The use of visible light to produce highly selective and potent drugs through photodynamic therapy (PDT) holds much potential in the treatment of cancer. PDT agents can be designed to follow an O2-dependent mechanism by producing highly reactive species such as 1O2 and/or an O2 independent mechanism through processes such as excited state electron transfer, covalent binding to DNA or photoinduced drug delivery. Ru(II)-polypyridyl and Rh2(II,II) complexes represent an important class of compounds that can be tailored to exhibit desired photophysical properties and photochemical reactivity by judicious selection of the ligand set. Complexes with relatively long-lived excited states and planar, intercalating ligands localize on the DNA strand and photocleave DNA through 1O2 production or guanine oxidation by the excited state of the chromophore. Photoinduced ligand substitution occurs through the population of triplet metal centered (3MC) excited states and facilitates covalent binding of the metal complex to DNA in a mode similar to cisplatin. Ligand photodissociation also provides a route to selective drug delivery. The ability to construct metal complexes with desired light absorbing and excited state properties by ligand variation enables the design of PDT agents that can potentially provide combination therapy from a single metal complex.

The population and reactivity of two low-lying excited states in [Ru(bpy)(2)(dppn)](2+) (bpy = 2,2'-bipyridine, dppn = benzo[i]dipyrido[3,2-a:2',3'-c]phenazine), a weakly emissive (3)MLCT state and a long-lived ligand-centered (3)pipi* state, lead to efficient photoinduced DNA damage. Irradiation with visible light results in nearly complete DNA cleavage within 30 s (lambda(irr) > or = 455 nm), likely from the combined action of guanine oxidation and the production of reactive oxygen species derived from (1)O(2).

The bimolecular quenching of the metal-to-ligand charge transfer (MLCT) excited state of Ru(phen)z-(dppz)2+ (phen = 1,lO-phenanthroline, dppz = dipyrido[3,2-a:2',3'-clphenazine) by proton transfer has been investigated in homogeneous acetonitrile solutions and in the presence of calf thymus DNA. In acetonitrile the monoexponential decay of the MLCT excited state emission of R~(phen)2(dppz)~+ is dynamically quenched by proton donors with pKa = 4.7-15.7. The emission lifetimes and quenching when the complex is bound to DNA have been measured for racemic mixtures, as well as A-and A-R~(phen)2(dppz)~+, and the values compared. In the presence of DNA the biexponential decay of the emission is quenched dynamically and three times slower than in acetonitrile when the quencher, in this case hydroquinone, is hydrophilic. Static quenching is observed in the presence of DNA when a hydrophobic proton donor, o-chlorophenol, is utilized. The static quenching with o-chlorophenol is shown to arise solely from the quenching of the long-lived component. These observations are explained in terms of two different modes of binding between the complex and DNA, as well as the different affinities for the aqueous medium of the quenchers.

CONSPECTUS Uncovering the factors that govern the electronic structure of Ru(II)–polypyridyl complexes is critical in designing new compounds for desired photochemical reactions, and strategies to tune excited states for ligand dissociation and 1O2 production are discussed herein. The generally accepted mechanism for photoinduced ligand dissociation proposes that population of the dissociative triplet ligand field (3LF) state proceeds through thermal population from the vibrationally cooled triplet metal-to-ligand charge transfer (3MLCT) state; however, temperature-dependent emission spectroscopy provides varied activation energies using the emission and ligand exchange quantum yields for [Ru(bpy)2(L)2]2+ (bpy = 2,2′-bipyridine; L = CH3CN or py). This suggests that population of the 3LF state proceeds from the vibrationally excited 3MLCT state. Because the quantum yield of ligand dissociation for nitriles is much more efficient than that for py, steric bulk was introduced into the ligand set to distort the pseudo-octahedral geometry and lower the energy of the 3LF state. The py dissociation quantum yield with 500 nm irradiation in a series of [Ru(tpy)(NN)(py)]2+ complexes (tpy = 2,2′:6′,2″-terpyridine; NN = bpy, 6,6′-dimethyl-2,2′-bipyridine (Me2bpy), 2,2′-biquinoline (biq)) increases by 2–3 orders of magnitude with the sterically bulky Me2bpy and biq ligands relative to bpy. Ultrafast transient absorption spectroscopy reveals population of the 3LF state within 3–7 ps when NN is bulky, and density functional theory calculations support stabilized 3LF states. Dual activity via ligand dissociation and 1O2 production can be achieved by careful selection of the ligand set to tune the excited-state dynamics. Incorporation of an extended π system in Ru(II) complexes such as [Ru(bpy)(dppn)(CH3CN)2]2+ (dppn = benzo[i]dipyrido[3,2-a:2′,3′-c]phenazine) and [Ru(tpy)(Me2dppn)(py)]2+ (Me2dppn = 3,6-dimethylbenzo[i]dipyrido[3,2-a:2′,3′-c]phenazine) introduces low-lying, long-lived dppn/Me2dppn 3ππ* excited states that generate 1O2. Similar to [Ru(bpy)2(CH3CN)2]2+, photodissociation of CH3CN occurs upon irradiation of [Ru(bpy)(dppn)(CH3CN)2]2+, although with lower efficiency because of the presence of the 3ππ* state. The steric bulk in [Ru(tpy)(Me2dppn)(py)]2+ is critical in facilitating the photoinduced py dissociation, as the analogous complex [Ru(tpy)(dppn)(py)]2+ produces 1O2 with near-unit efficiency. The ability to tune the relative energies of the excited states provides a means to design potentially more active drugs for photochemotherapy because the photorelease of drugs can be coupled to the therapeutic action of reactive oxygen species, effecting cell death via two different mechanisms. The lessons learned about tuning of the excited-state properties can be applied to the use of Ru(II)–polypyridyl compounds in a variety of applications, such as solar energy conversion, sensors and switches, and molecular machines.

The introduction of steric bulk to the bidentate ligand in [Ru(tpy)(bpy)(py)]2+ (1; tpy = 2,2′:2′,6″-terpyridine; bpy = 2,2′-bipyridine; py = pyridine) to provide [Ru(tpy)(Me2bpy)(py)]2+ (2; Me2bpy = 6,6′-dimethyl-2,2′-bipyridine) and [Ru(tpy)(biq)(py)]2+ (3; biq = 2,2′-biquinoline) facilitates photoinduced dissociation of pyridine with visible light. Upon irradiation of 2 and 3 in CH3CN (λirr = 500 nm), ligand exchange occurs to produce the corresponding [Ru(tpy)(NN)(NCCH3)]2+ (NN = Me2bpy, biq) complex with quantum yields, Φ500, of 0.16(1) and 0.033(1) for 2 and 3, respectively. These values represent an increase in efficiency of the reaction by 2–3 orders of magnitude as compared to that of 1, Φ500 < 0.0001, under similar experimental conditions. The photolysis of 2 and 3 in H2O with low energy light to produce [Ru(tpy)(NN)(OH2)]2+ (NN = Me2bpy, biq) also proceeds rapidly (λirr > 590 nm). Complexes 1–3 are stable in the dark in both CH3CN and H2O under similar experimental conditions. X-ray crystal structures and theoretical calculations highlight significant distortion of the planes of the bidentate ligands in 2 and 3 relative to that of 1. The crystallographic dihedral angles defined by the bidentate ligand, Me2bpy in 2 and biq in 3, and the tpy ligand were determined to be 67.87° and 61.89°, respectively, whereas only a small distortion from the octahedral geometry is observed between bpy and tpy in 1, 83.34°. The steric bulk afforded by Me2bpy and biq also result in major distortions of the pyridine ligand in 2 and 3, respectively, relative to 1, which are believed to weaken its σ-bonding and π-back-bonding to the metal and play a crucial role in the efficiency of the photoinduced ligand exchange. The ability of 2 and 3 to undergo ligand exchange with λirr > 590 nm makes them potential candidates to build photochemotherapeutic agents for the delivery of drugs with pyridine binding groups.