Deciphering lentiviral Vpr/Vpx determinants required for HUSH and SAMHD1 antagonism highlights the molecular plasticity of these evolutionary conflicts.

Abstract

SAMHD1 and the HUSH complex constitute two blocks during primate lentivirus infection, the first by limiting reverse transcription and the second by inhibiting proviral expression. Vpr and Vpx of specific lentiviral lineages have evolved to antagonize these antiviral proteins. While the antagonism of SAMHD1 has been well characterized, the evolutionary and molecular determinants of the antagonism against HUSH are unknown. We used chimeric Vpr proteins between SIVagm.Ver and SIVagm.Gri lentiviruses infecting two African green monkey species to investigate viral determinants involved in HUSH and SAMHD1 antagonisms. We found that different interfaces of closely related Vpr proteins are engaged to degrade different SAMHD1 haplotypes. In addition, we identified distinct viral determinants in SIVagm.Ver Vpr for SAMHD1 and HUSH degradation. The substitution of one residue in SIVagm.Gri Vpr is sufficient to gain the capacity to degrade SAMHD1, while the substitution of α-helix 3 confers HUSH antagonism. We also found that Vpx proteins of HIV-2 from people living with HIV have different abilities to degrade HUSH. These phenotypes rely on small changes in either the N or C terminal part of Vpx, depending on the context. On the host side, we found that HIV-2 and SIVsmm Vpx degrading HUSH from human and vervet monkey cells cannot degrade HUSH in owl monkey cells, suggesting some host species specificity. Altogether, we highlight the molecular plasticity and constraints of viral proteins to adapt to host restrictions. HUSH, like SAMHD1, may have been engaged in ancient and more recent coevolution conflicts with lentiviruses and a player in viral fitness.IMPORTANCEAntiviral host proteins, the so-called restriction factors, block lentiviruses at different steps of their viral replication cycle. In return, primate lentiviruses may counteract these immune proteins to efficiently spread in vivo. HIV-2 and some simian immunodeficiency viruses (SIVs), but not HIV-1, inactivate SAMHD1 and HUSH, two host antiviral proteins, thanks to their Vpx or Vpr viral proteins. First, we uncovered viral determinants involved in the function of closely related Vpr proteins from SIVs of African green monkeys and of HIV-2 Vpx alleles from people living with HIV-2. We show how these small viral proteins differently adapted to SAMHD1 polymorphism or to HUSH restriction and highlight their molecular plasticity. Finally, the capacity of divergent lentiviral proteins, including HIV-2 Vpx, to induce the degradation of HUSH depends on the cell/host species. Altogether, our results suggest that HUSH has been engaged in a molecular arms race along evolution, and therefore is a key player in host-pathogen interaction.