The role of nuclear envelope associated proteins in early HIV 1 infection steps

Abstract

There is increasing evidence that HIV-1 may interact with components associated with the nuclear envelope (NE) during the infection of dividing and non-dividing cells. This ensures correct nuclear import and integration, suggesting that NE may be of greater importance than is currently appreciated. Previous studies have shown that HIV 1 interacts with the nuclear pore complex, followed by nuclear import of the pre-integration complex and preferential integration into genomic areas that are topologically in close proximity to the inner nuclear membrane. To identify host proteins that may contribute to these processes, we performed an overexpression screen of known membrane-associated NE proteins. Two nuclear membrane associated proteins SUN1/UNC84A and SUN2/UNC84B, members of the Linker of Nucleoskeleton and Cytoskeleton complex, were shown to efficiently block nuclear import of certain HIV-1 laboratory strains (HIV-1NL4.3 and HIV-1IIIB) as well as natural strains upon overexpression. The amino-terminal 85-90 amino acid residues were identified as being required for the SUN1-mediated block and it was further demonstrated that the amino-terminal domains of SUN1 and SUN2 interact with HIV-1 in a capsid (CA)-specific way. To test whether depletion of endogenous SUN proteins causes differences in HIV-1 infection, SUN1-/- and SUN2-/- cells were generated with CRISPR/Cas9 and it was found that SUN1 absence did not have any detectable effect on HIV-1 infectivity, whereas the loss of SUN2 resulted in a modest suppressive effect in the accumulation of viral cDNA in the nucleus. The analysis with HIV 2 and other retroviruses suggests that SUN2 gene disruption affects HIV 1 specifically and does not involve any unspecific block to nuclear import. This block to infection was further analyzed in U87MG CD4 / CXCR4 cells with shRNA-reduced SUN2 expression. In this case, the reduction of SUN2 levels resulted in a 5-fold decrease in HIV-1 infection after 24h, in comparison to control cells while infection increased to wild type levels 48h post infection. Overall, the data suggest that SUN2 may help promote the early stages of HIV-1 infection, while the contribution of SUN1 needs to be further investigated. The role of the CA protein and its connection to IFN-α-induced suppression was also investigated, by analyzing the infectivities of HIV-1 CA mutants N74D, A105T, as well as P90A. Despite their relative resistance to ectopically expressed MX2, these CA mutants showed an increased sensitivity to the IFN-α-induced post entry block, which was not dependent on MX2 antiviral activity. The data suggests that CA protein and the capsid core may protect incoming HIV-1 nucleic acids not only from being detected by cytoplasmic DNA sensors, but also from IFN-α-induced effectors, thereby providing dual protection against host defense mechanism.