Maturation of the [Ni–4Fe–4S] active site of carbon monoxide dehydrogenases

Maturation of the [Ni–4Fe–4S] active site of carbon monoxide dehydrogenases Nickel-containing enzymes are diverse in terms of function and active site structure. In many cases, the biosynthesis of the active site depends on accessory proteins which transport and insert the Ni ion. We review and discuss the literature related to the maturation of carbon monoxide dehydrogenases (CODH) which bear a nickel-containing active site consisting of a [Ni–4Fe–4S] center called the C-cluster. The maturation of this center has been much less studied than that of other nickel- containing enzymes such as urease and NiFe hydrogenase. Several proteins present in certain CODH operons, including the nickel-binding proteins CooT and CooJ, still have unclear functions. We question the conception that the maturation of all CODH depends on the accessory protein CooC described as essential for nickel insertion into the active site. The available literature reveals biological variations in CODH active site biosynthesis. Keywords Carbon monoxide dehydrogenase · Active site · Iron–sulfur cluster · Maturation Introduction cofactor (pincer) linked to a Lys of the protein backbone and methyl-coM reductase in which Ni is part of coenzyme Nine nickel-containing enzymes have been discovered and F430. Depending on the enzyme, Ni participates in cataly- characterized so far, but there exists certainly others [1, 2]. sis either by acting as a Lewis acid or by promoting redox Among them, seven consume or produce small molecules chemistry. (hydrogenase, carbon monoxide dehydrogenase, superox- Although the insertion of Ni seems spontaneous in gly- ide dismutase, urease, acireductone dioxygenase, methyl- oxylase I and acireductone dioxygenase, it requires dedi- coM reductase and acetyl-CoA synthase). The other two cated biological machineries in the other cases. The present (glyoxylase I and lactate racemase) are involved in lactate review focuses on the carbon monoxide dehydrogenases metabolism. The structures of their active sites are diverse from anaerobic microorganisms (Ni-containing CODH) in terms of nature and number of ligands to the Ni. In most which catalyze the reversible oxidation of CO with high cases, the Ni is coordinated by acidic residues (Cys, His, turnover frequencies [3–6]. These enzymes bear a nickel- Glu, Asp, carbonylated Lys) or water molecules. The three containing active site, the so-called C-cluster, which consists exceptions are carbon monoxide dehydrogenase, in which of a [Ni–3Fe–4S] cubane connected to a unique iron site Ni is also coordinated to inorganic sulfur in a [Ni–4Fe–4S] through a linking sulfide [ 7–9]. cluster, lactate racemase in which Ni is part of a non-protein Generalities The original version of this article was revised due to a retrospective Open Access order. Some microorganisms can grow in the presence of carbon monoxide, which they use as a source of carbon and/or * Sébastien Dementin energy [10, 11]. The oxidation of CO to CO by these micro- dementin@imm.cnrs.fr organisms is catalyzed by carbon monoxide dehydrogenases Aix-Marseille Université, CNRS, BIP UMR 7281, Institut (CODH). Nevertheless, CODH from aerobic and anaerobic de Microbiologie de la Méditerranée, 31 Chemin J. Aiguier, bacteria are not phylogenetically related and have distinct 13402 Marseille Cedex 20, France Vol.:(0123456789) 1 3 614 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 structures and kinetic properties. Most aerobic CO-utilizing to the xanthine oxidase family; their active site a binuclear bacteria (carboxydotrophs) oxidize CO in their respiratory cluster of Mo and Cu (MoCu–CODH) (Fig. 1a, d). These chain [12] using a variety of acceptors such as O (Oligo- MoCu–CODH only catalyze the oxidation of CO (not the −1 tropha carboxidovorans [13]) or nitrate for dissimilatory reduction of C O ) with a turnover frequency of up to 100 s nitrate reduction (Burkholderia xenovorans LB400 [11]). [13]. Some photosynthetic bacteria, such as Rhodopseudomonas Here we focus on the CODH from anaerobic microor- gelatinosa, can use CO as a carbon source by first convert- ganisms (Ni-containing CODH), which we will abbrevi- ing it into CO , which is then reduced into carbohydrate ate CODH in the following text for clarity. These enzymes through the Calvin–Benson–Bassham cycle [14]. CODH contain a unique nickel-containing active site [Ni–4Fe–4S], from aerobic bacteria are heterotrimeric enzymes and belong called the C-cluster. This cluster consists of a [Ni–3Fe–4S] Fig. 1 Structures of prototypi- cal Ni-containing CODH and their active sites. a Structure of the MoCu-CODH from Oligotropha carboxidovorans, PDB: 1N63 [30]. The L-subunit (89 kDa) in yellow, contains the active site. The S-subunit (18 kDa) in green, contains the two [2Fe–2S] clusters and the M-subunit (30 kDa) binds a FAD cofactor. b Structure of the CODH-II from Carboxydother- mus hydrogenoformans (Ch), PDB: 3B53 [31]. Each subunit (67 kDa) of this homodimer is colored in red or in orange. c Structure of the CODH from Methanosarcina barkeri, PDB: 3CF4 [32]. Subunit α (89 kDa) containing the nine iron–sul- fur clusters is colored in blue and the subunit ε (20 kDa) is colored in black. Structures of the active sites of the MoCu– CODH from Oligotropha carboxidovorans (d) and of the CODH-II from Ch (e). The colors for the atoms in d, e are green for Ni, orange for Fe, yel- low for S, red for O, blue for N, turquoise for Mo, light orange for Cu and white for C 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 615 cubane connected to a unique iron site through a linking pathway [26–29]. The major function of CODH in this meta- sulfide (Fig.  1e) [7–9]. CODH catalyze the oxidation of CO bolic pathway requires coordination of CO reduction at the with turnover frequencies ranging from hundreds to tens of C-cluster with CO channeling and reaction with a methyl thousands turnovers per second [3–6]. CODH-I from Car- group and CoA at the A-cluster active site of ACS. boxydothermus hydrogenoformans (Ch) is the most active, This article reviews the mechanisms and accessory pro- −1 with a turnover rate of 39,000 s at 70 °C, pH 8 [3]. Con- teins/chaperones involved in the maturation of this unique trary to Mo–CODH, CODH also catalyze the reverse reac- C-cluster. There has been a general agreement that CODH tion (the reduction of C O into CO). These enzymes are maturation depends on the accessory protein CooC, but on divided in four classes depending on their structures and the basis of the recent literature, it appears that the matu- functions [15]. CODH of classes I, II and III are bifunc- ration of CODH may differ from one enzyme to another, tional, they form a complex with acetyl-CoA synthase (ACS/ and that in some cases, it is independent of any specific CODH), while class IV CODH are monofunctional. Class I maturase. and II CODH form heterotetramers containing nine iron–sul- fur clusters (Fig. 1c). CODH of class III and IV are homodi- meric, and contain three iron–sulfur clusters (Fig. 1b). Maturation of the [Ni–4Fe–4S] C‑cluster As exemplified by the photosynthetic bacterium Rho- of CODH dospirillum rubrum (Rr), CO can be used as an alternative to light as a source of energy [16]. It is hypothesized that The diversity of the operons encoding for CODH (Fig. 2) cytoplasmic oxidation of CO is coupled to the reduction of suggests a variability in the maturation mechanisms. Some protons. The product dihydrogen would diffuse through the operons encoding for a CODH do not contain any maturase internal membrane and be oxidized by a periplasmic hydro- (e.g., CODH-II from Ch [27]), whereas many of them con- genase, leading to the formation of a proton motive force tain the maturation gene cooC. Other accessory genes can [3, 17, 18]. Some anaerobic sulfate reducing microorgan- be present in the CODH operons in addition to cooC: for isms couple the oxidation of CO to the reduction of sulfate example, in Rhodospirillum rubrum (Rr) [16] and Citro- [19–21] and some methanogenic archaea use CO as carbon/ bacter amalonaticus Y19 (CaY19) [30], the operons also electron donor for methane formation [22, 23]. contain the cooJ and cooT genes. In cases where CODH is in complex with acetyl-CoA synthase (ACS), and is therefore called bifunctional, it reduces CO to CO which is channeled to ACS, whose active Physiological importance of accessory site (called A-cluster) is a bi-Ni center attached to an Fe–S proteins cluster [24, 25]. This CODH/ACS complex is found in some anaerobic microorganisms, such as acetogenic and sulfate- The physiological functions of cooC, cooT and cooJ, when reducing bacteria and methanogenic archaea, where CO is present in CODH operons, have mostly been studied in the used as a source of carbon through the Wood–Ljungdahl photosynthetic bacterium Rhodospirillum rubrum. The Fig. 2 Operons encoding for CODH in Desulfovibrio vulgaris encoding for maturation proteins, in blue genes encoding for the CO- Hildenborough (Dv), Carboxydothermus hydrogenoformans Z-2901 dependent transcriptional activator cooA, in dark gray genes encod- (Ch), Rhodospirillum rubrum ATCC 11170 (Rr), Moorella ther- ing for ferredoxin-like proteins, in green genes encoding for hydroge- moacetica and Citrobacter amalonaticus Y19 (CaY19). The operons nase subunits, in gray-blue genes encoding for acetyl-CoA synthase coding for Ch CODH I and II are represented in this figure. Arrows subunits and in light gray genes encoding for hypothetical proteins of in red represent the cooS genes encoding for CODH, in yellow genes unknown functions 1 3 616 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 group of Ludden has shown that, in the absence of light, the Walker A motif (GKGGVGKTT) in the P-loop (phosphate- bacterium grows using CO as energy source. This growth binding loop). Zinc is bound covalently at the interface of depends on CODH: a Rr strain in which cooS has been inac- two monomers by two strictly conserved cysteines of each tivated grows normally under light but does not grow in the monomer (Cys112 and Cys114) (Fig.  3b). The structure presence of CO in the dark [16]. Strains of Rr in which the of Ch CooC-1 in the presence of nickel and/or ATP has genes coding for CooC, CooT or CooJ have been inactivated not been determined. However, Jeoung et al. showed that do not grow unless nickel is added to the growth medium CooC-1 binds one nickel per dimer with a K of 0.4 µM [35] [16]. The presence of CooC is particularly crucial, since only and that Zn and Ni compete for the same tetracoordination a large excess of nickel (650 µM) restores the growth of the site at the interface of the dimer [37]. Whether all CooC cooC-inactivated strain. Therefore, these accessory proteins proteins bind nickel is, however, unclear: the protein from are important to ensure that CODH is fully Ni-loaded and Rr does not tightly bind nickel (Ni content lower than 0.1 active when nickel is scarce. However, their precise action Ni ion per dimer) under the conditions used in the study of and their interplay remain unknown. They may be directly Jeon et al. [36]. involved in nickel delivery into the active site of CODH The available structures of Ch CooC-1 show that ADP or act indirectly by being involved in intracellular nickel binding induces conformational changes: the distance transport. Note that they are probably exclusively dedicated between the metal (Zn) and the thiolate group of cysteine to CODH maturation: they are not involved in import of 114 becomes shorter (from 2.5 to 2.2 Å) and a flexible loop, exogenous nickel [31] and inactivating these genes has no the CAP loop, deviates from the metal binding site, making effect of the Ni–Fe hydrogenase activity [32, 33]. it more open to metals [37]. Thus, the presence of ADP seems to favor nickel binding. The group of Dobbek has constructed a structural model of ATP-bound CooC, based Biochemical and structural characterization on the structure of an ATPase of the same family involved of the accessory proteins in chromosome segregation [38]. They proposed that ATP binding increases the distances between the cysteines of the CooC metal binding site, which induces the release of nickel [37]. CooC is a 30 kDa ATPase which belongs to the MRP/MinD CooT family of the SIMIBI class (Signal recognition GTPases, MinD superfamily and BioD superfamily) [34]. Rr CooC Timm et al. have recently resolved the crystal structure of and Ch CooC-1 hydrolyze ATP at a slow rate in vitro (k CooT from Rr [39]. It is a small protein of 7.1 kDa, which cat −1 of 5 nmol min for Ch CooC) [35, 36] Rr CooC also hydro- dimerizes and binds one nickel per dimer with a K of 9 nM. lyzes GTP. The in vitro NTPase rates, although low, are in The crystal structure of CooT shows that the protein is com- the same range as those of other Ni-metallochaperones posed of seven β-sheets. Site-directed mutagenesis and which hydrolyze GTP (e.g., HypB and ureG, which are circular dichroism experiments showed that the position 2 involved in the maturation of NiFe hydrogenase and urease, cysteine is involved in nickel binding. The authors hypoth- respectively) but hydrolysis might occur at a faster rate in esized that Cys2 from each monomer coordinates nickel. the cellular context. However, no crystal structure of Ni-bound CooT could be Ch CooC-1 is monomeric [35, 37] and forms a dimer in obtained. 2+ the presence of ADP and/or Zn (Fig. 3a). According to the Using a bioinformatic search on genomes, Timm et al. structure of Ch CooC-1 (Fig. 3), ADP binds to the deviant have identified 111 proteins homologous to CooT. Among Fig. 3 a Structure of CooC-1 from Carboxydothermus hydrogenoformans, PDB: 3KJI [40]. b An atom of zinc in gray is at the interface of the two monomers, bound by cysteines 112 and 114. ADP is bound to the deviant Walker A motif (GKGGVGKTT), shown in red 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 617 the CooT-containing proteomes, 85 also contain CODH [39], The heterologous production of the complex carbon mon- showing that CooT-dependent maturation of CODH is prob- oxide dehydrogenase/acetyl-coA synthase from Clostridium ably more widespread than was previously expected. carboxidivorans (Cc) in Clostridium acetobutylicum induces the production of CO by the host organism. In the absence CooJ of added nickel, the presence of Cc CooC enhances CO production. On the contrary, the CO production does not Rr CooJ is a small protein of 12.6 kDa which has a histidine- depend on Cc CooC when the medium is supplemented in rich domain (16 histidines in the C-terminal domain). Bio- nickel (50–100 µM). This study supports the idea that CooC chemical studies revealed that CooJ binds four nickel per facilitates nickel insertion into the active site of CODH when monomer with a dissociation constant of 4.3 µM [40], but nickel is at trace levels [42]. its precise role in Ni cluster assembly is unknown. CooC is not specific to the maturation of monofunctional CODH, the heterologous maturation of the CODH from Moorella thermoacetica, which is in complex with acetyl- CooC‑dependent maturation CoA synthase, also seems to depend on the presence of this accessory protein, based on the results of experiments where Let us first focus on the cases in which CODH are encoded the bifunctional enzyme was produced in E. coli [6]. in a operon also containing cooC. Proposed mechanisms of CooC‑driven nickel Production of carbon monoxide dehydrogenases insertion into the active site in the absence of CooC Two different mechanisms for the maturation of the active The heterologous production of Ch CODH-I in E. coli, in the site of CODH in which CooC is essential are proposed in absence of CooC-1 leads to the formation of a an enzyme the literature. In a first mechanism, CooC binds nickel and that contains three times less nickel, and is three times less inserts it into the active site of CODH in reaction that is cou- active than when it is expressed in the presence of CooC pled to ATP hydrolysis [37]. In a second mechanism, CooC [41]. Similarly, the  CODH from Desulfovibrio vulgaris acts as a chaperone that induces a conformational change of Hildenborough (Dv) produced heterologously in D. fruc- the active site of CODH by hydrolyzing ATP, and then the tosovorans in the absence of CooC contains less than 0.5 folded active site of CODH spontaneously binds Ni [36]. nickel per dimer, compared to 0.8–1.8 Ni/dimer when it is We argued that the latter mechanism is more likely to be co-produced with CooC [5]. Dv CODH is inactive when pro- operational, since the Dv CODH produced in the absence duced without CooC and cannot be activated with exogenous of its maturase CooC, contains hardly any nickel, is inactive Ni. In both cases, it therefore appears that CooC is important and cannot be activated in vitro (with nickel under reducing for nickel insertion into the active site, even if the effect is conditions) contrary to the enzyme that has been co-pro- more pronounced for the Dv enzyme. duced with CooC [5]. Similar observations were reported for Jeon et al. showed that the hydrolysis of ATP by CooC Rr CODH: a strain producing a deficient CooC produces a in Rr is necessary for nickel delivery and the production of nickel-depleted CODH which can only be partially activated a fully matured CODH [36]. Indeed, a mutated Rr strain with nickel under reducing conditions (to approximately in which CooC has no ATPase activity (K13Q CooC) pro- 15% of the wild-type CODH activity); these CODH cannot duces a nickel-deficient and almost inactive CODH. Addi- be fully activated in vitro, which suggests that the active tion of nickel to the culture medium does not compensate site has not the right conformation to bind nickel [5, 36]. for the deficiency of CooC [36]. As mentioned before, the As a matter of fact, conversely, when WT Rr is grown in a inactivation of cooC prevents the CODH-dependent growth Ni-depleted medium, the purified CODH does not contain unless a large concentration of nickel is added in the medium Ni but activates upon incubation with NiCl under reducing [16]. This suggests that growth can be sustained even if the conditions. This shows that the presence of CooC influences CODH is only partly Ni-loaded. the ability of CODH to bind Ni. CooC is the Ni donor in the The UV visible spectra of Dv CODH obtained in the case of Ch CooC-1, but it may be that this function is per- absence or in the presence of CooC are similar [5]. The formed by accessory proteins (CooJ and CooT) in the case absence of CooC does not affect the iron content of CODH, of organisms such as Rr, whose CooC does not bind Ni. In which suggests that CooC is not involved in the biosynthesis the case of Dv and Rr, free Ni can be delivered directly in the of the iron–sulfur clusters of CODH (including that of the active site of CODH, at least in vitro, provided the enzyme active site). These iron–sulfur clusters are probably produced has been co-produced with CooC. through generic iron–sulfur cluster assembly machineries, We depict the current hypothesis in Fig. 4. In the first while CooC is specifically devoted to the delivery of nickel. step, CooC acts as a chaperone which properly folds the 1 3 618 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 Maturation without any specific accessory protein Surprisingly, whereas the maturation of some CODH is strictly dependent on the presence of one or several matu- rases, other CODH can be fully matured in the absence of any specific accessory protein. In organisms that express several CODH, operons coding for certain CODH do not contain genes encoding for acces- sory proteins, which suggests that these CODH may either depend on accessory proteins encoded by other CODH oper- ons (for example, the Ch CODH-II and -IV operons do not contain a cooC copy but the CODH-I and III operons do) or that their maturation is not assisted by any accessory protein [27]. The heterologous production in E. coli of Ch CODH-II or -IV, for instance, leads to the formation of nickel-loaded, mature and active CODH [8, 46, 47]. It is important to note that in these studies E. coli is grown in the presence of high concentrations of nickel (more than 0.5 mM) to favor nickel insertion into CODH. It may be that the high concentration of nickel compensates for the absence of accessory proteins. This is probably the case for the CODH from Citrobacter amalonaticus Y19 (CaY19). The cooS gene from CaY19 Fig. 4 Proposed mechanisms of CooC-dependent CODH maturation. Step I: CooC derives energy from the hydrolysis of ATP to induce a precedes four genes encoding for accessory proteins: CooC, conformational change of the active site. Step II: the active site is in CooT, CooJ and HypB (Fig. 2) [48]. Some CODH activ- a favorable folding to receive nickel. The active site of CODH may ity could be measured in crude extracts of an E. coli strain acquire nickel from a nickel-loaded CooC, b other nickel-loaded in which the CaY19 CODH was produced in the absence accessory proteins (CooJ and CooT) or c free nickel available in the environment of the accessory proteins, suggesting that the latter are not strictly necessary for Ni delivery to the active site. How- ever, the actual metal content of the heterologously produced CODH was not determined so that it is unknown whether the enzyme was fully Ni-loaded. Surprisingly, among the active site in a conformation that makes it ready to bind proteins encoded in the coo operon, only CooF is essential Ni, at the cost of ATP hydrolysis. In a second step, nickel for the formation of an active CaY19 CODH. CooF is a binds to the active site according to three possible routes small iron–sulfur-containing protein (22 kDa for CooF of depending on the organism: (a) it can be delivered by Ni- Rr) which probably shuttles electrons from CODH to the loaded CooC itself if its affinity for Ni is high (Ch CODH- CO-induced Ni–Fe hydrogenase Coo [17, 49]. Its potential I), (b) if CooC has low affinity for Ni (Rr CODH), CooT function in CaY19 CODH maturation remains unknown. and CooJ assist it for the delivery of nickel into the folded Regarding heterologous expression, another hypothesis is active site, and (c) free Ni can possibly be inserted without that accessory proteins of other Ni-containing enzymes, such the need for another protein (Dv and Rr CODH). as Ni–Fe hydrogenases, from the heterologous host may be Like what we propose for some CooC proteins, matu- involved in the maturation of CODH [6, 48]. In Helicobacter rases with double function have already been described. pylori, it is well known that Ni–Fe hydrogenase accessory NifH, involved in the maturation of nitrogenase, for proteins can maturate another nickel enzyme, urease. Dele- instance, is an ATPase that belongs to the MinD family. It tions of hypA and hypB cause a decrease in urease activity binds a [4Fe–4S] cluster at the interface of the two mono- (40- and 200-fold, respectively) compared to the wild-type mers [43]. First, hydrolysis of ATP by NifH induces a strain [50]. Site-directed mutagenesis studies showed that conformational change which results in electron transfer the GTPase activity of HypA and the nickel-binding site from the [4Fe–4S] cluster to the nitrogenase. Then NifH of HypB are involved in the maturation of the urease of H. is involved in the biosynthesis of the molybdenum iron pylori [51, 52]. On the contrary, accessory proteins of urease cofactor, by delivering homocitrate and molybdenum [44, from H. pylori cannot maturate Ni–Fe hydrogenase [53, 54]. 45]. More in-depth studies and comparisons of CooC from Similar conclusions have been drawn regarding the metal- different sources are required to support our hypothesis. lochaperones CooC, CooT and CooJ from Rr which are not 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 619 Open Access This article is distributed under the terms of the Crea- involved in the maturation of the CO-induced Ni–Fe hydro- tive Commons Attribution 4.0 International License (http://creat iveco genase (referred to as Coo hydrogenase) [32]. Although mmons .org/licen ses/by/4.0/), which permits use, duplication, adapta- the cross-talk between maturation machineries of different tion, distribution and reproduction in any medium or format, as long Ni-containing enzymes is established, the involvement of as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes hydrogenase accessory proteins, such as HypA and HypB, were made. in the Ni-acquisition of CODH remains to be demonstrated. Last, we note that there are also examples of genomes that contain a CODH but no cooC gene, which definitely estab- lishes that CODH maturation can be CooC-independent. It is References so in C. acetobutylicum, whose gene CA_C0116 is annotated as a CODH and codes for a protein that has all amino acids 1. Ragsdale SW (2009) J Biol Chem 284:18571–18575 known to be essential for CODH function (this is unlike the 2. Boer JL, Mulrooney SB, Hausinger RP (2014) Arch Biochem CA_C2498 gene which is annotated as a CODH but lacks Biophys 544:142–152 3. Svetlitchnyi V, Peschel C, Acker G, Meyer O (2001) J Bacteriol one cysteine residue of the C-cluster and probably codes for 183:5134–5144 a hybrid cluster protein). Preliminary results in our group 4. Ensign SA, Bonam D, Ludden PW (1989) Biochemistry show that indeed, expression of CA_C0116 in the absence 28:4968–4973 of any CooC leads to a functional CODH. 5. Hadj-Said J, Pandelia ME, Leger C, Fourmond V, Dementin S (2015) Biochim Biophys Acta Bioenerg 1847:1574–1583 6. Loke HK, Bennett GN, Lindahl PA (2000) Proc Natl Acad Sci USA 97:12530–12535 7. Drennan CL, Heo J, Sintchak MD, Schreiter E, Ludden PW (2001) Concluding remarks Proc Natl Acad Sci USA 98:11973–11978 8. Dobbek H, Svetlitchnyi V, Gremer L, Huber R, Meyer O (2001) Although small in numbers, Ni-containing enzymes exhibit Science 293:1281–1285 9. Darnault C, Volbeda A, Kim EJ, Legrand P, Vernede X, Lindahl a remarkable diversity in terms of active site architecture PA, Fontecilla-Camps JC (2003) Nat Struct Biol 10:271–279 and catalytic properties. They are often involved in crucial 10. Ragsdale SW (2004) Crit Rev Biochem Mol Biol 39:165–195 metabolic pathways, catalyzing reactions related to carbon 11. King GM, Weber CF (2007) Nat Rev Microbiol 5:107–118 fixation, energy conversion or pathogenicity in many micro- 12. Cypionka H, Meyer O (1983) J Bacteriol 156:1178–1187 13. Hille R, Dingwall S, Wilcoxen J (2015) J Biol Inorg Chem organisms. The structure of the active site of some of these 20:243–251 enzymes is so elaborated that accessory proteins are neces- 14. Uffen RL (1983) J Bacteriol 155:956–965 sary to bind the Ni ion and insert it. This is the case for 15. Lindahl PA (2002) Biochemistry 41:2097–2105 NiFe hydrogenase, urease, lactate racemase and methyl-coM 16. Kerby RL, Ludden PW, Roberts GP (1995) J Bacteriol 177:2241–2244 reductase [55–58]. As also noted by Zeer-Wanklyn and Zam- 17. Singer SW, Hirst MB, Ludden PW (2006) Biochim Biophys Acta ble [59], the common point of these maturation processes 1757:1582–1591 is that they depend on an NTP hydrolysis, which supplies 18. Soboh B, Linder D, Hedderich R (2002) Eur J Biochem the required energy. Focusing on CODH, CooC (sometimes 269:5712–5721 19. Henstra AM, Dijkema C, Stams AJ (2007) Environ Microbiol assisted by CooJ and CooT), is an ATPase involved in Ni 9:1836–1841 insertion in the active site but the mechanism involved 20. Parshina SN, Sipma J, Henstra AM, Stams AJ (2010) Int J Micro- remains undetermined. Examination of the literature reveals biol 2010:319527 that not all CODH are CooC-dependent, which is surprising 21. Hocking WP, Roalkvam I, Magnussen C, Stokke R, Steen IH (2015) Archaea 2015:235384 considering that the active site of all Ni-containing CODH is 22. Diender M, Stams AJ, Sousa DZ (2015) Front Microbiol 6:1275 believed to be the same. It is not clear in these cases whether 23. Oelgeschlager E, Rother M (2009) FEMS Microbiol Lett CooC is substituted by the Ni insertion machinery of other 292:254–260 Ni-enzymes or if Ni spontaneously binds the active site. 24. Drennan CL, Doukov TI, Ragsdale SW (2004) J Biol Inorg Chem 9:511–515 Understanding how CooC works and how and why some 25. Volbeda A, Fontecilla-Camps JC (2004) J Biol Inorg Chem CODH do not depend on this accessory protein is certainly 9:525–532 part of the most fascinating questions in the field. 26. Seravalli J, Ragsdale SW (2000) Biochemistry 39:1274–1277 27. Wu M, Ren Q, Durkin AS, Daugherty SC, Brinkac LM, Dodson Acknowledgements The authors acknowledge financial support from RJ, Madupu R, Sullivan SA, Kolonay JF, Haft DH, Nelson WC, CNRS, Aix Marseille Université, Agence Nationale de la Recherche Tallon LJ, Jones KM, Ulrich LE, Gonzalez JM, Zhulin IB, Robb (ANR-12-BS08-0014, ANR-14-CE05-0010, ANR-15-CE05-0020, FT, Eisen JA (2005) PLoS Genet 1:e65 ANR-17-CE11-0027) and the A*MIDEX Grant (ANR-11- 28. Ragsdale SW, Pierce E (2008) Biochim Biophys Acta IDEX-0001-02) funded by the French Government “Investissements 1784:1873–1898 d’Avenir” program. M.L. thanks the Erasmus program for funding. 29. Can M, Armstrong FA, Ragsdale SW (2014) Chem Rev M.M., M.B., C.L., V.F. and S.D. are members of FrenchBIC (http:// 114:4149–4174 frenc hbic.cnrs.fr). 30. Ainala SK, Seol E, Park S (2015) J Biotechnol 211:79–80 1 3 620 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 31. Watt RK, Ludden PW (1999) J Bacteriol 181:4554–4560 46. Inoue T, Yoshida T, Wada K, Daifuku T, Fukuyama K, Sako Y 32. Fox JD, Kerby RL, Roberts GP, Ludden PW (1996) J Bacteriol (2011) Biosci Biotechnol Biochem 75:1392–1394 178:1515–1524 47. Domnik L, Merrouch M, Goetzl S, Jeoung JH, Leger C, Demen- 33. Fox JD, He Y, Shelver D, Roberts GP, Ludden PW (1996) J Bac- tin S, Fourmond V, Dobbek H (2017) Angew Chem Int Ed Engl teriol 178:6200–6208 56:15466–15469 34. Leipe DD, Wolf YI, Koonin EV, Aravind L (2002) J Mol Biol 48. Sekar BS, Raj SM, Seol E, Ainala SK, Lee J, Park S (2014) Int J 317:41–72 Hydrogen Energ 39:15446–15454 35. Jeoung JH, Giese T, Grunwald M, Dobbek H (2009) Biochemistry 49. Ensign SA, Ludden PW (1991) J Biol Chem 266:18395–18403 48:11505–11513 50. Hube M, Blokesch M, Bock A (2002) J Bacteriol 184:3879–3885 36. Jeon WB, Cheng J, Ludden PW (2001) J Biol Chem 51. Mehta N, Olson JW, Maier RJ (2003) J Bacteriol 185:726–734 276:38602–38609 52. Mehta NS, Benoit S, Mysore JV, Sousa RS, Maier RJ (2005) 37. Jeoung JH, Giese T, Grunwald M, Dobbek H (2010) J Mol Biol Infect Immun 73:5311–5318 396:1165–1179 53. Olson JW, Mehta NS, Maier RJ (2001) Mol Microbiol 39:176–182 38. Draper GC, Gober JW (2002) Annu Rev Microbiol 56:567–597 54. Benoit SL, Zbell AL, Maier RJ (2007) Microbiology 39. Timm J, Brochier-Armanet C, Perard J, Zambelli B, Ollagnier-de- 153:3748–3756 Choudens S, Ciurli S, Cavazza C (2017) Metallomics 9:575–583 55. Lacasse MJ, Zamble DB (2016) Biochemistry 55:1689–1701 40. Watt RK, Ludden PW (1998) J Biol Chem 273:10019–10025 56. Farrugia MA, Macomber L, Hausinger RP (2013) J Biol Chem 41. Inoue T, Takao K, Fukuyama Y, Yoshida T, Sako Y (2014) Biosci 288:13178–13185 Biotechnol Biochem 78:582–587 57. Desguin B, Soumillion P, Hausinger RP, Hols P (2017) FEMS 42. Carlson ED, Papoutsakis ET (2017) Appl Environ Microbiol Microbiol Rev 41:S71–S83 83:e00829–17 58. Zheng K, Ngo PD, Owens VL, Yang XP, Mansoorabadi SO 43. Georgiadis MM, Komiya H, Chakrabarti P, Woo D, Kornuc JJ, (2016) Science 354:339–342 Rees DC (1992) Science 257:1653–1659 59. Zeer-Wanklyn CJ, Zamble DB (2017) Curr Opin Chem Biol 44. Rubio LM, Ludden PW (2005) J Bacteriol 187:405–414 37:80–88 45. Hu Y, Ribbe MW (2016) Annu Rev Biochem 85:455–483 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JBIC Journal of Biological Inorganic Chemistry Springer Journals

Maturation of the [Ni–4Fe–4S] active site of carbon monoxide dehydrogenases

Free
8 pages

Loading next page...
 
/lp/springer_journal/maturation-of-the-ni-4fe-4s-active-site-of-carbon-monoxide-Sl4Gi2Aocd
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Biochemistry, general; Microbiology
ISSN
0949-8257
eISSN
1432-1327
D.O.I.
10.1007/s00775-018-1541-0
Publisher site
See Article on Publisher Site

Abstract

Nickel-containing enzymes are diverse in terms of function and active site structure. In many cases, the biosynthesis of the active site depends on accessory proteins which transport and insert the Ni ion. We review and discuss the literature related to the maturation of carbon monoxide dehydrogenases (CODH) which bear a nickel-containing active site consisting of a [Ni–4Fe–4S] center called the C-cluster. The maturation of this center has been much less studied than that of other nickel- containing enzymes such as urease and NiFe hydrogenase. Several proteins present in certain CODH operons, including the nickel-binding proteins CooT and CooJ, still have unclear functions. We question the conception that the maturation of all CODH depends on the accessory protein CooC described as essential for nickel insertion into the active site. The available literature reveals biological variations in CODH active site biosynthesis. Keywords Carbon monoxide dehydrogenase · Active site · Iron–sulfur cluster · Maturation Introduction cofactor (pincer) linked to a Lys of the protein backbone and methyl-coM reductase in which Ni is part of coenzyme Nine nickel-containing enzymes have been discovered and F430. Depending on the enzyme, Ni participates in cataly- characterized so far, but there exists certainly others [1, 2]. sis either by acting as a Lewis acid or by promoting redox Among them, seven consume or produce small molecules chemistry. (hydrogenase, carbon monoxide dehydrogenase, superox- Although the insertion of Ni seems spontaneous in gly- ide dismutase, urease, acireductone dioxygenase, methyl- oxylase I and acireductone dioxygenase, it requires dedi- coM reductase and acetyl-CoA synthase). The other two cated biological machineries in the other cases. The present (glyoxylase I and lactate racemase) are involved in lactate review focuses on the carbon monoxide dehydrogenases metabolism. The structures of their active sites are diverse from anaerobic microorganisms (Ni-containing CODH) in terms of nature and number of ligands to the Ni. In most which catalyze the reversible oxidation of CO with high cases, the Ni is coordinated by acidic residues (Cys, His, turnover frequencies [3–6]. These enzymes bear a nickel- Glu, Asp, carbonylated Lys) or water molecules. The three containing active site, the so-called C-cluster, which consists exceptions are carbon monoxide dehydrogenase, in which of a [Ni–3Fe–4S] cubane connected to a unique iron site Ni is also coordinated to inorganic sulfur in a [Ni–4Fe–4S] through a linking sulfide [ 7–9]. cluster, lactate racemase in which Ni is part of a non-protein Generalities The original version of this article was revised due to a retrospective Open Access order. Some microorganisms can grow in the presence of carbon monoxide, which they use as a source of carbon and/or * Sébastien Dementin energy [10, 11]. The oxidation of CO to CO by these micro- dementin@imm.cnrs.fr organisms is catalyzed by carbon monoxide dehydrogenases Aix-Marseille Université, CNRS, BIP UMR 7281, Institut (CODH). Nevertheless, CODH from aerobic and anaerobic de Microbiologie de la Méditerranée, 31 Chemin J. Aiguier, bacteria are not phylogenetically related and have distinct 13402 Marseille Cedex 20, France Vol.:(0123456789) 1 3 614 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 structures and kinetic properties. Most aerobic CO-utilizing to the xanthine oxidase family; their active site a binuclear bacteria (carboxydotrophs) oxidize CO in their respiratory cluster of Mo and Cu (MoCu–CODH) (Fig. 1a, d). These chain [12] using a variety of acceptors such as O (Oligo- MoCu–CODH only catalyze the oxidation of CO (not the −1 tropha carboxidovorans [13]) or nitrate for dissimilatory reduction of C O ) with a turnover frequency of up to 100 s nitrate reduction (Burkholderia xenovorans LB400 [11]). [13]. Some photosynthetic bacteria, such as Rhodopseudomonas Here we focus on the CODH from anaerobic microor- gelatinosa, can use CO as a carbon source by first convert- ganisms (Ni-containing CODH), which we will abbrevi- ing it into CO , which is then reduced into carbohydrate ate CODH in the following text for clarity. These enzymes through the Calvin–Benson–Bassham cycle [14]. CODH contain a unique nickel-containing active site [Ni–4Fe–4S], from aerobic bacteria are heterotrimeric enzymes and belong called the C-cluster. This cluster consists of a [Ni–3Fe–4S] Fig. 1 Structures of prototypi- cal Ni-containing CODH and their active sites. a Structure of the MoCu-CODH from Oligotropha carboxidovorans, PDB: 1N63 [30]. The L-subunit (89 kDa) in yellow, contains the active site. The S-subunit (18 kDa) in green, contains the two [2Fe–2S] clusters and the M-subunit (30 kDa) binds a FAD cofactor. b Structure of the CODH-II from Carboxydother- mus hydrogenoformans (Ch), PDB: 3B53 [31]. Each subunit (67 kDa) of this homodimer is colored in red or in orange. c Structure of the CODH from Methanosarcina barkeri, PDB: 3CF4 [32]. Subunit α (89 kDa) containing the nine iron–sul- fur clusters is colored in blue and the subunit ε (20 kDa) is colored in black. Structures of the active sites of the MoCu– CODH from Oligotropha carboxidovorans (d) and of the CODH-II from Ch (e). The colors for the atoms in d, e are green for Ni, orange for Fe, yel- low for S, red for O, blue for N, turquoise for Mo, light orange for Cu and white for C 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 615 cubane connected to a unique iron site through a linking pathway [26–29]. The major function of CODH in this meta- sulfide (Fig.  1e) [7–9]. CODH catalyze the oxidation of CO bolic pathway requires coordination of CO reduction at the with turnover frequencies ranging from hundreds to tens of C-cluster with CO channeling and reaction with a methyl thousands turnovers per second [3–6]. CODH-I from Car- group and CoA at the A-cluster active site of ACS. boxydothermus hydrogenoformans (Ch) is the most active, This article reviews the mechanisms and accessory pro- −1 with a turnover rate of 39,000 s at 70 °C, pH 8 [3]. Con- teins/chaperones involved in the maturation of this unique trary to Mo–CODH, CODH also catalyze the reverse reac- C-cluster. There has been a general agreement that CODH tion (the reduction of C O into CO). These enzymes are maturation depends on the accessory protein CooC, but on divided in four classes depending on their structures and the basis of the recent literature, it appears that the matu- functions [15]. CODH of classes I, II and III are bifunc- ration of CODH may differ from one enzyme to another, tional, they form a complex with acetyl-CoA synthase (ACS/ and that in some cases, it is independent of any specific CODH), while class IV CODH are monofunctional. Class I maturase. and II CODH form heterotetramers containing nine iron–sul- fur clusters (Fig. 1c). CODH of class III and IV are homodi- meric, and contain three iron–sulfur clusters (Fig. 1b). Maturation of the [Ni–4Fe–4S] C‑cluster As exemplified by the photosynthetic bacterium Rho- of CODH dospirillum rubrum (Rr), CO can be used as an alternative to light as a source of energy [16]. It is hypothesized that The diversity of the operons encoding for CODH (Fig. 2) cytoplasmic oxidation of CO is coupled to the reduction of suggests a variability in the maturation mechanisms. Some protons. The product dihydrogen would diffuse through the operons encoding for a CODH do not contain any maturase internal membrane and be oxidized by a periplasmic hydro- (e.g., CODH-II from Ch [27]), whereas many of them con- genase, leading to the formation of a proton motive force tain the maturation gene cooC. Other accessory genes can [3, 17, 18]. Some anaerobic sulfate reducing microorgan- be present in the CODH operons in addition to cooC: for isms couple the oxidation of CO to the reduction of sulfate example, in Rhodospirillum rubrum (Rr) [16] and Citro- [19–21] and some methanogenic archaea use CO as carbon/ bacter amalonaticus Y19 (CaY19) [30], the operons also electron donor for methane formation [22, 23]. contain the cooJ and cooT genes. In cases where CODH is in complex with acetyl-CoA synthase (ACS), and is therefore called bifunctional, it reduces CO to CO which is channeled to ACS, whose active Physiological importance of accessory site (called A-cluster) is a bi-Ni center attached to an Fe–S proteins cluster [24, 25]. This CODH/ACS complex is found in some anaerobic microorganisms, such as acetogenic and sulfate- The physiological functions of cooC, cooT and cooJ, when reducing bacteria and methanogenic archaea, where CO is present in CODH operons, have mostly been studied in the used as a source of carbon through the Wood–Ljungdahl photosynthetic bacterium Rhodospirillum rubrum. The Fig. 2 Operons encoding for CODH in Desulfovibrio vulgaris encoding for maturation proteins, in blue genes encoding for the CO- Hildenborough (Dv), Carboxydothermus hydrogenoformans Z-2901 dependent transcriptional activator cooA, in dark gray genes encod- (Ch), Rhodospirillum rubrum ATCC 11170 (Rr), Moorella ther- ing for ferredoxin-like proteins, in green genes encoding for hydroge- moacetica and Citrobacter amalonaticus Y19 (CaY19). The operons nase subunits, in gray-blue genes encoding for acetyl-CoA synthase coding for Ch CODH I and II are represented in this figure. Arrows subunits and in light gray genes encoding for hypothetical proteins of in red represent the cooS genes encoding for CODH, in yellow genes unknown functions 1 3 616 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 group of Ludden has shown that, in the absence of light, the Walker A motif (GKGGVGKTT) in the P-loop (phosphate- bacterium grows using CO as energy source. This growth binding loop). Zinc is bound covalently at the interface of depends on CODH: a Rr strain in which cooS has been inac- two monomers by two strictly conserved cysteines of each tivated grows normally under light but does not grow in the monomer (Cys112 and Cys114) (Fig.  3b). The structure presence of CO in the dark [16]. Strains of Rr in which the of Ch CooC-1 in the presence of nickel and/or ATP has genes coding for CooC, CooT or CooJ have been inactivated not been determined. However, Jeoung et al. showed that do not grow unless nickel is added to the growth medium CooC-1 binds one nickel per dimer with a K of 0.4 µM [35] [16]. The presence of CooC is particularly crucial, since only and that Zn and Ni compete for the same tetracoordination a large excess of nickel (650 µM) restores the growth of the site at the interface of the dimer [37]. Whether all CooC cooC-inactivated strain. Therefore, these accessory proteins proteins bind nickel is, however, unclear: the protein from are important to ensure that CODH is fully Ni-loaded and Rr does not tightly bind nickel (Ni content lower than 0.1 active when nickel is scarce. However, their precise action Ni ion per dimer) under the conditions used in the study of and their interplay remain unknown. They may be directly Jeon et al. [36]. involved in nickel delivery into the active site of CODH The available structures of Ch CooC-1 show that ADP or act indirectly by being involved in intracellular nickel binding induces conformational changes: the distance transport. Note that they are probably exclusively dedicated between the metal (Zn) and the thiolate group of cysteine to CODH maturation: they are not involved in import of 114 becomes shorter (from 2.5 to 2.2 Å) and a flexible loop, exogenous nickel [31] and inactivating these genes has no the CAP loop, deviates from the metal binding site, making effect of the Ni–Fe hydrogenase activity [32, 33]. it more open to metals [37]. Thus, the presence of ADP seems to favor nickel binding. The group of Dobbek has constructed a structural model of ATP-bound CooC, based Biochemical and structural characterization on the structure of an ATPase of the same family involved of the accessory proteins in chromosome segregation [38]. They proposed that ATP binding increases the distances between the cysteines of the CooC metal binding site, which induces the release of nickel [37]. CooC is a 30 kDa ATPase which belongs to the MRP/MinD CooT family of the SIMIBI class (Signal recognition GTPases, MinD superfamily and BioD superfamily) [34]. Rr CooC Timm et al. have recently resolved the crystal structure of and Ch CooC-1 hydrolyze ATP at a slow rate in vitro (k CooT from Rr [39]. It is a small protein of 7.1 kDa, which cat −1 of 5 nmol min for Ch CooC) [35, 36] Rr CooC also hydro- dimerizes and binds one nickel per dimer with a K of 9 nM. lyzes GTP. The in vitro NTPase rates, although low, are in The crystal structure of CooT shows that the protein is com- the same range as those of other Ni-metallochaperones posed of seven β-sheets. Site-directed mutagenesis and which hydrolyze GTP (e.g., HypB and ureG, which are circular dichroism experiments showed that the position 2 involved in the maturation of NiFe hydrogenase and urease, cysteine is involved in nickel binding. The authors hypoth- respectively) but hydrolysis might occur at a faster rate in esized that Cys2 from each monomer coordinates nickel. the cellular context. However, no crystal structure of Ni-bound CooT could be Ch CooC-1 is monomeric [35, 37] and forms a dimer in obtained. 2+ the presence of ADP and/or Zn (Fig. 3a). According to the Using a bioinformatic search on genomes, Timm et al. structure of Ch CooC-1 (Fig. 3), ADP binds to the deviant have identified 111 proteins homologous to CooT. Among Fig. 3 a Structure of CooC-1 from Carboxydothermus hydrogenoformans, PDB: 3KJI [40]. b An atom of zinc in gray is at the interface of the two monomers, bound by cysteines 112 and 114. ADP is bound to the deviant Walker A motif (GKGGVGKTT), shown in red 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 617 the CooT-containing proteomes, 85 also contain CODH [39], The heterologous production of the complex carbon mon- showing that CooT-dependent maturation of CODH is prob- oxide dehydrogenase/acetyl-coA synthase from Clostridium ably more widespread than was previously expected. carboxidivorans (Cc) in Clostridium acetobutylicum induces the production of CO by the host organism. In the absence CooJ of added nickel, the presence of Cc CooC enhances CO production. On the contrary, the CO production does not Rr CooJ is a small protein of 12.6 kDa which has a histidine- depend on Cc CooC when the medium is supplemented in rich domain (16 histidines in the C-terminal domain). Bio- nickel (50–100 µM). This study supports the idea that CooC chemical studies revealed that CooJ binds four nickel per facilitates nickel insertion into the active site of CODH when monomer with a dissociation constant of 4.3 µM [40], but nickel is at trace levels [42]. its precise role in Ni cluster assembly is unknown. CooC is not specific to the maturation of monofunctional CODH, the heterologous maturation of the CODH from Moorella thermoacetica, which is in complex with acetyl- CooC‑dependent maturation CoA synthase, also seems to depend on the presence of this accessory protein, based on the results of experiments where Let us first focus on the cases in which CODH are encoded the bifunctional enzyme was produced in E. coli [6]. in a operon also containing cooC. Proposed mechanisms of CooC‑driven nickel Production of carbon monoxide dehydrogenases insertion into the active site in the absence of CooC Two different mechanisms for the maturation of the active The heterologous production of Ch CODH-I in E. coli, in the site of CODH in which CooC is essential are proposed in absence of CooC-1 leads to the formation of a an enzyme the literature. In a first mechanism, CooC binds nickel and that contains three times less nickel, and is three times less inserts it into the active site of CODH in reaction that is cou- active than when it is expressed in the presence of CooC pled to ATP hydrolysis [37]. In a second mechanism, CooC [41]. Similarly, the  CODH from Desulfovibrio vulgaris acts as a chaperone that induces a conformational change of Hildenborough (Dv) produced heterologously in D. fruc- the active site of CODH by hydrolyzing ATP, and then the tosovorans in the absence of CooC contains less than 0.5 folded active site of CODH spontaneously binds Ni [36]. nickel per dimer, compared to 0.8–1.8 Ni/dimer when it is We argued that the latter mechanism is more likely to be co-produced with CooC [5]. Dv CODH is inactive when pro- operational, since the Dv CODH produced in the absence duced without CooC and cannot be activated with exogenous of its maturase CooC, contains hardly any nickel, is inactive Ni. In both cases, it therefore appears that CooC is important and cannot be activated in vitro (with nickel under reducing for nickel insertion into the active site, even if the effect is conditions) contrary to the enzyme that has been co-pro- more pronounced for the Dv enzyme. duced with CooC [5]. Similar observations were reported for Jeon et al. showed that the hydrolysis of ATP by CooC Rr CODH: a strain producing a deficient CooC produces a in Rr is necessary for nickel delivery and the production of nickel-depleted CODH which can only be partially activated a fully matured CODH [36]. Indeed, a mutated Rr strain with nickel under reducing conditions (to approximately in which CooC has no ATPase activity (K13Q CooC) pro- 15% of the wild-type CODH activity); these CODH cannot duces a nickel-deficient and almost inactive CODH. Addi- be fully activated in vitro, which suggests that the active tion of nickel to the culture medium does not compensate site has not the right conformation to bind nickel [5, 36]. for the deficiency of CooC [36]. As mentioned before, the As a matter of fact, conversely, when WT Rr is grown in a inactivation of cooC prevents the CODH-dependent growth Ni-depleted medium, the purified CODH does not contain unless a large concentration of nickel is added in the medium Ni but activates upon incubation with NiCl under reducing [16]. This suggests that growth can be sustained even if the conditions. This shows that the presence of CooC influences CODH is only partly Ni-loaded. the ability of CODH to bind Ni. CooC is the Ni donor in the The UV visible spectra of Dv CODH obtained in the case of Ch CooC-1, but it may be that this function is per- absence or in the presence of CooC are similar [5]. The formed by accessory proteins (CooJ and CooT) in the case absence of CooC does not affect the iron content of CODH, of organisms such as Rr, whose CooC does not bind Ni. In which suggests that CooC is not involved in the biosynthesis the case of Dv and Rr, free Ni can be delivered directly in the of the iron–sulfur clusters of CODH (including that of the active site of CODH, at least in vitro, provided the enzyme active site). These iron–sulfur clusters are probably produced has been co-produced with CooC. through generic iron–sulfur cluster assembly machineries, We depict the current hypothesis in Fig. 4. In the first while CooC is specifically devoted to the delivery of nickel. step, CooC acts as a chaperone which properly folds the 1 3 618 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 Maturation without any specific accessory protein Surprisingly, whereas the maturation of some CODH is strictly dependent on the presence of one or several matu- rases, other CODH can be fully matured in the absence of any specific accessory protein. In organisms that express several CODH, operons coding for certain CODH do not contain genes encoding for acces- sory proteins, which suggests that these CODH may either depend on accessory proteins encoded by other CODH oper- ons (for example, the Ch CODH-II and -IV operons do not contain a cooC copy but the CODH-I and III operons do) or that their maturation is not assisted by any accessory protein [27]. The heterologous production in E. coli of Ch CODH-II or -IV, for instance, leads to the formation of nickel-loaded, mature and active CODH [8, 46, 47]. It is important to note that in these studies E. coli is grown in the presence of high concentrations of nickel (more than 0.5 mM) to favor nickel insertion into CODH. It may be that the high concentration of nickel compensates for the absence of accessory proteins. This is probably the case for the CODH from Citrobacter amalonaticus Y19 (CaY19). The cooS gene from CaY19 Fig. 4 Proposed mechanisms of CooC-dependent CODH maturation. Step I: CooC derives energy from the hydrolysis of ATP to induce a precedes four genes encoding for accessory proteins: CooC, conformational change of the active site. Step II: the active site is in CooT, CooJ and HypB (Fig. 2) [48]. Some CODH activ- a favorable folding to receive nickel. The active site of CODH may ity could be measured in crude extracts of an E. coli strain acquire nickel from a nickel-loaded CooC, b other nickel-loaded in which the CaY19 CODH was produced in the absence accessory proteins (CooJ and CooT) or c free nickel available in the environment of the accessory proteins, suggesting that the latter are not strictly necessary for Ni delivery to the active site. How- ever, the actual metal content of the heterologously produced CODH was not determined so that it is unknown whether the enzyme was fully Ni-loaded. Surprisingly, among the active site in a conformation that makes it ready to bind proteins encoded in the coo operon, only CooF is essential Ni, at the cost of ATP hydrolysis. In a second step, nickel for the formation of an active CaY19 CODH. CooF is a binds to the active site according to three possible routes small iron–sulfur-containing protein (22 kDa for CooF of depending on the organism: (a) it can be delivered by Ni- Rr) which probably shuttles electrons from CODH to the loaded CooC itself if its affinity for Ni is high (Ch CODH- CO-induced Ni–Fe hydrogenase Coo [17, 49]. Its potential I), (b) if CooC has low affinity for Ni (Rr CODH), CooT function in CaY19 CODH maturation remains unknown. and CooJ assist it for the delivery of nickel into the folded Regarding heterologous expression, another hypothesis is active site, and (c) free Ni can possibly be inserted without that accessory proteins of other Ni-containing enzymes, such the need for another protein (Dv and Rr CODH). as Ni–Fe hydrogenases, from the heterologous host may be Like what we propose for some CooC proteins, matu- involved in the maturation of CODH [6, 48]. In Helicobacter rases with double function have already been described. pylori, it is well known that Ni–Fe hydrogenase accessory NifH, involved in the maturation of nitrogenase, for proteins can maturate another nickel enzyme, urease. Dele- instance, is an ATPase that belongs to the MinD family. It tions of hypA and hypB cause a decrease in urease activity binds a [4Fe–4S] cluster at the interface of the two mono- (40- and 200-fold, respectively) compared to the wild-type mers [43]. First, hydrolysis of ATP by NifH induces a strain [50]. Site-directed mutagenesis studies showed that conformational change which results in electron transfer the GTPase activity of HypA and the nickel-binding site from the [4Fe–4S] cluster to the nitrogenase. Then NifH of HypB are involved in the maturation of the urease of H. is involved in the biosynthesis of the molybdenum iron pylori [51, 52]. On the contrary, accessory proteins of urease cofactor, by delivering homocitrate and molybdenum [44, from H. pylori cannot maturate Ni–Fe hydrogenase [53, 54]. 45]. More in-depth studies and comparisons of CooC from Similar conclusions have been drawn regarding the metal- different sources are required to support our hypothesis. lochaperones CooC, CooT and CooJ from Rr which are not 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 619 Open Access This article is distributed under the terms of the Crea- involved in the maturation of the CO-induced Ni–Fe hydro- tive Commons Attribution 4.0 International License (http://creat iveco genase (referred to as Coo hydrogenase) [32]. Although mmons .org/licen ses/by/4.0/), which permits use, duplication, adapta- the cross-talk between maturation machineries of different tion, distribution and reproduction in any medium or format, as long Ni-containing enzymes is established, the involvement of as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes hydrogenase accessory proteins, such as HypA and HypB, were made. in the Ni-acquisition of CODH remains to be demonstrated. Last, we note that there are also examples of genomes that contain a CODH but no cooC gene, which definitely estab- lishes that CODH maturation can be CooC-independent. It is References so in C. acetobutylicum, whose gene CA_C0116 is annotated as a CODH and codes for a protein that has all amino acids 1. Ragsdale SW (2009) J Biol Chem 284:18571–18575 known to be essential for CODH function (this is unlike the 2. Boer JL, Mulrooney SB, Hausinger RP (2014) Arch Biochem CA_C2498 gene which is annotated as a CODH but lacks Biophys 544:142–152 3. Svetlitchnyi V, Peschel C, Acker G, Meyer O (2001) J Bacteriol one cysteine residue of the C-cluster and probably codes for 183:5134–5144 a hybrid cluster protein). Preliminary results in our group 4. Ensign SA, Bonam D, Ludden PW (1989) Biochemistry show that indeed, expression of CA_C0116 in the absence 28:4968–4973 of any CooC leads to a functional CODH. 5. Hadj-Said J, Pandelia ME, Leger C, Fourmond V, Dementin S (2015) Biochim Biophys Acta Bioenerg 1847:1574–1583 6. Loke HK, Bennett GN, Lindahl PA (2000) Proc Natl Acad Sci USA 97:12530–12535 7. Drennan CL, Heo J, Sintchak MD, Schreiter E, Ludden PW (2001) Concluding remarks Proc Natl Acad Sci USA 98:11973–11978 8. Dobbek H, Svetlitchnyi V, Gremer L, Huber R, Meyer O (2001) Although small in numbers, Ni-containing enzymes exhibit Science 293:1281–1285 9. Darnault C, Volbeda A, Kim EJ, Legrand P, Vernede X, Lindahl a remarkable diversity in terms of active site architecture PA, Fontecilla-Camps JC (2003) Nat Struct Biol 10:271–279 and catalytic properties. They are often involved in crucial 10. Ragsdale SW (2004) Crit Rev Biochem Mol Biol 39:165–195 metabolic pathways, catalyzing reactions related to carbon 11. King GM, Weber CF (2007) Nat Rev Microbiol 5:107–118 fixation, energy conversion or pathogenicity in many micro- 12. Cypionka H, Meyer O (1983) J Bacteriol 156:1178–1187 13. Hille R, Dingwall S, Wilcoxen J (2015) J Biol Inorg Chem organisms. The structure of the active site of some of these 20:243–251 enzymes is so elaborated that accessory proteins are neces- 14. Uffen RL (1983) J Bacteriol 155:956–965 sary to bind the Ni ion and insert it. This is the case for 15. Lindahl PA (2002) Biochemistry 41:2097–2105 NiFe hydrogenase, urease, lactate racemase and methyl-coM 16. Kerby RL, Ludden PW, Roberts GP (1995) J Bacteriol 177:2241–2244 reductase [55–58]. As also noted by Zeer-Wanklyn and Zam- 17. Singer SW, Hirst MB, Ludden PW (2006) Biochim Biophys Acta ble [59], the common point of these maturation processes 1757:1582–1591 is that they depend on an NTP hydrolysis, which supplies 18. Soboh B, Linder D, Hedderich R (2002) Eur J Biochem the required energy. Focusing on CODH, CooC (sometimes 269:5712–5721 19. Henstra AM, Dijkema C, Stams AJ (2007) Environ Microbiol assisted by CooJ and CooT), is an ATPase involved in Ni 9:1836–1841 insertion in the active site but the mechanism involved 20. Parshina SN, Sipma J, Henstra AM, Stams AJ (2010) Int J Micro- remains undetermined. Examination of the literature reveals biol 2010:319527 that not all CODH are CooC-dependent, which is surprising 21. Hocking WP, Roalkvam I, Magnussen C, Stokke R, Steen IH (2015) Archaea 2015:235384 considering that the active site of all Ni-containing CODH is 22. Diender M, Stams AJ, Sousa DZ (2015) Front Microbiol 6:1275 believed to be the same. It is not clear in these cases whether 23. Oelgeschlager E, Rother M (2009) FEMS Microbiol Lett CooC is substituted by the Ni insertion machinery of other 292:254–260 Ni-enzymes or if Ni spontaneously binds the active site. 24. Drennan CL, Doukov TI, Ragsdale SW (2004) J Biol Inorg Chem 9:511–515 Understanding how CooC works and how and why some 25. Volbeda A, Fontecilla-Camps JC (2004) J Biol Inorg Chem CODH do not depend on this accessory protein is certainly 9:525–532 part of the most fascinating questions in the field. 26. Seravalli J, Ragsdale SW (2000) Biochemistry 39:1274–1277 27. Wu M, Ren Q, Durkin AS, Daugherty SC, Brinkac LM, Dodson Acknowledgements The authors acknowledge financial support from RJ, Madupu R, Sullivan SA, Kolonay JF, Haft DH, Nelson WC, CNRS, Aix Marseille Université, Agence Nationale de la Recherche Tallon LJ, Jones KM, Ulrich LE, Gonzalez JM, Zhulin IB, Robb (ANR-12-BS08-0014, ANR-14-CE05-0010, ANR-15-CE05-0020, FT, Eisen JA (2005) PLoS Genet 1:e65 ANR-17-CE11-0027) and the A*MIDEX Grant (ANR-11- 28. Ragsdale SW, Pierce E (2008) Biochim Biophys Acta IDEX-0001-02) funded by the French Government “Investissements 1784:1873–1898 d’Avenir” program. M.L. thanks the Erasmus program for funding. 29. Can M, Armstrong FA, Ragsdale SW (2014) Chem Rev M.M., M.B., C.L., V.F. and S.D. are members of FrenchBIC (http:// 114:4149–4174 frenc hbic.cnrs.fr). 30. Ainala SK, Seol E, Park S (2015) J Biotechnol 211:79–80 1 3 620 JBIC Journal of Biological Inorganic Chemistry (2018) 23:613–620 31. Watt RK, Ludden PW (1999) J Bacteriol 181:4554–4560 46. Inoue T, Yoshida T, Wada K, Daifuku T, Fukuyama K, Sako Y 32. Fox JD, Kerby RL, Roberts GP, Ludden PW (1996) J Bacteriol (2011) Biosci Biotechnol Biochem 75:1392–1394 178:1515–1524 47. Domnik L, Merrouch M, Goetzl S, Jeoung JH, Leger C, Demen- 33. Fox JD, He Y, Shelver D, Roberts GP, Ludden PW (1996) J Bac- tin S, Fourmond V, Dobbek H (2017) Angew Chem Int Ed Engl teriol 178:6200–6208 56:15466–15469 34. Leipe DD, Wolf YI, Koonin EV, Aravind L (2002) J Mol Biol 48. Sekar BS, Raj SM, Seol E, Ainala SK, Lee J, Park S (2014) Int J 317:41–72 Hydrogen Energ 39:15446–15454 35. Jeoung JH, Giese T, Grunwald M, Dobbek H (2009) Biochemistry 49. Ensign SA, Ludden PW (1991) J Biol Chem 266:18395–18403 48:11505–11513 50. Hube M, Blokesch M, Bock A (2002) J Bacteriol 184:3879–3885 36. Jeon WB, Cheng J, Ludden PW (2001) J Biol Chem 51. Mehta N, Olson JW, Maier RJ (2003) J Bacteriol 185:726–734 276:38602–38609 52. Mehta NS, Benoit S, Mysore JV, Sousa RS, Maier RJ (2005) 37. Jeoung JH, Giese T, Grunwald M, Dobbek H (2010) J Mol Biol Infect Immun 73:5311–5318 396:1165–1179 53. Olson JW, Mehta NS, Maier RJ (2001) Mol Microbiol 39:176–182 38. Draper GC, Gober JW (2002) Annu Rev Microbiol 56:567–597 54. Benoit SL, Zbell AL, Maier RJ (2007) Microbiology 39. Timm J, Brochier-Armanet C, Perard J, Zambelli B, Ollagnier-de- 153:3748–3756 Choudens S, Ciurli S, Cavazza C (2017) Metallomics 9:575–583 55. Lacasse MJ, Zamble DB (2016) Biochemistry 55:1689–1701 40. Watt RK, Ludden PW (1998) J Biol Chem 273:10019–10025 56. Farrugia MA, Macomber L, Hausinger RP (2013) J Biol Chem 41. Inoue T, Takao K, Fukuyama Y, Yoshida T, Sako Y (2014) Biosci 288:13178–13185 Biotechnol Biochem 78:582–587 57. Desguin B, Soumillion P, Hausinger RP, Hols P (2017) FEMS 42. Carlson ED, Papoutsakis ET (2017) Appl Environ Microbiol Microbiol Rev 41:S71–S83 83:e00829–17 58. Zheng K, Ngo PD, Owens VL, Yang XP, Mansoorabadi SO 43. Georgiadis MM, Komiya H, Chakrabarti P, Woo D, Kornuc JJ, (2016) Science 354:339–342 Rees DC (1992) Science 257:1653–1659 59. Zeer-Wanklyn CJ, Zamble DB (2017) Curr Opin Chem Biol 44. Rubio LM, Ludden PW (2005) J Bacteriol 187:405–414 37:80–88 45. Hu Y, Ribbe MW (2016) Annu Rev Biochem 85:455–483 1 3

Journal

JBIC Journal of Biological Inorganic ChemistrySpringer Journals

Published: Feb 14, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off