Steinar Mannsverk: Why Do So Many Viruses Infect Cells via Endosomes?: Dissecting Host Determinants of Viral Membrane Fusion via Single-Virus Microscopy

Abstract

Enveloped viruses enter host cells by fusing their lipid envelope with a host membrane, thereby releasing their genome into the host cytoplasm. While fusion can occur directly at the plasma membrane following attachment, most viruses preferentially fuse with the endosomal membrane after endocytosis. This thesis addresses the central question of why so many enveloped viruses exploit endosomal entry, using influenza A virus (IAV) as a model. I hypothesised that the endosomal membrane is generally more permissive to viral fusion, compared to the plasma membrane, partly due to differences in their lipid composition and the resulting physicochemical membrane properties.

To test this, I developed and applied single-virus fusion assays that report on either viral lipid or content mixing events. In Paper I, we established a novel genome exposure assay by loading target liposomes with a nucleic acid–sensitive dye, enabling direct detection of productive pore formation events. In Paper II, we leveraged this assay to demonstrate that the endosomally enriched phospholipid bis(monoacylglycero)phosphate (BMP) increases the likelihood that a hemifusion intermediate can progress to pore formation, during IAV fusion. This effect was largely attributable to BMP’s negatively charged headgroup, with a contribution from its unique chemical structure. In Paper III, we developed a plasma membrane content mixing assay using dye-loaded plasma membrane vesicles (PMVs). IAV fusion with PMVs was markedly less efficient than with BMP-containing liposomes, but BMP supplementation in PMVs did not restore fusion, indicating additional restrictive factors at the plasma membrane. Modest cholesterol depletion in PMVs enhanced IAV fusion kinetics and reduced membrane order, whilst modest cholesterol depletion in whole cells increased productive IAV infection at the plasma membrane. These findings demonstrate that the specific characteristics of the plasma membrane in terms of lipid order, largely driven by cholesterol and sphingomyelin, constitutes a significant barrier to influenza virus fusion at the cell surface.

Together, this work suggests that the lipid composition of the endosomal and plasma membranes promotes and inhibits fusion of the endosome-adapted influenza virus, respectively. I propose that influenza, and possibly other enveloped viruses, have evolved to exploit the inherently higher permissiveness of the endosomal membrane during cell entry.