The fusion of viral envelopes with endosomal membranes and the release of nucleocapsids into the cytoplasm of host cells were thought to occur in a single step during viral infection. However, in Nature Cell Biology, Gruenberg and colleagues describe how this one step can be two.

Vesicular stomatitis virus (VSV) takes advantage of the endocytic pathway to infect host cells, and it was thought that nucleocapsid release occurs immediately after viral-envelope fusion with endosomal membranes, which is triggered by the acidic pH of these endocytic compartments. However, the authors found that viral fusion and viral infection (nucleocapsid release into the cytosol and subsequent viral protein synthesis) are two distinct steps. Viral fusion was triggered before most virions reached late endocytic compartments — the drug-induced depolymerization of microtubules, which are required for transport from early to late endosomes, had no significant effect on viral fusion. By contrast, viral infection was affected by drug-induced microtubule depolymerization.

VSV infection, but not fusion, was also blocked by endocytosed antibodies against lysobisphosphatidic acid (LBPA), which selectively interfere with late endosome function. Together, these observations indicate that viral fusion occurs in multivesicular bodies (MVBs) — transport intermediates between early and late endosomes — and that viral envelopes fuse with intralumenal vesicles, rather than with the endosomal membrane. Electron-microscopy studies confirmed that fusion released viral nucleocapsids into the intralumenal vesicles of MVBs, where they remained hidden, rather than into the cytosol. Viral infection, on the other hand, requires transport to late endosomes and the subsequent delivery of viral nucleocapsids to the cytosol.

Focusing on the latter step, Gruenberg and co-workers showed that, in addition to LBPA, the delivery of viral nucleocapsids to the cytosol depends on the putative LBPA effector Alix. This delivery also seems to be regulated by phosphatidylinositol-3-phosphate (PtdIns3P) signalling (PtdIns3P is abundant in endosomes) through the PtdIns3P-binding sorting nexin Snx16. They conclude that “...the nucleocapsid is exported into the cytoplasm after the back-fusion of internal vesicles with the limiting membrane of late endosomes, and that this process is controlled by the phospholipids LBPA and PtdIns(3)P and their effectors”.

This work has shown that VSV entry into cells requires two steps, not one. And, although it is important to remember that this route is not followed by all viruses, it might have been hijacked by other pathogenic agents such as anthrax toxin. In the future, it will be interesting to elucidate the molecular mechanisms that underlie the back fusion of intralumenal vesicles with the limiting membrane of MVBs, which will provide further insights into this increasingly versatile compartment.