What’s the trajectory of a bullet-shaped virus?

by Mike

There hasn’t been much about viruses on this blog yet. Both labs I’ve worked in have been virus labs, not bacteria. But I don’t feel like a virologist, since everything we’ve looked at is the effect of the virus on the organism, or the effect of the virus on how the immune cells function. More like using viruses as a tool for immunology.

A true “virologist” virologist would dissect how the virus operates within the cell, on a more molecular level. I’ve never worked in a cold room or purified a protein. So it’s not easy for me to look at a virology paper and see its significance.

One 1968 paper in the Journal of Virology is clearly making a point, though. 


As you can see, this was just the second issue of J. Virol. At this time a considerable proportion of its papers consisted largely of a bunch of electron microscope images, plus interpretation. Hackett et al. (1) is no exception, though they also include a graph of the virus’s growth curve to contrast two cell types and two temperatures.


Vesicular stomatitis virus (VSV) is now, as it was in 1968, a very popular virus for experiments. It naturally infects humans and other mammals, mostly large animals like cows and horses for whatever reason. It’s transmitted by arthropods (e.g. sand fleas). Rabies virus and VSV are the most well-known examples of Rhabdoviruses, which are single-stranded RNA viruses that form small bullet-shaped particles enveloped by a membrane. “Vesicular” means blisters, and “stomatitis” means swelling of the mouth area. Those symptoms are particularly seen in horses, as well as swelling at the “coronal band” where leg meets hoof.

There is no DNA involved in the rhabdovirus life cycle, only RNA. Inside the cell, rhabdoviruses create pools of ribonucleoprotein (complexes of their RNA and protein, ready to be packaged into budding virus particles). These pools are visible as the “Negri bodies” that help diagnose rabies. For more on rhabdovirus structure, go to Vincent Racaniello’s blog post about a 2010 paper (2) with great images and movies.

In immunology, VSV is popular as a vector for potential vaccines, and VSV G protein (the primary protein in the viral membrane) is popular for coating other viruses to let them enter a wide range of cells. And in molecular virology, VSV is popular as a model for virus entry, virus assembly, virus budding, etc. It’s also photogenic, as Hackett and colleagues demonstrate. Here’s an infected chicken embryo cell, and a close-up of the membrane-bound compartment where viruses are forming.

VSV particles are accumulating, visible as what look like bundles, and also visible as circular cross-sections.

VSV particles are accumulating, visible as what look like bundles, and also visible as circular cross-sections.

The authors of this paper disagree with the consensus that has been developing, which states that VSV mostly forms at the outer membrane of the cell. What they see  is that the virus mostly forms in vacuoles, compartments inside the cell. They barely see any at the cell surface. They think this is because the viruses at the cell surface are actually viruses that attached from the outside. And why don’t they see viruses attaching from the outside? Because their cells are floating in suspension, instead of being fixed to the surface of the flask. And they use a very low number of cells, which do not border each other because they are floating. So it’s unlikely that a lot of virus would be released by one cell and find its way to the surface of others. Whereas if all the cells are lined up next to each other and some are releasing virus while dying, that can easily happen.

The experimental design used by the previous investigators would not preclude release and adsorption of newly produced virus and, to our knowledge, adsorption of VSV to infected cells has not been ruled out. Cells in a monolayer, such as was used by Mussgay and Weibel (3) and David-West and Labzoffsky (4), present readily available surfaces for adsorption of newly released virus from neighboring cells. Howatson and Whitmore (5) do not discuss the possibility of reinfection, but the late time of observation of infected cells would suggest that the virus rods attached to the cell surfaces may not be virus particles being released, but may be adsorbed virus.

The experiments in this study were designed with these problems in mind. Cells were infected in suspension, thoroughly washed after treatment with specific antiserum to remove the unabsorbed inoculum, diluted to 10,000 cells per ml, and incubated in suspension. Adsorption of newly produced virus under these conditions would be minimal. Vesicular stomatitis virus was found within cytoplasmic vacuoles, thus confirming the observations of Mussgay and Weibel and David-West and Labzoffsky. Virus was not found to accumulate at the cell surface, nor to bud out from the cell surface.

Shots fired! Perhaps this system, which uses small numbers of cells, and looks at cells soon after they have been infected (5 hours), is the best way to see where VSV truly accumulates.

* * *

What did the VSV community make of this hypothesis?

Not much. About a dozen citations in the ensuing two decades. Some papers cite it in passing (6, 7), as an example of virus accumulation being visible under the microscope. One cites it (8) as evidence that vacuoles form during VSV infection, without opining on whether these vacuoles are where the virus forms. A review of rabies biology (9), surprisingly, is fully convinced (maybe too convinced), saying:

Synthesis.–Rabies virus and vesicu!ar stomatitis virus are synthesized in the cytoplasms of infected cells. Their release mechanisms, however, appear to differ; most newly-formed particles of rabies virus are retained within the cell and only a few are released by budding at cell membranes in a manner similar to the release of myxoviruses (Hummeler et al., 1967). The vervet monkey agent also seems to bud from cytoplasmic membranes (Kissling et al., 1968). Vesicular stomatitis virus, on the other hand, matures like an arbovirus on the outer surface of the membranes lining intracytoplasmic vacuoles (Hackett, Zee, Schaffer & Talens, 1968).

Only one paper (10) directly addresses the issue; they agree that virus is found not just at the cell membrane but at other membranes, but they describe Hackett et al.’s system as “permissive cells of other origins in which no accumulation of nucleoprotein had been observed”. Which is to say… it’s odd to see VSV replication without seeing the large areas of ribonucleoprotein in the cytoplasm. Maybe their system was unrealistic.

So the combination of Adeline J. Hackett, Yuan Chung Zee, and Luciano Talens produced one other study about VSV (11). This one has the self-explanatory title “Vesicular Stomatitis Virus Maturation Sites in Six Different Host Cells”. And this one established, in a direct comparison, that some cells produce VSV intracellularly and some produce it at the cell surface.

It is concluded that the site of maturation of vesicular stomatitis virus is a host-dependent phenomenon.

This one has been cited much more often. A happy medium!

* * *

However, reviews (12) still say things like this:

For rhabdoviruses, virus budding takes place primarily at the plasma membrane, which was initially thought to be the default site for virus budding. …  There is some evidence that VSV can also bud from internal membranes in some cell types (Zee et al., 1970), but this aspect of VSV budding has not been studied in detail and will not be included here.

Meanwhile, outside the realm of either  VSV or rabies, scientists describing other rhabdoviruses kept citing the original 1968 paper when they saw similar accumulation of virus intracellularly. Whether it was infectious hematopoietic necrosis virus (IHNV) of salmon (13)wheat striate mosaic virus (14)or ecdysal gland virus (EGV-2) of the blue crab (15), people kept seeing rhabdoviruses form in intracellular compartments. All the way into September 2013, the discoverers of Niakha virus (16) establish that it gets stacked up in compartments inside the cell, and cite the 1968 Hackett paper as evidence that this is common in rhabdoviruses.


Look, it’s Niakha virus.

In fact, it seems like a rhabdovirus that doesn’t bud from intracellular membranes is more the exception than the rule. New ones are discovered all the time, and the electron micrographs very often show virus particles budding into vacuoles. Look at Farmington virus (17), or Durham virus (18), or Spring viremia of carp virus (19).

So when we see a statement like “X happens most of the time, but Y also happens”, what it means is probably “Either X or Y can happen, depending on the conditions”.

Maybe the number of experiments that see VSV budding exclusively from the cell surface is greater than the number of all other rhabdovirus budding experiments put together. But that doesn’t mean the former is the typical scenario.

* * *

1. Hackett AJ, Zee YC, Schaffer FL, Talens L (1968). Electron microscopic study of the morphogenesis of vesicular stomatitis virus. J Virol 2(10):1154-1162.

2. Ge P, Tsao J, Schein S, Green TJ, Luo M, Zhou ZH (2010). Cryo-EM model of the bullet-shaped vesicular stomatitis virus. Science 327(5996):689-693.

3. Mussgay M, Weibel J (1963). Electron microscopic studies on the development of vesicular stomatitis virus in KB cells. J Cell Biol 16:119-129.

4. David-West TA, Labzoffsky NA (1968). Electron microscopic studies on the development of vesicular stomatitis virus Arch Gesamte Virusforsch 23:105-125.

5. Howatson AF, Whitmore GF (1962). The development and structure of vesicular stomatitis virus. Virology 16:466-478.

6. Cohen GH, Atkinson PH, Summers DF (1971). Interactions of vesicular stomatitis virus structural proteins with HeLa plasma membranes. Nature New Biol 231:121-123.

7. Wagner RR, Snyder RM, Yamazaki S (1970). Proteins of vesicular stomatitis virus: Kinetics and cellular sites of synthesis. J Virol 5(5):548-558.

8. Dille BJ, Hughes JV, Johnson TC, Rabinowitz SG, Dal Canto MC (1981). Cytopathic effects in mouse neuroblastoma cells during a nonpermissive infection with a mutant of vesicular stomatitis virus. J Gen Virol 55:343-354.

9. Chalmers AW, Scott GR (1969). Ecology of rabies. Trop Anim Health Prod 1(1):33-55.

10. Zajac B, Hummeler K (1970). Morphogenesis of the nucleoprotein of vesicular stomatitis virus. J Virol 6(2):243-252.

11. Zee YC, Hackett AJ, Talens L (1970). Vesicular stomatitis virus maturation sites in six different host cells. J Gen Virol 7:95-102.

12. Jayakar HR, Jeetendra E, Whitt MA (2004). Rhabdovirus assembly and budding. Virus Research 106(2):117-132.

13. McAllister PE, Fryer JL, Pilcher KS (1974). Further characterization of infectious hematopoietic necrosis virus of salmonid fish (Oregon strain). Arch Gesamte Virusforsch 44(3):270-279.

14. Lee PE (1970). Developmental stages of wheat striate mosaic virus. J Ultrastructure Res 31:282-290.

15. Yudin AI, Clark WH Jr (1979). A description of rhabdovirus-like particles in the mandibular gland of the blue crab, Callinectes sapidus. J Invertebrate Pathol 33(2):133-147.

16. Vasilakis N et al (2013). Niakha virus: A novel member of the family Rhabdoviridae isolated from phlebotomine sandflies in Senegal. Virology 444:80-89.

17. Palacios G et al (2013). Characterization of Farmington virus, a novel virus from birds that is distantly related to members of the family Rhabdoviridae. Virol J 10:219.

18. Allison AB et al (2011). Characterization of Durham virus, a novel rhabdovirus that encodes both a C and SH protein. Virus Res 155(1):112-122.

19. Granzow H, Weiland F, Fichtner D, Enzmann PJ (1997). Studies of the ultrastructure and morphogenesis of fish pathogenic viruses grown in cell culture. J Fish Dis 20(1):1-10.

* * *

If you liked this post, follow via email or RSS! (links for that are at the bottom of the page.)

Or follow on twitter @AmboceptorBlog.