Skip to main content
NCBI home page
Journal of Virology logoLink to Journal of Virology
. 2024 Nov 7;98(12):e01639-24. doi: 10.1128/jvi.01639-24

Examination of respiratory syncytial virus fusion protein proteolytic processing and roles of the P27 domain

Hadley E Neal 1, Chelsea T Barrett 1, Kearstin Edmonds 1, Carole L Moncman 1, Rebecca Ellis Dutch 1,✉
Editor: Stacey Schultz-Cherry2
PMCID: PMC11650970  PMID: 39508603

ABSTRACT

The respiratory syncytial virus (RSV) fusion protein (F) facilitates virus–cell membrane fusion, which is critical for viral entry, and cell–cell fusion. In contrast to many type I fusion proteins, RSV F must be proteolytically cleaved at two distinct sites to be fusogenic. Cleavage at both sites results in the release of a 27 amino-acid fragment, termed Pep27. We examined proteolytic processing and the role of Pep27 for RSV F from both RSV A2 and RSV B9320 laboratory-adapted strains, allowing important comparisons between A and B clade F proteins. F from both clades was cleaved at both sites, and pulse-chase analysis indicated that cleavage at both sites occurs early after synthesis, most likely within the secretory pathway. Mutation of either site to alter the furin recognition motif blocked cell–cell fusion activity. To assess the role of Pep27 in F processing and expression, we deleted the Pep27 fragment, but preserved the cleavage sites. Deletion of Pep27 reduced F surface expression and cell–cell fusion. Two conserved N-linked glycosylation sites within Pep 27 are present in both the RSV A2 and RSV B9320 F. Randomization of the Pep27 sequence, while conserving the two N-liked glycosylation sites, did not significantly change surface expression, and only modestly reduced cell–cell fusion. However, the disruption of either Pep27 glycosylation site reduced cell–cell fusion. This work clarifies the timing of RSV F proteolytic cleavage and offers insight into the crucial role the N-linked glycosylation sites within Pep27 play in the biological function of F.

KEYWORDS: respiratory syncytial virus, fusion protein, membrane fusion, proteolytic enzyme cleavage

INTRODUCTION

Respiratory syncytial virus (RSV) is a significant cause of hospitalizations and death worldwide for young children, the immunocompromised, and adults over the age of 65 (1–3). There are two RSV clades, A and B, which co-circulate worldwide (4–6). Although the dominant clade varies from season to season between the two, RSV A has been the primary focus of a majority of molecular studies. The global burden of RSV led to the need for research and has led to a concerted effort to develop an RSV vaccine spanning 60 years (7). Three vaccines have been recently approved, primarily for the elderly population (8). Ongoing RSV research has revealed many critical details about the RSV F protein, but key aspects of mechanisms of entry, triggering, and fusion remain unclear.

The RSV fusion glycoprotein (F) facilitates fusion between viral and host cell membranes. RSV F is a class I fusion protein synthesized as an inactive precursor (F0) in the endoplasmic reticulum (ER), and it must undergo cleavage by a host cell protease, furin, to become fusion competent (9–12). RSV F differs from other paramyxovirus/pneumovirus class I fusion proteins in that cleavage must occur at two sites, rather than one, for fusogenic activity. These two sites both contain a canonical furin recognition motif (R-X-K/R-R), and are separated by a 27 amino-acid peptide fragment, termed Pep27 (P27) (6, 13). Furin cleavage site 1 (FCS1) is located between residues 105–109, with cleavage occurring following residue 109, while furin cleavage site 2 (FCS2) is between residues 130–136, with cleavage occurring after residue 136. A number of questions related to the formation and function of P27 remain to be clearly elucidated (14).

Studies of human RSV agree that cleavage at both sites is needed for membrane fusion, though interestingly a report using recombinant bovine RSV indicated that cleavage at FCS2 was dispensable for that virus (15). There are, however, significant differences in the reported studies on when cleavage at the two sites occurs (13, 16, 17). Work from Zimmer et al. examining cells transiently expressing RSV F (13) provided evidence that cleavage at both sites occurs in a similar time frame early after F protein synthesis. In contrast, Krzyzaniak et al., in studies utilizing infectious RSV, found evidence that the first cleavage event occurs prior to release of the progeny virus, but that the second cleavage event occurs after the new viral particle has been endocytosed into the next host cell during the viral entry process (16). Thus, the timing of when each cleavage event occurs remains to be clarified.

Second, recent studies have found that after RSV infection, children and young adults display a strong immune response to the P27 region of F (18, 19). If cleavage at both sites occurs during initial transport through the secretory pathway, the non-covalently attached P27 would likely not be present with the rest of F on the cellular or viral surface. Thus, authors suggested that the P27 immune response is likely due to epitope exposure from immature virions or dying infected cells (18). Interestingly, a recent study detected P27 on the surface of infected A549 cells and lung tissue samples from infected mice (20), which the authors suggested was due to a large proportion of unprocessed F0 on the cell surface, consistent with a model where part of F proteolytic processing occurs after viral entry (16). Recent examination of surface expressed RSV F indicated that the efficiency of P27 cleavage may be related to both the cell line utilized and the strain of RSV (21), and interestingly, the data also suggested that higher percentages of associated P27 correlated with higher stability of prefusion F.

A large body of work in many different viruses has shown a vital role for the glycosylation of viral proteins in proper folding and transport (22–28), interesting questions remain about the role of glycosylation within P27 on the folding and function of RSV F. In the RSV A2 F protein, there are five putative N-linked glycosylation sites within the F protein, two of which lie within P27. Early studies indicated that the removal of the RSV F N-linked glycosylation sites outside of P27 had no effect on F protein surface expression (29). More recently, removal of each of the five glycosylation sites individually in a recombinant viral system was shown to not block viral replication, though varying levels of infection efficiency were seen (30). Rescue of a virus lacking all five glycans was not possible. Interestingly, removal of the glycan at N116 within P27 resulted in a significantly higher neutralizing antibody titer and reduced syncytia formation, supporting a role for glycosylation within P27 (30). Additionally, another study indicated that immunizing mice with a plasmid encoding a mutation to block N-glycosylation at site 116 led to an increased immune response (31). These data suggest that glycosylation within the P27 region may be important in entry, infection, or immune evasion, but the mechanisms remain poorly understood.

In order to better understand the formation and function of the RSV F P27 peptide, we explored the timing of RSV F cleavage, at both recognition sites, using mutagenesis to disrupt furin recognition motifs and time course experiments to finely map the timing of cleavage. Importantly, our data indicate that both RSV A2 F and RSV B9320 F proteins undergo cleavage at both sites early after synthesis in a similar time frame, regardless of whether F is expressed transiently or during viral infection. We also show that RSV A2 and B9320 F proteins require both furin recognition sites to be fusogenically active, a finding that aligns with previous work for the A2 strain, but is novel for RSV F B9320 (6, 13). The impact of the presence and glycosylation of the P27 fragment was also evaluated for each subtype. We conclude that in the absence of P27, F remains stably expressed on the cell surface, but cell–cell fusion is reduced. Furthermore, we show that the glycosylation of P27 impacts cell–cell fusion differently in RSV F A2 and RSV F B9320. These findings add significantly to our understanding of this important region of the RSV F protein.

MATERIALS AND METHODS

Cell lines, viruses, and culture

Vero cells (ATCC CCL-81) were maintained in Dulbecco’s modified Eagle’s medium (DMEM, GE Healthcare), with 10% fetal bovine serum (FBS; MilliporeSigma) and 1% penicillin/streptomycin. A549 cells (ATCC CCL-81) were cultured in F12 Kaighn’s modification media (GE Healthcare) with 10% FBS and 1% penicillin/streptomycin. All cells were maintained at 37°C and 5% CO2. Recombinant respiratory syncytial virus A2 expressing green fluorescent protein (GFP, rgRSV-A2), a kind gift from MedImmune/Astrazeneca, was propagated in HEp-2 cells. rgRSV MOI 0.1 was added to HEp-2 cells in Opti-MEM. After a 3-h incubation, Opti-MEM + 5% FBS was added, and cells were incubated for 4 to 5 days until changes in cell morphology developed. Cells were scraped after one freeze–thaw, debris were spun at 2,500 rpm, and supernatant was mixed with sucrose phosphate from a 10× stock to yield 1× final concentration (Hyclone, special order from Astrazeneca/Medimmune). Virus was flash frozen, then titers were determined by infecting Vero cells and counting fluorescent infected cells.

Plasmids, antibodies, and mutagenesis

The P27 mouse monoclonal antibody, mAb RSV7.10, was kindly provided by Dr. Gale Smith (Novavax, Gaithersburg, MD). The mouse monoclonal antibody, synagis, was a kind gift from MedImmune/Astrazeneca. Goat anti-human IgG alexa fluor 647 (Jackson ImmunoResearch) was used as the secondary antibody. RSV F, laboratory adapted, A2, and B9320 strains were kind gifts from MedImmune/Astrazeneca in the mammalian expression vector pVAX. All RSV F constructs were created in pVAX using the QuikChange site-directed mutagenesis kit (Stratagene) with primers, designed in house, purchased from Eurofins. The RSV F A2 and B9320 randomized P27 mutants in pVAX were designed in house and produced by GenScript using supplied pVAX starting material. Each mutant construct was fully sequenced through ACGT DNA sequencing services.

Syncytia assay and syncytia counts

Vero cells were seeded in six-well plates, to be 75% confluent the following day, and transiently transfected with 2 µg of either wild-type (WT) or mutant RSV F, plasmid using Lipofectamine 3000 (Invitrogen) at a ratio of 1:2:2 DNA: P3000: Lipofectamine 3000. Syncytia formation was imaged at 24- and 48-h post−-transfection (hpt) on a Nikon Ti2 at 20× magnification. Nuclei were counted by defining the total syncytia area and dividing that by the area of the field. In some cases, nuclei were scored and counted as total nuclei in syncytia formation or total nuclei. The fusion index was expressed as 1 − (total nuclei − nuclei in syncytia + number of syncytia)/ total nuclei.

Surface biotinylation

WT or mutant F protein was transfected, using 2 μg, into Vero cells using the Lipofectamine 3000 system (Invitrogen; ratios described above). Cells were starved in Cys-/Met- media (Gibco) for 45 min 24 hpt, and metabolically labeled for 3 h using 50 µCi of 35S bound to cystine and methionine (PerkinElmer) diluted in Cys-/Met- media (35S Cys/Met). After the label, cells were washed once with PBS (pH 8) and incubated with 1 mg/mL of EZ-link Sulfo-NHS-biotin (Thermo Fisher) in PBS (pH 8) at 4°C for 35 min, and again at room temperature for 15 min. After incubation, the cells were lysed in 500 µL of RIPA buffer (100 mM Tris-HCl [pH 7.4], 0.1% SDS, 1% Triton X-100, 1% deoxycholic acid) containing 150 mM NaCl, protease inhibitors (1 TIU [trypsin inhibitor unit] aprotinin and 1 mM PMSF Sigma-Aldrich), 5 mM iodoacetamide, and complete EDTA-free protease inhibitor cocktail tablets (Sigma-Aldrich), and frozen overnight. Cell lysates were centrifuged at 55,000 rpm for 10 min, and supernatant was incubated with palivizumab at 4°C for 3 h. Following incubation, 30 µL protein A conjugated to Sepharose beads (Cytiva) were added to the samples and incubated at 4°C for an additional 30 min. Post-incubation, samples were washed twice with RIPA Buffer + 0.3M NaCl, RIPA Buffer + 0.15M NaCl, and once with SDS-Wash II buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, and 2.5 mM EDTA). After buffer aspiration and addition of 10% SDS, samples were boiled for 10 min. The supernatant was moved to a separate tube. Then, 15 µL of supernatant was removed and added to an equal portion of 2× SDS loading buffer and labeled “TOTAL”. Biotinylation buffer (20 mM Tris [pH 8], 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, and 0.2% BSA) and streptavidin-conjugated beads were added to the remaining supernatant, and incubated at 4°C for 1 h. Samples were again washed as described above, and 2× SDS loading buffer was added following the washes. Samples were boiled for 15 min, and proteins were separated on a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel. Gels were dried and exposed on a phosphoscreen for 2–4 days, then visualized using a Typhoon Imaging System (GE Healthcare). Bands were quantified using band densitometry using the ImageQuant software (GE Healthcare).

Radiolabel immunoprecipiation

Two micrograms of WT, or mutant F, was transfected into Vero or A549 cells using the Lipofectamine 3000 system (Invitrogen; ratios described above). At 24 hpt, cells were starved in Cys-/Met- media (Gibco) for 45 min, and metabolically labeled for 1 h using 50 µCi of 35S Cys/Met. After the 1-h label, cells were washed once with PBS, then lysed in 500 µL of RIPA lysis buffer and frozen overnight. Palivizumab was used to immunoprecipitate the RSV F protein, as described above, for analysis on a 10% SDS-PAGE gel. Gels were dried and exposed on a phosphoscreen for 2–4 days and visualized using a Typhoon Imaging System (GE Healthcare). Bands were quantified via band densitometry using the ImageQuant software (GE Healthcare).

Time course immunoprecipitation

Two micrograms of WT or mutant F was transfected into Vero or A549 cells using the Lipofectamine 3000 system (Invitrogen; ratios described above), or infected with rgRSV-A2 at MOI 0.3. At 24 hpt or infection, cells were starved in Cys-/Met- media (Gibco) for 45 min, and metabolically labeled for 15 min using 50 µCi of 35S Cys/Met. After the label, cells were washed once with PBS, and then DMEM + 10% FBS was added for indicated times. The DMEM + 10% FBS was aspirated, and cells were washed once with PBS followed by lysis with 500 µL of RIPA lysis buffer before freezing overnight. Palivizumab was used to immunoprecipitate the RSV F protein as described above, and the protein was analyzed on a 10% SDS-PAGE gel. Gels were dried and exposed on a phosphoscreen for 2–4 days and visualized using a Typhoon Imaging System (GE Healthcare). Bands were quantified using band densitometry using the ImageQuant software (GE Healthcare).

Immunofluorescence experiments

Vero cells were seeded into six-well plates with 12 mm #1 coverslips (Fisher, NC1418755), to have approximately 4.9 × 105 cells on the day of infection. Cells were transfected with WT A2 F or a 6× lysine mutant. At 24 hpt, cells were stained, either post-fixation or live cell, with wheat germ agglutinin 488 (Life Technologies, W11261) by diluting to 5 ug/mL in HBSS, or MemBrite 488 (Biotium, 30,093 T) per manufacturer instruction. Samples were then fixed with 4% PFA at 4°C for 15 min, and washed with PBSN (0.02% NaN3), before being moved to a humidified chamber. Coverslips were blocked in PBSN with 1% natal goat serum (block buffer) for 1 h, also at 4°C. Primary anti-pep27 antibody was added at 1:1,000 in block buffer for 1 h at room temperature. Samples were washed seven times with PBSN + 0.05% Tween prior to secondary goat anti-human 647 (prepared in block buffer), then incubated 1 h at 4°C. Finally, PBSN +0.05% Tween washes were repeated, and coverslips were mounted to slides using 10 µL of Everbrite hard-set mounting media (Biotium, 23002), and left covered overnight to solidify. Slides were imaged on the Axiovert 200M 63× or 100× oil objectives, as well as the Nikon Confocal 60× oil objective. Experiments preformed in duplicate.

Statistical analysis

Statistical analysis was performed using Graphpad Prism 7, considering P < 0.05 as statistically significant. Multiple comparison tests were generated using one-way or two-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test. GraphPad (GP): ns = P > 0.05, * =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001.

RESULTS

RSV F has two predicted furin cleavage sites, FCS1 (RARR109) and FCS2 (KKRKRR136) (Fig. 1A). Mutants were created to remove the furin cleavage motif at either FCS1 or FCS2 for both RSV F A-2 and RSV F B9320, as the only previous similar study had solely examined RSV A (13). Cleavage at only FCS1 generates F1+ (the F1 subunit with P27) and F2, while cleavage at only FCS2 generates F1 and F2+ (the F2 subunit with P27) (Fig. 1B). In both constructs and all constructs in this report, the disulfide bonds remain intact. Radiolabeling combined with surface biotinylation was used to evaluate how cleavage at each site impacts protein total and surface expression. As shown in Fig. 1C and F, proteins with an altered FCS1 (RARA) generated a band corresponding to the F1 cleavage product in the surface and total protein populations for both the RSV F A2 and B9320 subtypes. This indicates that cleavage still occurs at FCS2 in the absence of cleavage at FCS1. F protein mutants with altered FCS2 showed a band corresponding to F1+ product formation for both F subtypes, for both surface and total protein populations. Although the FCS2 mutations introduced are not consistent with the canonical furin recognition motif, furin or an alternative protease was able to inefficiently cleave at these sequences, as a small amount of F1 product formation was observed. These F protein mutants showed varying levels of protein expression in surface and total populations. In both RSV F subtypes, the KKKKKK mutant had the highest level of protein observed while effectively generating the F1+ cleavage product, potentially because this mutant was more stable than the other FCS2 mutants. In subsequent experiments, the F proteins with RARA and KKKKKK mutations were exclusively used. Bands for surface and total protein populations were quantified and indicated that altering FCS1 did not significantly change the ratio of F1 or F1+ product formation compared with what was observed for the WT F protein, for either surface and total protein for either subtype (Fig. 1D and E). Additionally, in the FCS2 mutants, there was a significant decrease in F1 product formation (Fig. 1D) and a significant increase in F1+ product formation (Fig. 1E) compared with WT. The F2 and F2+ products are too small to accurately be measured through this assay.

Fig 1.

Diagrams show sequence alignment and schematic cleavage of wild-type and mutant at FCS1 and FCS2, with the Western blots and bar graphs showing F1 and F1 plus protein expression across wild-type and mutant variants.

Open in a new tab

Mutations at the second cleavage site alter product formation in both surface and total populations. (A) Alanine or lysine mutations were made at each of the RSV cleavage sites in both subtypes, disrupting the furin recognition motif. (B) Diagramatic representation of the products formed depending on which cleavage site is active. (C) Vero cells were transfected with plasmids, expressing WT RSV F or the RSV F cleavage mutants, before being metabolically labeled for 3 h, followed by surface protein biotinylation to allow analysis of the surface and total populations. Using band densitometry, percent F1 (D) and F1+ (E) cleavage product formation was measured for both surface and total populations. All measurements represent the average of three independent experiments ± SD. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

RSV F cleavage is nonsequential

The relative timing of cleavage at the two sites remains unclear, given the differing results from several previous studies (13, 16). To assess if the F1 and F1+ cleavage products appear at the same or at different timepoints during F protein processing, the FCS1 and FCS2 mutants for both RSV F A2 and RSV F B9320 were evaluated through a pulse-chase time course. Vero cells were transfected with plasmids expressing WT, the FCS1 mutant (RARA), or the (FCS2) mutant KKKKKK, for either subtype. For both the WT RSV F A2 and WT RSV F B9320, bands corresponding to F1+ and F1 were present beginning at the 0.5-h chase mark, becoming more prevalent at the 1-h chase mark (Fig. 2A). This indicates that cleavage at FCS1 and FCS2 occurs within the same timeframe, suggesting that cleavage is nonsequential. F proteins with the RARA mutations (altering FCS1) produced bands corresponding to F1 at the 0.5-h chase mark, again becoming more prevalent at the 1-h chase mark, while F proteins with the KKKKKK mutations (altering FCS2) produced bands corresponding to F1+ at the same timeframe. This supports the independent nature of cleavage at either site, as the timing of cleavage was similar in F proteins with both cleavage sites or F proteins with only one.

Fig 2.

Western blots and bar graphs show time-dependent F1 and F1 plus product formation in wild-type and mutant variants across several time points.

Open in a new tab

RSV F cleavage is nonsequential. (A) Vero cells transfected with plasmids expressing WT RSV F or the RSV F cleavage mutants were metabolically labeled for 15 min and chased for the times, indicated in hours. Using band densitometry, percent F1 (B, C, and D) and F1+ (E, F, and G) cleavage product formation was measured. All measurements represent the average of three independent experiments ± SD. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

When bands were quantified, there were no significant differences in F1 or F1+ product formation between WT A2 and WT B9320 at any timepoint (Fig. 2B and E), indicating that cleavage timing is similar in F proteins from the two clades. There was a significant decrease in F1 product formation in the FCS2 mutant (KKKKKK), compared with WT of either respective subtype starting at 0.5 h, as predicted from the mutation of the cleavage site needed to generate F1. However, a significant increase in F1+ product formation was seen in the FCS2 mutant (KKKKKK) compared with WT at the same timepoint (Fig. 2F and G). These results suggest that cleavage at FCS1 and FCS2 occur independently and within the same timeframe. Figure 2C and D again show that although inefficient, cleavage can occur at the FCS2 site in the KKKKKK mutants, likely either due to inefficient furin processing or due to cleavage by other cellular proteases.

Cleavage dynamics are consistent in transfection and infection models

The model proposing cleavage of the second site after viral internalization into target cells was based on data from an infection with RSV A2 (16). To determine if cleavage is nonsequential in both transfection and infection systems, a time course immunoprecipitation experiment was performed using rgRSV A2 infected A549 or Vero cells, and this was directly compared with transfected cells expressing WT A2 F. As shown in Fig. 3A, in both transfected and infected Vero cells, bands corresponding to the F1 and F1+ products began to appear at the 0.5-h timepoint, becoming more prominent at the 1-h timepoint. These bands were quantified and showed no significant differences in F1 or F1+ product formation (Fig. 3C and E). In A549 cells, the same pattern was seen (Fig. 3B, D and F). Thus, for both RSV infected cells and transfected cells expressing RSV F, products of cleavage at both sites are detectable within the first hour after protein synthesis. These findings strongly support a model where furin cleavage of both sites occurs at similar times during transport through the secretory pathway.

Fig 3.

Western blots and bar graphs compare F1 and F1 plus protein formation in Vero and A549 cells after transfection or long infection with RSV A2.

Open in a new tab

Cleavage dynamics are consistent between transfection and infection models. Vero cells (A) and A549 cells (B) were transfected with plasmids expressing WT RSV A2 F or infected with RSV A2 were metabolically labeled for 15 min and chased for the times indicated in hours. Using band densitometry, percent F1 (C and D) and F1+ (E and F) cleavage product formation was measured for both cell types. All measurements represent the average of three independent experiments ± SD. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

Cell–cell fusion is abolished with the disruption of either cleavage site

Previous studies have indicated that cleavage at both FCS1 and FCS2 is necessary for cell–cell fusion activity (6, 13). However, this has exclusively been shown for a clade A virus. To evaluate the impact of cleavage site alterations in both clades, Vero cells were transfected with plasmids expressing WT F or F with a cleavage site mutation, and cells were imaged at 48 hpt. Syncytia formation was used as an indicator for cell–cell fusion. In either subtype, F proteins with a mutation in either FCS1 or FCS2 were unable to promote detectable syncytia formation (Fig. 4A). Images were analyzed, and syncytia were counted, as displayed in Fig. 4B. Thus, in either subtype, processing at both cleavage sites is required for cell–cell fusion competency.

Fig 4.

Micrographs show cell fusion in wild-type and mutant variants, including A2-RARA, A2-KKKKKK, B9320-RARA, and B9320-KKKKKK, with a bar graph comparing fusion indices.

Open in a new tab

Cell–cell fusion is abolished with the disruption of either cleavage site. (A) Vero cells were transfected with plasmids expressing WT F or a cleavage site mutation and imaged at 48 h post-transfection on a Nikon Ti2 for the presence of syncytia formation (denoted by white arrows). (B) Nuclei were counted by defining the total syncytia area and dividing that by the area of the field. In some cases, nuclei were scored and counted as either nuclei in syncytia or total nuclei. The fusion index was expressed as 1 − (total nuclei − nuclei in syncytia + number of syncytia)/total nuclei. All measurements represent the average of three independent experiments ± SD.

Randomization of P27 decreases F1+ product formation

The necessity of cleavage at both FCS1 and FCS2 for fusogenic activity raises questions regarding the role of P27, as it would not be covalently attached to the rest of RSV F after cleavage at both sites occurs. It has been generally assumed that P27 separates from F and does not serve a further role in viral infection. However, recent studies indicate an immunogenic response to the P27 region in children and young adults, which was theorized to be due to post-cleavage P27 in unstable particles (18). However, it is possible that the observed immune response could be due to a larger role for p27 in the infection process or immune evasion. Within the P27 fragment, two N-linked glycosylation sites are conserved among subtypes. The glycosylation of these specific sites has also recently been defined as immunogenic, further indicating that P27 is more important for RSV than previously thought (30, 31).

To evaluate how N-linked glycosylation at these sites impacts protein expression, cleavage, surface expression, and cell–cell fusion, mutants were created to remove the glycosylation site at either site 116 or 126, in both RSV F A2 and RSV F B9320 (Fig. 5A). Additionally, a mutant was made to randomize the residues within P27, with the exception of the N-linked glycosylation sites, to explore the sequence specificity of P27 on its potential functions. A P27 deletion mutation was also generated, keeping both FCS1 and FCS2 intact (Fig. 5A).

Fig 5.

A diagram shows sequence alignment of wild-type and mutant RSV F proteins, including N116Q, N126Q, P27 deletion, and randomized P27 variants, with the Western blot comparing the cleavage products F0, F1 plus, and F1 in WT and mutant variants.

Open in a new tab

Mutations within P27 targeting N-linked glycosylation and the deletion of P27 do not drastically impact protein expression in either subtype. (A) Mutations were made in both subtypes to delete P27, randomize the P27 sequence while preserving N-linked glycosylation sites, and disrupt either N-linked glycosylation site. (B) Vero cells were transfected with plasmids expressing WT F or the mutant constructs. A 35S metabolic label was performed to determine if the mutant constructs impact protein expression.

The expression and stability from degradation of these mutations were tested using a metabolic radiolabel (Fig. 5B) in transfected Vero cells. All F protein P27 mutations were expressed (Fig. 5B). Some decrease was observed in the overall protein expression levels for some of the mutants compared to the WT F proteins, but none of the mutations drastically disrupted the stability from degradation of the F protein.

A time course immunoprecipitation experiment was performed to identify how these mutations impact F cleavage and stability over time (Fig. 6A). In Vero cells transfected with plasmids expressing a P27 deletion, bands corresponding to F1 appeared at 0.5 h, similar to what was observed in the WT F protein of each respective subtype. This suggests that the timing and efficiency of cleavage are not impacted by the deletion of P27. Additionally, the protein expression over time was similar to that of WT, indicating that the deletion of P27 does not drastically change stability from degradation in either subtype. As expected, there was no formation of the F1+ cleavage product as the integral P27 fragment had been deleted. In the N116Q and N126Q mutations, stability over time, cleavage timing, and cleavage efficiency were similar to that of WT F (Fig. 6A). Interestingly, compared with WT F for either subtype, the P27 random mutant showed less F1+ product formation (Fig. 6A), indicating that cleavage at FCS1 prior to FCS2 does not occur as frequently in this mutant. In addition, in both subtypes, an upward shift was seen in F0 starting at the 1-h timepoint potentially indicating changes in post-translational modifications.

Fig 6.

Western blots and bar graphs show F1 and F1 plus product formation over time in wild-type and mutant RSV F protein variants, including N116Q, N126Q, P27 deletion, and randomized P27.

Open in a new tab

The randomization of P27 decreases F1+ product formation. (A) Vero cells transfected with plasmids expressing WT RSV F or the RSV F P27 mutants were metabolically labeled for 15 min and chased for the indicated time in hours. Using band densitometry, percent F1 (B and D) and F1+ (C and E) cleavage product formation was measured. All measurements represent the average of three independent experiments. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

Bands corresponding to F1 were quantified for both subtypes (Fig. 6B and D). In RSV F A2 (Fig. 6B), there was a small but statistically significant increase in F1 product formation for the P27 deletion mutation compared with WT at 0.5- and 1-h timepoints. However, in RSV F B9320 (Fig. 6D), this significant increase was detected at the 0.25- and 0.5-h timepoints. This slight increase in proportion of F1 product formation compared with WT can be attributed to the absence of P27, and therefore, the lack of F1+ formation. The F1+ cleavage products were quantified for both subtypes as shown in Fig. 6C and E. As expected, no F1+ was formed in the P27 deletion mutant for either subtype. Interestingly, in the RSV F A2 randomized P27 mutant, there was a significant decrease in F1+ formation compared to WT at hours 1 and 2 (Fig. 6C). This suggests that the formation of F1+ as an intermediate product may be affected by the elements in the specific P27 sequence.

P27 mutations impact cell–cell fusion differently between subtypes

Previous work with bovine RSV (BRSV) concluded that the deletion of P27 or hybridization of BRSV P27 with arbitrarily chosen peptide sequences resulted in reduced cell–cell fusion (32). This indicated that, in the case of BRSV, P27 is sequence specific. To understand how similar mutations impact cell–cell fusion in human RSV, Vero cells were transfected with the P27 mutations designed in Fig. 5A, and syncytia formation imaged at 48 hpt. As shown in Fig. 7A, in RSV F A2, the disruption of either N-linked glycosylation site reduced syncytia formation compared to WT. The deletion of P27 also reduced the size and frequency of syncytia (Fig. 7A). Interestingly, in the randomized A2 P27 construct, significant syncytia formation was still observed, though reduced from that of WT A2 (Fig. 7A; quantified in Fig. 7B). Different effects were observed for RSV F B930 mutations. As shown in Fig. 7A, WT B9320 did not form syncytia as efficiently as WT A2. In the N116Q and N126Q mutants, syncytia detection was similar to that of WT B9320. In contrast to what was seen in A2, in both the RSV F B9320 P27 deletion and randomized constructs, very little syncytia formation was detected (Fig. 7A). This indicates that glycosylation within P27 may impact cell–cell fusion differently between subtypes.

Fig 7.

Micrographs show cell fusion in wild-type and mutant RSV F protein variants, including N116Q, N126Q, P27 deletion, and randomized P27, with a bar graph comparing fusion indices.

Open in a new tab

Mutations within P27 impact cell–cell fusion. (A) Vero cells were transfected with plasmids expressing WT F or a P27 mutation and imaged at 48 h post-transfection on a Nikon Ti2 for the presence of syncytia formation (denoted by white arrows). (B) Nuclei were counted by defining the total syncytia area and dividing that by the area of the field. In some cases, nuclei were scored and counted as either nuclei in syncytia or total nuclei. The fusion index was expressed as 1 − (total nuclei − nuclei in syncytia + number of syncytia)/total nuclei. All measurements represent the average of three independent experiments ± SD. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

Low levels of P27 but little uncleaved F0 can be detected on the cell surface

In several recent studies, some P27 was detected on the cell surface of infected A549 cells (20). This could be due to immature, uncleaved F0 on the cell surface, though earlier studies found that F0 was inefficient in reaching, or unable to reach the surface (12, 33, 34). To determine the proportion of F0, F1, and F1+ on the cell surface in our system, a surface biotinylation experiment was performed. Vero cells were transfected with plasmids expressing WT RSV F, the FCS2 mutation KKKKKK (Fig. 1A) or the P27 mutations (Fig. 5A). Bands corresponding to F0, F1+, and F1 were quantified for both total and surface levels (Fig. 8B through D). Very little F0 was detected in the surface or total protein populations, indicating that during the 3-h label period, most of the F protein was cleaved (Fig. 8B). As anticipated, the FCS2 mutants KKKKKK formed significantly less F1 and significantly more F1+ than WT for either subtype (Fig. 8C and D), and this form was expressed on the cell surface, indicating that lack of cleavage at FCS2 did not prevent cellular transport. As shown in Fig. 8C, there was a small but significant decrease in A2-N116Q F1+ surface expression compared with WT A2. A significant decrease in total F1+ expression for A2-random compared with WT A2 was also observed. In both subtypes, there was no F1+ formation detected in the P27 deletion mutation (Fig. 8C). In WT F and the remaining mutants, nearly 80% of surface and total protein was detected as F1 (Fig. 8D). These results indicate that in WT RSV F A2 and B9320, uncleaved F accounts for less than 10% of F expressed on the cell surface in a transfection model. Despite the low level of F0 on the cell surface, immunofluorescence experiments with a P27 monoclonal antibody revealed P27 can be detected on the surface of transfected Vero cells (Fig. 9), with significantly more P27 observed for the A2-KKKKK mutant, as expected. This suggests that either there may be a mechanism outside of uncleaved F allowing P27 to be expressed on the cell surface, or that the low levels of uncleaved or partially cleaved F are sufficient for detection.

Fig 8.

Western blots and bar graphs compare total and surface levels of F0, F1 plus, and F1 in wild-type and mutant RSV F protein variants, including A2-KKKKKK, N116Q, N126Q, P27 deletion, and randomized P27.

Open in a new tab

Low levels of P27 and uncleaved F0 can be detected on the cell surface. (A) Vero cells transfected with plasmids expressing WT RSV F or the RSV F P27 mutants were metabolically labeled for 3 h, and surface proteins were biotinylated to analyze the surface and total populations. Using band densitometry, percent F0 (B), F1 (C), and F1+ (D) product formation was measured for both surface and total populations. All measurements represent the average of three independent experiments ± SD. Significance was determined by GraphPad two-way ANOVA (* =P ≤ 0.05, ** =P ≤ 0.01, *** ≤0.001,**** =P ≤ 0.0001).

Fig 9.

Fluorescence micrographs show DAPI, WGA-488, and P27 staining in mock, A6, and WT A2 samples, with merged channels, highlighting differences in protein distribution and cellular structure across the conditions.

Open in a new tab

P27 can be detected on the surface of transfected Vero cells. Vero cells were transfected with plasmids expressing WT RSV F, the cleavage mutant -KKKKKK, or empty vector (mock). Immunofluorescence labeling was performed 24 h post-transfection. Cells were labeled with a 488-surface membrane stain, as well as a nuclei DAPI stain. The P27 monoclonal antibody, tagged with a goat anti-human 647 secondary, is present on the cell surface of cells transfected with both F constructs. However, a general variation in the quantity of F present on the surface was observed, but this was not quantified. Images were taken on the Nikon Confocal 60× oil objective, and adjusted equally in NIS elements. Experiments performed in duplicate.

DISCUSSION

The unique cleavage pattern of the RSV F protein was first recognized more than 20 years ago (6, 13), but a number of important questions remain about the cleavage events that result in the P27 peptide fragment, and the potential functions of P27 (14). In this study, we present a detailed evaluation of RSV F cleavage timing and kinetics, as well as a functional characterization of the role of P27 in RSV F-mediated cell–cell fusion, F surface expression, and F stability over time. Consistent with previous work, we show that cleavage at both furin cleavage sites is necessary for cell–cell fusion in RSV A2 (6, 13). Importantly, we confirmed that both the A and B subtypes require cleavage at FCS1 and FCS2 for cell–cell fusion by evaluation of the strain RSV-B9320. No closely related viruses require cleavage at two sites, though cleavage at two sites occurs for other distantly related viral fusion proteins, including the coronavirus spike protein (S), where S cleavage at both sites is essential in primary airway cells, but not essential in cultured cell lines. Similar to RSV F, S is cleaved by a furin-like host cell protease to generate two disulfide linked subunits, and when one cleavage site is rendered inactive, fusion competency suffers (35–37).

The timing of proteolytic cleavage at the two sites in RSV F has been an unresolved issue in the field. An initial study with transiently expressed RSV F (13, 16). We utilized mutagenesis and pulse-chase analyses to evaluate cleavage in both transfected and RSV-infected cells. Our findings that cleavage at both sites occurs within the same time frame soon after protein synthesis are consistent with those of Zimmer et al. (13), and demonstrate that similar F cleavage kinetics are observed whether F is transiently expressed alone or is present in an infected cell. A very recent study from Rezende et al. (21) of RSV F cleavage state on the cell surface and virions found differences in the level of partial cleavage depending on the RSV strain, cell type, and post-infection, with significantly higher levels of uncleaved/partially cleaved F protein observed in very late times (5 days post-infection). Combined with our data, this suggests that early in infection, the vast majority of F protein is fully cleaved soon after synthesis, but that in very late stages, this tilt towards more uncleaved/partially cleaved F protein and a higher level of P27 on cell surface and virions, potentially due to saturation of the proteases within the secretory pathway.

Our studies found that less than 10% of the cell surface F protein is in the uncleaved, F0 form (Fig. 8B), with an additional small percentage present in the F1+ form that results from cleavage only at the first site. Consistent with this, we observed a low level of P27 on the cell surface by immunofluorescence with a P27 monoclonal antibody (Fig. 9). This may be because the small amount of uniformly distributed uncleaved F0 or partially cleaved F1+ is sufficient for significant antibody recognition.

P27 has recently been shown to play a significant role in immune recognition in mouse model studies (18, 30, 31).

Interestingly, in the unenveloped human adenovirus, capsid protein cleavage at two sites produces a 33 amino-acid peptide fragment (38). Recently, the N-terminal cleavage product, termed pVIn, has been shown to remain non-covalently associated with mature adenovirus, where it aids in the release of the newly synthesized virus from the endosome (38). It was plausible that P27 could similarly remain non-covalently associated with either F subunit through protein–protein interactions, contributing to expression of P27 on cell surfaces and virions. However, no significant changes in total or surface expression were detected with our F mutants that randomized the P27 sequence (Fig. 5 and 6), making the possibility of non-covalent continued association unlikely.

Posttranslational modifications to viral fusion proteins can impact protein expression, intracellular transport, proteolytic cleavage, receptor binding, and biological activity (39–42). P27 contains two of the five N-linked glycosylation sites within RSV A2 F, and these are conserved in B9320 F. Our studies therefore examined the role of the P27 glycans in both A2 and B9320 F. For RSV A2 F, a significant decrease in syncytia formation was seen in A2 N116Q, and a more dramatic decrease was observed with N126Q compared with WT. These results are consistent with a previous study, which saw a decrease in syncytia formation when infecting cells with recombinant RSV A2 expressing disruptions to these N-linked glycosylation motifs (30). Our studies of the impact of N-linked glycans in P27 suggest there are strain specific differences, however. For RSV B9320 F, there was a slight increase in syncytia formation compared with the WT protein for F N116Q and a slight decrease in syncytia formation for F N126Q, in contrast to what was observed for RSV F A2. Interestingly, studies evaluating the glycosylation of P27 have revealed that disrupting the N-linked glycosylation motif at site 116 increases the level of F protein immune recognition in an RSV infection (30, 31). The mechanisms by which glycans within P27 impact cell–cell fusion and immune recognition remain unclear.

The work presented here provides detailed biochemical characterization of RSV F cleavage and the fate of the P27 fragment in both RSV A2 and RSV B9320. While our work and that of others are providing clarity on P27 production, the biological function of this peptide remains to be fully determined. In the closely related BRSV, P27 produced from F cleavage is further processed and secreted by cells as virokinin, a peptide hormone in the tachykinin family. The constant secretion of virokinin desensitizes the G-protein coupled receptor tachykinin receptor 1 (TACR1) (43) that is prominently expressed in immune cells, potentially attenuating immune responses to BRSV infection. While human RSV does not contain a similar tachykinin motif, the BRSV studies illuminate the major effects that can result from peptides, such as P27. Additional studies are needed to fully understand the biological roles of P27 during RSV infection.

ACKNOWLEDGMENTS

We gratefully acknowledge Dr. Gale Smith, Novavax, Gaithersburg, MD, for the kind gift of the P27 mouse monoclonal antibody RSV7.10, and MedImmune/Astrazeneca for kindly providing the mouse monoclonal antibody synagis and the RSV F A2 and B9320 F protein genes in the expression vector pVAX.

Footnotes

This article was submitted via the Active Contributor Track (ACT). Rebecca Ellis Dutch, the ACT-eligible author, secured reviews from Mark E. Peeples, The Research Institute at Nationwide Children's Hospital, and Anne Moscona, Columbia University.

Contributor Information

Rebecca Ellis Dutch, Email: rebecca.dutch@uky.edu.

Stacey Schultz-Cherry, St. Jude Children's Research Hospital, Memphis, Tennessee, USA.

REFERENCES

  • 1. Rha B, Curns AT, Lively JY, Campbell AP, Englund JA, Boom JA, Azimi PH, Weinberg GA, Staat MA, Selvarangan R, et al. 2020. Respiratory syncytial virus–associated hospitalizations among young children: 2015-2016. Pediatrics 146:e20193611. doi: 10.1542/peds.2019-3611 [DOI] [PubMed] [Google Scholar]
  • 2. McLaughlin JM, Khan F, Schmitt H-J, Agosti Y, Jodar L, Simões EAF, Swerdlow DL. 2022. Respiratory syncytial virus-associated hospitalization rates among US infants: a systematic review and meta-analysis. J Infect Dis 225:1100–1111. doi: 10.1093/infdis/jiaa752 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Boyoglu-Barnum S, Chirkova T, Anderson LJ. 2019. Biology of infection and disease pathogenesis to guide RSV vaccine development. Front Immunol 10:1675. doi: 10.3389/fimmu.2019.01675 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Esposito S, Piralla A, Zampiero A, Bianchini S, Di Pietro G, Scala A, Pinzani R, Fossali E, Baldanti F, Principi N. 2015. Characteristics and their clinical relevance of respiratory syncytial virus types and genotypes circulating in northern Italy in five consecutive winter seasons. PLoS One 10:e0129369. doi: 10.1371/journal.pone.0129369 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Pandya MC, Callahan SM, Savchenko KG, Stobart CC. 2019. A contemporary view of respiratory syncytial virus (RSV) biology and strain-specific differences. Pathogens 8:67. doi: 10.3390/pathogens8020067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. González-Reyes L, Ruiz-Argüello MB, García-Barreno B, Calder L, López JA, Albar JP, Skehel JJ, Wiley DC, Melero JA. 2001. Cleavage of the human respiratory syncytial virus fusion protein at two distinct sites is required for activation of membrane fusion. Proc Natl Acad Sci U S A 98:9859–9864. doi: 10.1073/pnas.151098198 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Otieno JR, Kamau EM, Agoti CN, Lewa C, Otieno G, Bett A, Ngama M, Cane PA, Nokes DJ. 2017. Spread and evolution of respiratory syncytial virus a genotype ON1, Coastal Kenya, 2010–2015. Emerg Infect Dis 23:264–271. doi: 10.3201/eid2302.161149 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Addo M, Cornely O, Denkinger M, Ertl G, Herold S, Pletz M, Rohde G, Welte T, Windisch W, Witzenrath M. 2024. RSV vaccination strategies for high-risk patients 2023: a collaborative position paper by leading German medical societies and organizations. Infection 52:285–288. doi: 10.1007/s15010-023-02141-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Lamb RA. 1993. Paramyxovirus fusion: a hypothesis for changes. Virology (Auckl) 197:1–11. doi: 10.1006/viro.1993.1561 [DOI] [PubMed] [Google Scholar]
  • 10. Lamb RA, Paterson RG, Jardetzky TS. 2006. Paramyxovirus membrane fusion: lessons from the F and HN atomic structures. Virology (Auckl) 344:30–37. doi: 10.1016/j.virol.2005.09.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Lamb RA, Jardetzky TS. 2007. Structural basis of viral invasion: lessons from paramyxovirus F. Curr Opin Struct Biol 17:427–436. doi: 10.1016/j.sbi.2007.08.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Collins PL, Mottet G. 1991. Post-translational processing and oligomerization of the fusion glycoprotein of human respiratory syncytial virus. J Gen Virol 72:3095–3101. doi: 10.1099/0022-1317-72-12-3095 [DOI] [PubMed] [Google Scholar]
  • 13. Zimmer G, Budz L, Herrler G. 2001. Proteolytic activation of respiratory syncytial virus fusion protein. Cleavage at two furin consensus sequences. J Biol Chem 276:31642–31650. doi: 10.1074/jbc.M102633200 [DOI] [PubMed] [Google Scholar]
  • 14. Rezende W, Neal HE, Dutch RE, Piedra PA. 2023. The RSV F p27 peptide: current knowledge, important questions. Front Microbiol 14:1219846. doi: 10.3389/fmicb.2023.1219846 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Zimmer G, Conzelmann K-K, Herrler G. 2002. Cleavage at the furin consensus sequence RAR/KR(109) and presence of the intervening peptide of the respiratory syncytial virus fusion protein are dispensable for virus replication in cell culture. J Virol 76:9218–9224. doi: 10.1128/jvi.76.18.9218-9224.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Krzyzaniak MA, Zumstein MT, Gerez JA, Picotti P, Helenius A. 2013. Host cell entry of respiratory syncytial virus involves macropinocytosis followed by proteolytic activation of the F protein. PLoS Pathog 9:e1003309. doi: 10.1371/journal.ppat.1003309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Srinivasakumar N, Ogra PL, Flanagan TD. 1991. Characteristics of fusion of respiratory syncytial virus with HEp-2 cells as measured by R18 fluorescence dequenching assay. J Virol 65:4063–4069. doi: 10.1128/JVI.65.8.4063-4069.1991 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Fuentes S, Coyle EM, Beeler J, Golding H, Khurana S. 2001. Antigenic fingerprinting following primary RSV infection in young children identifies novel antigenic sites and reveals unlinked evolution of human antibody repertoires to fusion and attachment glycoproteins. PLoS Pathog 12:e1005554. doi: 10.1371/journal.ppat.1005554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Blunck BN, Aideyan L, Ye X, Avadhanula V, Ferlic-Stark L, Zechiedrich L, Gilbert BE, Piedra PA. 2022. Antibody responses of healthy adults to the p27 peptide of respiratory syncytial virus fusion protein. Vaccine (Auckl) 40:536–543. doi: 10.1016/j.vaccine.2021.11.087 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Lee J, Lee Y, Klenow L, Coyle EM, Tang J, Ravichandran S, Golding H, Khurana S. 2022. Protective antigenic sites identified in respiratory syncytial virus fusion protein reveals importance of p27 domain. EMBO Mol Med 14:e13847. doi: 10.15252/emmm.202013847 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Rezende W, Ye X, Angelo LS, Carisey AF, Avadhanula V, Piedra PA. 2023. The efficiency of p27 cleavage during in vitro respiratory syncytial virus (RSV) infection is cell line and RSV subtype dependent. J Virol 97:e0025423. doi: 10.1128/jvi.00254-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Vigerust DJ, Shepherd VL. 2007. Virus glycosylation: role in virulence and immune interactions. Trends Microbiol 15:211–218. doi: 10.1016/j.tim.2007.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Wagner R, Wolff T, Herwig A, Pleschka S, Klenk H-D. 2000. Interdependence of hemagglutinin glycosylation and neuraminidase as regulators of influenza virus growth: a study by reverse genetics. J Virol 74:6316–6323. doi: 10.1128/JVI.74.14.6316-6323.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Beyene A, Basu A, Meyer K, Ray R. 2004. Influence of N-linked glycans on intracellular transport of hepatitis C virus E1 chimeric glycoprotein and its role in pseudotype virus infectivity. Virol (Auckl) 324:273–285. doi: 10.1016/j.virol.2004.03.039 [DOI] [PubMed] [Google Scholar]
  • 25. Chu F, Wen H, Hou G, Lin B, Zhang W, Song Y, Ren G, Sun C, Li Z, Wang Z. 2013. Role of N-linked glycosylation of the human parainfluenza virus type 3 hemagglutinin-neuraminidase protein. Virus Res 174:137–147. doi: 10.1016/j.virusres.2013.03.012 [DOI] [PubMed] [Google Scholar]
  • 26. Hanna SL, Pierson TC, Sanchez MD, Ahmed AA, Murtadha MM, Doms RW. 2005. N-linked glycosylation of west nile virus envelope proteins influences particle assembly and infectivity. J Virol 79:13262–13274. doi: 10.1128/JVI.79.21.13262-13274.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Machamer CE, Rose JK. 1988. Vesicular stomatitis virus G proteins with altered glycosylation sites display temperature-sensitive intracellular transport and are subject to aberrant intermolecular disulfide bonding. J Biol Chem 263:5955–5960. [PubMed] [Google Scholar]
  • 28. Gallagher PJ, Henneberry JM, Sambrook JF, Gething MJ. 1992. Glycosylation requirements for intracellular transport and function of the hemagglutinin of influenza virus. J Virol 66:7136–7145. doi: 10.1128/JVI.66.12.7136-7145.1992 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Zimmer G, Trotz I, Herrler G. 2001. N-glycans of F protein differentially affect fusion activity of human respiratory syncytial virus. J Virol 75:4744–4751. doi: 10.1128/JVI.75.10.4744-4751.2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Leemans A, Boeren M, Van der Gucht W, Martinet W, Caljon G, Maes L, Cos P, Delputte P. 2019. Characterization of the role of N-glycosylation sites in the respiratory syncytial virus fusion protein in virus replication, syncytium formation and antigenicity. Virus Res 266:58–68. doi: 10.1016/j.virusres.2019.04.006 [DOI] [PubMed] [Google Scholar]
  • 31. Leemans A, Boeren M, Van der Gucht W, Pintelon I, Roose K, Schepens B, Saelens X, Bailey D, Martinet W, Caljon G, Maes L, Cos P, Delputte P. 2018. Removal of the N-glycosylation sequon at position N116 located in p27 of the respiratory syncytial virus fusion protein elicits enhanced antibody responses after DNA immunization. Viruses 10:426. doi: 10.3390/v10080426 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. König P, Giesow K, Schuldt K, Buchholz UJ, Keil GM. 2004. A novel protein expression strategy using recombinant bovine respiratory syncytial virus (BRSV): modifications of the peptide sequence between the two furin cleavage sites of the BRSV fusion protein yield secreted proteins, but affect processing and function of the BRSV fusion protein. J Gen Virol 85:1815–1824. doi: 10.1099/vir.0.80010-0 [DOI] [PubMed] [Google Scholar]
  • 33. Bolt G, Pedersen LO, Birkeslund HH. 2000. Cleavage of the respiratory syncytial virus fusion protein is required for its surface expression: role of furin. Virus Res 68:25–33. doi: 10.1016/s0168-1702(00)00149-0 [DOI] [PubMed] [Google Scholar]
  • 34. Sugrue RJ, Brown C, Brown G, Aitken J, McL Rixon HW. 2001. Furin cleavage of the respiratory syncytial virus fusion protein is not a requirement for its transport to the surface of virus-infected cells. J Gen Virol 82:1375–1386. doi: 10.1099/0022-1317-82-6-1375 [DOI] [PubMed] [Google Scholar]
  • 35. Grishin AM, Dolgova NV, Landreth S, Fisette O, Pickering IJ, George GN, Falzarano D, Cygler M. 2022. Disulfide bonds play a critical role in the structure and function of the receptor-binding domain of the SARS-CoV-2 spike antigen. J Mol Biol 434:167357. doi: 10.1016/j.jmb.2021.167357 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Barrett CT, Neal HE, Edmonds K, Moncman CL, Thompson R, Branttie JM, Boggs KB, Wu CY, Leung DW, Dutch RE. 2021. Effect of clinical isolate or cleavage site mutations in the SARS-CoV-2 spike protein on protein stability, cleavage, and cell-cell fusion. J Biol Chem 297:100902. doi: 10.1016/j.jbc.2021.100902 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Coutard B, Valle C, de Lamballerie X, Canard B, Seidah NG, Decroly E. 2020. The spike glycoprotein of the new coronavirus 2019-nCoV contains a furin-like cleavage site absent in CoV of the same clade. Antiviral Res 176:104742. doi: 10.1016/j.antiviral.2020.104742 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Snijder J, Benevento M, Moyer CL, Reddy V, Nemerow GR, Heck AJR. 2014. The cleaved N-terminus of pVI binds peripentonal hexons in mature adenovirus. J Mol Biol 426:1971–1979. doi: 10.1016/j.jmb.2014.02.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Alkhatib G, Shen SH, Briedis D, Richardson C, Massie B, Weinberg R, Smith D, Taylor J, Paoletti E, Roder J. 1994. Functional analysis of N-linked glycosylation mutants of the measles virus fusion protein synthesized by recombinant vaccinia virus vectors. J Virol 68:1522–1531. doi: 10.1128/JVI.68.3.1522-1531.1994 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Bernardi A, Huang Y, Harris B, Xiong Y, Nandi S, McDonald KA, Faller R. 2020. Development and simulation of fully glycosylated molecular models of ACE2-Fc fusion proteins and their interaction with the SARS-CoV-2 spike protein binding domain. PLoS One 15:e0237295. doi: 10.1371/journal.pone.0237295 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Carter JR, Pager CT, Fowler SD, Dutch RE. 2005. Role of N-linked glycosylation of the hendra virus fusion protein. J Virol 79:7922–7925. doi: 10.1128/JVI.79.12.7922-7925.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Fischer K, Topallar S, Kraatz F, Groschup MH, Diederich S. 2022. The role of N-linked glycosylation in proteolytic processing and cell surface transport of the Cedar virus fusion protein. Virol J 19:136. doi: 10.1186/s12985-022-01864-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Zimmer G, Rohn M, McGregor GP, Schemann M, Conzelmann K-K, Herrler G. 2003. Virokinin, a bioactive peptide of the tachykinin family, is released from the fusion protein of bovine respiratory syncytial virus. J Biol Chem 278:46854–46861. doi: 10.1074/jbc.M306949200 [DOI] [PubMed] [Google Scholar]

Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES