ALIX/AIP1 functions in enveloped virus budding, endosomal protein sorting, and many other cellular processes. Retroviruses, including HIV-1, SIV, and EIAV, bind and recruit ALIX through YPXnL late domain motifs (X=any residue, n=1-3). This collaborative study from the Hill Sundquist labs investigated the structure and function of the cellular ALIX protein.  Crystal structures revealed that human Alix is composed of an N-terminal Bro1 domain and a central domain composed of two extended three helix bundles, which form elongated arms that fold back into a V.  The structures also revealed conformational flexibility in the arms and suggested that the V domain may act as a flexible hinge in response to ligand binding. YPXnL late domains bind in a conserved hydrophobic pocket on the second arm near the apex of the V, whereas CHMP4/ESCRT-III proteins bind a conserved hydrophobic patch on the Bro1 domain, and both interactions are required for virus budding.  ALIX therefore serves as a flexible, extended scaffold that connects retroviral Gag proteins to ESCRT-III and other cellular budding machinery.

Structure of ALIXBro1-V (A) shows domain structure of human ALIX. Sequences that compose the different elements are color coded here (and throughout) as follows: Bro1, turquoise; Bro1-V linker, brown; V-domain arm1, green; V-domain loop, salmon; V-domain arm2, blue; and Proline-rich region (PRR), gray. (B) Ribbon representation of ALIXBro1-V is shown. Residues implicated in CHMP4 binding (Ile212; yellow) or YPXnL binding (Phe495, Val498, Ala509, Phe676, Leu680, and Ile683; red) are highlighted. The Bro1 and V domains are connected by a short hydrophobic linker (359VPV361). The limited set of interdomain hydrophobic packing interactions is made by Phe24 (a1, Bro1), Val359, Pro360, Val361, Val363 (a11, arm1, and V), and Leu585 (a19, arm1, and V).