Apolonia, L. | bioRxiv Science

bioRxiv2026

MX2 Mediates Collapse of the HIV-1 Capsid

The HIV-1 capsid core encapsulates the viral genome and mediates its delivery into the host cells nucleus. It is composed of multiple copies of the Capsid (CA, p24Gag) protein, assembled into hexamers and pentamers to create a lattice that forms a fullerene-like cone. Myxovirus resistance 2 (MX2) is an HIV-1 restriction factor that binds to the capsid core and blocks nuclear import of the viral genome. Here, we define a minimal region of MX2 required for HIV-1 restriction and produce a corresponding functional recombinant protein. We have used cryo-electron microscopy to determine the structure of this MX2 fragment bound to the tri-hexamer interface of the capsid lattice, revealing a large, buried interface combining electrostatic and hydrophobic interactions. This structure, together with assays that measure capsid core destabilisation, shows that MX2 binding induces conformational rearrangements in the capsid lattice that culminate in a loss of integrity. These results support a model whereby MX2 exerts its antiviral activity by disrupting the capsid lattice, inducing premature fragmentation and preventing HIV-1 nuclear import. By revealing the structural basis for MX2-mediated restriction, this work also provides the framework for the development of anti-HIV molecules that mimic MX2 restriction.

microbiology↗

bioRxiv2021

The lysine-rich intracellular loop and cell type-specific co-factors are required for IFITM3 antiviral immunity in hematopoietic stem cells

The interferon-induced transmembrane protein 3 (IFITM3) inhibits lentiviral gene therapy vector entry into hematopoietic stem cells and can be overcome by Cyclosporine H (CsH), but underlying mechanisms remain unclear. Here, we show that mutating the evolutionarily conserved lysines of the IFITM3 intracellular loop abolishes its antiviral activity without affecting either its localization or its degradation by CsH through non-canonical lysosomal pathways. When confined to the plasma membrane, the lysine-competent IFITM3 lost restriction against VSV-G pseudotyped viral vectors but gained antiviral activity against vectors that fuse directly at the plasma membrane. Interestingly, altering the lysines did not alter IFITM3 homodimerization but impacted higher-order protein complex formation, suggesting loss of interaction with cellular co-factors. In agreement, IFITM3 expression was not sufficient to restrict viral vectors in myeloid K562 cells as opposed to promonocytic THP1 or primary HSC. We exclude the involvement of previously identified factors affecting IFITM3 biology and propose a novel model for IFITM3 restriction that depends on the presence of cellular co-factor(s) that may interact with IFITM3 through the intracellular loop lysine residues. Overall, our work provides significant insight into the mechanisms of action of IFITM3 and CsH that can be exploited for improved gene therapies and broadly acting antiviral strategies.

microbiology↗