Organometallic Chemistry Tools for Building Biologically-Relevant Nanoscale Systems

Abstract

The recent emergence of organometallic chemistry for modification of biomolecular nanostructures has begun to re-write the long-standing assumption among practitioners that small-molecule organometallics are fundamentally incompatible with biological systems. This perspective sets out to clarify some of the existing misconceptions by focusing on the growing organometallic toolbox for biomolecular modification. Specifically, we highlight key organometallic transformations in constructing complex biologically-relevant systems on the nanomolecular scale, and the organometallic synthesis of hybrid nanomaterials composed of classical nanomaterial components combined with biologically relevant species. As research progresses, many of the challenges associated with applying organometallic chemistry in this context are rapidly being reassessed. Looking to the future, the growing utility of organometallic transformations will likely make them more ubiquitous in the construction and modification of biomolecular nanostructures.

Graphical Abstract

Introduction

Organometallic chemistry, defined simply as the chemistry of metal complexes containing a metal- or metalloid-carbon bond,¹ represents a powerful manifold of reagents and transformations harnessed by researchers working across a diverse range of fields including many sub-fields of chemistry. At first, many organometallic compounds were developed as mere curiosities, only later was their full potential in synthetic chemistry unleashed. A classical example is the Grignard reaction: introduced over a century ago, the reaction had very limited scope, however its use in modern chemistry is now ubiquitous.²

Whilst the use of metal catalysis to modify biologically-relevant systems is well-established, the application and apparent advantages of stoichiometric organometallic reagents in this field has only recently begun to emerge. Often, the use of organometallic species in a chemical transformation requires rigorous exclusion of oxygen, moisture, or both.³ For the modern synthetic chemist, these requirements may seem like a minor inconvenience considering the abundance of equipment and tools present in the chemistry laboratory. However, when applying organometallic chemistry to systems that contain biomolecules, these requirements can emerge as severe limitations considering that the practitioners who handle them may not have the ability to generate inert conditions. On the other hand, biomolecules normally necessitate handling in aqueous-containing media, which many traditional organometallic techniques do not tolerate. Therefore, on the surface, organometallic chemistry may seem as fundamentally incompatible with biomolecules. Recently, several groups have demonstrated that this long-standing dogma is not entirely true, as a number of reactive organometallic reagents exhibit the necessary stability under aqueous conditions to facilitate biomolecule functionalisation.⁴ The development of these novel strategies has provided a bridge between what was seemingly two disparate fields of chemistry, and the power of organometallic synthetic methods can now be applied towards the construction of post-synthetically modified biomolecules.

With the recent emergence of nanotechnology, researchers have developed a number of powerful concepts and approaches of controlling molecules and materials on the nanometer scale. Considering that the majority of biomolecules represent nature’s nanoscale building blocks, the ability to manipulate them in an abiotic fashion represents a very attractive direction for those interested in generating novel structures on the nanometer scale. Indeed, the evolution and functions of many biological systems heavily rely on natural nanomaterials and nanostructures. Some particularly elegant examples of these naturally occurring hierarchical nanoscale systems include kinesin traveling along a microtubule filament⁵ or the protein machinery used to replicate strands of DNA.⁶ Therefore, the challenge is to identify appropriate abiotic modifications of existing biological nanostructures, which can be performed with the same level of precision that nature can accomplish via recognition and post-translational modification. This perspective examines a recent emergence of organometallic-based transformations that have enabled the construction or modification of biologically-relevant systems on the nanomolecular scale. In particular, we will highlight the emerging toolbox of organometallic methods used to create or modify biologically-relevant nanostructures that builds upon existing organometallic strategies for modification of small-molecules. We will also discuss how these, and other, methods can be used to build hybrid nanostructures. Finally, we will highlight some key existing challenges and emerging opportunities in this area of research.

Organometallic Chemistry in Nature

Organometallic chemistry is often thought of as something invented by the synthetic chemist within the confines of the research laboratory, whereas the majority of metal-mediated transformations exhibited in nature occur by means of coordination chemistry.⁷ However, there exists a handful of known examples where nature harnesses organometallic chemistry to achieve modification on the nanomolecular scale. The seminal discovery was vitamin B₁₂, which is responsible for various methylation and carbon atom rearrangement reactions. These transformations occur at the Co-based center of the enzymatic co-factor, which facilitates making and breaking of Co-C bonds.⁸ Since the discovery of vitamin B₁₂ in 1955, there have been many more natural organometallic processes discovered, birthing the field of bioorganometallic chemistry.^4,9 A more recent discovery is that of the interstitial C^4– carbide ion central to the FeMo cluster in the quintessential nitrogenase family of enzymes (Figure 1). This field represents a number of chemical processes too vast for discussion within this perspective, the design principles of which have not yet been transferred to synthetic organometallic modifications of biomolecules. However, many of them serve as inspiration to the modern synthetic chemist when considering organometallic processes in the formation of nanomolecular structures.

Figure 1.

a) Crystallographic evidence of the placement of the FeMo-co cluster along with other relevant clusters and cofactors within the active site of nitrogenase enzyme as observed by x-ray diffraction (PDB: 3U7Q). Colors correspond as follows: yellow = sulfur; orange = iron, teal = molybdenum, black = carbon, red = oxygen, and purple = nitrogen. b) Proposed synthetic scheme of the formation of the FeMo-co cluster. In the first step, S-adenosyl-l-methionine (SAM) methylates the FeMo-co cluster, forming an organometallic intermediate. This is then progressively deprotonated and, through rearrangement of the cluster, the interstitial C^4- carbide ion is generated.

Organometallic Methods for the Modification of Nanostructures

Organometallic synthetic tools have recently started to gain a widespread interest for their use in the modification of biomolecules such as peptides, proteins, antibodies, and DNA. Although organometallic complexes are often considered highly oxygen- and water-sensitive, judicious selection of metal and the design of supporting ligand frameworks can have profound effects on the electronic profile, reactivity, and stability of such systems in biologically-relevant media. Control over these properties has provided a rich platform for exploring the use of organometallic complexes in biologically-relevant chemistry, and has enabled researchers to expand upon and apply existing knowledge and methodologies to novel bioconjugation approaches. Moreover, these organometallic bioconjugation methods are particularly applicable toward the creation and modification of biological nanostructures due to their rapid reaction kinetics and high chemoselectivity, both of which help to overcome the challenges concomitant with the sterically demanding and chemically complex reaction environments associated many of such systems. Here, we highlight several prominent examples of organometallic synthetic tools used to modify biological systems and applicable to researchers in the fields of chemical biology, biotechnology, and medicinal chemistry.

Broadly, the use of organometallic complexes for bioconjugation can be divided into two distinct categories: modification via metal catalysis, and modification through the use of stoichiometric metal complexes. Initially, we will focus our attention on transition metal catalysis, where the relevant organometallic complex is not necessarily isolated, but instead thought to be generated in situ within a catalytic cycle.

Cu-catalyzed Cycloaddition Chemistry (CuAAC “Click” Chemistry)

The copper(I)-catalyzed azide-alkyne [3+2]-dipolar cycloaddition (CuAAC) reaction represents one of the most widely used and versatile bioconjugation techniques in the currently available toolbox.^10–12 This chemical transformation, commonly referred to as a copper-catalyzed “click” reaction, has proven to be a successful bioconjugation tool due in part to its high chemoselectivity.¹³ Additionally, click reactions are typically irreversible as they result in the formation of covalently-linked five-membered triazole rings that have displayed stability under physiological conditions (Figure 2a).¹² Because both azide and alkyne are kinetically stable functional groups, the Cu(I) catalyst serves a critical role in substrate activation through the formation of a key organometallic Cu(I)-acetylide intermediate, which primes the alkyne for cycloaddition and thus accelerates the reaction rates.¹⁴ CuAAC click chemistry was first reported in 2002 by the Meldal¹⁵ and Sharpless¹⁶ groups, who both independently demonstrated its use in various settings. These initial reports demonstrated the regiospecificity, compatibility with aqueous media, broad substrate tolerance, and exceptionally mild reaction conditions of CuAAC click bioconjugation, which spurred immediate widespread interest in its applications relevant to medicinal chemistry, material science, proteomics, and synthetic biology, among others.^12,17–22 The utility of CuAAC click chemistry has been further realized for other bioconjugation applications including the in vitro and in vivo modification of biological nanostructures including proteins,^23–25 nucleic acids, polysaccharides,¹⁹ and lipid membranes,²⁶ and has been applied towards the functionalization of the outer and inner surfaces of complex virus particles.^10,27–30

Figure 2.

Selected examples of transition metal catalysis that have been applied to bioconjugation. a) Schematic of Cu-Catalyzed cycloaddition chemistry performed on peptides functionalized with alkynes or azides. b) Ring closing metathesis and olefin cross metathesis catalyzed by Grubbs I and Grubbs II complexes facilitates the formation of stapled peptides and allows the functionalization of alkene-labelled biomolecules. c) Copper-, nickel-, and palladium-catalyzed cross coupling reactions on biomolecules, including Chan-Lam-, Suzuki-Miyaura-, Sonogashira-, and Buchwald-Hartwig-type cross coupling reactions. Suzuki-Miyaura and Sonogashira reactions are facilitated by encoding the biomolecule with an unnatural aryl iodide or alkyne residue, whereas the Buchwald-Hartwig process is able to functionalize natural cysteine residues.

Metal-Catalyzed Olefin Metathesis

The olefin metathesis reaction has proven to be a cornerstone method for carbon-carbon bond formation in synthetic chemistry on account of its functional group tolerance, mild reaction conditions, and accessibility of commercially available and bench-stable transition-metal catalysts.^31,32 Since its discovery, metathesis chemistry has been used as a bioconjugation tool on peptides and proteins bearing olefinic side-chains, and has proven to be a powerful approach towards achieving selective synthetic transformations in complex biological environments (Figure 2b).^33–35 Olefin metathesis on biomolecular nanostructures was first introduced by the work of Ghadiri, in which two peptides containing homoallylglycine residues were linked through hydrogen-bond-promoted ring forming olefin metathesis in the presence of a Grubbs ruthenium catalyst³⁶ under mild reaction conditions.³⁷ Since this initial demonstration, olefin metathesis has been applied to the modification of biomolecules bearing unnatural olefinic amino acid residues, and has even been extended to heteroatom derivatives including S-allyl cysteine and Se-allyl selenocysteine substrates.^34,38 Early in this development, the Grubbs laboratory established that ring closing metathesis (RCM) can be used on peptide substrates to provide access to conformationally restricted peptide macrocycles that are constrained through an internal alkene-based “staple”.^39–41 The Verdine laboratory has extensively explored the use of RCM to prepare hydrocarbon cross-linked peptides, and demonstrated that the length and position of the cross-linking unit can have profound effects on the properties of the corresponding peptide, such as enhanced stability towards proteolytic degradation (Figure 3).^42,43 This form of covalent helix-stabilization has also been shown to facilitate peptide cell permeability and binding affinity to therapeutic targets.^35,44–47 This work ultimately set the stage for a commercial platform of peptide stapling that has produced several clinically relevant peptide-based macrocyclic nanostructures.⁴⁸

Figure 3.

a) Cartoon representation of an Axin-derived stapled peptide bound to a β-catenin (PDB ID 1QZ7). The staple is formed via an organometallic Ruthenium-catalyzed olefin cross-metathesis reaction. A scale bar is added to emphasize the nanoscale sizes of the system. b) A schematic expansion of the structure of the Axin-derived stapled peptide. The olefin “staple” is shown in red.

Ruthenium-catalyzed olefin cross-metathesis has also been developed as a powerful tool for the site-selective modification of complex proteins containing reactive olefin tags.³³ In large part, this work was pioneered by Bernardes and Davis, who identified the unique ability of allyl sulfide protein side chains to promote olefin cross-metathesis with monosaccharide allyl glycoside reaction partners in the presence of Hoveyda-Grubbs II catalyst.^49–51 This work was extended to the modification of allylselenocysteine⁵² and O-crotylserine⁵³ mutated protein scaffolds, which demonstrates a generality of this approach and contributes towards progress enabling Ru-catalyzed cross-metathesis to be used more routinely in protein bioconjugation.^{33,34,38,51,54}

An exciting development reported by Lu et al. is the use of Ru-mediated RCM and olefin cross-metathesis in DNA modification.⁵⁵ This approach could potentially serve as a viable tool for generating DNA-encoded libraries for hit identification and target validation, although improvements in reaction conditions such as lower [Ru] loading and reduction of the amount of MgCl₂ additive required would significantly improve the applicability of this method.

Metal Catalyzed Cross-Coupling Chemistry

Late transition-metal-mediated cross-coupling reactions rank among some of the most versatile transformations in organic synthesis and have recently gained widespread utility as instrumental tools for the functionalization of biomolecules.^56–59 Sonogashira, Heck, Buchwald-Hartwig, and Suzuki-Miyaura coupling reactions typically take place according to well-studied catalytic cycles, mostly proceeding via the elementary steps of oxidative addition, transmetalation, followed by C–C or C–X bond-forming reductive elimination from the metal center to generate the cross-coupled bioconjugate.⁶⁰ The first Suzuki-Miyaura cross coupling reaction that was successfully applied to a protein substrate was reported by Davis et al., in which a p-iodobenzyl cysteine mutant of Bacillus lentus was coupled with a variety of aryl- and vinylboronic acids using Pd(OAc)₂ in the presence of 2-amino-4,6-dihydroxypyrimidine (ADHP).⁶¹ This exemplary work overcame the barriers associated with previous reports employing palladium-mediated cross coupling reactions on biomolecules that suffered from low conversion, or required organic solvents and denaturing agents,^62–65 and demonstrated successful bioconjugation in buffered aqueous media at ambient temperature. This work laid the foundation for metal-mediated cross-coupling reactions on biomolecule substrates, including Sonogashira coupling on proteins in E. coli cells,⁶⁶ Suzuki-Miyaura cross coupling on cell surfaces,^67,68 tyrosine alkylation in bovine superoxide dismutase,⁶⁹ Buchwald-Hartwig-type cross-couplings on modified oligonucleotides and cysteine residues,^70–72 Tsuji-Trost allylation of cysteine residues,⁷³ among other examples. Work by the Ball group has applied copper- and nickel-catalyzed processes to biomolecules. Copper-catalyzed N-arylation and Chan-Lam-type couplings have been used for the N-arylation of N-terminal amines and amide nitrogens bearing an adjacent histidine residue, respectively.^74,75 The application of nickel-catalyzed processes also enabled the selective S-arylation of cysteine,⁷⁶ and histidine-directed N-terminal proline N-arylation.⁷⁷ Nickel-catalyzed bioconjugation methodologies have also been used to construct biologically-relevant nanostructures such as peptide macrocycles and protein-protein conjugates.^78,79 Most recently, Merck published an aqueous nickel-catalyzed cross-coupling methodology which enabled the arylation of cysteine residues on peptides, proteins, and antibodies.⁸⁰

Other Catalytic Organometallic Methods

Although representing the vast majority of metal-catalyzed processes for building and modifying biologically-relevant systems on the nanomolecular scale, there exists a handful of highly creative and elegant examples that do not use the above-discussed methods.

A particularly interesting example is the use of gold nanoparticles as both a template and chemical catalysis towards the building of hollow structures on the nanomolecular scale. The formation of such hollow nanostructures presents a significant challenge for chemists; however, their potential applicability to drug and gene delivery⁸¹ warrants their pursuit. Mirkin⁸² developed a particularly elegant example of the formation of hollow nanospheres or “nanopods” utilizing a gold nanoparticle as a templating agent (Figure 4). Utilizing the alkynophilicty of gold ions on the surface of a gold nanoparticle, terminal alkyne-bearing polymers were adsorbed onto the surface of a citrate stabilized gold nanoparticle. The nanoparticle itself then catalyzes the crosslinking of the polymer chains forming a densely crosslinked polymer shell around the nanoparticle core. After removal of the nanoparticle core through etching with potassium cyanide, the hollow polymer “nanopod” remains. Mirkin later applied gold nanoparticles as a template and catalyst towards the construction of polyvalent nucleic acid nanostructures (PNANs).⁸³ These structures were constructed by first synthesising oligonucleotides with a sequence of thymine bases functionalized with propargyl ether groups. Catalyzed by the nanoparticle template, the propargyl ether groups were crosslinked and, after template removal, PNANs were generated. Depending on the size of the template, PNANs between 5 nm and 30 nm could be constructed with remarkable cell permeability.

Figure 4.

Synthesis of a hollow nanopod via a gold nanoparticle template that is removed through a chemical etching process.

Palladium-catalyzed deprotection processes were applied to the activation of proteins in living mammalian cells by Chen and coworkers, modulating the function of a target protein.⁸⁴ By treating cells containing genetically encoded proteins with lysine residues masked with propargyl carbamate groups with catalytic quantities of Pd(II), free lysine was liberated. This allowed the in vitro activation of proteins within their native context, enabling probing of a protein’s lysine-dependent activity. Chen later adopted a similar approach to trigger the unmasking of genetically-encoded allene-caged tyrosine residues in vivo using palladium catalysis.⁸⁵ One of the major obstacles encountered in this work was the discrepancy between catalytic efficiency in vitro and in vivo, with the authors postulating that low cellular palladium uptake was likely the culprit. Some interesting approaches towards tackling the problem of cellular palladium uptake have been to employ peptides as a ligand, creating a catalytically-active palladium-containing biomolecular nanostructure. In one case, the palladium non-covalently staples the two residues together.⁸⁶ These bis-histidine(Pd) peptides showed excellent cellular uptake and once inside cells, could be used to promote the catalytic activation of fluorophores. Bradley and co-workers also used a peptide-ligated palladium complex to penetrate cells and perform catalysis to unmask a fluorophore.⁸⁷ These solutions are yet to be applied to the in vitro modification of proteins, but they represent exciting progress in the field of palladium-mediated biomolecule activation within cells.

The above methods for precise modification of nature’s nanostructures show the power of catalytic organometallic processes in modern chemistry. However, they are not without their compromises and challenges, and work remains ongoing to further optimize many forms of transition metal catalysis to ensure biocompatibility. Arguably one of the major drawbacks to using many of these methods is the requirement for installing or encoding unnatural reactive handles onto the biomolecules themselves: alkynes or azides for CuAAC chemistry,¹¹ alkenes for olefin metathesis chemistry,⁵⁴ and alkenes or even aryl halides to facilitate Suzuki-Miyaura or Buchwald-Hartwig-type chemistry.⁶² Whilst this requirement is amenable to certain applications, and is an intrinsic tradeoff of such biorthogonal techniques, the prevalent dogma that catalysis is inherently more desirable over a stoichiometric transformation may not be applicable in this case. Recent works in this area have shown that using stoichiometric organometallic reagents can enable practitioners with more powerful and practical bioconjugation techniques, specifically towards the functionalization of natural amino acid residues. The section below outlines the inception and recent advances in this area.

Stoichiometric Organometallic Complexes for Bioconjugation

Most of the palladium-catalyzed processes named above proceed via a similar catalytic cycle. Firstly, the active organometallic species is generated, most commonly by oxidative addition or C-H activation. This is followed by transmetalation and reductive elimination, then the cycle repeats.⁶⁰ The formation of the active metal complex necessary to begin the cross-coupling catalytic cycle is generally accepted as the rate-limiting step and often the first hurdle encountered when applying many cross-coupling reactions to biologically-relevant substrates.^88,89 Often, this step is only achieved at elevated temperatures or is only possible in solvents that are incompatible with many biologically derived substrates.⁹⁰ For catalytic reactions that do occur at low temperatures, irreversible poisoning of the catalytically active species by the substrate itself is common,⁹¹ leading to the requirement for high catalyst loading.⁹² In search of a solution to this problem, multiple approaches have been developed which pre-form and isolate the active organometallic species before reacting with a biomolecule. These methods use pre-formed oxidative addition or cyclometalated complexes as stoichiometric reagents to achieve bioconjugation, often using fewer equivalents of transition metal than their catalytic counterparts. These metal-mediated arylation reactions have emerged as a robust and efficient method of preparing bioconjugates through the formation of covalent carbon-carbon or carbon-heteroatom bonds.⁵⁸

Early examples of applying transition metal complexes to bioconjugation chemistry are sparse, the first example of which being the work of Francis and Tilley,⁹³ who reported electrophilic π-allyl palladium complexes capable of alkylating tyrosine residues. Later, the Myers group described a method for arylation of alkenes by a Heck-type coupling using Pd(II)-aryl reagents prepared via decarboxylative palladation of electron-rich benzoic acids.⁹⁴ Palladacycles derived from acetanilide have been shown to arylate biomolecules labelled with alkynes,^95,96 and in 2015, Leung and Wong demonstrated the reductive elimination from cyclometalated Au(III)-thiol complexes to arylate cysteine residues.⁹⁷

Palladium-Mediated Organometallic Chemistry

The above reactions all make use of organometallic reagents that were prepared using cyclometalation or insertion processes, where the oxidation state of the metal does not change upon formation of the organometallic species. However, a major leap in the technology arose when organometallic species prepared via oxidative addition processes were first applied to bioconjugation of natural amino acid residues. Unlike previously described organometallic techniques, preparation of these complexes using oxidative addition allows the chemist to make use of the vast library of aryl halides, facilitating bioconjugation using a mechanism akin to a Buchwald-Hartwig process (Figure 5).

Figure 5.

Modification of nucleophilic amino acid residues in biomolecular nanostructures with isolated stoichiometric organometallic complexes. Pictured are the general structures of oxidation addition complexes applied to bioconjugation including: Pd(II)-aryl reagents ligated with bulky RuPhos (ligand scope has since been expanded to include commercially-available ligands such as t-BuXPhos,⁹⁸ t-BuBrettPhos,⁹⁹ SPhos,¹⁰⁰ sSPhos,¹⁰¹ and bsSPhos¹⁰²), Au(III)-aryl reagents ligated with Me-DalPhos, and Pt(IV)F-aryl reagents ligated with bulky 2-[bis(adamant-1-yl)phosphino]phenoxide. Cyclometalated complexes include Au(III)-aryl reagents ligated with msen, and Pd(II)-aryl reagents ligated with bidentate XantPhos.

In 2015, Buchwald, Pentelute, and coworkers introduced an auxiliary-free, metal-assisted approach to chemoselective cysteine S-arylation using air-stable organometallic Pd(II) complexes supported by commercially available bulky biarylphosphine ligands.¹⁰³ The bioconjugation reactions proceed at ambient temperature, low micromolar concentrations of both biomolecule and bioconjugation reaction, and within a wide pH range (2–10), enabling the introduction of an array of substrates (e.g. drug molecule, affinity tag, fluorophores, bioconjugation handles) to biomolecules under mild conditions. The authors later showed that careful tuning of reaction conditions and chemical modification of the phosphine ligand could facilitate the S-arylation of cysteine residues in fully aqueous conditions,¹⁰¹ or the N-arylation of lysine residues.⁹⁸ In 2022, Frost and co-workers reported the use of cyclometalated Pd(II) complexes ligated with RuPhos and Xantphos, noting the facile preparation of cyclometalated complexes vis-à-vis the air-free synthesis of oxidative addition complexes.¹⁰⁴ Remarkably, both the oxidative addition and cyclometalated complexes display excellent stability towards long-term storage, with no loss of bioconjugation efficiency after months of storage under air in the fridge¹⁰³ or the benchtop,¹⁰⁴ respectively.

This stoichiometric Pd(II)-mediated approach has since been expanded to building and modifying a range of biomolecular nanostructures (Figure 6). Shortly after their seminal work, which included the preparation of peptide-drug (Figure 6a) and antibody-drug (Figure 6g) conjugates, Buchwald, Pentelute and co-workers expanded on their work by demonstrating divergent peptide macrocyclization of unprotected peptides (Figure 6d).¹⁰⁵ The methodology tolerated a wide variety of aryl cross-linkers and was used to produce a range of macrocyclic peptides. The physiochemical properties and binding affinities to HIV-1 C-terminal capsid protein were assessed, with the macrocyclised peptides demonstrating highly tunable physiochemical properties whilst maintaining binding affinity to the target protein. Radiochemical labelling of nanostructures was reported by Buchwald and Hooker in 2017, where they described a one-pot, two-step synthetic strategy to synthesize [¹¹C]cyano-containing peptides (Figure 6b).¹⁰⁶ In 2018, the work of Buchwald and Pentelute applied their well-established methodology to the cross-linking of biomolecules.⁹⁹ These bifunctional reagents combined the cysteine-selective palladium complexes with a lysine-selective phenyl carbamate reagent, allowing the stapling of cysteine to lysine residues both intra- and intermolecularly.

Figure 6.

Examples of biomolecular nanostructures constructed using Pd(II)-aryl complexes, including, but not limited to: a) peptide-drug conjugates¹⁰³ b) [¹¹C]-radiolabeled peptides¹⁰⁶ c) protein-carborane conjugates¹⁰⁰ d) stapled peptides^103,105 e) protein-polymer conjugates¹⁰² f) protein-protein heterodimers^107,108 g) antibody-drug conjugates¹⁰³

Further expanding the application of Pd(II) oxidative addition complexes, Buchwald and Pentelute focused their efforts on the synthesis of palladium-protein complexes, reporting two conceptually different approaches towards their synthesis. The first approach developed used a ‘ring walking’ mechanism, which is commonly harnessed in catalyst-transfer polymerization processes.¹⁰⁹ They achieved this by using Pd(II)-aryl halides as their bioconjugation substrates, which upon reductive elimination to form the C-S bond with a biomolecule,¹⁰⁷ produces a transient Pd(0) species which re-inserts into the intramolecular carbon-halogen bond. This rendered a protein with a functional organometallic oxidative addition complex appended to the biomolecule which could be isolated, and reacted further to produce protein heterodimers (Figure 6f). The second approach for synthesizing palladium-protein complexes used bifunctional reagents, with an NHS-ester linked to a Pd(II) oxidative addition complex.¹⁰⁸ A lysine-containing protein was reacted with the NHS-ester, producing the palladium-protein complex in excellent isolated yields. These palladium protein complexes were then reacted with a small-molecule dithiol, producing a protein homodimer, or a different cysteine-containing protein to form heterodimers. In 2022, Pd(II) oxidative addition complexes were applied to the incorporation of carboranes into biomolecules (Figure 6f) by linking cysteine-containing proteins with sodium borocaptate,¹⁰⁰ the preparation of palladium-peptide oxidative addition complexes from peptides containing 4-halophenylalanines,¹¹⁰ and the synthesis of protein-polyarene conjugates using catalyst transfer polymerization using a grafting from approach (Figure 6e).¹⁰² Further work has seen this technology applied to the preparation of transcription factor analogues for applications in biopharmaceuticals.^111,112 The examples above highlight a remarkable functional group tolerance of the organometallic Pd(II) reagents which enable the creation of sophisticated multifunctional biomolecular nanostructures.

Stoichiometric palladium complexes prepared via a cyclopalladation process have also been applied to address some of the challenges associated with activation of proteins using catalytic palladium, discussed above.^84,85 Thus, stoichiometric palladium complexes are desirable in such settings as they are highly tunable even compared to nano-palladium,¹¹³ or on-resin approaches,^114,115 as the phosphine ligand can be manipulated to impart desirable properties on the organometallic species, increasing cellular uptake.^116,117 Despite this, cyclopalladated Pd(II) complexes have not yet been applied to the uncaging of amino acid residues to modify biomolecular nanostructures, and it will be exciting to see how researchers harness these organometallics in this context.

Gold-Mediated Organometallic Chemistry

In 2018, Spokoyny, Maynard and coworkers introduced auxiliary-free aryl gold(III) reagents based on the hemilabile aminophosphine Me-DalPhos ligand framework,^118–120 which favors rapid reductive elimination of Ar-S_Cys groups from the Au(III) center to engage in stoichiometric cysteine arylation chemistry.¹²¹ This approach enabled the introduction of a drug molecule (Figure 7a), fluorophore, affinity tag, oligo(ethylene glycol), peptide staples (Figure 7c), and numerous other aryl substrates to the cysteine residues of biomolecule coupling partners. Au(III) aryl complexes for bioconjugation display high functional group tolerance, rapid reaction kinetics, and are compatible within a wide pH range. Unlike the Pd(II) oxidative addition complexes, the Au(III) reagents maintain competence within a wide pH range. and in the presence of a variety of buffers and solvents. The remarkable pH range tolerated by the Au(III)-mediated cysteine bioconjugation process compared to the analogous Pd(II) chemistry (and classical bioconjugation strategies utilizing various organic-based reagents) has been demonstrated experimentally: at pH 8.0, Au(III) oxidative addition complexes quantitatively arylate cysteine residues in less than five minutes.¹²¹ Conversely, Pd(II) complexes display significantly retarded reaction rates at pH 2.0, taking over seven hours to achieve 59% conversion to arylated cysteine.¹⁰³ Unpublished data suggests the formation of a dimeric species at low pH renders the Pd(II) oxidative addition complexes much less reactive towards nucleophiles. The utility of these compounds is also supported by their stability: much like the chemistries harnessing Pd(II) complexes, Au(III) reagents display no decrease in reaction efficiency after months of benchtop storage.^121,122

Figure 7.

Examples of biomolecular nanostructures constructed using Au(III)-aryl complexes including, but not limited to: a) peptide-drug conjugates¹²¹ b) PEGylation of proteins^123–125 c) stapled peptides¹²¹ d) highly constrained bicyclic peptides¹²⁶ e) [¹⁸F] radiolabelling of peptides¹²⁷ f) atomically-precise hybrid nanomaterials¹²⁸ g) complex biomolecular heterostructures¹²²

In a recent work,¹²² Spokoyny, Maynard, and Houk demonstrated that the Au(III)(Me-DalPhos) system was the fastest reported abiotic strategy for cysteine bioconjugation using a combination of competition and stopped-flow experiments. The fastest of these reagents demonstrated its second order rate constant of bioconjugation to be in excess of 16,000 M⁻¹s⁻¹, meaning that a reaction of 1 mm cysteine-containing biomolecule with 1 mm Au(III) reagent reaches 95% completion to bioconjugate in less than a third of a second. These outstanding reaction rates facilitated the first bioconjugation of a cysteine-containing protein at picomolar concentration.

The Au(III)-aryl system has been further applied to biomolecular nanomaterials (Figure 7): Spokoyny and Maynard demonstrated the utility of this cysteine arylation strategy through the conjugation of PEG polymers of 2, 5, and 10 kDa in size to a DARPin protein (Figure 7b).¹²³ This cysteine arylation strategy has also been applied to the introduction of radiolabeled species for the purposes of positron emission tomography (PET), an important biomedical technology (Figure 7 e).¹²⁹ The Murphy and Spokoyny groups used the Me-Dalphos ligand framework in a bioconjugation strategy to couple ¹⁸F containing aryl species to a biomolecule.¹²⁷ The rapid kinetics of the Au(III) bioconjugation makes this technology an attractive approach for [¹⁸F] radiolabeling given the short half-life of this isotope (110 minutes). Using a dodecaborate core functionalized at each vertex with an Au(III)-Ar species, the construction of atomically precise biologically-relevant nanomaterials was possible (Figure 7f).¹²⁸ Furthermore, constrained cyclic peptides have emerged as potential therapeutic targets and the use of organometallics to synthesize these species is of growing interest.¹³⁰ Mudd and coworkers demonstrated the use of a trimetallic Au(III)-aryl scaffold towards the construction of highly strained (bi)cyclic peptides. In this work, the rapid creation of a library of bicyclic peptides in good yield with high selectivity (up to 48%) was exhibited (Figure 7 d).¹²⁶

Alongside measuring the unmatched bioconjugation kinetics of Au(III) aryl species, the recent work described above demonstrated the ability to form complex nanomolecular heterostructures by careful control of the kinetics at the Au(III) center (Figure 7g). As a result, the bimetallic Au(III) species synthesized showed a selectivity of over 99:1 for the more reactive Au(III) center, producing conversions of up to 83%. The Au(III)-DalPhos system can also be applied towards the functionalization of polymer end groups and the synthesis of block-copolymers. Polymers prepared via ring opening polymerization, ring opening metathesis polymerization, and atom transfer radical polymerization were prepared with aryl iodide end groups, which were then converted into Au(III)-aryl reagents via oxidative addition of Au(I) into the C-I bond. These polymer-Au(III) reagents were then reacted with biologically relevant thiol-functionalized compounds such as fluorophores or biotin to functionalize their end groups, or reacted with thiol-functionalized polymers to produce various complex block copolymers.¹²⁴ This work for the first time demonstrates the use of organometallic chemistry in rapidly stitching together two large polymer components.

Other Stoichiometric Organometallic Reactions

The vast majority of stoichiometric organometallic reagents for bioconjugation are either palladium- or gold-based. However, recent works have shown platinum to be a promising candidate for expanding the organometallic toolbox for the construction of diverse biologically relevant nanostructures. The first example being the report of cysteine S-borylation using Pt(II)-carborane complexes by the Spokoyny group in 2021.¹³¹ These complexes were shown to efficiently borylate cysteine residues on peptides and proteins, furnishing biomolecular nanostructures bearing a carborane moiety. Notably, this report also contained the first observation of reductive elimination to form a B-S bond from Pt(II) complexes, thought to be typically less reactive than analogous Pt(IV) complexes. In 2022, the Vigalok group reported a more reactive, yet promiscuous, Pt(IV)-based system that is reactive to cysteine thiols, N-terminal amines, lysine amines, and tryptophan nitrogens.¹³² The works above highlight an opportunity to target stable organometallic reagents with reactivities not observed in Pd(II) systems.

The importance of robust chemistry in the field of bioconjugation is proven by the further development of organometallic complexes of palladium, platinum, and gold by other groups, who continue to apply them to new and emerging challenges.

Conclusions and Outlook

Nature’s successful incorporation and use of organometallic motifs within naturally occurring nanostructures, such as enzymes, remains a significant source of inspiration and challenge for modern chemists. Unlike nature, chemists have only had the past few centuries to develop methods of harnessing organometallics in a biological context. While many organometallic species are incompatible with the environments encountered within biological systems, careful engineering and design have enabled the use of certain small molecule organometallics for a number of applications. Furthermore, just like in the case of late transition metal cross-coupling chemistry, we believe that there currently exists known catalytic and stoichiometric transformations that would be amenable towards aqueous conditions with potential application in biomolecule modification. Alongside the cutting edge stoichiometric organometallic reagents that have emerged within the last decade, metal catalysis also continues to evolve and bring new opportunities to the area.^80,133

Historically, the majority of organometallic chemists overlook aqueous buffers as potential reaction media, for obvious stability reasons. For example, the original work by Bourissou et al. on Au(III) oxidative addition was conducted under rigorous air-free conditions, presuming high sensitivity of these species.¹²⁰ Despite these assumptions, Au(III) complexes were later shown to be stable in air and in aqueous buffers as well as in aqueous solutions in a wide pH range (1–14) and high salt concentrations.¹²¹ Clearly there exists an opportunity to reevaluate the stability profiles of the existing portfolio of reactive organometallic complexes, potentially uncovering novel reactivity with biomolecules. Along these lines, the majority of work so far has leveraged late transition metals which are more selective to soft ligands (e.g., sulfur) and more remains to be explored in terms of developing systems capable of modifying other residues in proteins.⁹⁸ Finally, arylation chemistry developed so far in the context of Cys bioconjugation with organometallic reagents shows extremely high reaction rates approaching some of the fastest “click” reactions.^103,134 This feature is made possible by eliminating the “slow” oxidative addition step from the reaction design ultimately showcasing that stochiometric organometallic transformations can carry a unique benefit over catalytic analogues. It remains to be seen how further ligand manipulation can affect the rates and selectivity of these transformations and what the ultimate rate ceiling for such reactions is. Importantly, an ability to conduct robust reactions with bimolecular rates faster than k_obs ≥ 10⁴ s⁻¹ should create unique opportunities to perform reactions at low concentrations amenable for applications in diagnostics and PET labeling,^38,135 which is a distinct advantage of the developed organometallic technology over more traditional methods. Furthermore, inherently fast reactions can allow one to overcome steric restrictions that are notorious for interfering with bioconjugation chemistry for fusing large biomacromolecules together.^{123,135–139} These transformations could also aid the development of new classes of atomically-precise nanomaterials, where dense functionalization can be complicated by steric repulsions at the nanomaterials surface.^140,141 Another untapped potential of these transformations is regioselectivity, where the unique properties of organometallic reagents can be used to target one reactive residue of a biomolecule, whilst leaving the same residue in a different location untouched. This level of selectivity would approach the level of sophistication normally associated with enzymes responsible for post-translational modifications,¹⁴² and are currently not present in the arsenal of man-made synthetic reagents.

Over the past several decades, organometallic chemists have established a powerful toolbox enabling the manipulation of ligands around metal centers and building selective recognition functions relevant to catalysis. Nevertheless, designing ligands to position biomolecular microenvironment recognition moieties that target specific amino acid residues in proteins remains in its infancy. For example, recent work by the Cohen group suggests that covalent organometallic inhibitors can target specific cysteine residues in the presence of other competing cysteine sites suggesting a largely untapped potential for engendering selectivity via this approach.^143,144 Recent work has also combined organometallic bioconjugation chemistry with more ‘traditional’ techniques, such as native chemical ligation, to enable the construction of highly functionalized and complex peptidic structures.^112,145,146 Furthermore, there exists many opportunities to leverage organometallic building blocks to create new nanoscale systems with emerging modalities such as proteolysis targeting chimeras (PROTACs),¹⁴⁷ conceptually new contrast agents and radiotherapies.¹⁴⁸

Main group organometallic chemistry is another largely underrepresented area of research when it comes to its relevance to hybrid nanoscale biomolecules. Several enticing opportunities exist, however, including incorporation of non-canonical residues within biomolecules that can alter their structure and function. For example, Liu and co-workers have recently realized the incorporation of an unnatural boron-containing heterocycles mimicking tryptophan residue.¹⁴⁹ Importantly, considering that natural biomolecular systems do not generally contain non-metallic Lewis acid sites, stable main group building blocks could potentially serve this unique purpose. These hybrid abiotic systems could open new opportunities for building novel catalysts, sensors and affinity labels.

Lastly, classical surface chemistry of hybrid nanomaterials has recently experienced a renaissance in terms of the development of new ligand systems allowing to stabilize and create new nanoscale features and objects. For example, an elegant work by Weiss and co-workers showed how icosahedral carborane thiols can be used to significantly stabilize and reduce surface defects on gold.^150,151 Similarly, N-heterocyclic carbene ligands have been demonstrated recently as an effective replacement for classical thiol ligands. Importantly, coordination of the C-atoms onto Au-based surfaces can result in a significant stabilization of the resulting system rendering it resistant towards harsh environmental conditions.^152,153 This concept has been further applied toward noble metal nanoparticles and clusters pioneered by Johnson^154–156 and Crudden¹⁵⁷ groups opening up new and exciting opportunities in using organometallic chemistry for building new nanoparticle architectures.

The rising importance of nanomaterials within the biological sciences has provided additional routes to combine the fields of biology and organometallic chemistry. Recently, a combination of nanomaterial and organometallic syntheses has resulted in the synthesis of novel and complex organometallic-based nanostructures which have been successfully applied to biological systems. We believe that this field will continue to grow as more and more creative methods of incorporating organometallic-based nanostructures into biological systems are discovered.