Catalytic drugs may be envisioned as a future holy grail in medical therapy offering low-dose therapy and selectivity based on targeted design for specific ailments. Additional advantages are specificity and the anticipated reduction or elimination of side effects caused by stoichiometric therapeutic doses. Progress in the design of catalytic drugs is moving rapidly forward with recent advances in the analysis and characterization of metal compounds in vitro. Large numbers of publications and reviews of the latest progress clearly display research into catalytic metallodrugs as an emerging and significant field in bioinorganic chemistry [1, 2, 3, 4, 5, 6, 7, 8, 9, ∗10, ∗∗11, ∗∗12].

Catalytic metal-based drugs encompass many different types of compounds and approaches to achieve different targets. Catalysis can be very successful targeting selected pathways exhibiting accelerated proliferation in cancer cells to achieve selective apoptosis and less cytotoxicity with increased therapeutic efficacy [1,8,13, 14, 15, 16, 17, ∗∗18, ∗∗19, ∗20, 21, 22, 23, 24, 25]. Approaches describing prodrugs delivery into cells, and their activation by various bond cleavage strategies have been reviewed recently, as well as advances in synthetic metallohydrolases [10,11].

Metal compounds are also studied in vitro in the hope of contributing toward numerous noncytotoxic therapeutic treatments (metabolic disorders, deficient metabolite synthesis), and to employ abiotic reactions for defenses against microbial, viral, diabetic, and neurodegenerative diseases, and to enhance imaging or diagnostic approaches, and even detoxification of poisons. Absence of cytotoxicity is an important aspect of these therapeutic approaches. Artificial enzymes have been reported to successfully achieve catalytic abiotic reactions [6,7,10,∗∗26, 27, 28, ∗∗29, 30, ∗∗31, ∗∗32, ∗∗33]. Sulfur-rich complexes modified from the amino terminal Cu Ni binding motif (ATCUN) motif were reported to show SOD activity, and a sulfur-rich molybdenum cluster proved able to mimic the rhodanese enzyme function by neutralizing poisoning by cyanide in vivo.

Artificial metalloenzymes may be employed as catalytic functional models to catalyze abiotic reactions [26]. In creating artificial metalloenzymes, two main approaches are preferred where either a) a homogeneous catalyst is exchanged with the cofactor in a natural metalloenzyme or b) a metal complex is covalently or supramolecularly anchored to a protein. A third emerging approach employs a site-specific mutagenesis in a protein with canonical amino acids to obtain residues in suitable geometries for metal coordination.

Compartmentalization may solve challenges associated with the metabolism of artificial metallocofactors by biomolecules en route to their target protein. This has been demonstrated in E. coli by using its periplasm to isolate the metallocofactors from potential reductants in cells, although lack of periplasm in human cells inspired further studies [26]. A successful example in human cells used the wild-type A2A adenosine receptor engineered to covalently bind a Cu(II) catalyst on the cell surface (Scheme 1a) [26]. The receptor was embedded in the cytoplasmic membrane of the cells to avoid metabolization of the complex. The resulting cells enantioselectively catalyzed the abiotic Diels–Alder cycloaddition reaction of cyclopentadiene and azachalcone. Interestingly, on the cell surface, the turnover number (TON) for the reaction was twice that obtained for the catalyzed reaction using the purified artificial enzyme.

Lower TON values for reactions in cells or bacteria compared to purified artificial enzymes are expected even when favorable kinetics are in place. The copper-catalyzed azide–alkyne cycloaddition (CuACC) reaction was successful in cells largely as a result of kinetic competency, as recently demonstrated for a metathesis reaction [10,28,34]. The promise of this approach lies in the in vivo preparation of receptor-based artificial metalloenzymes for the catalysis of reactions exogenous to the human metabolism [26]. An innovative application of this concept was demonstrated (Scheme 1b) by the creation of an artificial enzyme to mediate bioorthogonal prodrug activation and cell-to-cell communication between cancer cells and natural killer cells, resulting in activation of an ADCC effect. The artificial enzyme, not only eliminated the primary tumor but also prevented tumor metastasis efficiently because of immune activation [29].

A clever way to avoid the extraction/deactivation of abiotic reactants by biomolecules includes the synthesis of aromatic N-heterocyclic drugs with antitumor properties [30,35]. A noncytotoxic gold complex bound in the hydrophobic pocket of bovine serum albumin (BSA) was used as a catalytic metal trigger for an abiotic reaction. The negative charge on the albumin surface protected the Au center, and it activated a prodrug with a TON of over 200 (Scheme 1c) [10,30].

Photocatalytic conversion of aryl azides into nitrene intermediates was successful in living cells using Ru(bipy)2+. Adjustment of the light wavelength was directly related to the reaction yields. The complex proved stable toward common reductants in cells, including glutathione, BSA, and ascorbate giving modest yields without displaying quantifiable cytotoxicity [31].

Quercetin-2,4-dioxygenase is a copper/nickel dependent enzyme that catalyzes incorporation of both atoms of molecular oxygen into flavonols [32,36]. Quercetinases are known to accommodate a few different metal ions, including Ni(II), Mn(II), Fe(II/III), Cu(I/II), and Co(II) [37]. The promising utility of artificial enzymes for abiotic reactions is illustrated in recent examples where an artificial enzyme or a small molecule mimics dioxygenase reactivity [30, ∗∗31, ∗∗32,40]. Artificial enzyme synthesis by engineering non-native metal active sites into proteins using canonical amino acids offers many advantages [38]. An artificial Cu(II) Diels–Alderase capable of catalyzing an enantioselective Diels–Alder reaction [39] demonstrated a successful example of this approach where a barrel-shaped protein, imidazole glycerol phosphate synthase (tHisF), was engineered to demonstrate flavonol dioxygenase reactivity (Scheme 1d) [32]. A variant with a Glu/His/His motif was reported with the crystal structures of the Zn(II) and Ni(II) revealing both ions binding histidine residues, albeit at distinct sites with differing geometries. Both differ from the natural quercetinase-2,4-dioxygenase, demonstrating the adaptability of tHisF. The Ni(II) version catalyzes the oxidative cleavage of quercetin and myricetin, providing an unprecedented example of an artificial metalloprotein with quercetinase activity [32]. The Ni(II):tHisFEHH activity was found to be lower than the reported activity for the natural Ni(II)-QueDFLA that was related to the coordination environment for the Ni(II) center, which has a fac-N2O donor geometry (Scheme 1d) compared to the fac-N3 environment of the natural Ni(II)-QueDFLA enzyme [32]. The artificial enzyme beautifully demonstrated dioxygenase reactivity as a proof of concept and showed selectivity towards the more electron rich substrates offered. A recent review further discusses abiotic Diels–Alder reactions focusing on RNA-ases [33]. A study of a small molecule mimic of copper-quercetinase with the desired octahedral fac-N3 donor set concluded that despite the favorable binding constant of Ni(II), the reaction with Co(II) was significantly faster and preferred a less electron-rich substrate [40].

Oxidative stress has been implicated in common kidney diseases, including glomerulosclerosis and tubulointerstitial fibrosis [41]. The accumulative symptoms are inflammation and build-up of deposits in tissue that is associated with specific SOD deficiency. A catalytic antioxidant treatment may be a promising therapeutic strategy to reduce reactive oxygen species resulting from SOD/catalase dysfunction [41, 42, 43]. The types of SOD models vary greatly, although they share commonalities of oxygen and nitrogen atom donor sets at the metal (Scheme 2a)). Advances and new techniques for characterizing SOD mimics in biological environments were reviewed recently [8,∗∗44, 45, 46]. Mn–1P complexes exhibited improved anti-inflammatory properties and antioxidant abilities along with kinetic inertness in biological media (Scheme 2b) [47]. These compounds show comparatively outstanding biocompatibility and stability under physiological conditions relative to many reported SOD mimics [48]. Peptidyl copper complexes demonstrating SOD-activity were reported displaying antioxidant properties [49,50]. A second generation of the peptidyl copper complex mimics a dinuclear catalase comprised of peptidyl units connected by a linker (Scheme 2b). The linker chosen was shown to influence both catalytic activity and physical properties.

Emerging new models include sulfur-rich complexes of Ni(II) and Cu(II) displaying catalytic SOD abilities with cysteine or novel sulfur donor atoms. A binuclear sulfur-rich Cu(II) complex prepared from tridentate bis(propylthiouracil) (PTU) sulfur-linked ligands, [Cu2(ptu-S-ptu)2] (Scheme 2c), was shown to be capable of peroxide and superoxide dismutase anion activity [51,19,54]. The motivation for increasing the complex sulfur content was correlated with the reported antioxidant properties of trisulfides. Insights gained into the complex reactivity by comparison with a family of other sulfur-containing Cu(II) metal complexes showed that the octahedral geometry negatively affected the catalytic reaction rates. The di- and tri-sulfur bridges of octahedral, or planar, geometry complexes were less active compared to a single sulfur bridge in a mononuclear Cu(II) complex in a square pyramidal geometry [51].

The ATCUN motif has been a topic of research since the infancy of catalytic metallodrug research [13,52,53]. Ni and Cu ATCUN motifs have been used for RNA cleavage, ROS formation, enzyme inhibition, redox-inflicted damage, and even for antimicrobial efficacy [13,52]. Just recently a sulfur-rich ATCUN motif with Ni(II), producing a catalytic antioxidant with N2S2 and N3S donor sets (Scheme 2c) was reported as a mimic of NiSOD [19,54,55]. A promising TON of 80 was reported at physiological pH, although the thiolate donors exhibited oxygen sensitivity, leading to their oxidation and potential deactivation [19,54].

Poisoning by cyanide is caused by the halting of oxidative phosphorylation when the Fe(III) center in heme a3 in cytochrome c oxidase binds cyanide. Cyanide poisoning antidotes as well as chemical and clinical aspects were recently covered (Scheme 3) [56, 57, 58, 59, 60].

A long standing supplemental treatment for cyanide poisoning is coadministration of thiosulfate [61,62]. An approach to increase bioavailability and increase cyanide detoxification efficiency of the rhodanese enzyme involved liposome encapsulation of the enzyme along with highly efficient sulfur donors [63,64]. Recent efforts aim to introduce encapsulated methemoglobin as a potential treatment [65]. Study of the encapsulated methemoglobin was reported showing physicochemical stability and efficacy as a cyanide antidote [65,66]. It should be noted that methemoglobin is a transport protein, and that the binding of cyanide to methemoglobin is not cooperative [67]. Despite ongoing research on several cobalt compounds as potential antidotes (Scheme 3b), only one approach using a molybdenum complex has been explored as a catalytic treatment (Scheme 3c) [56,68,69]. The catalytic approach neutralizes cyanide in vivo by mimicking the rhodanese enzyme functionality outside the liver. The compound can enter cells and mop up cyanide in many different tissues [70,71].

One reported catalytic approach employs a dinuclear bioinspired molybdenum complex (Scheme 3c) [71]. The [Mo2O2(μ-S)2]2+ core of this complex was studied as a functional model for xanthine oxidase (XO) [72]. It acts both as a sulfur donor and a prodrug with an unusually polar sulfide ligand that readily disproportionates into S2− and an S atom easily forming a MoS moiety [73,74]. The catalytic mechanism is initiated by a reaction of the compound with cyanide to form thiocyanate, followed by catalysis by the active species of the reaction between cyanide and a sulfur donor [56,58,71].

Several compounds of the type [(L)Mo2O2(μ-S)2(S2)]—have been reported to exhibit low cytotoxicity, and stability in air, water, and acidic solutions [71,75]. The reported TON is a modest 8 using a high catalyst loading of 10% [71]. Optimization of the TON has not been reported, although eventual deactivation was reported to take place via dimerization to form the tetranuclear species [Mo4O4S6(CN)4]4- [76]. The thiosulfate, sulfite, or thiocyanate formed in the reaction do not deactivate the complex. Investigation of the biological and physiological effects of the compounds demonstrated their ability to enter cells and portioning into the cytosol, mitochondria, and nucleus. Cell volume regulation was found unperturbed, and negligible interaction with DNA was reported [70].