Unravelling the HIV-1 envelope determinants of macrophage tropism and resistance to the CCR5 antagonist Maraviroc
thesisposted on 07.02.2017, 23:55 authored by Salimi, Hamid
Thesis abstract HIV-1 macrophage tropism (M-tropism) evolves over time, with the emergence of highly M-tropic variants in late disease, suggesting a role for macrophages in HIV-1 pathogenesis. Additionally, infected macrophages are important viral reservoirs acting as a barrier to curing HIV-1 infection. M-tropism has been associated with neurotropism, a viral feature that leads to a myriad of neurological disorders in HIV-1 patients such as HIV-associated dementia (HAD). Despite combination antiretroviral therapy (cART), milder forms of neurological disorders continue to persist imposing a substantial personal, economic and societal burden. Neurotropic HIV-1 variants are highly M-tropic, but the molecular mechanisms underlying efficient macrophage entry are incompletely understood. In chapters 3 and 4, the genetic, structure and functional properties were compared between a panel of HIV-1 R5 Envs that were isolated from the brain and lymph nodes of three AIDS patients. Brain derived Envs were first shown to be highly M-tropic, whereas those isolated from lymph nodes were non-M-tropic. Highly M-tropic brain derived Envs displayed reduced CD4 dependence, which was associated with increased sensitivity to neutralization by sCD4 and b12. Additionally, brain derived Envs showed an altered mechanism of CCR5 engagement. In majority of cases, this was associated with an increased exposure of the CD4-induced epitope in gp120 and also with an enhanced dependence on the CCR5 N-terminus and on certain charged residues in the CCR5 ECLs. An increased dependence on the CCR5 N-terminus was further confirmed by structural modeling studies, where brain Envs relied on more atomic contacts with the CCR5 N-terminus compared to lymph node Envs, to mediate entry. Increased exposure of CD4-induced epitopes in brain derived Envs was associated with amino acid variations within the V3 region of gp120, such that introduction of the V3 loop from a lymph node Env (M2L2) into a brain Env (M2B2) led to a dramatic decrease in the exposure of the CD4-induced epitope, however it had no impact on the Env’s ability to enter macrophages. Rather, macrophage tropism was modulated by mutations within the V2 and V4 regions of gp120. Interestingly, quantitative analysis of affinofile data using VERSA (Viral Entry Receptor Sensitivity Analysis) metrics revealed inter-subject phenotypic segregation of brain- and lymph node derived Envs based solely on their CD4- and CCR5-usage profiles, which occurred independent of a genetic signature. CCR5-using HIV-1 variants can become resistant to the CCR5 antagonist maraviroc (MVC) by acquiring the ability to utilize the drug-bound form of CCR5. However, the molecular mechanisms involved in this process are poorly understood. In chapter 5, the functional properties of two patient derived MVC-resistant Envs were compared to those of their parental sensitive Envs. Resistant Envs displayed broadly divergent levels of resistance to MVC both in PBMCs and NP2-CD4/CCR5 cells, with approximately 70% difference in their maximum percent inhibition (MPI). The data from 293-Affinofile assays suggest that HIV-1 Envs with low or high MPI, display a highly efficient or a very inefficient interaction with the MVC-bound CCR5, respectively. Interestingly, a common molecular pathway of altered CCR5 engagement was employed by both the resistant Envs to escape MVC, regardless of their divergent MPIs and distinct resistant mutants. This pathway is characterized by increased dependence upon the CCR5 N-terminus and on charged histidine residues in the CCR5 ECLs. The altered mechanism of CCR5 engagement rendered both the resistant Envs vulnerable to inhibition by a sulfated peptide based on the CCR5 N-terminus sequence. Neither of the resistant Envs showed broad cross resistance to other CCR5 antagonists (TAK-779 and Vicriviroc) tested, suggesting that HIV-1 variants with resistance to MVC may still remain susceptible to inhibition by new CCR5 antagonists should they eventually reach the clinic. Finally it was shown that MVC is able to protect macrophages from infection by weakly resistant Envs (i.e. Envs with high MPI) but not the strongly resistant Envs (i.e. Envs with low MPI), which has potentially important clinical implications.