Background:  Human Apobec3F (h-A3F) is closely related to the broad antiviral factor Apobec3G (h-A3G) within a family of cytidine deaminases. H-A3F has recently been shown to exhibit antiviral activity against HIV-1 that is suppressed by vif protein. We previously showed that h-A3G is degraded by HIV-1 viral infectivity factor (Vif) through the Cullin5 (Cul5) containing E3 ubiquitin ligase. However, the mechanism of Vif-mediated suppression of h-A3F is still unclear.

Methods:  We tested the sensitivities of wild-type WT) and mutant h-A3F (E128K) to be degraded by various lentiviral vif proteins in an immunoblot assay, and their abilities to suppress WT or vif-defective HIV-1, SIVagm, and SIVmac viruses in a Magi infectivity assay. The effect of a dominant negative Cul5 mutant on polyubiquitination and degradation of h-A3F and HIV-1 Vif was tested by in vivo ubiquitination and immunoblot assays.

Results:  We demonstrate that h-A3F has broad antiviral potential against the tested primate lentiviruses in the absence of vif protein. HIV-1 vif, but not SIVagm vif, binds h-A3F and recruits Cul5-containing E3 ubiquitin ligase, through a SOCS box to induce polyubiquitination and degradation of h-A3F. Interestingly, Vif itself is also regulated by the same Cul5-E3 ligase. Unlike h-A3G, h-A3F showed no change in its species specificity against HIV-1 or SIVagm vif when a negatively charged amino acid at position 128 was substituted with lysine(E128K).

Conclusions:  Our results indicate that h-A3F may provide a barrier against diverse primate lentiviruses in humans possessing a different determinant of species specificity than that of h-A3G. Furthermore, Cul5-E3 ligase appears to be a common pathway hijacked by HIV-1 vif to defeat both h-A3F and h-A3G. Developing inhibitors to disrupt the interaction between vif and Cul5-E3 ligase could be therapeutically useful, allowing multiple host antiviral factors to suppress HIV-1.

Keywords: Apobec3F; Vif; Cullin5