Summary
Enzymes that activate dioxygen at carboxylate-bridged non-heme diiron clusters residing within ferritin-like, four-helix-bundle protein architectures have crucial roles in, among other processes, the global carbon cycle (e.g., soluble methane monooxygenase), fatty acid biosynthesis [plant fatty acyl-acyl carrier protein (ACP) desaturases], DNA biosynthesis [the R2 or β2 subunits of class Ia ribonucleotide reductases (RNRs)], and cellular iron trafficking (ferritins). Classic studies on class Ia RNRs showed long ago how this obligatorily oxidative di-iron/O2 chemistry can be used to activate an enzyme for even a reduction reaction, and more recent investigations of class Ib and Ic RNRs, coupled with earlier studies on dimanganese catalases, have shown that members of this protein family can also incorporate either one or two Mn ions and use them in place of iron for redox catalysis. These two strategies – oxidative activation for non-oxidative reactions and use of alternative metal ions – expand the catalytic repertoire of the family, probably to include activities that remain to be discovered. Indeed, a recent study has suggested that fatty aldehyde decarbonylases (ADs) from cyanobacteria, purported to catalyze a redox-neutral cleavage of a Cn aldehyde to the Cn−1 alkane (or alkene) and CO, also belong to this enzyme family and are most similar in structure to two other members with heterodinuclear (Mn-Fe) cofactors. Here, we first briefly review both the chemical principles underlying the O2-dependent oxidative chemistry of the “classical” di-iron-carboxylate proteins and the two aforementioned strategies that have expanded their functional range, and then consider what metal ion(s) and what chemical mechanism(s) might be employed by the newly discovered cyanobacterial ADs.
Di-iron-carboxylate oxidases and oxygenases: operational principles, chemical limitations, and strategies for making them more versatile
Non-heme di-iron oxidases and oxygenases, designated by Nordlund and Eklund as “diiron-carboxylate” proteins [1], include eukaryotic and bacterial ferritins [2]; the β2 (R2) subunits of class Ia ribonucleotide reductases (Ia-RNR-β2s) [3]; the hydroxylase components of bacterial multicomponent monooxygenases (BMMs), such as soluble methane monooxygenase (sMMO) [4,5], toluene/o-xylene monooxygenase (ToMO) [6], phenol hydroxylase (PhOH) [7], and alkene monooxygenase (AMO) [8]; plant soluble fatty acyl–acyl carrier protein (ACP) desaturases [9]; and the nitro-group-installing bacterial N-oxygenase AurF in the aureothin biosynthetic pathway [10]. Discovery of new members continues apace [11]. These proteins possess adjacent (~ 4 Å apart) sites to bind two metal ions in a coordination environment rich in carboxylate residues (most often four glutamates) and in most cases having one histidine ligand per metal sub-site. A conserved “four-helix-bundle” architecture provides the ligands. Additional helices provide structural stability, and in some cases the dimetal-binding subunits associate with additional polypeptides (as in several BMMs, which are α2β2γ2 hexameric complexes). There have been a number of reviews of structural and evolutionary aspects of this protein family (e.g. [1,12]); our brief account will focus on their diverse chemistry.
Each of the aforementioned proteins functions by reaction of its carboxylate-bridged di-iron cluster with dioxygen. Mechanistic studies have suggested that [μ-(hydro)peroxo]-Fe2(III/III) intermediates, formed by the bridging, oxidative addition of O2 to the Fe2(II/II) cofactors, are common to most, if not all, of the reactions [13–20]. However, significant differences in the spectroscopic characteristics of the intermediates suggest potentially important structural differences among them [18,19,21]. These differences, coupled with divergent downstream control of the reaction pathways by the individual proteins, lead to an impressive array of outcomes (Scheme 1). In the ferritins, protonation and dissociation of the peroxide prevents either further oxidation of the iron ions or oxidation of a substrate [22]. This “null” outcome suits the function of the reaction, which is simply to oxidize the iron to initiate its “mineralization” and storage. In the Ia-RNR-β2s, O-O-bond scission coupled with injection of an electron (by a specific, outer-sphere, electron-relay network involving a tryptophan residue that undergoes transient oxidation) [23] lead to formation of an Fe2(III/IV) complex, X [24–27], which then oxidizes a nearby tyrosine residue by one electron to a stable radical (Y•) [24]. This step activates the Ia-RNR-β2 for catalysis of nucleotide reduction with its partner subunit, α2, as discussed below [28]. In the BMM, sMMO, non-reductive O-O-bond scission leads to formation of the Fe2(IV/IV) complex, Q [13,29,30], which hydroxylates methane (as well as other hydrocarbons). Although not yet detected, an Fe2(IV/IV) complex is also a likely candidate to initiate fatty acid dehydrogenation (by abstracting hydrogen) in the soluble acyl-ACP desaturases from plants [9]. In the BMM ToMO [18], in the nitro-group-installing N-oxygenase AurF [19,31], and probably in other arene-, olefin-, and N-oxygenating systems (e.g., PhOH and AMO), nucleophilic attack by π or non-bonding electrons directly on the (hydro)peroxide complex results in hydroxylation of the substrates.
Scheme 1.
Diverse reaction pathways of the peroxo-Fe2(III/III) intermediates in reactions of the ferritin-like di-iron-carboxylate proteins. For simplicity, we depict the peroxo-Fe2(III/III) intermediate as having a μ-1,2 bridging mode, although many other binding modes have been discussed. Aromatic hydroxylation is believed to proceed by a mechanism similar to that labeled as —epoxidation.
Although the diversity of the above reaction outcomes is considerable, spanning zero-electron (peroxide release) [22], one-electron (Y• formation) [3] and three chemically distinct two-electron oxidations (hydroxylation [4–6], dehydrogenation [9], and epoxidation [8]), it would appear that the general strategy has important inherent limitations. With the exception of the recently discovered four-electron N-oxidation of an aryl hydroxylamine to the corresponding nitroaryl compound by AurF [31], each of the reactions generates a stable Fe2(III/III) form of the cofactor. Thus, (except for the AurF and ferritin reactions) catalysis requires reduction of the cofactor by two electrons back to the O2-reactive Fe2(II/II) state. The source of the required electrons varies among the different systems. In the BMMs, a dedicated reductase protein, containing one-electron-mediating cofactors [iron-sulfur (FeS) clusters and flavin], accepts a hydride from a reduced nicotinamide co-substrate and delivers the two electrons, one at a time, to the di-iron center of the hydroxylase component [4,5]. In the acyl-ACP desaturases, a small FeS-cluster-containing protein, ferredoxin, delivers the electrons one at a time and is in turn reduced by ferredoxin reductase with electrons from NADPH [9]. In the Ia-RNR-β2s, in which the Y• can undergo adventitious reduction to inactivate the protein, a flavin/ferric ion reductase was initially shown to use NADPH to reduce the Fe2(III/III) cluster to the Fe2(II/II) state [32], allowing for regeneration of the Y•. However, more recent studies have suggested that the ferredoxin-like Fe2S2 protein, YfaE, is more efficient at this reduction process and is probably the physiologically relevant Y•-regenerating factor [33].
The modus operandi of these proteins – reaction of the Fe2(II/II) cluster with O2 to form a [μ-(hydro)peroxo)]-Fe2(III/III) complex, reaction of this intermediate or its successor with the oxidation target to generate the Fe2(III/III) form of the cofactor, and reduction of the Fe2(III/III) state back to Fe2(II/II) by the relevant reducing system – would thus seem to limit them to exclusively oxidative outcomes, in which at least two of the four electrons required for O2 reduction are provided by reducing co-substrates and at most two are extracted from the actual oxidation target (with the one exception of the AurF reaction noted above). Two distinct variations that expand their catalytic repertoire have been discovered. The first one was originally exemplified by the class Ia RNRs. As noted, the di-iron-O2 chemistry in the β2 subunits of these enzymes targets a conserved tyrosine residue near the cluster for one-electron oxidation to a stable Y•. Once formed, the Y• is then used catalytically to generate protein and substrate radicals in the partner subunit, α2, leading to a two-electron reduction of the nucleoside 5′-diphosphate substrate to the corresponding 2′-deoxyribonucleoside 5′-diphosphate product [28]. The proximal source of the electrons for this reduction is two cysteine residues in α2, which are temporarily oxidized to a disulfide [3]. Thus, the oxidative di-iron chemistry activates the protein by producing a stable, yet potent one-electron oxidant that is subsequently used to promote the difficult reduction reaction by providing the means to generate a substrate radical.
The second variation expanding the repertoire of this class of proteins involves the use of metal ions other than iron. It was initially exemplified by the bacterial dimanganese catalases, also known as “pseudocatalases” [34,35]. Although not yet thoroughly understood, the catalytic cycle of these enzymes is thought to involve oxidation of the Mn2(II/II) form of the cofactor to Mn2(III/III) by H2O2 (which is reduced to two equiv H2O) and subsequent reduction of the Mn2(III/III) form to Mn2(II/II) by a second equivalent of H2O2 (which is oxidized to O2) [35,36]. It is difficult to imagine that a di-iron cofactor could function analogously. The Fe2(II/II) clusters can readily reduce H2O2, undergoing conversion to the corresponding Fe2(III/III) complexes [37], but reduction of the latter state by H2O2 has, to our knowledge, not been demonstrated and seems unlikely to be capable of supporting an efficient catalase cycle due to the stability of the Fe2(III/III) state. In principle, cycling between III/III and IV/IV states might be more efficient, but, again, the Fe2(III/III) complexes have generally shown sluggish reactivity toward oxidation by H2O2 [38]. It thus appears that the catalases’ use of Mn represents the selection of a transition metal that has more appropriate redox potentials (and perhaps also Lewis acidity and coordination dynamical properties) for the required transformation.
These two distinct versatility-enhancing variations converge in the recently elucidated case of the class Ic RNR from Chlamydia trachomatis (Ct) [39–41]. The Ic-RNR-β2 protein lacks the conserved tyrosine residue and thus cannot form the essential Y• [42]. Although it remains to be confirmed that the enzyme functions this way in vivo, in vitro studies suggest that the initiating amino acid radical is functionally replaced by the Mn(IV) ion of a stable heterodinuclear Mn(IV)/Fe(III) cluster [39–41,43]. The protein can bind Mn(II) in place of one of the two Fe(II) ions normally taken up by the Ia-RNR-β2s. Reaction of the Mn(II)/Fe(II) cofactor with O2 generates a Mn(IV)/Fe(IV) intermediate [44], which is reduced by one-electron at the Fe site via an electron-relay apparatus involving a tryptophan and tyrosine residue [45]. The Mn(IV) site of the resulting stable Mn(IV)/Fe(III) cofactor accepts an electron from the α2 subunit (in the same manner as the Y• in the Ia-RNRs), forming a transient Mn(III)/Fe(III) cluster [39–41]. Again here, the dimetal-O2 chemistry generates a stable, yet potent, oxidant that is used catalytically for substrate radical production. The difference is that the oxidant resides on the dimetal cluster rather than on an adjacent amino acid. It would appear that replacement of the one Fe ion by Mn is again necessary to tune the reduction potential, in this case to permit the high-valent state to be stabilized within the interior of the protein while remaining reactive enough to support formation of a substrate radical. The corresponding homodinuclear Fe2(III/IV) complex (equivalent to the X intermediate that generates the Y• in the Ia-RNR-β2s) might, in principle, be capable of this function, and, indeed, was initially proposed to be the active cofactor [46]. However, it was subsequently shown to be incompetent or, at best, much less competent at supporting nucleotide reduction by α2 [39]. Part of the reason seems to be that the Fe2(III/IV) complex is too strongly oxidizing: it is stable for only seconds or (at most) minutes, whereas the Mn(IV)/Fe(III) complex is stable for many hours. A recent computational study validated this intuition by suggesting that the Mn(IV)/Fe(III) → Mn(III)/Fe(III) reduction potential should nearly match that for the Y• → Y reduction in the Ia-RNR-β2s, whereas the Fe2(III/IV) → Fe2(III/III) potential is considerably higher and probably too high to support controlled, efficient, reversible generation of amino acid and substrate radicals in α2 [47].
It is noteworthy that, after years of controversy [48,49], it has finally been established that the class Ib RNR-β2 proteins (designated NrdFs) use dimanganese clusters to generate their Y•s [50,51]. The data suggest that the Mn2(II/II) form of this cofactor is not reactive toward O2 (as was also shown to be the case for the Mn2(II/II) form of the Ct Ic-RNR-β2), thus requiring that an activation protein, a flavoprotein called NrdI that can bind tightly to NrdF (the Ib-RNR-β2), supply a reduced form of O2 (probably HO2−) to support the oxidation chemistry necessary to generate the Y• [50,52]. In this system, the rationale for the metal substitution appears to be more biological than chemical. In Escherichia coli (for example), the Ib-RNRs are functional only under conditions of iron limitation or oxidative stress (which is known to limit iron bioavailability) [53,54]. Thus, Mn is used when Fe is unavailable. From a chemical-mechanistic standpoint, the Fe2-Y•-dependent Ia-RNRs and Mn2-Y•-dependent Ib-RNRs seem to be essentially analogous. This topic is discussed in depth by Cotruvo and Stubbe in this issue.
The Ct Ic-RNR-β2 represented the first examples of (i) a Mn/Fe protein redox cofactor of any kind and (ii) a heterodinuclear redox cofactor of any kind in this class of proteins. The results hinted at the possibility that the combination of oxidative activation for non-oxidative reactions and use of metal ions other than iron might permit a range of catalytic reactions much broader than had been recognized to that point. Soon after the Ct RNR work, the structure of a Mycobacterium tuberculosis (Mt) protein (Rv0233; Protein Data Bank accession code 3EE4) potentially important in Mt virulence revealed that it also belongs to the structural family and contains a Mn/Fe cluster [55]. The protein was found to bind a ligand, assigned as myristate, with its carboxylate coordinated directly to the cluster in a bridging mode and the linear (putatively) alkyl chain projecting away from the cofactor through a hydrophobic tunnel in the protein (Figure 1). The presence of this ligand, combined with the observation of a novel (two-electron) oxidative modification involving covalent coupling of tyrosine and valine residues via the Tyr phenolic oxygen and Val methine carbon atoms, were the basis for the tentative assignment of the protein as an oxidase: no catalytic or stoichiometric oxidative chemistry was directly demonstrated in vitro. Nevertheless, the study hinted at potentially new fatty acid chemistry mediated by a novel member of this protein family and possibly combining the use of a heterodinuclear (Mn/Fe) cofactor with oxidative activation (the novel crosslink).
Figure 1.
Comparison of the dimetal-carboxylate sites in Pm AD (pdb code: 2OC5) and the Mn/Fe-containing putative oxidase from Mycobacterium tuberculosis (pdb code: 3EE4). Amino acid residues of Pm AD and the Mt protein are colored in green and purple, respectively. The non-protein ligands, tentatively assigned as stearate and myristate, are shown in yellow and cyan, respectively. The Fe and Mn ions are shown in orange and purple, respectively. The oxidized crosslink between Val71 and Tyr162 in the Mt protein is shown in gray. Amino acid numbers given first refer to the Pm AD, while those in parentheses refer to the Mt protein.
Discovery of a fundamentally new type of reactivity for a member of this protein family: alkane synthesis in cyanobacteria
The very recent discovery of a two-step pathway in cyanobacteria for conversion of saturated fatty acids to alkanes and unsaturated fatty acids to alkenes suggests an astounding further expansion of the catalytic repertoire of this protein family into a fundamentally different type of chemistry [56]. The existence and utilization of alkanes and alkenes has been best appreciated in higher eukaryotes [57], where their hydrophobic nature is exploited to prevent desiccation in plants [58] and insects [59] and to provide waterproofing in birds [60]. They also serve as the structural core of certain insect pheromones [61] and the myelin sheath of peripheral nerves in vertebrates [62]. Although alkane production in diverse microorganisms has been reported—with most convincing observations in cyanobacteria—their biological roles and mechanism of formation remain obscure [56,63]. Taking advantage of their phylogenetic homogeneity and the availability of a large number of sequenced cyanobacterial genomes, Schirmer, et al. used a clever subtractive genomics approach to gain insight into this novel biochemical transformation [56]. Among eleven strains of cyanobacteria with genomes of known sequence, ten were found to produce alkanes—predominantly heptadecane and pentadecane, among others—when cultured under defined conditions, while one did not. The genomes of the ten alkane-producing strains were then intersected to find genes common to all, and the genome of Synechococcus sp. PCC7002—the cyanobacterial strain not found to produce alkanes—was then subtracted from this intersected genome. Seventeen genes common only to the alkane-producing strains were identified, of which ten had assigned functions. Of the remaining seven, one (orf1594 of Synechococcus elongatus PCC7942) encoded a protein with sequence similarity to short-chain fatty acid dehydrogenases, while another (orf1593 of Synechococcus elongatus PCC7942) encoded a protein that shared significant sequence similarity with ferritin-like proteins containing carboxylate-bridged dimetal clusters [64].
Earlier genetic and biochemical studies in plants had indicated that, in these higher eukaryotes, alkanes are produced in a two-step process, involving initial reduction of a long-chain fatty acid to a fatty aldehyde and subsequent decarbonylation of the fatty aldehyde to the corresponding alkane or alkene. In fact, CER1, a protein involved in wax formation in Arabidopsis thaliana and identified as a possible AD [65], contains the histidine-rich sequences common to a family of integral membrane non-heme-diiron oxidases and oxygenases [66] that have functions that overlap considerably with those of the soluble ferritin-like diiron-carboxylate proteins. Consistent with this activity, stem wax of cer1 mutants of A. thaliana was found to lack alkanes but be especially rich in aldehydes [65]. This finding, and the observation of fatty aldehyde decarbonylation in other organisms, allowed formulation of the hypothesis that orf1594 of S. elongatus encodes a protein that catalyzes reduction of activated fatty acids to fatty aldehydes, while orf1593 encodes a fatty aldehyde decarbonylase [56]. In agreement with this hypothesis, the co-expression of these two genes in E. coli conferred to the heterologous host the ability to produce fatty aldehydes and alkanes. When the gene encoding orf1593 was omitted, the ability to make alkanes was abrogated; however, the bacterium still produced fatty aldehydes. When the gene encoding orf1594 was omitted, neither fatty aldehydes nor alkanes were produced. Moreover, when orthologs of S. elongatus PCC7942 orf1593 and orf1594 in Synechocystis sp. PCC6803 were disrupted by kanamycin-resistance cassettes, this alkane producing organism was no longer capable of this activity, indicating that these two orfs are necessary and sufficient to produce alkanes from fatty acids.
Subsequent in vitro biochemical studies verified the roles of orf1594 and orf1593. Because bacterial fatty acids are biosynthesized or metabolized while attached covalently in thioester linkages to acyl carrier protein (ACP) or coenzyme A (CoA), both fatty acyl ACPs (oleoyl–ACP) and CoAs (oleoyl–CoA) were evaluated as substrates for orf1594, by assessment of their reduction to fatty aldehydes with concomitant production of NADP+ from NADPH. Although Vmax for oleoyl–CoA was 2.6-fold greater than that for oleoyl–ACP, Vmax/KM for the latter substrate was six-fold greater than that for the former, suggesting that, in vivo, orf1594 is an acyl–ACP reductase rather than an acyl–CoA reductase. An ortholog of orf1593 from Nostoc punctiforme (Np) was used to study the in vitro reaction of the putative AD, because it was shown to yield the highest level of alkanes among the eleven tested by heterologous expression in E. coli [56]. A variant Np AD containing an N-terminal hexahistidine tag was overproduced in E. coli, isolated by immobilized metal affinity chromatography, and shown to catalyze scission of the C1–C2 bond of octadecanal, affording heptadecane. Interestingly, the co-product, proposed to be carbon monoxide (CO), was never identified. Formation of octadecanal required a reducing system—similar to that required by other diiron-carboxylate oxidases and oxygenases—which could be satisfied by NADPH along with commercially available ferredoxin and ferredoxin reductase from spinach [56] (hereafter, the reducing system is denoted N/F/FR).
Structural similarities between the Prochlorococcus marinus MIT 9313 (Pm) AD and a Mn/Fe fatty-acid-binding putative oxidase from Mycobacterium tuberculosis
The amino acid sequences of the cyanobacterial ADs suggested that they belong to the ferritin-like dimetal-caboxylate protein family [56]. The structure of the AD from Prochlorococcus marinus MIT9313 (Pm), which was solved by the Joint Center of Structural Genomics before the function of the protein was known, confirmed this deduction. Schirmer and co-workers noted [56] that the structure of the Pm AD is similar to that of the Ia-RNR-β2 protein from E. coli [67]. However, the Pm AD structure is actually even more similar to that of the Mn/Fe-dependent Ic-RNR-β2 from Ct, including matches at two sequence positions that positively distinguish Ia β2 proteins from their Ic counterparts [68]. As noted above, the Ic-β2s lack the Y•-harboring tyrosine residue of the Ia-β2s, having phenylalanine (F) in its place [42,68,69]. In addition, the unique aspartate (D) ligand to Fe1 that is conserved in the Ia-β2s is the more usual glutamate (E) in the Ic-β2s [68,69]. The Pm AD protein, and by sequence alignment, the other cyanobacterial ADs, have the F and E residues of the Ic-RNR-β2s at the corresponding two positions [56]. Moreover, the Pm AD is structurally most similar to the aforementioned Mn/Fe-containing putative fatty acid oxidase from Mt [55,68,70]. Figure 1 shows an illuminating comparison of the active sites of these proteins. Although not discussed by Schirmer, et al. [56], the structure of the Pm AD, like that of the Mt putative oxidase, also apparently has a non-protein ligand bound to the di-metal cluster. This ligand is, most likely, also a fatty acid (most simply, stearate), despite the fact the atoms that would be carbons 1–3 and 7–18 are annotated as oxygen atoms in the PDB file. Interestingly, carbons 1–9 of the putative myristate ligand in the Mt protein and the putative stearate ligand in Pm AD nearly overlay. Thus, the structural similarities between the Mn/Fe-containing putative oxidase and Pm AD include not only the protein and dimetal-site architecture but also the presence of a hydrophobic tunnel leading to the dimetal site, which can be occupied by a (presumed) fatty acid.
Conundrum raised by the structure of the cyanobacterial AD and the reported biochemical observations
The exciting discovery and initial characterization of the ADs raised an obvious conundrum. Their purported catalytic reaction, conversion of the Cn aldehyde to Cn−1 alk(a/e)ne and CO (Scheme 2, lower left pathway), is not a net redox reaction, although one may envisage it as involving, formally, one-electron oxidation of C1 balanced by one-electron reduction of C2. Paradoxically, the structure of the Pm ortholog places the cyanobacterial ADs in a structural family normally associated with oxidative chemistry. Moreover, the reported requirement for the N/F/FR reducing system could be rationalized most simply according to an oxidase/oxygenase-like catalytic cycle for the AD reaction, as outlined above for the BMMs and acyl-ACP desaturases. It is, therefore, crucial to recall that the C1-derived co-product of alkane formation was not identified. If the product is, as suggested, CO or, alternatively, formate (HCO2−, which has the same redox state as CO and would be the product of a formally hydrolytic reaction; Scheme 2, top left pathway), then perhaps the principle outlined above of oxidative activation for a non-oxidative reaction can resolve the conundrum. In this case, the N/F/FR system might be required only to mediate reactivation of the AD by reducing its cofactor to permit subsequent reaction with O2, analogously to the reactivation of the Ia-RNR-β2 proteins by YfaE discussed above. However, it is also conceivable that the AD reaction is authentically oxidative, generating the more oxidized co-product, CO2, from C1 of the aldehyde (Scheme 2, top right pathway). Consistent with this possibility, a previous study of an aldehyde-cleaving enzyme system from the common house fly suggested that it does, in fact, yield CO2 as the C1-derived co-product [61]. In this case, the N/F/FR system required for the activity of the cyanobacterial AD could be fulfilling the usual role of the analogous reducing systems in the BMMs and acyl-ACP desaturases. A variation of this reactivity would involve the use of O2 as a co-substrate, but with all four electrons required for its reduction being derived from the N/F/FR reducing system (i.e., a NADPH:aldehyde stoichiometry of 2). Such a mechanism would yield the same products as the redox-neutral reactions (alkane and either CO or formate). We refer to this as a cryptically redox pathway (Scheme 2, top). In principle, one might even consider C1-derived co-products more reduced than CO and formate, such as formaldehyde (H2CO, Scheme 2, bottom left pathway) or methanol (CH3OH). In this case, the N/F/FR system would serve to provide the electrons for the reductive C1-C2 scission, with the dimetal cofactor presumably serving as conduit. However, in light of the fact that this structural framework is invariably used for oxidative chemistry [3–5,9] (even in the aforementioned case of the class I RNRs, in which the oxidized cofactor subsequently initiates a reduction reaction at a remote site), we view this latter possibility as extremely unlikely and focus discussion on the more likely possibilities of redox-neutral and oxidative outcomes. Nevertheless, it should be abundantly clear that the most urgent need to advance our understanding of the AD reaction is identification of its co-product and definition of its redox balance.
Scheme 2.
Possible reaction pathways and co-products in conversion of fatty aldehydes to alk(a/e)nes by cyanobacterial aldehyde decarbonylases.
A number of previous studies on aldehyde decarbonylase enzyme systems might potentially provide clues as to the mechanism of the cyanobacterial AD reaction. Indeed, like the cyanobacterial AD, all other ADs studied to date appear to be metalloenzymes [58–61,63,71]. However, in no case has the nature of the metallocofactor or the mechanism of the reaction been definitively elucidated. This fact, combined with the aforementioned uncertainty in the redox balance of the cyanobacterial AD reaction, limits the value of clues that might be gleaned from these enzymatic precedents.
If one assumes that the C1-derived co-product is, as proposed, CO, then one is directed toward important mechanistic precedents from both inorganic/organometallic aldehyde decarbonylation catalysts and several organometallic-like enzyme systems that handle CO or alkyl groups in their catalytic cycles for clues as to possible mechanisms of the AD reaction and identities of the metal ions in the enzyme’s cofactor.
Potential mechanistic clues from inorganic/organometallic aldehyde decarbonylation catalysts
Ohno and Tsuji first showed that aldehyde decarbonylation can be carried out at a Rh(I) center [72]. Subsequent studies by Doughty and Pignolet improved the efficacy of the aldehyde decarbonylation reaction by using a chelating bis-phosphine ligand to coordinate the Rh(I) center [73]. Recent density functional theory (DFT) calculations by Madsen and co-workers provided further insight into the mechanism of aldehyde decarbonylation [74]. It its believed that the reaction entails three key chemical steps (Scheme 3A): (i) the oxidative addition of the aldehyde to the reduced Rh(I) center, yielding a Rh(III)-acyl-hydride complex; (ii) extrusion of CO to yield a Rh(III)-alkyl-CO-hydride complex; and (iii) reductive elimination of alkane to form the Rh(I)-CO complex. We hereafter refer to this three-step mechanism as the organometallic (OM) mechanism.
Scheme 3.
Proposed organometallic mechanisms for aldehyde decarbonylation. (A) Mechanism of the Rh(I)-catalyzed aldehyde decarbonylation, adapted from [74]; (B) Possible OM-like mechanism for aldehyde decarbonylation at a hypothetical dinuclear Fe/Ni cofactor in cyanobacterial ADs.
Other transition metal-catalyzed aldehyde decarbonylation reactions have also been reported in the literature. These have involved porphyrin complexes of Fe [75], Co [63], Cu [71], and Ru [76]. Belani et al. suggested that the mechanism of the Fe-porphyrin-catalyzed aldehyde decarbonylation is distinct from the OM mechanism, due to the fact that the porphyrin ligand precludes the necessary cisoid coordination of the substrate functional groups through the reaction [75]. However, recent work on methyl-coenzyme M-reductase (see below) suggests that two cisoid coordination sites are available at the Ni(I) center ligated by the tetradentate tetrapyrrole ligand, coenzyme F430 [77]. In the latter case, the reduced porphyrin ring is significantly ruffled, allowing for two ligands to bind on one face of the tetrapyrrole ligand.
Potential mechanistic clues from Ni-containing enzymes
While Rh is almost certainly not the metal used by the cyanobacterial ADs, it is interesting to note that chemical steps analolgous to those of the OM mechanism have been observed in the bifunctional Ni-containing enzyme acetyl-CoA synthase/CO dehydrogenase (ACS/CODH) and the Ni-containing enzyme methyl-coenzyme M-reductase (MCR). These analogies suggest that aldehyde decarbonylation could, in principle, be carried out via an OM-like mechanism at a Ni-containing active site.
MCR utilizes coenzyme F430 [78–80], a Ni(I)-hydroporphinoid prosthetic group, to convert coenzyme M (CH3-S-CoM) and coenzyme B (CoB-SH) to CH4 and the CoB-S-S-CoM heterodisulfide [81]. This reaction is the last step in methane formation in all methanogenic archaea [81]. Recent work by Thauer, Jaun, and co-workers elegantly demonstrated that this reaction is reversible and allows for the anaerobic oxidation of methane [77]. Although the mechanistic details remain to be elucidated, the key step in methane formation reaction is likely to be reductive elimination of CH4 from a Ni(III)-hydride-methyl intermediate [82]. The reverse of this reaction corresponds to the oxidative addition of CH4 to a coordinatively unsaturated Ni(I) center to yield to hypothetical Ni(III)-hydride-methyl intermediate. These reactions are remarkably similar to the proposed oxidative addition and reductive elimination steps of the OM mechanism for aldehyde decarbonylation.
The generation of acetyl-S-CoA by ACS is carried out at its A-cluster, a Fe4S4 cluster bridged by one of the cysteine cluster ligands to a Ni2 cluster [81,83,84]. It is proposed that the acetyl group to be incorporated into acetyl-CoA is generated on the Ni site proximal to the Fe4S4 cluster (Nip). This reaction is believed to entail (i) transfer of a methyl group from the corrinoid Fe/S protein to the Nip; (ii) binding of CO (generated at the enzyme’s Fe4S4Ni C-cluster (see below) and channeled to the A-cluster [85,86]) to Nip; (iii) insertion of CO into the Ni-CH3 bond; and (iv) transfer of the acetyl group to CoA-S−. The C-C-bond formation step in the ACS mechanism is the reverse of the CO-extrusion step proposed for aldehyde decarbonylation by the OM mechanism.
Interestingly, there are further possible parallels between steps catalyzed by ACS/CODH and the cyanobacterial AD. The CODH reaction at the Fe4S4Ni cluster is believed to involve (i) binding of CO to the Ni center, (ii) attack of a Fe-coordinated hydroxide on the C-atom of CO to yield CO2, and (iii) elimination of CO2, leading to the two-electron reduced form of the cluster [81]. Oxidation of the cluster by two electrons, accepted by ferredoxin, completes the cycle. As discussed below, oxidation of coordinated CO to CO2 could theoretically be part of the cyanobacterial AD cycle.
Hypothetical OM-like mechanism for CO production by the cyanobacterial ADs
Again assuming that CO is the co-product of the reaction, one can envisage an OM-like mechanism at the dinuclear cluster of the cyanobacterial AD, as shown in Scheme 3B. Such a mechanism would seem to favor Ni as the active metal (as shown), due to the extensive precedent of similar steps in Ni-containing enzymes (see above), although other metals could also be capable of carrying out these steps. Oxidative addition of the fatty aldehyde to the Ni(I) center would lead to a Ni(III)-acyl-hydride complex. Extrusion of CO would yield a Ni(III)-alkyl-hydride-CO complex. Finally, reductive elimination of the alkane product and dissociation of CO would complete the catalytic cycle. In this mechanism, there is no overt role for the second metal ion, but it might be required to stabilize the low-valent state of the active metal (e.g., redox tuning) or to control its coordination geometry. Another possible role of the second metal may be envisaged if one assumes that oxidation of CO is required for its release (discussed below).
Hypothetical “RNR-like” mechanism for the cyanobacterial ADs
A second type of mechanism for CO and alkane production can be formulated on the basis of the structural similarity of the cyanobacterial AD to the class Ic RNR-β2 from Ct and the mechanism by which the (μ-O)(μ-OH)-Mn(IV)/Fe(III) cofactor [87] functions in that system (discussed above). An analogous (μ-O)(μ-OH)-Mn(IV)/Fe(III) cluster in the AD could abstract a hydrogen atom from C1 of the fatty aldehyde substrate (homolytic bond dissociation energy ≈ 88 kcal/mol [88]) to yield an acyl radical and (μ-OH)2-Mn(III)/Fe(III) form of the cofactor. Subsequent C1-C2 fragmentation of the acyl radical would yield a primary alkyl radical and CO [89]. Re-abstraction of the hydrogen atom from the (μ-OH)2-Mn(III)/Fe(III) cluster by the alkyl radical would then yield the alkane product and regenerate the starting (μ-O)(μ-OH)-Mn(IV)/Fe(III) cluster. The cofactor would not necessarily have to be Mn(IV)/Fe(III) for the reaction to proceed accordingly: other clusters capable of cycling between a stable, yet strongly oxidizing, form and a protonated, one-electron-more-reduced form could also support such a mechanism. However, as argued above in discussion of the Ct RNR system, it seems unlikely that a di-iron cofactor could function in this capacity.
The other possible co-product of the non-redox C1-C2 scission of the fatty aldehyde is formate, HCO2−. An adaptation of the RNR-like C1-hydrogen-abstraction mechanism, involving removal by the cofactor of a proton and electron from the hydrated form of the fatty aldehyde substrate, provides a plausible route to this alternative co-product (Scheme 4A). Homolysis of the second O-H bond by the dinuclear cluster, either by hydrogen-atom abstraction or by concerted or step-wise proton and electron transfers, could again be followed by C1-C2 radical fragmentation, giving a formate ligand and an alkyl radical. Return of the hydrogen atom equivalent from the cluster to the alkyl radical would yield the alkane product and regenerate the starting form of the cofactor.
Scheme 4.
Possible mechanisms for aldehyde decarbonylation leading to generation of formate as co-product. (A) A hypothetical —RNR-like mechanism initiated by formation of a gem-diol-yl radical from the hydrated form of the fatty aldehyde. The active form of the dinuclear form of the cofactor is shown as (μ-O)-Mn(IV)/Fe(III), by analogy to the cofactor of the Ct Ic-RNR-β2 [39]. (B) A hypothetical —O2 activation— mechanism involving nucleophilic attack of a peroxide on the substrate carbonyl group to initiate O-O and C-C fragmentation. The O2-reactive form of the dinuclear cofactor is shown as Fe2(II/II), due to the extensive precedent for this cofactor to form peroxo intermediates of diverse reactivity.
Other possible strategies for generation of a substrate radical
The central principle of the above hypothetical RNR-like mechanisms for production of either CO or formate is the initial one-electron oxidation of the substrate to promote a radical-type C1-C2 fragmentation. One can envisage at least three variations on this strategy, involving different ways of generating the substrate radical. Firstly, the dimetal cluster could, as in the class Ia RNR-β2 proteins, generate a stable amino acid radical that would remove a hydrogen atom equivalent from the substrate, subsequently returning it to the alkyl radical to form the product. Recognizing this possibility, Schirmer, et al. substituted the two residues seemingly most likely to support an initiating radical, the two tyrosine residues in the vicinity of the dimetal center, with phenylalanine, and found that the variant proteins are still active [56]. Therefore, if an amino acid radical is involved as initiator, it must reside on a different residue. Secondly, a reduced metal ion (Mm+) of the cofactor could bind O2, forming a superoxo-M(m+1)+ complex that could transiently accept a hydrogen atom equivalent from the substrate and return it to the alkyl radical after C1-C2 fragmentation. In this case, O2 would be acting as a cofactor. This strategy would be directly analogous to the mechanism that the Mn-dependent oxalate decarboxylase is thought to employ [90]. Thirdly, the reduced form of the dimetal cofactor could activate O2 in each turnover. As noted, production of either CO or formate would make the AD reaction redox-neutral. Thus, the four electrons required for complete reduction of O2 would all be supplied by the N/F/FR system. A plausible mechanism for this case is shown in Figure 4B. Although we depict a di-iron cofactor, in recognition of precedent for the early steps in the mechanism (discussed below), other di-metal cofactors might also be competent to function in this pathway. The pathway begins with addition of O2 to the Fe2(II/II) cofactor to generate a peroxo-Fe2(III/III) intermediate, which attacks the carbonyl of the substrate, yielding a peroxyhemiacetal-Fe2(III/III) intermediate. Reductive cleavage of the O-O bond of the peroxyhemiacetal-Fe2(III/III) intermediate would yield a gem-diolyl radical and Fe2(III/III) cofactor. The former could undergo C1-C2 homolysis to give formate and an alkyl radical, which could abstract a hydrogen atom from the Fe2(III/III) cofactor to generate the alkane product and an Fe2(III/IV) form of the cofactor. A total of three additional electrons would then be required to convert the Fe2(III/IV) form to the O2-reactive Fe2(II/II) form and complete the cycle. Other sequences of bond-cleaving/forming and cofactor-reduction steps are also conceivable. The first steps of this mechanism have precedent. The addition of O2 to Fe2(II/II) cofactors to generate peroxo-Fe2(III/III) intermediates has been observed for numerous enzymes [13,14,16–19]. While there is extensive precedent for peroxo-Fe2(III/III) enzyme intermediates to act either as electrophiles toward aromatic [18] or amine substrates [19] or to undergo reductive or non-redox O-O scission to yield high-valent Fe2(III/IV) or Fe2(IV/IV) complexes that can cleave strong O-H and C-H bonds [13,17], we are not aware of an enzymic peroxo-Fe2(III/III) intermediate acting as a nucleophile. However, nucleophilic attack of a peroxo-Fe2(III/III) inorganic complex on CO2 has been reported [91]. In addition, there is extensive precedent for the nucleophilic nature of mononuclear peroxo complexes. Of particular relevance are complexes that attack the carbonyl C-atom of aldehydes, thereby initiating their deformylation. These include various P450-type heme enzymes (e.g. P450 aromatase, lanosterol-14α-demethylase, and progesterone-17αhydroxylse/17,20-lyase [92]), as well as mononuclear heme-iron-peroxo [93], non-heme-iron-peroxo [94], Ni-peroxo [95], and Mn-peroxo complexes [96]. However, the above examples all initiate the oxidative decarbonylation of aldehydes to yield formate and either two- or four-electron oxidized co-products. Thus, the pathway for decay of the proposed peroxyhemiacetal complex proposed for cyanobacterial ADs would necessarily be distinct from that occurring in previously reported reactions. However, these steps are similar to those proposed for the —RNR-like mechanism.
Hypothetical mechanisms for CO2 production by the cyanobacterial ADs
Given the structural similarity of the cyanobacterial ADs to other oxygenases and the precedent of the house fly AD for CO2 as co-product [61], it is logical also to consider this possible oxidative outcome. One conceivable mechanism would simply extend the OM-like mechanism (Scheme 3B). Following reductive elimination of the alkane, CO could remain stably bound, and its oxidation to CO2 might be necessary to complete the cycle. Addition of O2, perhaps to a Fe(II) ion of the cofactor (as shown in Scheme 3B), would lead to a superoxide complex (here {FeO2}8), of which the distal O-atom might attack the carbonyl C-atom, generating CO2 and an oxidized cluster [here Fe(III)/Ni(II)]. Re-reduction of that cluster form by the N/F/FR reducing system would then regenerate the aldehyde-reactive reduced cluster [here Fe(II)/Ni(I)] to complete the catalytic cycle.
A second conceivable class of mechanisms for production of CO2 along with the alkane is suggested by previous studies on a putatively heme-dependent aldehyde decarbonylase from the common house fly [61]. This enzyme reportedly converts an aldehyde substrate to the alkane and CO2. Reed, et al. suggested that an Fe(III)-heme peroxo could attack the carbonyl to form the peroxyhemiacetal complex, which could undergo O-O-bond homolysis to form an intermediate containing an Fe(IV)-oxo heme complex (compound II) and a gem-diolyl substrate radical. Radical fragmentation of the substrate C1-C2 bond would then yield formate and the alkyl radical. The alkyl radical would abstract the hydrogen atom from formate, generating the alkane and the potently reducing CO2− radical anion, which would reduce the Fe(IV) heme to the resting Fe(III) state [61]. The Reed, et al. mechanism can readily be adapted to include a dimetal-peroxo complex in place of the heme Fe(III)-O-O− heme intermediate and an Fe2(III/IV) complex in place of the Fe(IV)-oxo complex.
Outlook
The diversity of the mechanisms in the preceding discussion highlights the fact that, before fine details of the catalytic mechanism of the cyanobacterial AD can be probed, several more fundamental questions about the nature of the reaction must be answered. Conversely, answering these questions will drastically limit the types of mechanisms that must be considered. Thus, the imperatives for furthering our understanding of this exciting new enzyme are (1) to identify the C1-derived co-product of alkane production, which will eliminate the possible oxidative mechanisms if it is found to be formate or CO, or, alternatively, eliminate the “RNR-like” mechanisms if it is found to be CO2; (2) to determine whether the reported requirement for reductant is stoichiometric, as one would expect for an oxidative di-iron-oxygenase-like reaction, or catalytic, as one would anticipate for the OM-like and RNR-like non-redox pathways (in which the reductant would function merely in a pre-activation sequence); (3) determine whether O2 is, along with the reducing system, required for activity (it might not be for a non-redox reaction via the OM-like mechanism) and, if so, whether the requirement is stoichiometric, indicating a role as co-substrate in oxidative di-iron-oxygenase-like chemistry, or catalytic, indicating a role in an RNR-like preactivation or as a cofactor, à la oxalate-decarboxylase; and (4) identify the metal ions and their oxidation states in the active form of the enzyme. The report that the Np AD can be expressed in E. coli in an active form is remarkable and suggests that this system should be more amenable to such biochemical and mechanistic studies than any previously reported AD system [56]. In light of recent enormous investments in developing renewable, fungible (possibly alk(a/e)ne) fuels from cyanobacteria (e.g., the joint venture between Exxon-Mobil and Synthetic Genomics, Inc.), it appears likely that an understanding of the nature of the AD cofactor, including how it is assembled into its active state, and the products and mechanism of its reaction might be of considerable interest.
Acknowledgments
This work was supported by the National Institutes of Health (GM-55365 to JMB and CK) and the Dreyfus Foundation (Teacher Scholar Award to CK). The authors thank Megan L. Matthews for preparation of Figure 1.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Carsten Krebs, Email: ckrebs@psu.edu, Department of Chemistry and Department of Biochemistry and Molecular Biology, Penn State University, 332 Chemistry Building, University Park, PA, 16802, USA.
J. Martin Bollinger, Jr., Email: jmb21@psu.edu, Department of Chemistry and Department of Biochemistry and Molecular Biology, Penn State University, 336 Chemistry Building, University Park, PA, 16802, USA
Squire J. Booker, Email: Squire@psu.edu, Department of Chemistry and Department of Biochemistry and Molecular Biology, Penn State University, 302 Chemistry Building, University Park, PA, 16802, USA
References
References of outstanding interest (two dots)
Reference of special interest (one dot)
- 1.Nordlund P, Eklund H. Di-iron-carboxylate proteins. Curr Opin Struct Biol. 1995;5:758–766. doi: 10.1016/0959-440x(95)80008-5. [DOI] [PubMed] [Google Scholar]
- 2.Theil EC, Matzapetakis M, Liu XF. Ferritins: iron/oxygen biominerals in protein nanocages. J Biol Inorg Chem. 2006;11:803–810. doi: 10.1007/s00775-006-0125-6. [DOI] [PubMed] [Google Scholar]
- 3.Stubbe J. Di-iron-tyrosyl radical ribonucleotide reductases. Curr Opin Chem Biol. 2003;7:183–188. doi: 10.1016/s1367-5931(03)00025-5. [DOI] [PubMed] [Google Scholar]
- 4.Wallar BJ, Lipscomb JD. Dioxygen activation by enzymes containing binuclear non-heme iron clusters. Chem Rev. 1996;96:2625–2657. doi: 10.1021/cr9500489. [DOI] [PubMed] [Google Scholar]
- 5.Merkx M, Kopp DA, Sazinsky MH, Blazyk JL, Müller J, Lippard SJ. Dioxygen activation and methane hydroxylation by soluble methane monooxygenase: a tale of two irons and three proteins. Angew Chem Int Ed. 2001;40:2782–2807. doi: 10.1002/1521-3773(20010803)40:15<2782::AID-ANIE2782>3.0.CO;2-P. [DOI] [PubMed] [Google Scholar]
- 6.Pikus JD, Studts JM, Achim C, Kauffman KE, Münck E, Steffan RJ, McClay K, Fox BG. Recombinant Toluene-4-monooxygenase: Catalytic and Mössbauer Studies of the Purified Diiron and Rieske Components of a Four-Protein Complex. Biochemistry. 1996;35:9106–9119. doi: 10.1021/bi960456m. [DOI] [PubMed] [Google Scholar]
- 7.Nordlund I, Powlowski J, Shingler V. Complete nucleotide sequence and polypeptide analysis of multicomponent phenol hydroxylase from Pseudomonas sp. strain CF600. J Bacteriol. 1990;172:6826–6833. doi: 10.1128/jb.172.12.6826-6833.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Small FJ, Ensign SA. Alkene monooxygenase from Xanthobacter strain Py2. Purification and characterization of a four-component system central to the bacterial metabolism of aliphatic alkenes. J Biol Chem. 1997;272:24913–24920. doi: 10.1074/jbc.272.40.24913. [DOI] [PubMed] [Google Scholar]
- 9.Fox BG, Lyle KS, Rogge CE. Reactions of the diiron enzyme stearoyl-acyl carrier protein desaturase. Acc Chem Res. 2004;37:421–429. doi: 10.1021/ar030186h. [DOI] [PubMed] [Google Scholar]
- 10.Choi YS, Zhang H, Brunzelle JS, Nair SK, Zhao H. In vitro reconstitution and crystal structure of p-aminobenzoate N-oxygenase (AurF) involved in aureothin biosynthesis. Proc Natl Acad Sci USA. 2008;105:6858–6863. doi: 10.1073/pnas.0712073105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Behan RK, Lippard SJ. The Aging-Associated Enzyme CLK-1 Is a Member of the Carboxylate-Bridged Diiron Family of Proteins. Biochemistry. 2010;49:9679–9681. doi: 10.1021/bi101475z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Andrews SC. The ferritin-like superfamily: Evolution of the biological iron storeman from a rubrerythrin-like ancestor. Biochim Biophys Acta. 2010;1800:691–705. doi: 10.1016/j.bbagen.2010.05.010. [DOI] [PubMed] [Google Scholar]
- 13.Liu KE, Wang D, Huynh BH, Edmondson DE, Salifoglou A, Lippard SJ. Spectroscopic detection of intermediates in the reaction of dioxygen with the reduced methane monooxygenase/hydroxylase from Methylococcus capsulatus (Bath) J Am Chem Soc. 1994;116:7465–7466. [Google Scholar]
- 14.Tong WH, Chen S, Lloyd SG, Edmondson DE, Huynh BH, Stubbe J. Mechanism of assembly of the diferric cluster-tyrosyl radical cofactor of Escherichia coli ribonucleotide reductase from the diferrous form of the R2 subunit. J Am Chem Soc. 1996;118:2107–2108. [Google Scholar]
- 15.Broadwater JA, Ai J, Loehr TM, Sanders-Loehr J, Fox BG. Peroxodiferric intermediate of stearoyl-acyl carrier protein Δ9 desaturase: Oxidase reactivity during single turnover and implications for the mechanism of desaturation. Biochemistry. 1998;37:14664–14671. doi: 10.1021/bi981839i. [DOI] [PubMed] [Google Scholar]
- 16.Pereira AS, Small GW, Krebs C, Tavares P, Edmondson DE, Theil EC, Huynh BH. Direct Spectroscopic and Kinetic Evidence for the Involvement of a Peroxodiferric Intermediate during the Ferroxidase Reaction in Fast Ferritin Mineralization. Biochemistry. 1998;37:9871–9876. doi: 10.1021/bi980847w. [DOI] [PubMed] [Google Scholar]
- 17.Yun D, Garcia-Serres R, Chicalese BM, An YH, Huynh BH, Bollinger JM., Jr (μ-1,2-Peroxo)diiron(III/III) complex as a precursor to the diiron(III/IV) intermediate X in the assembly of the iron-radical cofactor of ribonucleotide reductase from mouse. Biochemistry. 2007;46:1925–1932. doi: 10.1021/bi061717n. [DOI] [PubMed] [Google Scholar]
- 18.Murray LJ, Naik SG, Ortillo DO, Garcia-Serres R, Lee JK, Huynh BH, Lippard SJ. Characterization of the arene-oxidizing intermediate in ToMOH as a diiron(III) species. J Am Chem Soc. 2007;129:14500–14510. doi: 10.1021/ja076121h. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Korboukh VK, Li N, Barr EW, Bollinger JM, Jr, Krebs C. A long-lived, substrate-hydroxylating peroxodiiron(III/III) intermediate in the amine oxygenase, AurF, from Streptomyces thioluteus. J Am Chem Soc. 2009;131:13608–13609. doi: 10.1021/ja9064969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Vu VV, Emerson JP, Martinho M, Kim YS, Munck E, Park MH, Que L. Human deoxyhypusine hydroxylase, an enzyme involved in regulating cell growth, activates O2 with a nonheme diiron center. Proc Natl Acad Sci USA. 2009;106:14814–14819. doi: 10.1073/pnas.0904553106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Skulan AJ, Brunold TC, Baldwin J, Saleh L, Bollinger JM, Jr, Solomon EI. Nature of the peroxo intermediate of the W48F/D84E ribonucleotide reductase variant: implications for O2 activation by binuclear non-heme iron enzymes. J Am Chem Soc. 2004;126:8842–8855. doi: 10.1021/ja049106a. [DOI] [PubMed] [Google Scholar]
- 22.Jameson GNL, Jin W, Krebs C, Perreira AS, Tavares P, Liu X, Theil EC, Huynh BH. Stoichiometric production of hydrogen peroxide and parallel formation of ferric multimers through decay of the diferric-peroxo complex, the first detectable intermediate in ferritin mineralization. Biochemistry. 2002;41:13435–13443. doi: 10.1021/bi026478s. [DOI] [PubMed] [Google Scholar]
- 23.Baldwin J, Krebs C, Ley BA, Edmondson DE, Huynh BH, Bollinger JM., Jr Mechanism of rapid electron transfer during oxygen activation in the R2 subunit of Escherichia coli ribonucleotide reductase. 1. Evidence for a transient tryptophan radical. J Am Chem Soc. 2000;122:12195–12206. [Google Scholar]
- 24.Bollinger JM, Jr, Edmondson DE, Huynh BH, Filley J, Norton JR, Stubbe J. Mechanism of assembly of the tyrosyl radical-dinuclear iron cluster cofactor of ribonucleotide reductase. Science. 1991;253:292–298. doi: 10.1126/science.1650033. [DOI] [PubMed] [Google Scholar]
- 25.Sturgeon BE, Burdi D, Chen S, Huynh BH, Edmondson DE, Stubbe J, Hoffman BM. Reconsideration of X, the diiron intermediate formed during cofactor assembly in E. coli ribonucleotide reductase. J Am Chem Soc. 1996;118:7551–7557. [Google Scholar]
- 26.Shanmugam M, Doan PE, Lees NS, Stubbe J, Hoffman BM. Identification of Protonated Oxygenic Ligands of Ribonucleotide Reductase Intermediate X. J Am Chem Soc. 2009;131:3370–3376. doi: 10.1021/ja809223s. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mitić N, Clay MD, Saleh L, Bollinger JM, Jr, Solomon EI. Spectroscopic and electronic structure studies of intermediate X in ribonucleotide reductase R2 and two variants: a description of the FeIV-oxo bond in the FeIII-O-FeIV dimer. J Am Chem Soc. 2007;129:9049–9065. doi: 10.1021/ja070909i. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Stubbe J, Nocera DG, Yee CS, Chang MCY. Radical initiation in the class I ribonucleotide reductase: long-range proton-coupled electron transfer? Chem Rev. 2003;103:2167–2202. doi: 10.1021/cr020421u. [DOI] [PubMed] [Google Scholar]
- 29.Lee S-K, Fox BG, Froland WA, Lipscomb JD, Münck E. A transient intermediate of the methane monooxygenase catalytic cycle containing an FeIVFeIV cluster. J Am Chem Soc. 1993;115:6450–6451. [Google Scholar]
- 30.Shu L, Nesheim JC, Kauffmann KE, Münck E, Lipscomb JD, Que L., Jr An Fe2IVO2 diamond core structure for the key intermediate Q of methane monooxygenase. Science. 1997;275:515–518. doi: 10.1126/science.275.5299.515. [DOI] [PubMed] [Google Scholar]
- 31•.Li N, Korboukh VK, Krebs C, Bollinger JM., Jr Four-electron oxidation of p-hydroxylaminobenzoate to p-nitrobenzoate by a peroxodiferric complex in AurF from Streptomyces thioluteus. Proc Natl Acad Sci USA. 2010;107:15722–15727. doi: 10.1073/pnas.1002785107. This paper reports the first example of a four-electron substrate oxidation by a peroxo-diferric intermediate, the stable complex previously identified in the nitro-group-installing di-iron-carboxylate N-oxygenase, AurF, from the aureothin pathways in Streptomyces thioluteus. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Fontecave M, Eliasson R, Reichard P. NAD(P)H:flavin oxidoreductase of Escherichia coli. A ferric ion reductase participating in the generation of the free radical of ribonucleotide reductase. J Biol Chem. 1987;262:12325–12331. [PubMed] [Google Scholar]
- 33.Wu CH, Jiang W, Krebs C, Stubbe J. YfaE, a ferredoxin involved in diferric-tyrosyl radical maintenance in Escherichia coli ribonucleotide reductase. Biochemistry. 2007;46:11577–11588. doi: 10.1021/bi7012454. [DOI] [PubMed] [Google Scholar]
- 34.Kono Y, Fridovich I. Isolation and characterization of the pseudocatalase of Lactobacillus plantarum - A new manganese-containing enzyme. J Biol Chem. 1983;258:6015–6019. [PubMed] [Google Scholar]
- 35.Dismukes GC. Manganese enzymes with binuclear active sites. Chem Rev. 1996;96:2909–2926. doi: 10.1021/cr950053c. [DOI] [PubMed] [Google Scholar]
- 36.Whittaker MM, Barynin VV, Igarashi T, Whittaker JW. Outer sphere mutagenesis of Lactobacillus plantarum manganese catalase disrupts the cluster core - Mechanistic implications. Eur J Biochem. 2003;270:1102–1116. doi: 10.1046/j.1432-1033.2003.03459.x. [DOI] [PubMed] [Google Scholar]
- 37.Krebs C, Chen S, Baldwin J, Ley BA, Patel U, Edmondson DE, Huynh BH, Bollinger JM., Jr Mechanism of rapid electron transfer during oxygen activation in the R2 subunit of Escherichia coli ribonucleotide reductase. 2. Evidence for and consequences of blocked electron transfer in the W48F variant. J Am Chem Soc. 2000;122:12207–12219. [Google Scholar]
- 38.Sahlin M, Sjöberg B-M, Backes G, Loehr T, Sanders-Loehr J. Activation of the iron-containing B2 protein of ribonucleotide reductase by hydrogen peroxide. Biochem Biophys Res Comm. 1990;167:813–818. doi: 10.1016/0006-291x(90)92098-k. [DOI] [PubMed] [Google Scholar]
- 39•.Jiang W, Yun D, Saleh L, Barr EW, Xing G, Hoffart LM, Maslak M-A, Krebs C, Bollinger JM., Jr A manganese(IV)/iron(III) cofactor in Chlamydia trachomatis ribonucleotide reductase. Science. 2007;316:1188–1191. doi: 10.1126/science.1141179. This paper describes the discovery of the Mn(IV)/Fe(III) cofactor in the class Ic RNR-β2 protein from Chlamydia trachomatis and provides evidence that the Mn(IV)-site functionally replaces the tyrosyl radical cofactor of the Ia/b-RNR-β2s proteins. It represents the first examples of a Mn/Fe redox cofactor in biology and a heterodinuclear redox cofactor in the di-iron-carboxylate protein family. [DOI] [PubMed] [Google Scholar]
- 40.Jiang W, Yun D, Saleh L, Bollinger JM, Jr, Krebs C. Formation and function of the manganese(IV)/iron(III) cofactor in Chlamydia trachomatis ribonucleotide reductase. Biochemistry. 2008;47:13736–13744. doi: 10.1021/bi8017625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Bollinger JM, Jr, Jiang W, Green MT, Krebs C. The manganese(IV)/iron(III) cofactor of Chlamydia trachomatis ribonucleotide reductase: Structure, assembly, radical initiation, and evolution. Curr Opin Struct Biol. 2008;18:650–657. doi: 10.1016/j.sbi.2008.11.007. [DOI] [PubMed] [Google Scholar]
- 42.Roshick C, Iliffe-Lee ER, McClarty G. Cloning and characterization of ribonucleotide reductase from Chlamydia trachomatis. J Biol Chem. 2000;275:38111–38119. doi: 10.1074/jbc.M006367200. [DOI] [PubMed] [Google Scholar]
- 43.Jiang W, Bollinger JM, Jr, Krebs C. The active form of Chlamydia trachomatis ribonucleotide reductase R2 protein contains a heterodinuclear Mn(IV)/Fe(III) cluster with S = 1 ground state. J Am Chem Soc. 2007;129:7504–7505. doi: 10.1021/ja072528a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Jiang W, Hoffart LM, Krebs C, Bollinger JM., Jr A manganese(IV)/iron(IV) intermediate in assembly of the manganese(IV)/iron(III) cofactor of Chlamydia trachomatis ribonucleotide reductase. Biochemistry. 2007;46:8709–8716. doi: 10.1021/bi700906g. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Jiang W, Saleh L, Barr EW, Xie J, Gardner MM, Krebs C, Bollinger JM., Jr Branched activation- and catalysis-specific pathways for electron relay to the manganese/iron cofactor in ribonucleotide reductase from Chlamydia trachomatis. Biochemistry. 2008:8477–8484. doi: 10.1021/bi800881m. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Voevodskaya N, Narvaez AJ, Domkin V, Torrents E, Thelander L, Gräslund A. Chlamydial ribonucleotide reductase: tyrosyl radical function in catalysis replaced by the FeIII-FeIV cluster. Proc Natl Acad Sci USA. 2006;103:9850–9854. doi: 10.1073/pnas.0600603103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Roos K, Siegbahn PEM. Density functional theory study of the manganese-containing ribonucleotide reductase from Chlamydia trachomatis: Why manganese is needed in the active complex. Biochemistry. 2009;48:1878–1887. doi: 10.1021/bi801695d. [DOI] [PubMed] [Google Scholar]
- 48.Willing A, Follmann H, Auling G. Ribonucleotide reductase of Brevibacterium ammoniagenes is a manganese enzyme. Eur J Biochem. 1988;170:603–611. doi: 10.1111/j.1432-1033.1988.tb13740.x. [DOI] [PubMed] [Google Scholar]
- 49.Huque Y, Fieschi F, Torrents E, Gibert I, Eliasson R, Reichard P, Sahlin M, Sjöberg B-M. The active form of the R2F protein of class Ib ribonucleotide reductase from Corynebacterium ammoniagenes is a diferric protein. J Biol Chem. 2000;275:25365–25371. doi: 10.1074/jbc.M002751200. [DOI] [PubMed] [Google Scholar]
- 50•.Cotruvo JA, Stubbe J. An active dimanganese(III)-tyrosyl radical cofactor in Escherichia coli class Ib ribonucleotide reductase. Biochemistry. 2010;49:1297–1309. doi: 10.1021/bi902106n. Work reported in this paper and in [51] provide evidence that the Ib-RNR-β2 proteins from E. coli and Corynebacterium ammoniagenes harbor Mn2(III/III)/Y• cofactors. In addition, the authors also demonstrate that the assembly of this cofactor requires the accessory protein NrdI and O2, putatively to generate a hydroperoxide (HO2−) oxidant. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51•.Cox N, Ogata H, Stolle P, Reijerse E, Auling G, Lubitz W. A tyrosyl-dimanganese coupled spin system is the native metalloradical cofactor of the R2F subunit of the ribonucleotide reductase of Corynebacterium ammoniagenes. J Am Chem Soc. 2010;132:11197–11213. doi: 10.1021/ja1036995. Work reported in this paper and in [50] provide evidence that the Ib-RNR-β2 proteins harbor Mn2(III/III)/Y• cofactors. [DOI] [PubMed] [Google Scholar]
- 52.Boal AK, Cotruvo JA, Stubbe J, Rosenzweig AC. Structural basis for activation of class Ib ribonucleotide reductase. Science. 2010;329:1526–1530. doi: 10.1126/science.1190187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Monje-Casas F, Jurado J, Prieto-Alamo MJ, Holmgren A, Pueyo C. Expression analysis of the nrdHIEF operon from Escherichia coli - Conditions that trigger the transcript level in vivo. J Biol Chem. 2001;276:18031–18037. doi: 10.1074/jbc.M011728200. [DOI] [PubMed] [Google Scholar]
- 54.McHugh JP, Rodriguez-Quinones F, Abdul-Tehrani H, Svistunenko DA, Poole RK, Cooper CE, Andrews SC. Global iron-dependent gene regulation in Escherichia coli - A new mechanism for iron homeostasis. J Biol Chem. 2003;278:29478–29486. doi: 10.1074/jbc.M303381200. [DOI] [PubMed] [Google Scholar]
- 55•.Andersson CS, Högbom M. A Mycobacterium tuberculosis ligand-binding Mn/Fe protein reveals a new cofactor in a remodeled R2-protein scaffold. Proc Natl Acad Sci USA. 2009;106:5633–5638. doi: 10.1073/pnas.0812971106. This paper reports the three-dimensional structure of the Rv0233 protein from Mycobacterium tuberculosis, which reveals the presence of a carboxylate-bridged Mn/Fe cofactor. Although the function of this protein was not (and, to our knowledge, still has not been) established, it is likely that this is the second example of a Mn/Fe redox cofactor in this protein family. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56••.Schirmer A, Rude MA, Li X, Popova E, del Cardayre SB. Microbial biosynthesis of alkanes. Science. 2010;329:559–562. doi: 10.1126/science.1187936. The authors report the identification of pairs of genes involved in alkane biosyntheses in cyanobacteria by a clever comparison of entire cyanobacterial genomes. The authors convincingly corroborate this hypothesis by experiments involving heterologous expression of the genes in Escherichia coli, disruption of the genes in a cyanobacterium, and in vitro verification of the proposed activities. Notably, they did not identify the C1-derived co-product of the second enzyme, the aldehyde decarbonylase. [DOI] [PubMed] [Google Scholar]
- 57.Kolattukudy PE. Chemistry and biochemistry of natural waxes. Amsterdam: Elsevier Scientific Publishing Co; 1976. [Google Scholar]
- 58.Cheesbrough TM, Kolattukudy PE. Alkane biosynthesis by decarbonylation of aldehydes catalyzed by a particulate preparation from Pisum sativum. Proc Natl Acad Sci USA. 1984;81:6613–6617. doi: 10.1073/pnas.81.21.6613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Yoder JA, Denlinger DL, Dennis MW, Kolattukudy PE. Enhancement of diapausing flesh fly puparia with additional hydrocarbons and evidence for alkane biosynthesis by a decarbonylation mechanism. Insect Biochem Mol Biol. 1992;22:237–243. [Google Scholar]
- 60.Cheesbrough TM, Kolattukudy PE. Microsomal preparation from an animal tissue catalyzes release of carbon monoxide from a fatty aldehyde to generate an alkane. J Biol Chem. 1988;263:2738–2743. [PubMed] [Google Scholar]
- 61.Reed JR, Quilici DR, Blomquist GJ, Reitz RC. Proposed mechanism for the cytochrome P450-catalyzed conversion of aldehydes to hydrocarbons in the house fly, Musca domestica. Biochemistry. 1995;34:16221–16227. doi: 10.1021/bi00049a038. [DOI] [PubMed] [Google Scholar]
- 62.Bourre JM, Cassagne C, Larrouquère-Régnier S, Darriet D. Occurence of alkanes in brain myelin - comparison between normal and quaking mouse. J Neurochem. 1977;29:645–648. doi: 10.1111/j.1471-4159.1977.tb07781.x. [DOI] [PubMed] [Google Scholar]
- 63.Dennis M, Kolattukudy PE. A cobalt-porphyrin enzyme converts a fatty aldehyde to a hydrocarbon and carbon monoxide. Proc Natl Acad Sci USA. 1992;89:5306–5310. doi: 10.1073/pnas.89.12.5306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Fox BG, Shanklin J, Ai J, Loehr TM, Sanders-Loehr J. Resonance Raman Evidence for an Fe-O-Fe Center in Stearoyl-ACP Desaturase. Primary Sequence Identity with Other Diiron-Oxo Proteins. Biochemistry. 1994;33:12776–12786. doi: 10.1021/bi00209a008. [DOI] [PubMed] [Google Scholar]
- 65.Aarts MGM, Keijzer CJ, Stiekema WJ, Pereira A. Molecular characterization of the CER1 gene of arabidopsis involved in epicuticular wax biosynthesis and pollen fertility. Plant Cell. 1995;7:2115–2127. doi: 10.1105/tpc.7.12.2115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Shanklin J, Whittle E, Fox BG. Eight Histidine Residues Are Catalytically Essential in a Membrane-Associated Iron Enzyme, Stearoyl-CoA Desaturase, and Are Conserved in Alkane Hydroxylase and Xylene Monooxygenase. Biochemistry. 1994;33:12787–12794. doi: 10.1021/bi00209a009. [DOI] [PubMed] [Google Scholar]
- 67.Nordlund P, Sjöberg B-M, Eklund H. Three-dimensional structure of the free radical protein of ribonucleotide reductase. Nature. 1990;345:593–598. doi: 10.1038/345593a0. [DOI] [PubMed] [Google Scholar]
- 68.Högbom M, Stenmark P, Voevodskaya N, McClarty G, Gräslund A, Nordlund P. The radical site in Chlamydial ribonucleotide reductase defines a new R2 subclass. Science. 2004;305:245–248. doi: 10.1126/science.1098419. [DOI] [PubMed] [Google Scholar]
- 69.Högbom M. The manganese/iron-carboxylate proteins: what is what, where are they, and what can the sequences tell us? J Biol Inorg Chem. 2010;15:339–349. doi: 10.1007/s00775-009-0606-5. [DOI] [PubMed] [Google Scholar]
- 70.Holm L, Sander C. The FSSP database: Fold classification based on structure structure alignment of proteins. Nucleic Acids Research. 1996;24:206–209. doi: 10.1093/nar/24.1.206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Schneider-Belhaddad F, Kolattukudy PE. Solubilization, partial purification, and characterization of a fatty aldehyde decarbonylase from a higher plant, Pisum sativum. Arch Biochem Biophys. 2000;377:341–349. doi: 10.1006/abbi.2000.1798. [DOI] [PubMed] [Google Scholar]
- 72.Ohno K, Tsuji J. Organic synthesis of noble metal compounds XXXV: Novel decarbonylation reactions of aldehydes and acyl halides using rhodium complexes. J Am Chem Soc. 1968;90:99–107. [Google Scholar]
- 73.Doughty DH, Pignolet LH. Catalytic decarbonylation of aldehydes. J Am Chem Soc. 1978;100:7083–7085. [Google Scholar]
- 74.Fristrup P, Kreis M, Palmelund A, Norrby P-O, Madsen R. The mechanism for the rhodium-catalyzed decarbonylation of aldehydes: A combined experimental and theoretical study. J Am Chem Soc. 2008;130:5206–5215. doi: 10.1021/ja710270j. [DOI] [PubMed] [Google Scholar]
- 75.Belani RM, James BR, Dolphin D, Rettig SJ. Catalytic decarbonylation of aldehydes using iron(II) porphyrin complexes, and the crystal structure of (5,10,15,20-tetraphenylporphinato)bis(tri-n-butylphosphine)iron(II) Can J Chem. 1988;66:2072–2078. [Google Scholar]
- 76.Domazetis G, Tarpey B, Dolphin D, James BR. Catalytic decarbonylation of alhehydes using ruthenium(II) porphyrin systems. J Chem Soc, Chem Comm. 1980:939–940. [Google Scholar]
- 77•.Scheller S, Goenrich M, Boecher R, Thauer RK, Jaun B. The key nickel enzyme of methanogenesis catalyses the anaerobic oxidation of methane. Nature. 2010;465:606–609. doi: 10.1038/nature09015. This paper provides evidence that the key step of anaerobic methane oxidation, the cleavage of one of the strong C-H bonds of methane, proceeds by —reverse methanogenesis at the Ni-center of the coenzyme F430 prosthetic group found in the enzyme, methyl coenzyme-M reductase. [DOI] [PubMed] [Google Scholar]
- 78.Diekert G, Klee B, Thauer RK. Nickel, a component of factor-F430 from Methanobacterium thermoautotrophicum. Arch Microbiol. 1980;124:103–106. doi: 10.1007/BF00407036. [DOI] [PubMed] [Google Scholar]
- 79.Pfaltz A, Jaun B, Fassler A, Eschenmoser A, Jaenchen R, Gilles HH, Diekert G, Thauer RK. Factor F-430 from methanogenic bacteria - structure of the porphinoid ligand system. Helv Chim Acta. 1982;65:828–865. [Google Scholar]
- 80.Ellefson WL, Whitman WB, Wolfe RS. Nickel-containing factor-F430 - chromophore of the methylreductase of methanobacterium. Proc Natl Acad Sci USA. 1982;79:3707–3710. doi: 10.1073/pnas.79.12.3707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 81.Ragsdale SW. Nickel and the carbon cycle. J Inorg Biochem. 2007;101:1657–1666. doi: 10.1016/j.jinorgbio.2007.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 82.Scheller S, Goenrich M, Mayr S, Thauer RK, Jaun B. Intermediates in the catalytic cycle of methyl coenzyme M reductase: Isotope exchange is consistent with formation of a σ-alkane-nickel complex. Angew Chem Int Ed. 2010;49:8112–8115. doi: 10.1002/anie.201003214. [DOI] [PubMed] [Google Scholar]
- 83.Xia JQ, Hu ZG, Popescu CV, Lindahl PA, Münck E. Mössbauer and EPR study of the Ni-activated alpha-subunit of carbon monoxide dehydrogenase from Clostridium thermoaceticum. J Am Chem Soc. 1997;119:8301–8312. [Google Scholar]
- 84.Brunold TC. Spectroscopic and computational insights into the geometric and electronic properties of the A-cluster of acetyl-coenzyme A synthase. J Biol Inorg Chem. 2004;9:533–541. doi: 10.1007/s00775-004-0566-8. [DOI] [PubMed] [Google Scholar]
- 85.Maynard EL, Lindahl PA. Evidence of a molecular tunnel connecting the active sites for CO2 reduction and acetyl-CoA synthesis in acetyl-CoA synthase from Clostridium thermoaceticum. J Am Chem Soc. 1999;121:9221–9222. [Google Scholar]
- 86.Seravalli J, Ragsdale SW. Channeling of carbon monoxide during anaerobic carbon dioxide fixation. Biochemistry. 2000;39:1274–1277. doi: 10.1021/bi991812e. [DOI] [PubMed] [Google Scholar]
- 87.Younker JM, Krest CM, Jiang W, Krebs C, Bollinger JM, Jr, Green MT. Structural analysis of the Mn(IV)/Fe(III) cofactor of Chlamydia trachomatis ribonucleotide reductase by extended X-ray absorption fine structure spectroscopy and density functional theory calculations. J Am Chem Soc. 2008;130:15022–15027. doi: 10.1021/ja804365e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 88.Berkowitz J, Ellison GB, Gutman D. 3 methods to measure RH bond energies. J Phys Chem. 1994;98:2744–2765. [Google Scholar]
- 89.Tsentalovich YP, Fischer H. Solvent effect on the decarbonylation of acyl radicals studied by laser flash-photolysis. J Chem Soc, Perkin Trans. 1994;2:729–733. [Google Scholar]
- 90.Svedružić D, Liu Y, Reinhardt LA, Wroclawska E, Cleland WW, Richards NGJ. Investigating the roles of putative active site residues in the oxalate decarboxylase from Bacillus subtilis. Arch Biochem Biophys. 2007;464:36–47. doi: 10.1016/j.abb.2007.03.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91.LeCloux DD, Barrios AM, Lippard SJ. The reactivity of well defined diiron(III) peroxo complexes toward substrates: Addition to electrophiles and hydrocarbon oxidation. Bioorg Med Chem. 1999;7:763–772. doi: 10.1016/s0968-0896(98)00270-3. [DOI] [PubMed] [Google Scholar]
- 92.Wertz DL, Valentine JS. Metal-Oxo and Metal-Peroxo Species in Catalytic Oxidations. Vol. 97 Springer-Verlag; Berlin: 2000. Nucleophilicity of iron-peroxo porphyrin complexes; pp. 37–60. Structure and Bonding. [Google Scholar]
- 93.Wertz DL, Sisemore MF, Selke M, Driscoll J, Valentine JS. Mimicking cytochrome P-450 2B4 and aromatase: Aromatization of a substrate analogue by a peroxo Fe(III) porphyrin complex. J Am Chem Soc. 1998;120:5331–5332. [Google Scholar]
- 94.Annaraj J, Suh Y, Seo MS, Kim SO, Nam W. Mononuclear nonheme ferric-peroxo complex in aldehyde deformylation. Chem Comm. 2005:4529–4531. doi: 10.1039/b505562h. [DOI] [PubMed] [Google Scholar]
- 95.Cho J, Sarangi R, Annaraj J, Kim SY, Kubo M, Ogura T, Solomon EI, Nam W. Geometric and electronic structure and reactivity of a mononuclear ‘side-on’ nickel(III)-peroxo complex. Nature Chemistry. 2009;1:568–572. doi: 10.1038/nchem.366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 96.Annaraj J, Cho JH, Lee YM, Kim SY, Latifi R, de Visser SP, Nam W. Structural Characterization and Remarkable Axial Ligand Effect on the Nucleophilic Reactivity of a Nonheme Manganese(III)-Peroxo Complex. Angew Chem - Int Ed. 2009;48:4150–4153. doi: 10.1002/anie.200900118. [DOI] [PubMed] [Google Scholar]