Exploring the Biological Activity of Rhodium Metalloinsertors

Abstract

Rhodium metalloinsertors are a unique family of potential anticancer agents that have been show to bind selectively to thermodynamically destabilized DNA base pair mismatches, abasic sites, and insertions/deletions (indels) in vitro . These metalloinsertors are also able to target mismatches in cells: metalloinsertors preferentially kill mismatch repair (MMR)-deficient cancer cells, which have a relative abundance of uncorrected DNA mismatches and indels, over MMR-proficient cells, which can repair these lesions. As such, these complexes have shown great promise as a potential treatment strategy for MMR-deficient cancers, which are often resistant to classic chemotherapies. Recently, a new class of metalloinsertors that bear a rhodium-oxygen bond was synthesized and shown to have remarkable potency and selectivity towards MMR-deficient cells. We have discovered many key differences between first generation metalloinsertors and these new metalloinsertors: (1) the MMR-selectivity of first generation metalloinsertors is heavily influenced by ancillary ligand bulk and lipophilicity, whereas the MMR-selectivity of metalloinsertors is strong regardless of ancillary ligand properties, (2) first generation metalloinsertors have toxicities in the micromolar range while metalloinsertors have toxicities in the nano molar range, and (3) first generation metalloinsertors can only bind DNA via the Δ-enantiomer while metalloinsertors can bind DNA via both the Δ- and Λ-enantiomers. Excitingly, the improved potency and selectivity of these Rh-O metalloinsertors brings them into a realm of clinical relevance. Here we examine the basis for the improved potency and selectivity of these new metalloinsertors. A family of six metalloinsertors that vary in the steric bulk and lipophilicity of an ancillary ligand was synthesized and characterized. Regardless of ancillary ligand identity, these metalloinsertors exhibit nanomolar or low-micromolar toxicities and all preferentially target MMR-deficient cancer cells over MMR-proficient cells. Notably, the off-target accumulation of these metalloinsertors in mitochondria is very low. This cellular distribution is in stark contrast with first generation metalloinsertors in which increased ligand lipophilicity led to increased mitochondrial uptake and ultimately non-selective mitochondrial-mediated cell death. We believe robust selectivity of these complexes is retained in part due to their low off-target accumulation in the mitochondria, which is further complemented by the low dosing requirements of these potent therapeutic agents. Our studies also suggest the high potency of these complexes may be due to a difference in DNA-binding abilities, which is supported by observed differences in which enantiomers can bind to DNA mismatches, differences in ligand buckling at physiological pH, and lipophilicity of the therapeutics, with metalloinsertors being dramatically more lipophilic than their first generation counterparts. To better understand the structural basis for this increased potency, crystallographic experiments are underway. A first generation metalloinsertor was previously crystallized with mismatched DNA, and the structure was pivotal in identifying the DNA binding mode of metalloinsertion. Using similar methods, we are working to produce a high-resolution crystal structure of an metalloinsertor with mismatched DNA in order to gain structural insights into the increased potency of these new complexes. A significant difference in DNA binding could result in different biological activation of proteins and overall higher potency of these metalloinsertors. Finally, as metalloinsertors are moved towards pre-clinical study, understanding their biological activity in diverse cell culture experiments is essential. We examined a metalloinsertor and the FDA approved chemotherapeutic agent cisplatin in 27 diverse colorectal cancer cell lines. The comparison of these drugs revealed the metalloinsertor to be on average five times more potent than cisplatin in this panel. The potency of the metalloinsertor in different cell lines spanned nearly three orders of magnitude and correlated with whole-cell uptake of rhodium. Additionally, a fluorescent metalloinsertor conjugate was used to quantify the number of lesions in DNA that could be targeted by metalloinsertion, a result that correlated well with the potency of a metalloinsertor across several cell lines, consistent with DNA mismatches as the effective biological target of the metalloinsertor. The experiments described within this thesis have allowed us to gain a better understanding of the biological activity of rhodium metalloinsertors. We have established that metalloinsertors are distinct from first generation metalloinsertors, and that these new metalloinsertors can serve as highly tunable, potent, and mismatch-selective anticancer agents. Furthermore, this potency is observed across diverse cell lines and has been shown to correlate with the number of genomic DNA lesions that can be bound by metalloinsertion. The unique biological activity of these complexes makes them ideal candidates for the treatment of MMR-deficient cancers, and the potency and tunability of metalloinsertors will allow for the development of previously unattainable diagnostic and therapeutic tools for MMR-deficiencies.