Application of Structure-based Methods to Analyze Resistance Mutations for Chemically Diverse Non-Nucleoside Reverse Transcriptase Inhibitors.

Abstract

Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are used in combination with antiretroviral therapy to suppress viral loads in HIV patients. The chemical design of NNRTIs has changed in recent years in response to resistance-associated mutations (RAMs) and resistance. NNRTIs are chemically diverse compounds that bind an allosteric site of HIV RT. Resistance- associated mutations (RAMs) identified in HIV patients are associated with NNRTI resistance. RAMs confer amino acid changes that alter both structural and physiochemical properties of the allosteric site. Ultimately, these changes reduce NNRTI affinity. Previously, we used a combination of computational and experimental methods to analyze and validate RAMs for 3 diarylpyrimidine (DAPY) NNRTIs. The objective of this study is to apply these methods to other chemically diverse, non- DAPY NNRTIs. We selected MIV-150 (experimental microbicide) and doravirine for this study. A computational and molecular modeling strategy was used to evaluate the effects of RAMs. Calculated changes in drug affinity and stability (ΔS + ΔA) were used to determine overall resistance levels: susceptible, low, intermediate, and high. The ΔS + ΔA values for K101P suggest that this mutation confers intermediate/high-level resistance to MIV-150, but remains susceptible to doravirine. Based on the determined resistance levels, we analyzed the models and used Molecular Dynamics (MD) to compare the interactions of MIV-150/doravirine with RT wild-type (WT) and RT (K101P). From MD, we found that key interactions were lost with RT (K101P), but were retained with doravirine. To experimentally validate our findings, we conducted a fluorescence-based reverse transcription assay for MIV-150 with RT (WT) and RT (K101P). IC50 values determined in assays showed a 101-fold change in potency for MIV-150, but essentially no change for doravirine. Our computational and experimental results are also consistent with antiviral data reported in the literature. We believe that this approach is effective for analyzing mutations to determine resistance profiles for chemically diverse NNRTIs in development.