Deisgn and engineering of ligand-induced conformational constraints in HIV-1 envelope gp120.

摘要

We designed and produced a novel peptide triazole (PT) / HIV gp120 covalent conjugate. Gp120 is the main viral protein on the surface of HIV-1 virions. PTs are promising inhibitors of HIV-1 gp120 binding to its target receptors. In absence of a crystal structure of the PT / gp120 complex, we used molecular dynamics (MD) to generate a model of the peptide/gp120 complex. We used this model to design reactive variants of gp120 and the peptide. We implemented the design in the wet-lab and provided evidence that the resulting product showed the properties expected of the peptide/gp120 bound state.;We first studied the conformational flexibility mechanisms of available gp120 structures as potential targets for ligand docking experiments. We showed that in contrast to the activated state gp120 core which displays a bridging sheet mini-domain, the b12-bound conformation of gp120 core is flexible in absence of this mini-domain. Further we showed that removal of a disulfide stitch in the engineered structure of the latter state greatly enhances its flexibility. Hence this part of the work provided some insight as to where the moving part of the gp120 machinary might be located.;We next used temperature-accelerated molecular dynamics (TAMD) to further explore the conformational landscape of gp120 core. Starting from the activated state, we were able to see opening of the bridging sheet and exposure of conserved residues behind it which we saw were transiently exposed in the previous part. Using this technique, we could exit the deep energy minimum of the activated state and sample a diverse range of different conformations. Interestingly we were able to sample multiple conformations similar to the recently resolved F105-bound gp120 conformation. We were not able to sample conformations similar to the two engineered non-activated states of gp120 in complex with the b12 and b13 antibodies. These results provided us with useful hypotheses as to where the PTs may bind, which crystal conformation might be a more plausible target, and how PTs might prevent bridging sheet formation.;We used this information, in combination with past experimental data, to propose a binding site for PTs on gp120. We hypothesized that PTs bind closely to the F43 pocket on gp120 and prevent formation of the bridging sheet. We generated stably bound PT/gp120 complexes which displayed properties consistent with more recent experimental data. Furthermore, our results were in agreement with saturation transfer nuclear magnetic resonance (STD NMR) experiments performed by our collaborators. The generated model both could help us test concrete ideas to verify its validity and also be used to design our desired covalent conjugate.;In the last part of the work, we used the generated model to design reactive variants of PTs and gp120. We characterized both variants and showed they are functional. We then combined the two and showed that the peptide indeed covalently linked to gp120. Furthermore, the covalent complex displayed conformational properties which we expected of a PT/gp120 complex. The generated covalent complexes can be used as stabilized systems for crystallization or as antibody baits for identification of new antibodies.