Molecular Biosciences
Sara Gnolek

Characterizing the oligomeric structure and fusion activity of PIV5 SH protein

  • Faculty Advisor

    Robert A. Lamb

Published On

May 2017

Originally Published

NURJ Online
Undergraduate Research Grant


Paramyxovirus parainfluenza virus 5 contains a 44-residue hydrophobic protein SH. SH has been shown to affect infected cell apoptosis, a crucial part of the PIV5’s ability to have effective infection. The mechanism of stopping cell apoptosis involving SH is not well understood. The structure of SH and its role in virus fusion could play a role in cell apoptosis, so SH structure and fusion activity was investigated. SH did not demonstrate any affect on viral fusion activity. There was also no evidence found of higher order oligomeric structures of SH.


Programmed cell death, apoptosis, is an important action for cells to defend themselves from being infected by viruses. Apoptosis can be activated from viruses signaling pathways that allow the infected host organisms avoid infection by killing off infected cells.[1] To counter this, viruses can have ways to prolong the life of infected cells in order to be able to live and infect more cells. Viruses can also trigger apoptosis to release and spread the virus from infected cells.[2] Paramyxovirus parainfluenza virus 5 contains a small membrane protein SH which has been shown to have an effect on limiting cell apoptosis so the virus is more easily spread. There is a lot unknown about the SH protein, especially about its structure.

Paramyxovirus parainfluenza virus 5 (PIV5, formerly known as simian virus 5, SV5) is a member of the Rubulavirus genus of the family Paramyxoviridae. Apoptosis is an important part of paramyxovirus pathogenesis. PIV5 is very good at infecting many cells without causing signaling apoptosis pathways and with little cytopathic effect, suggesting PIV5 has a mechanism to avoid cell apoptosis.[3] PIV5 encodes a protein SH, located between the genes for the fusion protein (F) and hemagglutinin-neuraminidase (HN). SH is a 44-residue hydrophobic integral membrane protein with its N terminus in the cytoplasm.[4] A recombinant PIV5 without the SH gene showed no difference in virus viability from the wild-type PIV5 virus, which gives the conclusion that SH is not necessary for virus replication.[5] However, the recombinant PIV5 without the SH gene caused more apoptosis and cytopathic effect in cells, which suggests that a function of SH is limiting virus induced infected cell apoptosis.[6]

The oligomeric structure and viral membrane fusion of SH was investigated. Evidence of a higher order structure of SH would further the understanding of how a very small protein affects apoptosis. The role of SH in membrane fusion is not known, and could also be a method of preventing cell death of infected cells.


The SH protein structure and function was analyzed using sucrose gradients, luciferase assays, immunofluorescence microscopy, cross-linking and fusion imagery.

First, SH protein was produced by transfecting cells with SH in pcaggs plasmid, and also by infecting cells with PIV5. In both cases, the SH protein was placed on top of a sucrose gradient (65%-20%-5% sucrose), and spun at a 60,000 rpm for 16hrs. The gradient was then fractionated into 14 fractions and run on a SDS-PAGE gel. The transfection produced protein was analyzed by western blot and the infected cell protein fractions were immunoprecipitated. Next, BHK cells were transfected with five different complexes; pcaggs or empty vector, F, F and HN, F and SH, and finally F HN and SH. The F protein is responsible for fusion, and HN protein interacts with F to promote fusion. The BHK transfected cells were imaged and showed more fusion in cells with F and HN, but did not seem dependent on whether SH was present.

To quantify if SH has an effect on fusion, a luciferase assay was performed. Vero cells transfected with the same complexes (plus the luciferase plasmid in all cases) as the BHK cells and were later overlaid with BSR cells. Four hours later, the luciferase reagent was added and the light emitted from the cells was recorded. If cells fuse, the luciferase plasmid in the vero cells can react with the reagent and light is emitted, so the fusion activity in cells can be quantified.

Immunofluorescence microscopy was used next to see if there is any colocalization of F and HN with SH. Hela cells were transfected with the same complexes as before, and F/HN and SH were stained and imaged under a microscope (F and HN use the same secondary antibody, so only one could be stained for on a slide).

To further explore whether SH has any higher order oligomeric structure, cross-linking experiments were performed. Five different crosslinkers were used with SH transfected cells, and then run on a gel and analyzed by western blot. Cross-linkers help form disulfide bonds between proteins if they are able to, but are not close enough together or in the right orientation. Disulfide bonds are what hold monomers together to form higher order oligomers, so cross-linking is just a way to ensure all possible disulfide bonds between proteins are formed.


The sucrose fraction gradients showed strong bands of protein in the fractions 2-4 of the gradient indicating that the SH protein is very small and does not form higher order oligomerization. The fraction density that SH settled at in consistent with a monomer of its size of 44 residues. The fractions were also run under non-reducing conditions to confirm the lack of higher oligomerization and the results were the same.

The luciferase results agreed with the BHK cells, SH did not increase fusion activity in any of the cells transfected with it compared to those without SH. Fusion occurred when F and HN were present, but SH had no effect on fusion in cells with HN and did not cause fusion in cells with only F. HN has been shown previously to initiate membrane fusion and cells with HN were expected to fuse.

In the above images, green is F or HN, red is SH, and blue is the nucleus stained with DAPI. Yellow is colocalization of F/HN and SH. The immunofluorescence images of SH show little to no colocalization with F and HN, which is to be expected since SH was already shown not to affect F and HN during fusion. SH is also shown to localize inside of the membrane which is further evidence it is not involved in membrane fusion. HN and F are both seen mostly near the membrane which is further evidence of their involvement in viral fusion.

No higher order oligomers were seen on the western blot with cross-linkers. SH was present but the bands indicated the protein existed only as a monomer for each crosslinker used. The SH protein with the different cross-linkers all had the same band as SH without cross-inker.


Plans for future experiments include cross-linking of SH with F and HN to further rule out any interactions of SH with fusion proteins. Imaging immunofluorescence of infected Hela cells would give something to compare the SH transfected cells with, and this could show that F or HN does affect the localization of SH. After these experiments, verifying SH effect on cell apoptosis could be next.


Sara Gnolek


I would like to thank Professor Robert Lamb for advising me and allowing me to work in his lab. I would also like to thank Dr. Reay Paterson for her mentoring, and the Office of Undergraduate Research for allowing me to work during the summer under a summer grant.