Decoding respiratory syncytial virus morphology: distinct structural and molecular signatures of spherical and filamentous particles

Abstract

Respiratory syncytial virus (RSV) is a pleomorphic enveloped virus that buds as both spherical and filamentous particles. The determinants of RSV particle morphology and the roles of these forms in transmission and pathogenicity are not clearly defined, owing to a complex interplay of viral proteins and host factors that remains poorly understood. To further characterize RSV morphology, we developed a sucrose gradient velocity sedimentation method to separate spherical and filamentous virions. Fluorescence microscopy and electron microscopy (EM) confirmed two distinct peaks containing predominantly spherical or filamentous particles, respectively. Notably, EM images revealed a distinctive "honeycomb" pattern on the RSV envelope, suggesting an ordered lattice of glycoproteins on the virion surface. Biochemical analyses of viral protein and lipid content showed that filamentous particles contained higher levels of uncleaved fusion protein F0 and exhibited distinct phospholipid profiles compared to spherical particles. Both forms were enriched in cholesterol and phospholipids characteristic of lipid rafts, consistent with the idea that RSV buds from lipid raft microdomains. This enrichment in raft lipids is linked to cell-to-cell fusion (syncytium) formation and virion assembly. Quantitative real-time PCR analysis indicated a high particle-to-PFU ratio (~4:1), meaning only about one in four RSV virions was infectious. Spherical particles contained on average ~3 genomic RNA copies per virion, whereas filamentous particles contained ~2 copies. These data reveal several structural and compositional differences between RSV particle morphologies that may influence viral pathogenesis, and they provide a foundation for new antiviral approaches targeting virion assembly and morphology.

Keywords: filamentous particles; morphology; respiratory syncytial virus; spherical particles; sucrose gradient velocity sedimentation.

Figure 1

Separation of spherical and filamentous RSV particles by sucrose gradient velocity sedimentation. Panel 1: Infectivity profile of gradient fractions. RSV was grown for 72 h and loaded onto a 15–60% continuous sucrose gradient. Twelve fractions were collected after centrifugation. Plaque assays of each fraction (PFU/mL on the y-axis) revealed two distinct peaks of infectious virus (major peaks at fraction 2 and fraction 6). Panel 2; (A, B): Immunofluorescence images of virus in peak fractions. Fraction 2 (A) shows mostly spherical particles (examples indicated by arrowheads), whereas fraction 6 (B) is enriched in filamentous particles (arrows). RSV particles were stained with anti-F monoclonal antibody and rhodamine-labeled secondary antibody (red). Scale bar = 10 µm. Panel 2; (C, D): Transmission EM of peak fractions confirms particle morphology. (C) Representative spherical RSV virions from fraction 2 (diameters ~200 nm). (D) Representative filamentous RSV from fraction 6 (lengths ≥1 µm). Data represent three independent experiments yielding similar results. Scale Bar =100 nm (C, D).

Figure 2

RSV protein composition in spherical versus filamentous virions. (A) Western blot of RSV proteins in gradient fractions 1–12. Blot was probed with anti-RSV polyclonal antibodies that detect major structural proteins. Key bands corresponding to F0 (unprocessed fusion protein), F1 (cleaved fusion protein subunit), N (nucleocapsid), P (phosphoprotein), and M (matrix) are indicated. (B) Densitometric analysis of envelope glycoproteins F0 and F1 in spherical (fraction 2) vs. filamentous (fractions 6–7) particles. The filamentous-rich fractions contain a higher proportion of uncleaved F0 relative to cleaved F1, whereas spherical particles have more F1. (C) Densitometry of internal proteins N, P, and M shows higher levels in filamentous particle fractions than in spherical. Data represent mean band intensity (arbitrary units) from three independent experiments (error bars indicate range).

Figure 3

RSV lipid composition of spherical vs. filamentous virions. Lipids were extracted from RSV particles using the Folch method (Folch et al., 1957) and analyzed by reversed-phase HPLC-electrospray ionization mass spectrometry with a C18 column (Kim et al., 1994). Lipid extracts, spiked with d35-labeled phospholipids as internal standards, were directly injected onto the column and separated using a mobile phase containing water and 0.5% ammonium hydroxide in methanol-hexane. Detection was performed on a Hewlett-Packard HPLC-MS Series 1100 MSD instrument. (A) Specific phospholipid molecular species differentially enriched in spherical (blue bars) versus filamentous (red bars) RSV particles. Examples include distinct species of phosphatidylserine (PLS), sphingomyelin (SM), phosphatidylcholine (PC), and phosphatidylinositol (PI) that show statistically significant differences between the two forms (p < 0.05). (B) Comparison of total phospholipid and cholesterol content per particle (arbitrary units normalized to spherical virion = 1). Filamentous virions have a higher total lipid content, consistent with their larger membrane area. Data represent the average of three independent experiments. Abbreviations: SM, sphingomyelin; PS, phosphatidylserine; PC, phosphatidylcholine; PI, phosphatidylinositol.

Figure 4

Genome copy number of RSV particles determined by quantitative real-time PCR: PCR reactions were carried out with 0.05 µl cDNA in a 50-µl reaction mixture containing Platinum Quantitative PCR SuperMix-UDG (Invitrogen), 0.5 µM N1 and N2 primers, and 0.2 µM probe, as recommended by the manufacturer. Real-time RT-PCR was performed using an Opticon 2 detection system (MJ Research). The cycling conditions were 50°C for 2 min (UDG incubation), 95°C for 2 min (UDG inactivation), followed by 45 cycles of 15 s at 95°C and 1 min at 60°C to amplify the RT products. Standard curve of quantitative real-time PCR for RSV N gene (positive control) was used to calculate genome copies in each sample. Negative controls and serial dilutions of positive controls (plasmid containing the N gene) were included in each PCR assay. Tenfold serial dilutions of the pTM-N plasmid were prepared, containing 4.8×10¹¹⁰ (a), 4.8×10⁹ (b), 4.8×10⁸ (c), 4.8×10⁶ (d), and 4.8×10³ (g) copies per reaction mixture. Additionally, 0.05 µl of reverse-transcribed cDNA from peak fraction 2 (f) and fraction 6 (e) was used per reaction, with RSV copy numbers of 3.8×10⁵ and 4.7×10⁵, respectively. Data represent three independent experiments.