Concluding remarks: Discussion on Natural and Artificial Enzymes including Synthetic Models

Abstract

This paper overviews the Final Remarks lecture delivered (by K.D.K.) at the end of this bioinorganic chemistry Faraday Discussion, held online for a worldwide audience, January 31 – February 3, 2022. This paper provides discussion in six sections: (1) the Introductory Lecture, from Ed Solomon, emphasized past and present uses of advanced spectroscopic methods and theoretical approaches to elucidate metalloenzyme active site structure, physical properties and function. (2) The Discussion topics are divided into groups having similar research themes, as seen from this author’s perspective. Emphasis is given to the non-heme iron group of articles with dioxygen activation research. (3) Small molecule activation (e.g., N₂, CO₂ and O₂ reduction; CH₄ or H₂O oxidation) is widely covered in this Discsussion; this authors’ view of the important reactions in bioinorganic chemistry is discussed. (4) We discuss current practice and vision for employing materials chemistry to widely apply to electroatalytic methods to effect small molecule activation (as above) to fulfill societal energy demands. 5) A discussion is given on the topic of synthetic models and the approach utilized therein. (6). New research on the authors’ ‘synthetic modeling is presented; preliminary results are given in the area of copper mediated peroxide activation.

1. Introduction

It was a great pleasure and honor to be invited to participate in this 2022 Faraday Discussion. This important scientific gathering was conceived and organized (via the Royal Society of Chemistry) by a long-time participant in this field, from the synthetic modelling side of research, Professor Rabindranath (Rabi) Mukherjee of the Indian Institute of Technology (IIT) Kanpur, India. The broader field is these days widely referred to as Bioinorganic Chemistry, or Metal Ions in Biology. It is worth noting that the last Faraday Discussion devoted to this field, and closely related or even the same specific topics (e.g., oxygen activation, artificial enzymes, hydrogenases) occurred in 2010, with the Discussion title ”Spectroscopy, Theory and Mechanism in Bioinorganic Chemistry”; for that, a themed collection of articles was published as Faraday Discuss., 2011, 148, pages 1–460.

As outlined on the RSC Web page for this Discussion, at https://www.rsc.org/events/detail/46673/natural-and-artificial-metalloenzymes-faraday-discussion, the main themes defined were:

1. Small molecule activation and synthetic analogues.

2. Techniques for studies of the kinetics and thermodynamics of metal-ligand exchange reactions.

3. Electron transfer, spectroscopy and theory

4. Natural & artificial enzymes and medicinal aspects

In fact, on the basis of the 20 short lectures presented, which were accompanied by considerable discussion (on-line, and recorded) and full manuscripts published in Faraday Discuss. (vide supra), topics 1, 3 and 4 were very well addressed, but topic 2 much less so. Bioinorganic chemistry is a very exciting field and it is much too broad in scope to cover in any one gathering of scientists in a single symposium format.

This Faraday Discussion included speakers, session chairpersons Poster presenters and attendees, from all over the globe, providing a challenge for an On-Line gathering. A good number of the presentations and discussions occurred at times which made it very difficult to attend (e.g., 4:00 to 7:00 am for K.D.K.). Thus, participants often had to rely on the published papers from speakers (which were made available prior to the actual On-Line event) or recordings of the presentations and discussion, to stay on top of the meeting.

The first scientific speaker, presenting the “Introductory Lecture”, was Edward Solomon of Stanford University, who provided a lecture entitled “Spiers Memorial Lecture: activating metal sites for biological electron transfer”.¹ This talk emphasized spectroscopic methods utilized in bioinorganic chemistry research, themed around elucidation of the structure and electronic structure (i.e.,, bonding and spectroscopic properties) of metalloenzyme active sites (Fig. 1). The article emphasized the finding that these often display distinctive spectroscopic characteristics which reflect coordination geometry and electronic structure (i.e., bonding) imposed by the protein matrix (and the specific ligation) that are critical to the enzyme’s reactivity.

Fig. 1.

Depiction of a multicopper oxidase (MCO) emphasizing the zoomed out active-site, the Type-1 (T1) His₂-Cys copper site which draws electrons from an external substrate and passes them on to the trinuclear Type-2 / Type 3 (Cu₂) center. The latter binds molecular oxygen and effects it’s reduction-protonation giving 2 H₂O. The point is to in the drawn out expanded view to depict ground state wave-functions emphasizing the T1 copper-cysteine sulfur atom bonding which is key to the electron-transfer process. Reproduced from Ref. 1 with permission from the Royal Society of Chemistry.

Solomon began his lecture with the very basics, the structure and bonding (using computational methods) spectroscopy (EPR, electronic absorption and charge-transfer) spectroscopy of square-planar [Cu^IICl₄]^2–. This led to the advanced subject matter,, electron-transfer sites in bioinorganic chemistry, especially the paradigmatic “blue” copper sites whose unique spectroscopic properties and Cu^II-cysteinate bonding were elucidated by Solomon and coworkers. However, the real emphasis of Solomon’s lecture and paper¹ was the delineation-description of the numerous physical and theoretical methods now available to interrogate metalloenzyme active sites. These include Cu L-edge or sulfur-ligand K-edge X-ray Absorption Spectroscopy (XAS), low-temperature magnetic circular dichroism (MCD) spectroscopy, resonant inelastic X-ray scattering (RIXS) and ultrafast XFEL spectroscopy. Solomon emphasizes the critical need to accompany all of these studies with theoretical-computational methods, such as Density Functional Theory (DFT). The material discussed in the Solomon lecture and paper, included discussion of perturbed or “entatic state” active site structures, resulting in distinctive spectroscopic properties contributing to unique reactivity. These concepts, accompanied by utilization of theoretical-computational methods, overlaps in one way or another with nearly all efforts in bioinorganic chemistry research, including the topics discussed by the other contributors to this Faraday Discussion, even including those whose studies/research involve synthetic chemistry.

2. Groupings and Survey of Lecture/Papers Presented

To a reasonable extent, one could group the lectures presented into various sub-topics, perhaps useful in order to have a view of the subject matter focused upon in this Discussion, given the incredibly broad scope of bioinorganic chemistry. The choice of groupings and the extent of the authors’ discussion of these lectures/papers is to a great extent dictated by the authors’ particular research interests and biases, those being especially the activation of molecular oxygen with copper and/or heme-iron coordination complexes. These of course lead to biases in the authors’ particular perspectives and preferences for discussion in this article.

These groupings can be labelled in the following manner: (a) Non-heme iron coordination complexes, elaborated upon below. (b) Copper bioinorganic chemistry, either synthetic modeling chemistry in relationship to dioxygen activation (T. D. P. Stack and coworkers,² Schindler and coworkers³ and our research (vide infra)). P. Walton and coworkers⁴ described aspects of the chemistry within a lytic polysaccharide monooxygenase. The contribution by G. Mugesh and coworkers⁵ was rather different, and focused on a new material which can deliver NO(g) as a drug, so possibly also coming under the label of medicinal chemistry. (c) Organometallic chemistry, with synthetic modeling of hydrogenase chemistry by V. Artero, C. Duboc and coworkers⁶, a DFT study of Ru or Os asymmetric transfer hydrogenation catalysis (J. A. Wolny, P. Sadler and coworkers)⁷ and application of a new/novel analytical physical technique (X-ray Absorption spectroelectrochemistry) to hydrogenases (O. Rudiger, S. DeBeer and coworkers).⁸ (d) Heme (bio)chemistry with contributions by A. Dey and coworkers⁹ on O₂-reduction and O. Shoji and coworkers¹⁰ on insights into previously unknown functions of a CYP fatty acid peroxygenase. (e) The topic of Artificial Enzymes consisted of very exciting computational contributions. The paper by A. Lledós, J.-D. Maréchal and coworkers¹¹ reported on multiscale model of an artificial enzyme, a POP-Rh₂ cyclopropanase, C. M. Jäger and coworkers¹² presented a computational perspective on iridium based alcohol dehydrogenases. There is considerable research activity in artificial enzymes; it is a very hot area and promising results for practical applications are apparent.

Two excellent Discussion presentations and accompanying papers which did not fall into the above categories, but which certainly belong in this bioinorganic Discussion are (i) the work by D. Parker and coworkers,¹³ and (ii) that by K. V. Lakshmi, G. W. Brudvig, and coworkers.¹⁴ The former study centered around lanthanide luminescence since such are involved in bioanalytical applications. Here, the scope and mechanisms involved in excited state quenching of 9-coordinate terbium and europium(III) were delineated. The latter paper involved the application of an advanced pulsed electron paramagnetic resonance (EPR) spectroscopic method, two-dimensional (2D) hyperfine sublevel correlation (HYSCORE) spectroscopy, to the study of isotopically labelled MeOH (as a probe molecule) binding to Mn₄Ca-oxo cluster in the S₂ state of the Oxygen Evolving Complex of a cyanobacterial Photosystem II (PSII).

With so many papers covering somewhat disparate subjects, we choose to here elaborate on that subtopic having the most speaker contributors, and possessing a number of highly exciting new results. This we label as the non-heme iron coordination complex area; five speakers gave lectures/papers in this area, with four emphasizing synthetic systems and reactivity studies. Sen Gupta and coworkers¹⁵ reported on the use of a later generation bTAML derived iron complex (TAML = tetra-amido macrocyclic ligand, that developed and championed by T. J. Collins)¹⁶ to carry out the efficient electrochemical oxidation of alcohols to aldehydes/ketones. The researchers were able to obtain considerable mechanistic information. They identified a high-valent Fe^V=O species as a reactive intermediate formed by PCET of a starting TAML aqua complex, which effects rate-determining (2^nd order) substrate C–H abstraction, in a process involving overall net hydride transfer. The use of electrochemistry is notable, as mentioned below.

D. Kumar, S. P. de Visser, C. V. Sastri and coworkers¹⁷ reported on much less common use of high-valent non-heme iron-oxo species. The Fe^IV=O complex with a quinolyl and pyridyl containing pentadentate ligand is able to oxidatively degrade halo-substituted phenols, which are nasty pollutants. Mechanistic studies revealed the occurrence of a phenol H-atom abstraction reaction and the authors suggest a new kind of species is involved in the overall reaction; here, a phenoxyl radical is proposed to bind to an iron(III)-hydroxide complex. Interestingly, o-quinones were found to be the major products following oxidative dehalogenation.

P. Comba and coworkers¹⁸ reported on the extraordinary reactivity, in fact record reactivity, of a new intermediate-spin (S = 1) Fe^IV=O complex employing a pentadentate bispidine ligand. For example, C-H abstraction on cyclohexane as a substrate occurs (in propionitrile solvent at –90 °C) at a rate comparable to non-heme iron enzymes. The authors complemented their work using ligand field theory to compare the properties of their bispidine complex with the well-known Fe^IV=O complex with tetramethylcyclam, toward providing information on complex spin-state gap size and its correlation to oxidative reactivity.

A most exciting advance was discussed and published by L. Que Jr. and coworkers as part of this Faraday Discussion,¹⁹ a breakthrough in the field of di-iron chemistry relevant to oxygen activating proteins possessing a μ-oxo diiron active site such as in the hydroxylase component of soluble methane monooxygenase (sMMOH). A powerful oxidant is generated derived from molecular oxygen, going through a diferric-peroxo intermediate P, which undergoes O–O cleavage to give a CH₄-oxidizing intermediate called Q which most spectroscopic studies have suggested to have a Fe^IV₂(μ-O)₂ “diamond core” structure. While the Que group was some time ago able to characterize a related and relevant high-valent (μ-O)₂Fe^IVFe^III complex, they only now (in 2021–2022) could describe the successful synthesis of a model compound with a diiron(IV) “diamond core”, and its X-ray structure (Fig. 2).

Fig. 2.

Representation of the X-ray structure of the [(TPA*)Fe^IV₂(μ-O)₂]⁴⁺ (TPA* = tris(3,5-dimethyl-4-methoxypyridyl-2-methyl)amine; crystallized with perchlorate as counter-anion). The Fe...Fe distance is 2.711(4) Å, which can be compared to other relevant diiron complexes and proteins.¹⁹ Reproduced from Ref. 19 with permission from the Royal Society of Chemistry.

The Faraday Discussion publication accompanying the Discussion short lecture delivered by L. Que Jr. is a very important paper, one which will be very highly cited over the next few years.¹⁹ It is very detailed and describes synthetic approaches/methods and it tabulates a variety of important structural (e.g., Fe...Fe distance) and spectroscopic data (e.g., UV-vis, resonance Raman, Mössbauer and EXAFS) for precursor complexes, the diiron(IV) product (Fig. 2), related high-valent or other diiron complexes, as well diiron proteins. Further, this paper gives reactivity comparisons which very interestingly show that this new complex [(TPA*)Fe^IV₂(μ-O)₂]⁴⁺ is amongst the least reactive of a related series of compounds, based on comparisons of oxidative reactivity with the substrate dihydroanthracene. The more highly reactive complexes, by many orders of magnitudes, have a partially or fully open core structure possessing a mono μ-oxo bridge. Many questions still remain, such as how to explain the reactivity variations observed in terms of structure, iron ion (and complex) spin-state and there are still differing interpretations of structural data or theoretical analyses of the actual enzyme Q intermediate. As already suggested, this work represents a significant advance and seminal contribution in bioinorganic oxidative chemistry.

The fifth paper in this group of non-heme iron papers is a strictly computational work on first-row metal ion high-valent oxo complexes by A. Sen and G. Rajaraman.²⁰ They use DFT and an ab initio method to explore the electronic structure of high-valent oxo complexes of cobalt in different geometries or spin states; when do Co^IV=O versus Co^III–O• descriptions apply, the former violating the Oxo-Wall concept long ago described by C. J. Ballhausen and H. B. Gray^{21, 22} for metal-oxo complexes, but only in tetragonal geometries. The authors present calculations indicate that Fe^IV=O and Mn^IV=O are the best descriptions for these species; they do exist and are pre-Oxo-Wall complexes. For cobalt, the authors studied different possible ligand architectures and varying cobalt ion spin states, and find that a mixing of the Co^IV=O <–> Co^III–O• electromer description is considerable in most geometries and also depends on the cobalt spin state. They conclude that discussion of the percent Co^III-O• character for certain complexes can be reasonable.

3. Small Molecule Activation and Key Reactions in (Bioinorganic) Chemistry

As mentioned, the organizers of the present Discussion enumerated small molecule activation as being one of the major themes, and they stated that abundant and mostly inert species include O₂, N₂, CO₂ and CH₄. Of course, molecular oxygen is not inert and lectures and papers on transition metal mediated oxygen activation are strongly represented in this Discussion. However, there are many other entities, neutral molecules or ions, which are biologically important and whose chemistry includes transition metal manipulation. These include CO and H₂O along with a good many nitrogen oxides, hydroxylamine (NH₂OH), N₂O, NO (nitrogen monoxide; nitric oxide), HNO (nitroxyl), nitrite (NO₂ –) and others including sulfur containing species. Clearly, the research represented by the presentations in this Discussion covered only a small fraction of possible bioinorganic chemistry involving small molecules.

We believe it is useful to consider a listing of perhaps the most important reactions in chemistry, those that are critically important in biological processes and which impact society in terms of energy conservation, energy needs and/or environmental demands. In fact, most of the listing below comes via the senior author of this article; however, the ideas here come from listening to lectures (e.g., at American Chemical Society national meetings) by one of the giants in inorganic and bioinorganic chemistry, Harry B. Gray, of Caltech:

1. O₂ → H₂O (fuel cell reaction)

2. H₂O → O₂ (water oxidation)

3. H₂O → H₂ + O₂ (water splitting)

4. CO₂ reduction → CO, formate, CH₄, C O ²⁻

5. 2 H⁺(aq) → H₂ (g) (cheap fuel production)

6. N₂ → 2 NH₃ (nitrogen fixation) (Recent: NH₃ as fuel)

7. CH₄ → MeOH (hydrocarbon functionalization)

Six of the reactions listed do occur in nature and require mediation or catalysis by metal ions such as in metalloenzyme active sites, enhancing broad interest in bioinorganic chemistry. The reduction of dioxygen to water (a) is an oxidase reaction and is coupled to some other process such as membrane proton translocation (which is then linked to bio-generation of adenosine triphosphate (ATP) from ADP) in cytochrome c oxidases, containing heme and copper ion centers.²³ Multicopper oxidases (4 copper’s per functional subunit)²⁴ also efficiently reduce O₂ to two waters coupling this reaction to substrate dehydrogenation (oxidation) chemistry.

Water oxidation is of course the reaction in photosynthesis so is mediated by the OEC Mn₄Ca-oxo cluster.^{25, 26} There is considerable current interest and activity by (bio)inorganic researchers to find and define catalysts for practical water oxidation (i.e., the oxygen evolution reaction; OER) (b); most of the research community employs electrocatalytic methods. Water splitting, akin to water electrolysis to give 2H₂ (g) + O₂ (g) (c) as a single process is not a reaction carried out by Nature, but there is interest in carrying out such chemistry via photocatalytic methods.

Carbon dioxide reduction to carbon monoxide (d) is carried out by protein CO dehydrogenases (CODHs).²⁷ Aerobic bacteria carry out this biochemistry using copper-molybdenum flavoenzymes having a Mo-[2Fe-2S]-FAD active site,²⁸ while anaerobes employ a Ni-Fe catalyst where the active site possesses a [Ni-3Fe-4S] cluster and another distal iron center.²⁹ Because of our societal environmental excess of CO₂, there are considerable efforts to develop electrocatalysts which can selectively carry out a reduction directly to methane or C₂ products such as ethylene (d). Treated copper-based material electrocatalysts seem to be particularly effective in facilitating CO₂ reduction with H₂(g).^{30, 31}

Two reduction reaction of great biological and societal importance are (e) production of H₂(g) as a fuel from water, the hydrogen evolution reaction (HER) and (f) dinitrogen reduction to give ammonia, as a useful precursor (e.g., as a fertilizer) for nitrogen needed for protein and nucleic acid biosynthesis. Hydrogenases efficiently interconvert protons and dihydrogen^{32, 33} and of course the latter is important as a fuel in energy applications. The key sulfur rich FeMo-cluster in nitrogenase enzymes found in cyanobacteria and rhizobacteria possesses a Fe₇MoS₉C stoichiometry; it has a somewhat complicated structure, where a central carbon atom connects to six (6) surrounding iron ions, where the latter are bridged by either two or three sulfide (S^2–) ions. In a highly energy demanding (via ATP) reaction, dinitrogen molecules react as follows:

$${\mathbf{N}}_{\mathbf{2}}+\mathbf{8}{\mathbf{H}}^{+}+\mathbf{8}{\mathbf{e}}^{-}+\mathbf{16}\mathbf{M}\mathbf{g}\mathbf{A}\mathbf{T}\mathbf{P}\to \mathbf{2}\mathbf{N}{\mathbf{H}}_{\mathbf{3}}+{\mathbf{H}}_{\mathbf{2}}+\mathbf{16}\mathbf{M}\mathbf{g}\mathbf{A}\mathbf{D}\mathbf{P}+\mathbf{16}{\mathbf{P}}_{\mathbf{i}}$$

Atmospheric dinitrogen conversion into ammonia gives us our primary source of bioavailable nitrogen. Ammonia oxidation is also a critical step in global nitrogen cycling, and research in ammonia oxidation has attracted great interest in clean energy and environmental pollution control.^34–37 NH₃ can be readily liquified and its oxidation (chemical or electrochemical) may yield dinitrogen plus H₂(g), thus NH₃ can be thought of as a hydrogen carrying fuel (f). Again illustrating the extremely broadly based subject matter in bioinorganic chemistry, none of the lectures/papers in this Discussion in any way pertained to nitrogenase or N₂-reduction.

Reaction (g) is the non-heme iron or copper catalyzed O₂ oxygenation of methane to give methanol, which is a well-known bioinorganic chemistry topic, falling under the oxygen-activation label. Thus, aerobic methanotrophic organisms are gram-negative bacteria that occur in a variety of environments and utilize methane as their sole carbon source. There is a soluble cytoplasmic enzyme form, sMMO, which possesses an active-site non-heme di-iron center. In this Faraday Discussion the lecture and paper from L. Que Jr., and coworkers detail aspects of both the protein and model chemistry efforts, see above).¹⁹ When environmental copper ion is present, the predominant form of methane monooxygenase is the particulate membrane-bound form, pMMO. The nature of the copper-containing active-site(s) has been controversial, and it is still not completely clear if the active site is comprised of a mono- or di-copper center.^{38, 39} More importantly, future studies must focus on mechanistic aspects of these enzymes and/or appropriate model compounds. Certainly, for copper ion mediated methane/hydrocarbon oxygenation chemistry, it is quite uncertain as to what sort of O₂-derived ligand-copper chemistry leading to reactive intermediates is required to hydroxylate methane, other hydrocarbons, or polysaccharides (vide infra).

4. Small Molecule Activation and Energy Concerns via Electrocatalysis, Materials / Solid-State Chemistry

As a practicing member of the synthetic bioinorganic chemistry community, of course I have a constant eye on the chemical literature reporting on the fast moving and considerable advances in the activation/manipulation of small molecules, in our case especially molecular oxygen. And it was with pleasure and admiration that I listened to the lecture presentations and read the accompanying papers and saw that these fit right in to the contemporary landscape of the field. However, in my normal perusing of prominent journals in chemistry, I have not been able to avoid noticing the huge number of high quality papers that involve small molecule manipulation but approach the issues and chemistry using materials/solid-state chemistry and even an engineering approach. Design or manipulation, including systematic variation, including use of theoretical-computational methods of evaluation, lead to new materials as “reagents” or catalysts for carrying out the relevant chemistry. Prominent among the approaches utilized include electrochemistry and electrocatalysis where the materials/solid-state chemists supply one or the other of the electrode materials.

The prominent use and discussions of the likely future employment of electrocatalysis led back to a key publication in Science magazine by T. F. Jaramillo and coworkers,⁴⁰ entitled “Combining theory and experiment in electrocatalysis: Insights into materials design” which was noted in our Final Remarks presentation (Fig. 3). Here, Seh et al. ⁴⁰ reviewed advancements in electrocatalyst development, which also necessitates complementary theoretical-computational research, which my enhance electrocatalysis of the water-splitting reaction (vide supra), that being the reverse of the reaction that underlies fuel cell chemistry. Related dioxygen, dinitrogen, and carbon dioxide reductions are also emphasized (Fig. 3). The authors presented a theoretical framework which highlights the need for catalyst design strategies that may stabilize distinctive reaction intermediates in a selective manner, relative to each other.

Fig. 3.

General research strategy for the development of electrode catalysts, via materials/solid-state chemical synthesis, to be utilized in electrocatalytic processes involving important small molecules such as H₂O, CO₂ and N₂. Reprinted from reference 40 with permission from The American Association for the Advancement of Science.

Seh et al.⁴⁰ also described how electrocatalysis research and implementation of such electrocatalytic processes can fit into the need for safekeeping in our energy future, given a rising global population, accompanying increased energy demands and impending climate change. Fig. 4 depicts conceivable sustainable production and transformation of fuels and chemicals such as hydrogen, hydrocarbons, oxygenates and ammonia. Water, carbon dioxide and dinitrogen are plentiful and available from earth’s atmosphere and may be transformed into the products mentioned above via electrocatalytic processes coupled to renewable energy (Fig. 4) if electrocatalysts with the requisite properties (e.g., efficiency and reaction selectivity) can be advanced.

Fig. 4.

Conceivable sustainable pathways for the production of essential fuels and chemicals. See the text. Reprinted from reference 40 with permission from The American Association for the Advancement of Science.

Seh et al.⁴⁰ has been heavily cited, especially including papers utilizing materials chemistry approaches to generate electrode materials for electrocatalysis, for example (i) promoting the use of “high-entropy materials” in catalysts,⁴¹ (ii) implementation of specialized proton shuttles to aid electrocatalytic reduction of N₂ to NH₃,⁴² or (iii), introduction of a nanoscale conductive copper-based metal-organic framework (MOF) layer supported over iron [Fe(OH)_x] nanoboxes to catalyze the HER.⁴³ Yet another recent publication emphasizes the use of “nature’s blueprint to expand catalysis with Earth-abundant metals”. It is notable that the discussion in this paper and composition of the coauthor pool is dominated by bioinorganic chemistry and bioinorganic research chemists.⁴⁴

A number of bioinorganic chemists participating in this Faraday Discussion employed electrocatalysis, notably (i) V. Artero, C. Duboc and coworker⁶ describing FeFe, FeNi or CoFe binuclear complexes which may or may not mediate the HER in an organic solvent and (ii) S. Sen Gupta and coworkers,¹⁵ in utilizing a non-heme iron complex to effect electrocatalytic alcohol oxidation. In fact, within the broader bioinorganic chemistry community, electrocatalysis is now a very popular research tool, especially for the HER, O₂-reduction, CO₂-reduction and water oxidation. A very new publication⁴⁵ presents a striking result, where a heme possessing synthetically derived tethered urea functions is a highly efficient photocatalyst for CO₂ reduction to CO; mechanistic insights suggest (i) the involvement of the urea functional groups to aid docking of the CO₂ substrate and (ii) transformation of an Fe^I-CO₂ adduct is involved in the rate determining reaction step.

5. Synthetic Models in Bioinorganic Chemistry & Synergy with Physical Methods-Spectroscopy-Theory

As per my Introduction, Edward I. Solomon and Anex Jose¹ emphasized elucidation of enzyme active sites by utilizing a variety of approaches: Spectroscopy, X-ray crystallography, Enzymology, Molecular Biology, and as per my own expertise and research activity, the synergy arising from the study of carefully designed and then well-defined synthetic models (also see below). Professors Solomon and Richard H. Holm from Harvard University have been instrumental in expressing these ideas, starting in 1996 with their conception and editorial organization of three thematic issues published in Chemical Reviews (ACS) (1996, 2004 and 2014). These publications featured a huge range of outstanding review articles from key researchers active in bioinorganic chemistry, presenting the latest results from many bioinorganic sub-topics, “Bioinorganic Enzymology” (Fig. 5). It should be kept in mind the this Faraday Discussion’s Solomon/Jose presentation & publication¹ emphasized utilization of a good many recently developed advanced physical and spectroscopic methods, now critical to the advancement of bioinorganic chemistry research.

Figure 5.

Subfield research contributing to the elucidation of a metalloenzyme’s active site structure and electronic structure, leading to an understanding of function. Results from computational-theoretical chemistry studies should be added as an important contribution. Reprinted (adapted) with permission from R. H. Holm and E. I. Solomon, Chem. Rev., 2014, 114, 3367–3368. Copyright 2014 American Chemical Society

Models in Bioinorganic Chemistry and in Science

As biased by our own research interests, we feel it important to provide some discussion concerning bioinorganic synthetic models. The building and consideration of models is critical in complementing research on the system of interest, e.g., the enzyme and its active site. “Models are the building blocks of science”.⁴⁶ “The model is the most basic element of the scientific method. Everything done in science is done with models. A model is any simplification, substitute or stand-in for what you are actually studying or trying to predict. Models are used because they are convenient substitutes”.⁴⁶ “A model is a substitute, but it is also similar to what it represents.” ⁴⁶ “Models are used in most if not all subfields of chemistry. With respect to biological research and the understanding of protein function, “Virtually everything that is understood about biological systems is derived from work with model systems”. ⁴⁷ And models are useful to many diverse fields such as Membrane Fusion,⁴⁸ or to understand “Lizard Tail Autotomy”.⁴⁹

Thus, as they pertain to synthetic models, biomimetic studies need not necessarily lead to the duplication of chemical or physical characteristics of the enzyme, but rather serve to sharpen or focus the scientific questions asked. By careful design and systematic variations (e.g., in the nature of the ligand employed), one might help identify those factors (e.g., donor atom type, metal redox potential, coordination number and/or geometry) contributing to the observed enzyme structure, spectroscopy and chemical mechanism of action. Chemical models can provide reference compounds for structural and coordination aspects, or provide insights into the plausibility of biochemically proposed ‘active’ intermediates, and their reaction competency. Bioinorganic models may also lead to compounds which mimic enzyme function and provide new reagents or catalysts for practical application.⁵⁰

The present authors would like to express some frustration and mention some comments from reviewers/referees received over the years, illustrating how many in the field just don’t understand concepts in bioinorganic chemistry (or any) modeling. (1) Reviewer: The studies were not performed in aqueous, physiologically relevant conditions, comparable to the natural reaction in the protein. This is a ridiculous comment; most metalloenzyme active sites are not hydrophilic, and may be highly hydrophobic. (2) Reviewer: Unfortunately, the results do not provide proof of the actual mechanism of action in the protein. By definition, the mechanism of action of any enzyme can only be absolutely proven on studies of the enzyme itself. Models have other purposes.

A classic example of a synthetic model having great impact in that subfield of dioxygen activation by copper and Cu-metalloproteins came in 1989 (Fig. 6).⁵¹ This was a breakthrough study. The binuclear dioxygen complex generated and characterized possessed structural features (i.e., a μ-η²:η²- (i.e., side-on) peroxo ligand binding, Cu...Cu distance and spectroscopic properties (UV-vis, resonance Raman, etc.)) which closely match those known for oxy-hemocyanin (arthropodal and molluscan O₂-carrier) and oxy-tyrosinase (monooxygenase). The matching of properties occur in spite of the fact that a highly non-biomimetic ligand was employed, a sterically hindered trispyrazolylborate anionic species (Fig. 6). This study in fact was published several years prior to confirmation of the oxy-hemocyanin active site structure determined by X-ray crystallography.⁵² The approach and results illustrate the synthetic model approach and how critically important results can come from such a study. Here in this 2022 Faraday Discussion, the successful generation and characterization of the [(TPA*)Fe^IV ₂(μ-O)₂ ]⁴⁺ model compound (vide supra)¹⁹ closely compares in importance and likely impact.

Fig. 6.

A classically known synthetic model from K. Fujisawa, N. Kitajima and coworkers. See text for further discussion.

6. New Synthetic Bioinorganic Chemistry Relevant to LPMO’s

As this paper’s authors are actively involved in synthetic bioinorganic chemistry relevant to dioxygen activation by copper, the senior author here chose to present new research during his Final Remarks lecture. The class of copper metalloproteins of interest/relevance are Lytic Polysaccharide Monooxygenase (LPMOs; Fig. 7).^53–60 The active site has a single copper ion, with three protein derived ligands, a terminal histidine forming a chelate via coordination of the amino terminus and an imidazole nitrogen atom (the “histidine brace”), along with a second histidine derived imidazole donor (Fig. 7). Recent research indicates that LPMOs possess both monooxygenase and peroxygenase activity, but the latter occurs at considerably faster rates. Thus, research aimed at providing insight to the peroxygenase reaction, via new Cu/H₂O₂ model chemistry, may provide interesting insights.

Fig. 7.

Depiction of one derivative form of the active site and polysaccharide oxidation reactions carried out by LPMOs. See the text for further discussion.

It has come to light that the classic Fenton reaction, the process where a reduced metal ion (Fe(II) or Cu(I)) reacts with hydrogen peroxide is not as simple as producing an oxidized metal ion (Fe(III) or Cu(II) plus hydroxide plus hydroxyl radical (•OH). For iron, depending on the ligand environment for Fe((II) and other reaction conditions (such as acidity), the product is usually a high-valent Fe^IV=O species.⁶¹ For copper ion, much less is known, and the reaction of a (ligand)Cu^I complex and H₂O₂ could lead to the following outcomes:⁶²

$${\text{LCu}}^{\text{I}}+{\text{H}}_2{\text{O}}_2\to {\text{LCu}}^{\text{II}}–\text{O}•+{\text{ H}}_{\text{2}}\text{O}$$ (1)

$${\text{LCu}}^{\text{I}}+{\text{H}}_{\text{2}}{\text{O}}_{\text{2}}\to {\text{LCu}}^{\text{II}}–\text{OH}+\text{•OH}$$ (2)

$${\text{LCu}}^{\text{I}}+{\text{H}}_{\text{2}}{\text{O}}_{\text{2}}\to {\text{LCu}}^{\text{III}}–{\left(\text{OH}\right)}_2$$ (3)

A combination of reactions might occur, such as reaction (2) being followed by Cu^II-bound hydroxide losing •H to the hydroxyl radical via HAT, giving the products in Eq. 1. In any case, a substrate R–H (with C–H bond) could undergo hydrogen-atom abstraction (HAT) by the cupryl species in Eq. (1), or by the freed hydroxyl radical in Eq. (2) or by the copper(III) species in Eq. (3), where a bound hydroxide could be a proton acceptor. In all cases, a substrate carbon radical would form but rebound of a copper bound •OH (formally) group would lead to hydroxylated product R-OH leaving behind copper(I).

The chemical system we’ve chosen to study is shown in Fig. 8. Here, the copper(I) complex [Cu^I(TMG₃-tren)]⁺, which otherwise has been shown to reversibly bind molecular oxygen ^{63, 64} also can under certain conditions be oxidized to give an alkoxide-copper(II) complex [Cu^II(TMG₃-tren)O]⁺, which has been crystallographically characterized.⁶⁵ In the present ongoing work, we have been using low-temperature UV-vis spectroscopic monitoring in organic solvents of the reaction of [Cu^I(TMG₃-tren)]⁺ with a dry source of hydrogen peroxide, n-(o-Tol₃POH₂O₂)₂, shown in Fig. 8.

Fig. 8.

The chemical system examined in this research wherein a known ligand-copper(I) complex is exposed to a ‘dry’ hydrogen peroxide source under cryogenic conditions in organic solvents. With excess H₂O₂, the reaction results in the hydroxylation of one of the exterior ligand methyl groups, which can lead to the copper(II)-alkoxide structure shown. See the text for further discussion.

Some of the experimental observations are: (i) use of an excess of the H₂O₂ reagent gives good yields of [Cu^II(TMG₃-tren)O]⁺, (ii) When the H₂O₂ reagent is added but the reaction quenched prior to alkoxide complex formation, stripping of copper and analysis of organics reveals that good yields of hydroxylated ligand are obtained; one of the ligand methyl groups has been converted to a -CH₂OH group, giving L-OH. (iii) When excess copper(I) complex is reacted with the H₂O₂ reagent, L-OH is still generated, in amounts correlating with the quantity of H₂O₂ added. These results are consistent with the nature of the reaction occurring as being [Cu^I(TMG₃-tren)]⁺ + H₂O₂ → [Cu^I(L-OH)]⁺, suggesting that one of the reactions (1), (2) or (3) (vide supra) took place, i.e., a ligand methyl group was attacked and •OH rebound may have ensued leaving copper(I). It may that we can show that LCu^I reacts with hydrogen peroxide to produce (L-OH)Cu^I plus H₂O, in line with literature proposals for catalytic peroxygenation. Many more experiments need to be carried out, including utilizing trapping reagents which possibly could show that •OH, for example, in produced as an intermediate.

Conclusions

It is clear that bioinorganic chemistry is alive and well and the 2022 Faraday Discussion on Natural and Artificial Enzymes including Synthetic Models provided contemporary and exciting contributions for a range of subtopics in the field. As mentioned, bioinorganic chemistry is a hugely broad discipline and it includes study of metals ions in biology which possesses contributions from spectroscopy (and as noted many very new advanced techniques and applications), X-ray crystallography, enzymology, molecular biology, computational/theoretical studies and synthetic modeling. The expectation is that bioinorganic chemistry will remain as a very vibrant field and there will be much more to come.