LC3-positive vesicles in influenza virus infection. Secretion or endocytosis? Rachel Ulferts1, Matthew Russell1, Elena Marcassa1, Mike Hollinshead2, Lucy Collinson1, and Rupert Beale1 1The Francis Crick Institute, London, UK; 2Department of Pathology, University of Cambridge, Cambridge, UK The M2 proton channel of influenza A virus (IAV) causes accumulation of the autophagy protein LC3 at intracellular membranes and the plasma membrane. We have recently shown that the molecular mechanism of M2-induced LC3-lipidation is distinct from canonical, starvation, induced autophagy. Recruitment of the ATG5-ATG12/ATG16L1 in IAV-induced LC3-lipidation critically depends on the C-terminal WD40 domain (WD40 CTD) of ATG16L1. This domain is dispensable for starvation-induced autophagy. The LC3 lipidation induced by drugs that raise the pH of vesicles, LC3-associated phagocytosis (LAP) and entosis similarly depend on the WD40 domain of ATG16L1. We proposed that this pathway is a cellular response to a failure to correctly acidify intracellular vesicles. LC3-positive vesicles in LAP, entosis and drug ionophore treatment are endolysosomal compartment, including phagosomes and lysosomes. Additionally, secretory autophagy has recently been shown to depend on the WD40 CTD. To characterise M2-induced LC3-positive vesicles we analysed their colocalization with various endolysosomal marker proteins. Consistent with published data, no co-localisation with lamp-1 was observed. Similarly, M2-induced LC3-II did not colocalise with “classical” endosomal markers including EEA1, rab5 and rab7. Live cell imaging revealed a population of highly dynamic vesicles that exhibits fusion–fission behaviour and a largely immobile perinuclear cluster of vesicles. Using 3D correlative light and focussed Ion Beam milling combined with scanning electron microscopy (CLEM) we show that these vesicles are single membrane vesicles located throughout the cytoplasm while the perinuclear cluster represents a post Golgi compartment.