Cells communicate via extracellular vesicles (EVs) containing functional RNAs, proteins, and lipids. Knowledge on the fate of internalized EVs, especially their capacity to fuse with target cell membranes and deliver luminal cargo, is limited. Currently available EV-cargo delivery assays are indirect and thus unlikely to uncover molecular players and conditions that specifically control the EV-fusion step. Here, we present a novel live-cell imaging assay for detection of EV-binding, -uptake, and -fusion in time and space. We employed the SunTag system for exceptional signal amplification. EV-donor cells were engineered to tag the luminal EV-membrane with a fluorescent label coupled to SunTag peptides. Recipient cells express fluorescent single-chain anti-SunTag antibody (STAb), which binds EV-enclosed SunTag upon its cytosolic exposure. Using SunTagged EVs carrying fusogen VSV-G, we visualize the EV-fusion process, quantify fusion kinetics and efficiency, and determine subcellular localization of fusion events. We term this methodology the Extracellular Vesicle Fusion Spatiotemporal Imaging Method (EV-FUSIM). In the future, this technology can support the identification of fusogenic EV-subsets, as well as molecular players and drugs that modulate EV-fusion, without confounding effects of post-fusion processes. This will extend knowledge on EV-biology and can aid in the engineering of EVs that efficiently deliver intraluminal therapeutic payloads.

Cell-to-cell heterogeneity in infection outcome is a general feature of most viruses, but the underlying mechanisms are poorly understood. Here, we developed a live-cell single-molecule imaging technology to visualize infection by unmodified influenza A viruses (IAVs) with unprecedented resolution. Using this approach, we generated a detailed kinetic map of IAV infection, which identified viral ribonucleoprotein (vRNP) replication, nuclear export, and virion budding as important sources of heterogeneity. Mechanistically, we show that infection heterogeneity is caused by differential viral gene expression signatures, resulting from widespread transcriptional defects and loss of viral genome segments. For example, loss of NS, but surprisingly not polymerase subunits, severely delays replication onset, and loss of M and NS, but not HA, underlies vRNP nuclear export defects. In summary, our work identifies the origin and consequences of infection heterogeneity and provides a broadly applicable technology that allows high-resolution phenotyping of unmodified IAVs and other negative-strand RNA viruses.

Antiviral signalling, which can be activated in host cells upon virus infection, restricts virus replication and communicates infection status to neighbouring cells. The antiviral response is heterogeneous, both quantitatively (efficiency of response activation) and qualitatively (transcribed antiviral gene set). To investigate the basis of this heterogeneity, we combined Virus Infection Real-time IMaging (VIRIM), a live-cell single-molecule imaging method, with real-time readouts of the dsRNA sensing pathway to analyse the response of human cells to encephalomyocarditis virus (EMCV) infection. We find that cell-to-cell heterogeneity in viral replication rates early in infection affect the efficiency of antiviral response activation, with lower replication rates leading to more antiviral response activation. Furthermore, we show that qualitatively distinct antiviral responses can be linked to the strength of the antiviral signalling pathway. Our analyses identify variation in early viral replication rates as an important parameter contributing to heterogeneity in antiviral response activation.