This paper is published as part of a Dalton Transactions themed issue on:

Bioorganometallic Chemistry

Guest Editor Charles Riordan University of Delaware, USA

Published in issue 22, 2009 of Dalton Transactions

Images reproduced with permission of John Peters (left) and Toshikazu Hirao (right) Papers published in this issue include: Perspective: Hydrogenase cluster biosynthesis: organometallic chemistry nature's wayShawn E. McGlynn, David W. Mulder, Eric M. Shepard, Joan B. Broderick and John W. Peters Dalton Trans., 2009, 4274, DOI: 10.1039/b821432h Role of aromatic substituents on the antiproliferative effects of diphenyl ferrocenyl butene compounds Ouardia Zekri, Elizabeth A. Hillard, Siden Top, Anne Vessières, Pascal Pigeon, Marie-Aude Plamont, Michel Huché, Sultana Boutamine, Michael J. McGlinchey, Helge Müller-Bunz and Gérard Jaouen, Dalton Trans., 2009, 4318, DOI: 10.1039/b819812h Resin-bound models of the [FeFe]-hydrogenase enzyme active site and studies of their reactivity Kayla N. Green, Jennifer L. Hess, Christine M. Thomas and Marcetta Y. Darensbourg Dalton Trans., 2009, 4344, DOI: 10.1039/b823152d Molecular structures of protonated and mercurated derivatives of thimerosalWesley Sattler, Kevin Yurkerwich and Gerard Parkin Dalton Trans., 2009, 4327, DOI: 10.1039/b823467a Critical aspects of [NiFe]hydrogenase ligand compositionKoji Ichikawa, Takahiro Matsumoto and Seiji Ogo Dalton Trans., 2009, 4304, DOI: 10.1039/b819395a

Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research

www.rsc.org/dalton

and for more bioinorganic and biological inorganic chemistry papers in Dalton Transactions: www.rsc.org/Publishing/Journals/dt/News/BIB.asp

1477-9226(2009)22;1-2

DaltonTransactionsAn international journal of inorganic chemistry

PERSPECTIVEPeters et al.Hydrogenase cluster biosynthesis: organometallic chemistry nature’s way

HOT ARTICLESchatzschneider et al.Sonogashira and “Click” reactions for the N-terminal and side-chain functionalization of peptides with CO releasing molecules

www.rsc.org/dalton Number 22 | 14 June 2009 | Pages 4261–4384

ISSN 1477-9226

Themed issue: Bioorganometallic chemistry

PERSPECTIVE www.rsc.org/dalton | Dalton Transactions

Hydrogenase cluster biosynthesis: organometallic chemistry nature’s way

Shawn E. McGlynn, David W. Mulder, Eric M. Shepard, Joan B. Broderick and John W. Peters*

Received 1st December 2008, Accepted 11th March 2009First published as an Advance Article on the web 30th March 2009DOI: 10.1039/b821432h

It has been over a decade now since it was revealed that the metal containing active sites ofhydrogenases possess carbonyl and cyanide ligands bound to iron. The presence of these ligands inhydrogenases came as a surprise and to-date these ligands have not been observed to be associated withany other enzymatic metallocenter. The elucidation of the structures of these unique metalloenzymesand their associated metal clusters created opportunity for a number of different lines of research. Forsynthetic chemists, the structures of hydrogenase active sites have provided attractive targets forsyntheses that advance our understanding of the electronic structure and reactivity of these uniqueenzyme active sites. These efforts contribute to the synthesis of first row transition metal catalysts forhydrogen oxidation and hydrogen production that could have significant impacts on alternative andrenewable energy solutions. Although effective synthetic approaches have been identified to generatemodels with a high degree of similarity to these active sites, the details of how these metal clusters aresynthesized biochemically have not been resolved. Since hydrogen metabolism is presumed to be anearly feature in the energetics of life and hydrogen metabolizing organisms can be traced very early inmolecular phylogeny, the metal clusters at hydrogenase active sites are presumed to be among theearliest of known co-factors. Comparison of mineral based precursors and synthetic cluster analogchemistry to what is observed in contemporary biological systems may shed light on howproto-metabolically relevant catalysts first arose prebiotically by the processes of adoption ofpre-existing functionality and ligand assisted catalysis.

Introduction

Hydrogenases occur in three forms [FeFe]-, [NiFe]-, and [Fe]-hydrogenase so named to reflect their respective metal clustercontent. The [FeFe]- and [NiFe]-hydrogenases formally catalyzereversible interconversion between dihydrogen gas and protonsand electrons where the [Fe]-hydrogenases which occur exclusively

Department of Chemistry and Biochemistry and the Astrobiology Biogeo-catalysis Research Center, Montana State University, Bozeman, Montana59715, USA. E-mail: john.peters@chemistry.montana.edu; Fax: +1 406-994-7470

Shawn E. McGlynn

Shawn McGlynn was born in1983. He received a B.S. degreein chemistry from Montana StateUniversity in 2005 and is cur-rently a Ph.D. candidate in Prof.John Peters research group atMontana State University and anIGERT fellow in GeobiologicalSystems. His research interestsinclude metal usage in biology,hydrogenase active site assembly,and the origins of life.

David W. Mulder

David Mulder was born in 1983.He studied chemistry at CalvinCollege, receiving his B.S. de-gree in chemistry in 2005 andis currently a Ph.D. candidatein Prof. John Peters researchgroup at Montana State Univer-sity where he received an INBREresearch fellowship. His scientificinterests are in iron–sulfur en-zyme biochemical and spectro-scopic characterization and hy-drogenase biosynthesis.

in the methanogenic Archaea are given the systematic name,H2-forming methylene-H4MPT dehydrogenase (Hmd), and cat-alyze the dehydrogenation of methylene-tetrahydromethanopterin(methlylene-H4MPT).

Hydrogenases are widely distributed in microorganisms.1,2 Hy-drogen oxidation is used to couple the generation of reducingequivalents to energy yielding reactions to support microbialgrowth, while hydrogen production is utilized as a mechanismto recycle reducing equivalents that accumulate during fermen-tative metabolism. In nature, hydrogen oxidizing and hydrogenproducing microorganisms frequently occupy the same or adjacent

4274 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

environmental niches and interspecies hydrogen transfer is ofparamount importance to the energetics and equilibrium processesof microbial systems. So large is the role of hydrogen in microbialecosystems that interspecies hydrogen transfer and the associatedinterdependence of hydrogen oxidizing and hydrogen producingorganisms has been suggested in two independent hypotheses ashaving been the driving force of the evolution of the first eukaryoticcell.3,4

The occurrence of hydrogenases in microbial hosts is aninteresting mystery of molecular evolution since to date [FeFe]-hydrogenases have been found to be associated with bacteria andlower eukaryotes such as algae and protists, but are not foundto be associated with archaea or cyanobacteria. In contrast, the[NiFe]-hydrogenases, which are found to be associated with a widevariety of bacteria, are also commonly found in many archaeaand cyanobacteria but are not found to be associated with anylower eukaryotes. Finally, the occurrence of the [Fe]-hydrogenasesis observed exclusively in some hydrogenotrophic methanogenicarchaea. These discreet differences in the occurrence of thedifferent classes of hydrogenases is surprising given the commonfeatures of their enzyme active site ligand architectures and

Eric M. Shepard

Eric Shepard was born in 1977.He received a B.S. degree inchemistry from Rocky MountainCollege. He studied copper andTPQ-containing amine oxidasesunder Dr David M. Dooley atMontana State University, wherehe was supported by a NSFIGERT fellowship on complexbiological systems. He receivedhis Ph.D. degree in biochemistryfrom MSU and is currently apostdoctoral research associateunder Dr. Joan B. Broderick. His

research interests are metal cluster assembly in the [FeFe] hydroge-nase system and [Fe–S] cluster spectroscopy and reactivity.

indicate that these enzymes have evolved to common ligandarchitectures and reactivity by convergent evolution.

Interest in hydrogen metabolism and hydrogenases has grownexponentially over the past decade in part due to the growinginterest in alternative and renewable energy solutions. Hydrogenproduction by microorganisms or the use of hydrogenase enzymesor mimetic compounds as hydrogen oxidation or hydrogenproduction catalysts has tremendous promise; however thereare significant barriers to surmount. Overcoming these barrierswill require an intimate understanding of the biosynthesis andmaturation of hydrogenase enzymes as well as their structure andreactivity.

As thoughts concerning hydrogenase structure, function, andreactivity continue to evolve, the relationships between the featuresof hydrogenase active sites and those of other complex iron-sulfurenzymes and iron-sulfur mineral structures and reactivity havealso been gaining interest. Given that complex iron-sulfur enzymesare involved with reactions such as reversible hydrogen oxidation,nitrogen fixation, and reversible carbon monoxide oxidation—reactions that would have been important in the prebiotic earthand necessary for the first stages of the emergence of life—it isrational to think about these enzymes as having an ancient origin.It is possible that the unique metal-containing prosthetic groups atthe active sites of these enzymes may represent the oldest of knownco-factors with what is observed today existing as highly evolvedversions of modified iron-sulfur mineral catalysts. Investigationsof hydrogenase structure, function, and biosynthesis may in factprovide some keys as to the origin of life and perhaps evencontribute to the development of signatures for life beyond Earth.Given the potential importance of hydrogenases at the origin oflife and the potential for hydrogenase based catalysis for renewableenergy and sustaining life on Earth, hydrogenase research is anideal combination of basic and applied research.

Active site metal cluster structure

In 1995, the first structure of a [NiFe]-hydrogenase was determinedby X-ray diffraction methods from the sulfate reducing bacteria(SRB) Desulfovibrio gigas5 and with the structure came some

Joan B. Broderick

Joan Broderick was born in 1965.She received a B.S. from Wash-ington State University and aPh.D. from Northwestern Univer-sity. She was an American Can-cer Society postdoctoral fellow atMIT before joining the facultyat Amherst College as AssistantProfessor in 1993. She movedto Michigan State University in1998 and to Montana State Uni-versity in 2005, where she iscurrently Professor of Chemistryand Biochemistry. Her research

interests are in mechanistic bioinorganic chemistry, with a particularfocus on enzymes utilizing iron–sulfur clusters to catalyze radicalreactions.

John W. Peters

John Peters was born in 1965. Hereceived a B.S. in Microbiologyfrom the University of Oklahomaand a Ph.D. from Virginia Tech.He was a NIH postdoctoral fel-low at Cal Tech before joiningUtah Sate University as Assis-tant Professor in 1997 where hewas the recipient of the CamilleDreyfus Teacher-Scholar Award.In 2002, he joined Montana StateUniversity and is currently Pro-fessor of Chemistry and Biochem-istry and Director of the NASA

funded Astrobiology Biogeocatalysis Research Center and the MSUThermal Biology Institute. His research interests are in complexiron-sulfur enzymes, specifically hydrogenase and nitrogenase struc-ture, function and biosynthesis.

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4275

surprises (Fig. 1a). From this work, it was revealed that theactive site possessed an Fe atom that was previously missed bybiochemical and spectroscopic work. Even more intriguing wasthat the Fe was coordinated by a number of non-protein ligandsthat could not be resolved by X-ray crystallography. The non-protein ligands were determined to be cyanide and carbonylligands by Fourier transform infrared (FTIR) spectroscopy andchemical analysis, which resolved long standing issues concerningthe interpretation of carbon monoxide inhibited states of thehydrogenase.6 Through this work, the active site of the D. gigas[NiFe]-hydrogenase was found to exist as a heterometallic clusterwith a Ni ion coordinated by four cysteine thiolates to an Fe ion bytwo bridging thiolates. Additional coordination to Fe is suppliedby three diatomic ligands (two cyanide and one carbonyl) and abridging species for which its composition has been proposed to bea m-oxo, sulfido, hydroxo, peroxide, or sulfoxide group7–9 (Fig. 1b).

Structural characterization of the [FeFe]-hydrogenase camelater from Clostridium pasteurianum (CpI) and Desulfovibriodesulfuricans10,11 (Fig. 1c) and revealed a similar iron coordi-nation environment yet a somewhat different overall active sitecomposition. Common to these hydrogenases are the presenceof several accessory iron–sulfur clusters, which presumably func-tion to shuttle electrons to and/or from external physiologicalelectron donors and/or acceptors. For the [FeFe]-hydrogenasefrom Clostridium pasteurianum the apparent bifurcated pathwaywith two separate accessory clusters approaching the surface atdifferent positions may suggest that the enzyme can accommodate

alternative electron donors and/or acceptors. The six iron catalytic“H-cluster” active site (Fig. 1d) is comprised of a [4Fe-4S] cubanebridged via a single cysteinyl thiolate to a 2Fe subcluster inwhich both irons are coordinated by a CO and CN- ligand,and bridged by a m-CO ligand and a non-protein dithiolate ofstill unknown composition having been proposed to exist aspropane dithiolate, dithiomethylamine12, or dithiomethylether.13

Knowledge as to the stereochemical position of the CO and CN-

ligands on the active site was based on CO and CN- ability forhydrogen bonding.11 Subsequent FTIR studies revealed that theseenzymes are reversibly inhibited by CO12,14 and the structure of theCO-inhibited enzyme was determined via X-ray crystallographyfrom crystals of the C. pasteurianum enzyme grown in the presenceof CO.14 The structure revealed reversible hydrogen oxidationlikely occurs at a ligand-exchangeable site at the distal iron atomof the 2Fe subcluster in relation to the [4Fe-4S] cluster. In thepresumed oxidized state of the C. pasteurianum enzyme, a watermolecule occupies the ligand-exchangeable site. However, thiswater molecule is not present in the D. desulfuricans [FeFe]-hydrogenase structure as it is believed to represent a more reducedstate of the active site.

The [Fe]-hydrogenases were first shown to contain iron lig-ated by CO ligands in 200415 and very recently, the structuraldescription of the [Fe]-hydrogenase was revealed with bound co-factor allowing for the first time side by side comparison ofthese unique active sites.16 The [Fe]-hydrogenase active site existswith iron coordinated by two CO ligands, a cysteinyl sulfur, the

Fig. 1 Ribbon/space filling diagram of the (A) [NiFe]-hydrogenase from D. gigas (PDB code 1YQ98). (B) [FeFe]-hydrogenase from C. pasteurianum(PDB code 3C8Y13). For both structures, the protein domains are represented with different colors. For D. gigas, a small and a large subunit exist, for whichthey can be divided into two and five separate domains, respectively, according to protein folds. Atomic models of the [NiFe]- and [FeFe]-hydrogenaseactive site metal clusters are shown for each (B and D), respectively, (Fe: dark red, S: orange, O: red, N: blue, C: dark grey, unknown atom of dithiolateligand: magenta). For the D. gigas model, a bridging peroxide ligand is shown and represents the oxidized “unready” state of the active site. TheC. pasteurianum active site depicted is presumed to represent the oxidized form as a water molecule is present at the ligand-exchangeable site of the distalFe atom. All ribbon diagrams and atomic models were generated in PyMOL.103

4276 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

nitrogen of a 2-pyridinol compound, and a ligand of unknowncomposition.

Collectively, these enzyme active sites and features presumablyallow for the stabilization of reduced iron in a form capable ofcatalyzing the hydrogenase reaction, and in the case of the [Fe]-hydrogenase center, the dehydrogenation of methylene-H4MPT.The nature of the evolutionary and biochemical origins of thesecomplexes, as well as the electronic and structural basis for theirreactivity, poses an array of intriguing questions, which are beingaddressed by research groups from a diverse range of backgroundsand expertise.

Active site cluster mimics

The syntheses of hydrogenase active site analogs began soonafter the first structural report of the [NiFe]-hydrogenase whenworkers were motivated by the prospects of creating biomimeticcatalysts and gaining further insight into the reactivity of thehydrogenase active site clusters. Significant progress has beenmade in generating active site cluster mimics for both [NiFe]- and[FeFe]-hydrogenases and this work has been the topic of recentreviews.17–20 A more topical summary of the progress in the field isdescribed here.

The development of synthetic techniques suitable for theformation of heterobimetallic sulfur bridged CO and CN- ligatedclusters posed unique challenges to the field of organometallicchemistry, and in the case of the [NiFe] center, the events leading tothe generation of sulfur coordinated Ni species ligated in bidentatefashion to iron coordinated by CO and CN- were major highlights.Within a year of the first structural description of the [NiFe]-hydrogenase, Darensbourg and co-workers21 reported a hetero-bimetallic structural analogue (Fig. 2a) obtained by the reactionof a sulfur coordinated Ni2+ complex with an iron carbonyl.The resultant structure consisted of a nickel moiety bound toa Fe(CO)4 fragment via a single bridging sulfur. In 1997, a [NiFe]-hydrogenases active site model compound in which nickel waslinked to iron by two sulfur atoms (Fig. 2b) was reported22

from the reaction of an Fe(CO)2(NO)2 with a nickel complexto give the formation of a heterometallic [NiFe] complex withFe ligated by NO. The formation of mimics with heterometallic[NiFe] complexes coordinated to CO ligated Fe came later withthe complex synthesized by the reaction of an iron thiolate withNiCl2(dppe) under an atmosphere of CO23 (Fig. 2c). In 2002, theformation of an analog with Ni ligated solely by thiolates wasachieved24 (Fig. 2d), marking a significant advance from othersyntheses, which utilized phosphines as soft donor ligands.23,25

An additional advancement in bimetallic heteroatomic [NiFe]analogue synthesis came in 2005; when it was realized that reactionof the pre-formed iron carbonyl [Fe(CO)3(CN)2Br]- with a nickelcomplex gave a fully sulfur coordinated Ni moiety bridged toiron ligated by two CO and two CN- molecules26 (Fig. 2e). Thisdevelopment represented the culmination of many years of workby a diverse group of chemists to develop the reaction specificityto yield organometallic compounds with three main qualitiesrequisite for [NiFe]-hydrogenase active site resemblance: sulfurcoordinated Ni bound via two sulfur thiolates to an iron centerligated by both CO and CN-. Synthetic efforts to mimic the [NiFe]site continue to advance the understanding of H2 activation by theactive site, with a recent report suggesting that coordination of

Fig. 2 Selected synthetic structures mimicking the active sites of the (left)[NiFe]-hydrogenase active site and (right) [FeFe]-hydrogenase active site(discussed in text).

hydrogen by the [NiFe] center stems from an ability of Ni in theactive site to shift from square pyramidal to octahedral geometry,this being based upon observations of various [NiFe] complexes.27

In the midst of efforts to develop [NiFe] active site models,structural descriptions of the [FeFe]-hydrogenases in 1998 and199910,11 catalyzed the development of a parallel vein of researchaimed at mimicking the related but distinct different active sitestructures. Similar challenges were faced in the case of [FeFe]-hydrogenase active site analog synthesis as in the case of the[NiFe]-hydrogenases but with some variations. Beginning workwas directed at the synthesis of the 2Fe-subcluster of the H-cluster.

The previous synthesis and characterization of various di-iron carbonyl complexes, dating back to 192828,29 (Fig. 2f)and furthered in the 1980’s, provided a starting point for thesynthesis of H-cluster mimics and thus allowed the commu-nity to hit the ground running. These syntheses demonstratedthe formation of the di-ironhexacarbonyldisulfide compound[Fe2(SCH2CH2CH2S)(CO)6] present with a propane dithiolate30–32

starting from either Fe3(CO)12 or (m-S2)Fe2(CO)6 (Fig. 2g left).Following structural elucidation of the [FeFe]-hydrogenase, twoof the CO ligands on this compound were substituted withCN- by three different groups33–35 via reaction of Fe2(pdt)(CO)6

(pdt = propyl dithiolate) with two equivalents of CN- (Fig. 2gright) to give a remarkable analog of the H-cluster 2Fe-subclusterexhibiting a short Fe-Fe distance (2.6 A), and the presence ofboth CO and CN- ligands. The parent compound for this reactionwas again a remnant of non-intentional hydrogenase active sitemimicry carried out some thirty years prior to the knowledge of

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4277

the active site structure; King showed in 196136,37 the developmentof chemistry suitable for the thiolation of iron carbonyls byreaction with alkyl sulfides, allowing m-sulfur-iron bond formationof di-iron centers. Subsequent work on the di-iron alkyl-sulfideligated compounds resulted in variations in which the dithiolatelinkage was varied with central atomic positions being occupiedby ether38,39 and amine40 (Fig. 2h) functional groups, as opposed tothe methylene group at this position in the initial propane dithiolligated compounds. These variations were elicited by the inabilityto assign the central atom of the bridging dithiolate based oncrystallographic data since the likely possibilities (CH2, O, NH,NH2

+) are isoelectronic.A further challenge presented in the development of structural

analogs included adding the presence of a m-CO ligand. This goalwas first met by Pickett and co-workers41 with the formationof a mixed valent Fe1+Fe2+ compound with a m-CO ligand,although this compound proved to be very unstable. Later workby two groups gave stable m-CO containing di-iron compounds42,43

(Fig. 2i) and in addition to these CO bridged compounds, a modelwith a bridging hydride capable of binding and activating H2 atan Fe2+ site has been reported.44,45 However, these compoundsdid not have CN- ligands and instead were coordinated by bulkyIMES, PMe3, and dppv ligands, respectively, which presumablyfunctioned to stabilize the unique geometry that the m-CO liganddemands. A major advancement in the synthesis of H-clustermimics came in 2005 with the synthesis of the complete 6Feframework of the H-cluster46 (Fig. 2j). The bridging of a di-ironcarbonyl subcluster to a [4Fe-4S] cubane was accomplished byfirst capping Fe3(CO)12 by reaction with a three sulfur containingthioester with the subsequent product being reacted with a thiolatecoordinated [4Fe-4S] cubane. This general approach allows signif-icant flexibility for the formation of additional analogue targets aswell as providing a means to probing cluster characteristics as hasbeen recently demonstrated for the investigation of CO binding ata 2Fe center.47

Mimetic synthesis of hydrogenase active sites metal clustersresulted in the development of a suite of chemical approaches.From a synthetic perspective, the starting materials for a numberof these compounds are simple iron carbonyl complexes such asFe(CO)5, Fe2(CO)9, and Fe3(CO)12, to which sulfur in the form ofsulfide or thiolates and CN- from compounds such as Et4NCN,and a variety of nickel complexes may be added. In terms ofcatalysts for hydrogen energy technologies, platinum chemistryis very effective at catalyzing reversible hydrogen oxidation atlittle or no overpotential, however, with the limited availabilityof platinum there is a real need for examining non-noble metalbased solutions to accomplishing this chemistry. Lessons learnedin synthetic efforts are thus valuable for the continued pursuitof hydrogenase active site mimics that may well yet prove to becatalytic solutions for the accumulation of H2 as a fuel currency,as well as providing a vantage point from which to consider thebiochemical events that led to the formation of these interestingcatalytic co-factors.

Biosynthesis and maturation

Biological utilization of metal-sulfide protein bound co-factors,which are ubiquitous in living systems, stems from the inherentelectronic and reactive properties of both iron and sulfur.48

The utilization and assembly of these co-factors necessitates theoperation of a complex and specialized set of proteins to allowfor the coordinated and regulated delivery of atoms in a way thatavoids the inherent toxicity of free aqueous metal ions.49–52 Theformation of the hydrogenases provides additional biosyntheticchallenges and has long puzzled biochemists as to the molecularmechanisms responsible for the harnessing of the unique ligandset of the active sites. The collective features of the hydrogenaseco-factors require the operation of a specialized set of proteins tobiosynthesize the active site cluster. Given the similarity in overallcomposition between the hydrogenase active sites, it is interestingif not surprising that the generation of the hydrogenase active sitesappears to occur by non-redundant chemistry, with distinct setsof maturation proteins being required for active site synthesis ofeach of the hydrogenases. A summary of the proposed activities forgene products involved in [NiFe]- and [FeFe]-hydrogenase activesite cluster biosynthetic pathways is provided in (Fig. 3) and isdiscussed further below.

[NiFe]-hydrogenase active site synthesis has been studied in anumber of organisms including Ralstonia eutropha, Bradyrhizo-bium japonicum, Rhizobium leguminosarum, Azotobacter species,and E. coli and proceeds by the action of at least six accessoryproteins, with seven being required in cases in which proteolyticprocessing is required at the final step of maturation.53 Theproposed roles of individual proteins in the maturation processfor E. coli hydrogenase 3 is summarized in (Fig. 4A) and beginswith the generation of cyanide from carbamoylphosphate bythe action of the two proteins HypF and HypE. In a twostep process that requires two ATP hydrolyzing events, HypFtransfers the carboxamide group of carbamoylphosphate to HypEas thiocarboxyamide where dehydration occurs to yield an enzymebound thiocyanate, which can later be donated to iron.54–57

Subsequent interaction of thiocyanate bound HypE with theHypD/HypC complex results in the donation of the CN- ligandto an iron bound to HypD/HypC.58 The CN- ligated iron boundby the HypD/HypC complex is then transferred by HypC tothe large subunit of the hydrogenase enzyme.54 Still in questionis the metabolic source of the CO ligand and at which stageit is incorporated. Carbamoylphosphate has been ruled out asthe source of CO by labelling studies59,60 and the utilization of apathway involving acetate has been suggested.61 With CO and CN-

coordinated Fe now present on the large subunit, Ni is insertedby the actions of the Ni binding GTPase HypB62,63 and the Nibinding HypA.64 The final step in [NiFe]-hydrogenase maturationinvolves proteolytic cleavage of a C-terminal extension, if present.

[FeFe]-hydrogenase active site biosynthesis shares many of thechemical challenges as just outlined for the [NiFe]-hydrogenasesite, but from what is known thus far appears to require fewergene products perhaps due to the presence of only a singletype of metal ion in the active site1. Indeed, [FeFe]-hydrogenasematuration was found to be dependent on three gene products(HydE, HydF and HydG) based on analysis of deletion mutantsof Chlamydomonas reinhardtii in which the gene products HydEFand HydG were identified to be required for the accumulation ofactive hydrogenase.65 Subsequent analysis revealed that the geneproducts HydE, HydF, and HydG are conserved in organismsharbouring an active [FeFe]-hydrogenase, with HydEF existing asa fusion product in C. reinhardtii. Based on sequence comparison,the two proteins HydE and HydG were predicted to belong to the

4278 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

Fig. 3 Overview of genes involved in the biosynthesis of the [NiFe]-hydrogenase and [FeFe]-hydrogenase active site and their proposed functions.

radical S-adenosylmethionine (AdoMet) family of enzymes, andHydF was predicted to bind an iron-sulfur cluster in additionto having the ability to hydrolyse nucleotide tri-phosphates.65

Following this discovery, an E. coli based heterologous expressionallowing the co-expression of the [FeFe]-hydrogenase structuralprotein with the three accessory proteins was developed andshown to be sufficient for heterologous maturation, paving theway for future biochemical studies. In subsequent work, the genebased predictions for the enzymatic activities of these enzymeswas corroborated by analysis of anaerobically reconstituted Hydaccessory protein homologues from the organism T. maritima.66, 67

The AdoMet class of enzymes comprises a large family of iron-sulfur cluster containing enzymes capable of catalyzing a widerange of free radical initiated reactions including carbon-carbonbond formation, isomerisations, and sulfur insertions68; thus anumber of chemistries may be envisaged as taking part in H-clusterbiosynthesis. Knowledge of the enzyme families to which the[FeFe]-hydrogenase assembly factors belong to led to a hypothesisbased off pre-established reactivity of AdoMet enzymes in whichthe activities of HydE and HydG, respectively, are proposed toact to generate the CO and CN- ligands, as well as the bridgingdithiolate ligand.69 The three maturation proteins are presumed tobe responsible in total for the generation of the ligand set of the[FeFe] subcluster with the possibility that CO could be generatedthrough the action of carbon monoxide dehydrogenase excluded as[FeFe]-hydrogenases are found in organisms which do not containthis enzyme. In our hypothesis for [FeFe]-hydrogenase maturation,the three gene products HydE, HydF, and HydG are thought tobe directed solely at the synthesis of the 2Fe subunit of the [FeFe]active site, as the [4Fe-4S] moiety presumably would not require theaction of a specific set of assembly proteins but instead be formedvia the action of the endogenous host iron-sulfur cluster assemblymachinery. HydE or HydG is envisioned as acting to generatethe bridging dithiolate linkage in a mechanism analogous to thatof the radical-AdoMet protein LipA, which catalyzes carbon–

sulfur bond formation; such C–S bond formation, occurring ata bridging sulfide of an iron–sulfur cluster, would be expectedto shift the reactivity from the sulfur atoms to the irons of thecluster. Following alkylation of the sulfides, a radical-initiatedamino acid decomposition of glycine is proposed to occur by theaction of HydE or HydG, resulting in the formation of CO andCN-. The first step of this latter proposed reaction is analogous tothe chemistry catalyzed by the radical-AdoMet enzyme pyruvateformate-lyase activating enzyme, lysine amino mutase, or ThiHinvolved in thiamine biosynthesis which involves the generation ofan amino acid radical intermediate. These ligand-forming eventsare hypothesized to result in assembly of the 2Fe subcluster ofthe H-cluster on one of the accessory proteins. At this stagethe role of GTP binding and hydrolysis by HydF is unknown,however, our results concerning in vitro activation of the [FeFe]-hydrogenase would suggest that these activities are not directed atcluster insertion as originally proposed.

Definitive knowledge as to the roles of the accessory proteinsin active site formation has been advancing rapidly, due in largepart to development of an in vitro system for the activation of thestructural hydrogenase protein.70 This in vitro system has led to therecent identification of HydF as a scaffold or carrier protein whichacts at the terminal step of H-cluster biosynthesis71, presumablydelivering the 2Fe sub-cluster to the immature structural protein(Fig. 4B). Currently unresolved issues relate to the enzymatic rolesand substrates of the two AdoMet proteins HydE and HydG, theprecise nature of the cluster precursor bound by HydF prior todelivery, and the role of GTP hydrolysis by HydF in the maturationprocess.

Very little is known regarding the biosynthetic route for the caseof the [Fe]-hydrogenases. As is the case between the [NiFe]- and[FeFe]-hydrogenases, however, phylogenetic evidence indicatesthat the structural protein is evolutionarily isolated and stemsfrom a separate lineage, suggesting a separate route for thegeneration of CO ligands at the iron center. One possible and

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4279

Fig. 4 Overview of the metal cluster assembly processes for the (A) [NiFe]- and (B) [FeFe]-hydrogenases. The [NiFe]-hydrogenase maturation schemeis based on hydrogenase-3 in E. coli and the [FeFe]-hydrogenase maturation model is thought to be general to organisms containing this enzyme.(A) CN- is generated by the action of HypE and HypF utilizing carbamoylphosphate and ATP forming thiocyanate bound by HypE. The CN- group onHypE is transferred to the iron coordinating HypD/HypC complex, which is then transferred to the large subunit of the structural hydrogenase protein.The source of CO is unknown. HypA and HypB act to deliver nickel, and in most cases a C-terminal peptide of the hydrogenase is endopeptidicallycleaved completing the maturation process. (B) The two SAM radical enzymes HydE and HydG are proposed to impose non-protein cluster ligands onan existing [2Fe-2S] rhomb in two separate steps: (i) the formation of the dithiolate linkage and (ii) CO and CN- ligands, possibly from an amino acidprecursor on the scaffold protein HydF. HydF subsequently delivers the ligand modified 2Fe unit to the immature structural protein, which is presumedto exist with the presence of a preformed [4Fe-4S] cluster, accomplishing activation.

4280 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

intriguing correlation does exist however, as the gene encodinga radical-AdoMet protein has been observed to cluster withthe gene encoding the [Fe]-hydrogenase structural protein in anumber of cases.72 Given this observation, it is interesting tospeculate about the commonality of radical-AdoMet chemistryin the generation of essential CO ligands for both the [Fe]-and the [FeFe]- hydrogenases and that the CO of both thesehydrogenases may be formed through a radical initiated aminoacid decomposition event.

Despite the distinct differences in the maturation systems forthe [NiFe]-, [FeFe]-, and [Fe]-hydrogenases, similar chemical andcatalytic functionality can be found. In the case of the [NiFe]-and [FeFe]-hydrogenases, a cluster precursor has been proposedto be assembled on a scaffold protein prior to insertion intothe immature hydrogenase structural protein, and in both casesGTPase activity has been implicated in having a role in thematuration process. These two hydrogenase maturation systemsdiffer however in their respective routes to the generation of thediatomic ligands CO and CN-. Insofar as organisms harbouringhydrogenases are able to occupy niches unavailable to organismswithout these enzymes, the functional, catalytic, and maturationpathway similarities between hydrogenases attests to the ability ofevolution to produce functionally similar yet distinct solutions toenvironmental problems. This points to an apparent optimum ofhydrogenase catalytic activity being associated with CO and CN-

ligated iron, presumably as a direct consequence of the ability ofthese ligands to stabilize metals in low oxidation states. In short,three distinct hydrogenase enzymatic systems have convergentlyevolved to contain, in whole or in part, iron coordinated by smallinorganic p acid ligands (CO and/or CN-), suggesting that the lowvalent iron stabilized by such p acids has been a central drivingforce in hydrogenase evolution.

Prebiotic mineral clusters, ligand accelerated catalysis andancient enzymes

Hydrogen occupies a central role as a mobile electron carrier inbiological systems. Given the wide diversity of organisms thatutilize hydrogen oxidation/reduction in metabolic processes, aswell as the potential to derive high concentrations of hydrogengeochemically via processes such as serpentinization73,74 and thehigh concentration of hydrogen present on the early earth75, it isreasonable to suppose that hydrogen as a mobile electron carrierand potent reductant has been coupled to metabolic processesfrom their inception.2,74,76 The wide distribution of hydrogenasesin organisms and the presence of hydrogen on the early earth thusmake the consideration of hydrogenases and the events which ledto the formation of hydrogen oxidation and reduction catalystsespecially relevant to the consideration of early energetics and theorigin of life.

In addition to reversible hydrogen oxidation, small moleculeinterconversions and cycling of the major atomic constituentsof life play a critical role in contemporary living systems andmust have also participated in the processes accompanying thetransitions from an abiotic to a biotic world. Given that processesincluding nitrogen fixation, reversible carbon monoxide oxidation,and hydrogenase activity occur biologically via the utilizationof metal clusters, it is possible that these clusters in biologicalsystems are signatures and even reflections of ancient metabolic

and prebiotic processes in which transition metal based catalystsoccupied salient roles. Metal sulfides and metallic transition metalshave been shown to function as catalysts for biologically relevantreactions and these catalytic activities may have been requisitefor the accumulation of molecules later incorporated into livingsystems. Activities associated with metallic, metal sulfide, andmetal oxide based catalysts include the reduction of N2 to NH3

77–79,the synthesis of pyruvate80 and amino acids, and peptides.81,82

These activities strongly suggest that transition metals, whichfeature so prominently in contemporary biology, were relevantto chemistry associated with the development of life.

As well as occupying an important role in individual reactions,a further important role of both prebiotic and contemporarycatalysts is that of the reaction network connector. In the sensethat catalysts bound the metabolic capacity and flexibility ofan organism, the formation of such catalysts prebiotically mayhave been a major barrier in the origin of life. Catalysts inmetabolic networks function to kinetically channel products andsubstrates at time scales allowing for sufficient interconnectivityto result in the ability to couple energy yielding reactions to thosethat are utilized by organisms to maintain function. With theseactivities in mind the importance of catalysis in the origins ofbiological complexity can perhaps not be overstated. Taking intoaccount the aforementioned chemistry underlying the creation ofhydrogenase active sites both biologically and synthetically resultsin a unique perspective from which to consider the origins ofprebiotic catalysts. A greater understanding of (i) the extent towhich minerals present on the early Earth may have harboured pre-existing catalytic functionality, (ii) the ability of metal clusters toform and the role of this in the formation of biologically accessiblemolecules, and (iii) the contribution of ligand modifying events tothe emergence of catalytic functionality, may together yield insightas to the nature or the original catalysts and catalytic events whichled to the emergence of life.

The present day biological utilization of the metal sulfide co-factors observed in hydrogenases and other enzymes to accomplishefficient reactivity hints at the possibility that these co-factorsthemselves may have been derived from geologically relatedmineral precursors and thus have been adopted, modified, andrefined through evolutionary time to become the extant clustersobserved today existing as products of geomimicry. Various iron-sulfur minerals exhibit structures similar to those seen in biology,and attention has been drawn to the resemblance of biologicalclusters such as that observed in the enzyme CODH and that ofthe mineral greigite83, as well as a similarity in geometry betweenthe mineral mackinawite and the di-iron center of the [FeFe]-hydrogenase (Fig. 5a).84 These “ready made” clusters have beenproposed to have provided a structural basis for the catalytic eventsrequired in the origin of life84, itself being a product at least in partof inherent geological functionalities.

Along with the similarities existing between mineral and enzymeactive site structure, an emerging body of work has contributedto the understanding of how ligand modifications, includingcarbonylation, can occur in prebiotic conditions. Iron carbonylcomplexes, which as described above form the reactive startingmaterial of many active site analogs, have been synthesizedfrom ferrous iron upon the addition of CO2

85, CO and ethanethiol86, and as well as from aqueous preparations in a simulatedhydrothermal reactor upon the reaction of iron sulfide with CO.80,87

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4281

Fig. 5 Possible functional roles of prebiotic metal sulfides existing as (A) “ready made” as described by Russell in which some metal sulfides exist withlattice structures of striking similarity to clusters observed in biology. Here, the iron sulfide mackinawite is shown, highlighting resemblance of the minerallattice structure to the 2Fe unit of the [FeFe]-hydrogenase, albeit existing without ligand modifications (adapted from ref. 84). (B) Possible formation ofthe amino acid glycine via the condensation of iron bound CO and CN- with H2O in a reverse reaction to that proposed to occur in the biosynthesisof CO and CN- from glycine. (C) An iron-sulfur mineral surface, shown to be capable of ligand-accelerated autocatalysis and feedback as describedby89 as a means of catalyst formation and diversification. The mineral surface is initially capable of carrying out catalytic formation of a molecule thatsubsequently binds to the mineral surface, and via ligand accelerated catalysis enhances the formation of more of the same molecule. In this way, thecatalyst is selected for kinetically, propagating itself through ligand-assisted autocatalysis.

In addition to these carbonylated iron species, iron cyanide speciesin the form of Prussian blue have been isolated from Miller–Ureytypes of experiments showing that spark discharge experimentsare sufficient for the generation of CN-, which reacts with ferrousiron.88 The observation of iron carbonyl and iron cyanide speciesindicate that the unique ligand set of the hydrogenase enzymes maybe accessible in prebiotic conditions. Furthermore, iron cyanidesynthesis has been shown to be feasible with nucleophilic ironand electrophilic CN- and in the reverse situation57, validating theproposed model for CN- transfer occurring by the action of HypEin [NiFe]-hydrogenase active site biosynthesis and drawing aninteresting parallel between geochemical and biochemical ligandaddition. A further possible parallel between geochemical andbiochemical events exists in the possibility of small moleculeformation by the condensation of these modifiers. In an extensionof chemistry in which CO and CN- become bound to iron, thereverse reaction could take part in the generation of the aminoacid glycine. In the [FeFe]-hydrogenase maturation hypothesisoutlined above69, glycine is proposed to be the CO and CN- ligandprecursor; a reverse process in which CO and CN- bound to ironcondense at the mineral–solution interface with water would resultin its formation (Fig. 5b), illustrating the possibility that both theformation as well as the degradation of prebiotic molecules atmineral surfaces could have had important consequences at theorigin of life. CO and CN- addition to iron in possible prebioticconditions may have thus acted as a source for the accumulationof biologically accessible molecules such as amino acids, and asdescribed further below, as a driving force for the formation ofdiverse catalytic functionality acting as metal surface modifyingagents via ligand addition.

The challenge of the formation of a diverse repertoire ofcatalysts has been addressed in part by a proposal based off ofligand modification. Here, the chemical reactions that lead tometal ligand modification are viewed as acting in a type of chemicalevolution where catalytically derived products may themselves addto metal centers in chemistry analogous to that described above,modifying their reactivity perhaps for their betterment and thus

act to diversify and enhance catalytic attributes.89 This process, inwhich the addition of a specific ligand leads to a quickening of aparticular reaction is termed “ligand accelerated catalysis” and isapplicable to both heterogeneous and homogeneous catalysts, be-ing especially relevant to transition metal based catalysts that mayexist in environments in which dynamic ligand exchange events arepossible.90 The series of events in which ligand modification mayserve as a route to positive feed back, termed ligand-acceleratedautocatalysis, has been proposed to operate as a most basic type ofreproductive mechanism89 and may function in a manner enablingcatalytic mineral core derived products to modify clusters therebyproducing generations and families of catalysts (Fig. 5c). Indeed,this type of phenomenon may occur in a system wide sense,with neighbouring mineral sites of differing catalytic proclivitycontributing to the formation of proficient neighbouring catalysts,thus forming a network basis for catalytic fecundity.

Also relevant to the discussion of what events could comprisecatalyst formation is the notion of “defect” sites or areas ofdiscontinuity of bulk structure on mineral surfaces. At theselocations, which represent approximately 1% of total surfacearea,91,92 defects in lattice structure result in unique electronic aswell as spatial localities, where molecules may transiently bindin orientations and environments amenable to reaction progress.Heterogeneous catalysis by mineral surfaces has indeed beenobserved to occur at these sites93,94 and highlights the observationof biological clusters existing as structurally, stoichiometrically,and molecularly tuned derivatives of minerals.

In addition to the possibility of pre-existing functionality ofminerals and ligand modification reactions as having playeda role in the synthesis of relevant biological constituents, thepossibility of individual iron sulfur clusters in solution exists.Iron–sulfur clusters of the nature [Fe4S4(SR)]2- have been shownto assemble in aqueous solvent from iron, sulfide, and thiols(RSH).95 Together with ligand modification schemes and reportedexperiments showing the generation of iron carbonyls and reactiveprecursors obtained through hydrothermal reaction conditions80,these experiments show that a suite of chemistries and possible

4282 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

catalyst generating events closely related to those observed inbiology are possible in putative prebiotic conditions and indicatethat relevant transition metal based chemistry can occur outsideof proteins. Given this, it is interesting to speculate on the catalyticproperties of these modified metal species and the possibility thatcluster assembly events similar to those that occur biologically tocreate the active site architecture of the hydrogenases and othermetalloproteins may at least be partly mimicked by naturallyoccurring geochemical processes.

Based on the above discussion of synthetic, biosynthetic, andprebiotic reactions, a putative route to biological catalyst creationand the steps associated with this are figuratively portrayed inFig. 6 starting from an assortment of FeS mineral surfaces. Broadlydivided into stages these transitions include the concepts of (i) amineral basis for early catalysis, (ii) the formation of individualcatalytic entities; ligand modification and cluster formation, and(iii) the encapsulation of the nascent metal sulfide clusters bypeptides. Specifically, the proposed process of catalyst evolution isportrayed to include the inherent activities of mineral surfaces inwhich the surface itself or its transformation (oxidation in the caseof pyrite formation) results in relevant reactivity (e.g. ref. 78,96–98). Also at this stage is represented the aforementioned conceptof “ready made”84 clusters and defect sites which may catalyzereactions contributing to the formation of biologically accessiblemolecules. With these roles in mind, a next step in catalystevolution may have been the formation of individual clusters andcluster components such as the di-iron hexacarbonyl species asobtained under hydrothermal conditions upon the reaction withCO and alkyl thiols80,99, the formation of iron sulfur cubanes, andthe formation of Fe–CN bonds as observed in spark dischargeexperiments. The integration of these concepts and chemistriesmay lead to the generation of proto-metallo-co-factors which bearsimilarity to those observed in biology today. These intermediaryclusters may, in a similar way as described by Milner-White andRussell, be bound by peptide “nests” comprised of simple aminoacids.100,101 Finally, the incorporation of these clusters into morecomplex peptide chains leads to protein bound catalysts capableof supporting catalytic activity from a robust mobile peptideplatform and the formation of an early proto-metabolism repletewith catalysts of sufficient diversity by which to harness chemicalenergy.

Prospectus

The structural and spectroscopic characterization of the hy-drogenase enzymes created new lines of research that, in aninteresting complementarity, are applicable both to future energysolutions and to understanding the past evolution of hydrogenredox catalysts and their role in the formation of life. With therapid pace of research in this area, the field is certain to yieldadditional new insights in the coming years. For example, muchis yet to be revealed regarding biosynthesis of the metal centersof the hydrogenases, particularly the [FeFe] and [Fe] classes. Inaddition, while synthetic studies have resulted in the developmentof creative and often elegant chemistries, an effective first rowtransition metal based catalyst for hydrogen production basedon the structure of hydrogenase active sites has not emerged,perhaps hinting that the proteinaceous environments surroundingthe active sites of the hydrogenases are critical for fine-tuning

Fig. 6 Possible set of major transition stages in the emergence ofprebiotically relevant catalysts from metal sulfide minerals. Evolutionaryroutes to protein bound transition metal sulfide catalysts are postulatedas having originated from the basis of iron and sulfur bearing minerals.As a first illustrative stage stemming from the inherent properties ofmetal-sulfides, the lattice structure of pyrite104 is shown as being derivedfrom FeS by oxidation which can serve as a driver for prebiotic catalysis.In addition, the structure of a nickel–iron sulfide is shown which mayharbour pre-existing functionality. Also depicted at this stage is theportrayal of a mineral surface defect site where atomic irregularitiesin the lattice structure may lead to relevant catalytic activity throughmodified electronic and small molecule binding characteristics. Continuingin catalyst evolution, the next stage involves individual cluster formationand possible ligand modifications in the forms of CO and CN- addition andthe creation of individual metal centers resulting in the iron hexacyanide,[4Fe-4S], and di-iron hexacarbonyl species (Fe: dark red, S: orange,O: red, C: dark gray, R group: magenta, N: blue) resulting in anintermediary step prior to the protein encapsulation of metal co-factors.In the final stage (top), the combination of these chemistries leads tosimilar metal clusters to those seen in biology today, being bound byrelatively simple peptides which serve to contribute the unique solventmicro-environment observed in contemporary metalloenzymes and existas ligand modifying agents themselves. Shown as examples are the metalcluster active sites of acetyl-CoA synthase105, the [FeFe]-hydrogenase,and carbon monoxide dehydrogenase.106 Together, these representationsillustrate only a few of the myriad possibilities and routes available in theformation of transition metal based catalytic species.

the catalytic potential of the metal centers, making studies inwhich synthetic metal co-factors are linked to designed peptidesparticularly attractive.102 Given their high catalytic efficiencies,hydrogenase enzymes themselves remain interesting prospects forbiotechnology for non-noble metal based hydrogen oxidation andhydrogen production catalysis in a hydrogen fuel economy and athorough understanding of the structural determinants of catalysisand the biosynthesis of these enzymes is required for the fullpotential to be realized. Finally, the ideas of preformed geolog-ically derived catalytic motifs, ligand accelerated autocatalysis,

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4283

and the chemical ability of catalysts derived as such, may togetherprove to yield deep insight into the origins of modern biologicalcomplexity. The brief account of possibilities concerning prebioticcatalysis described herein indicates that much is possible in relevantprimordial conditions with concern to cluster assembly and ligandmodification. The observations presented demonstrate a diversearray of modifications that minerals and individual metal ions mayundergo and indicate that studies into transition metal mineralsand the possible transformations possible from these may bea highly insightful area of research with concern to prebioticcatalysis. Continued work directed at the areas described hereinwill shed new and fascinating light on hydrogenases; biology’sresponse to the hydrogen redox couple.

Acknowledgements

The work was supported by an Air Force Office of ScientificResearch Multidisciplinary University Research Initiative Award(FA9550–05-01–0365) to J. P. and the NASA AstrobiologyInstitute-Montana State University Astrobiology BiogeocatalysisResearch Center (NNA08CN85A) award to J. B. and J. P. S. M. issupported by an NSF IGERT Fellowship by the MSU Programin Geobiological Systems (DGE 0654336).

References

1 P. M. Vignais, B. Billoud and J. Meyer, FEMS Microbiol. Rev., 2001,25, 455–501.

2 P. M. Vignais, Results Probl. Cell Differ., 2008, 45, 223–252.3 W. Martin and M. Muller, Nature, 1998, 392, 37–41.4 D. Moreira and P. Lopez-Garcia, J. Mol. Evol., 1998, 47, 517–530.5 A. Volbeda, M. H. Charon, C. Piras, E. C. Hatchikian, M. Frey and

J. C. Fontecilla-Camps, Nature, 1995, 373, 580–587.6 R. P. Happe, W. Roseboom, A. J. Pierik, S. P. Albracht and K. A.

Bagley, Nature, 1997, 385, 126.7 H. Ogata, S. Hirota, A. Nakahara, H. Komori, N. Shibata, T. Kato,

K. Kano and Y. Higuchi, Structure, 2005, 13, 1635–1642.8 A. Volbeda, L. Martin, C. Cavazza, M. Matho, B. W. Faber, W.

Roseboom, S. P. J. Albracht, E. Garcin, M. Rousset and J. C.Fontecilla-Camps, J. Biol. Inorg. Chem., 2005, 10, 239–249.

9 S. E. Lamle, S. P. J. Albracht and F. A. Armstrong, J. Am. Chem. Soc.,2004, 126, 14899–14909.

10 J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science,1998, 282, 1853–1858.

11 Y. Nicolet, C. Piras, P. Legrand, C. E. Hatchikian and J. C. Fontecilla-Camps, Structure, 1999, 7, 13–23.

12 Y. Nicolet, A. L. de Lacey, X. Vernede, V. M. Fernandez, E. C.Hatchikian and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 2001,123, 1596–1601.

13 A. S. Pandey, T. V. Harris, L. J. Giles, J. W. Peters and R. K. Szilagyi,J. Am. Chem. Soc., 2008, 130, 4533–4540.

14 B. J. Lemon and J. W. Peters, Biochemistry, 1999, 38, 12969–12973.15 E. J. Lyon, S. Shima, R. Boecher, R. K. Thauer, F. W. Grevels, E. Bill,

W. Roseboom and S. P. J. Albracht, J. Am. Chem. Soc., 2004, 126,14239–14248.

16 S. Shima, O. Pilak, S. Vogt, M. Schick, M. S. Stagni, W. Meyer-Klaucke, E. Warkentin, R. K. Thauer and U. Ermler, Science, 2008,321, 572–575.

17 S. Canaguier, V. Artero and M. Fontecave, Dalton Trans., 2008, 315–325.

18 T. W. Chiou and W. F. Liaw, C. R. Chimie, 2008, 11, 818–833.19 M. Y. Darensbourg, E. J. Lyon, X. Zhao and I. P. Georgakaki, Proc.

Natl Acad. Sci. U. S. A., 2003, 100, 3683–3688.20 X. M. Liu, S. K. Ibrahim, C. Tard and C. J. Pickett, Coord. Chem.

Rev., 2005, 249, 1641–1652.21 C. H. Lai, J. H. Reibenspies and M. Y. Darensbourg, Angew. Chem.,

1996, 35, 2390–2393.

22 F. Osterloh, W. Saak, D. Haase and S. Pohl, Chem. Commun., 1997,979–980.

23 S. C. Davies, D. J. Evans, D. L. Hughes, S. Longhurst and J. R. Sanders,Chem. Commun., 1999, 1935–1936.

24 D. Sellmann, F. Geipel, F. Lauderbach and F. W. Heinemann, Angew.Chem., 2002, 41, 632–634.

25 M. C. Smith, J. E. Barclay, S. P. Cramer, S. C. Davies, W. W. Gu,D. L. Hughes, S. Longhurst and D. J. Evans, Dalton Trans., 2002,2641–2647.

26 Z. L. Li, Y. Ohki and K. Tatsumi, J. Am. Chem. Soc., 2005, 127,8950–8951.

27 Y. Ohki, K. Yasumura, K. Kuge, S. Tanino, M. Ando, Z. Li and K.Tatsumi, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 7652–7657.

28 H. Reihlen, A. Grul and G. Hessling, Liebigs Ann. Chem., 1929, 472.29 H. Reihlen, A. von Friedolsheim and J. Oswald, Liebigs Ann. Chem.,

1928, 465.30 D. Seyferth, R. S. Henderson and L. C. Song, J. Organomet. Chem.,

1980, 192, C1–C5.31 D. Seyferth, G. B. Womack, M. K. Gallagher, M. Cowie, B. W. Hames,

J. P. Fackler and A. M. Mazany, Organomet., 1987, 6, 283–294.32 A. Winter, L. Zsolnai and G. Huttner, Z. Naturforsch., B, 1982, 37,

1430–1436.33 A. Le Cloirec, S. P. Best, S. Borg, S. C. Davies, D. J. Evans, D. L.

Hughes and C. J. Pickett, Chem. Commun., 1999, 2285–2286.34 E. J. Lyon, I. P. Georgakaki, J. H. Reibenspies and M. Y. Darensbourg,

Angew. Chem., 1999, 38, 3178–3180.35 M. Schmidt, S. M. Contakes and T. B. Rauchfuss, J. Am. Chem. Soc.,

1999, 121, 9736–9737.36 R. B. King, J. Am. Chem. Soc., 1962, 84, 2460.37 R. B. King, P. M. Treichel and F. G. A. Stone, J. Am. Chem. Soc.,

1961, 83, 3600–3604.38 H. X. Li and T. B. Rauchfuss, J. Am. Chem. Soc., 2002, 124, 726–727.39 L. C. Song, Z. Y. Yang, H. Z. Bian and Q. M. Hu, Organometallics,

2004, 23, 3082–3084.40 J. D. Lawrence, H. X. Li, T. B. Rauchfuss, M. Benard and M. M.

Rohmer, Angew. Chem., 2001, 40, 1768–1771.41 M. Razavet, S. J. Borg, S. J. George, S. P. Best, S. A. Fairhurst and C.

J. Pickett, Chem. Commun., 2002, 700–701.42 B. E. Barton and T. B. Rauchfuss, Inorg. Chem., 2008, 2261–2263.43 T. Liu and M. Y. Darensbourg, J. Am. Chem. Soc., 2007, 129, 7008–

7009.44 X. Zhao, I. P. Georgakaki, M. L. Miller, R. Mejia-Rodriguez, C. Y.

Chiang and M. Y. Darensbourg, Inorg. Chem., 2002, 41, 3917–3928.45 X. Zhao, I. P. Georgakaki, M. L. Miller, J. C. Yarbrough and M. Y.

Darensbourg, J. Am. Chem. Soc., 2001, 123, 9710–9711.46 C. Tard, X. M. Liu, S. K. Ibrahim, M. Bruschi, L. De Gioia, S. C.

Davies, X. Yang, L. S. Wang, G. Sawers and C. J. Pickett, Nature,2005, 433, 610–613.

47 F. F. Xu, C. Tard, X. F. Wang, S. K. Ibrahim, D. L. Hughes, W. Zhong,X. R. Zeng, Q. Y. Luo, X. M. Liu and C. J. Pickett, Chem. Commun.,2008, 606–608.

48 H. Beinert, Eur. J. Biochem., 2000, 267, 5657–5664.49 C. Ayala-Castro, A. Saini and F. W. Outten, Microbiol. Mol. Biol.

Rev., 2008, 72, 110–125.50 M. Fontecave and S. Ollagnier-de-Choudens, Arch. Biochem. Biophys.,

2008, 474, 226–237.51 R. Lill and U. Muhlenhoff, Annu. Rev. Biochem., 2008, 77, 669–700.52 X. M. Xu and S. G. Moller, Chembiochem, 2008, 9, 2355–2362.53 A. Bock, P. W. King, M. Blokesch and M. C. Posewitz, Adv. Microb.

Physiol., 2006, 51, 1–71.54 M. Blokesch and A. Bock, J. Mol. Biol., 2002, 324, 287–296.55 M. Blokesch, A. Paschos, A. Bauer, S. Reissmann, N. Drapal and A.

Bock, Eur. J. Biochem., 2004, 271, 3428–3436.56 A. Paschos, R. S. Class and A. Bock, FEBS Lett., 2001, 488, 9–12.57 S. Reissmann, E. Hochleitner, H. F. Wang, A. Paschos, F. Lottspeich,

R. S. Glass and A. Bock, Science, 2003, 299, 1067–1070.58 M. Blokesch, S. P. J. Albracht, B. F. Matzanke, N. M. Drapal, A.

Jacobi and A. Bock, J. Mol. Biol., 2004, 344, 155–167.59 L. Forzi, P. Hellwig, R. K. Thauer and R. G. Sawers, FEBS Lett.,

2007, 581, 3317–3321.60 O. Lenz, I. Zebger, J. Hamann, P. Hildebrandt and B. Friedrich, FEBS

Lett., 2007, 581, 3322–3326.61 W. Roseboom, M. Blokesch, A. Bock and S. P. J. Albracht, FEBS

Lett., 2005, 579, 469–472.62 J. W. Olson and R. J. Maier, J. Bacteriol., 2000, 182, 1702–1705.

4284 | Dalton Trans., 2009, 4274–4285 This journal is © The Royal Society of Chemistry 2009

63 R. Waugh and D. H. Boxer, Biochimie, 1986, 68, 157–166.64 R. Bohm, M. Sauter and A. Bock, Mol. Microbiol., 1990, 4, 231–

243.65 M. C. Posewitz, P. W. King, S. L. Smolinski, L. P. Zhang, M. Seibert

and M. L. Ghirardi, J. Biol. Chem., 2004, 279, 25711–25720.66 X. Brazzolotto, J. K. Rubach, J. Gaillard, S. Gambarelli, M. Atta and

M. Fontecave, J. Biol. Chem., 2006, 281, 769–774.67 J. K. Rubach, X. Brazzolotto, J. Gaillard and M. Fontecave, FEBS

Lett., 2005, 579, 5055–5060.68 P. A. Frey, A. D. Hegeman and F. J. Ruzicka, Crit. Rev. Biochem. Mol.

Biol., 2008, 43, 63–88.69 J. W. Peters, R. K. Szilagyi, A. Naumov and T. Douglas, FEBS Lett.,

2006, 580, 363–367.70 S. E. McGlynn, S. S. Ruebush, A. Naumov, L. E. Nagy, A. Dubini,

P. W. King, J. B. Broderick, M. C. Posewitz and J. W. Peters, JBIC, J.Biol. Inorg. Chem., 2007, 12, 443–447.

71 S. E. McGlynn, E. M. Shepard, M. A. Winslow, A. V. Naumov, K. S.Duschene, M. C. Posewitz, W. E. Broderick, J. B. Broderick and J. W.Peters, FEBS Lett., 2008, 582, 2183–2187.

72 M. Schick, Diplomarbeit, Philipps Universitat Marburg, 2008.73 N. H. Sleep, A. Meibom, T. Fridriksson, R. G. Coleman and D. K.

Bird, Proc Natl. Acad. Sci. U. S. A., 2004, 101, 12818–12823.74 M. Schulte, D. Blake, T. Hoehler and T. Mccollom, Astrobiology, 2006,

6, 364–376.75 F. Tian, O. B. Toon, A. A. Pavlov and H. De Sterck, Science, 2005,

308, 1014–1017.76 W. Martin and M. J. Russell, Philos. Trans. R. Soc. London, Ser. B,

2007, 362, 1887–1925.77 J. A. Brandes, N. Z. Boctor, G. D. Cody, B. A. Cooper, R. M. Hazen

and H. S. Yoder, Jr., Nature, 1998, 395, 365–367.78 M. A. Schoonen and Y. Xu, Astrobiology, 2001, 1, 133–142.79 A. Smirnov, D. Hausner, R. Laffers, D. R. Strongin and M. A. A.

Schoonen, Geochem. Trans., 2008, 9.80 G. D. Cody, N. Z. Boctor, T. R. Filley, R. M. Hazen, J. H. Scott, A.

Sharma and H. S. Yoder, Science, 2000, 289, 1337–1340.81 C. Huber and G. Wachtershauser, Tetrahedron Lett., 2003, 44, 1695–

1697.82 C. Huber and G. Wachtershauser, Science, 1998, 281, 670–672.83 M. J. Russell and W. Martin, Trends Biochem. Sci., 2004, 29, 358–

363.84 M. J. Russell, Acta. Biotheor., 2007, 55, 133–179.

85 P. I. Volkers, C. A. Boyke, J. Z. Chen, T. B. Rauchfuss, C. M. Whaley,S. R. Wilson and H. J. Yao, Inorg. Chem., 2008, 47, 7002–7008.

86 T. B. Rauchfuss, S. M. Contakes, S. C. N. Hsu, M. A. Reynolds andS. R. Wilson, J. Am. Chem. Soc., 2001, 123, 6933–6934.

87 G. D. Cody, Annu. Rev. Earth Planet Sci., 2004, 32, 569–599.88 M. Ruiz-Bermejo, C. Menor-Salvan, S. Osuna-Esteban and S.

Veintemillas-Verdaguer, Origins Life Evol. Biosphere, 2007, 37, 507–521.

89 G. Wachtershauser, Chem. Biodivers., 2007, 4, 584–602.90 D. J. Berrisford, C. Bolm and K. B. Sharpless, Angew. Chem., 1995,

34, 1059–1070.91 M. A. A. Schoonen, Y. Xu and D. R. Strongin, J. Geochem. Explor.,

1998, 62, 201–215.92 G. A. Somorjai, Introduction to Surface Chemistry and Catalysis,

Wiley, New York, 1994.93 K. Andersson, M. Nyberg, H. Ogasawara, D. Nordlund, T.

Kendelewicz, C. S. Doyle, G. E. Brown, L. G. M. Pettersson andA. Nilsson, Phys. Rev. B, 2004, 70.

94 A. Stirling, M. Bernasconi and M. Parrinello, J. Chem. Phys., 2003,119, 4934–4939.

95 F. Bonomi, M. T. Werth and D. M. Kurtz, Inorg. Chem., 1985, 24,4331–4335.

96 M. Keller, E. Blochl, G. Wachtershauser and K. O. Stetter, Nature,1994, 368, 836–838.

97 E. Drobner, H. Huber, G. Wachtershauser, D. Rose and K. O. Stetter,Nature, 1990, 346, 742–744.

98 D. Hafenbradl, M. Keller, G. Wachtershauser and K. O. Stetter,Tetrahedron Lett., 1995, 36, 5179–5182.

99 E. W. Abel and B. C. Crosse, Organometal. Chem. Rev, 1967, 443–494.100 E. J. Milner-White and M. J. Russell, Origins Life Evol. Biosphere,

2005, 35, 19–27.101 E. J. Milner-White and M. J. Russell, Biology Direct, 2008, 3.102 A. K. Jones, B. R. Lichtenstein, A. Dutta, G. Gordon and P. L. Dutton,

J. Am. Chem. Soc., 2007, 129, 14844–14845.103 W. L. Delano, Delano Scientific, Palo Alto, CA, USA.

http://www.pymol.org, 2002.104 P. Bayliss, Am. Mineral., 1977, 62, 369–373.105 T. I. Doukov, T. M. Iverson, J. Seravalli, S. W. Ragsdale and C. L.

Drennan, Science, 2002, 298, 567–572.106 H. Dobbek, V. Svetlitchnyi, J. Liss and O. Meyer, J. Am. Chem. Soc.,

2004, 126, 5382–5387.

This journal is © The Royal Society of Chemistry 2009 Dalton Trans., 2009, 4274–4285 | 4285