Theoretical Studies of Nickel-Dependent Enzymes

inorganics

Review Theoretical Studies of Nickel-Dependent Enzymes

Per E. M. Siegbahn 1,*, Shi-Lu Chen 2 and Rong-Zhen Liao 3 1 Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, SE-10691 Stockholm, Sweden 2 School of Chemistry and Chemical Engineering, Beijing Institute of Technology, Beijing 100081, China 3 School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, China * Correspondence: [email protected]

Received: 10 June 2019; Accepted: 23 July 2019; Published: 29 July 2019

Abstract: The advancements of quantum chemical methods and computer power allow detailed mechanistic investigations of metalloenzymes. In particular, both quantum chemical cluster and combined QM/MM approaches have been used, which have been proven to successfully complement experimental studies. This review starts with a brief introduction of nickel-dependent enzymes and then summarizes theoretical studies on the reaction mechanisms of these enzymes, including NiFe hydrogenase, methyl-coenzyme M reductase, nickel CO dehydrogenase, acetyl CoA synthase, acireductone dioxygenase, quercetin 2,4-dioxygenase, urease, lactate racemase, and superoxide dismutase.

Keywords: nickel enzymes; reaction mechanism; quantum chemical calculations

1. Introduction Nature harnesses transition metals to catalyze many different types of biological reactions that are crucial to life. Although about one-third of the enzymes are metalloenzymes, the use of nickel as a cofactor is not very common [1–9]. It was in 1975 that Zerner discovered the first nickel-dependent enzyme, namely urease [10]. Until now, ten different types of nickel enzymes (Table1) have been reported with diverse biological functions. Among the known nickel enzymes, four of them are involved in the processing of anaerobic gases, namely, H2, CO, CO2, and CH4, which have been proposed to be crucial to the origin of life [7]. Ni–Fe hydrogenase mediates the reversible two-electron reduction of protons to H2 [11]. Methyl-coenzyme M reductase catalyzes the reversible transformation of methyl-coenzyme M plus coenzyme B into CH4 and heterodisulfide CoM–S–S–CoB [12]. CO dehydrogenase catalyzes the reversible oxidation of CO to CO2 using water as the oxygen source [13]. Acetyl-CoA synthase interacts tightly with CO dehydrogenase and uses the CO generated by CO dehydrogenase, CoA-SH, and a methyl group from a corrinoid/FeS protein to synthesize acetyl-CoA [14]. Acireductone dioxygenase [15], and quercetin 2,4-dioxygenase [16] are both nickel-dependent dioxygenases, and produce CO as one of the products. Ni-dependent superoxide dismutase targets the toxic reactive oxygen species superoxide and catalyzes its conversion to O2 and H2O2 [17]. All these enzymes belong to the oxidoreductase class, while the remaining three enzymes are involved in the hydrolysis of urea (urease) [10], the isomerization of methylglyoxal to lactate (glyoxylase I) [18], and the racemization of lactate (lactate racemase, or LarA) [19]. Understanding the reaction mechanisms of nickel-dependent enzymes is of fundamental and practical importance. A particularly interesting question is why these enzymes select nickel as a catalytic center. One important advantage of using nickel has been suggested to be related to its flexible coordination geometry [8]. For redox reactions, the ligand environment has been demonstrated to play a crucial role in tuning the redox potentials of nickel, ranging from +0.89 V in superoxide

Inorganics 2019, 7, 95; doi:10.3390/inorganics7080095 www.mdpi.com/journal/inorganics Inorganics 2019, 7, 95 2 of 29 dismutase to 0.60 V in Methyl-coenzyme M reductase and CO dehydrogenase [6]. Sulfur-donor − ligands, such as sulﬁdes, thiolates of cysteine residues, are commonly used to fulﬁll this purpose.

In addition, the Lewis acidicInorganics character 2019, 7, ofx FOR nickel PEER has REVIEW also been suggested to be important for urease, 3 of 40 glyoxalase I, acireductone dioxygenase,Inorganics 2019, 7 and, x FOR quercetin PEER REVIEW 2,4-dioxygenase [8]. Importantly, these enzymes 3 of 40 Inorganics 2019, 7, x FOR PEER REVIEW 3 of 40 preferentially adopt O/N-donor ligands. AnotherTable important 1. Reactions question catalyzed is by how Ni-dependent the enzyme enzymes. catalyzes Inorganics 2019, 7, x FOR PEERTable REVIEW 1. Reactions catalyzed by Ni-dependent enzymes. 3 of 40 the reaction. To understandInorganics 2019 the, mechanistic7, x FOR PEER REVIEW details, especially the structures of transition states and 3 of 40 InorganicsInorganics 2019 2019, 7, ,x 7 InorganicsFOR, x FOREnzyme PEER PEER 2019 REVIEW REVIEW,Name 7, x FOR Table PEER REVIEW 1. Reaction Reactions catalyzed by Ni-dependent enzymes. 3 of3 40of 40 3 of 40 intermediatesInorganics for all 2019 elementary, 7, x FOR PEER steps, REVIEW quantum chemical calculations have proven to be useful to3 of 40 EnzymeInorganics Name 2019, 7, x FOR PEERTable Reaction REVIEW 1. Reactions catalyzed by Ni-dependent enzymes. 3 of 40 Enzyme Name Reaction complement experimental work.TableTable 1. Reactions1.Table ReactionsTable 1. catalyzedReactions catalyzed1. Reactions by catalyzed by Ni-dependent Ni-dependentcatalyzed by Ni-dependent by enzymes. Ni-dependent enzymes. enzymes. enzymes. HydrogenaseTable 1. Reactions catalyzed by Ni-dependent enzymes. Enzyme Name 1Table Reaction 1. Reactions catalyzed by Ni-dependent enzymes. Enzyme NameEnzyme Hydrogenase Name Reaction Reaction1 Enzyme NameHydrogenaseTableEnzyme 1. Reactions Name Reaction catalyzed Reaction by Ni-dependent enzymes. Enzyme NameMethyl-coenzyme Reaction M1 Enzyme Name Reaction Methyl-coenzymeHydrogenase M 2 1 Enzyme NameHydrogenaseMethyl-coenzymereductase M Reaction HydrogenaseHydrogenase Hydrogenase1 1 1 2 Hydrogenase reductase11 2 Hydrogenase COreductaseMethyl-coenzymeHydrogenase dehydrogenase M 3 Methyl-coenzymeMethyl-coenzymeCO M dehydrogenase M 3 1 Methyl-coenzymeMethyl-coenzymeAcetyl-CoA Mreductase M 2 Methyl-coenzymeCO dehydrogenase M2 2 3 Methyl-coenzymereductase M reductase reductaseAcetyl-CoAMethyl-coenzyme 2 M2 reductasereductaseAcetyl-CoA COreductasesynthase dehydrogenase 2 4 3 synthase 2 COCO dehydrogenase dehydrogenaseCO dehydrogenaseCO dehydrogenase3reductase 3 4 COCO dehydrogenasedehydrogenasesynthaseAcetyl-CoA 33 4 3 Acetyl-CoAAcetyl-CoA AcireductoneCO dehydrogenase Acetyl-CoAAcetyl-CoA Acetyl-CoAAcireductonesynthase 43 Acetyl-CoAsynthase synthase Acireductonesynthasedioxygenase 4 synthase synthaseAcetyl-CoAdioxygenase4 4 4 5 synthase 4 5 dioxygenaseAcireductonesynthase 4 AcireductoneAcireductone 5 AcireductoneAcireductoneAcireductone Quercetindioxygenase 2,4- Acireductone dioxygenaseQuercetin 2,4- 5 dioxygenasedioxygenasedioxygenase dioxygenaseAcireductone dioxygenaseQuercetin dioxygenase 52,4-5 5 5 dioxygenasedioxygenase5 6 dioxygenaseQuercetin 2,4- 6 5 Quercetin 2,4-QuercetinSuperoxide 2,4- 6 QuercetinQuercetinQuercetin 2,4-dioxygenase 2,4-2,4-QuercetinSuperoxidedioxygenase 2,4- 7 dioxygenaseSuperoxidedioxygenase dismutase 76 dioxygenase dioxygenaseQuercetindismutase 2,4- dioxygenase 6 6 6 7 6 dismutaseSuperoxidedioxygenase6 SuperoxideSuperoxide dismutaseSuperoxide 76 SuperoxideSuperoxide SuperoxideUreasedismutase Urease7 7 7 7 dismutasedismutase dismutase dismutaseSuperoxide 7 8 UreasedismutaseUrease 8 7 Ureasedismutase 8 Urease Urease Urease UreaseGlyoxylase I 8 Urease 8 8 Glyoxylase I GlyoxylaseUrease88 I 8 9 Glyoxylase I 9 8 Glyoxylase I 9 GlyoxylaseGlyoxylase I I LactateGlyoxylaseGlyoxylase racemase II Glyoxylase Lactate racemase I 9 LactateGlyoxylase9 racemase9 I 9 9 Lactate racemase9 10 109 Lactate racemase 10 Two popularLactateLactate but racemase racemaseLactate differentLactate Tworacemase approaches racemasepopular but have different been10 approaches developed have in the been modeling developed of in metalloenzymes. the modeling of metalloenzymes. Lactate racemaseTwo 10 popular but10 different approaches have been developed in the modeling of metalloenzymes. The first one is termed theTwo quantumThe popularLactate first1010 chemical onebutracemase differentis termed cluster 10 approaches the approach,quantum have chemical developed been developed cluster by approach, Siegbahn in the modeling developed and Himo of bymetalloenzymes. etSiegbahn and Himo The first one is termed the10 quantum chemical cluster approach, developed by Siegbahn and Himo al. [20–26]. More than 30The years etfirstTwo al. of one[20–26] experiencepopular is termed. Morebut hasdifferentthe than qua demonstrated 30ntum approachesyears chemical of experience thathave cluster abeen relatively hasapproach, developed demonstrated small developed in the model that modeling by a ofrelaSiegbahn thetively of metalloenzymes. smalland Himo model Two popularTwo but popular differentet al. [20–26] but approaches different. More approaches havethan 30been years developedhave of beenexperience indeveloped the modelinghas demonstratedin the of modeling metalloenzymes. that of a metalloenzymes. relatively small model active site, fromTwoTwo about popularpopular 30et atoms Twoal. butbutof [20–26]The differentthe differentpopular initially firstactive. Moreone approachesbutapproaches site, to is different morethan termedfrom 30 than about have haveyearstheapproaches 200qua been30been of ntumatoms experience developeddeveloped have chemical initially nowadays, been has inin developedtocluster the thedemonstrated more modelingmodeling is approach, capablethan in 200 the ofof that metalloenzymes. modelingofmetalloenzymes.atomsdeveloped a describing rela nowadays,tively of by metalloenzymes. smallSiegbahn is modelcapable and Himoof The first oneThe is firsttermedof oneTwothe the activeis popular termedqua ntumsite, butthe from chemical differentqua aboutntum cluster approaches30chemical atoms approach, clusterinitially have developedapproach,been to more developed than bydeveloped Siegbahn 200 in atomsthe by modeling and Siegbahn nowadays, Himo of and metalloenzymes. is Himocapable of all important mechanisticTheThe firstfirst oneoneof features The istheisdescribing termedtermedet activefirst al. of[20–26]one thesite,the the all is quaqua from catalysis.termedimportant. ntumMorentum about chemical thechemicalthan mechanistic Thequa30 30 atomsntum alternativeyears clustercluster chemicalinitially of features approach,experienceapproach, approachto cluster ofmore the developeddeveloped has approach,thancatalysis. isdemonstrated called200 byby atoms Thedeveloped SiegbahnSiegbahn the alternative nowadays, combinedthat andabyand rela Siegbahn HimoapproachHimo tivelyis capable small and is ofcalled Himomodel quantum mechanicset al. [20–26]/molecularet al.. More [20–26]describing mechanicsThethan. firstMore30 yearsall one thanimportant (QM is of termed30 experience/MM) years mechanistic the method,of qua experiencehasntum demonstrated features which chemical has was ofdemonstrated the clusterthat first catalysis. a proposedrela approach,tively that The a small rela bydevelopedalternative Warsheltively model small by approach Siegbahn model is called and Himo etet al.al. [20–26][20–26]describinget.. Morethe Moreal.of combined [20–26]the than thanall active important 3030. More yearsquantumyearssite, than from ofofmechanistic experience experience mechanics/molecular30 about years 30 featuresof hasatomshas experience demonstrateddemonstrated initially of the mechanics has catalysis. to demonstrated more thatthat a(QM/MM) a than The relarela tively200alternativetively that atoms meth smallsmall a rela od,nowadays,approach modelmodeltively which small iswas is called capable modelfirst of and Levitt inof 1976 the active [27].of site, the It has activefromtheet beencombined al.about site, [20–26] shown from 30 quantumatoms .about More by initially Liao30 thanmechanics/molecular atoms and30 to years initiallymore Thiel of than thatexperienceto 200more withmechanics atoms than properhas 200nowadays, demonstrated (QM/MM) atoms selection nowadays,is capablemeth that andod, a relaof aniswhich capabletively was small of first model ofof thethe activeactivethe site,ofsite, combinedproposed thedescribing fromfrom active aboutabout byquantum site, allWarshel 3030 importantfrom atomsatoms mechanics/molecular aboutand initiallyinitially Levittmechanistic 30 atoms tointo more1976more initially features [27]thanmechanicsthan. 200toIt200 of hasmore the atomsatoms been(QM/MM) catalysis. than nowadays, nowadays,shown 200 Theatomsmeth by alternativeLiao isisod, nowadays, capablecapable whichand Thiel of approachofwas is that capablefirst with is called of increase of thedescribing number ofdescribingall atomsimportantproposedof in all the the mechanisticimportant active by QM Warshel region,site, mechanistic features from and both aboutLevitt of approaches featuresthe 30in catalysis. atoms1976 of [27] theinitially gave The .catalysis. It similarhasalternative to beenmore The results shown than alternativeapproach 200 andby atomsLiao theis approach called sameand nowadays, Thiel is called that is withcapable of describingdescribing proposed allalldescribing importantimportantproperthe combinedby selection Warshelall mechanisticmechanistic important quantum and and an Levitt featuresfeaturesmechanistic increase mechanics/molecular in of1976of of thethe thefeatures [27] catalysis.catalysis. number. It ofhas the mechanicsTheofThe been atomscatalysis. alternativealternative shown in (QM/MM) the The by approachQMapproach Liao alternative region, andmeth is isThiel od, calledboth calledapproach which that approaches with wasis called first gave conclusion [28the,29 combined]. the quantumcombinedproperdescribing mechanics/molecular selectionquantum all mechanics/molecularandimportant an increase mechanisticmechanics of the mechanics(QM/MM)number features of of (QM/MM)atomsmeth the catalysis.od, in whichthe meth QM The wasod, region, alternative whichfirst both was approachapproaches first is gavecalled thethe combinedcombinedproperthe quantumquantumsimilar proposedcombined selection results mechanics/molecularmechanics/molecular by quantumand Warsheland an the increase mechanics/molecular same and Levitt conclusionof mechanics mechanicsthe in number 1976 [28,29] (QM/MM) (QM/MM)[27] ofmechanics atoms.. It has in meth methbeen (QM/MM)theod, od, QMshown whichwhich region, methby was wasLiao bothod, firstfirst andwhich approaches Thiel was that first gave with In the present review, wesimilar summarize results and the the mechanistic same conclusion studies [28,29] on nickel. enzymes by quantum proposedproposed byproposed byWarshel Warshelproposed theby and Warshelcombinedand byLevitt LevittWarshel inand quantumin1976 Levittand1976 [27] Levitt [27]inmechanics/molecular. 1976It. inhasIt has1976 [27]been been. [27] Itshown has shown. It been has by mechanics by Liao beenshown Liao andshown andby (QM/MM)Thiel Liao Thiel by that Liaoand that with Thielmethand with Thielod, that which withthat withwas first chemical calculations.proposed bysimilar Warshelproper results and selection and Levitt the andinsame 1976 an conclusion increase [27]. It hasof [28,29] the been number .shown of by atoms Liao in and the Thiel QM thatregion, with both approaches gave properproper selection selectionproperproper andIn selection and the proposedan selection anpresentincrease increase and by review, an andof Warshel increaseofthe an the numberweincrease number and summarizeof theLevitt of of ofatomsnumber the atoms in thenumber 1976 in ofmechanistic inthe atoms [27]the QMof QM.atoms It region,in has region,the studies inbeen QM theboth both shownregion,onQM approaches nickel approachesregion, by both enzymesLiao bothapproaches gave and gave approaches byThiel quantum gave that gavewith proper selectionInsimilar and the anpresent resultsincrease review, and of thethe we samenumber summarize conclusion of atoms the [28,29]mechanistic in the QM. region, studies both on nickel approaches enzymes gave by quantum similarsimilar results resultssimilarIn and similarthe andchemical resultsthe presentproper the same results sameand calculations.review,selection conclusion the andconclusion same the we and same summarize conclusion [28,29] an [28,29] conclusionincrease. . the[28,29] of mechanistic [28,29]the. number. studies of atoms on innickel the QM enzymes region, by both quantum approaches gave similar results andchemical the same calculations. conclusion [28,29]. chemicalInsimilar the calculations. present results review, and the we same summarize conclusion the mechanistic[28,29]. studies on nickel enzymes by quantum In the presentIn thereview, present we review, summarize we summarize the mechanistic the mechanistic studies on nickelstudies enzymes on nickel by enzymes quantum by quantum InIn thethe presentpresent2.In review,NiFereview, thechemical present Hydrogenase wewe summarizecalculations.summarize review, we thethe summarize mechanisticmechanistic the studiesstudiesmechanistic onon nickelnickel studies enzymesenzymes on nickel byby quantumquantumenzymes by quantum chemical calculations.chemical2.chemical NiFe Incalculations. theHydrogenase calculations. present review, we summarize the mechanistic studies on nickel enzymes by quantum chemicalchemical2. calculations. calculations.NiFe Hydrogenase chemical calculations. 2. NiFe Hydrogenase 2. NiFe Hydrogenase2. NiFe2. NiFe Hydrogenase Hydrogenase 2.2. NiFeNiFe HydrogenaseHydrogenase 2. NiFe Hydrogenase

Inorganics 2019, 7, 95 3 of 29

2. NiFe Hydrogenase In nature, hydrogenases are the enzymes designed to form dihydrogen from protons and electrons in a reversible process: + Inorganics 2019, 7, x FOR PEER REVIEW 2H + 2e− H2 3 of(1) 29

The mostmost commoncommon forms forms of of hydrogenases hydrogenases are are NiFe- NiFe and- and FeFe-hydrogenases. FeFe-hydrogenases. The The NiFe-enzymes NiFe-enzymes are primarilyare primarily used forused hydrogen for hydrogen oxidation, oxidation, and the FeFe-enzymes and the FeFe- forenzymes proton reduction.for proton NiFe-hydrogenase reduction. NiFe- ishydrogenase probably the is Ni-enzyme probably thatthe Ni has-enzyme been most that intensively has been studied most intensively by theoretical studied methods. by theoretical The active sitemethods. [30] is The shown active in Figure site [130.] Apart is shown from in the Figure four cysteines, 1. Apart therefrom are the three four additional cysteines, ligandsthere are on three iron, twoadditional cyanides, ligands and one on iron, carbon two monoxide, cyanides, which and one are carbonvery unusual monoxide, in biological which are systems, very unusual since they in arebiological potentially systems, poisonous. since they A complicated are potentially machinery poisonous. involving A complicated many enzymes machinery is, therefore, involving used many to constructenzymes is this, therefore complex., used to construct this complex.

Figure 1. The X X-ray-ray structure structure of of the the active active site site of of NiFe NiFe hydrogenase hydrogenase (P (Proteinrotein D Dataata B Bankank (PDB) (PDB) code code:: 1YRQ)1YRQ) [[3030].].

The theoretical work prior to 2007 has been reviewed [31,,32].]. A variety of different different models had been used, ranging from a minimal one, with only the direct ligands of the NiFe complex, to larger ones, with with some som chargede charged second second shell shell residues. residues. All studies All studies agreed agreed on the general on the mechanism general mechanism described indescribed Figure2 .in It Figure contains 2. threeIt contains intermediate three intermediate states observed states experimentally, observed experimentally, Ni a–C*, Nia–S Ni anda–C*, Ni Nia–SR.a–S Nianda–C* Nia is–SR. the Ni electrona–C* is paramagnetic the electron resonanceparamagnetic (EPR)-active resonance resting (EPR state,)-active which resting has astate, bridging which hydride has a betweenbridging thehydride metals. between The EPR-silent the metals. Nia The–SR EPR state-silent is reached Nia–SR by reductionstate is reached of Nia –C*,by reduction while Ni aof–S, Ni whicha–C*, iswhile also Ni EPR-silent,a–S, which is is obtained also EPR by-silent, oxidation. is obtained To complete by oxidation. the cycle, To Nicompletea–SR can the be cycle, reached Nia– bySR adding can be dihydrogenreached by adding to Nia –S.dihydrogen Several otherto Nia states–S. Several have other also beenstates observed have also experimentally been observed experimentally under varying conditions.under varying The conditions. TS for H–H The cleavage TS for H was–H generallycleavage was described generally as heterolytic, described as leading heterolytic, to a bridging leading hydrideto a bridging and a hydride protonation and of a aprotonation cysteine, usually of a cysteine, Cys543. Thisusually mechanism Cys543. couldThis mechanism be regarded could as being be inregarded consensus as being among in the consensus theoretical among groups the [ 31theoretical,32], and hadgroups strong [31, support32], and from had experiments.strong support For from the energyexperiments. profile, For itcould the energy be concluded profile, that it could entropy be and conclu dispersionded that played entropy important and dispersion roles [31 ],played even thoughimportant the roles dispersion [31], e evenffect could though not the be dispersionroutinely included effect could in the not calculations, be routinely as it includedcan be now. in the calculations, as it can be now.

Inorganics 2019, 7, 95 4 of 29 Inorganics 2019, 7, x FOR PEER REVIEW 4 of 29

FigureFigure 2. Schematic drawing of the catalytic cycles cycles for for NiFe NiFe-hydrogenases,-hydrogenases, suggested by experiments. experiments. AdaptedAdapted fromfrom [[3131],], CopyrightCopyright ©© 2007,2007, American ChemicalChemical Society.Society.

InIn thethe yearsyears followingfollowing thethe earlyearly periodperiod beforebefore 2007,2007, somesome newnew possibilitiespossibilities andand aspectsaspects werewere investigated.investigated. ThisThis developmentdevelopment hashas alsoalso been brieflybriefly reviewedreviewed [[2626,,3333].]. First, from experiments, itit waswas suggestedsuggested thatthat aa variantvariant ofof thethe mechanismmechanism inin FigureFigure2 2 might might be be operative, operative, leading leading to to a a mechanism mechanism in in FigureFigure3 3[ [3434].]. The main didifferencefference toto thethe mechanismmechanism ofof FigureFigure2 2 described described above above was was that that Ni Niaa–S–S doesdoes notnot participateparticipate inin thethe activeactive cycle.cycle. Instead, catalysiscatalysis startsstarts onlyonly afterafter anan initialinitial reactionreaction betweenbetween NiNiaa–S–S andand dihydrogen,dihydrogen, wherewhere NiNiaa–C*–C* isis created.created. Nia––C*C* is then onlyonly inin equilibriumequilibrium withwith NiNiaa–SR–SR duringduring turnover.turnover. Some supportsupport forfor thethe alternativealternative mechanismmechanism waswas givengiven inin oneone studystudy [[3535].]. AnAn interestinginteresting statestate inin thethe context context of of that that mechanism mechanism was was aNi(I)-state a Ni(I)-state (Ni (Nia–R*),a–R*) appearing, appearing at the at endthe end of the ofsingle the single step heterolyticstep heterolytic mechanism. mechanism. A TS forA TS oxidative for oxidative addition addition was found was leading found toleading two hydrides, to two hydrides, one bridging one andbridging one bound and one to bound nickel (Nito nickela–X*). (Nia–X*). Another development was that the interest switched over to the oxidized states. The reason for this interest is the possibility to generate fuel by combining water oxidation with H2 formation. A major problem in this context is that most hydrogenases are inactivated with O2 present. Two main oxidized states of the NiFe cofactor have been identified. In Ni–B, there is a bridging hydroxide between Ni(III) and Fe(II). It is inactive for H2 cleavage, but is relatively easy to reactivate. The Ni–A state is more difficult to reactivate. At the highest resolution (1.1 Å) it was initially concluded that Ni–A is a peroxide with one oxygen bridging between the metals and the other oxygen on nickel [30,36]. The early modeling studies were reviewed in 2007 [31]. While there was no problem obtaining the Ni–B structure with a bridging hydroxide [37,38], the modeling of Ni–A was more problematic. A structure similar to the one found experimentally could rather easily be obtained, but the energy was too high. A detailed analysis of the density of the experimental structure showed that it was likely to be a mixture of states, and the structure was probably also over reduced [39]. Calculations on the mechanism for the reactivation of the initially suggested structure for Ni–A were made [37]. It was suggested that the full reactivation of the enzyme requires two additional electrons and protons and this is the reason that it was slow. A reactivation mechanism for an oxidized cysteine ligand was also investigated. In 2015, a

Inorganics 2019, 7, 95 5 of 29 re-analysis of the experimental structure for Ni–A was performed, and it was concluded that Ni–A instead has a bridging hydroxide and an oxidized sulfenated form of Cys75 [40]. DFT calculations were then done for the new structure using larger models than before [41,42], and a revised mechanism forInorganics the reactivation 2019, 7, x FOR was PEER made. REVIEW 5 of 29

FigureFigure 3.3. AA variantvariant of the mechanism for NiFe NiFe-hydrogenases.-hydrogenases. Adapted Adapted from from [31 [31],], Copyright Copyright ©© 2007,2007, AmericanAmerican ChemicalChemical Society.Society.

SeveralAnother oxygen development insensitive was that NiFe the hydrogenases interest switched have over been to found. the oxidized In these states. enzymes, The reason the Ni–A for statethis interest is activated is the by possibility two electrons to generate supplied fuel from by combining the proximal water FeS oxidation cluster [43 wi].th A H few2 formation. DFT studies A havemajor been problem performed in this tocontext determine is that the most details hydrogenases of the mechanism are inactivated of how with the O electrons2 present. are Two released main fromoxidized the FeS states cluster of the [44 NiFe–46]. Thecofactor oxygen have tolerant been identified. hydrogenases In Ni have–B, there been found is a bridging to have hydroxide speciﬁcally designedbetween proximalNi(III) and Fe–S Fe(II). clusters, It is inactive where for a sulﬁde H2 cleavage, in a corner but is is relatively missing and easy replaced to reactivate. by two The cysteines. Ni–A Thisstate allows is more the difficult Fe–S cluster to reactivate. to deliver At two the electronshighest resolution to the NiFe-cluster. (1.1 Å) it was Together initially with concluded two electrons that fromNi–A the is activea peroxide site, thewith accumulation one oxygen ofbridging the Ni–A between state is th diminished.e metals and DFT the calculationsother oxygen were on usednickel to determine[30,36]. The electronic early modeling and structural studies were changes reviewed of the in Fe–S 2007 cluster. [31]. While there was no problem obtaining the Ni–B structure with a bridging hydroxide [37,38], the modeling of Ni–A was more problematic. A structure similar to the one found experimentally could rather easily be obtained, but the energy was too high. A detailed analysis of the density of the experimental structure showed that it was likely to be a mixture of states, and the structure was probably also over reduced [39]. Calculations on the mechanism for the reactivation of the initially suggested structure for Ni–A were made [37]. It was suggested that the full reactivation of the enzyme requires two additional electrons and protons and this is the reason that it was slow. A reactivation mechanism for an oxidized cysteine

Inorganics 2019, 7, 95 6 of 29

After these initial modeling studies, the interest in the mechanism of NiFe-hydrogenase has decreased. The studies mentioned above gave mechanisms that are quite similar with minor diﬀerences, and in good agreement with available experiments [31]. However, in recent years there was a set of studies by Ryde et al. in 2014–2018, where many of the results obtained previously were seriously criticized [47–51]. In particular, the accuracy of the Becke, 3-parameter, Lee–Yang–Parr exchange-correlation functional (B3LYP), used in practically all previous studies, was claimed to be insuﬃcient. Comparisons between B3LYP and more advanced methods of ab initio type seemed to give results, showing that there were large errors using B3LYP. For the bare cofactor without hydrogen, the singlet–triplet splitting (S–T) error was tolerable with a usual error for B3LYP of a few kcal/mol compared to the advanced methods [47]. However, when H2 was added, the S–T splitting showed large errors for B3LYP [48], with a value of 13.2 kcal/mol compared to high level studies that gave 2–3 kcal/mol [49], but there are severe problems with that comparison. The most important one is that the structures were not optimized, but taken directly from a QM/MM study. This meant that the structures, as used in the comparison, are about 40 kcal/mol higher in energy than the optimized ones. The result most emphasized in the new DFT studies was that Nia–R* should not be an intermediate in the reaction sequence [51], as it was in some previous studies [35]. However, there is an interesting experimental result in this context using electro-chemical methods, where the conclusion was that Nia–R* was indeed observed as an intermediate [52]. Another more important aspect of the Ryde et al. studies is that they showed that it is actually possible to do calculations with the most accurate methods in quantum chemistry, such as the coupled cluster single-double and perturbative triple method (CCSD(T)) and the density matrix renormalization group method (DMRG). They have not yet been used for the reaction mechanism but it will be interesting to follow that development in the future.

3. Methyl-Coenzyme M Reductase

Methane, acting as the second most important anthropogenic greenhouse gas after CO2 [53,54], has a significant influence on the global carbon cycling and the climate [53]. With a methane emission annually of around 500–600 Tg, about 69% is biologically released by microbial metabolisms, i.e., methanogenic archaea [55]. Meanwhile, a large amount of methane is oxidatively consumed by aerobic methanotrophic bacteria and anaerobic methane oxidizing archaea (ANME) [56,57]. The anaerobic methane oxidation is usually coupled with the reduction of sulfate (sulfate-reducing δ-proteobacteria) [58,59] iron [60,61], manganese [61], nitrate [62], soluble metal complexes [63], etc. [64], which may critically affect the Proterozoic and Neoproterozoic global climate systems [65] and the breeding of animals [66]. The final step of microbial methane formation and the first step of anaerobic methane oxidation are both performed by the Ni-dependent methyl-coenzyme M reductase (MCR) [56,67], which is capable of reversibly catalyzing the conversion of methyl-coenzyme M (CH3–SCoM) and coenzyme B (CoB–SH) to a heterodisulfide (CoM–S–S–CoB) and methane using a Ni(I) porphyrinoid cofactor (F430) (Figure4)[ 6,9]. It is further found that MCR-like enzymes may constitute a family of alkyl-coenzyme M reductases, which are essential for the microbial anaerobic oxidation of ethane [68], propane [56,68], and butane [69] following the same reversible reaction pattern as MCR, i.e., alkane + CoM–S–S–CoB alkyl–SCoM + CoB–SH. Thus, the MCR mechanism may serve as a template for the family of alkyl-coenzyme M reductases [68], rendering it significant for the understanding of the alkane biosynthesis and the development of biomimetic catalysts for the alkane conversion. Inorganics 2019, 7, x FOR PEER REVIEW 7 of 29

reversible reaction pattern as MCR, i.e., alkane + CoM–S–S–CoB ⇌ alkyl–SCoM + CoB–SH. Thus, the MCR mechanism may serve as a template for the family of alkyl-coenzyme M reductases [68], Inorganicsrendering2019 it, 7significant, 95 for the understanding of the alkane biosynthesis and the development7 of 29 of biomimetic catalysts for the alkane conversion.

FigureFigure 4.4. TheThe reactionreaction catalyzedcatalyzed byby methyl-coenzymemethyl-coenzyme MM reductasereductase (MCR)(MCR) usingusing aa FF430430 cofactor.

MainlyMainly four four mechanisms mechanisms for forMCR MCR have have been proposed been proposed and examined and examined by experiments by experiments or theoretical or calculationstheoretical calculations (Figure5). Accompanying (Figure 5). Accompanying the crystallization the crystallizationof MCR [ 12], mechanism of MCR [1 I2 (Figure], mechanism5a) was I hypothesized(Figure 5a) was to includehypothesized an organometallic to include Ni(III)-methylan organometallic intermediate Ni(III)-methyl (referred intermediate to as MCRMe (referred) resulting to fromas MCR theMe Ni-activated) resulting C–Sfrom bond the Ni dissociation-activated of C– CHS bond3–SCoM dissociation and the proton of CH transfer3–SCoM fromand CoB–SH the proton to thetransfer sulfur from of CH CoB3–SCoM.–SH to In the mechanism sulfur of I, CH the3– firstSCoM. step In of mechanism C–S bond dissociation I, the first step is most of Clikely–S bond not rate-limiting,dissociation is since most the likely following not rate electron-limiting transfer, since the was following suggested electron to lead transfer to an unstable was suggested CoM–S Hto · radical.lead to an Methane unstable is CoM formed–S·H in radical. the next Methane step of is S–S formed bond in formation. the next step This of mechanismS–S bond formation. received This the strongestmechanism support received from the the strongest direct synthesis support from of MCR theMe directfrom synthesis the active of MCR MCR (MCRMe fromred1 the) with active methyl MCR bromide(MCRred1 [)70 with] or methyl iodide [bromide71], although [70] or its iodide formation [71], from although the native its formation substrate from (CH the3–SCoM) native has substrate never been(CH3 found.–SCoM) However, has never a rough been found. estimation However, by means a rough of the estimation B3LYP bond by dissociation means of the energies B3LYP (BDEs) bond predicteddissociation a veryenergies large (BDEs) endothermicity predicted fora very the large MCR enMedothermicityformation in for the the native MCR MCRMe formation reaction in [ 72the]. Anative more MCR elaborate reaction B3LYP [72] investigation. A more elaborate based B3LYP on a chemical investigation model based built on from a chemical the crystal model structure built (Figurefrom the6) cry showedstal structure that the (Figure MCR Me 6) formationshowed that is the dependent MCRMe formation on the acidity is dependent of the methyl-X on the acidity leaving of group,the methyl i.e.,- theX leaving C–X bond group, dissociation i.e., the C– energyX bond [ 73dissociation]. The MCR energyMe formation [73]. The is MCR facileMe with formation a barrier is facile of a fewwith kcal a barrier/mol when of a few X is akcal/mol halide (I when−, Br− X, or is Cla −halide), while (I−, it Br has−, or a veryCl−), highwhile barrier it has whena very starting high barrier from the native substrate with a stronger C–S bond, which was predicted to have a large endothermicity of 23.5 kcal/mol [73]. This endothermicity was then updated to 21.8 kcal/mol with dispersion and entropy Inorganics 2019, 7, x FOR PEER REVIEW 8 of 29 when starting from the native substrate with a stronger C–S bond, which was predicted to have a Inorganics 2019, 7, 95 8 of 29 large endothermicity of 23.5 kcal/mol [73]. This endothermicity was then updated to 21.8 kcal/mol with dispersion and entropy effects added [74], which is sufficient to rule out mechanism I, econsideringffects added the [74 possible], which additional is sufficient barriers to rule in out the mechanism subsequent I, consideringsteps (proton the transfer possible and additional electron barrierstransfer). in the subsequent steps (proton transfer and electron transfer).

CoB a S H a H3C S CoM a b Ni(I) H+

a MCR d

CoB b c CoB S (15.1) H S CoM CoB S H S CoB CH3 H S H CoM H S Ni(III) [21.8] •CH3 CoM CoM CH3 S H3C S d CoB Ni(II) Ni(III) Ni(I) [22.9] S H S CoM [44.5] CH3

Ni(II) CH4 CoB S CoB H CH4 S• H S CoM S CoM CoB CH3 S S CoM Ni(II) [-0.8] Ni(III) •

Ni(II) (9.2)

MCR + CoB-S-S-CoM [-0.6]

Figure 5. 5. FourFour proposed proposed mechanisms mechanisms for forthe MCR the MCR reaction. reaction. (a) Mechanism (a) Mechanism I, including I, including a methyl a-

methyl-Ni(III)FNi(III)F430 intermediate.430 intermediate. (b) Mechanism (b) Mechanism II, involving II, involving a methyl a methyl radical radical and and a a CoM–S–Ni(II)F CoM–S–Ni(II)F430430 intermediate. ( (c)) Mechanism III ,including,including aa ternaryternary intermediateintermediate withwith side-onside-on methylmethyl andand CoM–SCoM–S−− d coordinating to to Ni. Ni. (d () Mechanism) Mechanism IV IV,, involving involving a proton a proton binding binding at the at thesulfur sulfur of CH of3– CHSCoM.3–SCoM. The Theenergies energies for minima for minima and andtransition transition states states are given are given in brackets in brackets and parentheses, and parentheses, respectively. respectively. The Theenergies energies for mechanisms for mechanisms I and I and II are II are adapted adapted from from [74 [74], ],copyright copyright ©© WILEYWILEY-VCH-VCH Verlag Verlag GmbH GmbH & Co. KGaA, Weinh Weinheim,eim, while thethe onesones forfor mechanismsmechanisms IIIIII andand IVIV areare adaptedadaptedfrom from [ [7575],], with with permissionpermission from the the PCCP PCCP Owner Owner Societies. Societies. A more detailed DFT cluster modeling of mechanism II (Figure5b) was then performed [ 72,76]. A more detailed DFT cluster modeling of mechanism II (Figure 5b) was then performed [72,76]. Mechanism II proceeds through a C–S bond dissociation forming Ni(II)–SCoM and a methyl radical, Mechanism II proceeds through a C–S bond dissociation forming Ni(II)–SCoM and a methyl radical, the immediate quenching of the methyl radical by CoB–SH leading to methane and a CoB–S radical. the immediate quenching of the methyl radical by CoB–SH leading to methane and a CoB–S·· radical. This is followed by a rebound mechanism between Ni(II)–SCoM and CoB–S to form a CoM–S–S–CoB This is followed by a rebound mechanism between Ni(II)–SCoM and CoB·–S· to form a CoM–S–S– and a regeneration of Ni(I) F430. This mechanism has the following characteristics that are compatible CoB and a regeneration of Ni(I) F430. This mechanism has the following characteristics that are with most of the mechanistic experiments: (i) The ﬁrst step of C–S bond dissociation is rate-limiting compatible with most of the mechanistic experiments: (i) The first step of C–S bond dissociation is with a barrier of 19.5 kcal/mol, which is supported by a substantial carbon kinetic isotope eﬀects rate12 -limiting with13 a barrier of 19.5 kcal/mol, which is supported by a substantial carbon kinetic ( CH3–SCoM/ CH3–SCoM) of 1.04 0.01 [77]. (ii) The methyl radical is captured rapidly by CoB–SH, isotope effects (12CH3–SCoM/13CH3–±SCoM) of 1.04 ± 0.01 [77]. (ii) The methyl radical is captured so that methane formation happens almost simultaneously with the C–S bond dissociation and the whole MCR reaction can be regarded as two chemical steps via two transition states. (iii) An inversion Inorganics 2019, 7, x FOR PEER REVIEW 9 of 29

Inorganicsrapidly 2019by CoB, 7, 95–SH, so that methane formation happens almost simultaneously with the C–S 9bond of 29 dissociation and the whole MCR reaction can be regarded as two chemical steps via two transition states. (iii) An inversion of configuration at the methyl carbon takes place via a trigonal planar of configuration at the methyl carbon takes place via a trigonal planar configuration during the methane configuration during the methane formation; step (iii) combined with step (ii) is consistent with the formation; step (iii) combined with step (ii) is consistent with the finding using an isotopically chiral finding using an isotopically chiral form of ethyl-coenzyme M [78] and a large α-secondary kinetic form of ethyl-coenzyme M [78] and a large α-secondary kinetic H/D isotope effect (kH/kD = 1.19 0.01), H/D isotope effect (kH/kD = 1.19 ± 0.01), which indicates a geometric change of the methyl group± from which indicates a geometric change of the methyl group from tetrahedral to trigonal planar at the tetrahedral to trigonal planar at the highest transition state [77]. However, the energetics obtained in highest transition state [77]. However, the energetics obtained in 2002 and 2003 [72,76] are not accurate 2002 and 2003 [72,76] are not accurate enough to give a quantitatively correct description of the MCR enough to give a quantitatively correct description of the MCR mechanism, mainly because of the use of mechanism, mainly because of the use of methanol as the model of Tyr333 and Tyr367, the lack of methanol as the model of Tyr333 and Tyr367, the lack of dispersion effects, and the omission of entropy dispersion effects, and the omission of entropy effects in the second step of the S–S bond formation effects in the second step of the S–S bond formation [74]. This leads to an excessively high barrier of [74]. This leads to an excessively high barrier of ~26 kcal/mol for the reverse reaction (i.e., the ~26 kcal/mol for the reverse reaction (i.e., the anaerobic oxidation of methane) and an excessively large anaerobic oxidation of methane) and an excessively large barrier difference of 10.0 kcal/mol between barrier difference of 10.0 kcal/mol between the two transition states of C–S bond dissociation and S–S the two transition states of C–S bond dissociation and S–S bond formation [72,76]. This process faced bond formation [72,76]. This process faced additional challenges posed by two new isotope exchange additional challenges posed by two new isotope exchange experiments. These new experiments experiments. These new experiments proved the feasibility of the reverse oxidation of methane [79] proved the feasibility of the reverse oxidation of methane [79] and showed a small barrier difference and showed a small barrier difference between the two chemical steps [80]. With dispersion and more between the two chemical steps [80]. With dispersion and more accurate entropy effects included, a accurate entropy effects included, a new DFT investigation with methylphenols as tyrosine models gave new DFT investigation with methylphenols as tyrosine models gave more reasonable barriers of 15.1 more reasonable barriers of 15.1 and 21.1 kcal/mol for the forward and reverse reactions, respectively, and 21.1 kcal/mol for the forward and reverse reactions, respectively, and a smaller barrier difference and a smaller barrier difference of 3.6 kcal/mol for the two transition states [74], which demonstrated of 3.6 kcal/mol for the two transition states [74], which demonstrated that mechanism II is still that mechanism II is still acceptable and in agreement with mechanistic experiments at that time. acceptable and in agreement with mechanistic experiments at that time. Interestingly, an ingenious Interestingly, an ingenious spectroscopic experiment using a shorter CoB–SH with one methylene less spectroscopic experiment using a shorter CoB–SH with one methylene less in order to elongate the in order to elongate the distance of CoB–SH to the reacting core, lowering the MCR reaction rate, leads distance of CoB–SH to the reacting core, lowering the MCR reaction rate, leads to a direct observation to a direct observation of the Ni(II)–SCoM intermediate [81]. This provided the clearest experimental of the Ni(II)–SCoM intermediate [81]. This provided the clearest experimental evidence for evidencemechanism for II mechanism and made IImechanism and made II mechanism a consensus II aMCR consensus mechanism MCR [ mechanism82]. [82].

Figure 6. Model used in the calculations of the methyl-coenzyme M reductase (MCR). Adapted from Figure 6. Model used in the calculations of the methyl-coenzyme M reductase (MCR). Adapted [74], copyright © WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. from [74], copyright © WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

AnAn oxidative oxidative addition addition mechanism mechanism (i.e., Mechanism III in Figure5 c) 5c) proposed proposed by by DFT DFT calculations [83] and experiments [80,84] involves a side-on C–S (for CH3–SCoM) [84] or C–H (for the calculations [83] and experiments [80,84] involves a side-on C–S (for CH3–SCoM) [84] or C–H (forreverse the reversemethane methane oxidation) oxidation) [80] coordination [80] coordination to the Ni toion, the forming Ni ion, a forming ternary aintermediate ternary intermediate with both methyl and CoM–S−/hydride coordinating to the metal. However, from a DFT study based on the with both methyl and CoM–S−/hydride coordinating to the metal. However, from a DFT study based onsame the cluster same cluster model model that was that used was usedin the in previous the previous calculations calculations of mechanism of mechanism II, II,the the formation formation of of a aternary ternary intermediate intermediate was was predicted predicted to to have have a a very very large large endothermicity endothermicity of of 44.5 44.5 kcal/mol kcal/mol starting from CH3–SCoM, and also a very high barrier starting from methane [75]. This can safely rule out CH3–SCoM, and also a very high barrier starting from methane [75]. This can safely rule out mechanism

Inorganics 2019, 7, 95 10 of 29

III. Another proposal (mechanism IV in Figure5d) suggested a protonation at the sulfur position of CH3–SCoM, which should facilitate the C–S bond activation and the subsequent methyl–Ni(III)F430 formation. Nevertheless, the DFT calculations showed that the substrate sulfur has a much smaller proton affinity than the F430 cofactor, in particular the carbonyl oxygen in F430, which means that a proton entering the MCR active site should prefer binding at F430 instead of CH3–SCoM, and that an additional energy of ~23 kcal/mol is required to move the proton from F430 to the substrate sulfur before the C–S bond activation [75]. This makes mechanism IV infeasible. In summary, mechanism II (Figure5b) is finally the most acceptable one for the MCR enzyme after two decades of computational and experimental studies. It fits reasonably well with several structural characteristics of the MCR active site. For example, a deeply-buried active site and the strict arrangement of F430, CH3–SCoM, and CoB–SH from bottom to top [12,85], ensuring the quick quenching of the methyl radical. In fact, the absence of CoB–SH cannot initiate the C–S bond cleavage and a CoB–SH analogue with one methylene less makes the MCR reaction much slower [86], which may be explained by the fast rebound between Ni(II)–SCoM and the unstable methyl radical. Even if the methyl radical escapes from CoB–SH, it cannot easily cause irreparable damage to the MCR active site, since the latter is hydrophobic and four active-site residues (His257, Arg271, Gln400, and Cys452) have been uncommonly methylated [12,85]. With this, MCR presents a perfect example of a mechanistic proposal by quantum chemical modeling finally leading to the convergence between theory and experiment. The understanding of the MCR mechanism could lead to new ways of producing methane/alkanes as fuels [87], of utilizing methane/alkanes as chemical feedstocks [82,87], and of breaking the C–S bond [88].

4. Nickel CO Dehydrogenase and Acetyl CoA Synthase Carbon dioxide fixation is an extremely important process in nature, with implications on present day discussions on the greenhouse effect. The dominant enzymes under anaerobic conditions are Ni–CODH and Acetyl CoA synthase, which both have an active site containing Ni. Structures of the enzymes were first determined in 2002–2003 [14,89,90]. In the latest structure [13], obtained at a high resolution of 1.1 Å, an intermediate with bound CO2 was found. The Ni–CODH and Acetyl CoA synthase enzymes are contained in one complex, and are connected by a 140 Å long channel in which CO is transported [91,92]. In Ni–CODH, CO can be reversibly transformed to CO2. The reduction potential used for oxidation of CO is 0.3 V, while for reduction of CO it has to be decreased to 0.6 V. Experimentally, the rate for − 2 − reduction is 10 s 1 at pH 7, and at pH 8 the oxidation rate is as high as 3.9 104. − × The active site of Ni–CODH has an unusual NiFe3S4 complex, termed the C-cluster, connected to another iron termed Feu. The experimental X-ray structure with a bound CO2 [13], termed Cred1, has carbon bound to nickel and one of the oxygens bound to Feu, as shown in Figure7. There are two important hydrogen bonds to CO2, with one being from the positively charged Lys563, and the other one from a His93–Asp219 couple, which is most probably protonated. The spin state is a doublet [93] indicating an oxidation state of Ni(II)Fe3(III)Fe(II) for the NiFe4 cluster. Theoretical studies have supported this assignment [94–96]. After adding two electrons, one C–O bond is cleaved and the state termed Cred2 is reached, which could be either Ni(I)Fe4(II) or Ni(0)Fe3(III)Fe(II). In the X-ray structure of Cred2, the CO, resulting from the C–O cleavage of CO2, has disappeared from the cluster leaving an empty site on nickel [13]. Inorganics 2019, 7, 95 11 of 29 Inorganics 2019, 7, x FOR PEER REVIEW 11 of 29

Figure 7. The X-ray structure of the active site of Ni–CODH with bound CO2 [13]. Figure 7. The X-ray structure of the active site of Ni–CODH with bound CO2 [13]. The first computational study of the mechanism of Ni–CODH was made by Amara et al. [96]. To explainThe first the computational X-ray structures, study they of suggested the mechanism an unusual of Ni– mechanism,CODH was made where by in Amara Cred2 a et hydride al. [96]. occupiesTo explain the emptythe X-ray site structures, on nickel. The they hydride suggested would an thenunusual have mechanism a role as an, electron where in reservoir Cred2 a inhydride Cred2. Thisoccupies would the mean empty that site nickel on nickel. always The stays hydride in the would Ni(II) then state. have A a bound role as hydridean electron would reservoir not be in seenCred2. inThis the would X-ray structuremean that and nickel might always explain stays the in observation the Ni(II) state. of small A bound amounts hydride of H 2wouldin the not experiments. be seen in Inthe the X- COray2 structurereduction and direction might explain from C red2the ,observation CO2 was suggested of small amounts to insert of into H2 the in the Ni–hydride experiments. bond. In Thethe proteinCO2 reduction matrix shoulddirection prevent from C formatered2, CO2 formation.was suggested Mössbauer to insert parameter into the Ni calculations,–hydride bond. and The an orbitalprotein and matr a chargeix should distribution prevent formate analysis, formation. were also Mössbauer performed parameter to support calculations, their mechanism. and an orbital and Quitea charge recently, distribution the energetics analysis, for were the fullalso catalytic performed cycle to ofsupport Ni–CODH their weremechanism computed. for the first time (FigureQuite recently,8)[ 97]. Energy the energetics diagrams for for the both full catalytic the reduction cycle ofof CONi–CODH, using awere redox computed potential for of the0.6 first V, 2 − andtime oxidation (Figure 8 of) [ CO,97]. usingEnergy a redoxdiagrams potential for both of the0.3 reduction V, were calculated. of CO2, using In line a redox with experiments, potential of the−0.6 − oxidationV, and oxidation is faster withof CO, a barrier using dia redoxfference potential of 4.8 kcal of/ mol,−0.3 V compared, were calculated. to experiments In line of with 6.2kcal experiments,/mol [92]. Thethe rate-limiting oxidation is step faster in both with directions a barrier is difference the cleavage of (formation) 4.8 kcal/mol, of the compared C–O bond to of experiments CO2. The strong of 6.2 hydrogenkcal/mol bond[92]. The from rate Lys563-limiting and thestep proton in both transfer directions from is the the His93–Asp219 cleavage (formation) couple areof the very C– important.O bond of ToCO obtain2. The accuratestrong hydrogen energetics, bond it wasfrom found Lys563 that and one the of proton the sulfides transfer in from the NiFeSthe His93 cluster–Asp219 should couple be protonated.are very important A water. moleculeTo obtain inaccurate the active energetics, site was it also was found found to that be reasonably one of the important.sulfides in the NiFeS clusterAn should important be diprotonated.fference between A water the molecule computed in andthe experimentalactive site was structures also found found to be was reasonably that CO wasimportant. bound to nickel in the reduced structure, in contrast to the experimental structure. To explain this result,An important it was suggested difference thatbetween the X-raythe computed structure and was experimental over-reduced. structures The calculations found was forthat the CO over-reducedwas bound to structure nickel in showed the reduced that COstructure, is initially in contrast exchanged to the with experimental a hydride. Thestructure. hydride To and explain the waterthis result, molecule it was then suggested form a hydrogen that the X molecule,-ray structure suggested was over to explain-reduced. the The presence calculations of small for amounts the over- ofreduced H2. structure showed that CO is initially exchanged with a hydride. The hydride and the water molecule then form a hydrogen molecule, suggested to explain the presence of small amounts of H2.

Inorganics 2019, 7, 95 12 of 29 Inorganics 2019, 7, x FOR PEER REVIEW 12 of 29

FigureFigure 8. 8. Mechanism of nickel CO dehydrogenase suggested on the the basis basis of of density density functional functional calculations.calculations. AdaptedAdapted fromfrom [[9797],], CopyrightCopyright ©© 2019, American Chemical Society.Society.

AsAs mentionedmentioned above,above, acetylacetyl CoACoA synthasesynthase (ACS)(ACS) isis locatedlocated inin thethe samesame complexcomplex asas Ni–CODH,Ni–CODH, atat thethe otherother endend ofof thethe longlong channelchannel throughthrough whichwhich COCO isis transported.transported. The active catalyst in ACS, termedtermed thethe A-cluster,A-cluster, isis aa nickelnickel dimerdimer connectedconnected byby twotwo sulfidesulfide bridgesbridges toto anan FeFe44SS44 cluster (Figure 9 9)).. TheThe nickelnickel closestclosest toto thethe FeSFeS cluster cluster is is termed termed Ni Nippand and thethe otherother one one Ni Niuu.. NiNipp is, apart from thethe sulfidesulfide bridgesbridges toto thethe FeSFeS cluster,cluster, connectedconnected byby two two bridging bridging cysteines cysteines to to Ni Niuu.. NiNiuu isis furthermorefurthermore ligatedligated toto twotwo backbonebackbone nitrogensnitrogens ofof twotwo connectedconnected cysteines.cysteines. NiNipp obtains a square planarplanar coordinationcoordination withwith aa Ni(II)Ni(II) oxidationoxidation statestate before before the the substrates substrates enter. enter. Ni Nip isp is five-coordinated five-coordinated and and has has been been suggested suggested to beto thebe the active active metal metal in thein the dimer. dimer. There There are are three three substrates, substrates, CO, CO, CH CH3,3 and, and coenzyme coenzyme A A (CoA). (CoA). CH CH33 isis donateddonated toto thethe clustercluster fromfrom aa cobalamine.cobalamine. TheThe orderorder ofof thethe enteringentering substratessubstrates isis debated.debated. ThereThere areare reasonsreasons to to believe believe that that the order the order might mightmatter, matter,since the since protein the has protein at least twohas di atff erent least conformations, two different ofconformations, which only one—the of which open only conformation—allowsone—the open conformation CO to— beallows transported CO to allbe transported the way to ACS. all the In way the closedto ACS. conformation In the closed the conformation transport of the CO transport is blocked. of CO is blocked. ThereThere has been one one theoretical theoretical study study of of the the mechanism mechanism of ofACS ACS [98 [].98 A]. cluster A cluster QM QMapproach approach was wasused. used. A problem A problem in the in themodeling modeling is that is that no no X- X-rayray structures structures with with bound bound substrates substrates are are available. PossiblePossible intermediatesintermediates inin thethe catalyticcatalytic cyclecycle werewere studied,studied, includingincluding theirtheir oxidationoxidation states.states. Ni(0)Ni(0) hadhad earlierearlier beenbeen suggested suggested as as a possiblea possible state state of Niofp Ni, butp, but the studythe study shows shows that thisthat is this unlikely. is unlikely. The zero-valent The zero- 1+ statevalent is state stabilized is stabilized only if theonly FeS if the cluster FeS iscluster reduced is reduced to a state to with a state S = with1/2 ([FeS =4 1/2S4] ([Fe). As4S4] CO1+). As binds, CO Nibinds,p(0) isNi stillp(0) favored.is still favored. If the FeS If the cluster FeS iscluster oxidized, is oxidized, Nip(I) is Ni insteadp(I) is favored.instead favored. It is suggested It is suggested that CO andthat CHCO3 andboth CH bind3 both to Ni bindp in to a mononuclearNip in a mononuclear mechanism mechanism on Nip. There on Ni arep. There major are remaining major remaining issues in theissues mechanism, in the mechanism, mostly because mostly of the because limited of structural the limited information. structural Ininformation. particular, the In particular,transfer of the the transfer of the methyl from cobalamine to the nickel dimer would be very interesting to study in the future, but also many questions remain regarding the final stages of CoA synthase.

Inorganics 2019, 7, 95 13 of 29 methyl from cobalamine to the nickel dimer would be very interesting to study in the future, but also many questions remain regarding the ﬁnal stages of CoA synthase. Inorganics 2019, 7, x FOR PEER REVIEW 13 of 29

FigureFigure 9. 9.Mechanism Mechanism of of Acetyl Acetyl CoA CoA suggested suggested onon thethe basis of of quantum quantum mechanics mechanics (QM (QM)) cluster cluster calculations.calculations. Adapted from Adapted [98], Copyrightfrom [98], ©Copyright2005, American © 2005, ChemicalAmerican Society. Chemical Society.

AA theoreticaltheoretical studystudy ofof thethe Ni-dimerNi-dimer inin ACSACS withwith emphasisemphasis onon pKpKaavalues values andand redoxredox potentialspotentials hashas alsoalso beenbeen performedperformed [ 99]. Based Based on on these these computed computed values, values, the the initial initial part part of ofa mechanism a mechanism for forACS ACS was was suggested. suggested. The The low low reduction reduction potential potential computed computed indicates indicates that that the the reduced form form is is protonated.protonated. BeforeBefore oror duringduring methylationmethylationof of thethe A-cluster,A-cluster, the the proton proton could could be be detached detached by by a a base. base. AA protonproton coupledcoupled electronelectron transfertransfer (PCET)(PCET) mechanismmechanism shouldshould bebe operationaloperational inin thethe methylationmethylation step.step. Furthermore,Furthermore, aa shortshort earlyearly notenote onon thethe ACSACS mechanismmechanism byby anotheranother groupgroup hashas beenbeen publishedpublished [[100100].]. TheirTheir calculations calculations suggest suggest that that Ni Nippis is betterbetter describeddescribed byby thethe oxidation oxidation state state Ni(I) Ni(I) than than Ni(0). Ni(0).

5.5. AcireductoneAcireductone DioxygenaseDioxygenase AcireductoneAcireductone dioxygenasesdioxygenases (ARDs)(ARDs) areare aa familyfamily ofof metalloenzymesmetalloenzymes thatthat areare capablecapable ofof usingusing 2+ 2+ didifferentﬀerent transition transition metal metal ions ions (Ni (Ni2+ andand FeFe2+)) to to catalyze catalyze the dioxyg dioxygenationenation of acireductone [[101101,,102102].]. 2+ 2+ 2+ FeFe2+-ARD-ARD and NiNi2+--ARDARD share share exactly the samesame proteinprotein sequence.sequence. However,However, Fe Fe2+-ARD-ARD catalyzescatalyzes thethe 2+ formationformation of of methylthioketobutyrate methylthioketobutyrate and and formate formate [101 [],101 whereas], whereas Ni Ni-ARD2+-ARD oxidizes oxidizes acireductone acireductone into threeinto three products, products, namely namely formate, formate, 3-methylthiopropionic 3-methylthiopropionic acid, along acid, withalong carbon with carbon monoxide monoxide [102]. It[102 has]. 2+ beenIt has previously been previously suggested suggested that the that substrate the substrate binding binding mode formode Fe for-ARD Fe2+- isARD a ﬁve-membered is a five-membered ring, ring, while there is a six-membered ring for Ni2+-ARD (Scheme 1) [103], and this should be the main reason for the formation of different products.

Inorganics 2019, 7, 95 14 of 29

Inorganics 2019, 7, x FOR PEER REVIEW 14 of 29 while there is a six-membered ring for Ni2+-ARD (Scheme1)[ 103], and this should be the main reason for the formation of diﬀerent products. Inorganics 2019, 7, x FOR PEER REVIEW 14 of 29

Scheme 1. The previously suggested mechanisms for ARDs. Adapted from [103], Copyright © 2001, SchemeScheme 1.1. TheThe previouslypreviously suggestedsuggestedAmerican mechanismsmechanisms Chemical forfor ARDs.ARDs. Society. Adapted from [103],], CopyrightCopyright © 2001, American Chemical Society. American Chemical Society. Sparta et al. performed discrete molecular dynamics simulations (QM/DMD) to investigate the hypothesisSpartaSparta of the etet al.differental. performedperformed substrate discretediscrete binding molecularmolecular modes dynamicsdynamics in Fe 2+ simulationssimulations-ARD and (QM(QM/DMD)Ni2+/-DMD)ARD [ toto104 investigateinvestigate]. The QM/DMD thethe 2+2+ 2+2+ hypothesishypothesis ofof thethe didifferentﬀerent substrate substrate2+ binding binding modes modes2+ in in Fe Fe -ARD-ARD andand Ni Ni -ARD-ARD [[104104].]. The QMQM/DMD/DMD simulations suggested that both Fe 2-+ARD and Ni2+-ARD exhibit a six-membered ring coordination, simulations suggested that both Fe 2+-ARD and Ni 2+ -ARD exhibit a six-membered ring coordination, with simulationsthe O1 and suggested O3 atoms that binding both Fe to the-ARD central and Ni metal-ARD ion exhibit (Scheme a six 2-membered, React). A ringsecond coordination,-shell residue withwith thethe O1O1 andand O3O3 atomsatoms bindingbinding toto thethe centralcentral metalmetal ionion (Scheme(Scheme2 2,, React). React). A A second-shell second-shell residue residue Arg154 forms a hydrogen bond to the O2 atom of the substrate, which stabilizes this coordination Arg154Arg154 formsforms aa hydrogenhydrogen bondbond toto thethe O2O2 atomatom ofof thethe substrate,substrate, whichwhich stabilizesstabilizes thisthis coordinationcoordination mode. One snapshot was then selected for DFT calculations in order to understand the mechanisms mode.mode. One snapshot was then selected forfor DFTDFT calculationscalculations inin orderorder toto understandunderstand thethe mechanismsmechanisms 2+ 2+ 2+ 2+ of Feofof- FeARD2+--ARDARD and and andNi Ni- NiARD.2+-ARD.-ARD. The The Theenergy energy energy barrier barrier barrier forfor for the the first ﬁrst dioxygendioxygen dioxygen attack attack attack was was wasnot not considered. not considered. considered. The The startingThestarting startingcomplex complex complex is a is dioxygen a isdioxygen a dioxygen adduct, adduct, adduct, with with with a a peroxyperoxy a peroxy bridging bridging C1 C1 C1 and andand C3 C3C3 of of ofthe the the substrate substrate substrate ( (reaction,react (rioneact, ion, 2+ SchemeSchemeScheme 2). For2 2).). For ForNi2+ Ni Ni-ARD,2+--ARD,ARD, the the the reaction reaction reaction takes takes place place in in the triplettriplet state. state. After After a concerteda concerted cleavage cleavage ofof thethe of the O–O,O–O,O C1–O,– C2 C1–C2,C1,– C2and, andand C2 – C2–C3C2C3– C3bonds, bonds,bonds, carbon carboncarbon monoxide monoxidemonoxide andand twotwotwo carboxylatescarboxylate carboxylates ares are generatedgenerated generated (production,(prod (puctionroduction, , SchemeSchemeScheme 2). The2 2).). The The energy energy energy barrier barrier barrier is is isonly only only 3.7 3.7 3.7 kcal/mol. kcal kcal/mol./mol.

Scheme 2. Reaction mechanism for Ni-acireductone dioxygenase supported by the results of a DFT SchemeSchemestudy. 2. AdaptedReaction 2. Reaction from mechanism mechanism [104], Copyright for for Ni Ni-acireductoneacireductone© 2013 Elsevier dioxygenase Ltd. supported supported by by the the results results of a of DFT a DFT study.study. Adapted Adapted from from [104 [104], Copyright], Copyright ©© 20132013 Elsevier Elsevier Ltd.Ltd. Recently, based on the structure of human acireductone dioxygenase, Borowski et al. have performedRecently, QM/MM based calculations on the structure to revisit of human the reaction acireductone mechanisms dioxygenase, of ARDs [105 Borowski]. For the et Ni al.2+-ARD, have Recently, based on the structure of human acireductone dioxygenase, Borowski2 + et al. have performedthe calculat QMions/MM suggested calculations that the to revisitmechanism the reaction involves mechanisms four major ofsteps ARDs (Scheme [105]. 3 For). A the proton Ni -ARD,is first performed QM/MM calculations to revisit the reaction mechanisms of ARDs [105]. For the Ni2+-ARD, thetransferred calculations from suggested the substrate that tothe His88 mechanism when the involves substrate four enters major into steps the (Scheme active3 site,). A and proton His88 is the calculations suggested that the mechanism involves four major steps (Scheme 3). A proton is first ﬁrstdissociates transferred from from the nickel the substrate ion. Dioxygen to His88 binds when to thethe substratemetal coupled enters with into one the electron active site, transfer and His88 from transferred from the substrate to His88 when the substrate enters into the active site, and His88 dissociatesthe substrate from to the the dioxygen nickel ion. moiety Dioxygen to generate binds to a the Ni2+ metal-superoxide coupled intermediate with one electron (complex transfer 1), which from 2+ dissociatestheis quite substrate from similar tothe theto nickel que dioxygenrcetin ion. 2,3 moietyDioxygen-dioxygenases to generate binds [ 106to a the Ni,107 metal]-superoxide. The coupledground intermediate state with of one comp (complexelectronlex 1 is transfer 1a), quintet which from the substrateisstate quite (high similar to-spin the to nickel(II)dioxygen quercetin ions, moiety 2,3-dioxygenases one electron to generate on [the106 a dioxygen, ,Ni1072+].-superoxide The one ground electron intermediate state on of the complex substrate) (complex1 is, awhile quintet 1), the which is quitestatefollowing similar (high-spin reaction to que nickel(II) isrcetin in the ions,2,3 triplet-dioxygenases one state electron (two on [unpaired106 the,107 dioxygen,] .electrons The ground one on electron the state nickel on of the (II)comp substrate), ion).lex Hence, 1 is whilea tquintethe statethe energy(high following- spinof the nickel(II) reactiontriplet comple isions, in the xone 1 triplet was electron taken state (twoonas thethe unpaired reference. dioxygen, electrons After one theelectron on distal the nickel onoxygen the (II) substrate) atom ion). Hence,attack, whiles thethe the 1 followingenergyC2 atom, reaction of thea peroxo triplet is in intermediate complex the tripletwas (complexstate taken (two as 2 the) isunpairedreference. formed. Thiselectrons After step the was distalon thefound oxygen nickel to atombe (II) barrierattacks ion).-less. Hence, the The C2 the energynext of step the istriplet the migration comple xof 1 the was C2 taken-bound as oxygen the reference. atom to the After C3 atomthe distal (complex oxygen 3), which atom is attack readys the C2 atom,for the a followingperoxo intermediate formation of the(complex O–O bridge 2) is betweenformed. C1This and step C3 (complexwas found 4). toThe be final barrier concerted-less. The next cleavagesstep is the of migration the O–O, C1 of– theC2, andC2- boundC2–C3 oxygenbonds were atom found to the to beC3 the atom rate (complex-limiting step 3), swhich (TS45) is for ready the substrate oxidation, with a barrier of only 6.8 kcal/mol. It is more likely that the substrate binding for the following formation of the O–O bridge between C1 and C3 (complex 4). The final concerted or product release is the rate-limiting step for the whole reaction. cleavages of the O–O, C1–C2, and C2–C3 bonds were found to be the rate-limiting steps (TS45) for the substrate oxidation, with a barrier of only 6.8 kcal/mol. It is more likely that the substrate binding or product release is the rate-limiting step for the whole reaction.

Inorganics 2019, 7, 95 15 of 29

atom, a peroxo intermediate (complex 2) is formed. This step was found to be barrier-less. The next step is the migration of the C2-bound oxygen atom to the C3 atom (complex 3), which is ready for the following formation of the O–O bridge between C1 and C3 (complex 4). The ﬁnal concerted cleavages of the O–O, C1–C2, and C2–C3 bonds were found to be the rate-limiting steps (TS45) for the substrate oxidation, with a barrier of only 6.8 kcal/mol. It is more likely that the substrate binding or product Inorganicsrelease 2019 is, the7, x rate-limitingFOR PEER REVIEW step for the whole reaction. 15 of 29

SchemeScheme 3. 3. Reaction mechanisms mechanisms for Ni-ARD for supportedNi-ARD by supported the results of by a quantum the results mechanics of/ molecular a quantum mechanics/molecularmechanics (QM/MM) mechanics study. Adapted (QM/MM from) study [105],. copyright Adapted© fromWILEY-VCH [105], copyright Verlag GmbH © WILEY & Co.-VCH VerlagKGaA, GmbH Weinheim. & Co. KGaA, Weinheim.

Alternative pathways have also been considered from complex 2. The C2-bound oxygen atom Alternative pathways have also been considered from complex 2. The C2-bound oxygen atom may attack the C1 atom to form complex 6. After the formation of a dioxethane intermediate (complex may attack the C1 atom to form complex 6. After the formation of a dioxethane intermediate (complex 7) and the cleavage of the O–O and C1–C2 bonds (TS78), methylthioketobutyrate and formate are 7) andgenerated the cleavage (complex of8 ).the However, O–O and the C1 barrier–C2 ofbonds TS78 (TS78), was found methylthioketobutyrate to be 11.2 kcal/mol, higher and than formate that are generatedof TS45. (complex Other pathways 8). However, that pass the through barrier a of Baeyer-Villiger TS78 was found type rearrangementto be 11.2 kcal/mol, (2 6 higher10 than11) that → → → of TS45.and the Other homolytic pathways O–O that cleavage pass ( 2through6 9 a) wereBaeyer also-Vill rulediger out type due rearrangement to their higher energy(2  6 barriers  10  11) → → andcompared the homolytic with the O– mostO cleavage favorable (2 mechanism. 6  9) were also ruled out due to their higher energy barriers compared with the most favorable mechanism. 6. Quercetin 2,4-Dioxygenase 6. QuercetinQuercetin 2,4-Dioxygenase 2,4-dioxygenase from Streptomyces sp. strain FLA is a nickel-dependent enzyme catalyzing the oxidative ring-cleavage of quercetin to produce CO and 2-protocatechuoylphloroglucinol Quercetin 2,4-dioxygenase from Streptomyces sp. strain FLA is a nickel-dependent enzyme carboxylic acid using O2 as the oxidant [16,108]. The mechanism for the dioxygen activation and catalyzingsubstrate oxidation the oxidative was ﬁrst investigated ring-cleavage by Liu andof co-workers quercetin [109 to]. From produce the 19 ns CO molecular and 2- protocatechuoylphloroglucinoldynamics simulations, one snapshot carboxylic was chosen acid forusing the followingO2 as the QM oxidant/MM calculations. [16,108]. The The mechanism ﬁrst-shell for the liganddioxygen Glu76 activation set to be deprotonated and substrate as aoxidation proton is transferredwas first investi from thegated neutral by quercetinLiu and substrate.co-workers For [109]. Fromthe the reactant 19 ns complexmolecular with dynamics the O2 binding simulations, to the Ni one ion insnapshot an end-on was fashion, chosen their for calculations the following showed QM/MM calculations.that the quintet The first is the-she groundll ligand state, Glu76 the triplet set liesto be 2.3 deprotonated kcal/mol higher as in a energy, proton while is transferred the singlet wasfrom the neutralfound quercetin to be much substrate. higher in For energy. the reactant It is likely complex that the with closed-shell the O2 binding singlet was to the obtained, Ni ion while in an the end-on fashion,open-shell their singletcalculations was not showed considered. that In the the tripletquintet state, is the one ground electron isstate, transferred the triplet from thelies substrate 2.3 kcal/mol 2+ higherto the in Oenergy,2 moiety, while leading the to singlet the formation was found of a Nito be-superoxide-substrate much higher in energy. radical It species.is likely Importantly, that the closed- shellthe singlet reaction was takes obtained, place in while the triplet the stateopen involving-shell singlet four majorwas not steps considered. (Scheme4). In First, the thetriplet terminal state, one electron is transferred from the substrate to the O2 moiety, leading to the formation of a Ni2+- superoxide-substrate radical species. Importantly, the reaction takes place in the triplet state involving four major steps (Scheme 4). First, the terminal oxygen of the superoxide attacks C2 of the substrate to form the first C–O bond, which has a barrier of only 5.0 kcal/mol. This is followed by a conformational change by a rotation around the newly-formed C2–O bond, and this step turned out to be rate-limiting, with a barrier of 19.9 kcal/mol. Subsequently, the terminal oxygen of the peroxide attacks C4 of the substrate, resulting in a peroxide-bridging intermediate. Finally, simultaneous cleavage of the O–O bond and two C–C bonds produces a CO molecule and the 2- protocatechuoylphloroglucinol carboxylic acid. When the central nickel ion is replaced by an iron ion, the reaction proceeds via a similar pathway in the quintet state. However, the last step becomes rate-limiting, associated with a very high barrier of 29.9 kcal/mol.

Inorganics 2019, 7, 95 16 of 29 oxygen of the superoxide attacks C2 of the substrate to form the ﬁrst C–O bond, which has a barrier of only 5.0 kcal/mol. This is followed by a conformational change by a rotation around the newly-formed C2–O bond, and this step turned out to be rate-limiting, with a barrier of 19.9 kcal/mol. Subsequently, the terminal oxygen of the peroxide attacks C4 of the substrate, resulting in a peroxide-bridging intermediate. Finally, simultaneous cleavage of the O–O bond and two C–C bonds produces a CO molecule and the 2-protocatechuoylphloroglucinol carboxylic acid. When the central nickel ion is replaced by an iron ion, the reaction proceeds via a similar pathway in the quintet state. However, the last step becomes rate-limiting, associated with a very high barrier of 29.9 kcal/mol. Inorganics 2019, 7, x FOR PEER REVIEW 16 of 29

SchemeScheme 4. 4Reaction. Reaction mechanism mechanism for for Ni-QueDNi-QueD suggestedsuggested by Liu Liu et et al. al. Adapted Adapted from from [109 [109], with], with permission from The Royalpermission Society of from Chemistry. The Royal Society of Chemistry.

Liao and co co-workers-workers independently investigated the reaction mechanism of this enzyme almost at the same time [107].]. In In their their QM/MM QM/MM calculations, calculations, the the model model sta startedrted directly directly from from the XX-ray-ray conformation. Importantly, Importantly, the the first ﬁrst-shell-shell ligand ligand Glu74 Glu74 was was considered considered to to be be both neutral and deprotonated. The The aim aim of of that that study study was was to to elucidate elucidate the the mechanism and also to rationalize the chemoselectivity, in which only 2,4-dioxygenolytic2,4-dioxygenolytic cleavage takes place, but not 2,3-dioxygenolytic2,3-dioxygenolytic cleavage. They They found found a a similar similar mechanism mechanism as as Liu Liu et et al. al. [ 109]] when Glu74 is protonated. However, However, the last last step step was was found found to to be be rate rate-limiting-limiting with with a barrier a barrier of of24.8 24.8 kcal/mol. kcal/mol. Unexpectedly, Unexpectedly, the 2,3 the- dioxygenolytic2,3-dioxygenolytic pathway pathway was was found found to tobe bemore more favorable, favorable, with with a abarrier barrier of of 21.8 21.8 kcal/mol. kcal/mol. Therefore, Therefore, the model model with with a protonateda protonated Glu74 Glu74 residue residue could could not reproduce not reprod theuce observed the observed chemoselectivity. chemoselectivity. Instead, Instead,a model witha model a deprotonated with a deprotonated Glu74 was Glu74 demonstrated was demonstrated to reproduce to thereproduce chemoselectivity the chemoselectivity (Scheme5). (InScheme this case, 5). the In thistotal case, barrier the for total the 2,4-dioxygenolytic barrier for the 2,4 pathway-dioxygenolytic decreases pathway to 17.4 decreases kcal/mol, while to 17.4 it kcal/mol,increases to while 30.6 kcal it increases/mol for the to 2,3-dioxygenolytic30.6 kcal/mol for pathway.the 2,3-dioxygenolytic The calculated pathway. barrier of The17.4 kcal calculated/mol is barrierin good o agreementf 17.4 kcal/mol with is the in experimental good agreement kinetic with data the [ 108experimental], which gave kinetic a barrier data of[108 about], which 15 kcal gave/mol. a barrier of about 15 kcal/mol.

Inorganics 2019, 7, 95 17 of 29

Inorganics 2019, 7, x FOR PEER REVIEW 17 of 29

SchemeScheme 5. 5Reaction. Reaction mechanism mechanism and and chemoselectivitychemoselectivity for Ni Ni-QueD-QueD suggested suggested by by Liao Liao etet al. al. Adapted Adapted fromfrom [107 [107], with], with permission permission from from the the PCCP PCCP OwnerOwner Societies.Societies.

7. Urease7. Urease UreaseUrease is is a nickel-containinga nickel-containing enzyme enzyme thatthat catalyzescatalyzes the hydrolysis hydrolysis of of urea urea to to yield yield ammonia ammonia and and carboncarbon dioxide dioxide in in the the last last step step of of nitrogen nitrogen mineralizationmineralization [ 110]].. The The uncatalyzed uncatalyzed reaction reaction has has been been suggestedsuggested by by Merz Merz and and Estiu Estiu [ 111[111]] to to proceed proceed viavia an elimination path pathway.way. For For the the enzyme enzyme catalyzed catalyzed reaction,reaction, there there have have been been three three di differentfferent mechanistic mechanistic proposals. proposals. First, First, Karplus Karplus et et al. al. suggested suggested a a mechanismmechanism in whichin which a Nickel-coordinated a Nickel-coordinated terminal terminal hydroxide hydroxide performs performs the thenucleophilic nucleophilic attack, attack, while anwhile adjacent an adjacent protonated protonated His320 residueHis320 residue delivers delivers a proton a proton to the leaving to the leaving amino amino group group [110,112 [110]., Second,112]. BeniniSecond, et al. Benini proposed et al. proposed a mechanism a mechanism involving involving a nucleophilic a nucleophilic attack attack of a bridging of a bridging hydroxide hydroxide on the carbonylon the carbon carbonyl of urea, carbon followed of urea, by followed the protonation by the protonation of the leaving of amino the lea groupving amino by this group hydroxide by this [113 ]. hydroxide [113]. Third, Lippard et al. suggested an elimination mechanism, similarly to the Third, Lippard et al. suggested an elimination mechanism, similarly to the uncatalyzed reaction [114]. uncatalyzed reaction [114]. However, this elimination pathway could not explain the experimental However, this elimination pathway could not explain the experimental observation of carbamate as observation of carbamate as the first product found by Zerner et al. [115]. A number of computational the first product found by Zerner et al. [115]. A number of computational studies have, therefore, been studies have, therefore, been conducted to elucidate the reaction mechanism of this enzyme [116– conducted to elucidate the reaction mechanism of this enzyme [116–123]. 123]. TheThe two two diff erent different hydrolytic hydrolytic mechanisms mechanisms were investigatedwere investigated by Merz by et Merz al. using et al. DFT using calculations DFT (Schemescalculations6 and (7Scheme)[ 120 ]. 6 Forand theScheme bridging 7) [120 hydroxide]. For the attack bridging mechanism, hydroxide the attack calculations mechanism, suggested the II thatcalculations the reaction suggested starts from that a the Ni reaction2 complex starts with from a bridging a NiII2 hydroxidecomplex with (Scheme a bridging6). In the hydroxide reactant complex(Scheme S3, 6 the). In urea the reactant substrate complex binds in S3, a bi-dentatethe urea substrate fashion, wherebinds in the a carbonylbi-dentate oxygen fashion, coordinates where the to II Ni1,carbonyl while oneoxygen of the coordinate amino groupss to Ni1 coordinate, while one of to the Ni2. amino Both group Ni ionss coordinate are in the to tripletNi2. Both state, NiII and ions the complexare in the is, thus,triplet a state, quintet and for the the complex ferromagnetically-coupled is, thus, a quintet for the state. ferromagnetically The bridging hydroxide-coupled state. performs The a nucleophilicbridging hydroxide attack on performs the carbonyl a nucleophilic carbon of the attack urea on substrate the carbonyl via TS1A, carbon leading of the tourea the substrate formation via of a tetrahedralTS1A, leading intermediate to the formation (I1A) with of the a tetrahedral oxyanion stabilizedintermediate by one(I1A) of with the two the nickeloxyanion ions. stabilized This pathway by is consistentone of the two with nickel the proposal ions. This by pathway Ciurli et is al.,consistent on the basiswith the of dockingproposal studiesby Ciurli [120 et ].al. Then,, on the a basis proton is deliveredof docking from studies the [120 bridging]. Then, hydroxidea proton is todelivered the amino from group the bridging of urea hydroxide (TS2A), facilitated to the amino by group Asp360. This leads to the production of an ammonia molecule and a carbamate. The second step was found

Inorganics 2019, 7, x FOR PEER REVIEW 18 of 29 InorganicsInorganics2019 2019, 7,, 957, x FOR PEER REVIEW 18 of18 29 of 29

of urea (TS2A), facilitated by Asp360. This leads to the production of an ammonia molecule and a tocarbamate.carbamate. be rate-limiting, TheThe secondsecond with stepstep a barrier waswas foundfound of 19.7 toto bebe kcal raterate/mol.--limiting,limiting, Protonation with a barrier of the of leaving 19.7 kcal/mol. ammonia Protonation group by a protonatedof the leaving His320 ammonia residue group has by also a protonated been taken His320 into residue account, has and also this been step taken is closeinto account, to isothermic. and this step is close to isothermic. It should be noted that the role of this histidine residue has very It shouldthis step be is noted close that to isothermic. the role of this It should histidine be residuenoted that has verythe role recently of this been histidine confirmed residue by modulatinghas very recently been confirmed by modulating its protonation state and conformation of the mobile flap itsrecently protonation been stateconfirmed and conformation by modulating of its the protonation mobile flap state [124 and]. Afterconformation the release of the of themobile ammonia flap [124]. After the release of the ammonia molecule, a water molecule binds to Ni2, and this water molecule,[124]. After a water the molecule release of binds the ammonia to Ni2, and molecule, this water a water molecule molecul thene binds performs to Ni2, a nucleophilic and this water attack molecule then performs a nucleophilic attack on the carbon atom of the carbamate anion (TS4A), onmolecule the carbon then atom performs of the carbamatea nucleophilic anion attack (TS4A), on the concertedly carbon atom with of a protonthe carbamate transfer anion from (TS4A), the water concertedly with a proton transfer from the water molecule to the nearby Asp360 residue. Finally, moleculeconcertedly to the with nearby a proton Asp360 transfer residue. from Finally, the water the protonated molecule to Asp360 the nearby residue Asp360 delivers residue. a proton Finally, to the thethe protonatedprotonated Asp360Asp360 residueresidue dedeliverslivers aa protonproton toto thethe bridgingbridging oxyanionoxyanion (TS6A),(TS6A), coupledcoupled withwith thethe bridgingC–O bond oxyanion cleavage. (TS6A), The di coupled-nickel complex with the (I7A) C–O with bond a cleavage.bridging hydroxide The di-nickel is regenerated. complex (I7A) It should with a bridgingC–O bond hydroxide cleavage. is The regenerated. di-nickel complex It should (I7A) be pointed with a bridging out that hydroxide this kind ofis mechanismregenerated. agreesIt should very be pointed out that this kind of mechanism agrees very well with the recent crystal structurere ofof thethe well with the recent crystal structure of the complex between urea and fluoride-inhibited urease, in complexcomplex betweenbetween ureaurea andand fluoridefluoride--inhibitedinhibited urease,urease, inin whichwhich anan unreactiveunreactive fluoridefluoride replacesreplaces thethe which an unreactive fluoride replaces the bridging hydroxide [125]. bridging hydroxide [125125].].

Lys217 Lys217 HN HN

O O His134 246His O O His134 246His Ni12+ Ni22+ His136 2+ 2+ His136 272His Ni1 O Ni2 272His O O O NH O C 2 O NH C 2 Asp360 HO N Asp360 H N H HO H H H I1A-H+ (10.7) H I1A-H+ (10.7) Lys217 Lys217 Lys217 Lys217 His320 Lys217 Lys217 Lys217 Lys217 His320 HN HN HN HN H+ HN HN HN HN His320 His320H+ O O His134 O O His134 O O O His134 O His134 246His O O His134 O 246His O His134 246His O His134 246His O O His134 O 2+ 2+ His136 2+ 2+ His136 2+ 2+ His136 246His Ni1 Ni2 Ni1 Ni2 2+ 2+ His136 246His Ni1 Ni2 246His 246His Ni1 Ni2 2+ 2+ His136 2+ 2+ His136 O TS1A 2+ H 2+ His136 TS2A 272His Ni1 O Ni2 272His Ni1 O Ni2 Ni12+ H Ni22+ His136 272His Ni1 O Ni2 TS3A 272His O H 272His O O TS1A 272His O TS2A O O TS3A 272His O O 272His O H O 8.3 O O 19.7 NH2 O NH2 O C Asp360 16.9 C O NH2 O O C O NH O O C 8.3 NH 19.7 C 2 Asp360 16.9 NH2 C 2 C C Asp360 Asp360 O O Asp360 N NH O Asp360 N O 2 O Asp360 N Asp360 N O O H H NH2 H N H O H N H H H H NH H H H H 3 H NH S3 (0.0) I1A (10.9) I2A (13.1) I3A (-5.1) 3 S3 (0.0) I1A (10.9) I2A (13.1) I3A (-5.1)

H2O H2O

NH3 NH3 Lys217 Lys217 Lys217 Lys217 Lys217 Lys217 Lys217 Lys217 HN HN HN HN HN HN HN HN

O O O His134 O O His134 O His134 246His O O His134 O O O His134 246His O O His134 246His O His134 TS6A 246His 246His O O His134 2+ 2+ His136 Ni12+ Ni22+ His136 TS5A 246His Ni12+ Ni22+ His136 TS4A 246His Ni1 Ni2 272His 246His Ni12+ Ni22+ His136 2+ 2+ His136 TS6A 2+ 2+ His136 TS5A 2+ 2+ His136 TS4A Ni1 Ni2 Ni1 Ni2 272His Ni1 Ni2 2+ 2+ His136 272His O 15.2 272His O 19.3 O 6.8 272His Ni1 O Ni2 H O O 272His O O O 15.2 O O 272His O O OH O OH O 19.3 O 6.8 272His O O O H O O C O C O OH O C H O Asp360 OH O H C OH C C H Asp360 Asp360 Asp360 H-O Asp360 C C OH H-O H O Asp360 H2N H2N O Asp360 H-O Asp360 NH2 H2N H-O O H2N H2N O NH2 H2N I6A (11.9) I5A (10.5) I7A (-3.6) I6A (11.9) I5A (10.5) I4A (-7.1) I7A (-3.6) I4A (-7.1) Scheme 6. The bridging hydroxide attack mechanism suggested by Merz et al. on the basis of DFT SchemeScheme 6. 6The. The bridging bridging hydroxide hydroxide attackattack mechanismmechanism suggested by by Merz Merz et et al. al. on on the the basis basis of ofDFT DFT calculations. Adapted from [120], Copyright © 2003, American Chemical Society. Energies are given calculations.calculations. Adapted Adapted from from [120 [120],], Copyright Copyright© ©2003, 2003, American AmericanChemical Chemical Society.Society. EnergiesEnergies areare givengiven in in kcal/mol relative to S3. kcalin/ molkcal/mol relative relative to S3. to S3.

Lys217 Lys217 Lys217 Lys217 Lys217 Lys217 HN HN HN HN HN HN

NH O O His134 O O His134 3 O O His134 O O His134 O His134 NH3 O 246His 246His O 246His O His134 2+ TS1B 246His Ni1 Ni22+ His136 246His Ni12+ Ni22+ His136 246His 2+ 2+ His136 2+ O 2+ His136 TS1B 2+ 2+ His136 Ni1 Ni2 Ni1 Ni2 Ni1 O Ni2 Ni12+ O Ni22+ His136 272His HO 24.1 272His HO TS2B 272His HO 272His H O 24.1 272His H O TS2B 272His H O H2O O O H O O OH O 24.6 O O 2 O OH 24.6 O O Asp360 C Asp360 C Asp360 H Asp360 O O C O C O C NH2 H N NH Asp360 O 2 O H2N C OH NH2 N NH Asp360 N H 2 H2N OH N H H H H H I1B(20.0) I2B (-5.2) H S1B (-7.1) H I1B(20.0) I2B (-5.2) S1B (-7.1) Scheme 7. The nickel bound water nucleophilic attack mechanism considered by Merz et al. on the SchemeScheme 7. 7The. The nickel nickel bound bound water water nucleophilic nucleophilic attackattack mechanism conside consideredred by by Merz Merz et etal. al. on on the the basis of DFT calculations [120], Copyright © 2003, American Chemical Society. Energies are given basisbasis of of DFT DFT calculations calculations [120 [120],], Copyright Copyright© ©2003, 2003, AmericanAmerican ChemicalChemical Society. EnergiesEnergies areare givengiven in in kcal/mol relative to S3 in Scheme 6. kcalin/ molkcal/mol relative relative to S3 to in S3 Scheme in Scheme6. 6.

Inorganics 2019, 7, 95 19 of 29 Inorganics 2019, 7, x FOR PEER REVIEW 19 of 29

NucleophilicNucleophilic attack attack by by the the nickel nickel bound bound water water molecule molecule was was also also taken taken into considerationinto consideration by Merz by etMe al.rz [ 120et al.], and[120 the], and starting the start complexing complex was a monodentate was a monodentate urea-bound urea- adductbound withadduct the with carbonyl the carbonyl oxygen coordinatedoxygen coordinated to Ni1 (S1B, to Ni1 Scheme (S1B,7 ).Scheme The calculations 7). The calculations suggested suggested that this mechanism that this mechanism involves only involves two stepsonly two (Scheme steps7 ):(Scheme the Ni2-bound 7): the Ni2 water-bound molecule water performsmolecule aperforms nucleophilic a nucleophilic attack on theattack urea on carbonyl the urea carbon,carbonyl which carbon, is coupledwhich is with coupled a proton with transfer a proton from transfer the water from moleculethe water to molecule the leaving to the ammonia, leaving assistedammonia, by assisted the bridging by th hydroxidee bridging (TS1B). hydroxide The next(TS1B). step The is the next C–N step bond is the cleavage C–N bond with cleavage the release with of anthe ammonia release of molecule. an ammonia The secondmolecule. step The is rate-limiting,second step is with rate a-limiting, barrier of with 24.6 a kcal barrier/mol, of which 24.6 kca is higherl/mol, thanwhich the is onehigher of the than bridging the one attack of the by bridging a hydroxide. attack by a hydroxide. TheyThey havehave alsoalso performedperformed DFTDFT calculationscalculations toto analyzeanalyze thethe eliminationelimination mechanismmechanism ofof ureaseurease (Scheme(Scheme8 )[8) [122122].]. TheyThey startedstarted fromfrom thethe enzymeenzyme withoutwithout thethe ureaurea substrate,substrate, andand thethe substratesubstrate bindingbinding waswas calculatedcalculated to to be be endothermic endothermic by by 5.0 5.0 kcal kcal/mol./mol. Then, Then, deprotonation deprotonation of theof the amino amino group group of ureaof urea by theby bridging the bridging hydroxide hydroxide should should take place take (TS3uw) place (TS3uw) and a new and bridging a new water bridging molecule water is molecule generated is (I3uw).generated In (I3uw). the next In step, the next a proton step, isa proton transferred is transferred from the from bridging the bridging water to water the other to the amino other groupamino ofgroup urea of (TS4uw), urea (TS4uw), with a with barrier a barrier of 30.5 of kcal 30.5/mol. kcal/mol. Subsequently, Subsequently, a water a water molecule molecule enters ent intoers into the activethe active site andsite and forms forms hydrogen hydrogen bonds bonds with with the twothe two Ni-bound Ni-bound water water molecules, molecules, which which facilitates facilitates the + releasethe release of NH of4 NHand4+ and OCN OCN−. This−. This mechanism mechanism has has a substantially a substantially higher higher barrier barrier than than those those for the for two the othertwo other mechanisms. mechanisms.

SchemeScheme 8.8The. The elimination elimination mechanism mechanism of urea of considered urea considered by Merz by et Merzal. on the et al. basis on of the DFT basis calculations. of DFT Adaptedcalculations from. Adapted [122], Copyright from [122©], 2007,Copyright American © 2007, Chemical American Society. Chemical Society.

8.8. LactateLactate RacemaseRacemase LactateLactate racemaseracemase (LarA)(LarA) hashas beenbeen identifiedidentified toto bebe thethe enzymeenzyme thatthat catalyzescatalyzes thethe finalfinal stepstep inin thethe d l racemizationracemization between between -D and- and-lactic L-lactic acids acids [126 ],[126 using], using a previously a previously unknown unknown Ni cofactor Ni (Ni-PTTMN), cofactor (Ni- pyridinium-3-thioamide-5-thiocarboxylicPTTMN), pyridinium-3-thioamide-5-thiocarboxylic acid mononucleotide acid mononucleotide Ni pincer complex Ni pincer [19 complex,127] (Figure [19, 12710).] Ni-PTTMN(Figure 10). isNi formed-PTTMN by is the formed coordination by the coordination between a Ni(II) between ion and a Ni(II) a prosthetic ion and group a prosthetic derived group from nicotinicderived acidfrom (Figure nicotinic 10 ). acid The ( latterFigure has 10 a). thioamide The latter (modified has a thioamide from the (modified nicotinic moiety) from the covalently nicotinic boundmoiety) to covalently Lys184 and bo aund thiocarboxylate, to Lys184 and and a thiocarboxylate, acts as a tridentate and pincer acts as ligand a tridentate with the pincer two sulfursligand with and onethe pyridinetwo sulfurs carbon and ligatingone pyridine to the carbon nickel, ligating where ato Ni–C the nickel, bond is where formed. a Ni The–C His200bond is worksformed. as The the fourthHis200 nickel works ligand. as the fourth The Ni-PTTMN nickel ligand. cofactor The inNi LarA-PTTMN serves cofactor as a rare in LarA example serves of aas pincer a rare complexexample foundof a pincer in enzymes complex and found its Ni–C in bondenzymes is the and first its case Ni– ofC abond metal-carbon is the first bond case originating of a metal from-carbon nicotinic bond originating from nicotinic acid in biology [128]. Considering the common and widespread NAD+- type cofactors (nicotinamide adenine dinucleotide and its phosphorylated form, NADP+) derived

Inorganics 2019, 7, 95 20 of 29

acid inInorganics biology 2019 [,128 7, x ].FOR Considering PEER REVIEW the common and widespread NAD+-type cofactors (nicotinamide20 of 29 adenine dinucleotide and its phosphorylated form, NADP+) derived from nicotinic acid, it is surprising from nicotinic acid, it is surprising that such a complicated Ni-PTTMN cofactor has been biologically that such a complicated Ni-PTTMN cofactor has been biologically evolved with extra consumption of evolved with extra consumption of resources. A question is whether it has evolutional advantages resources. A question is whether it has evolutional advantages compared with NAD+ [128]. compared with NAD+ [128].

Figure 10. A modified proton-coupled hydride transfer mechanism for lactate racemase (LarA) Figure 10. A modified proton-coupled hydride transfer mechanism for lactate racemase (LarA) proposed by Chen et al. Adapted from [129], copyright © WILEY-VCH Verlag GmbH & Co. KGaA, proposed by Chen et al. Adapted from [129], copyright © WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Tyr294 and Lys298 play crucial roles in orienting substrates and stabilizing the negative Weinheim. Tyr294 and Lys298 play crucial roles in orienting substrates and stabilizing the negative charge developing at the substrate hydroxyl oxygen in the transition states, thus lowering reaction chargebarriers developing significantly. at the substrate hydroxyl oxygen in the transition states, thus lowering reaction barriers significantly. A proton-coupled hydride transfer (PCHT) mechanism using a Ni(II) ion (Figure 10) has been Averified proton-coupled by two DFT hydride cluster modeling transfer investigations (PCHT) mechanism by Chung using et al. a[130 Ni(II)] and ion Chen (Figure et al. 10[129) has]. In been verifiedthis bymechanism, two DFT a clusterhydridemodeling is transferred investigations to the Ni-PTTMN by Chung pyridine et carbon al. [130 from] and the Chen substrate et al. α- [129]. In thiscarbon mechanism,, coupled a with hydride a proton is transferred transfer from to the the substrate Ni-PTTMN hydroxyl pyridine to a histidine, carbon from followed the substrate by a α-carbon,transfer coupled back to withthe α- acarbon proton of transferthe resultant from pyruvate the substrate from the hydroxyl opposite side to a. histidine,This is achieved followed via a by a transferrotation back of to the the pyruvateα-carbon acetyl of the moiety. resultant Chung pyruvate et al. used from an active the opposite-site model side. of 139 This atoms is achievedto give a via a rotationbarrier of of the ~18 pyruvatekcal/mol for acetyl the hydri moiety.de transfer Chung at etthe al. B3LYP used-D3 an level active-site [130]. Based model on small of 139 models atoms to (mainly involving cofactors and a proton-acceptor imidazole) and low-polarity solvation effects with give a barrier of ~18 kcal/mol for the hydride transfer at the B3LYP-D3 level [130]. Based on small a dielectric constant (ε) of 4, the further comparison of Ni-PTTMN with NAD+-like cofactors showed models (mainly involving cofactors and a proton-acceptor imidazole) and low-polarity solvation effects that the Ni-PTTMN cofactor has a higher racemization barrier than NAD+ [130], which makes it + with ainteresting dielectric to constant explore the (ε) LarA of 4, activity the further in the comparison full active-site of Ni-PTTMNenvironment. withIn the NAD subsequent-like cofactorsDFT + showedcalculations that the by Ni-PTTMN Chen et al. using cofactor a 200 has-atom a highermodel with racemization Tyr294 and barrier full side than chains NAD of the [residues130], which makesincluded, it interesting a lower to rate explore-limiting the LarAbarrier activity of 12 kcal/mol in the full for active-sitethe hydride environment. transfer was predicted In the subsequent. The DFT calculationscatalytic acceleration by Chen effect et al.is achieved using a via 200-atom the stabilization model withof the Tyr294 transition and states full sideby Tyr294 chains and of the residuesLys298 included, [129]. Further a lower calculations rate-limiting were barrier performed of 12 with kcal various/mol formodified the hydride Ni-PTTMN transfer cofactors was predicted.in the The catalytic acceleration effect is achieved via the stabilization of the transition states by Tyr294 and

Lys298 [129]. Further calculations were performed with various modiﬁed Ni-PTTMN cofactors in the Inorganics 2019, 7, 95 21 of 29

LarA active site. They showed that the barrier of 12 kcal/mol for the native LarA with a Ni-PTTMN cofactor is much lower than those (at least by 19 kcal/mol) for the LarA enzymes with NAD+-like cofactors [129], indicating an enhanced racemization activity of Ni-PTTMN. To further analyze the acceleration effects of Ni-PTTMN, the hydride affinities of the Ni-PTTMN and NAD+-like cofactors were calculated [129]. It was revealed that compared with NAD+-like cofactors, Ni-PTTMN has a hydride affinity less sensitive to the environmental polarity, thus keeping a stronger hydride-addition reactivity in moderately and highly polar surroundings (ε 8), while a weaker one in a low-polarity ≥ environment (ε < 8) [129]. This not only explains why a higher barrier for Ni-PTTMN was obtained in the calculations by Chung et al. [130], but also indicates the evolutionary advantage of Ni-PTTMN using an architecture with a Ni pincer, which fits perfectly in the moderately polar active site of LarA, in which a dielectric constant of around 14–18 was estimated by a barrier analysis [130]. Inspired by the Ni-PTTMN-dependent LarA reaction, a series of Ni-PTTMN-like metal pincer complexes (metal = Ni and Pd) were proposed by Yang et al. using DFT calculations to be potential catalysts for lactate racemization with a barrier as low as ~26 kcal/mol, adopting the same PCHT mechanism as LarA [131], which is a step closer to utilizing the novel Ni-PTTMN cofactor and its related chemistry in man-made catalysis. Another mechanism starting from a Ni(III) state was proposed in a QM/MM study by Shaik et al. [132]. It includes the electron transfer from the lactate carboxylate to Ni(III), the C–C bond dissociation leading to a CO2 radical anion and an acetaldehyde, the rotation of acetaldehyde, and then a C–C rebound step and the electron transfer back to the carboxylate. In this mechanism, the Ni-PTTMN cofactor works as a reversible electrode to accept and donate back an electron and does not result in covalent bonding to the lactate moiety. However, it is not clear if a Ni(II) Ni-PTTMN cofactor can be oxidized to the Ni(III) state in the LarA active site [133].

9. Superoxide Dismutase Superoxide dismutases (SOD) catalyze the disproportionation of the toxic superoxide radicals into the less toxic hydrogen peroxide and molecular oxygen,

+ 2O − + 2H O + H O (2) 2 · → 2 2 2 They have an important protective role, particularly in aerobic metabolism and photosynthesis. There are at least three different types of SOD’s, and in one of them there is an active nickel complex. The metal complex in Ni-SOD has a mononuclear nickel, coordinated to two cysteines and two backbone nitrogens in a square planar configuration (Figure 11). This is very similar to the coordination of Niu in acetyl CoA synthase (see above). However, there is one additional, axial, histidine ligand, which stabilizes a Ni(III) oxidation state. The disproportionation reaction occurs in two consecutive half-reactions, each one with one superoxide radical substrate. In the oxidative phase, the first superoxide radical binds and is transformed to dioxygen,

+ Ni(III)-L + O − + H Ni(II)-LH + O (3) 2 · → 2 In the reductive phase, the second superoxide substrate binds and is transformed to hydrogen peroxide, + Ni(II)-LH + O − + H Ni(III)-L + H O (4) 2 · → 2 2 Shortly after the first high-resolution X-ray structure appeared in 2004 [17], there were two independent quantum chemical studies of the mechanism [134,135]. Essentially two different mechanisms had been suggested based on experiments. The most popular one was based on the findings on the structure formed by X-ray reduction [17,136]. It was found that upon reduction, the axial histidine lost its coordination to nickel. Therefore, it was suggested that the His became protonated as the proton mediator in the mechanism. The alternative is that one of the cysteines Inorganics 2019, 7, 95 22 of 29 becomes protonated instead. The calculations used essentially minimal cluster models, including only the directly coordinating ligands and the backbone connecting the two cysteines. Inorganics 2019, 7, x FOR PEER REVIEW 22 of 29 The mechanism obtained in the first study [134] is shown in Figure 11. It was suggested that the superoxide substrate enters as an O H radical and binds in the empty axial site, where the cost for The mechanism obtained in the2 fi·rst study [134] is shown in Figure 11. It was suggested that the the protonationsuperoxide substrate was included enters inas thean O energetics.2H· radical Anti-ferromagneticand binds in the empty coupling axial site, between where the nickel cost andfor the radicalthe was protonation found to was be included best. After in the substrate energetics. binding, Anti-ferromagnetic the proton movescoupling over between to Cys2 nickel and and O 2theleaves. Nickelradical has now was beenfound reduced to be best. to After Ni(II). substrate The second binding, substrate the proton radical moves then over enters to Cys2 and and binds O2 atleaves. the same Nickel has now been reduced to Ni(II). The second substrate radical then enters and binds at the same axial position. The proton on Cys2 is now well suited to move over to the O2H− ligand to form H2O2, - and theaxial disproportionation position. The proton is on completed. Cys2 is now The well protonation suited to move of His1 over was to the found O2H to ligand be much to form less H favorable.2O2, and the disproportionation is completed. The protonation of His1 was found to be much less The barrier for the first half-reaction was found to be 9.7 kcal/mol and for the second one 11.5 kcal/mol, favorable. The barrier for the first half-reaction was found to be 9.7 kcal/mol and for the second one in good11.5 agreement kcal/mol, in with good expectations agreement with based expectations on experiments. based on experiments.

FigureFigure 11. Reaction 11. Reaction mechanism mechanism for for superoxide superoxide dismutase dismutase (Ni-SOD)(Ni-SOD) suggested by by P Pelmenschikovelmenschikov et et al. al. Adapted from [134], Copyright © 2006, American Chemical Society. Adapted from [134], Copyright © 2006, American Chemical Society.

Inorganics 2019, 7, 95 23 of 29

In the second study, a different mechanism was found for the first half-reaction, which was the only one studied [135]. A few second shell residues were included in the model apart from the directly binding ones. It was suggested that Cys6 started out protonated and donated its proton to the first superoxide. H2O2 can then be formed by a donation of a proton from the medium via His1 and Tyr9. A proton donating role of His1 was, therefore, suggested in line with earlier experimental suggestions, based on the X-ray reduced structure [17,136]. Transition states were not determined due to the large size of the model. Two additional studies can be mentioned. Several years later in 2011, a study was made [137] that reached similar conclusions as the earlier ones. An energy diagram was calculated showing a rate-limiting step with a barrier for one of them of around 19 kcal/mol. In another study on a mimic of SOD, multireference effects were studied with the use of the complete active space self-consistent field method (CASSCF) [138]. It was found that the near-degeneracy effects were small, which was considered surprising at the time. Multi-reference effects were expected, since a radical became bound to a transition metal.

10. Conclusions In this review, we have presented the progress of theoretical studies on the reaction mechanisms of nickel-dependent enzymes. Both quantum chemical cluster and QM/MM approaches have been successfully used to answer mechanistic questions in these enzymes. Detailed information can be obtained on the structures, energies, electron densities, and spin densities of all relevant intermediates and transition states. Nickel is redox active and the common oxidation states are NiI, NiII, and NiIII in Ni enzymes. Due to its ﬂexible coordination geometry, its redox potential can be tuned by diﬀerent ligand coordinations with a span from +0.89 V to 0.60 V. Therefore, it can be used for challenging redox reactions, such as − for superoxide dismutase, selective reduction of CO2, and methane formation. In addition, the nickel ion can also function as a Lewis acid to stabilize anionic intermediates and transition states during the reactions.

Author Contributions: Writing—original draft preparation, P.E.M.S., S.-L.C., and R.-Z.L.; writing—review and editing, P.E.M.S., S.-L.C., and R.-Z.L. Funding: This research was funded by the Swedish Research Council, the Knut and Alice Wallenberg Foundation, the National Natural Science Foundation of China, grant numbers 21673019 and 21873031. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Walsh, C.T.; Orme-Johnson, W.H. Nickel Enzymes. Biochemistry 1987, 26, 4901–4906. [CrossRef][PubMed] 2. Ermler, U.; Grabarse, W.; Shima, S.; Goubeaud, M.; Thauer, R.K. Active sites of transition-metal enzymes with a focus on nickel. Curr. Opin. Struct. Biol. 1998, 8, 749–758. [CrossRef] 3. Maroney, M.J. Structure/function relationships in nickel metallobiochemistry. Curr. Opin. Chem. Biol. 1999, 3, 188–199. [CrossRef] 4. Mulrooney, S.B.; Hausinger, R.P. Nickel uptake and utilization by microorganisms. FEMS Microbiol. Rev. 2003, 27, 239–261. [CrossRef] 5. Li, Y.; Zamble, D.B. Nickel Homeostasis and Nickel Regulation: An Overview. Chem. Rev. 2009, 109, 4617–4643. [CrossRef] 6. Ragsdale, S.W. Nickel-based Enzyme Systems. J. Biol. Chem. 2009, 284, 18571–18575. [CrossRef] 7. Fontecilla-Camps, J.C.; Amara, P.; Cavazza, C.; Nicolet, Y.; Volbeda, A. Structure–function relationships of anaerobic gas-processing metalloenzymes. Nature 2009, 460, 814–822. [CrossRef][PubMed] 8. Maroney, M.J.; Ciurli, S. Nonredox Nickel Enzymes. Chem. Rev. 2014, 114, 4206–4228. [CrossRef][PubMed] 9. Boer, J.L.; Mulrooney, S.B.; Hausinger, R.P. Nickel-dependent metalloenzymes. Arch. Biochem. Biophys. 2014, 544, 142–152. [CrossRef][PubMed] Inorganics 2019, 7, 95 24 of 29

10. Dixon, N.E.; Gazzola, C.; Blakeley, R.L.; Zerner, B. Jack Bean Urease (EC 3.5.1.5). A Metalloenzyme. A Simple Biological Role for Nickel? J. Am. Chem. Soc. 1975, 97, 4131–4133. [CrossRef] 11. Volbeda, A.; Charon, M.H.; Piras, C.; Hatchikian, E.C.; Frey, M.; Fontecilla-Camps, J.C. Crystal structure of the nickel–iron hydrogenase from Desulfovibrio gigas. Nature 1995, 373, 580–587. [CrossRef][PubMed] 12. Ermler, U.; Grabarse, W.; Shima, S.; Goubeaud, M.; Thauer, R.K. Crystal Structure of Methyl-Coenzyme M Reductase: The Key Enzyme of Biological Methane Formation. Science 1997, 278, 1457–1462. [CrossRef] [PubMed] 13. Jeoung, J.H.; Dobbek, H. Carbon dioxide activation at the Ni, Fe-cluster of anaerobic carbon monoxide dehydrogenase. Science 2007, 318, 1461–1464. [CrossRef][PubMed] 14. Doukov, T.; Iverson, T.M.; Seravalli, J.; Ragsdale, S.W.; Drennan, C.L. A Ni–Fe–Cu Center in a Bifunctional Carbon Monoxide Dehydrogenase/Acetyl-CoA Synthase. Science 2002, 298, 567–572. [CrossRef][PubMed] 15. Pochapsky, T.C.; Pochapsky, S.S.; Ju, T.; Hoeﬂer, C.; Liang, J. A reﬁned model for the structure of acireductone dioxygenase from Klebsiella ATCC 8724 incorporating residual dipolar couplings. J. Biomol. NMR. 2006, 34, 117–127. [CrossRef][PubMed] 16. Jeoung, J.H.; Nianios, D.; Fetzner, S.; Dobbek, H. Quercetin 2,4-Dioxygenase Activates Dioxygen in a Side-On

O2–Ni Complex. Angew. Chem. Int. Ed. 2016, 55, 3281–3284. [CrossRef] 17. Wuerges, J.; Lee, J.W.; Yim, Y.I.; Yim, H.S.; Kang, S.O.; Carugo, K.D. Crystal structure of nickel-containing superoxide dismutase reveals another type of active site. Proc. Natl Acad. Sci. USA 2004, 101, 8569–8574. [CrossRef] 18. He, M.M.; Clugston, S.L.; Honek, J.F.; Matthews, B.W. Determination of the Structure of Escherichia coli Glyoxalase I Suggests a Structural Basis for Diﬀerential Metal Activation. Biochemistry 2000, 39, 8719–8727. [CrossRef] 19. Desguin, B.; Zhang, T.; Soumillion, P.; Hols, P.; Hu, J.; Hausinger, R.P. A tethered niacin-derived pincer complex with a nickel–carbon bond in lactate racemase. Science 2015, 349, 66–69. [CrossRef] 20. Siegbahn, P.E.M.; Blomberg, M.R. Transition-metal systems in biochemistry studied by high-accuracy quantumchemical methods. Chem. Rev. 2000, 100, 421–437. [CrossRef] 21. Himo, F.; Siegbahn, P.E.M. Quantum chemical studies of radical containing enzymes. Chem. Rev. 2003, 103, 2421–2456. [CrossRef] 22. Siegbahn, P.E.M.; Borowski, T. Modeling enzymatic reactions involving transition metals. Acc. Chem. Res. 2006, 39, 729–738. [CrossRef] 23. Siegbahn, P.E.M.; Himo, F. Recent developments of the quantum chemical cluster approach for modeling enzyme reactions. J. Biol. Inorg. Chem. 2009, 14, 643–651. [CrossRef] 24. Siegbahn, P.E.M.; Himo, F. The quantum chemical cluster approach for modeling enzyme reactions. WIREs Comput. Mol. Sci. 2011, 1, 323–336. [CrossRef] 25. Himo, F. Recent trends in quantum chemical modeling of enzymatic reactions. J. Am. Chem. Soc. 2017, 139, 6780–6786. [CrossRef] 26. Blomberg, M.R.; Borowski, T.; Himo, F.; Liao, R.Z.; Siegbahn, P.E.M.Quantum chemical studies of mechanisms for metalloenzymes. Chem. Rev. 2014, 114, 3601–3658. [CrossRef] 27. Warshel, A.; Levitt, M. Theoretical studies of enzymatic reactions dielectric, electrostatic, and steric stabilization of carbonium-ion in reaction of lysozyme. J. Mol. Biol. 1976, 103, 227–249. [CrossRef] 28. Liao, R.Z.; Thiel, W. Comparison of QM-only and QM/MM models for the mechanism of tungsten-dependent acetylene hydratase. J. Chem. Theory Comput. 2012, 8, 3793–3803. [CrossRef] 29. Liao, R.Z.; Thiel, W. Convergence in the QM-only and QM/MM modeling of enzymatic reactions: A case study for acetylene hydratase. J. Comput. Chem. 2013, 34, 2389–2397. [CrossRef] 30. Volbeda, A.; Martin, L.; Cavazza, C.; Matho, M.; Faber, B.; Roseboom, W.; Albracht, S.; Garcin, E.; Rousset, M.; Fontecilla-Camps, J.C. Structural diﬀerences between the ready and unready oxidized states of [NiFe] hydrogenases. J. Biol. Inorg. Chem. 2005, 10, 239–249. [CrossRef] 31. Siegbahn, P.E.M.; Tye, J.W.; Hall, M.B. Computational studies of [NiFe] and [FeFe] hydrogenases. Chem. Rev. 2007, 107, 4414–4435. [CrossRef] 32. Bruschi, M.; Zampella, G.; Greco, Y.; Bertini, L.; Fantucci, P.; Gioia, D.L. Computational Inorganic and Bioinorganic Chemistry; Solomon, E.I., King, B., Scott, R., Eds.; John and Wiley and Sons: Chichester, UK, 2009. 33. Siegbahn, P.E.M.; Blomberg, M.R.A. Quantum Chemical Studies of Proton-Coupled Electron Transfer in Metalloenzymes. Chem. Rev. 2010, 110, 7040–7061. [CrossRef] Inorganics 2019, 7, 95 25 of 29

34. Kurkin, S.; George, S.J.; Thorneley, R.N.F.; Albracht, S.P.J. Hydrogen-induced activation of the [NiFe]-hydrogenase from Allochromatium vinosum as studied by stopped-ﬂow infrared spectroscopy. Biochemistry 2004, 43, 6820–6831. [CrossRef] 35. Nilsson-Lill, S.O.; Siegbahn, P.E.M. An autocatalytic Mechanism for NiFe-Hydrogenase: Reduction to Ni(I) Followed by Oxidative Addition. Biochemistry 2009, 48, 1056–1066. [CrossRef] 36. Ogata, H.; Hirota, S.; Nakahara, A.; Komori, H.; Shibata, N.; Kato, T.; Kano, K.; Higuchi, Y. Activation process of [NiFe] hydrogenase elucidated by high-resolution X-ray analyses: Conversion of the ready to the unready state. Structure 2005, 13, 1635–1642. [CrossRef] 37. Siegbahn, P.E.M. Hybrid density functional study of the oxidized states of NiFe-hydrogenase. C. R. Chim. 2007, 10, 766–774. [CrossRef] 38. Pardo, A.; Lacey, A.L.D.; Fernandez, V.M.; Fan, Y.B.; Hall, M.B. Characterization of the active site of catalytically inactive forms of [NiFe] hydrogenases by density functional theory. J. Biol. Inorg. Chem. 2007, 12, 751–760. [CrossRef] 39. Söderhjelm, P.; Ryde, U. Combined computational and crystallographic study of the oxidised states of [NiFe] hydrogenase. Theochem 2006, 770, 108–116. [CrossRef] 40. Volbeda, A.; Martin, L.; Barbier, E.; Gutiérrez-Sanz, O.; De Lacey, A.L.; Liebgott, P.P.; Dementin, S.; Rousset, M.; Fontecilla-Camps, J.C. Crystallographic studies of [NiFe]-hydrogenase mutants: Towards consensus structures for the elusive unready oxidized states. J. Biol. Inorg. Chem. 2015, 20, 11–22. [CrossRef] 41. Breglia, R.; Greco, C.; Fantucci, P.; De Gioia, L.; Bruschi, M. Theoretical investigation of aerobic and anaerobic oxidative inactivation of the [NiFe]-hydrogenase active site. Phys. Chem. Chem. Phys. 2018, 20, 1693–1706. [CrossRef] 42. Breglia, R.; Greco, C.; Fantucci, P.; De Gioia, L. Reactivation of the Ready and Unready Oxidized States of [NiFe]-Hydrogenases: Mechanistic Insights from DFT Calculations. Inorg. Chem. 2019, 58, 279–293. [CrossRef] 43. Goris, T.; Wait, A.F.; Saggu, M.; Fritsch, J.; Heidary, N.; Stein, M.; Zebger, I.; Lendzian, F.; Armstrong, F.A.; Friedrich, B.; et al. A unique iron–sulfur cluster is crucial for oxygen tolerance of a [NiFe]-hydrogenase. Nat. Chem. Biol. 2011, 7, 310–318. [CrossRef] 44. Mouesca, J.M.; Fontecilla-Camps, J.C.; Amara, P. The Structural Plasticity of the Proximal [4Fe3S] Cluster is

Responsible for the O2 Tolerance of Membrane-Bound [NiFe] Hydrogenases. Angew. Chem. Int. Ed. 2013, 52, 2002–2006. [CrossRef] 45. Pandelia, M.E.; Bykov, D.; Izsak, R.; Infossi, P.; Giudici-Orticoni, M.T.; Bill, E.; Neese, F.; Lubitz, W. Electronic 57 structure of the unique [4Fe–3S] clusterin O2-tolerant hydrogenases characterized by Fe Mössbauer and EPR spectroscopy. Proc. Natl Acad. Sci. USA 2013, 110, 483–488. [CrossRef] 46. Pelmenschikov, V.; Kaupp, M. Redox-Dependent Structural Transformations of the [4Fe–3S] Proximal

Cluster in O2-Tolerant Membrane-Bound [NiFe]-Hydrogenase: A DFT Study. J. Am. Chem. Soc. 2013, 135, 11809–11823. [CrossRef] 47. Delcey, M.G.; Pierloot, K.; Quan, M.P.; Vancoille, S.; Lind, R.; Ryde, U. Accurate calculations of geometries and singlet–triplet energy diﬀerences for active-site models of [NiFe] hydrogenase. Phys. Chem. Chem. Phys. 2014, 16, 7927–7938. [CrossRef]

48. Dong, G.; Quan, M.P.; Hallaert, S.D.; Pierloot, K.; Ryde, U. H2 binding to the active site of [NiFe] hydrogenase studied by multiconﬁgurational and coupled-cluster methods. Phys. Chem. Chem. Phys. 2017, 19, 10590–10601. [CrossRef] 49. Dong, G.; Ryde, U. Protonation states of intermediates in the reaction mechanism of [NiFe] hydrogenase studied by computational methods. J. Biol. Inorg. Chem. 2016, 21, 383–394. [CrossRef]

50. Dong, G.; Ryde, U.; Jensen, H.J.; Hedegård, E.D. Exploration of H2 binding to the [NiFe]-hydrogenase active site with multiconﬁgurational density functional methods. Phys. Chem. Chem. Phys. 2018, 20, 794–801. [CrossRef] 51. Dong, G.; Quan, M.P.; Pierloot, K.; Ryde, U. Reaction Mechanism of [NiFe] Hydrogenase Studied by Computational Methods. Inorg. Chem. 2018, 57, 15289–15298. [CrossRef] 52. Murphy, B.J.; Hidalgo, R.; Roessler, M.M.; Evans, R.M.; Ash, P.A.; Myers, W.K.; Vincent, K.A.; Armstrong, F.A. Discovery of Dark pH-Dependent H+ Migration in a [NiFe]-Hydrogenase and Its Mechanistic Relevance: Mobilizing the Hydrido Ligand of the Ni–C Intermediate. J. Am. Chem. Soc. 2015, 137, 8484–8489. [CrossRef] 53. Murrell, J.C.; Jetten, M.S.M. The microbial methane cycle. Environ. Microbiol. Rep. 2009, 1, 279–284. [CrossRef] Inorganics 2019, 7, 95 26 of 29

54. Hogan, K.B.; Hoffman, J.S.; Thompson, A.M. Methane on the greenhouse agenda. Nature 1991, 354, 181–182. [CrossRef] 55. Conrad, R. The global methane cycle: Recent advances in understanding the microbial processes involved. Environ. Microbiol. Rep. 2009, 1, 285–292. [CrossRef] 56. Evans, P.N.; Boyd, J.A.; Leu, A.O.; Woodcroft, B.J.; Parks, D.H.; Hugenholtz, P.; Tyson, G.W. Methane on the greenhouse agenda. Nat. Rev. Microbiol. 2019, 17, 219–232. [CrossRef] 57. Thauer, R.K.; Kaster, A.K.; Seedorf, H.; Buckel, W.; Hedderich, R. Methanogenic archaea: Ecologically relevant differences in energy conservation. Nat. Rev. Microbiol. 2008, 6, 579–591. [CrossRef] 58. Nauhaus, K.; Albrecht, M.; Elvert, M.; Boetius, A.; Widdel, F. In vitro cell growth of marine archaeal-bacterial consortia during anaerobic oxidation of methane with sulfate. Environ. Microbiol. 2007, 9, 187–196. [CrossRef] 59. Nauhaus, K.; Boetius, A.; Krüger, M.; Widdel, F. In vitro demonstration of anaerobic oxidation of methane coupled to sulphate reduction in sediment from a marine gas hydrate area. Environ. Microbiol. 2002, 4, 296–305. [CrossRef] 60. Ettwig, K.F.; Zhu, B.; Speth, D.; Keltjens, J.T.; Jetten, M.S.M.; Kartal, B. Archaea catalyze iron-dependent anaerobic oxidation of methane. Proc. Natl Acad. Sci. USA 2016, 113, 12792–12796. [CrossRef] 61. Beal, E.J.; House, C.H.; Orphan, V.J. Manganese-and iron-dependent marine methane oxidation. Science 2009, 325, 184–187. [CrossRef] 62. Haroon, M.F.; Hu, S.; Shi, Y.; Imelfort, M.; Keller, J.; Hugenholtz, P.; Yuan, Z.; Tyson, G.W. Anaerobic oxidation of methane coupled to nitrate reduction in a novel archaeal lineage. Nature 2013, 500, 567–570. [CrossRef] 63. Scheller, S.; Yu, H.; Chadwick, G.L.; McGlynn, S.E.; Orphan, V.J. Artificial electron acceptors decouple archaeal methane oxidation from sulfate reduction. Science 2016, 351, 703–707. [CrossRef] 64. Rotaru, A.E.; Thamdrup, B. A new diet for methane oxidizers. Science 2016, 351, 658–659. [CrossRef] 65. Shen, B.; Dong, L.; Xiao, S.H.; Lang, X.G.; Huang, K.J.; Peng, Y.B.; Zhou, C.M.; Ke, S.; Liu, P.J. Molar tooth carbonates and benthic methane fluxes in Proterozoic oceans. Nat. Commun. 2016, 7, 10317. [CrossRef] 66. Planavsky, N.J.; Reinhard, C.T.; Wang, X.; Thomson, D.; McGoldrick, P.; Rainbird, R.H.; Johnson, T.; Fischer, W.W.; Lyons, T.W. Low Mid-Proterozoic atmospheric oxygen levels and the delayed rise of animals. Science 2014, 346, 635–638. [CrossRef] 67. Thauer, R.K. The Wolfe cycle comes full circle. Proc. Natl Acad. Sci. USA 2012, 109, 15084–15085. [CrossRef] 68. Chen, S.C.; Musat, N.; Lechtenfeld, O.J.; Paschke, H.; Schmidt, M.; Said, N.; Popp, D.; Calabrese, F.; Stryhanyuk, H.; Jaekel, U.; et al. Anaerobic oxidation of ethane by archaea from a marine hydrocarbon seep. Nature 2019, 568, 108–111. [CrossRef] 69. Pérez, R.L.; Wegener, G.; Knittel, K.; Widdel, F.; Harding, K.J.; Krukenberg, V.; Meier, D.V.; Richter, M.; Tegetmeyer, H.E.; Riedel, D.; et al. Thermophilic archaea activate butane via alkyl-coenzyme M formation. Nature 2016, 539, 396–401. [CrossRef] 70. Yang, N.; Reiher, M.; Wang, M.; Harmer, J.; Duin, E.C. Formation of a nickel-methyl species in methyl-coenzyme M reductase, an enzyme catalyzing methane formation. J. Am. Chem. Soc. 2007, 129, 11028–11029. [CrossRef] 71. Dey, M.; Telser, J.; Kunz, R.C.; Lees, N.S.; Ragsdale, S.W.; Hoffman, B.M. Biochemical and spectroscopic studies of the electronic structure and reactivity of a methyl-Ni species formed on methyl-coenzyme m reductase. J. Am. Chem. Soc. 2007, 129, 11030–11032. [CrossRef] 72. Pelmenschikov, V.; Blomberg, M.R.A.; Siegbahn, P.E.M.; Crabtree, R.H. A mechanism from quantum chemical studies for methane formation in methanogenesis. J. Am. Chem. Soc. 2002, 124, 4039–4049. [CrossRef] 73. Chen, S.L.; Pelmenschikov, V.; Blomberg, M.R.A.; Siegbahn, P.E.M. Is There a Ni-Methyl Intermediate in the Mechanism of Methyl-Coenzyme M Reductase? J. Am. Chem. Soc. 2009, 131, 9912–9913. [CrossRef] 74. Chen, S.L.; Blomberg, M.R.A.; Siegbahn, P.E.M. How is methane formed and oxidized reversibly when catalyzed by Ni-containing methyl-coenzyme M reductase? Chem. Eur. J. 2012, 18, 6309–6315. [CrossRef] 75. Chen, S.L.; Blomberg, M.R.A.; Siegbahn, P.E.M. An investigation of possible competing mechanisms for Ni-containing methyl-coenzyme M reductase. Phys. Chem. Chem. Phys. 2014, 16, 14029–14035. [CrossRef] 76. Pelmenschikov, V.; Siegbahn, P.E.M. Catalysis by methyl-coenzyme M reductase: A theoretical study for heterodisulfide product formation. J. Biol. Inorg. Chem. 2003, 8, 653. [CrossRef] 77. Scheller, S.; Goenrich, M.; Thauer, R.K.; Jaun, B. Methyl-coenzyme M reductase from methanogenic archaea: Isotope effects on label exchange and ethane formation with the homologous substrate ethyl-coenzyme M. J. Am. Chem. Soc. 2013, 135, 14975–14984. [CrossRef] Inorganics 2019, 7, 95 27 of 29

78. Ahn, Y.; Krzycki, J.A.; Floss, H.G. Steric course of the reduction of ethyl coenzyme M to ethane catalyzed by methyl coenzyme m reductase from methanosarcina barkeri. J. Am. Chem. Soc. 1991, 113, 4700–4701. [CrossRef] 79. Scheller, S.; Goenrich, M.; Boecher, R.; Thauer, R.K.; Jaun, B. The key nickel enzyme of methanogenesis catalyses the anaerobic oxidation of methane. Nature 2010, 465, 606–609. [CrossRef] 80. Scheller, S.; Goenrich, M.; Mayr, S.; Thauer, R.K.; 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. [CrossRef] 81. Wongnate, T.; Sliwa, D.; Ginovska, B.; Smith, D.; Wolf, M.W.; Lehnert, N.; Raugei, S.; Ragsdale, S.W. The radical mechanism of biological methane synthesis by methyl-coenzyme M reductase. Science 2016, 352, 953–958. [CrossRef] 82. Lawton, T.J.; Rosenzweig, A.C. Methane-make it or break it. Science 2016, 352, 892–893. [CrossRef] 83. Duin, E.C.; Mckee, M.L. A new mechanism for methane production from methyl-coenzyme M reductase as derived from density functional calculations. J. Phys. Chem. B 2008, 112, 2466–2482. [CrossRef] 84. Harmer, J.; Finazzo, C.; Piskorski, R.; Ebner, S.; Duin, E.C.; Goenrich, M.; Thauer, R.K.; Reiher, M.; Schweiger, A.; Hinderberger, D.; et al. A nickel hydride complex in the active site of methyl-coenzyme m reductase: Implications for the catalytic cycle. J. Am. Chem. Soc. 2008, 130, 10907–10920. [CrossRef] 85. Grabarse, W.; Mahlert, F.; Duin, E.C.; Goubeaud, M.; Shima, S.; Thauer, R.K.; Lamzin, V.; Ermler, U. On the mechanism of biological methane formation: Structural evidence for conformational changes in methyl-coenzyme M reductase upon substrate binding. J. Mol. Biol. 2001, 309, 315–330. [CrossRef] 86. Horng, Y.C.; Becker, D.F.; Ragsdale, S.W. Mechanistic studies of methane biogenesis by methyl-coenzyme M reductase: Evidence that coenzyme B participates in cleaving the C–S bond of methyl-coenzyme M. Biochemistry 2001, 40, 12875–12885. [CrossRef] 87. Borman, S. Carbon dioxide hydrogenated to methanol on large scale. Chem. Eng. News 2016, 94, 7. 88. Shima, S. The Biological Methane-Forming Reaction: Mechanism Conﬁrmed Through Spectroscopic Characterization of a Key Intermediate. Angew. Chem. Int. Ed. 2016, 55, 13648–13649. [CrossRef] 89. Dobbek, H.; Svetlitchnyi, V.; Gremer, L.; Huber, R.; Meyer, O. Crystal Structure of a Carbon Monoxide Dehydrogenase Reveals a [Ni–4Fe–5S] Cluster. Science 2001, 293, 1281–1285. [CrossRef] 90. Darnault, C.; Volbeda, A.; Kim, E.J.; Legrand, P.; Vernede, X.; Lindahl, P.A.; Fontecilla-Camps, J.C.

Ni–Zn–[Fe4–S4] and Ni–Ni–[Fe4–S4] clusters in closed and open α subunits of acetyl-CoA synthase/carbon monoxide dehydrogenase. Nat. Struct. Biol. 2003, 10, 271–279. [CrossRef] 91. Appel, A.M.; Bercaw, J.E.; Bocarsly, A.B.; Dobbek, H.; DuBois, D.L.; Dupuis, M.; Ferry, J.G.; Fujita, E.; Hille, R.;

Kenis, P.J.A.; et al. Frontiers, Opportunities, and Challenges in Biochemical and Chemical Catalysis of CO2 Fixation. Chem. Rev. 2013, 113, 6621–6658. [CrossRef] 92. Can, M.; Armstrong, F.A.; Ragsdale, S.W. Structure, Function, and Mechanism of the Nickel Metalloenzymes, CO Dehydrogenase, and Acetyl-CoA Synthase. Chem. Rev. 2014, 114, 4149–4174. [CrossRef] 93. Lindahl, P.A.; Munck, E.; Ragsdale, S.W. CO Dehydrogenase from Clostridium thermoaceticum: EPR and

electrochemical studies in CO2 and argon atmospheres. J. Biol. Chem. 1990, 265, 3873–3879. 94. Cao, Z.-X.; Mo, Y.-R. Computational Characterization of the Elusive C-Cluster of Carbon Monoxide Dehydrogenase. J. Theor. Comput. Chem. 2008, 7, 473–484. [CrossRef] 95. Xie, H.-J.; Cao, Z.-X. Computational Characterization of Reactive Intermediates of Carbon Monoxide Dehydrogenase. Chin. J. Struct. Chem. 2009, 28, 1525–1532. 96. Amara, P.; Mouesca, J.-M.; Volbeda, A.; Fontecilla-Camps, J.C. Carbon Monoxide Dehydrogenase Reaction

Mechanism: A Likely Case of Abnormal CO2 Insertion to a Ni–H− Bond. Inorg. Chem. 2011, 50, 1868–1878. [CrossRef] 97. Liao, R.-Z.; Siegbahn, P.E.M. Energetics for the mechanism of nickel-containing CO-dehydrogenase. Inorg. Chem. 2019.[CrossRef] 98. Amara, P.; Volbeda, A.; Fontecilla-Camps, J.C.; Field, M.J. A quntum chemical study of the reaction mechanism of Acetyl-Coenzyme A synthase. J. Am. Chem. Soc. 2005, 127, 2776–2784. [CrossRef] 99. Chmielowska, A.; Lodowski, P.; Jaworska, M. Redox Potentials and Protonation of the A-Cluster from Acetyl-CoA Synthase. A Density Functional Theory Study. J. Phys. Chem. 2013, 117, 12484–12496. [CrossRef] Inorganics 2019, 7, 95 28 of 29

100. Schenker, R.P.; Brunold, T.C. Computational studies on the A cluster of acetyl-coenzyme A synthase: Geometric and electronic properties of the NiFeC species and mechanistic implications. J. Am. Chem. Soc. 2003, 125, 13962–13963. [CrossRef] 101. Dai, Y.; Wensink, P.C.; Abeles, R.H.J. One protein, two enzymes. Biol. Chem. 1999, 274, 1193–1195. [CrossRef] 102. Wray, J.W.; Abeles, R.H. The Methionine Salvage Pathway in Klebsiella pneumoniae and Rat Liver identification and characterization of two novel dioxygenases. J. Biol. Chem. 1995, 270, 3147–3153. [CrossRef] 103. Dai, Y.; Pochapsky, T.C.; Abeles, R.H. Mechanistic studies of two dioxygenases in the methionine salvage pathway of Klebsiella pneumonia. Biochemistry 2001, 40, 6379–6387. [CrossRef] 104. Sparta, M.; Valdez, C.E.; Alexandrova, A.N. Metal-dependent activity of Fe and Ni acireductone dioxygenases: How two electrons reroute the catalytic pathway. J. Mol. Biol. 2013, 425, 3007–3018. [CrossRef] 105. Miłaczewska, A.; Kot, E.; Amaya, J.A.; Makris, T.M.; Zaj ˛ac,M.; Korecki, J.; Chumakov, A.; Trzewik, B.; K˛edracka-Krok, S.; Minor, W.; et al. On the Structure and Reaction Mechanism of Human Acireductone Dioxygenase. Chem. Eur. J. 2018, 24, 5225–5237. [CrossRef] 106. Siegbahn, P.E.M. Hybrid DFT study of the mechanism of quercetin 2,3-dioxygenase. Inorg. Chem. 2004, 43, 5944–5953. [CrossRef] 107. Wang, W.J.; Wei, W.J.; Liao, R.Z. Deciphering the chemoselectivity of nickel-dependent quercetin 2, 4-dioxygenase. Phys. Chem. Chem. Phys. 2018, 20, 15784–15794. [CrossRef] 108. Merkens, H.; Kappl, R.; Jakob, R.P.; Schmid, F.X.; Fetzner, S. Quercetinase QueD of Streptomyces sp. FLA, a Monocupin Dioxygenase with a Preference for Nickel and Cobalt. Biochemistry 2008, 47, 12185–12196. [CrossRef] 109. Li, H.; Wang, X.; Tian, G.; Liu, Y. Insights into the dioxygen activation and catalytic mechanism of the nickel-containing quercetinase. Catal. Sci. Technol. 2018, 8, 2340–2351. [CrossRef] 110. Karplus, P.A.; Pearson, M.A.; Hausinger, R.P. 70 years of crystalline urease: What have we learned? Acc. Chem. Res. 1997, 30, 330–337. [CrossRef] 111. Estiu, G.; Merz, K.M. The hydrolysis of urea and the proficiency of urease. J. Am. Chem. Soc. 2004, 126, 6932–6944. [CrossRef] 112. Pearson, M.A.; Park, I.S.; Schaller, R.A.; Michel, L.O.; Karplus, P.A.; Hausinger, R.P. Kinetic and Structural Characterization of Urease Active Site Variants. Biochemistry 2000, 39, 8575–8584. [CrossRef] 113. Benini, S.; Rypniewski, W.R.; Wilson, K.S.; Miletti, S.; Ciurli, S.; Mangani, S. A new proposal for urease mechanism based on the crystal structures of the native and inhibited enzyme from Bacillus pasteurii: Why urea hydrolysis costs two nickels. Structure 1999, 7, 205–216. [CrossRef] 114. Barrios, A.M.; Lippard, S.J. Interaction of urea with a hydroxide-bridged dinuclear nickel center: An alternative model for the mechanism of urease. J. Am. Chem. Soc. 2000, 122, 9172–9177. [CrossRef] 115. Blakeley, R.L.; Hinds, J.A.; Kunze, H.E.; Webb, E.C.; Zerner, B. Jack bean urease (EC 3.5.1.5) demonstration of a carbamoyl-transfer reaction and inhibition by hydroxamic acids. Biochemistry 1969, 8, 1991–2000. [CrossRef] 116. Smyj, R.P. A conformational analysis study of a nickel(II) enzyme: Urease. J. Mol. Struct. THEOCHEM 1997, 391, 207–223. [CrossRef] 117. Zimmer, M. Molecular mechanics evaluation of the proposed mechanisms for the degradation of urea by urease. J. Biomol. Struct. Dyn. 2000, 17, 787–797. [CrossRef] 118. Zimmer, M. Are classical molecular mechanics calculations still useful in bioinorganic simulations? Coord. Chem. Rev. 2009, 253, 817–826. [CrossRef] 119. Musiani, F.; Arnofi, E.; Casadio, R.; Ciurli, S. Structure-based computational study of the catalytic and inhibition mechanisms of urease. J. Biol. Inorg. Chem. 2001, 6, 300–314. [CrossRef] 120. Suárez, D.; Díaz, N.; Merz, K.M. Ureases: Quantum chemical calculations on cluster models. J. Am. Chem. Soc. 2003, 125, 15324–15337. [CrossRef] 121. Estiu, G.; Merz, K.M. Catalyzed decomposition of urea. Molecular dynamics simulations of the binding of urea to urease. Biochemistry 2006, 45, 4429–4443. [CrossRef] 122. Estiu, G.; Merz, K.M. Competitive hydrolytic and elimination mechanisms in the urease catalyzed decomposition of urea. J. Phys. Chem. B 2007, 111, 10263–10274. [CrossRef] 123. Carlsson, H.; Nordlander, E. Computational Modeling of the Mechanism of Urease. Bioinorg. Chem. Appl. 2010, 2010, 8. [CrossRef] Inorganics 2019, 7, 95 29 of 29

124. Mazzei, L.; Cianci, M.; Benini, S.; Ciurli, S. The impact of pH on catalytically critical protein conformational changes; the case of the urease, a nickel enzyme. Chem. Eur. J. 2019.[CrossRef] 125. Mazzei, L.; Cianci, M.; Benini, S.; Ciurli, S. The structure of the elusive urease-urea complex unveils the mechanism of a paradigmatic nickel-dependent enzyme. Angew. Chem. Int. Ed. 2019, 131, 7493–7497. [CrossRef] 126. Desguin, B.; Goffin, P.; Viaene, E.; Kleerebezem, M.; Diaconescu, V.M.; Maroney, M.J.; Declercq, J.P.; Soumillion, P.; Hols, P. Lactate racemase is a nickel-dependent enzyme activated by a widespread maturation system. Nat. Commun. 2014, 5, 3615. [CrossRef] 127. Xu, T.; Bauer, G.; Hu, X. A novel nickel pincer complex in the active site of lactate racemase. ChemBioChem 2016, 17, 31–32. [CrossRef] 128. Zamble, D. It costs more than a nickel. Science 2015, 349, 35–36. [CrossRef] 129. Yu, M.J.; Chen, S.L. From NAD+ to Nickel Pincer Complex: A Significant Cofactor Evolution Presented by Lactate Racemase. Chem. Eur. J. 2017, 23, 7545–7557. [CrossRef] 130. Zhang, X.; Chung, L.W. Alternative Mechanistic Strategy for Enzyme Catalysis in a Ni-Dependent Lactate Racemase (LarA): Intermediate Destabilization by the Cofactor. Chem. Eur. J. 2017, 23, 3623–3630. [CrossRef] 131. Qiu, B.; Yang, X.Z. A bio-inspired design and computational prediction of scorpion-like SCS nickel pincer complexes for lactate racemization. Chem. Commun. 2017, 53, 11410–11413. [CrossRef] 132. Wang, B.; Shaik, S. The Nickel-Pincer Complex in Lactate Racemase Is an Electron Relay and Sink that acts through Proton-Coupled Electron Transfer. Angew. Chem. Int. Ed. 2017, 56, 10098–10102. [CrossRef] 133. Zhang, T.H.; Zhang, X.Y.; Chung, L.W. Computational Insights into the Reaction Mechanisms of Nickel-Catalyzed Hydrofunctionalizations and Nickel-Dependent Enzymes. Asian J. Organ. Chem. 2018, 7, 522–536. [CrossRef] 134. Pelmenschikov, V.; Siegbahn, P.E.M. Nickel superoxide dismutase reaction mechanism studied by hybrid density functional methods. J. Am. Chem. Soc. 2006, 128, 7466–7475. [CrossRef] 135. Prabhakar, R.; Morokuma, K.; Musaev, D.G. A DFT study of the mechanism of Ni superoxide dismutase (NiSOD): Role of the active site cysteine-6 residue in the oxidative half-reaction. J. Comput. Chem. 2006, 27, 1438–1445. [CrossRef] 136. Barondeau, D.P.; Kassmann, C.J.; Bruns, C.K.; Tainer, J.A.; Getzoff, E.D. Nickel superoxide dismutase structure and mechanism. Biochemistry 2004, 43, 8038–8047. [CrossRef] 137. Wang, Q.; Chen, D.; Liu, X.; Zhang, L.F. Theoretical mechanisms of the superoxide radical anion catalyzed by the nickel superoxide dismutase. Comput. Theor. Chem. 2011, 966, 357–363. [CrossRef] 138. Shearer, J.; Schmitt, J.C.; Clewett, H.S. Adiabaticity of the Proton-Coupled Electron-Transfer Step in the Reduction of Superoxide Effected by Nickel-Containing Superoxide Dismutase Metallopeptide-Based Mimics. J. Phys. Chem. B, 2015, 119, 5453–5461. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).