g. HIV infection/therapy), resulting in active tuberculosis. It is thus clear that a successful long term strategy against M. tuberculosis requires antibiotics targeting the bacteria also in the persistent state. Latency is associated with nonreplicating or very slow growth of M. tuberculosis, and several experimental in vitro models for the dormant phase of the bacilli have been developed (4Wayne L.G. Hayes L.G. Infect. Immun. 1996; 64: 2062-2069Crossref PubMed Google Scholar, 5Betts J.C. Lukey P.T. Robb L.C. McAdam R.A. Duncan K. Mol. Microbiol. 2002; 43: 717-731Crossref PubMed Scopus (1131) Google Scholar, 6Voskuil M.I. Schnappinger D. Visconti K.C. Harrell M.I. Dolganov G.M. Sherman D.R. Schoolnik G.K. J. Exp. Med. 2003; 198: 705-713Crossref PubMed Scopus (769) Google Scholar, 7Hu Y. Mangan J.A. Dhillon J. Sole K.M. Mitchison D.A. Butcher P.D. Coates A.R. J. Bacteriol. 2000; 182: 6358-6365Crossref PubMed Scopus (139) Google Scholar). Comparison of gene expression profiles and proteome analyses of active versus nonreplicating bacteria have identified a number of genes that are up-regulated in the dormant phase. Genes involved in oxidative stress, anaerobic respiration, and the metabolism of sulfur have consistently been identified as up-regulated in response to limited access to oxygen and nutrient starvation (8Hu Y. Coates A.R.M. FEMS Microbiol. Lett. 2001; 202: 59-65Crossref PubMed Google Scholar, 9Schnappinger D. Ehrt S. Voskuil M.I. Liu Y. Mangan J.A. Monahan I.M. Dolganov G. Efron B. Butcher P.D. Nathan C. Schoolnik G.K. J. Exp. Med. 2003; 198: 693-704Crossref PubMed Scopus (1143) Google Scholar, 10Hampshire T. Soneij S. Bacon J. James B.W. Hinds J. Laing K. Stabler R.A. Marsh P.D. Butcher P.D. Tuberculosis. 2004; 84: 228-238Crossref PubMed Scopus (174) Google Scholar, 11Starck J. Källenius G. Marklund B.-I. Andersson D.I. Åkerlund T. Microbiology. 2004; 150: 3821-3829Crossref PubMed Scopus (128) Google Scholar). One of the genes active in the dormant phase of M. tuberculosis is nirA (Rv2391) (9Schnappinger D. Ehrt S. Voskuil M.I. Liu Y. Mangan J.A. Monahan I.M. Dolganov G. Efron B. Butcher P.D. Nathan C. Schoolnik G.K. J. Exp. Med. 2003; 198: 693-704Crossref PubMed Scopus (1143) Google Scholar, 10Hampshire T. Soneij S. Bacon J. James B.W. Hinds J. Laing K. Stabler R.A. Marsh P.D. Butcher P.D. Tuberculosis. 2004; 84: 228-238Crossref PubMed Scopus (174) Google Scholar). Himar1 transposon mutagenesis has further shown that nirA is an essential gene (12Sasetti C.M. Boyd D.H. Rubin E.J. Mol. Microbiol. 2003; 48: 77-84Crossref PubMed Scopus (1998) Google Scholar). The amino acid sequence, derived from the nirA gene, shows homology to a family of ferredoxin-dependent sulfite/nitrite reductases. These enzymes are found in archaea, bacteria, fungi, and plants (for a review, see Refs. 13Crane B.R. Getzoff E.D. Curr. Opin. Struct. Biol. 1996; 6: 744-756Crossref PubMed Scopus (122) Google Scholar and 14Nakayama M. Akashi T. Hase T. J. Inorg. Biochem. 2000; 82: 27-32Crossref PubMed Scopus (79) Google Scholar). The sulfite reductases catalyze the reduction of sulfite to sulfide, one step in the biosynthesis of sulfur-containing amino acids and cofactors. Nitrite reductases participate in the assimilation of nitrogen via nitrate in plants and can also act in anaerobic energy metabolism. This class of sulfite/nitrite reductases generally accepts both nitrite and sulfite as substrate, but the particular metabolic function is reflected in pronounced differences in the kinetic parameters. These enzymes contain a unique combination of cofactors, a Fe4-S4 iron-sulfur cluster and a siroheme (15Krueger R.J. Siegel L.M. Biochemistry. 1982; 21: 2892-2904Crossref PubMed Scopus (112) Google Scholar). NirA also shows weak amino acid sequence similarity to bacterial NADPH-dependent Fe4-S4- and siroheme-containing sulfite reductases (e.g. 23% sequence identity to CysI from Escherichia coli). Unlike the monomeric ferredoxin-dependent enzymes, the NADPH-dependent sulfite reductases are oligomeric complexes consisting of four or eight subunits of the hemoprotein component (CysI) and eight flavoprotein subunits that deliver electrons derived from NADPH to the redox centers of the hemoprotein subunit (16Siegel L.M. Davis P.S. Kamin H. J. Biol. Chem. 1974; 249: 1572-1586Abstract Full Text PDF PubMed Google Scholar, 17Zeghouf M. Fontecave M. Coves J. J. Biol. Chem. 2000; 275: 37651-37656Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar). In the ferredoxin-dependent enzymes, the electron donor binds transiently and delivers electrons to the Fe4-S4 cluster, one at a time (18Knaff D.B. Hirasawa M. Biochim. Biophys. Acta. 1991; 1056: 93-125Crossref PubMed Scopus (229) Google Scholar). The electrons are then transferred to the which the The available for this enzyme family is the structure of a truncated form of the hemoprotein subunit of NADPH-dependent sulfite reductase from amino acids at the been B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google Scholar, B.R. Siegel L.M. Getzoff E.D. Biochemistry. PubMed Scopus Google Scholar). structure for a of the ferredoxin-dependent sulfite/nitrite reductases is that the nirA gene from M. tuberculosis encodes Fe4-S4 and sulfite The structure of the a potential target for against human persistent tuberculosis an covalent bond between the side chains of Tyr69 and Cys161 in the of the siroheme and the binding The of this in the active site of NirA have been by site-directed and gene for the NirA (Rv2391) is an the of the for NirA (Rv2391) in the sequence of M. tuberculosis The the first of the sequence as the in the the is as of the NirA The it in a sequence, and a with the for further from M. tuberculosis by using with the and the The and the and the sequence of the by expression using a of expression and several coli strains at in of the siroheme for and of for an active in the biosynthesis of has been shown to the of siroheme-containing Siegel L.M. J. Bacteriol. 1991; PubMed Google Scholar, M. H. Biochim. Biophys. Acta. 2000; PubMed Scopus Google Scholar). The gene from and the expression the of and with all of nirA expression of in to of the of NirA using either a or an B. T. B. L. A. C. M. PubMed Scopus Google the and expression to the enzyme with coli as expression in 2 of at and in phase by and and the of to coli and in a and by and The by The to an acid and the by an on a the to by using a and The and at The of from of determined to M. Biochem. PubMed Scopus Google Scholar). mutagenesis using the the and for and for and for and and for In all the genes to the of and of the enzymes as for and NirA for sulfite reductase using the electron donor which by (15Krueger R.J. Siegel L.M. Biochemistry. 1982; 21: 2892-2904Crossref PubMed Scopus (112) Google Scholar). The and in a of at The by the of and as the of at The of the time to the for and The of of the enzymatic reaction. The at this K. J. Biol. Chem. Full Text PDF PubMed Google Scholar). One of sulfite reductase is as that of enzyme that the of of (15Krueger R.J. Siegel L.M. Biochemistry. 1982; 21: 2892-2904Crossref PubMed Scopus (112) Google Scholar). The sulfide, by the of the of the of the or the of a The by the of and the with a of to the The identified by the of or 2 and for the L.M. Biochem. PubMed Scopus Google to the The for nitrite reductase of the as that nitrite the of Nitrite reductase by or by the of the the at time and the of the nitrite determined at the of and and C. T. in of Scholar). that of the with the of nitrite using sequence H. in Scopus Google Scholar). The by time of on a using acid as and the for up to with the of via and of NirA by the 2 of the in with 2 of and and against of the of at for x-ray structure by of for against the from a nitrogen at at and at The x-ray and with the and from the D. 1994; PubMed Scopus Google Scholar). to the with a and form in the with a In both the contains two with a of The of the are in of and of of unique in are for the from of the to and in the of of of residues in of residues in in are for the of the to and in the in a and structure by using the A. A. J. Scopus Google Scholar), in of CysI from coli B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google with residues and as a The for the NirA a of and an of with two in the of using G. E.J. D. PubMed Scopus Google in a of the by The siroheme molecule and the iron-sulfur cluster from these and the of the by electron for the cofactors at the of the with the J. S. M. A. 1991; PubMed Scopus Google Scholar), on and electron R.J. A. Scopus Google Scholar), and with the G. E.J. D. PubMed Scopus Google Scholar). symmetry in to the number of for a with The structure for the NirA form in a that an NirA from the for the with residues from the This of the P.D. G.M. J. M. R.J. L.M. T. D. PubMed Scopus Google to at the of The models with R.A. J. J. Google in to the of the are in using J.D. 1994; PubMed Scopus Google Scholar). with the S. J. C. A. M. L. 2001; PubMed Scopus Google Scholar), G. J. 2000; Scopus Google Scholar), and the in J. S. M. A. 1991; PubMed Scopus Google Scholar), parameters. using J. Mol. Scopus Google and with D. 1994; PubMed Scopus Google Scholar). The and structure for NirA have been with the with and of NirA from M. tuberculosis has a and shows the of Fe4-S4- and siroheme-containing sulfite/nitrite reductases in the state (15Krueger R.J. Siegel L.M. Biochemistry. 1982; 21: 2892-2904Crossref PubMed Scopus (112) Google Scholar, M. H. Biochim. Biophys. Acta. 2000; PubMed Scopus Google Scholar, S. D.B. Hirasawa M. Biochemistry. 2004; 43: PubMed Scopus Google NirA is active as a sulfite reductase and is able to sulfite to using the electron donor The identified by the and by the L.M. Biochem. PubMed Scopus Google Scholar). The enzyme has a of with sulfite reductases from other B.R. Getzoff E.D. Curr. Opin. Struct. Biol. 1996; 6: 744-756Crossref PubMed Scopus (122) Google Scholar, R.J. Siegel L.M. Biochemistry. 1982; 21: 2892-2904Crossref PubMed Scopus (112) Google Scholar). in the of nitrite in this of the electron donor by in the of of sulfite as a which with nitrite as a The of the enzyme to nitrite these in shows that NirA has a pronounced for sulfite nitrite as of the family are able to both sulfite and at of NirA in the of nitrite or a in the reduction of the as for nitrite reductase S. D.B. Hirasawa M. Biochemistry. 2004; 43: PubMed Scopus Google Scholar, Kamin H. J. Biol. Chem. Full Text PDF PubMed Google Scholar). This that NirA is able to nitrite and and thus able to nitrite at and suggest that NirA is a monomer in a that is also with the The and of the structure of NirA by in two two in the of the the iron-sulfur cluster and the siroheme are in electron The to the electron is in the of for both chains and The and the first residues of NirA are in the electron of the several and The consists of chains comprising residues of two Fe4-S4 two siroheme two and The of the is as for this of the models of NirA from the two an of with the located in the of M. tuberculosis and in the NirA is as a The for the enzyme in both are on the of to of the This is for for a The NirA molecule consists of three α/β the B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google domain and the domain and the domain and the of the the against the of the domain and through a number of and with residues from that The of the domain is built up of two with a ferredoxin-like fold T. C. J. Mol. Biol. PubMed Scopus Google related by a pseudo-2-fold symmetry and consists of an by two on one The two to form the major interface residues the in which form an to the and thus to a of of this domain form a long between and against the and access to the active site, and residues that are in to the siroheme and thus form part of the active site The and domains are very in domain consists of a with the to the One side of the part of the binding site for the or the Fe4-S4 cluster, the other side of the is by four the of the molecule. The fold of these domains is unique to the sulfite/nitrite reductase family. of the structure of NirA is the pseudo-2-fold This also in the structure of CysI and to the that the enzyme has as a of gene B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google Scholar). of NirA can on residues with an of for The pseudo-2-fold symmetry is pronounced in NirA in coli of the CysI structure residues with an of The number of residues is in part to the of the in which have from the of the The nirA gene thus also to have as a of gene with of residues it can at the amino acid sequence the and residues in NirA and that are between the and four are in both enzymes, and These residues are located at the or of and to for Comparison with from the of sequence between NirA and CysI the are very of the two models an of on differences are found at the very of the and are to several either or have in structure is the of the active The of the which been truncated in CysI and in the CysI the in NirA and the access to the active site from the In the fold of the ferredoxin-like of the to the of the active site, and is close to the siroheme In the residues fold in a binds in the at the interface between the three domains of NirA The is in the and two oxygen from one of the are to are interactions of the siroheme with siroheme which can act as donor or are in or residues and form to the of the siroheme and to that are in the enzyme family are In several other residues form with the siroheme and is the bond between the of the siroheme and the of This is also to one of the of the Fe4-S4 cluster. It thus the for the of the two redox cofactors, shown in of sulfite/nitrite reductases using a of as J.A. Siegel L.M. J. Biol. Chem. Full Text PDF PubMed Google Scholar), Siegel L.M. Biochemistry. 1982; 21: PubMed Scopus Google Scholar), Siegel L.M. Biochemistry. PubMed Scopus Google Scholar), and J. Siegel L.M. Biochemistry. PubMed Scopus Google the in electron from the iron-sulfur center to the siroheme B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google Scholar, B.R. Siegel L.M. Getzoff E.D. Biochemistry. PubMed Scopus Google Scholar). Fe4-S4 cluster is by residues from the domain and is located at the side of the siroheme the cluster is in the it is close to the The four are via covalent to the of four and These residues are located on two of the the of the first of the domain and the and a part of the between the and This one of the sequence of the sulfite/nitrite reductase family B.R. Getzoff E.D. Curr. Opin. Struct. Biol. 1996; 6: 744-756Crossref PubMed Scopus (122) Google and the cluster, it from the The side of the is from the and involved in electron from ferredoxin to the Fe4-S4 cluster. The a of the siroheme is from the through a that is by The potential at the of this which the active site, the of the enzyme with the and the of these the of the binding the of the siroheme an to the have this electron as a to the of the and the that and in This site is located at the and other are in CysI B.R. Siegel L.M. Getzoff E.D. PubMed Scopus Google Scholar, B.R. Siegel L.M. Getzoff E.D. Biochemistry. PubMed Scopus Google Scholar). close to the in NirA are and In the residues are involved in binding of the substrate, and both residues also involved in of these with one are in the family of sulfite/nitrite reductases. The of by an has been associated with a in from sulfite to nitrite B.R. Siegel L.M. Getzoff E.D. Biochemistry. PubMed Scopus Google Scholar). in the structure of NirA is the covalent bond between the of Cys161 and the of Tyr69 to this covalent by The of of NirA a of the sequence the in the of the bond a with the of the the covalent This of has been found in the active site of redox C. 1991; PubMed Scopus Google Scholar). The amino acids are located to a redox in this and act as a in the of C. 1991; PubMed Scopus Google Scholar, Chem. 2003; PubMed Scopus Google Scholar). The of a side to the siroheme redox in NirA to further the of this amino acid The of Tyr69 is in close proximity to the site, further a of this in Tyr69 by and Cys161 to and of the and a with the of the covalent The of the covalent bond also in the in of the versus enzyme in as in J. Biol. Chem. 2003; Full Text Full Text PDF PubMed Scopus Google Scholar). The of the four but differences that the of the redox cofactors are by these amino acid The to with the and the The in that the covalent between these two residues in NirA is of for It in that the electron donor has been in the and the in the of the electron The of the further that this proximity of the at the site, participate in The sulfite/nitrite the electrons from the external electron donor via the Fe4-S4 cluster to the siroheme (18Knaff D.B. Hirasawa M. Biochim. Biophys. Acta. 1991; 1056: 93-125Crossref PubMed Scopus (229) Google Scholar), and the for the site is in the of the iron-sulfur cluster. This redox center is located the enzyme with the about from the of the of NirA and its NADPH-dependent CysI shows that several of the major differences between these two enzymes are located in this the in the structure of the external electron In two at the enzyme are The which with in the is the between and is in NirA by residues and It is that the of for these two is also in other bacterial ferredoxin-dependent sulfite/nitrite reductases with that the binding site for the external electron donor is located at the enzyme in the of these two which are also close to the in with the iron-sulfur cluster. This from the the on the enzyme with a potential all from M. tuberculosis are the of residues binding of ferredoxin for electron of have shown that the nirA gene of M. tuberculosis encodes a Fe4-S4- and sulfite It also that NirA is the sequence of M. tuberculosis that is to CysI of This suggests that NirA in the assimilation of sulfur for the biosynthesis of sulfur-containing amino of the of of the genes in coli and M. tuberculosis also this In the genes for the flavoprotein subunit and the hemoprotein component CysI of sulfite reductase are by In M. tuberculosis, NirA is by and gene to the flavoprotein subunit of the NADPH sulfite reductase of coli is found at this J. PubMed Scopus Google of the whole of M. tuberculosis R. J. T. C.M. K. S. K. D. D. T. R. K. T. S. S. T. K. A. J. S. S. J. J. S. K. S. S. R. K. S. PubMed Scopus Google for of the gene from coli in suggesting that in M. tuberculosis, the NADPH-dependent sulfite reductase is by a ferredoxin-dependent in a gene expression of dormant M. tuberculosis, nirA but also several other genes involved in biosynthesis shown to up-regulated T. Soneij S. Bacon J. James B.W. Hinds J. Laing K. Stabler R.A. Marsh P.D. Butcher P.D. Tuberculosis. 2004; 84: 228-238Crossref PubMed Scopus (174) Google Scholar). One of these genes is an enzyme of siroheme further the between NirA and the Nitrite sequence of ferredoxin-dependent reductases suggests that these enzymes can two In one of amino acid Tyr69 and involved in the covalent in common with the NADPH-dependent sulfite residues and with the are In the other these residues are consistently by and that these two sequence to enzymes with a for sulfite and nitrite as substrate, This with available on of of this family all sulfite reductase to the first nitrite reductases the This of genes of the ferredoxin-dependent reductase family as or of M. tuberculosis NirA the first structure of a ferredoxin-dependent sulfite/nitrite the of sequence the of the molecule is in structure to the NADPH-dependent sulfite reductase from The part of the comprising the first amino in structure, form a domain but is part of the first ferredoxin-like of the This of the also to the of the active It access to the of the siroheme and residues the active The covalent bond between the side chains of residues Tyr69 and Cys161 in the active site of NirA but is essential for the enzymatic with electron This from that found in a covalent bond is essential for enzyme function Chem. 2003; PubMed Scopus Google Scholar). suggest that this is for ferredoxin-dependent sulfite reductases as to nitrite reductases. The site is located at a of the close to the iron-sulfur cluster. are to and Coates for of M. tuberculosis and to for the S. access to at and and the at these for