Open in a separate window Figure 3 Validation of enfuvirtide resistance mutants using a TZM-bl inhibition assay

Open in a separate window Figure 3 Validation of enfuvirtide resistance mutants using a TZM-bl inhibition assay. on the diverse mechanisms of enfuvirtide resistance and highlights the utility of using deep mutational scanning to comprehensively map potential drug resistance mutations. = 13 of 670 mutagenized sites, 9 of which are in this gp120 structure) are shown with spheres. Residues 1 to 18 of CCR5 are shown with sticks to indicate bridging sheet interactions. PDB:6MEO. There was also modest, but reproducible enrichment of mutations at other Env sites outside of the NHR domain. One such mutation was P76Y, which interacts with NHR sites L555 and L556 in the prefusion conformation (Figure 2B). Other potential resistance mutations occurred at sites 424C436 in the 20/21 strand of C4, as well as sites 119, 121, and 207 in the V1/V2 stem. While the V1/V2 stem is distant from 20/21 in the prefusion Env conformation, it shifts upon CD4 binding to form the 4-stranded bridging sheet along with the 20/21 strand, creating the portion of the co-receptor binding site that interacts with the N-terminus of CCR5 [42]. This cluster of potential resistance mutations extended to site 111 present below the bridging sheet in Envs CD4- and CCR5-bound state. To validate that our high-throughput mapping accurately identifies mutations that increase resistance to enfuvirtide in cell culture, we generated and tested individual BG505 Env pseudoviruses bearing single mutations for enfuvirtide sensitivity. We selected both previously characterized and novel resistance KIR2DL5B antibody mutations from each of the clusters of resistance mutations. The V549E and Q552R mutations increased resistance, shifting the IC50 by 150-fold (Figure 3). Other mutations that were modestly enriched (P76Y, C119R, K121P, and K207L) had little effect on IC50 but instead altered the slope and/or decreased the maximal inhibition plateau at the 8 g/mL enfuvirtide concentration used in Nisoldipine resistance profiling (Figure 3), suggesting these mutations may result in a subpopulation of resistant viruses. This agrees with prior work characterizing how enfuvirtide resistance can affect the inhibition Nisoldipine curve slope [43]. Notably, both these validation experiments and the resistance profiling itself were performed with a high concentration of infection enhancer (100 g/mL DEAE-dextran). When the assays were repeated with 10 g/mL DEAE-dextran, some of the resistance phenotypes were less prominent (Figure S3). Open in a separate window Figure 3 Validation of enfuvirtide resistance mutants using a TZM-bl inhibition assay. TZM-bl inhibition assays were performed in the presence of 100 g/mL DEAE-dextran, similar to the resistance profiling. (A) Inhibition curves are the average of two biological replicates, each performed in duplicate. (B) The IC50, the fold change in IC50 relative to wildtype (WT), and the maximum percent inhibition for each Nisoldipine mutant, determined from the fit four-parameter logistic curves. WT virus was run on each plate, and each mutant virus curve was compared to the plate internal WT control. The standard error of the mean is also shown. H330R, which was not enriched in the resistance profiling, was included as a control. In (A,B), mutant pseudoviruses are colored according to groups (black: WT; green: control mutant not expected to affect enfuvirtide sensitivity; blue: mutants in the V1/V2 Stem/co-receptor binding site; red: mutants in/near NHR binding site). 4. Discussion We Nisoldipine have quantified the effect of all single-amino-acid mutations to the extracellular and transmembrane ectodomain of BG505 Env on resistance to the fusion inhibitor enfuvirtide in cell culture. This map of resistance mutations included both previously characterized and numerous novel resistance mutations. The comprehensive aspect of these data defined clusters of mutations that likely alter enfuvirtide sensitivity via different mechanisms and at different steps during fusion. Even within the NHR, the selected mutations also help elucidate multiple potential mechanisms of resistance. While some NHR mutations may directly disrupt interactions with enfuvirtide (e.g., site 551), others appear to introduce positive charges or bulky amino acids at the center of the NHR coiled-coil (e.g., sites 548 and 552). These mutations may slightly alter the coiled-coil structure to disrupt enfuvirtide binding or favor the intramolecular binding of the CHR domain over binding to enfuvirtide. This hypothesis is supported by a study showing that the Q552R mutation results in an asymmetric six-helix bundle structure with the positively charged 552R residues oriented away from the coiled-coil.