HIV-1 and host cell proteins interactions: role of Env and HLA-C in viral infectivity and molecular analysis of Nef and ACOT8 association

HIV-1 relies on several host cell proteins to carry out its infective cycle. In this work I investigated the interactions between HIV-1 proteins and human proteins, in particular between Env and HLA-C and between Nef and the thioesterase 8 (ACOT8). HLA-C presence on HIV-1 virions increases viral infectivity; in addition a correlation between HLA-C expression levels and HIV-1 replication control has been reported. In this study we demonstrated that: HIV-1 presence increases the amount of HLA-C open conformers (not bound to β2m) on infected cells membrane; HIV-1 Env relocates HLA-C on the cell surface of β2m-negative cells; HLA-C alternatively associates with Env or with β2m; Env replaces β2m in the transport of HLA-C open conformers to the cell surface; HIV-1 pseudoviruses are less infectious in the presence of β2m, suggesting a competitive mechanism in HLA-C binding to Env or β2m. Taken together, our data suggest that the enhanced infectivity conferred to HIV-1 by HLA-C specifically involves HLA-C molecules as open conformers. In addition this work draws attention to HLA-C stability as a discriminatory factor for the association with Env. Differences observed between HLA-C alleles may be the consequence of different binding affinities to β2m, which may account for distinct surface expression levels of HLA-C allelic variants. Differences in binding affinity to β2m or to alternative ligands such as HIV-1 Env may confer to allelic variants of HLA-C the ability to preferentially act either as a conventional immune-competent molecule or as an accessory molecule involved in HIV-1 infectivity. HIV-1 Nef interacts with several cellular proteins, among which ACOT8. This interaction may modulate lipid composition in membrane rafts during HIV-1 infection. Currently, the regions involved in the interaction have been experimentally characterized on Nef but not on ACOT8. The lack of structural information for ACOT8 hampers a deep characterization of the putative interaction regions. In this work we modelled, through in silico predictions, the ACOT8 structure in order to identify the aminoacids putatively involved in the interaction with Nef. Our data demonstrated a high charge complementarity between Nef and ACOT8 surface, which allowed the identification of the ACOT8 putative contact points involved in the interaction. The predictions were then validated by in vitro assays. Several ACOT8 deletion mutants were prepared. Through immunofluorescence and co-immunoprecipitation assays we demonstrated that the ACOT8 p∆PK mutation (K91S) is sufficient to abrogate the association with Nef, indicating that K91 plays a fundamental role for this interaction. In addition also the ACOT8 ∆PAK deletion (R45- F55), as well as the ∆PK deletion (R86-P93) resulted to be determinant for the Nef association, suggesting that other residues might be involved. No effects on Nef binding were observed upon ACOT8 p∆PAK deletion (K52) indicating that this aminoacid is not significantly involved in the interaction. Our findings open the way to further investigations for the designing of new inhibitory molecules that interfere with the Nef activity and thus reduce HIV-1 infectivity.