α‐hydroxytropolones Inhibit RNase H1 Noncompetitively Despite Binding in the Active Site

Abstract

Human immunodeficiency virus (HIV) and Hepatitis B virus (HBV) ribonucleases H (RNase H) are type 1 RNases H that are promising drug targets because inhibiting their activity blocks viral replication. RNases H cleave RNA in RNA/DNA hybrids. Eukaryotic RNase H1 is an essential protein and probable off‐target enzyme for viral RNase H inhibitors. α‐hydroxytropolones (αHTs) comprise an anti‐RNase H inhibitor class that can inhibit the HIV, HBV, and human RNases H1. These compounds work by binding the RNase H active site by chelating the catalytic divalent metal cofactors. We hypothesized that a better understanding of RNase H1 inhibition will help development of compounds selective for the viral RNases H. To this end, we expressed and purified recombinant human RNase H1 and determined its inhibition mechanism(s) in steady‐state kinetics by two αHTs, 110 and 404 (Fig. 1). Inhibition was not competitive with a 12‐mer RNA/DNA substrate, but the turnover rate was reduced despite inhibitor binding to the active site (Fig. 2). 110 and 404displayed inhibition constants of 9 μM and 3 μM in saturating substrate concentrations, respectively, and these values were elevated 2‐3‐fold in very low substrate. Saturating 110and 404concentrations modestly reduced the apparent substrate binding constant (KM) from 90 nM to ~30 nM, while reducing the turnover rate (kcat = 0.17 s‐1) ~20‐fold. We found that 110enhanced affinity of RNase H1 for substrate by 4‐fold using a fluorescence polarization (FP) substrate binding assay with Ca2+ instead of Mg2+ to prevent RNA cleavage. 404, on the other hand, competed with substrate in binding assays, raising the substrate's KD~7‐fold from 24 nM without compound to ~150 nM. Induced fit docking studies in the Schrödinger suite suggest 110 binds to the active site metals as expected, while the substrate is still capable of binding via RNase H1’s high‐affinity auxiliary RNA/DNA hybrid binding domain (HBD) and substrate binding groove within the RNase H domain. 110 made favorable contacts with both enzyme and substrate, stabilizing an ESI complex. 404, on the other hand, occupies much of the substrate binding groove as well as the active site due to its larger structure. This would explain why 404 competes with substrate binding, while 110enhances substrate binding. The reason the KM decreased with 404 despite its competitive behavior in substrate binding assays is not clear. However, we hypothesize that 404locally competes with the substrate for the RNase H1 active site and the substrate binding groove within the RNase H domain without interfering with the HBD:substrate interface. This could lower the overall ES affinity. However, we speculate that if substrate release is slow relative to RNA hydrolysis and product release, the compound could behave uncompetitively in kinetics assays by inhibiting the breakdown of the ES complex through catalysis, only permitting enzyme‐substrate dissociation via the putatively slow substrate release pathway. Thus, these results illustrate that non‐competitive steady‐state kinetics may be observed without an allosteric inhibitor binding site. This may be common to active‐site inhibitors of enzymes with auxiliary substrate binding domains.