Retroviruses encode a number of genes that are crucial for their replication in mammalian cells. Gag, Pol and Env are the main components of virions;during viral assembly, Gag and Gag-Pol polyproteins are targeted to the membrane of infected cells where they interact with Env and encapsidate genomic RNA to form a nascent viral particle. Gag molecules multimerize at the cell surface to form the shell of the virion. This step requires multiple domains in Gag: the N-terminal domain Matrix (MA) that targets Gag to the plasma membrane, an intact central capsid domain (CA) that assembles to form the retroviral central core, a nucleocapsid domain (NC) that binds genomic RNA, and a C-terminal p6 domain that carries highly conserved motifs that play essential roles during budding and release of virions. My laboratory investigates the mechanisms that control the late stages of viral morphogenesis, budding and release and the host cell proteins involved in these steps. The deletion of the p6 domain of Gag has been shown to arrest HIV-1 in late stages of viral release suggesting that the p6 domain contains cis-acting regions required for HIV-1 virions release and spread. A short motif located in the N-terminal end of p6, the PT/SAP motif, plays an essential role in late stages of budding and was then called Late domain or L domain. We and others have previously reported that HIV-1 p6 domain binds the cellular protein Tsg101 via direct interaction with the PT/SAP motif and this interaction is essential for the separation of the nascent HIV-1 virions from the surface of infected cells (Verplank et al., 2001). Tsg101 binding to HIV-1 p6 helps to recruit a complex set of cellular factors that are normally used for endosomal protein sorting, membrane invagination, and fission at the MultiVesicular Body (MVB). The proteins involved in these cellular processes are organized in multi-protein complexes called Endosomal Complex Required for Transport (ESCRT). Other retroviruses also used this machinery to bud out from the cell;the Moloney Murine Leukemia Virus (MoMLV) and the Rous Sarcoma Virus (RSV) are both a type C Retrovirus that bud and release particles from the plasma membrane. To do so, they require a short motif within their Gag polyprotein, the PPPY motif, that binds E3 ubiquitin ligases of the family of Nedd4 (Kikonyogo et al.,2001). ESCRT complexes, I, II and III, are sequentially recruited to the surface of the endosome through interaction between the cellular protein Tsg101 and its natural partner in the cell, the Hrs protein. The current model hypothesizes that HIV-1 Gag mimics Hrs to bind Tsg101 and usurps the ESCRT machinery that facilitates HIV-1 virions release. Overexpression of fragments of the Hrs protein potently inhibited HIV-1 particle release. Yeast two hybrid assays were used to identify new and independent Tsg101 binding sites in Hrs. Mutants of Hrs that interfered with Tsg101 binding to HIV-1 Gag were identified and, as a consequence, potently inhibited HIV-1 particle release. Scanning electron microscopy of cells expressing a mutant Hrs protein showed an accumulation of HIV-1 particles in abnormal structures at the cell surface. These findings are described in Bouamr et al., published in Journal of Virology 2007. The p6 domain of HIV-1 Gag polyprotein was also described to bind a cellular protein called Alg-2 Interacting Protein or AIP-1 also called Alix. This direct interaction with Gag occurs via a conserved motif, the LYXPL motif, located in the C-terminal region of the p6 domain. Yeast two hybrid analysis showed that Alix binds both HIV-1 Gag and Tsg101;AIP-1 also binds a component of the ESCRT-III, the CHMP 4 protein. Overexpression of a dominant negative version of CHMP 4 inhibited the release of HIV-1 virions. This indicated that Alix links HIV-1 Gag to components of the ESCRT machinery that facilitate HIV-1 viral release. To understand the role of Alix in HIV-1 particle morphogenesis and release, we followed a dominant negative approach: fragments of AIP-1 using fragments various regions including the N-terminal portion called Bro1 domain. Its overexpression inhibited HIV-1 budding and release. Other analyses led to the discovery that Bro1 binds a new region in Gag located within the Nucleocapsid (NC). Importantly, such interaction is a new route via which Gag gains access to members of the ESCRT pathway. These findings were published in PLos Pathogens in 2009 (Dussupt et al., 2009). Recent studies in our laboratory indicated that NC is also required for virus release through the Tsg101/PTAP pathway. We found that mutation in NC arrested virus release in late stages of virus release even though Gag retained the ability to recruit Tsg101. These studies suggested that NC is involved in the recruitment of members of ESCRT that act downstream of Tsg101 and are required for virus separation from cells. In our program, we also study the role of Nedd-like ubiquitin ligases in virus budding. In particular, we examined the role of Itch in virus release. We found that Itch interacts with the model retrovirus MoMLV Gag and stimulates virus release in an L domain independent manner. We published these findings in Journal of Virology in 2009 (Jadwin et al., 2010). Additionally, we found that Nedd4-1 interacts and ubiquitinates the ESCRT-associated protein Alix. Remarkably, cellular Nedd4-1 is required for Alix mediated virus release. These data were published in Journal of Virology in 2010 (Sette et al., 2010). The molecular details of Alix Nedd4-1 interactions are being deciphered, as well as the mechanism utilized in Nedd4-1 driven control of Alix's function in HIV-1 release.
