Nitrous oxide reductase

Coordination Chemistry Reviews 257 (2013) 332–349

Contents lists available at SciVerse ScienceDirect

Coordination Chemistry Reviews journal homepage: www.elsevier.com/locate/ccr

Review

Nitrous oxide reductase Sofia R. Pauleta a,∗ , Simone Dell’Acqua a,b , Isabel Moura a,∗∗ a b

REQUIMTE-CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2529-516 Caparica, Portugal Dipartimento di Chimica, Università di Pavia, Via Taramelli 12, 27100 Pavia, Italy

Contents 1. 2.

3. 4. 5.

6.

7.

The environmental relevance of nitrous oxide and the denitrification pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The enzyme nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Primary sequence of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Biogenesis and regulation of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biochemical and spectroscopic properties of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Structure of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enzymatic activity of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Activation of N2 O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Catalytic properties of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Binding of N2 O to the catalytic center of nitrous oxide reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Intermediates in the catalytic cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The electron transfer route . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. The intramolecular electron transfer: CuA–CuZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. The intermolecular electron transfer: electron donor-enzyme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and future perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

a r t i c l e

i n f o

Article history: Received 4 March 2012 Received in revised form 19 May 2012 Accepted 23 May 2012 Available online 31 May 2012 Keywords: Denitrification Nitrous oxide reductase CuZ center CuA center Copper-sulfide center Nitrous oxide

333 334 334 335 337 341 342 342 342 343 344 345 345 346 347 347 347

a b s t r a c t Nitrous oxide is a potent greenhouse gas, whose atmospheric concentration has been increasing since the introduction of the Haber Bosch process led to the widespread use of nitrogenous fertilizers. One of the pathways to its destruction is reduction to molecular nitrogen by the enzyme nitrous oxide reductase found in denitrifying bacteria. This enzyme catalyzes the last step of the denitrification pathway. It has two copper centers, a binuclear CuA center, similar to the one found in cytochrome c oxidase, and the CuZ center, a unique tetranuclear copper center now known to possess either one or two sulfide bridges. Nitrous oxide reductase has been isolated in different forms, depending on the oxidation state and molecular forms of its Cu centers. Recently, the structure of a purple form, which has both centers in the oxidized state, revealed that the CuZ center has the form [Cu4 S2 ]. This review summarizes the biogenesis and regulation of nitrous oxide reductase, and the spectroscopic and kinetic properties of nitrous oxide reductase. The proposed activation and catalytic mechanism, as well as, electron transfer pathways are discussed in the light of the various structures of the CuZ center. © 2012 Published by Elsevier B.V.

Abbreviations: Ac, Achromobacter; B, Bacillus; CD, circular dichroism; Dnr, dissimilative nitrate respiration regulator; EPR, electron paramagnetic resonance; Fnr, fumarate and nitrate reductase regulator; Ma, Marinobacter; MCD, magnetic circular dichroism; N2 OR, nitrous oxide reductase; nir, nitrite reductase; nar, nitrate reductase; Nnr, nitrite and nitric oxide reductases regulator; nor, nitric oxide reductase; nos, nitrous oxide reductase; Pa., Paracoccus; Ps., Pseudomonas; RR, resonance Raman; Sec, secretory pathway; Tat, twin-arginine translocation; Wo., Wolinella. ∗ Corresponding author. Tel.: +351 212 948 300x10967; fax: +351 212 948 550. ∗∗ Corresponding author. Tel.: +351 212 948 300x10916; fax: +351 212 948 550. E-mail addresses: [email protected], sofi[email protected] (S.R. Pauleta), [email protected] (S. Dell’Acqua), [email protected] (I. Moura). 0010-8545/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.ccr.2012.05.026

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

333

1. The environmental relevance of nitrous oxide and the denitrification pathway Nitrous oxide is a potent greenhouse gas that has a global warming potential more than 300 times higher than that of carbon dioxide [1]. Even though it comprises only 0.03% of the total greenhouse gases (carbon dioxide, methane and nitrous oxide) emissions it contributes substantially to global warming [1]. Moreover, in the last 100 years, the atmospheric concentration of nitrous oxide relative to air has increased almost 20%, and is currently 322 ppb [2,3]; on average there has been a yearly increase of 0.25% in its concentration [4]. Nitrous oxide has a long atmospheric lifetime, taking approximately 120 years to decrease the initial emission by 63% [4,5]. Removal from the stratosphere occurs through photolysis [5] followed by reaction with excited oxygen atoms, reactions that are also responsible for the depletion of the ozone layer [2,6]. The emission of nitrous oxide into the atmosphere has several sources, with the majority being produced by microbial metabolism of nitrogen compounds in soils and oceans, and with the human activities also contributing significantly for its emission [3,4]. From these activities, agriculture is the main contributor due to the increased use of fertilizers and application of live-stock manure in crop lands and pasture [5]. Other human activities that have a major impact in nitrous oxide emissions are fuel combustion and, to a lesser extend or not fully accounted for, human sewage, waste water treatment plants, and burning of biomass and biofuels [5]. The emission of nitrous oxide from microbial processes is derived mainly from two metabolic pathways belonging to the nitrogen biogeochemical cycle: nitrification, which is an oxic process and denitrification, an anoxic or near anoxic process [7–9] (Fig. 1). These pathways are carried out mainly by proteobacteria, though methanotrophic bacteria and fungi also contribute to the release of nitrous oxide using metabolic pathways still only poorly explored [10–15]. One of the bacterial pathways, denitrification, that leads to the release of nitrous oxide to the atmosphere involves the reduction of the nitric oxide intermediate [15]. In the case of nitrification (an autotrophic pathway involved in the aerobic oxidation of ammonia to nitrate, Fig. 1) nitric oxide is produced by the aerobic oxidation of hydroxylamine [17,18], which is favored under low nitrite and high ammonia concentrations [19]. Also, under oxygen-limiting conditions or under high nitrite concentrations, ammonium oxidizing bacteria can reduce nitrite to nitrous oxide in combination with ammonia oxidation (nitrifier denitrification, Fig. 1) [20–22] (Fig. 1). However, under anaerobic conditions, denitrifying organisms perform the reduction of inorganic nitrate or nitrite in sequential steps that involve the abstraction of an oxygen atom at each step, with the production of gaseous molecules at intermediate stages: NO3 − → NO2 − → NO → N2 O → N2 . Each reaction is catalyzed in a concerted way by a different enzyme: nitrate reductase (Nar), nitrite reductase (Nir), nitric oxide reductase (Nor) and nitrous oxide reductase (N2 OR), either located in the internal membrane or in the periplasm of several proteobacteria (alpha, beta, gamma and epsilon division). In archaea a similar pathway has been proposed, though the four reductases are membrane bound and dependent on menaquinol as the electron donor [23]. The denitrification pathway allows the bacteria to survive under anaerobic conditions, since there is the generation of an electrochemical gradient across the cytoplasmic membrane, and thus leads to energy conservation [24]. Recently, it has been shown that Gram-positive bacteria, such as some Bacillus (B.) strains, are also denitrifying organisms able to reduce nitrate or nitrite with the production of nitrous oxide or molecular nitrogen [25,26]. However, the molecular systems involved in this reduction have only been partially identified at the protein level for B. azotoformans [27]. The reasons for the

Fig. 1. Biological pathways involved in the nitrogen cycle: denitrification (blue), nitrification (red), anaerobic ammonium oxidation (Anammox) (green), nitrogen fixation (brown) and dissimilatory nitrite reduction to ammonium (DNRA) (black), and dissimilatory nitrite reduction pathway of aerobic nitrifier denitrification (gray). The gene encoding the catalytic subunit of the enzyme(s) that catalyses each reaction is indicated on top of each arrow: different nitrate reductases (nas—nitrate reductase cytoplasmic prokaryotic-assimilatory, euk-nr—nitrate reductase cytoplasmic eukaryotic-assimilatory, narG—nitrate reductase membrane bound dissimilatory, napA—periplasmic nitrate reductase dissimilatory), nitrite reductases (nirK, nirS), nitric oxide reductase (cnorB, qnorB), N2 OR (nosZ), nitrogenase (nif), hydrazine hydrolase (hh), hydrazine oxidoreductase (hzo), dissimilatory nitrite reductase (nrf), ammonium monooxygenase (amo), hydroxylamine oxidoreductase (hao) and nitrite oxidoreductase (nxr). Figure adapted from [16].

denitrification pathways being overlooked in these organisms has been attributed to the low number of completely sequenced Grampositive genomes identified as denitrifying organisms, the low DNA sequence homology of the putative proteins with the Gramnegative counterparts and also the possibility of new unknown genes being involved in this pathway [28]. Surprisingly, in the case of N2 OR, a gene that shares around 35% sequence homology with those found in proteobacteria, coding for N2 OR has been identified [29] (see Section 2.2). On the other hand, in Gram-negative bacteria, these molecular systems have been extensively studied for many years. The different genes that code for the enzymes and electron shuttle proteins involved in denitrification have been identified and those required for its assembly are for the most part known (Table 1). However, their catalytic mechanism, metal cluster assembly, and regulatory systems are not yet completely unraveled. Recently, one more step toward a better understanding of the denitrification pathway has been achieved with the threedimensional structure determination of nitric oxide reductase from Pseudomonas (Ps.) aeruginosa (a Gram-negative bacteria), a c-type nitric oxide reductase (cNor) [30] and a quinol-dependent nitric oxide reductase from Geobacillus stearothermophilus (a Grampositive bacteria) [31]. Each step of the denitrification pathway can be catalyzed by more than one type of enzyme (Table 1). In the case of nitrite reductase, two different enzymes have been isolated, both with two redox centers in two separate structural domains, one being the electron transfer center and the other the catalytic center, but

334

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Table 1 Genes that code for the catalytic subunits of the enzymes involved in the denitrification pathway. The nas and euk-nr genes are not mentioned in this table, since they are part of an assimilatory pathway. Reaction

Enzyme

Gene encoding the catalytic subunit

Ref.

NO3 − + 2H+ + 2e− → NO2 − + H2 O

Membrane nitrate reductase (Nar) Periplasmic nitrate reductase (Nap)

narG napA

[40] [41]

NO2 − + 2H+ + e− → NO + H2 O

Nitrite reductase cytochrome cd1 (cNir) Copper nitrite reductase (CuNir) Three domain T1Cu CuNira Three domain cytc CuNirb

nirS nirK Cu nirK Cyt nirK

[42] [43] [44] [45]

2NO + 2H+ + 2e− → N2 O + H2 O

Cytochrome c nitric oxide reductase (cNor) Quinol nitric oxide reductase (qNor) Quinol CuA nitric oxide reductase (qCuA Nor)

cnorB qnorB (norZ) Not identified

[46] [47,48] [49]

N2 O + 2H+ + 2e− → N2 + H2 O

Nitrous oxide reductase Cytochrome c nitrous oxide reductase

nosZ cnosZ

[50] [51]

a b

This is a hexameric nitrite reductase isolated from Hyphomicrobium denitrificans, that contains an additional N-terminal cupredoxin domain. The gene coding for this enzyme has been identified at the genome level and is proposed to present an additional C-terminal class I c-type cytochrome domain.

differing in the transition metal present in those centers. In the copper nitrite reductase, CuNir, there are two copper centers, a type 1 copper center, and a type 2 copper center as catalytic center, while in cytochrome cd1 nitrite reductase, cNir, both redox centers are heme iron centers, a c-type heme, as electron transferring heme, and a d-type heme as the catalytic center. These two enzymes do not co-exist in a given strain and are distributed differently [32,33]. cNir has been mostly found in Pseudomonads, while CuNir has been identified in many different physiological groups, though globally the number of isolated organisms containing the gene coding for cNir (nirS) is higher [29,34]. Nevertheless, this observation may only reflect the methodologies and sources available for the isolation of denitrifying organisms [35] and this distribution may also change in time with the increasing number of complete sequenced genomes. Similarly, homologous enzymes that catalyze other steps of the denitrification pathway have different electron transfer domains and cofactors but unlike nitrite reductase, the other enzymes share a similar catalytic center, Mo-bis-molybdopterin guanine dinucleotide cofactor in nitrate reductases [36], heme b3 – FeB in nitric oxide reductases [37,38], and the “CuZ center” in N2 OR (Table 1), the focus of this review, though, they share the presence of two domains: an electron transfer domain and a catalytic domain [39]. 2. The enzyme nitrous oxide reductase The reduction of nitrous oxide to molecular nitrogen requires two protons and two electrons, according to Eq. (1) [35]: N2 O + 2H+ + 2e− → N2 + H2 O, E◦ (pH7.0) = 1.35 V; G0 = −339.5 kJ/mol

(1)

This is a challenging reaction to be catalyzed by a metalloenzyme, as nitrous oxide is not just kinetically inert to decomposition but also a poor transition metal ligand, due to its weak ␴-donating and ␲-accepting properties [52,53]. Although, there have been some reports of other enzymes, such as nitrogenase [54,55] and a multicopper oxidase from Pyrobaculum aerophilum [56], able to catalyze this reaction in vitro, the only one, whose primary role is the reduction of nitrous oxide, is N2 OR. This enzyme has been isolated and biochemically characterized from different classes of Proteobacteria that carry out the complete denitrification pathway and is the focus of this review. The importance of this reaction has led not just to the detailed biochemical, spectroscopic and structural characterization of N2 OR itself but also to the discovery of other enzymes and proteins that assist in its biological function and assembly. The genes that code for these proteins are organized in an operon, conserved in most of

the organisms that have the nosZ gene, which codes for N2 OR, as will be briefly discussed in Section 2.2. N2 OR encoded by nosZ is a copper-containing enzyme with two domains, an electron transferring domain that binds a binuclear copper center, the CuA center, similar to that of cytochrome c oxidase, and a catalytic domain which binds a tetranuclear copper-sulfide center, unique in biology, named “CuZ center” (see Section 4). 2.1. Primary sequence of nitrous oxide reductase The alignment of the primary sequence of N2 ORs of known structure highlights the presence of several conserved residues. Most of these residues are involved in the coordination of the two redox centers, the CuA and the “CuZ center”, or located at the surface and proposed to be involved in the electron transfer complex with the redox partner, or in the putative electron transfer route from the CuA center, electron-transferring center, to the catalytic “CuZ center” [57] (Fig. 2). The analysis of these sequences also reveals the presence of a N-terminus signal peptide, comprising around 35–40 residues and containing the consensus SRRXF/L “twin-arginine” motif, recognized by the twin-arginine translocation (Tat) system [58,59], which is responsible for the translocation of folded proteins, across the cytoplasmic membrane, to the periplasmic space. The seven conserved histidine residues, found in the N-terminal region are the residues that coordinate the “CuZ center”, while in the C-terminal region there are two cysteine, two histidine, a methionine and a tryptophan residues, which are conserved and involved in the coordination of the CuA center (see Section 4). Other conserved residues are also highlighted and will be discussed at a later stage, as being involved in the putative electron transfer pathway from small electron carriers to the “CuZ center”, passing through the CuA center as an intermediate center (see Section 6). A search for homologous sequences in all the sequenced genomes deposited in the database (BioProject, NCBI) highlights the presence of two types of N2 OR. The so-called simple Z-type N2 OR, which is the focus of this review and encoded by the nos gene, and another type, that has an additional C-terminus domain (of around 200 residues), with the canonical c-type heme binding motif –CXXCH– (Fig. 2). This three-domain Z-type N2 OR sequences have been detected only in a few organisms from the Campylobacter, Sulfurimonas, Wolinella (Wo.) and Denitrovibrio genera, with the first three being host-associated organisms from the ␧-proteobacteria group. The primary sequence analysis of N2 OR from these organisms has a signal peptide with Sec (secretory pathway) characteristics. This, difference can be attributed to the

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Ma.h Ps.s Ac.c Pa.d Wo.s Ma.h Ps.s Ac.c Pa.d Wo.s Ma.h Ps.s Ac.c Pa.d Wo.s Ma.h Ps.s Ac.c Pa.d Wo.s Ma.h Ps.s Ac.c Pa.d Wo.s Ma.h Ps.s Ac.c Pa.d Wo.s Wo.s Wo.s

335

1 20 40 60 80 100 120 | | | | | | | | | | | | | MKKRDDLTKDTPEVSEG-GLSRRRFMGAAA----LAGVAG--------ATGLGTSVMSRETWAAAAEEARNKAHVAPGELDEYYGFWSGGHQGEVRVLGVPSMRELMRIPVFNVDSATGW 107 MSDKD--SKNTPQVPEKLGLSRRGFLGASA----VTGAAVA-------ATALGGAVMTRESWAQAVKESKQKIHVGPGELDDYYGFWSGGHQGEVRVLGVPSMRELMRIPVFNVDSATGW 107 MESKE----------HK-GLSRRALFSATAGSAILAGTVG------PAALSLGAAGLATPARA----ATGADGSVAPGKLDDYYGFWSSGQTGEMRILGIPSMRELMRVPVFNRCSATGW 99 MESKQ----------EK-GLSRRALLGATAGGAAVAGAFGGRLALGPAALGLGTAGVATVAGSGAALAASADGSVAPGQLDDYYGFWSSGQSGEMRILGIPSMRELMRVPVFNRCSATGW 109 : : ***** ::.*:* ::*. * .** : :: : : : . *.**:**:******.*: **:*:**:********:**** ***** ----------------MQRLLKQSLVVTASLLALGTASLAS---------SDLQTIMKERKLTEKDVLAAAKTYQPSGRKDEFVVFSSGGQSGQILVYGVPSMRIYKYIGVFTPEPWQGY 95 * :: :. ::: :. . : : : : . .*. *:: * *.*: *:: : *:**** : **. . *: GITNESKEILG--------------GDQQYLNGDCHHPHISMTDGRYDGKYLFINDKANTRVARIRLDIMKTDKITHIPNVQAIHGLRLQKVPKTNYVFCNAEFVIPQPNDGTDFSLD-GLTNESRHIMG--------------DSAKFLNGDCHHPHISMTDGKYDGKYLFINDKANSRVARIRLDIMKCDKMITVPNVQAIHGLRLQKVPHTKYVFANAEFIIPHPNDGKVFDLQDE GQTNESIRIHQRTMTEKTKKQLAANGKKIHDNGDLHHVHMSFTDGKYDGRYLFMNDKANTRVARVRCDVMKTDAILEIPNAKGIHGMRPQKWPRSNYVFCNGEDEAPLVNDGSTMTDVAGQTNESVRIHERTMSERTKKFLAANGKRIHDNGDLHHVHMSFTEGKYDGRFLFMNDKANTRVARVRCDVMKCDAILEIPNAKGIHGLRPQKWPRSNYVFCNGEDETPLVNDGTNMEDVA* **** .* .. . *** ** *:*:*:*:***::**:*****:****:* *:** * : :**.:.***:* ** *:::***.*.* * ***. : GFDDDSKKVLRQ----------GDIRGREINWGDTHHPNFTEKNGEYVGDYLFINDKANPRIAVVNLHDFETTQIVVNPIMKSEHGG-SFVTPNTEYVIEASQYAAPLDHQYHPIEEYEA ** ** ::: .:*.* * :**:*****.*:* :. . :: : * :. ** *.::**: .: * :: : * ::* .:

213 213 219 229

NSYTMFTAIDAETMDVAWQVIVDG---------NLDNTDADYTGKYATSTCYNSERAVD---------LAGTMRNDRDWVVVFNVERIAAAVKAG-NFKTIGDSKVPVVDGRG----ESE NSYTMYNAIDAETMEMAFQVIVDG---------NLDNTDADYTGRFAAATCYNSEKAFD---------LGGMMRNERDWVVVFDIHAVEAAVKAG-DFITLGDSKTPVLDGRKKDGKDSK TYVNIFTAVDADKWEVAWQVKVSG---------NLDNCDADYEGKWAFSTSYNSEMGMT---------LEEMTKSEMDHVVVFNIAEIEKAIKAG-QYEEING--VKVVDGRKE--AKSL NYVNVFTAVDADKWEVAWQVLVSG---------NLDNCDADYEGKWAFSTSYNSEKGMT---------LPEMTAAEMDHIVVFNIAEIEKAIAAG-DYQELNG--VKVVDGRKE--ASSL **** **** *::* :*.**** .. * : * :***:: : *: ** :: :.. . *:*** .* . .::.*:**:. ::*:** *.* VFRGAVTLWKFDYAKGKIDEKASFSLEFPPYMQDLSDAGKGESFGWAFTNSFNSEMYTGGIEKGLPPFEAGMSRNDTDYMHVYNWQMLEKLAQDPKNYKIYHGHRVISIEAAVK----AG . . : . : .. :*.: . . :* :..:*** : * : *:: : :: . . ::. :

308 314 315 325

FTRYIPVPKNPHGLNTSPDGKYFIANGKLSPTVSVIAIDKLDDLFEDKIE------------LRDTIVAEPELGLGPLHTTFDGR-GNAYTTLFIDSQVCKWNIADAIKHYNGDRVNYIR FTRYVPVPKNPHGCNTSSDGKYFIAAGKLSPTCSMIAIDKLPDLFAGKLAD-----------PRDVIVGEPELGLGPLHTTFDGR-GNAYTTLFIDSQVVKWNMEEAVRAYKGEKVNYIK FTRYIPIANNPHGCNMAPDRKHLCVAGKLSPTVTVLDVTKFDALFYDNAE------------PRSAVVAEPELGLGPLHTAFDGR-GNAYTSLFLDSQVVKWNIDEAIRAYAGEKINPIK FTRYIPIANNPHGCNMAPDKKHLCVAGKLSPTVTVLDVTRFDAVFYENAD------------PRSAVVAEPELGLGPLHTAFDGR-GNAYTSLFLDSQVVKWNIEDAIRAYAGEKVDPIK *..:*.***********:**** *****:**:**** ***: :*:: * *:::: *: ****:*:.:**** * :.* *:: . ****** ::: : :: :* : ALFLIPEPKSPHGVDVSPDGRYIVVGGKLDTHASVYDFRKIKQLIDKKEFIGADPYGIPILDMKKTLHGQVELGLGPLHHTYDAQDGIIYTSLYVDSQIVKWDYKNLK----------VL :* .:.*** : :.* ::: . ***.. :: . :: :: : :..: .: ******** ::*.: * **:*::***: **: : :

415 422 422 432

QKLDVQYQPGHNHASLTESRDADGKWLVVLSKFSKDRFLPVGPLHPENDQLIDISGEEMKLVHDGPT-YAEPHDCILVRRDQIK-TKKIYERNDPYFASCRAQAEKDGVTLES-DNKVIR QKLDVHYQPGHLHASLCETNEADGKWLVALSKFSKDRFLPVGPLHPENDQLIDISGDEMKLVHDGPT-FAEPHDCIMARRDQIK-TKKIWDRNDPFFAPTVEMAKKDGINLDT-DNKVIR DKLDVQYQPGHLKTVMGETLDAANDWLVCLCKFSKDRFLNVGPLKPENDQLIDISGDKMVLVHDGPT-FAEPHDAIAVSPSILPNIRSVWDRNDPLWAETRKQAEADEVDIDEWTEAVIR DKLDVHYQPGHLKTVMGETLDATNDWLVCLSKFSKDRFLNVGPLKPENDQLIDISGDKMVLVHDGPT-FAEPHDAIAVHPSILSDIKSVWDRNDPMWAETRAQAEADGVDIDNWTEEVIR :.:::**** :* *: * : :: : *** :****:***** :: : *: :* ..*** *.******** ****:***********::* ******* :*****.* . . : DRVNVHYNIGHLDSMEGKSAKPKGKYALALDKLSIDRFNPVGPLHPQNHQLIDIGGPKMELIYDLPIPLGEPHDVISIAADKLK-----PQVTYPMGTNSRTGKQHEAMTLAG-QERVER ::::*:*: ** .: :: .. ..: : * *:* *** ****:*:*.*****.*: * *::* * .**** * . : : . * : : : : : : * *

532 539 541 551

DGNKVRVYMTSVAPQYGMTDFKVKEGDEVTVYITNLDMVEDVTHGFCMVNHGVSMEISPQQTASVTFTAGKPGVYWYYCNWFCHALHMEMVGRMLVEAA--------------------DGNKVRVYMTSMAPAFGVQEFTVKQGDEVTVTITNIDQIEDVSHGFVVVNHGVSMEISPQQTSSITFVADKPGLHWYYCSWFCHALHMEMVGRMMVEPA--------------------DGNKVRVYMTSVAPSFSQPSFTVKEGDEVTVIVTNLDEIDDLTHGFTMGNHGVAMEVGPQQTSSVTFVAANPGVYWYYCQWFCHALHMEMRGRMFVEPKGA------------------DGNKVRVYMSSVAPSFSIESFTVKEGDEVTVIVTNLDEIDDLTHGFTMGNYGVAMEIGPQMTSSVTFVAANPGVYWYYCQWFCHALHMEMRGRMLVEPKEA------------------*********:*:** :. .*.**:****** :**:* ::*::*** : *:**:**:.** *:*:**.* :**::****.********** ***:**. KGNEVKIYGTLIRSHINPEHVTVNKGDKVTFYLTNLERAQDETHGFAVSGYNVHASVEPGKTVAVTFTADEEGVFPYYCTEFCSALHLEMMGYLYVKDPKKKYESVKELKLQKMSKEQLE .**:*::* : : . . ..*::**:**. :**:: :* :*** : .:.* .: * * ::**.* : *:. *** ** ***:** * : *:

631 638 642 652

204

320

430

544

664

SEYKKVIATNKATDDVIQSVVKFLKDKNYAKYPKVKSLVEDALDQYGKIGEVKAKADESYKKGDVNGAILWEYQVWQYMVKTADVGLRAKNNLAKELATPMKPAAQKGEEAYLKGGCNGC 784 HVIGQVSSGPDLTGVLSRHENAEKWVFDFIKNPASKYEEDYVKTMINYFNLRMPNQHMNDQEIKDIIEYLKWIDENAGLF---------------------------------------- 864

Fig. 2. Primary sequence alignment of N2 ORs for which the three-dimensional structure is known with the primary sequence of the enzyme isolated from the ␧-proteobacteria Wolinella succinogenes. In violet and pink are highlighted the residues that coordinate “CuZ” and CuA center respectively. The c-type heme binding motif is in red, and the tat-motif is underlined. In red are highlighted residues proposed to be involved in electron transfer. Ma.h—Ma. hydrocarbonoclasticus, Ps.s—Ps. stutzeri, Ac.c—Ac. cycloclastes, Pa.d—Pa. denitrificans, Wo.s—Wo. succinogenes. Asterisks, colons or stops below the sequence indicate identity, high conservation or conservation of the amino acids, respectively. Mature Ma. N2 OR starts at A52 (identified with a square).

requirement of the periplasmic cytochrome c maturation system for its biosynthesis [60]. In addition, with the exception of the tryptophan residue, located in between the two cysteine residues that coordinates the CuA center, and that is absent in the sequence of N2 OR from ␧-proteobacteria, the other ligands of the two copper centers are conserved (Fig. 2). 2.2. Biogenesis and regulation of nitrous oxide reductase The gene cluster containing nosZ codes for several accessory proteins required for the transcription and assembly of the copper centers of N2 OR, specifically in the “CuZ center” assembly. Although, most of the studies on the biogenesis of the “CuZ center” have been performed in Ps. stutzeri, the gene organization of the gene cluster that includes nosZ, from the different sub-divisions of proteobacteria and other bacteria, share several common aspects (Fig. 3). These genes are usually located within a single locus, with the gene arrangement in the cluster being nosRZDFYL for the majority of the organisms, and in some denitrifying bacteria, mainly from the ␣-proteobacteria group, followed or preceded by nosX. The location of nosR is also variable as it can be found after nosZ or at the end of the gene cluster (Fig. 3), and in some cases there

is more than one copy of the accessory genes. Another feature is the presence of small electron carrier proteins, either c-type cytochromes or type 1 copper proteins, in the vicinity of this locus, that might function as electron donor proteins to N2 OR in those organisms. Besides the diversity in gene organization of the nosZ locus, the number of transcriptional units transcribed from the N2 OR gene cluster differ between organisms. In the case of Ps. aeruginosa there is only one hexa-cistronic transcriptional unit [61], while in the other systems studied, there are three variable transcriptional units. In fact, in Ps. stutzeri, three transcripts were identified, nosR, nosZ and nosDFYLtatE [62–64], and a similar organization has been proposed for Sinorhizobium meliloti [65], and Paracoccus (Pa.) denitrificans, though in the later case comprising nosCR, nosZ and nosDFYLX genes [66,67] (Fig. 3). The regulation of nos genes has not been extensively characterized in any organism, although, it has been shown that expression of nos genes is only weakly induced by N2 O [61]. Although, the promoter of nosRZDY is strongly up-regulated by NO [61]. In fact, the promoter region upstream of the nos genes has a sequence similar to the consensus Fnr-binding motif (usually named FNR box, TTGAT-N4 -ATCAA) [68], which indicates that the nos genes are under the control of a Fnr-like regulator [61,69,70].

336

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Fig. 3. Organization of the nos gene cluster from different proteobacteria divisions and also two Gram-positive bacteria identified as complete denitrifiers. The identified transcriptional units and promoter regions are identified as arrows and dots, respectively. PA—pseudoazurin, Az—azurin, C—protein containing c-type heme binding motif –CXXCH or thioredoxin-like protein, C*—similar size protein to C, but with no c-type heme binding motif, identified as a putative thioredoxin. The arrows in black correspond to hypothetical proteins, dnr—dissimilative nitrate respiration regulator, tat—twin-arginine translocation.

In Ps. aeruginosa, the transcription factor Dnr (dissimilative nitrate respiration regulator) controls the expression of the nos genes, in a NO-dependent way [61]. The specific binding of NO to the non-covalently bound heme, in the effector domain of Dnr, is proposed to induce a conformational change in the protein [71–73], increasing its affinity to the promoter region and activating the transcription of the nos genes [71,74]. In Pa. denitrificans and Ps. stutzeri a similar role has been attributed to the Dnr homologue Nnr (nitrite and nitric oxide reductases regulator) and DnrD, respectively [69,75], though recently it has been shown that in Pa. denitrificans FnrP, an oxygen sensor, also regulates the transcription of N2 OR [69]. The nitric oxide (the substrate of nitric oxide reductase, the preceding denitrifying enzyme) dependence observed in the transcription activation of the nos genes [70], with nitrous

oxide being only a weak inducer, leads to the synthesis of N2 OR prior to the accumulation of N2 O, and has been proposed to be crucial to attain a higher energy conservation (ATP production). In Ps. stutzeri, the transcription of nosZ is dependent on nosR [38,62,70], which is also involved in the transcription of nosDFYLtatE operon [63]. The gene coding for this protein is conserved in all the organisms with exception of ␧-proteobacteria and Grampositive bacteria. NosR isolated from Ps. stutzeri is a multidomain membrane-bound iron-sulfur flavoprotein, confirming the bioinformatic analysis of its primary sequence [38]. NosR has a large periplasmic domain that binds covalently a flavin cofactor to the conserved T163 residue, and a cytoplasmic domain with a polyferredoxin signature, which is proposed to bind two [4Fe–4S] centers [38].

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

These features are not of a transcription regulator, though NosR is essential for the expression of nosZ, whose function might be exerted through the interaction with the transcription activator DnrD [38]. Moreover, NosR is essential for full activity of N2 OR in vivo and for the correct insertion of the CuA and the “CuZ” centers. In fact, it is possible that NosR is involved in different aspects of the biosynthesis and activity of N2 OR. The flavin domain is essential for whole-cell N2 O-reduction [38], presenting a putative role as electron/proton donor (or electron sink) from the quinol pool. However, NosR might also be involved in maintaining N2 OR in a redox active state, since the enzyme isolated from a background in which NosR is mutated in one intermembrane CX3 CP motif has “CuZ center” as CuZ* in the [1Cu2+ –3Cu+ ] redox state (redox inert1 and catalytically inactive, see Sections 3 and 5), but the organism still presents whole-cell N2 O reduction activity [76]. It is proposed that the intermembrane CX3 CP motif is involved in reversible metal binding or –SH redox chemistry which might properly orient NosR transmembrane helixes [76]. Moreover, in a background in which NosR does not present the cytoplasmic iron-sulfur domain, the whole-cell N2 O reduction activity is lower than in that of the previously mention Ps. stutzeri NosR mutant [38]. Though, in this case the isolated N2 OR has the “CuZ center” as CuZ (redox active, as it can be 1-electron reduced and then 1-electron oxidized) [38]. The other group of conserved genes, nosDFYL, have been proposed to be involved in the assembly of the “CuZ center”, which together with the CuA center are assembled in the periplasm after translocation of the folded polypeptide through the Tat system, as mentioned before [77]. In contrast to N2 OR, NosD and NosL are transported to the periplasm by the Sec system, while NosF and NosY are located in the cytoplasm and in the cytoplasmic membrane [78]. NosDFY are proposed to form a ABC transporter involved in the assembly of the “CuZ center”, as in the absence of these genes, the enzyme is isolated only with the CuA center [78–80]. Therefore, since copper insertion does not seem to depend on these genes, it has been proposed that this system is involved in sulfur transport and insertion into N2 OR to assemble the “CuZ center” [67]. The ATPase activity of NosF has been observed, which could couple hydrolysis of ATP/GTP with energy-dependent transfer of sulfur or other small molecules across the cytoplasmic membrane pore, NosY [63]. This other component, a transport protein, is a 30 kDa five-span membrane protein, that is propose to contact with NosD, located in the periplasmic space. NosD is a 45 kDa protein of unknown function, with a predicted ␤-helical domain, base on the sequence homology with carbohydrate binding and sugar hydrolase enzymes [67]. Although, not essential for the “CuZ center” assembly, nosL is in most organisms co-transcribed with nosDFY, pointing out its involvement in N2 OR biogenesis. NosL is a 20 kDa lipoprotein located in the outer-membrane, that binds specifically one Cu+ and is proposed to be a copper chaperone [81,82]. Another common nos gene in the genome of ␣- and some ␤-proteobacteria is nosX (Fig. 3). This gene encodes a 34 kDa periplasmic flavoprotein that is proposed to be involved similarly to NosR, not in the “CuZ center” assembly itself, but probably in maintaining a fully functional N2 OR. Pa. denitrificans, besides nosX, has a homologue nirX (as part of nirSIX operon) [83], thus in order to study the effect of this gene in the spectroscopic properties of the isolated N2 OR, it was used a nosXnirX double mutant [84]. In N2 OR isolated from the double mutant background, the “CuZ center” is present as redox inert, CuZ*. Whole cells have no N2 O-reducing

1 CuZ* is classified as redox inert due to the fact that it cannot be oxidized nor reduced with the usual reducing or oxidizing agents, though in the presence of reduced methyl viologen it can be further reduced on prolonged incubation.

337

activity despite the in vitro specific activity of the isolated enzyme being of the same magnitude as the wild-type [84]. Therefore, NosX, a member of the ApbE protein family (involved in iron-sulfur synthesis) [85] and binding a FAD, might be involved in electron transfer to N2 OR, and maintaining the “CuZ center” in a catalytically active state. Since reductive activation is essential to attain in vitro high catalytic activity (see Section 5), this might be accomplished in the periplasm with the assistance of NosR and/or NosX. In ␧-proteobacteria, the locus coding for N2 OR has additional genes, two coding for c-type cytochromes, and two other, nosH and nosG (homologous to napHG quinol dehydrogenase [86]), that together could constitute the electron transport system from menaquinol to the periplasmic N2 OR [51]. The primary sequence of NosH and NosG reveals the presence of conserved cysteine motifs that bind [4Fe–4S] centers, four in the case of NosG, and two in the case of the membrane bound NosH [51]. These group of genes might have a similar function to NosR, as using the quinol pool to transfer electrons either directly or indirectly to N2 OR through the two c-type cytochromes that are also encoded in this gene cluster [51]. Since some of these proteins are proposed to be involved in the assembly of the copper centers into N2 OR, process that occurs in the periplasm, the presence of a Tat signal peptide, indicates that this enzyme is an exception for Tat-transported proteins, as it is transported across the membrane folded but in the apo-form [58,59]. 3. Biochemical and spectroscopic properties of nitrous oxide reductase In the previous section, two families of N2 OR were described: the Z-type enzyme (encoded by nosZ gene) and the enzyme from Wo. succinogens that has an additional c-type heme containing domain [87–90]. The presence of the c-type heme hampers the observation of the typical absorption features of the Z-type N2 OR domain. Homologous genes to the latter enzyme are present in other ␧-proteobacteria [57]. However, most of the biochemical and structural characterization has been directed to the Z-type enzyme, and is the focus of this review. N2 OR was first isolated in 1972 from Alcaligenes faecalis as being a new type of copper binding protein, but neither its function nor enzymatic activity was established [91]. Ten years later, a protein with similar visible spectrum was isolated from Ps. stutzeri (classified at the time as Ps. perfectomarinus) and its N2 OR activity identified [92]. Since then, this enzyme has been isolated from different denitrifying bacteria, such as Achromobacter (Ac.) xyloxidands [93], Pa. denitrificans [94], Ac. cycloclastes [95], Ps. aeruginosa [96], Pa. pantotrophus [97], Thiobacillus denitrificans [98], Marinobacter (Ma.) hydrocarbonoclasticus (previously named as Ps. nautica) [99], and Hyphomicrobium denitrificans [100]. As described in the previous section, depending on the genomic background and on the purification procedure used, different redox forms have been identified. These redox forms were named according to the color of the isolated protein and reflect the redox state of the CuA and “CuZ” centers (Table 2).2 However, this classification can be misleading

2 Please note that whenever the authors refer to the CuZ center itself, regardless of the redox state or atomic composition, it will be referred to as the “CuZ center”, while CuZ and CuZ* will be used to differentiate two forms of the center: CuZ is redox active while CuZ* is redox inert. The recent structure of Ps. stutzeri N2 OR shows the presence of a second sulfur in CuZ. However, the differences between CuZ and CuZ* are still an object of debate and might be solved in the future by more detailed spectroscopic analysis and reactivity comparison (see Section 4).

338

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Table 2 The different redox forms of nitrous oxide reductase. CuA center

CuZ center

Fully oxidized/As-isolated (purple form) Ascorbate reduced Dithionite reduced

[Cu1.5+ –Cu1.5+ ] [Cu1+ –Cu1+ ] [Cu1+ –Cu1+ ]

[2Cu2+ –2Cu+ ] [2Cu2+ –2Cu+ ] [1Cu2+ –3Cu+ ]

CuZ* center

Fully oxidized/As-isolated (purple/pink forma ) Dithionite reduced/ascorbate reduced (blue formb )

[Cu1.5+ –Cu1.5+ ] [Cu1+ –Cu1+ ]

[1Cu2+ –3Cu+ ] [1Cu2+ –3Cu+ ]

Fully reduced/activated formc

[Cu1+ –Cu1+ ]

[4Cu+ S]

a

Some authors have named this form also as purple form. In the “blue form” of nitrous oxide reductase, “CuZ center” is found mainly in the CuZ* state, which has also been named “resting CuZ”, a redox inert state [101]. The two dithionite-reduced forms of nitrous oxide reductase are different, although both copper centers have the same redox state. c This form has only been shown to be formed from CuZ*. b

as it is a subjective classification, and in this review we chose to always define the redox state of each copper center. In addition to multifrequency EPR and visible spectroscopy, other spectroscopic techniques, such as resonance Raman, MCD, circular dichroism (CD) and extended X-ray absorption fine structure (EXAFS) have been important in the identification and characterization of the different redox states of the “CuZ center”, as well as, of the CuA center [80,102–108]. The first purifications of N2 OR were performed in the presence of oxygen, and the analysis of the visible spectra of the isolated enzymes, indicated that the CuA center was oxidized and the “CuZ center” in what is now known to correspond to the redox inert [1Cu2+ –3Cu+ ] state (termed CuZ*). In fact, the CuA center is a binuclear mixed-valence copper center similar to the one found in cytochrome c oxidase, and thus shares several of its spectroscopic features [109]. This center contributes to the visible spectrum with absorption bands at 480 nm, 540 nm and 800 nm, while the CuZ* center has an absorption band at 640 nm (Fig. 4A, ii). This form of the enzyme with the CuA center in the oxidized state and the “CuZ center” as CuZ* has been named by different authors as either “pink” or “pink/purple”. In a few cases a “blue form” has also been isolated, in which the CuA center is reduced and the “CuZ center” is CuZ*, thus the visible spectrum of this redox state is characterized by a single strong absorption band at 640 nm (Fig. 4A, iii). The isolation of the enzyme in the absence of oxygen led to the identification of another redox state of the enzyme, the “purple form”, in which the “CuZ center” is in the [2Cu2+ –2Cu+ ] redox state, which is redox active (but not catalytically competent in vitro, Table 3, Section 5) [79,94,96,103], with a midpoint potential of 60 mV, at pH 7.5 [103]. The visible spectrum of N2 OR in this redox state is the sum of the contributions of the two copper centers: the oxidized CuA center, with absorption bands at 480 nm, 540 nm and 800 nm, and the “CuZ center” with absorption bands at 550 nm and a shoulder at 635 nm (Fig. 4A, i). The features of the “CuZ center” alone are only observed when the CuA center is reduced with an equimolar amount of sodium ascorbate (Fig. 4A, dashed line). The X-band electron paramagnetic resonance (EPR) spectrum of this form of the enzyme is characterized by an axial signal with g|| = 2.18 and g⊥ ∼ 2.03 and by the presence of 7-line hyperfine coupling (A|| = 38 G) from the contribution of the CuA center, as the “CuZ center” in the [2Cu2+ –2Cu+ ] state, was shown, by magnetic circular dichroism (MCD), to be diamagnetic (S = 0) [108] (Fig. 4B, i). In this state, the unpaired electron of the CuA center is delocalized between the two copper centers, in a mixed valence [Cu1.5+ –Cu1.5+ ] state, with a spin S = 1/2, and the observed hyperfine splitting is due to the electron spin interaction with the nuclei of the Cu ions (with a nuclear spin of I = 3/2) [103,110–112]. As indicated above, when the enzyme is isolated in the presence of oxygen other redox states are observed, which differ on the redox state of the CuA center (oxidized or reduced). In this case, when the CuA center is reduced, the EPR spectrum is characterized by a broad

and poorly resolved hyperfine splitting signal, with g-values at 2.16 and 2.04 [107,113] (Fig. 4B, iii), which masks the seven-line EPR spectrum of the CuA center in the oxidized state [103,107] (Fig. 4B, ii), and corresponds to the redox inert CuZ* [1Cu2+ –3Cu+ ] state. Since the first spectroscopic characterization of N2 OR, it has been suggested that in any enzyme preparation the “CuZ center” is not in a single form [99,103,107,108]. Usually the CuA center is reduced or oxidized, and the “CuZ center” is a mixture of CuZ and CuZ*, that usually is richer in CuZ when the enzyme is isolated in

Fig. 4. The UV–visible absorption (A) and EPR (B) spectra of different redox forms of Ma. hydrocarbonoclasticus N2 OR. (i) Enzyme with the “CuZ center” as CuZ in the [2Cu2+ –2Cu+ ] state and the CuA center oxidized, (ii) enzyme with the “CuZ center” as CuZ* in the [1Cu2+ –3Cu+ ] state and the CuA oxidized, (iii) enzyme with the “CuZ center” as CuZ* in the [1Cu2+ –3Cu+ ] state and the CuA reduced. The absorption spectrum of the enzyme with the “CuZ center” as CuZ in the [2Cu2+ –2Cu+ ] state and CuA reduced is shown in (A) as dashed line. The EPR spectra were recorded at 9.66 GHz (X-band), at 30 K [114].

[99,160] 1QNI [115,116] [118,157]

[95] 2IWF [122]

0.01 (as-isolated) 160 (MV incubation) 14 (N2 O) 12 (MV) 50 (cyt c552 ) 86 7 (as-isolated) 124 (MV incubation) 25 (N2 O) 2 (as-isolated)

[79,112] 3SBP, 3SBQ, 3SBR [124]

4 (as-isolated)

[97,103] [94]

Note: ED—electron donor, HH—horse heart, MV—methyl viologen. a U/mg = ␮mol N2 O/min × mg nitrous oxide reductase.

Specific activity, Vmax (U/mg)a Km (␮M) (electron donor) Reference Structure

[121] 1FWX [121]

6 (HH cyt c) 7 (N2 O)

9 3 122

2.180 2.050 2.180 2.030 35 2.180 2.030 38 2.180 2.046 43

Catalytic properties

550 (8000)

2.181 2.038 38

535 645 (5000) 550 (1700) 640 550 (10,600)

2.224 2.042 35 g|| g⊥ A (G) EPR

625 (3200) 530 (6700) 620 (4200) 540 (14,000)

480 (5400) 480 480 Visible spectrum (nm) (ε, M−1 cm−1 )

2.18 2.02 40

540 640 (7000) 530 (5000)

480 480

Blue/purple (aerobic) Pink (aerobic) Blue (anaerobic) Pink (aerobic) Purple (anaerobic) Pink (aerobic) Purple (anaerobic) Blue (aerobic) Purple (anaerobic) Form (purification)

Paracoccus denitrificans

Bacterial source Property

Table 3 Properties of the most studied nitrous oxide reductases.

Paracoccus pantotrophus

Pseudomonas stutzeri

Achromobacter cycloclastes

Mariobacter hydrocarbonoclasticus

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

339

the absence of oxygen or the enzyme (or the cells prior to purification) was not subjected to many freeze–thaw cycles [114]. In the case of Ps. stutzeri N2 OR isolated under exclusion of oxygen [107], the amount of CuZ* is very small, while the enzyme isolated from Pa. pantotrophus, using a similar procedure, has a ratio between CuZ* and CuZtotal is 0.29, that increases to 0.66 when the enzyme is isolated in the presence of oxygen [103]. Therefore, due to the instability and the difficulty in isolating N2 OR with its catalytic center completely as CuZ, most studies have been performed on enzyme preparations with the “CuZ center” as CuZ*, which can be obtained by reducing the enzyme isolated in the presence of oxygen with sodium ascorbate or dithionite. Moreover, until very recently all the available X-ray structures of N2 OR were obtained from enzyme preparations with the “CuZ center” as CuZ* (Section 4). The reduction process of CuZ and CuZ* with dithionite is still not completely clarified. In the enzyme containing only CuZ* the reduction only affects the CuA center, since the catalytic center is in the [1Cu2+ –3Cu+ ], a redox inert state (Table 2); on the other hand, for the CuZ-containing N2 OR, reduction occurs in two distinct kinetic steps: a fast reduction of the CuA center and a slow process of reduction that corresponds to the reduction of CuZ from the [2Cu2+ –2Cu+ ] state to the [1Cu2+ –3Cu+ ] state. The resulting reduced form has a different visible spectrum from the [1Cu2+ –3Cu+ ] state of CuZ*, since the maximum absorption is shifted from 640 nm to higher values (680 nm for Pa. pantotrophus N2 OR [103], or 660 nm for Ma. hydrocarbonoclasticus N2 OR [114]) and the band is much broader (spectrum not shown). Moreover, the redox behavior is also different, since CuZ* in the [1Cu2+ –3Cu+ ] state cannot be further oxidized, while CuZ in the [1Cu2+ –3Cu+ ] state can be reoxidized by potassium ferricyanide [103,114]. Several experimental or theoretical studies aiming to explain the electronic structure and the electron storage capacity of the “CuZ center”, required for the reduction of nitrous oxide to dinitrogen, and to propose a catalytic mechanism for this process have been reported. Although the structure of “CuZ center” was only unraveled with the determination of the first X-ray structure of this enzyme [115,116], earlier spectroscopic studies had already pointed out that it would be the first and up-to-now unique example of a biological tetranuclear copper center containing inorganic sulfur [107,108]. The first evidence for the presence of an inorganic sulfur in the “CuZ center” was first pointed out by Farrar and co-authors [107,108], as the magnetic circular dichroism transitions of “CuZ center” as CuZ* (inactive and redox inert [1Cu2+ –3Cu+ ] state) exhibited a pair of perpendicular polarized bands centered at 640 nm, which was an indication of a metal–ligand covalency, and the position and intensity of the optical bands also pointed out for CuZ having thiolate ligands [108]. The other evidence was the presence of inorganic sulfur in the Ps. stutzeri and Pa. pantotrophus enzyme, in a 6.2 Cu/S ratio [117], which later was also quantified in the other most studied enzymes isolated from Ma. hydrocarbonoclasticus [115], Pa. denitrificans [116] and Ac. cycloclastes [118] as 6.3, 7.8 and 3.8, respectively. Moreover, in the resonance Raman spectrum was identified a Cu–S vibrational mode in the RR spectrum of the CuA deficient N2 OR [117]. These data were obtained after the determination of the first X-ray structure of N2 OR (Section 4), that although inorganic sulfur was present in a 6.2 Cu/S ratio, the “CuZ center” was modeled with an oxygen atom [115]. The spin state of CuZ* was assigned S = 1/2 through the analysis of variable temperature, variable field MCD, which in combination with Cu K-edge X-ray absorption spectroscopy, led to the identification of the oxidation state of the four copper atoms as being [1Cu2+ –3Cu+ ] [102,105]. Moreover, although it was proposed that there is only one Cu atom in the oxidized form, this did not explain the odd number of lines of the metal hyperfine pattern observed

340

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Fig. 5. Structure of N2 OR and of its copper centers in the different redox states. (A) Representation of Ma. hydrocarbonoclasticus N2 OR functional dimer. On the left, the dimer is colored according to subunit, with one monomer colored gray and the other purple, the CuA and the “CuZ” centers are colored in dark blue and light blue, respectively. On the right, the surface of the dimer is colored according to subunit showing the functional dimer. The figure was created with Chimera using 1QNI coordinates. (B) The CuA and CuZ* centers have their ligands colored according to element and the copper ions in the CuA and the “CuZ” center are numbered I and II, or I, II, III and IV, respectively. The “CuZ center” is in the [1Cu2+ –3Cu+ ] redox state. The figure was created with Chimera using 1QNI coordinates. (C) The CuA and the “CuZ” centers have their ligands colored according to element and the copper ions in the CuA and the “CuZ” centers are numbered I and II, or I, II, III and IV, respectively. The “CuZ center” is in the [2Cu2+ –2Cu+ ] redox state. The figure was created with Chimera using 3SBQ coordinates.

in the X-band EPR spectrum [105]. However, the features of this spectrum could be described considering the presence of two oxidized copper atoms with one dominating the hyperfine splitting (A|| = 61 G) over the other (A|| = 24 G), and with the ratio between these hyperfine constants, of approximately 5:2, giving the ratio of the spin densities of these two copper atoms [104,105]. Therefore, CuZ* could be described as having the spin density partially delocalized over the four atoms and the sulfide bridging ion. In the case

of Ma. hydrocarbonoclasticus CuZ*, the spin density distribution was estimated by density functional theory (DFT) calculations as being mainly over CuI (0.42) and CuII (0.18) with 0.15 on the sulfur atom [102,104] (see Fig. 5 for identification of the copper atoms). In the case of Pa. pantotrophus CuZ center (also in the [1Cu2+ :3Cu+ ] state but redox active), the spin density is more widely distributed, and the copper atoms have the following spin density: CuI (0.20), CuII (0.10), CuIII (0.05) and CuIV (0.09), with 0.18 on the sulfur atom

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

and considering that there is 0.29 over the ligand that is present in between CuI and CuIV [113], identified at the time as being a water/hydroxyl molecule (see Section 4 and Fig. 5). Indeed, the electron delocalization over the “CuZ center” is one of the important features of this center as it is proposed to contribute to a low reorganization energy of the redox processes during the catalytic cycle and also to the stabilization of the oxidized form of the “CuZ center” after nitrous oxide reduction [102]. The “CuZ center” of Pa. pantotrophus, Ps. stutzeri and Ma. hydrocarbonoclasticus N2 OR has also been characterized using resonance Raman (RR) spectroscopy [104,106,117]. The RR spectrum of CuZ* excited at 624.4 nm (close to the maximum absorption of this form at 640 nm) presents three vibrational modes at 366, 386 and 415 cm−1 [105], which shift upon 34 S isotope labeling to lower frequencies, and thus were assigned to Cu–S stretching vibrations [106,119]. A single 384 cm−1 mode is observed in the RR spectrum of Ps. stutzeri “CuZ center” in the [1Cu2+ –3Cu+ ] state, when excited at 647 nm, while the oxidized CuZ, in the [2Cu2+ –2Cu+ ] state, presents four sulfur isotope sensitive modes, at 293, 347, 352, and 408 cm−1 , when excited at 568 nm (close to the maximum absorption of this form at 550 nm) [106]. The assignment of the spectroscopic features to the CuA, CuZ and CuZ* centers could have only been achieved through the characterization of the several mutants of Ps. stutzeri N2 OR (on the ligands of either CuA or “CuZ center”) or for its accessory genes, which lacked either the CuA or the “CuZ center” or had the “CuZ center” as either CuZ or CuZ* (even when purified under anaerobic conditions) [78,84,120]. 4. Structure of nitrous oxide reductase The structure of N2 OR was first determined from Ma. hydrocar˚ in 2000 [115]. This structure bonoclasticus at a resolution of 2.4 A, was determined for the enzyme with the “CuZ center” as CuZ* (redox inert and catalytically inactive) and revealed that the catalytic center of this enzyme is an unique and novel metal center in biology, composed by four copper atoms bound to seven histidine residues [115], and a bridging atom, that Brown et al. [115] identified at the time as an oxygen. The nature of this bridging atom was a matter of discussion since it did not agree with some of the spectroscopic and biochemical data for this enzyme [107,108,117]. In fact, the structure was re-evaluated in the same year and the oxygen atom was then assigned to be a sulfur [116]. Since 2000, two other crystal structures of the enzyme with the “CuZ center” as CuZ*, one isolated from Pa. denitrificans [116,121] and the other from Ac. cycloclastes [122], have been determined. These two crystal structures were obtained at a higher resolution ˚ respectively), and differ in the redox state of the CuA (1.6 and 1.86 A, center. In the case of Pa. denitrificans N2 OR, the CuA center was in the reduced state [121], similarly to the enzyme preparation from Ma. hydrocarbonoclasticus, while for Ac. cycloclastes N2 OR the CuA center was in the oxidized state [122]. The overall folding of N2 OR is similar in all the solved structures. N2 OR is a homodimer, with a contact surface between the monomers of 6250 A˚ 2 , which corresponds to around 26% of the total solvent-accessible surface area of the monomer [115,122]. Each monomer is composed of two domains, each containing one of the copper centers. The structure of the N-terminal domain has a seven-bladed ␤-propeller fold and binds the “CuZ center”, the catalytic center, while the C-terminal has a cupredoxin fold and binds the electron transferring copper center, CuA (Fig. 5A). The arrangement of the two monomers is such that the CuA center from one monomer is located at 10 A˚ from the “CuZ center” of the other subunit, making N2 OR a functional dimer, as the two copper centers in one subunit are 40 A˚ apart (Fig. 5A), a distance that would hinder intramolecular electron transfer in a monomer [123]. The two

341

residues M5703 and F564 from the C-terminal domain bridge the solvent channel between the two centers. As will be discussed in Section 6, these residues might be involved in the intramolecular electron transfer pathway between the two centers. The cupredoxin domain (480–580) is composed by nine ␤strands, forming an antiparallel ␤-sandwich in a Greek key motif, which is also found in the cytochrome c oxidase domain that binds CuA center. CuA center is located in the loop region between two ␤strands near the C-terminus, and is composed of two copper atoms coordinated by two S␥ atoms of C561 and C565, one S␦ of M572, two N␧2 of H526 and H569 and the carbonyl atom of W563. The two cysteine residues bridge the two copper atoms (CuAI and CuAII ), while the other residues bind only CuAI (H526 and M572) or CuAII (H569 and W563) (Fig. 5B). The ␤-propeller domain (10–440) has the blades radially arranged around a central solvent channel, which has in one end a 11 A˚ diameter and in the other a 4.5 A˚ diameter, with “CuZ center” being located in the middle of this channel. The four copper atoms of “CuZ center” are coordinated by seven conserved histidine residues through their N␧2 atom (H80, H128, H270, H325, H376) or N␦1 atom (H79 and H437), with a bridging inorganic sulfide (Fig. 5B), adopting the shape of a distorted tetrahedron. All copper atoms, with exception of CuIV , are coordinated by two histidine ligands, which lead to the proposal of this being the substrate binding site. In fact, in all these structures, it was observed the presence of a bridging ligand between CuI and CuIV , though its nature is still under discussion. In Ma. and Pa. N2 OR structures only one oxygen atom of a hydroxyl group or a water molecule was bound to CuIV , while in Pa. N2 OR, two oxygen atoms, of a water molecule and a hydroxyl group, were observed, bridging CuI and CuIV (Fig. 5B). Moreover, in the case of Ac. cycloclastes N2 OR this bridging position was also the binding site observed for the inhibitor iodide [122]. The first N2 OR structures were determined with the “CuZ center” as CuZ* (redox inert and catalytically inactive) and with CuA center either oxidized or reduced. However, “CuZ center” can also be isolated in the oxidized [2Cu2+ :2Cu+ ] state, which is a redox active but also non-catalytically competent form of the enzyme [114] (Section 5). This difference in redox activity has been proposed to be due to conformation/coordination alterations of/around “CuZ center”, which might now be explained by the crystal structure of Ps. stutzeri N2 OR in the fully oxidized state, that was recently determined at a resolution of 1.7 A˚ and 2.1 A˚ [124,125]. The global fold of the fully oxidized Ps. stutzeri N2 OR (with the CuA center oxidized and the “CuZ center” as CuZ in the [2Cu2+ :2Cu+ ] state) is identical to that of the other enzymes, but the coordination sphere of both copper centers is different (Fig. 5C) [124]. In the case of the “CuZ center”, one striking difference is the presence of an additional sulfide ion bridging CuI and CuIV , in the ␮4 -sulfidebridged [4Cu:S] center. This additional sulfide ion is in a similar position to the one of the oxygen atoms observed in the other structures. Moreover, in this structure, the side-chain of the histidine, corresponding to H526, is not coordinating CuA1 as its imidazole ring is rotated 130◦ and its N␦1 is forming a hydrogen bond with the ˚ and its N␧2 is close to the ␤-carboxy group of D519 O␥ of S493 (2.6 A) ˚ (2.8 A), which has been proposed to be located in the electron entry region [57,126] (see Section 6) and are conserved residues (Fig. 2). Addition of substrate to the crystals of Ps. stutzeri N2 OR in the fully oxidized state led to changes mainly in the coordination sphere of the CuA center. In this other structure, the imidazole ring of H526 is rotated back and now coordinates CuAI , though at a ˚ [124] longer than the one reported in the other distance (2.6 A) structures (2.1 A˚ for Ma. hydrocarbonoclasticus [115], 2.0 A˚ for Pa.

3 Unless otherwise stated the numbering is according to Ma. hydrocarbonoclasticus N2 OR sequence.

342

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Scheme 1. Representation of the resonance structures of the linear N2 O molecule.

denitrificans [121] and 2.2 A˚ in the case of Ac. cycloclastes N2 ORs [122]). Unfortunately the visible spectra of the crystals, prior and after pressurization with nitrous oxide, were not acquired and thus it is difficult to determine whether the coordination sphere of the CuA center, especially around CuAI , is not coupled to CuA reduction and is simply due to substrate binding, though the CuA center does not seem to be affected by nitrous oxide addition when both copper centers are oxidized [124]. 5. Enzymatic activity of nitrous oxide reductase The enzymatic properties of the different redox states of N2 OR described in Sections 2–4 comprise a useful basis to expand our analysis regarding the catalytic aspects of this enzyme. Indeed, the data obtained in in vitro enzymatic assays will be discussed in the light of the structural and electronic properties of the enzyme and in particular of the catalytic “CuZ center”. 5.1. Activation of N2 O Nitrous oxide is a linear asymmetrical molecule, and its electronic and structural properties can be described with the resonance structures shown in Scheme 1, in which the average ˚ respecinteratomic N–N and N–O distance are 1.128 A˚ and 1.184 A, tively, shorter than the average double bond values. As mentioned in Section 2, nitrous oxide reduction is highly exergonic (Go = −339.5 kJ mol−1 ) [35], and this molecule is a stronger oxidant than molecular nitrogen, as seen by its redox potential (E◦ (pH 7.0) = 1.35 V). Although this reaction is thermodynamically favorable, a high activation barrier makes this process kinetically unfavorable, with an activation energy of around 250 kJ/mol [127], consistent with a spin-forbidden process [128]. The kinetic barrier of this reduction can be overcome by coordination and activation by metal ions. However, the coordination chemistry of this molecule with transition metals is quite poor due to the weak ligand properties [53], a problem that was solved by nature through the use of the unique tetranuclear copper-sulfide bridged center, the “CuZ center” of N2 OR. The most common reactivity of a N2 O–metal complex with a redox active metal, reported to date, involves the breaking of the O N bond with release of molecular nitrogen and formation of a O Mn+2 species [129–133]. Other reactions involve the net insertion of an oxygen atom into metal–hydride [134,135] or metal–carbon bonds [128,136,137] to yield hydroxyl–, alkoxyl–, or aryloxyl–metal complexes. Moreover, the reductive denitrification of N2 O to yield metal nitride and nitrosyl complexes has also been proposed [138,139]. The high reactivity and the transient nature of the metal–N2 O are the reasons why it is very difficult to obtain X-ray structure for such complexes. However, recently the first structure of a M–N2 O has been reported, in which a vanadium complex supported by pyrrolide ligands can reversibly bind N2 O in a N-bound endon coordination mode [140]. However, the better characterized metal–N2 O complex is the [RuII (NH3 )5 (N2 O)]2+ complex, that was first presented by Armor and Taube [141–143], and subsequently prepared with different counter ions [144,145]. The spectroscopic and electronic properties of this complex have been recently characterized by infrared and Raman spectroscopy [146], indicating that the N2 O molecule is coordinated to the ruthenium ion in a linear end-on mode through the terminal nitrogen atom. A similar binding

geometry has also been proposed for another ruthenium complex (RuCl2 (␩1 -N2 O) (o-(N,N-dimethylamino) phenyl) diphenylphosphine) (PPh3 ) [147]. These results are in agreement with precedent theoretical studies on the [RuII (NH3 )5 (N2 O)]2+ complex that suggested that the most favorable binding mode is a coordination in a Ru–N–N–O fashion with N–N–O angles close to 180◦ [148]. An increase in the electron-donation (back-bonding) from the metal to the ␩1 -N–N–O coordinated molecule might favor the activation of N2 O, as illustrated by a recent osmium-complex [149]. More recently, in an attempt to mimic the “CuZ center” of N2 OR, several multinuclear Cu–sulfide complexes supported by N-donor ligands have been obtained [53,150,151]. As will be discussed in Section 5.4, the binding of nitrous oxide to some of these complexes has been proposed based on DFT calculations as being in a ␮-1,1O fashion, inducing the scission of the N O bond, and leaving the terminal oxygen bridging two copper atoms. An interesting observation was that the only cluster capable of reducing nitrous oxide to molecular nitrogen, at low temperatures is a localized mixed-valent [Cu3 S2 ]2+ cluster [152]. 5.2. Catalytic properties of nitrous oxide reductase Different experimental assays have been used in order to evaluate both the enzymatic and cellular activity toward nitrous oxide reduction. In particular, methodologies using gas-chromatography [153] and Clark-type electrodes [154] were proposed, but the most used assay to estimate the enzymatic activity was the spectrophotometric assay developed by Kristjansson and Hollocher [155], that follows the oxidation of methyl or benzyl viologen cation radical upon reduction of N2 O to N2 . The strong reducing conditions used in the spectrophotometric assays, and the increase in catalytic activity observed during incubation with reduced methyl viologen, suggested the need for enzyme activation [156]. In fact, as described in Section 3, dithionite reduced CuZ or CuZ*, both in the [1Cu2+ –3Cu+ ] state, cannot be further reduced, and CuZ* is also redox inert, though in in vitro assays, N2 OR with the “CuZ center” in either form present similar catalytic activity [103]. Therefore, since reduction of nitrous oxide requires two electrons it has been proposed that the “CuZ center” must undergo conformational changes to become redox and catalytically active. The work of Moura’s, Dooley’s and Solomon’s groups [156,157] identified the activation mechanism as being the complete reduction of the “CuZ center” to a [4Cu+ ] state, by correlating the increase in activity with a decrease in the intensity of the EPR signal of the enzyme (specifically of the “CuZ center”, as the CuA center is already reduced by dithionite), from a paramagnetic [1Cu2+ –3Cu+ ] state to a diamagnetic spectroscopically silent [4Cu+ ] state. This process involves a slow reaction, since the kinetic rate constant was determined to be 0.07 min−1 [156]. As mentioned in the previous sections, N2 OR from Ps. stutzeri, Pa. pantotrophus, Pa. denitrificans, and recently also from Ma. hydrocarbonoclasticus can be isolated with the “CuZ center” in the [2Cu2+ –2Cu+ ] oxidation state (as CuZ) [103,106,108,114]. Although the “CuZ center” in this state is redox active, the catalytic activity of the enzyme is low, and similar to the one determined for the enzyme with the “CuZ center” as CuZ* [103]. Moreover, the activity of Ac. cycloclastes N2 OR, with “CuZ center” mainly in the [2Cu2+ –2Cu+ ] oxidation state, increased from 8 to 125 U/mg after incubation with reduced methyl viologen, in agreement with an activation process in which the “CuZ center” must be fully reduced for the enzyme to attain maximum activity [157]. The pH profile of the catalytic activity of Ac. cycloclastes nitrous oxide reductase activated at different pHs, opened the question as whether the activation involves not only the complete reduction of the “CuZ center”, but also protonation/deprotonation of a ligand

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

or the transfer of two hydrides, since the proton concentration has multiple effects on both activation and catalytic activity [118,158]. Moreover, spectroscopic, computation and kinetic studies on both Ac. cycloclastes and Ma. hydrocarbonoclasticus N2 ORs have shown that the effect of pH on the rate of reduction of the “CuZ center” has a pKa of 9.0 [119]. This pKa has been proposed to correspond to the protonation of a conserved lysine residue close to the “CuZ center” (K412 and K397 for Ac. cycloclastes and Ma. hydrocarbonoclasticus N2 ORs, respectively) [119], which would raise the redox potential of the “CuZ center” and provide a proton to lower the barrier for both the N–O cleavage and reduction of the “CuZ center” to the fully [4Cu1+ ] state during the turnover cycle [159]. Although it is now recognized that in vitro high catalytic activity is only attained after enzyme activation, most of the kinetic studies were performed with non-activated enzyme. Kinetic parameters of N2 OR isolated from different sources were determined using methyl- or benzyl-viologen as an artificial electron donor, estimating an apparent Km in the micromolar range, while Vmax varied between bacterial sources (Table 3). The activation process results in activity being time-dependent on enzyme incubation with the artificial electron donor, making the comparison between reported data difficult. In fact, in the first reported assays, the enzyme was incubated with reduced methyl or benzyl-viologen (during an usually non-reported time) and the assay was initiated by addition of substrate. However, recently a modified version of this assay was reported, enabling the separation between the activation process and the catalytic activity [160], as the assay is initiated by the addition of the activated enzyme (and the reducing agent used is removed from the fully reduced enzyme prior to the assay in an anaerobic chamber) [160]. Although most of the kinetic parameters were obtained using artificial electron donors, several small electron donor proteins, such as c-type cytochromes, either physiological or non-physiological (mitochondrial cytochrome c), have also been reported as competent electron donors to N2 OR in assays in vitro. In the case of N2 OR purified from Rhodobacter capsulatus, Rhodobacter sphaeroides f. sp. denitrificans, Ma. hydrocarbonoclasticus and Pa. pantotrophus [97,160–162], the physiological donor is proposed to be a periplasmic c-type cytochrome, but Pa. pantotrophus N2 OR can also accept electrons from a periplasmic type 1 copper protein, pseudoazurin [97], and from the mitochondrial horse heart cytochrome c [163]. In the case of Ac. cycloclastes N2 OR [118], bovine heart cytochrome c is able to reduce the enzyme but its physiological electron donor is proposed to be a pseudoazurin [164]. Although Wo. succinogenes N2 OR, has an additional c-type heme containing domain with a proposed role in electron transfer, it can accept electrons from a periplasmic c-type cytochrome isolated from the same organism, a putative electron donor [87,89,165]. A comparative study between the physiological (cytochrome c552 ) and artificial electron donor (methyl viologen) to Ma. hydrocarbonoclasticus N2 OR, showed that the rate limiting step in the reaction is the intermolecular electron transfer for cytochrome c552 , whereas electron donation is faster than the substrate reduction when methyl viologen is used as electron donor ([160], with turnover numbers of 4 s−1 and 300 s−1 , respectively). Moreover, a difference is also observed on the pH profile of the activity with these two electron donors, with a pKa of 6.6 for methyl viologen with high activity being maintained at high pH values, while a pKa of 8.3 was determined for cytochrome c552 , and the activity decreases at high pH values. Indeed, the optimum pH activity for several N2 ORs was determined to be in the range of pH 8.0–9.5, when artificial electron donor were used as electron donors [97,100,112,160]. Since in the proposed electron donor – enzyme interface there is no residue to which a pKa of 8.3 could be attributed, attention must be turned to the reduction process of the “CuZ center”. This pKa has been proposed to be associated with a water ligand in the

343

Cu2+ –H2 O–Cu+ state, as the deprotonation of this state would lower the redox potential making it difficult for the CuA center to reduce the “CuZ center”. This problem would not exist when methyl viologen is used as electron source, considering that this small reductant can deliver electrons directly to the “CuZ center” without the need to go through the CuA center [160].

5.3. Binding of N2 O to the catalytic center of nitrous oxide reductase As mentioned in Section 4, in the structures of Ma. hydrocarbonoclasticus [115], Pa. denitrificans [116] and Ac. cycloclastes [122] N2 OR, the “CuZ center” was modeled with a water-derived molecule bound in between CuI and CuIV atoms, a position that has been proposed to be the substrate binding site, since CuIV is coordinated only by one histidine side-chain. Before the structure of Ps. stutzeri N2 OR with nitrous oxide bound was available, the model for the binding of N2 O to the “CuZ center” proposed by Solomon and co-workers [156] was based on the earlier structures and on the available experimental data. Although, as mentioned for the binding of N2 O to model compounds (Section 5.1), the substrate could bind to the “CuZ center” in different modes. The lowest energy structure for the binding mode of N2 O to the “CuZ center” in the fully reduced [4Cu1+ ] state was determined by DFT calculations to be the ␮-1,3-N2 O bridging mode [102,156] (Fig. 6B). However, in the redox state [1Cu2+ –3Cu1+ ], N2 O is proposed to bind through the N-atom to CuI in an end-on mode [156]. Solomon and co-workers [156] proposed that binding of N2 O in the bent ␮-1,3 bridging mode is most efficient due to the strong back-bonding from CuI and CuIV atoms. The CuZ → N2 O charge transfer and strong interaction of the ␮-1,3-N2 O ligand with the two redox active CuI and CuIV atoms provide a low activation energy pathway for the N O cleavage process, which involves two electrons. The N O bond is weakened, which facilitates the cleavage through simultaneous transfer of two electrons from the “CuZ center”. Moreover, the N2 O molecule bound to the CuI CuIV edge is bent with an angle of approximately 139◦ . This configuration makes the N2 O ligand a very good electron acceptor since the N2 O ␲* orbital loses in antibonding character and is stabilized close to the fully occupied d orbitals of the CuZ (4CuI ) cluster [156]. The created electron holes on CuI and CuIV by the two-electron reduction of N2 O can be efficiently delocalized through the bridging sulfide exchange pathways, contributing to the thermodynamics and kinetics of this process, by lowering the geometric and electronic reorganization energies between the fully reduced and oxidized “CuZ center” states. As described previously in Section 4, a three-dimensional structure of the purple N2 OR from Ps. stutzeri indicates that the “CuZ center” has an additional sulfur atom, forming a [Cu4 S2 ] cluster [124] (Fig. 6A). Since the additional sulfur occupies the edge position between CuI and CuIV , similar position to the water-derived molecule, the binding site for N2 O must be located elsewhere or a rearrangement has to occur in the catalytic form of the enzyme. The first structural data for the mode of binding of nitrous oxide was a structure from the same authors obtained after pressurization of the crystals in which nitrous oxide is shown to bind near the “CuZ cluster”, in a side-on manner at the cluster face built by the atoms CuII , CuIV and S1 [124]. Since the interatomic distance between N2 O ˚ between N1 of nitrous oxide and the “CuZ center” is too long (3.3 A, and CuIV of the “CuZ center”) to indicate a direct coordination of N2 O to the cluster, the authors suggested the important role of the protein backbone in the substrate coordination and orientation, in particular through residues Phe621, Met627 and His626. The N2 O is placed at the bottom of a hydrophobic channel that allows the

344

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Fig. 6. Nitrous oxide binding mode to the catalytic “CuZ center” of N2 OR. (A) Substrate binding mode proposed based on the crystallographic structure of Ps. stutzeri N2 OR obtained after pressurization with nitrous oxide. The figure was created with Chimera and 3SBR coordinates. (B) Model for substrate binding mode to CuZ* center proposed for Ma. hydrocarbonoclasticus N2 OR based on structural, electronic and theoretical calculations [156]. This information was used to model the binding of nitrous oxide in between CuI and CuIV atoms of the “CuZ center” in a 1,3-N2 O bridging mode. The figure was created with Chimera and 2IWF coordinates. Atoms colored as in Fig. 5.

entry of the substrate from the surface and the release of the product N2 . The distance between nitrous oxide and the “CuZ center” suggests weak interaction and a conformational change in the protein structure must be assumed to enable the coordination of N2 O to the “CuZ center” during the catalytic cycle, since the “CuZ center” in this redox state is catalytically inactive. However, no further data or theoretical models for binding of N2 O to the Cu4 S2 cluster are available at this moment. In the light of this new “CuZ center” structure, the recently reported copper-sulfide models reported by Tolman and coworkers [152], gain interest as not only it has nitrous oxide reduction activity, but also has two sulfide ions bridging the two copper atoms. Moreover, this tri-nuclear copper-sulfide complex is proposed to bind N2 O in an alternative mode, compared with that proposed by Solomon and co-workers [156], in a ␮-1,1-O coordination after which a fragment of Cu(I) dissociates yielding a dicopper (I,II) species, able to promote the N O bond scission. Interestingly, the estimated activation barrier presented is similar to the one calculated for the enzyme-catalyzed process [159]. 5.4. Intermediates in the catalytic cycle The enzymatic reaction catalyzed by N2 OR is a multi-step process in which several reactions, including electron-transfer and protonation steps, contribute to the whole catalytic cycle. The identification of these intermediate species is crucial for us to understand what the active form of the enzyme is, even if its characterization is often difficult due to the transient nature of the catalytic intermediates. As mentioned in Section 5.2, N2 OR requires a prolonged incubation before reaching high activity through a process that is very slow compared with the fast catalytic cycle (k = 0.07 min−1 for the activation process compared with a k = 300 s−1 for the catalytic cycle in the presence of methyl viologen) [156,160]. However, the fully reduced form of the “CuZ center” ([4Cu+ ]) obtained after the activation process is a catalytically active form of the enzyme and can be considered as a starting point in the catalytic cycle. Solomon and co-authors [159] have proposed, based on density functional theory (DFT) calculations of the reaction thermodynamics and potential energy surfaces, that after N O bond cleavage,

the catalytic cycle consists in a sequence of protonation and oneelectron reduction steps which return the “CuZ center” to the fully reduced [4Cu+ ] state for further turnover. Recently, single-turnover experiments between the preactivated N2 OR and a stoichiometric amount of nitrous oxide, monitored by UV–visible spectroscopy, allowed the identification of an intermediate in the turnover cycle of the enzyme, named CuZ◦ [166]. This intermediate species has CuA center in the oxidized state (with absorption bands at 480 and 540 nm) and the “CuZ center” in the [1Cu2+ –3Cu1+ ] state, with an absorption band at 680 nm, but its catalytic activity is identical to the enzyme in the fully reduced state. The life-time of this intermediate species is short (decay with k = 0.3 min−1 ), and is correlated with the decay in catalytic activity [166] and increase of the CuZ* state, characterized by the 640 nm absorption band. It was proposed that the redox state of the “CuZ center” in this new intermediate CuZ◦ is identical to the one of CuZ*, [1Cu2 –3Cu1+ ]. However, subtle but important local structural rearrangements must occur during this process to account for the different catalytic activity and probably redox properties. With the identification of CuZ◦ species as an intermediate in the catalytic cycle of N2 OR, the mechanism of reduction, catalysis and inactivation of the enzyme was revised (Fig. 7). In this mechanism, CuZ* does not participate in the catalytic cycle, since its reduction is very slow (requiring an activation process in the presence of strong reducing agents), and the fast turnover cycle implies that the re-reduction of the N2 O-oxidized copper center must be fast, thus excluding the involvement of this species. This is also in agreement with the fact that all the as-purified forms of N2 OR present very low catalytic activity (Table 3), unless subjected to a prolonged reduction–activation mechanism. The conversion of CuZ◦ to CuZ* is a slow process indicating that a simple proton transfer and/or structural changes in “CuZ center” might probably be required to meet the optimized structure for the catalytic cycle to continue after product release. However, structural information regarding the active fully reduced and CuZ◦ forms are still absent and represent an important goal for further research. As illustrated in Fig. 7, the first intermediate after the 2-electron reduction of nitrous oxide has been proposed to be a [2Cu2+ –2Cu+ ] state (Fig. 7, box with dashed line). However, this intermediate

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

345

Fig. 7. Mechanism of activation and proposed catalytic cycle of N2 OR with the “CuZ center” as CuZ*. Copper ions are represented as blue circles in the (II) oxidation state or gray in the (I) oxidation state. In the CuZ* form, the CuI 2+ ion is represented as a blue square to indicate that this form is redox inert. However, in either the “CuZ” or the CuA centers, the unpaired electrons are delocalized within the cluster through the sulfur(s) atom(s), represented as yellow circles. ED—electron donor.

has not yet been trapped under turnover conditions, probably because fast delivery of one electron from the CuA center makes this form very short lived. The study of this key intermediate in turnover using fast-kinetic techniques will be surely an important step in the characterization of the catalytic cycle of the enzyme. 6. The electron transfer route The characterization of the transient electron transfer complex between small electron shuttle proteins and N2 OR, and the identification of the pathways involved in the electron route between the different metal cofactors are also useful to understand the global enzymatic mechanism. The whole electron transfer pathway required for a single turnover can be analyzed by separating the intramolecular electron transfer within N2 OR and the intermolecular electron transfer, that consists in the electron delivery from an electron shuttle protein to the CuA center. 6.1. The intramolecular electron transfer: CuA–CuZ As described in Section 4, N2 OR is a functional homodimer, with the CuA center from one monomer at 10 A˚ from the “CuZ center” of the other monomer (Fig. 5A). Up to now the experimental data regarding this electron transfer reaction are scarce; however, it is possible to extend the analysis made for cytochrome c oxidase,

which contains an analogous CuA center [167,168], as an electron transferring center, and for which several parameters and properties of the electron transfer between the metal cofactors within the enzyme have been determined [35,109]. In cytochrome c oxidase, the CuA center receives electrons from a cytochrome c, which are then transferred to the electron transfer center, heme a, and subsequently to the dinuclear catalytic center heme a3 – CuB center [169]. An alternative pathway, represented by the direct electron transfer from the CuA center to heme a3 , is proposed to be a physiological short circuit [170551], though this secondary route is slower than the CuA – heme a route, due to the ˚ respechigher distance between the redox cofactors (19 A˚ and 16 A, tively – distances based on Pa. denitrificans cytochrome c oxidase structure [171]). The electron transfer between the CuA center and heme a depends on the source of the enzyme and the experimental methodology used, but in general is characterized by a relatively high rate constant in the range between 2 × 104 s−1 and 10 × 104 s−1 [172–174], despite the distance of approximately 16 A˚ between the two metal centers. The estimated rate is in excellent agreement with the rate constant of 8.7 × 104 s−1 predicted from the structure by the Moser-Dutton equation [170]. Fast kinetic experiments on Rhodobacter sphaeroides cytochrome c oxidase mutants showed the crucial role of a histidine residue that binds to CuA center in the electron transfer to heme a [175]. The structure of the enzyme shows that the electrons can be delivered from the histidine to two adjacent arginine residues that are connected to heme a propionates via hydrogen bonds, an electron transfer

346

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

Fig. 8. The proposed intramolecular electron transfer pathway across the dimer interface between the CuA center and the “CuZ center”. (A) The two intramolecular electron transfer pathways from the CuA to the “CuZ center” proposed for Ma. hydrocarbonoclasticus N2 OR. The first route, indicated as ET1, involves Trp563, which is a CuA ligand, and the neighbor Phe564 (Trp563 and Phe564 are represented in purple); the electron is then transferred to the oxygen of a water molecule or a hydroxyl group bound between CuI and CuIV of the “CuZ center” as CuZ*. An alternative route, indicated as ET2, involves the electron transfer from the CuA center to His569, then to Met570, and subsequently to His128, which is a ligand of CuII of the “CuZ center” (His569, Met570, and His128 are represented in orange). The ligands of the “CuZ center” (His79, His80, His128, His270, His325, His376, His437) are represented in green, while the CuA ligands (His526, Cys561, Cys565, His569, Met572) are represented in red. The copper atoms of the CuA and the “CuZ center” are represented as blue spheres, and the sulfur atom in the “CuZ center” is represented in yellow. (B) Intramolecular electron transfer pathway proposed for Ps. stutzeri N2 OR in the presence of nitrous oxide. The most favorable route involves the electron transfer from the CuA to the ligand His626, subsequently to Met627 (represented in light blue) and finally to nitrous oxide located in the proximity of the “CuZ center”. The ligands of the “CuZ center” (His129, His1300, His178, His326, His382, His433, His494) are represented in green, while CuA ligands (His583, Cys618, Cys622, Trp620, Met629) are represented in red. The copper atoms of the CuA and the “CuZ center” are represented as blue spheres, and the sulfur atom in the “CuZ center” is represented in yellow. In both panels the surfaces of the two monomers of N2 OR are represented in light green and light pink, respectively.

pathway that has been recently confirmed by a quantum mechanics/molecular mechanics (QM/MM) study [176]. Regarding N2 OR, no kinetic parameters are available for the electron transfer from the CuA center to the catalytic “CuZ center”, though a few theoretical models have been proposed based on the different crystal structures. An investigation performed on Ma. hydrocarbonoclasticus N2 OR with the “CuZ center” as CuZ* identified two possible electron transfer routes [57]. One involves Trp563, which is a CuA ligand and the neighbor Phe564 (Trp563 and Phe564 are represented in purple in Fig. 8A); the electron is then transferred to the oxygen of a water molecule or hydroxyl group bound in between CuI and CuIV of the “CuZ center”. The alternative route involves the electron transfer from the CuA center to His569, which is a CuA ligand, then to Met570 and subsequently to His128, which is a ligand of CuII of the “CuZ center” (Fig. 8A). An analog analysis using Pa. denitrificans N2 OR structure identified a pathway where the electron jumps directly from His639, the CuA ligand corresponding to His569 in Ma. hydrocarbonoclasticus N2 OR, to the water ligand located near the “CuZ center” [126]. However, the presence of an additional sulfur atom in the “CuZ center” of Ps. stutzeri N2 OR [124] and the identification of a substrate binding cavity in between the two metal cofactors, raised the hypothesis that both the CuA and the “CuZ” center cooperate in nitrous oxide reduction [124]. In this model, nitrous oxide is located in a binding pocket in which His626 (numbered according to Ps. stutzeri N2 OR primary sequence), a ligand of CuA, and Met627 assists in substrate orientation (Fig. 8B), and since nitrous oxide is placed in the direct route for the electron transfer between the CuA and the “CuZ” center, the authors [124] suggested that both centers are required for the reduction of nitrous oxide during catalysis. 6.2. The intermolecular electron transfer: electron donor-enzyme Small electron carrier proteins, either c-type cytochromes or type 1 copper proteins, have been used as electron donors to N2 OR in order to reproduce the physiological conditions in the in vitro

assays, as described in Section 5.2. Moreover, the electron transfer from these small electron donor proteins is specific to the CuA center since the catalytic “CuZ center” is not reduced by these proteins [160,163]. On the contrary, small reduced dyes (methyl viologen or benzyl viologen) are able to reduce both the CuA and the “CuZ” centers, at different rates: reduction of the CuA center is fast, while the “CuZ center” is slowly reduced [35]. Therefore, it has been proposed that these small dyes are able to interact directly with both the “CuZ” and the CuA centers. The recognition and orientation of the encounter electron transfer complex is driven by the nature and composition of the surface of both proteins. In the case of Ma. hydrocarbonoclasticus N2 OR, ionic strength dependence of the catalytic activity indicated that the complex with the physiological donor is driven by hydrophobic interactions, whereas for Pa. denitrificans and Ac. cycloclastes N2 OR, the interaction is mainly governed by electrostatic effects [57], as suggested by the experimental data [160,163,177,178]. This information has been used to sort the putative model structures of the electron transfer complexes between N2 OR from different organisms and its respective physiological electron donors [57]. In those model structures, the small electron donor (Ma. hydrocarbonoclasticus cytochrome c552 , both pseudoazurin and cytochrome c550 from Pa. denitrificans and pseudoazurin from Ac. cycloclastes) was proposed to bind at the N2 OR surface near the CuA center [57], in accordance with the direct electron transfer experiments [97,126,160]. In fact, the exposed heme methyl of cytochrome c or the imidazolic ring of the solvent exposed histidine ligand of the copper center of pseudoazurin, are positioned at about 5 A˚ of a conserved patch in the enzyme surface, proposed to be the electron delivery site. However, the authors did not identify a single electron transfer pathway on the enzyme and instead a set of well-conserved residues have been suggested to be involved in the electron transfer route to the CuA center [57]. The similarity between the CuA center of N2 OR and cytochrome c oxidase guided the analysis of the model structures. Structural studies asserted that in cytochrome c oxidase a surface-exposed

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

tryptophan is essential for the electron transfer from a soluble cytochrome c to the CuA center [175,179]. In N2 OR, there is no surface-exposed tryptophan residue in a similar position that could be involved in a similar electron transfer route. However, Pro496 and His566 (residues numbered according to Ma. hydrocarbonoclasticus N2 OR primary sequence) occupy an analogous position to Trp121 in Pa. denitrificans cytochrome c oxidase, and thus have been suggested to be involved in the electron transfer pathway [126,160]. The conserved patch is composed by Ala495–Pro496, Asp519 and His566 (residues numbered according to Ma. hydrocarbonoclasticus N2 OR primary sequence), in which the carboxyl group of Asp519 is hydrogen bonded to His526, the terminal ligand of CuA center (Fig. 5, conserved residues highlighted in Fig. 2) [57,126,160]. In the recently determined structure of the fully oxidized Ps. stutzeri N2 OR, the side chain of this histidine residue is infact not coordinating the CuA center (Fig. 5, Section 4), in the absence of substrate [124]. Einsle and collaborators [124] suggested that only in the presence of substrate His526 will flip to coordinate the CuA center and enables its reduction by the electron donor protein.

7. Conclusions and future perspectives Since its discovery and its first characterization, the enzyme N2 OR has been a fascinating subject for bioinorganic chemists. In particular, the study of the two unusual multicopper centers, the CuA and the “CuZ center”, is an excellent challenge for the copper coordination chemistry community. Ten years after the determination of the first structure of N2 OR, many scientific advances have occurred specially on the characterization of the unique “CuZ center”, as it provided a structural basis for the interpretation of the spectroscopic data supported by theoretical calculations; the identification of the different oxidation states; the proposal of inter- and intra-molecular electron transfer routes and a better understanding of the catalytic cycle; as well as the characterization of active species involved in the enzyme turnover cycle. However, several aspects of the “CuZ center”, activation mechanism, substrate binding mode and catalysis are still unclear and remain to be answered. For many years, and it still remains a question of debate, it has been proposed by the groups lead by Thomson, Kroneck and Zumft that the purple form of N2 OR (with both the CuA and the “CuZ center” in the oxidized form) could only be isolated in the absence of oxygen. Recently, our group isolated for the first time N2 OR in the presence of oxygen, with the “CuZ center” mainly as CuZ in the [2Cu2+ –2Cu1+ ] state, from both a microaerophilic and a strictly anaerobic growth of Ma. hydrocarbonoclasticus [114]. In addition, it was observed that during prolonged (weeks) freezing at −80 ◦ C, the “CuZ center” as CuZ decays to CuZ*, which raises the question as to what is the difference between these two forms of the “CuZ center”. Recently, the structure of N2 OR with the “CuZ center” in the [2Cu2+ –2Cu1+ ] state has been determined, which may provide some insights into what this difference might be, though without further experimental data several questions about this difference remain open. The presence of an extra-sulfur in the coordination sphere of the CuZ cluster represents another big step in the characterization of this center, opening new frontiers in the correlation of its structural, spectroscopic and electronic properties. On the subject of the catalytic properties of N2 OR and the activation of the thermodynamically inert nitrous oxide more is now known. In particular, the identification of the fully reduced state CuZ [4Cu+ ] and CuZ◦ as highly reactive forms of the enzyme, when compared with any of the as-isolated forms, led to the proposal of a mechanism for the enzyme activation and catalytic cycle. However, all the available structures seem to reflect low-activity enzyme

347

forms, compared with the high-reactive fully reduced state or CuZ◦ . For this reason, the structural differences between CuZ◦ intermediate (as well as the “CuZ center” in the [4Cu+ ] state) and the “CuZ center” in the [1Cu2+ –3Cu+ ] state, and also the structure of the proposed catalytically competent intermediate in the [2Cu2+ –2Cu+ ] state still remain to be fully characterized. Moreover, the structural changes that must occur in the enzyme, either around the “CuZ center” and/or the CuA center during the activation to allow substrate binding have also to be clarified. The slow activation of the “CuZ center” and the fact that its fully reduced form has never been isolated, even when its purification is performed under exclusion of oxygen, raises the question has how this activation is attained and maintained in vivo. The hypothesis that this function may be exerted by NosR or/and NosX accessory proteins is still unexplored. Regarding the electron transfer route and the activation of the enzyme, the extensive characterization of the interaction of N2 OR with its physiological is a crucial aspect, since the results obtained so far with the strong reductants (methyl and benzyl viologen), argue that the electron route promoted is not physiologically relevant. Finally, N2 OR plays a central role in the detoxification of the green-house gas, nitrous oxide. Therefore, the understanding of the catalytic cycle and activation mechanism of this enzyme will open, for sure, a door for new ideas to improve or implement novel biotechnological applications of this enzyme, such as biosensors and removal of nitrogen compounds from wastewater, as recently discussed by Thomson et al. [180]. Acknowledgements We would like to thank the financial support provided by Fundac¸ão para a Ciência e Tecnologia to IM (PTDC/QUIBIQ/116481/2010), SRP (PTDC/BIA-PRO/098882/2008) and to SD in the form of a Ph.D. scholarship (SFRH/BD/30414/2006), and by Conselho de Reitores Universidades Portuguesa – DAAD to SRP. The C.I.R.C.M.S.B. is gratefully acknowledged for a fellowship to SD. References [1] The Environmental Chemistry of Trace Atmospheric Gases, U.S. Environmental Protection Agency, Washington, 2011. [2] A.R. Ravishankara, J.S. Daniel, R.W. Portmann, Science 326 (2009) 123. [3] K.L. Denman, G. Brasseur, A. Chidthaisong, P. Ciais, P.M. Cox, R.E. Dickinson, D. Hauglustaine, C. Heinze, E. Holland, D. Jacob, U. Lohmann, S. Ramachandran, P.L.d.S. Dias, S.C. Wofsy, X. Zhang, in: S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor, H.L. Miller (Eds.), Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge Univ. Press, Cambridge, 2007, p. 499. [4] D.J. Wuebbles, Science 326 (2009) 56. [5] K. Lassey, M. Harvey, Water Atmos. 15 (2007) 10. [6] M.J. Prather, Science 279 (1998) 1339. [7] M. Mengis, R.G.Ä. Chter, B. Wehrli, Biogeochemistry 38 (1997) 281. [8] R.R. Eady, S.S. Hasnain, in: J.A. McCleverty, T.J. Meyer (Eds.), Denitrification, in Comprehensive Coordination Chemistry II, Pergamon, Oxford, 2003, p. 759. [9] W.C. Trogler, J. Chem. Educ. 72 (1995) 973. [10] H. Shoun, D.H. Kim, H. Uchiyama, J. Sugiyama, FEMS Microbiol. Lett. 73 (1992) 277. [11] K.F. Ettwig, M.K. Butler, D. Le Paslier, E. Pelletier, S. Mangenot, M.M. Kuypers, F. Schreiber, B.E. Dutilh, J. Zedelius, D. de Beer, J. Gloerich, H.J. Wessels, T. van Alen, F. Luesken, M.L. Wu, K.T. van de Pas-Schoonen, H.J. Op den Camp, E.M. Janssen-Megens, K.J. Francoijs, H. Stunnenberg, J. Weissenbach, M.S. Jetten, M. Strous, Nature 464 (2010) 543. [12] M. Kobayashi, Y. Matsuo, A. Takimoto, S. Suzuki, F. Maruo, H. Shoun, J. Biol. Chem. 271 (1996) 16263. [13] E. Blagodatskaya, M. Dannenmann, R. Gasche, K. Butterbach-Bahl, Biogeochemistry 97 (2010) 55. [14] S.W. Kim, S. Fushinobu, S. Zhou, T. Wakagi, H. Shoun, Biosci. Biotechnol. Biochem. 74 (2010) 1403. [15] L.Y. Stein, Methods Enzymol. 486 (2011) 131. [16] D.E. Canfield, A.N. Glazer, P.G. Falkowski, Science 330 (2010) 192. [17] L.Y. Stein, Y.L. Yung, Annu. Rev. Earth Planet. Sci. 31 (2003) 329. [18] R. Yu, M.J. Kampschreur, M.C. van Loosdrecht, K. Chandran, Environ. Sci. Technol. 44 (2010) 1313.

348

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349

[19] P. Wunderlin, J. Mohn, A. Joss, L. Emmenegger, H. Siegrist, Water Res. 46 (2012) 1027. [20] J.J. Cantera, L.Y. Stein, Arch. Microbiol. 188 (2007) 349. [21] D.M. Kool, J. Dolfing, N. Wrage, J.W. Van Groenigen, Soil Biol. Biochem. 43 (2011) 174. [22] M. Poth, D.D. Focht, Appl. Environ. Microbiol. 49 (1985) 1134. [23] P. Cabello, M.D. Roldan, C. Moreno-Vivian, Microbiology 150 (2004) 3527. [24] W.G. Zumft, Microbiol. Mol. Biol. Rev. 61 (1997) 533. [25] C.M. Jones, A. Welsh, I.N. Throback, P. Dorsch, L.R. Bakken, S. Hallin, FEMS Microbiol. Ecol. 76 (2011) 541. [26] I. Verbaendert, N. Boon, P. De Vos, K. Heylen, Syst. Appl. Microbiol. 34 (2011) 385. [27] Suharti, S. de Vries, Biochem. Soc. Trans. 33 (2005) 130. [28] I. Verbaendert, P. De Vos, N. Boon, K. Heylen, Biochem. Soc. Trans. 39 (2011) 254. [29] C.M. Jones, B. Stres, M. Rosenquist, S. Hallin, Mol. Biol. Evol. 25 (2008) 1955. [30] T. Hino, Y. Matsumoto, S. Nagano, H. Sugimoto, Y. Fukumori, T. Murata, S. Iwata, Y. Shiro, Science 330 (2010) 1666. [31] Y. Matsumoto, T. Tosha, A.V. Pisliakov, T. Hino, H. Sugimoto, S. Nagano, Y. Sugita, Y. Shiro, Nat. Struct. Mol. Biol. 19 (2012) 238. [32] K. Enwall, I.N. Throback, M. Stenberg, M. Soderstrom, S. Hallin, Appl. Environ. Microbiol. 76 (2010) 2243. [33] C.M. Jones, S. Hallin, ISME J. 4 (2010) 633. [34] M.S. Coyne, A. Arunakumari, B.A. Averill, J.M. Tiedje, Appl. Environ. Microbiol. 55 (1989) 2924. [35] W.G. Zumft, P.M. Kroneck, Adv. Microb. Physiol. 52 (2007) 107. [36] P.J. Gonzalez, C. Correia, I. Moura, C.D. Brondino, J.J. Moura, J. Inorg. Biochem. 100 (2006) 1015. [37] T. Hino, S. Nagano, H. Sugimoto, T. Tosha, Y. Shiro, Biochim. Biophys. Acta 1817 (2012) 680. [38] P. Wunsch, W.G. Zumft, J. Bacteriol. 187 (2005) 1992. [39] I. Moura, S.R. Pauleta, J.J. Moura, J. Biol. Inorg. Chem. 13 (2008) 1185. [40] J.F. Ghiglione, L. Philippot, P. Normand, R. Lensi, P. Potier, J. Bacteriol. 181 (1999) 5099. [41] A.J. Gates, R.O. Hughes, S.R. Sharp, P.D. Millington, A. Nilavongse, J.A. Cole, E.R. Leach, B. Jepson, D.J. Richardson, C.S. Butler, FEMS Microbiol. Lett. 220 (2003) 261. [42] A. Jungst, C. Braun, W.G. Zumft, Mol. Gen. Genet. 225 (1991) 241. [43] J.J. Cantera, L.Y. Stein, Environ. Microbiol. 9 (2007) 765. [44] M. Nojiri, Y. Xie, T. Inoue, T. Yamamoto, H. Matsumura, K. Kataoka, Deligeer, K. Yamaguchi, Y. Kai, S. Suzuki, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 4315. [45] M.J. Ellis, J.G. Grossmann, R.R. Eady, S.S. Hasnain, J. Biol. Inorg. Chem. 12 (2007) 1119. [46] H. Arai, Y. Igarashi, T. Kodama, Biochim. Biophys. Acta 1261 (1995) 279. [47] J. Hendriks, A. Oubrie, J. Castresana, A. Urbani, S. Gemeinhardt, M. Saraste, Biochim. Biophys. Acta 1459 (2000) 266. [48] R. Cramm, A. Pohlmann, B. Friedrich, FEBS Lett. 460 (1999) 6. [49] Suharti, H.A. Heering, S. de Vries, Biochemistry 43 (2004) 13487. [50] A. Viebrock, W.G. Zumft, J. Bacteriol. 170 (1988) 4658. [51] J. Simon, O. Einsle, P.M. Kroneck, W.G. Zumft, FEBS Lett. 569 (2004) 7. [52] R.G.S. Banks, R.J. Henderso, J.M. Pratt, Chem. Commun. 8 (1967) 387. [53] W.B. Tolman, Angew. Chem. Int. Ed. Engl. 49 (2010) 1018. [54] B.B. Jensen, R.H. Burris, Biochemistry 25 (1986) 1083. [55] J. Christiansen, L.C. Seefeldt, D.R. Dean, J. Biol. Chem. 275 (2000) 36104. [56] A.T. Fernandes, J.M. Damas, S. Todorovic, R. Huber, M.C. Baratto, R. Pogni, C.M. Soares, L.O. Martins, FEBS J. 277 (2010) 3176. [57] S. Dell’acqua, I. Moura, J.J. Moura, S.R. Pauleta, J. Biol. Inorg. Chem. 16 (2011) 1241. [58] T. Palmer, F. Sargent, B.C. Berks, Trends Microbiol. 13 (2005) 175. [59] B.C. Berks, T. Palmer, F. Sargent, Adv. Microb. Physiol. 47 (2003) 187. [60] J. Simon, L. Hederstedt, FEBS J. 278 (2011) 4179. [61] H. Arai, M. Mizutani, Y. Igarashi, Microbiology 149 (2003) 29. [62] H. Cuypers, A. Viebrock-Sambale, W.G. Zumft, J. Bacteriol. 174 (1992) 5332. [63] U. Honisch, W.G. Zumft, J. Bacteriol. 185 (2003) 1895. [64] H. Cuypers, J. Berghofer, W.G. Zumft, Biochim. Biophys. Acta 1264 (1995) 183. [65] P. Holloway, W. McCormick, R.J. Watson, Y.K. Chan, J. Bacteriol. 178 (1996) 1505. [66] F.U. Hoeren, B.C. Berks, S.J. Ferguson, J.E. McCarthy, Eur. J. Biochem. 218 (1993) 49. [67] W.G. Zumft, J. Mol. Microbiol. Biotechnol. 10 (2005) 154. [68] R. Veldman, W.N. Reijnders, R.J. van Spanning, Biochem. Soc. Trans. 34 (2006) 94. [69] L. Bergaust, R.R. van Spanning, A. Frostegard, L.R. Bakken, Microbiology 158 (2012) 826. [70] K.U. Vollack, W.G. Zumft, J. Bacteriol. 183 (2001) 2516. [71] G. Giardina, S. Rinaldo, K.A. Johnson, A. Di Matteo, M. Brunori, F. Cutruzzola, J. Mol. Biol. 378 (2008) 1002. [72] G. Giardina, S. Rinaldo, N. Castiglione, M. Caruso, F. Cutruzzola, Proteins 77 (2009) 174. [73] G. Giardina, N. Castiglione, M. Caruso, F. Cutruzzola, S. Rinaldo, Biochem. Soc. Trans. 39 (2011) 294. [74] N. Castiglione, S. Rinaldo, G. Giardina, F. Cutruzzola, Microbiology 155 (2009) 2838. [75] R.J. Van Spanning, E. Houben, W.N. Reijnders, S. Spiro, H.V. Westerhoff, N. Saunders, J. Bacteriol. 181 (1999) 4129.

[76] W. Zumft, H. Korner, in: H. Bothe, S.J. Ferguson, W.E. Newton (Eds.), Biology of the Nitrogen Cycle, Elsevier, 2007, p. 67. [77] N. Minagawa, W.G. Zumft, Biometals 1 (1988) 117. [78] P. Wunsch, M. Herb, H. Wieland, U.M. Schiek, W.G. Zumft, J. Bacteriol. 185 (2003) 887. [79] J. Riester, W.G. Zumft, P.M. Kroneck, Eur. J. Biochem. 178 (1989) 751. [80] D.M. Dooley, M.A. McGuirl, A.C. Rosenzweig, J.A. Landin, R.A. Scott, W.G. Zumft, F. Devlin, P.J. Stephens, Inorg. Chem. 30 (1991) 3006. [81] M.A. McGuirl, J.A. Bollinger, N. Cosper, R.A. Scott, D.M. Dooley, J. Biol. Inorg. Chem. 6 (2001) 189. [82] L.M. Taubner, M.A. McGuirl, D.M. Dooley, V. Copie, Biochemistry 45 (2006) 12240. [83] N.F. Saunders, J.J. Hornberg, W.N. Reijnders, H.V. Westerhoff, S. de Vries, R.J. van Spanning, J. Bacteriol. 182 (2000) 5211. [84] P. Wunsch, H. Korner, F. Neese, R.J. van Spanning, P.M. Kroneck, W.G. Zumft, FEBS Lett. 579 (2005) 4605. [85] J.M. Boyd, J.A. Endrizzi, T.L. Hamilton, M.R. Christopherson, D.W. Mulder, D.M. Downs, J.W. Peters, J. Bacteriol. 193 (2011) 887. [86] D.J. Richardson, B.C. Berks, D.A. Russell, S. Spiro, C.J. Taylor, Cell. Mol. Life Sci. 58 (2001) 165. [87] S. Teraguchi, T.C. Hollocher, J. Biol. Chem. 264 (1989) 1972. [88] C.S. Zhang, T.C. Hollocher, A.F. Kolodziej, W.H. Orme-Johnson, J. Biol. Chem. 266 (1991) 2199. [89] C. Zhang, A.M. Jones, T.C. Hollocher, Biochem. Biophys. Res. Commun. 187 (1992) 135. [90] C. Mukonoweshuro, T.C. Hollocher, Arch. Biochem. Biophys. 306 (1993) 195. [91] T. Matsubara, H. Iwasaki, J. Biochem. 71 (1972) 747. [92] W.G. Zumft, T. Matsubara, FEBS Lett. 148 (1982) 107. [93] S. Ferretti, J.G. Grossmann, S.S. Hasnain, R.R. Eady, B.E. Smith, Eur. J. Biochem. 259 (1999) 651. [94] S.W. Snyder, T.C. Hollocher, J. Biol. Chem. 262 (1987) 6515. [95] C.L. Hulse, B.A. Averill, Biochem. Biophys. Res. Commun. 166 (1990) 729. [96] C.K. SooHoo, T.C. Hollocher, J. Biol. Chem. 266 (1991) 2203. [97] B.C. Berks, D. Baratta, J. Richardson, S.J. Ferguson, Eur. J. Biochem. 212 (1993) 467. [98] U.H. Hole, K.U. Vollack, W.G. Zumft, E. Eisenmann, R.A. Siddiqui, B. Friedrich, P.M. Kroneck, Arch. Microbiol. 165 (1996) 55. [99] M. Prudencio, A.S. Pereira, P. Tavares, S. Besson, I. Cabrito, K. Brown, B. Samyn, B. Devreese, J. Van Beeumen, F. Rusnak, G. Fauque, J.J. Moura, M. Tegoni, C. Cambillau, I. Moura, Biochemistry 39 (2000) 3899. [100] K. Yamaguchi, A. Kawamura, H. Ogawa, S. Suzuki, J. Biochem. 134 (2003) 853. [101] S. Dell’acqua, S.R. Pauleta, I. Moura, J.J. Moura, J. Biol. Inorg. Chem. 16 (2011) 183. [102] P. Chen, S.I. Gorelsky, S. Ghosh, E.I. Solomon, Angew. Chem. Int. Ed. Engl. 43 (2004) 4132. [103] T. Rasmussen, B.C. Berks, J.N. Butt, A.J. Thomson, Biochem. J 364 (2002) 807. [104] P. Chen, S. DeBeer George, I. Cabrito, W.E. Antholine, J.J. Moura, I. Moura, B. Hedman, K.O. Hodgson, E.I. Solomon, J. Am. Chem. Soc. 124 (2002) 744. [105] P. Chen, I. Cabrito, J.J. Moura, I. Moura, E.I. Solomon, J. Am. Chem. Soc. 124 (2002) 10497. [106] M.L. Alvarez, J. Ai, W. Zumft, J. Sanders-Loehr, D.M. Dooley, J. Am. Chem. Soc. 123 (2001) 576. [107] J.A. Farrar, A.J. Thomson, M.R. Cheesman, D.M. Dooley, W.G. Zumft, FEBS Lett. 294 (1991) 11. [108] J.A. Farrar, W.G. Zumft, A.J. Thomson, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 9891. [109] H. Beinert, Eur. J. Biochem. 245 (1997) 521. [110] P.M. Kroneck, W.A. Antholine, J. Riester, W.G. Zumft, FEBS Lett. 242 (1988) 70. [111] W.E. Antholine, D.H. Kastrau, G.C. Steffens, G. Buse, W.G. Zumft, P.M. Kroneck, Eur. J. Biochem. 209 (1992) 875. [112] C.L. Coyle, W.G. Zumft, P.M. Kroneck, H. Korner, W. Jakob, Eur. J. Biochem. 153 (1985) 459. [113] V.S. Oganesyan, T. Rasmussen, S. Fairhurst, A.J. Thomson, Dalton Trans. (2004) 996. [114] S. Dell’acqua, S.R. Pauleta, J.J.G. Moura, I. Moura, Philos. Trans. R. Soc. Lond. B: Biol. Sci. 367 (2012) 1204. [115] K. Brown, M. Tegoni, M. Prudencio, A.S. Pereira, S. Besson, J.J. Moura, I. Moura, C. Cambillau, Nat. Struct. Biol. 7 (2000) 191. [116] K. Brown, K. Djinovic-Carugo, T. Haltia, I. Cabrito, M. Saraste, J.J. Moura, I. Moura, M. Tegoni, C. Cambillau, J. Biol. Chem. 275 (2000) 41133. [117] T. Rasmussen, B.C. Berks, J. Sanders-Loehr, D.M. Dooley, W.G. Zumft, A.J. Thomson, Biochemistry 39 (2000) 12753. [118] K. Fujita, J.M. Chan, J.A. Bollinger, M.L. Alvarez, D.M. Dooley, J. Inorg. Biochem. 101 (2007) 1836. [119] S. Ghosh, S.I. Gorelsky, S.D. George, J.M. Chan, I. Cabrito, D.M. Dooley, J.J. Moura, I. Moura, E.I. Solomon, J. Am. Chem. Soc. 129 (2007) 3955. [120] J.M. Charnock, A. Dreusch, H. Korner, F. Neese, J. Nelson, A. Kannt, H. Michel, C.D. Garner, P.M. Kroneck, W.G. Zumft, Eur. J. Biochem. 267 (2000) 1368. [121] T. Haltia, K. Brown, M. Tegoni, C. Cambillau, M. Saraste, K. Mattila, K. DjinovicCarugo, Biochem. J 369 (2003) 77. [122] K. Paraskevopoulos, S.V. Antonyuk, R.G. Sawers, R.R. Eady, S.S. Hasnain, J. Mol. Biol. 362 (2006) 55. [123] J.R. Winkler, Curr. Opin. Chem. Biol. 4 (2000) 192. [124] A. Pomowski, W.G. Zumft, P.M. Kroneck, O. Einsle, Nature 477 (2011) 234. [125] A. Pomowski, W.G. Zumft, P.M. Kroneck, O. Einsle, Acta Crystallogr. F 66 (2010) 1541.

S.R. Pauleta et al. / Coordination Chemistry Reviews 257 (2013) 332–349 [126] K. Mattila, T. Haltia, Proteins 59 (2005) 708. [127] K. Jones, in: J.C. Bailar, H.J. Emelöus, R. Nyholm, A.F. Trottman-Dickenson (Eds.), Comprehensive Inorganic Chemistry, Pergamon Press, Oxford, UK, 1975, p. 147. [128] G.A. Vaughan, P.B. Rupert, G.L. Hillhouse, J. Am. Chem. Soc. 109 (1987) 5538. [129] F. Bottomley, I.J.B. Lin, M. Mukaida, J. Am. Chem. Soc. 102 (1980) 5238. [130] F. Bottomley, D.E. Paez, P.S. White, J. Am. Chem. Soc. 104 (1982) 5651. [131] F. Bottomley, G.O. Egharevba, I.J.B. Lin, P.S. White, Organometallics 4 (1985) 550. [132] J.T. Groves, J.S. Roman, J. Am. Chem. Soc. 117 (1995) 5594. [133] D.J. Berg, C.J. Burns, R.A. Andersen, A. Zalkin, Organometallics 8 (1989) 1865. [134] A.W. Kaplan, R.G. Bergman, Organometallics 16 (1997) 1106. [135] A.W. Kaplan, R.G. Bergman, Organometallics 17 (1998) 5072. [136] P.T. Matsunaga, G.L. Hillhouse, A.L. Rheingold, J. Am. Chem. Soc. 115 (1993) 2075. [137] K. Koo, G.L. Hillhouse, A.L. Rheingold, Organometallics 14 (1995) 456. [138] C.E. Laplaza, A.L. Odom, W.M. Davis, C.C. Cummins, J.D. Protasiewicz, J. Am. Chem. Soc. 117 (1995) 4999. [139] A.R. Johnson, W.M. Davis, C.C. Cummins, S. Serron, S.P. Nolan, D.G. Musaev, K. Morokuma, J. Am. Chem. Soc. 120 (1998) 2071. [140] N.A. Piro, M.F. Lichterman, W.H. Harman, C.J. Chang, J. Am. Chem. Soc. 133 (2011) 2108. [141] J.N. Armor, H. Taube, J. Am. Chem. Soc. 91 (1969) 6874. [142] J.N. Armor, H. Taube, J. Am. Chem. Soc. 92 (1970) 2560. [143] J.N. Armor, H. Taube, J. Am. Chem. Soc. 93 (1971) 6476. [144] A.A. Diamantis, G.J. Sparrow, J. Chem. Soc., D. 13 (1970) 819. [145] A.A. Diamantis, G.J.A. Sparrow, M.R.A. Snow, T.R.A. Norman, Aust. J. Chem. 28 (1975) 1231. [146] F. Paulat, T. Kuschel, C. Nather, V. Praneeth, O. Sander, N. Lehnert, Inorg. Chem. 43 (2004) 6979. [147] C.B. Pamplin, E.S.F. Ma, N. Safari, S.J. Rettig, B.R. James, J. Am. Chem. Soc. 123 (2001) 8596. [148] D.F.T. Tuan, R. Hoffmann, Inorg. Chem. 24 (1985) 871. [149] J.-H. Lee, M. Pink, J. Tomaszewski, H. Fan, K.G. Caulton, J. Am. Chem. Soc. 129 (2007) 8706. [150] J.T. York, I. Bar-Nahum, W.B. Tolman, Inorg. Chim. Acta 361 (2008) 885. [151] R. Sarangi, J.T. York, M.E. Helton, K. Fujisawa, K.D. Karlin, W.B. Tolman, K.O. Hodgson, B. Hedman, E.I. Solomon, J. Am. Chem. Soc. 130 (2008) 676. [152] I. Bar-Nahum, A.K. Gupta, S.M. Huber, M.Z. Ertem, C.J. Cramer, W.B. Tolman, J. Am. Chem. Soc. 131 (2009) 2812. [153] R.T. St John, T.C. Hollocher, J. Biol. Chem. 252 (1977) 212. [154] P.R. Alefounder, S.J. Ferguson, Biochem. Biophys. Res. Commun. 104 (1982) 1149.

349

[155] J.K. Kristjansson, T.C. Hollocher, J. Biol. Chem. 255 (1980) 704. [156] S. Ghosh, S.I. Gorelsky, P. Chen, I. Cabrito, J.J. Moura, I. Moura, E.I. Solomon, J. Am. Chem. Soc. 125 (2003) 15708. [157] J.M. Chan, J.A. Bollinger, C.L. Grewell, D.M. Dooley, J. Am. Chem. Soc. 126 (2004) 3030. [158] K. Fujita, D.M. Dooley, Inorg. Chem. 46 (2007) 613. [159] S.I. Gorelsky, S. Ghosh, E.I. Solomon, J. Am. Chem. Soc. 128 (2006) 278. [160] S. Dell’acqua, S.R. Pauleta, E. Monzani, A.S. Pereira, L. Casella, J.J. Moura, I. Moura, Biochemistry 47 (2008) 10852. [161] D.J. Richardson, L.C. Bell, A.G. McEwan, J.B. Jackson, S.J. Ferguson, Eur. J. Biochem. 199 (1991) 677. [162] M. Itoh, K. Matsuura, T. Satoh, FEBS Lett. 251 (1989) 104. [163] T. Rasmussen, T. Brittain, B.C. Berks, N.J. Watmough, A.J. Thomson, Dalton Trans. 21 (2005) 3501. [164] K. Fujita, F. Ijima, Y. Obara, M. Hirasawa, D.E. Brown, T. Kohzuma, D.M. Dooley, J. Biol. Inorg. Chem. 14 (Suppl. 1) (2009) S11. [165] C.S. Zhang, T.C. Hollocher, Biochim. Biophys. Acta 1142 (1993) 253. [166] S. Dell’Acqua, S.R. Pauleta, P.M. Paes de Sousa, E. Monzani, L. Casella, J.J. Moura, I. Moura, J. Biol. Inorg. Chem. 15 (2010) 967. [167] S. Iwata, C. Ostermeier, B. Ludwig, H. Michel, Nature 376 (1995) 660. [168] T. Tsukihara, H. Aoyama, E. Yamashita, T. Tomizaki, H. Yamaguchi, K. Shinzawa-Itoh, R. Nakashima, R. Yaono, S. Yoshikawa, Science 272 (1996) 1136. [169] V.R.I. Kaila, M.I. Verkhovsky, M. Wikström, Chem. Rev. 110 (2010) 7062. [170] C.C. Moser, C.C. Page, P.L. Dutton, Philos. Trans. R. Soc. Lond. B 361 (2006) 1295. [171] A. Harrenga, H. Michel, J. Biol. Chem. 274 (1999) 33296. [172] J.R. Winkler, B.G. Malmström, H.B. Gray, Biophys. Chem. 54 (1995) 199. [173] L.M. Geren, J.R. Beasley, B.R. Fine, A.J. Saunders, S. Hibdon, G.J. Pielak, B. Durham, F. Millett, J. Biol. Chem. 270 (1995) 2466. [174] M. Ruitenberg, A. Kannt, E. Bamberg, B. Ludwig, H. Michel, K. Fendler, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 4632. [175] K. Wang, L. Geren, Y. Zhen, L. Ma, S. Ferguson-Miller, B. Durham, F. Millett, Biochemistry 41 (2002) 2298. [176] M.F. Lucas, D.L. Rousseau, V. Guallar, Biochim. Biophys. Acta 1807 (2011) 1305. [177] M. Kukimoto, M. Nishiyama, T. Ohnuki, S. Turley, E.T. Adman, S. Horinouchi, T. Beppu, Protein Eng. 8 (1995) 153. [178] K. Kataoka, K. Yamaguchi, M. Kobayashi, T. Mori, N. Bokui, S. Suzuki, J. Biol. Chem. 279 (2004) 53374. [179] H. Witt, F. Malatesta, F. Nicoletti, M. Brunori, B. Ludwig, J. Biol. Chem. 273 (1998) 5132. [180] A.J. Thomson, G. Giannopoulos, J. Pretty, E.M. Baggs, D.J. Richardson, Philos. Trans. R. Soc. Lond. B: Biol. Sci. 367 (2012) 1157.