SERINC5 is a restriction factor for retroviruses, antagonized by Nef of primate lentiviruses, by glycoGag of Moloney Murine Leukaemia Virus (MoMLV) and by S2 of Equine Infectious Anaemia virus (EIAV). In addition, SERINC5 sensitizes HIV-1 to neutralizing antibodies (nAbs) targeting the MPER in gp41. However, since the identification of SERINC5 as an inhibitor of retrovirus infectivity, many features of the host factor await clarification, notably the molecular mechanisms of restriction and viral counteraction. Furthermore, SERINC5 cellular role beyond restriction is still obscure. This thesis explores multiple aspects of the mutual antagonism governing the SERINC5 interplay with retroviruses. We first describe a contribution towards the determination of the structure of SERINC5 and the identification of the determinants crucial for antiviral activity, virus sensitization to neutralization and counteraction by retroviruses. By performing a structure-based mutagenesis screening, we identified SERINC5 ECL3, ECL5 and the interface between subdomains as regions essential for inhibition of HIV-1 infectivity and virus sensitization to 4E10 and 2F5 nAbs. The simultaneous impairment of both SERINC5 antiviral effects indicates that they are mechanistically related and support the hypothesis of a SERINC5-mediated impairment of the envelope glycoproteins. We included a comparative analysis of the antiviral activity of human SERINC paralogs and their sensitivity to retroviral counteraction. It has been previously established that SERINC3 inhibits HIV-1 infectivity less potently than SERINC5, while SERINC2 has no antiviral effects. We report here that similarly to SERINC3, SERINC1 is endowed with a modest antiviral activity; in contrast, SERINC4 severely inhibits HIV-1 infectivity, despite being poorly expressed. Irrespectively of their antiretroviral potency, all SERINC proteins are incorporated into virus particles. Interestingly, we observed that virion-associated SERINC2 is specifically cleaved by the viral protease, but proteolysis does not explain the lack of antiretroviral effects. Furthermore, SERINC5 and SERINC2 have different glycomic profiles, but diverse post-translational modification is irrelevant for their opposite activity against HIV-1. In addition, we reported that human SERINCs are differently targeted by retroviral counteracting factors, with SERINC5 being the paralog most efficiently downregulated, while SERINC1 being completely resistant. A cysteines cluster within ICL4 emerged as the major determinant of SERINC5 responsiveness to different nef alleles, while it proved irrelevant for internalization by MoMLV glycoGag and EIAV S2, indicating that diverse retroviral counteractors likely target the host factor differently. Though SERINC5 ICL4 harbours multiple motifs governing SERINC5 sensitivity to antagonization, insertion of this loop within SERINC2 was not enough to transfer susceptibility to Nef activity, suggesting that the overall conformation of the protein is essential for downregulation by Nef. Importantly, the cysteine stretch within ICL4 is palmitoylated, suggesting that this modification may be important for counteraction by the lentiviral factor. SERINC5 and CD4 downregulation by Nef are functionally related, as they both require AP-2 mediated endocytosis. However, regions in Nef selectively governing SERINC5 internalization are unknown. We reported here that Phe90 within Nef αA-helix genetically uncouples the activities on SERINC5 and CD4, being selectively involved in SERINC5 downregulation. In parallel, we explored SERINC5 antagonization by different glycoGag alleles and observed that the ability to target the host factor is not conserved across divergent γ-retroviruses. Finally, we observed that HIV-1 may evade SERINC5 restriction by direct cell-to-cell infection, suggesting that the host factor may have a broader role in retroviral spreading, requiring the evolution and the conservation of active viral counteraction. To this end, we preliminary investigated a positive contribution of SERINC5 to intracellular signalling.

The accessory protein Nef of human immunodeficiency virus type 1 (HIV-1) has long been known to enhance the infectivity of HIV-1 progeny virions. The multipass transmembrane proteins serine incorporator 3 (SERINC3) and SERINC5 were recently identified as novel antiviral proteins that restrict HIV-1 infectivity. Nef enhances HIV-1 infectivity by removing SERINCs from the plasma membrane, which prevents their incorporation into progeny HIV-1 virions. To exploit this potent intrinsic antiretroviral factor for potential therapy development, it is critical to explore the determinants in SERINC5 that govern its downregulation by Nef and its restriction on HIV-1 infectivity. Here I report that the ability to inhibit HIV-1 infectivity is conserved among vertebrate SERINC5 proteins, whereas the sensitivity to downregulation by Nef is not. However, a Nef-resistant SERINC5 became Nef-sensitive when its intracellular loop 4 (ICL4) was replaced by that of Nef-sensitive human SERINC5. Conversely, human SERINC5 became resistant to Nef when its ICL4 was replaced by that of a Nef-resistant SERINC5. In general, ICL4 regions from SERINCs that exhibited resistance to a given Nef conferred resistance to the same Nef when transferred to a sensitive SERINC, and vice versa. I demonstrate that human SERINC5 can be modified to restrict HIV-1 infectivity even in the presence of Nef. Moreover, by generating chimeras between SERINC5 and SERINC2, which does not exhibit antiretroviral activity, I demonstrate that SERINC5’s inhibitory function, unlike the sensitivity to Nef, requires the participation of more than one region. Helix 4 and extracellular loop 5 (ECL5) of SERINC5 are both required for the potent restriction of HIV-1 infectivity. In contrast, a large amino-terminal portion of SERINC5 is not required for its antiretroviral activity of SERINC5. The determinants in ECL5 disperse throughout the loop. Furthermore, the ECL5 of SERINC5 is a hotspot region that determines the Env-dependent antiretroviral activity of SERINC5.