The coronavirus spike protein (S) allows the virus to begin infecting a cell. The S protein is made of two parts: the S1 and S2 domains. S1 is responsible for letting the virus bind to receptors on the surface of the cell. S2 helps the viral membrane fuse with the cell’s membrane. In certain cases, there are proteins that can cut these two pieces apart. Research has shown that viral entry into the cell requires this proteolysis by cathepsin L, but treatment with trypsin also allows the virus to enter. It is still unclear how the proteolysis of the SARS-CoV S protein works exactly. A mutation at site R797 on the S protein made the virus unable to fuse with the cell after trypsin treatment. Researchers also inserted cleavage site for a protein called furin. This allowed for cellular fusion even without trypsin. It was noted that the addition of a second cleavage site increased this cell-to-cell fusion. This data suggests that these two cleavage sites may work together to allow the virus to fuse with the cell.
There are three kinds of S proteins. The type discussed in this paper is a class 1 fusion protein. Class 1 fusion proteins require proteolytic cleavage to activate. This activation can also occur with a low pH, receptor binding, or a combination of these two. This activation causes a shape change that exposes the part of the protein that can insert itself into the cell membrane.
Coronaviruses have continued to be a threat since SARS-CoV first emerged in 2003. The SARS-CoV S protein displays many characteristics of a class 1 fusion protein. The cleavage of S1 and S2 can be enhanced by increasing the expression of furin enzymes.
Research on SARS-CoV has shown that acidifying the inside of the viral endosome and the protease cathepsin L are important steps in viral infection. Proteins besides cathepsin L have been shown to be able to cleave the S protein as well. Other research has also shown that expression of factor Xa and other proteases like trypsin, thermolysin, and elastase can induce and enhance SARS-CoV infection. These proteases seem to cleave the S protein along the S1/S2 boundary. The use of trypsin identified position R667 as a cleavage site. This suggests there may be an entry route into the cell for SARS-CoV that doesn’t require fusion and the formation of an endosome.
Role of the S1-S2 Boundary in Trypsin-Mediated Activation of SARS-CoV S Membrane Fusion
Researchers did experiments to understand the role of S1/S2 cleavage in allowing S protein fusion. A normal SARS-CoV S protein was cleaved at the S1/S2 boundary by trypsin. However, a mutant SARS-CoV S protein (mutated at position R667) was not able to be cleaved by trypsin. Using cells referred to as Vero E6, researchers used immunofluorescence microscopy to observe the fusion. Researchers inserted the genes for SARS-CoV S protein into these Vero E6 cells. When these two cells fuse into one, the cell with now two nuclei is called syncytia. Both cells with a normal SARS-CoV S protein gene and ones with a mutated S protein gene showed normal syncytia formation with immunofluorescence microscopy. To exactly measure these two groups, researchers used a luciferase assay. Researchers used another cell line (BHK cells) for this part of the experiment. They found that more trypsin caused more fusion in cells with a normal S protein. Mutant cells were able to fuse, but not as efficiently. This suggests that S protein cleavage is likely not the most important during fusion for SARS-CoV.
Identification of Additional Cleavage Sites within the Coronavirus S Protein
Researchers used a ProP 1.0 server to look for other cleavage sites in classes 1, 2, and 3 coronaviruses. Researchers by default looked for site that furin enzymes can cleave. Researchers found a cleavage site in the S2 domain of the IBV (infectious bronchitis virus) strain called Beaudette. The Beaudette strain is the only IBV strain with a cleavage site in S2. Researchers named this cleavage site S2’. Researchers used a Western blot analysis to verify that S2’ is a cleavage site. The antibodies were specific to the S2 domain of the S protein. The results showed the S protein was cleaved at the S1/S2 boundary as well as at S2’.
Further analysis showed that SARS-CoV may also have this S2’ cleavage site.
Introduction of a Furin Site at S2’ Induces SARS-CoV-Mediated Membrane Fusion
Researchers inserted furin cleavage recognition sites at the S1/S2 boundary and at S2’ of the SARS-CoV S protein, specifically 667 and 797. The furin cleavage recognition sites allow for the furin to identify these sites as where it is supposed to cut. Researchers did this by making specific mutations in the S protein. Researchers used Vero E6 cells again for these experiments. 797 mutants showed much more fusion than 667 mutants. Cells with 667 and 797 mutations had even more fusion than 797 (>95%). This was measured using a luciferase assay.
Data additionally shows that S1/S2 cleavage facilitates cleavage at S2’ and the subsequent S2’-activated fusion.
Mutation of R797 at SARS-CoV S2’ Inhibits Trypsin-Induced Membrane Fusion
Researchers mutated K796 and R797 to further investigate the role of the S2’ position in membrane fusion. These became K796A and R797N. They had three cell groups: a K796A mutant group, an R797N mutant group, and a group with both mutations. Cells with both mutations had lower amounts of the S protein in their membrane, however, incubating the cells at 32 C rescued them. Researchers used immunofluorescence microscopy to look at syncytia formation. Using trypsin and Vero E6 cells with normal SARS-CoV S protein led to many syncytia. Cells with both positions mutated had almost undetectable amounts of syncytia formation. Mutant K796A cells still were able to efficiently fuse with trypsin treatment. A luciferase assay supported what researchers found with the microscopy experiments. The double mutation strongly inhibited trypsin-induced membrane fusion and that K796A mutations only partially affected the fusion.
SARS-CoV S-Mediated Virus Entry Is Dependent on Cleavage at S2’, as well as at the S1-S2 Boundary
To analyze the S1/S2 and S2’ cleavage sitesin vivo, researchers used a mouse leukemia virus (MLV) system that makes particles to activate luciferase expression. Researchers are now using viruses to try and infect cells, instead of looking at two cells fusing like in prior experiments. In this experiment, luciferase expression would mean the virus has infected the cell. Cells were treated with NH4Cl, which will prevent viral infection through the endosomal pathway. Researchers activated the fusion using a short trypsin treatment. The K796A and R797N mutations completely inhibited trypsin activation and infection. This suggests that these positions may play a significant role in infection. K796A mutations had a very weak effect, nearing the same levels of fusion as normal cells. R797N mutants had barely any fusion. Even though the R667N mutation decreased cell-to-cell fusion by trypsin, this in vivo experiment showed that these mutants could not be recovered by trypsin treatment. This suggests that cleavage at the S1/S2 boundary (the R667 position) has a more important role for viral infection than it does when two cells are fusing. When researchers took away the NH4Cl and trypsin, R797N mutant viruses were still able to easily infect the cell, suggesting that the R797 position is not important for the endosomal route of SARS-CoV entry.
The data from these experiments suggests that SARS-CoV S protein cleavage at the S1/S2 position triggers cleavage at the S2’ position and then membrane fusion. The S2’ position was found when research used bioinformatics to search for another possible cleavage position on the S protein that is involved in viral entry. Both mutations discussed in these experiments seemed to have no effect on virus entry if there was no endosomal acidification or cathepsin L inhibitors. This may be because cathepsin L recognition sites are not as specific as furin and trypsin sites. Therefore, cathepsin L may not be involved in S2’ cleavage, or there are other “satellite” sites around S2’ that are cathepsin recognition sites. Future experiments on these spike protein cleavage sites may have implications on viral host range or pathogenesis.
Methods and Materials
Cell-Cell Fusion Assays
Vero E6 cells were grown on 24-well plates. Invitrogen Lipofectamine 2000 was used to transfect them with the S proteins. Cells were then placed in serum-free media with trypsin. These cells were then transferred to a media with 10% FBS, incubated, and fixed in place. Immunofluorescence was used to observe the fixed cells.
For syncytia formation, cells were fixed without protease treatment. Researchers then counted the number of cells with two or more nuclei.
For quantitative assays, BHK cells were also grown in 24-well plates. Researchers then transfected them with plasmids having either normal or mutant SARS-CoV S protein and a plasmid with luciferase controlled by a T7 promoter. BHK cells were then incubated the same way that Vero E6 cells were and incubated for 6 hours before lysis. Luciferase activity was then measured
Production of Pseudotyped Virions
Researchers used plasmids to produce pseudovirions. The transduction of the pseudoparticles involved cotrasnfecting BHK cells with plasmid DNA. Following preparation and periods of incubation, luciferase activity was measured.