Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Apr 5.
Published in final edited form as: Structure. 2016 Mar 24;24(4):567–575. doi: 10.1016/j.str.2016.02.011

Crystal Structure and Proteomics Analysis of Empty Virus-like Particles of Cowpea Mosaic Virus

Nhung T Huynh 1, Emma L Hesketh 3, Pooja Saxena 2,5, Yulia Meshcheriakova 2, You-Chan Ku 1, Linh T Hoang 4, John E Johnson 1, Neil A Ranson 3, George P Lomonossoff 2, Vijay S Reddy 1,*
PMCID: PMC4828962  NIHMSID: NIHMS764299  PMID: 27021160

SUMMARY

Empty virus-like particles (eVLPs) of Cowpea mosaic virus (CPMV) are currently being utilized as reagents in various biomedical and nanotechnology applications. Here, we report the crystal structure of CPMV eVLPs determined using X-ray crystallography at 2.3 Å resolution and compare it with previously reported cryo-electron microscopy (cryo-EM) of eVLPs and virion crystal structures. Although the X-ray and cryo-EM structures of eVLPs are mostly similar, there exist significant differences at the C terminus of the small (S) subunit. The intact C terminus of the S subunit plays a critical role in enabling the efficient assembly of CPMV virions and eVLPs, but undergoes proteolysis after particle formation. In addition, we report the results of mass spectrometry-based proteomics analysis of coat protein subunits from CPMV eVLPs and virions that identify the C termini of S subunits undergo proteolytic cleavages at multiple sites instead of a single cleavage site as previously observed.

INTRODUCTION

Cowpea mosaic virus (CPMV), the type member of the Comoviridae subfamily of the plant-infecting Secoviridae, consists of single-stranded, positive-sense, bipartite RNA genomes of RNA-1 (6 kb) and RNA-2 (3.5 kb). The two RNAs are encapsidated into separate particles and both RNAs are required for virus infection. The virus particles are ~310 Å in diameter and display pseudo T = 3 icosahedral symmetry (Figures 1A and 1B). Each particle consists of 60 copies of a protomer comprising large (L) and small (S) coat protein (CP) subunits derived from a single CP precursor (VP60) encoded by the RNA-2 and processed by the RNA-1-encoded 24K proteinase. The L subunit (41 kDa) is composed of two jellyroll β-barrel domains, and the S subunit (23 kDa) consists of a single jellyroll β barrel (Figure 1C).

Figure 1. Structure and Organization of CPMV eVLPs.

Figure 1

Open in a new tab

(A) An icosahedral (pseudo) T = 3 cage showing the geometric organization of the large (L) and small (S) subunits in a CPMV capsid, shown in green and blue, respectively. The two β barrels (β1 and β2) of the L subunit encircle the 3-fold axis denoted by a yellow triangle. The β barrel (β3) of the S subunit surrounds the icosahedral 5-fold vertices, identified by a yellow pentagon. The yellow diamond represents the location of icosahedral 2-fold axis.

(B) A surface representation of the subunit organization of the L and S subunits. The C-terminal residues (561–567) of the S subunit are shown as a red-colored surface.

(C) A ribbon diagram of the CPMV protomer depicting the secondary and tertiary structures of the L and S subunits. The N termini of the L and S subunits are labeled as L-N and S-N, respectively, and the visible C termini are correspondingly labeled as L-C and S-C, respectively. Figures were generated using Chimera (Pettersen et al., 2004) and PyMOL (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC.)

Due to the simplicity and robustness of its capsid and the availability of an atomic resolution structure (Lin et al., 1999; Stauffacher et al., 1987), CPMV has been widely used in bio-nanotechnology applications. These include the display of antigenic peptides on the particle surface (Dalsgaard et al., 1997; Porta et al., 1994; Usha et al., 1993), the chemical coupling of a variety of moieties (Aljabali et al., 2010, 2012; Steinmetz et al., 2006a, 2006b), and the deposition of minerals on the particle surface (Aljabali et al., 2011; Shah et al., 2009; Steinmetz et al., 2009), as well as development of particles to deliver drugs coupled to the outer surface of the cells (Aljabali et al., 2013). A particular advantage of CPMV for in vivo applications is that particles are well tolerated in mammals (Destito et al., 2009; Rae et al., 2005). However, a disadvantage of using CPMV particles produced via infection is that approximately 90% of such particles contain one or other genomic RNAs. As a result, preparations of such particles retain their infectivity to plants, with attendant bio-containment concerns, and the particles cannot be efficiently loaded with cargo. To address these issues, a system for the production of RNA-free empty virus-like particles (eVLPs) has been developed. These can be generated by co-expression of the CP precursor VP60 along with the 24K viral proteinase in Nicotiana benthamiana (Montague et al., 2011; Saunders et al., 2009). CPMV eVLPs generated this way have already proven useful as reagents for bio- and nanotechnology applications (Lebedev et al., 2016; Sainsbury et al., 2011, 2014; Wen et al., 2012).

To gain insight into the structural similarities between eVLPs and native CPMV virions, we have conducted structural studies on CPMV eVLPs using X-ray crystallography. Furthermore, this allows comparison with the recently determined cryo-electron microscopy (cryo-EM) structure of eVLPs at 3.0 Å, performed independently (Hesketh et al., 2015), and facilitates the correlation of eVLP structures determined using two different methods. Hence, here we report the crystal structure of eVLPs at 2.3 Å resolution and compare it with the cryo-EM structure of eVLPs (Hesketh et al., 2015) and the crystal structure of CPMV virion (PDB: 1NY7) (Lin et al., 1999; Stauffacher et al., 1987). In addition, we have carried out mass spectrometry-based proteomics analysis on the S subunits from CPMV eVLPs and virions to investigate in detail the location of the proteolytic cleavage sites of the S subunit that result in the occurrence of the slow and fast electrophoretic forms of the virus (Niblett and Semancik, 1969). Each protomer of CPMV that contains 587 amino acid (aa) residues undergoes proteolytic processing at residues Gln374-Gly375 by the viral protease prior or during assembly generating the large (L) and small (S) subunits (Franssen et al., 1982). The full-length L and S subunits contain 374 and 213 aa, respectively. It has been shown that the intact C terminus of the S subunit is required for the efficient assembly of both CPMV virions and eVLPs (Sainsbury et al., 2011; Taylor et al., 1999). After particle assembly, the surface-exposed C terminus is proteolysed up to residue Leu563 (189) (Taylor et al., 1999). In the virion crystal structure (PDB: 1NY7), the last visible C-terminal residue of the S subunit, Leu563 (189) is partially exposed (Lin et al., 1999) (we have chosen to use the continuous numbering system for the S subunit to be consistent with the sequence databases, instead of the old numbering that is shown in parentheses and starts from residue number 1, which was used in the description of the virion structure; Lin et al., 1999). However, the rate of loss of the C-terminal amino acids appears to be slower with eVLPs than with virus (Sainsbury et al., 2011), raising the possibility that some of residues may be visible in the X-ray structure of eVLPs.

The results presented here demonstrate that the overall structure of eVLPs is extremely similar to that of virions produced via infection, whose crystal structure was determined previously (Lin et al., 1999). Furthermore, with the exception of the C terminus of the S subunit, the crystal structure of eVLPs reported in the current study shows close agreement with the independently obtained cryo-EM structure (Hesketh et al., 2015). The proteomics analysis of eVLPs and of virions has revealed that the C terminus of the S subunit of CPMV undergoes proteolytic processing at multiple sites suggesting the possibility that CPMV eVLPs and virions are composed of S subunits of variable lengths. This is in agreement with previous observations that the presence of different proportions of processed and unprocessed C termini of S subunits are likely to result in fast and slow electrophoretic forms of CPMV virions (Niblett and Semancik, 1969).

RESULTS

Crystal Structure Determination of eVLPs

Crystals of CPMV eVLPs were grown under the following conditions: 0.1 M sodium acetate (pH 4.68), 3%–4% (w/v) PEG3350, 0.3 M ammonium sulfate, and 5% (v/v) glycerol, at 20°C. The crystals belong to space group I41 (a = 655.97 Å, b = 655.97 Å, c = 571.45 Å, α = 90°, β = 90°, γ = 90°) with two particles in the crystallographic asymmetric unit. The eVLP crystals are distinct from the crystals of CPMV virions (of space group I23), grown under different conditions: 0.05 M potassium phosphate (pH 7.0), 0.4 M ammonium sulfate, 2% (w/v) PEG8000 (Lin et al., 1999). The statistics associated with diffraction data and model refinement are shown in Table 1. Bin-wise statistics for various resolution bins are provided in Table S1. The details of data collection, data processing, structure determination, and model refinement are described in the Experimental Procedures. Briefly, the structure was determined using a molecular replacement method using the program Phaser (McCoy et al., 2007) and refined using the program CNS (Brunger et al., 1998). The R factor of the eVLP structure is relatively high (Rwork, 0.357; Rfree, 0.359), but is consistent with the molecular replacement R factor (0.351) calculated between the Fo and the Fc derived from back-transformed electron density obtained by imposing 120-fold non-crystallographic symmetry (NCS). The high model R factor is likely due to imposition of 120-fold strict NCS constraints, in a high-symmetry space group (I41), resulting from two particles present in the crystallographic asymmetric unit of a high-symmetry space group (I41). Further discussion on this topic can be found in the Experimental Procedures.

Table 1.

Data Collection and Refinement Statistics

X-Ray source 23-ID-D, GMCA, APS
Wavelength (Å) 0.858
Space group I41
Cell parameters a = 655.97 Å, b = 655.97 Å, c = 571.45 Å, α = 90°, β = 90°, γ = 90°
Number of particles 16/unit cell; 2/crystal asymmetric unit
VM (Å3/Da) 2.75
Percent solvent (%) 55
Resolution (Å) 50–2.3 (2.42–2.3)
Number of unique reflections 3,192,399 (383,718)
Completeness (%) 60.3 (49.6)
Rmergea (%) 0.29 (0.452)
I/σ 3.7 (1.7)
Particle orientations (Phaser solution) Particle-1: EULER 39.90, 71.85, 333.96
FRAC 0.252, 0.00, 0.0000
Particle-2: EULER 74.83, 89.08, 189.73
FRAC 0.500, 0.249, 0.253
Ordered residues 1–369 (L); 375–567 (S)
Waters 58
Refinement Statistics
Resolution (Å) 20–2.3
Number of reflections 1,861,715
Degrees of NCS 120
Rworkb 0.357
Rfreec 0.359
Rmsd bond length (Å) 0.009
Rmsd bond angle (°) 1.527
Ramachandran Plot Statistics
Most favorable (%) 78.2
Allowed (%) 2.6
Disallowed (%) 0
Open in a new tab
a

Rmerge = [ΣhΣi|Ih – Ihi|/ΣhΣiIhi] where Ih is the mean of Ihi observations of reflection h. Numbers in parentheses are the statistics for the highest-resolution shell.

b

Rwork = Σ||Fo| – |Fc||/Σ|Fo| × 100 for 95% of recorded data.

c

Rfree = Σ||Fo| – |Fc||/Σ|Fo| × 100 for 5% of data.

A majority of the crystal structure of CPMV eVLP was built in strong electron density at a contour level of 1.0σ (Figure 2A), with the exception of the last two residues (566–567) at the C terminus of S subunit. The contour level was reduced to 0.5σ in the omit maps to model residues 566–567 (Figures 2B and 2C). However, all the reported residues, 1–369 (L subunit) and 375–567 (S subunit) are clearly visible in the final refined maps (Figure 2B). The C-terminal residues 370–374 of the L subunit and residues 568–587 (194–213) of the S subunit are disordered in the crystal structure. It is possible that some of the latter unidentified residues of the S subunit could have been proteolytically removed prior to crystallization and hence likely to be absent in the crystals of eVLPs.

Figure 2. Representative Electron Density of the eVLP Crystal Structure.

Figure 2

Open in a new tab

(A) The (3Fo – 2Fc) electron density illustrated by blue mesh at residue Tyr100 contoured at 1.0σ.

(B) The continuous (3Fo – 2Fc) density at the C terminus of the S subunit comprising the terminal 16 residues (551–567) in the crystal structure.

(C) An omit map showing the density at the C terminus of the S subunit, calculated by omitting residues 554–567. The βX″-strand of residues 421–425, adjacent to the C-terminal residues is shown as reference.

CPMV eVLP Structure and Particle Organization

The overall structure, size, and organization of CPMV eVLPs is similar to that of the CPMV virion structure (Lin et al., 1999). The CPMV protomer contains 587 aa residues and 60 copies of the protomer form an icosahedral particle with T = 1 symmetry and an average diameter of ~310 Å. However, the same (T = 1) icosahedral architecture can be described as displaying pseudo T = 3 (pT = 3) icosahedral symmetry, as the three β-barrel domains of the protomer occupy comparable positions seen in a T = 3 quasi-equivalent surface lattice (Figures 1A and 1B). Each protomer undergoes proteolytic processing at residues Gln374-Gly375 by the 24K viral protease prior or during assembly generating the large (L) and small (S) subunits (Figure 1C). In the eVLP structure, the L subunit (41 kDa) is composed of ordered residues 1–369 that form two jellyroll β-barrel domains, while the S subunit (23 kDa) consisting of ordered residues 375–567 (1–193) forms a single β-barrel domain. The β barrel (β3) of the small subunit is located near the 5-fold axis, with the β barrels (β1 and β2) of the large subunit surrounding the 3-fold axis (Figures 1A and 1B).

Comparison of CPMV eVLP and Virion Structures

The crystal and cryo-EM structures of eVLP show a high degree of similarity with each another and also with the CPMV virion crystal structure. The structures have an overall root-mean-square deviation (rmsd) of 0.518 Å between eVLP-X-ray and eVLP-EM; 0.527 Å between eVLP-X-ray and CPMV virion (X-ray); and 0.445 Å between eVLP-EM and CPMV virion (X-ray) for 550 aligned Cα atoms (Figure 3A). The structure superpositions and calculation of rmsd values were carried out using the UCSF Chimera program (Pettersen et al., 2004) and the graphics program O (Jones et al., 1991). Figures 4A–4F show the rmsd values for the large and small subunits between the crystal and cryo-EM structures. Other than at the C terminus of the S subunit, minor differences are observed at the residue Lys176 of the large subunit and Asp418 (44) of the small subunit between the eVLP and virion structures due to differences in rotamer and alternate backbone conformations, respectively (Figures 4G and 4H). However, the similar side chain conformation seen for the residue Lys176 in both the eVLP structures compared with the virion structure (Figure 4G) and its location on the capsid interior, near the icosahedral 2-fold axis, suggests that it might potentially interact with genomic RNA in CPMV virions. Other small structural differences with rmsds in the range of 1–2 Å occur due to small changes in the residue locations (Figure S2). Notwithstanding these differences, the eVLP and virion structures of CPMV appear to be very similar. We also compared the average B factors per residue of eVLP and virion structures. Even though, the magnitudes of B factors differ between the two capsids, the variation of B factors along the polypeptide chain is very similar between the two structures (Figure S3). This suggests that there are no significant differences between in the dynamics of eVLPs and virions, influenced by the absence or presence of genomic RNA, respectively.

Figure 3. Superposition of CPMV Protomers.

Figure 3

Open in a new tab

(A) A superposition of the eVLP-X-ray (yellow), eVLP-cryoEM (blue), and native CPMV virion-X-ray (1NY7; green) structures.

(B) Close-up view of the superposition at the C termini of the S subunits from three structures. Residues are labeled in corresponding colors of the structures. A dashed line represents disordered residues, 558–563 (184–189) in the eVLP-cryoEM structure.

Figure 4. Graphs Showing the Root-Mean-Square Deviation Values between the Crystal and cryoEM Structures Highlighting Structural Similarities and Differences.

Figure 4

Open in a new tab

(A–C) The rmsd values at each residue between the L subunits of the eVLP-Xray and CPMV (virion)-Xray (1NY7), eVLP-Xray and eVLP-cryoEM, and eVLP-cryoEM and CPMV-Xray (1NY7) structures, respectively. The Cα atom deviations are shown by blue line and the average deviation of all atoms in each residue is indicated by a red line.

(D–F) The rmsd values at each residue between the S subunits for the corresponding pairs as in graphs (A–C), respectively. The residue numbers of the S subunits in graphs (D–F) are shown from 1 to 193, which corresponds to residues 375–567.

(G) The side-chain (rotamer) differences at residue Lys176 of the L subunit between the three structures: eVLP-Xray (yellow), eVLP-cryoEM (blue), and CPMV-Xray (1NY7; green).

(H) The conformational differences in the backbone of the structures at residue Asp418 (44) of the S subunit.

The major difference between the X-ray and cryo-EM structures of eVLPs is seen at the C terminus of the S subunit. A helical segment of residues 564–576 (190–202) is ordered only in the eVLP cryo-EM structure, with preceding residues 558–563 (184–189) and succeeding residues 577–587 (203–213) that could not be modeled due to weak or disordered density (Figure 3B) (Hesketh et al., 2015). In contrast, the C termini of the S subunit are continuously ordered until residues Arg567 (193) and Leu563 (189) in the crystal structures of eVLPs and CPMV virion, respectively (Figure 3B), and in spite of different conditions used to grow crystals of eVLPs (pH 4.7) and virions (pH 7.0). Of note, the cryo-EM structure of eVLPs was determined at pH 7.0 (Hesketh et al., 2015). Furthermore, the coiled structure formed by the visible C-terminal residues 557–567 of the S subunit in the eVLP crystal structure is consistent with secondary structure predictions (Buchan et al., 2013) (Figure S1). Notably, the ordered helical segment of residues 568–576 (194–202) in the cryo-EM structure spatially superimposes onto residues 562–567 (188–193) in the eVLP crystal structure (Figure 3B). These structural differences seen at the C terminus of the S subunit between the crystal and cryo-EM structures of the eVLPs may be attributable to different pH values of 4.7 and 7.0 and/or different extents of proteolysis undergone by the eVLPs used in the respective studies.

Mass Spectrometry and Proteomics

The S subunit of CPMV eVLPs and virions is present in slow and fast electrophoretic forms. The slow form contains the full-length S subunits, while the fast form consists of proteolysed form(s) of S subunits. In freshly prepared samples of eVLPs and virions, which were dialyzed into 50 mM potassium phosphate buffer (pH 7.0), both the slow and fast forms are present (Figure 5A). However, within 2–3 weeks after purification of the eVLPs stored at 4°C, the S subunit contains mostly the fast (cleaved) form (Sainsbury et al., 2011). The extent of proteolysis is also found to be different based on the type of plant host used for the expression of CPMV or eVLPs, such as cowpea (Vigna unguiculata) compared with N. benthaminana (Figure 5B). Mass spectrometry and proteomics analysis of the electrophoretic slow and fast forms (bands) of the S subunit reveal that the slow (uncleaved) form contains the native C terminus ending with the sequence Thr-Ala-Ala (Figure 5C), while the fast form contains peptides with different C termini ending with residue Leu563 (189), Leu564 (190), or Phe566 (192). The proteomics analysis on the fast form of S subunit of CPMV virions has identified proteolysis at Ph566 in addition to Leu563 and Leu564 (Figure 5C). It is important to emphasize that the slow electrophoretic band of the S subunit contains only the full-length S subunit (213 aa), unlike the fast form that displays different C termini and is composed of a varying number of amino acids between 189 and 192.

Figure 5. SDS-PAGE and Proteomic Analysis of CPMV eVLPs and Virions.

Figure 5

Open in a new tab

(A) CPMV eVLPs, purified from N. benthamiana: lane 1, molecular weight marker; lane 2, CPMV eVLPs fresh preparation; and lane 3, CPMV eVLPs after storage at 4°C for more than 1 month (less slow forms of S subunit). The L subunit has a molecular weight of 41 kDa and the S subunit of 23 kDa. The lower case s and f letters (in gray) refer to slow (uncleaved) and fast (cleaved) bands of the S subunit.

(B) CPMV virion: lane 1, molecular weight marker; lanes 2–4, samples purified from N. benthamiana; and lanes 5–7, samples purified from cowpea plants. Different levels of proteolytic cleavage of S subunit are observed depending on the host plant. Samples from N. benthamiana look more intact compared with samples from cowpea plants. The lower case s and f letters (in gray) refer to slow (uncleaved) and fast (cleaved) bands of the S subunit.

(C) List of identified peptides of the S subunit of CPMV virus and eVLPs after GluC protease digestion using proteomics mass spectrometry analysis. The cleavage site is indicated by a period (dot). The slow and fast forms of the S subunit were individually digested and the peptide sequences were identified. The cleavages at non-glutamic acid residues downstream of the cleavage site are due to non-specific cleavages. The cases where both ends of the peptide do not contain glutamic acid residues represent termini of the protein. Selected amino acids are identified by their residue numbers. The sequence ending with TAA (587) corresponds to the native C terminus of the full length of the S subunit (slow form).

DISCUSSION

CPMV particles are being used as reagents for various biomedical and nano-technology applications. More recently, these applications have turned to using eVLPs rather than the infectious virus (Lebedev et al., 2016; Sainsbury et al., 2011, 2014; Wen et al., 2012). Therefore, it is of importance to structurally characterize eVLPs and analyze any significant differences that occur between the wild-type virus and eVLPs of CPMV. Previous studies have shown that the C-terminal 24 residues of the S subunit are important for particle assembly (Hesketh et al., 2015; Sainsbury et al., 2011), RNA encapsidation (Taylor et al., 1999), and the suppression of gene silencing (Canizares et al., 2004). Based on the location of the visible C terminus (567) of the S subunit (Figure 1B), the C-terminal 24 residues are likely to play a crucial role in stabilizing the inter-protomer contacts around the 5-fold axis during assembly. These protomer-protomer associations lead to the formation of the pentamers of the protomers, and 12 such pentamers assemble together to form CPMV particles. Since RNA encapsidation of CPMV is likely to involve assembling of pentamers around the viral RNA, the C-terminal residues would also play an indirect role in RNA packaging. This assessment is in agreement with the assembly mechanism proposed by other cryo-EM studies (Hesketh et al., 2015). After assembly, the C termini of S subunits of CPMV virions are prone to proteolysis and lose up to 24 residues without adversely affecting the particle integrity (Geelen et al., 1973; Niblett and Semancik, 1970; Taylor et al., 1999). In freshly prepared samples, a significant proportion of the C termini of the S subunits are intact (Sainsbury et al., 2011), resulting in the occurrence of slow and fast electrophoretic forms of CPMV particles (Niblett and Semancik, 1969). However, most of the S subunits lose up to 24 aa from their C termini by about 2–3 weeks after the purification, when stored at 4°C (Sainsbury et al., 2011).

The proteomics analysis on different electrophoretic forms (bands) of the S subunits of CPMV eVLPs as well as of virions suggests that the fast (cleaved) form of S subunits contains different or multiple C-terminal residues: Leu563 (189), Leu564 (190), and Phe566 (192), as opposed to a single cleavage site as previously reported (Taylor et al., 1999). The previous results were designed to ascertain the limit of cleavage and were obtained from a preparation of virions, produced in cowpea, which was deliberately aged for several weeks. Furthermore, the technique used, electrospray mass spectrometry of the entire S protein, did not permit the resolution of the results reported here. The observation that the fast form contains different C-terminal residues implies that the CPMV eVLPs or virions might consist of the S subunits with their C termini processed to varying lengths. However, the final cleavage site cannot be too far from Leu563 (189), otherwise the residues would be surface exposed and susceptible to further proteolysis. This is consistent with the observation that none of the peptides from the fast form of the S subunit contain C-terminal residues beyond Phe566 (192) (Figure 5C).

The superposition of eVLP structures onto the virion structure of CPMV reveals a high degree of structural similarity with the exception of differences at the C terminus of the S subunit (Figure 3A). This shows that the CPMV eVLPs are able to assemble independently in the absence of nucleic acid and form capsids virtually identical to CPMV virion particles. The eVLP crystal structure shows continuous electron density until Arg567 (193), four residues beyond the last cleavage seen at residue Leu563 (189) of the S subunit, identified by proteomics analysis (Figure 5C). A bifurcated salt bridge is observed in the eVLP crystal structure between residue Glu559 (185) at the C terminus of the reference S subunit and residues Arg469 (95) and Asp478 (104) of the S subunit of the 5-fold-related (clockwise) protomer that stabilizes the pentameric interface. The identification of a few additional residues beyond the limit of the cleavage site found in virions indicates that the X-ray structure represents an average structure of eVLPs that are composed of cleaved, partially cleaved, and possibly uncleaved forms of the S subunit. The remaining residues beyond Arg567 (193) are either proteolysed and totally absent from the crystals used for structure determination or, if present, are disordered in the crystal structure. In addition, the identification of peptide fragments with residues downstream of Leu563 (189) in the proteomics analysis (Figure 5C) suggests that the proteolysis may be either sequential in nature or may occur at several different sites and independently on different S subunits, as there are 60 such sites per particle. It is likely that after particle assembly, the C termini of the small subunit are continually proteolysed by host proteases at various basic or aromatic residues until the last cleavable/accessible residue Leu564 (190) or a residue close to it is reached.

In the cryo-EM structure of eVLPs, a helical segment of 13 residues 564–576 (190–202) of the S subunit is found to be ordered beyond the last proteolysed residue Leu563 (189) (Hesketh et al., 2015). However, some of the residues, Leu564 (190)-Arg569 (195), in this peptide segment that are known to undergo proteolysis, according to above proteomics results, are buried and inaccessible in the cryo-EM structure. Hence, if these terminal residues as seen in the cryo-EM structure were to become accessible and undergo proteolytic processing, it would require conformational reorganization of the C terminus. Indirect evidence that such a conformational change occurs on proteolysis is provided by the observation that deliberate digestion of eVLPs with chymotrypsin, which cuts after aa 565 (191), 567 (193), and 569 (195) of the S protein, sites similar to those identified in this study, significantly enhances their permeability to ions compared with the untreated eVLPs (Sainsbury et al., 2011).

The results presented in this manuscript demonstrate that the overall structure of eVLPs is extremely similar to that of virions produced via infection, whose crystal structure was determined previously (Lin et al., 1999). Furthermore, based on the residue-wise B factor comparisons along the polypeptide chain, the particle dynamics of CPMV virions and eVLPs appear to be similar irrespective of whether or not the genomic RNA is packaged (Figure S3). With the exception of the C terminus of the S subunit, the crystal structure of eVLPs reported in this study shows close agreement with the independently obtained cryo-EM structure (Hesketh et al., 2015). Taken together, the information obtained from X-ray and cryo-EM structures along with the proteomics analysis of eVLPs has provided new insights into the structure and processing of the C terminus of the S subunit of CPMV as well as particle assembly. Such knowledge is important for developing eVLPs for biotechnological applications as this region has been known to control assembly and accessibility (Sainsbury et al., 2011).

EXPERIMENTAL PROCEDURES

Expression, Purification, and Crystallization of CPMV eVLPs

CPMV eVLPs were expressed and purified as previously described (Sainsbury et al., 2014). CPMV eVLPs, at a concentration of 6.0 mg/ml, were crystallized by hanging-drop vapor diffusion by mixing 2 μl of eVLPs with 2 μl of reservoir solution containing 3%–4% (w/v) PEG3350, 0.3 M ammonium sulfate, 5% (v/v) glycerol, and 0.1 M sodium acetate (pH 4.68), and stored at room temperature (20°C). Diamond-shaped crystals appeared within 1–2 days and grew to a maximum size of ~0.3 mm in about 1–2 weeks. The crystals were sequentially transferred into a reservoir solution containing 10%, 15%, and 20% (v/v) and flash cooled in liquid nitrogen prior to data collection.

Data Collection, Structure Determination, Model Building, and Refinement

The diffraction data from the CPMV eVLP crystals were collected on a Pilatus detector at the beam-line 23-ID-D, GMCA at APS. The crystals diffracted better than 2.3 Å resolution. The data were processed using HKL2000 (Otwinowski and Minor, 1997). The final data scaling and merging of the unmerged data derived from SCALEPACK (HKL2000) was done using the SCALA program (Evans, 2006). All the reflections with partiality >0.6 were included in the dataset. Details of the data statistics are shown in Table 1 and Table S1. The CPMV eVLP crystals belong to the space group I41 with unit cell dimensions a = b = 655.97 Å, c = 571.45 Å, α = β = γ = 90°, and contain two virus particles in the crystallographic asymmetric unit (16 particles in the unit cell), resulting in 120-fold NCS. The CPMV eVLP structure was determined by the molecular replacement method employing the program Phaser (McCoy et al., 2007) and using the polyalanine model derived from the CPMV structure (PDB: 1NY7). The details of the particle orientations and positions are given in Table 1. The latter polyalanine model was oriented and positioned according to the Phaser solution and rigid body refined using the program X-PLOR (Brunger, 1992). The initial electron density was generated with the phases from the rigid-body-refined polyalanine model and improved by employing the phase-refinement procedure with 120-fold NCS averaging using the RAVE and CCP4 suite of programs (CCP4, 1994; Jones, 1992). This resulted in a readily interpretable map at 2.3 Å with a final R factor of 0.351 and a correlation coefficient of 0.68, calculated between the Fo and the Fc values derived from back-transformed electron density. An initial model of the CPMV protomer that occupies the icosahedral asymmetric unit was built automatically using the BUCCANEER program (Cowtan, 2008). The final model was adjusted manually into the electron density map contoured at 1.0σ using the programs COOT (Emsley et al., 2010) and O (Jones et al., 1991). The structure was refined using the program CNS imposing (strict) 120-fold NCS constraints (Brunger et al., 1998). Water molecules were identified using the program WATPEAK (CCP4, 1994) based on the difference (Fo – Fc) densities and later manually verified and adjusted using the graphics program COOT (Emsley et al., 2010). Finally, 58 water molecules were included in the eVLP structure. The final refinement statistics are shown in Table 1. Model quality was analyzed using PROCHECK (Laskowski et al., 1993), and Stride (Heinig and Frishman, 2004) was used for the assignment of secondary structures.

The high model R factor is likely due to imposition of 120-fold NCS constraints, resulting from two particles present in the crystallographic asymmetric unit of the high-symmetry space group (I41). To improve the R factor, it may be necessary to release the strict NCS constraints and impose them as restraints at high resolutions (e.g., 2.3 Å), particularly when more than one particle is present in the crystallographic asymmetric unit. This phenomenon is even more pronounced in a high-symmetry space group relative to low-symmetry space group P1 as seen in a few cases (Khayat et al., 2011; Tars et al., 2000a, 2000b). Even though NCS restraints were used in the refinement of simple (T = 1) viruses, where 60-fold NCS was present (Larson et al., 2014), it is difficult to implement the same with 120-fold NCS because of a greater number of chains (120 × 2 = 240) involved. Finally, this may not significantly improve the model except for slightly better statistics (i.e., R factor), as the model R factor (0.359) is very similar to the molecular replacement R factor (0.351) calculated between the Fo and the Fc derived from back-transformed electron density obtained by imposing 120-fold NCS. We tested if the imposition of 120-fold NCS constraints is the reason for the high R factor at 2.3 Å, by calculating the R factors at different high-resolution limits of the data. The R factors calculated at various high-resolution limits of 3, 4, and 5 Å yielded increasingly lower R factors 0.325, 0.294, and 0.291, respectively. This indicates that the imposition of 120-fold strict NCS is likely the cause of relatively higher R factors at 2.3 Å resolution.

Mass Spectrometry and Proteomics

The capsid protein subunits (L and S) of CPMV virus particles from Cowpea and CPMV eVLPs were separated on an SDS-PAGE gel as previously described (Sainsbury et al., 2011). The S subunit of CPMV virions and that of eVLPs contained two bands corresponding to the slow and fast forms (Figures 5A and 5B). The bands corresponding to the slow and fast forms of the S subunit of eVLPs and of the fast form (only) of the virions were excised and submitted for proteomics mass spectrometry analysis by digestion with GluC protease. The excised gel bands were destained with a mixture of a 50:50 (v/v) dilution of acetonitrile-25 mM ammonium bicarbonate and dried with a SpeedVac system. The protein samples in the gel were then reduced in 25 μl of 10 mM d,l-DTT (Sigma) for 1 hr and alkylated with 25 μl of 55 mM iodoacetamide (Sigma) for 30 min in the dark prior to an 18-hr GluC digestion at 37°C using a 1:30 (w/w) enzyme-to-substrate ratio. The resulting peptides were extracted twice with a 50:45:5 (v/v) volume of acetonitrile-water-formic acid and concentrated to 30 μl before being analyzed. Peptides were analyzed by reverse-phase chromatography prior to mass spectrometry analysis and tandem mass spectrometry samples were analyzed by using Mascot (version 2.1.04; Matrix Science) as described in Powers et al. (2011).

Supplementary Material

1
2

Highlights.

  • The crystal structure of eVLPs is extremely similar to that of CPMV virions

  • After assembly, the C termini of S subunits undergo proteolysis at multiple sites

  • Varied composition of processed S subunits ensues slow and fast electrophoretic forms

Acknowledgments

We acknowledge support from NIH grant AI070771 (to V.S.R.), UK Biotechnological and Biological Sciences Research Council (BBSRC) grant BB/L020955/1, the Institute Strategic Program grant “Understanding and Exploiting Plant and Microbial Secondary Metabolism” (BB/J004596/1) and the John Innes Foundation (to G.P.L). GM/CA has been funded with US federal funds from the NIH (Y1-CO-1020, Y1-GN-11040) and the APS is supported by the US Department of Energy under contract DE-AC02-06CH11357. G.P.L. declares that he is a named inventor on granted patent WO 29087391 A1, which describes the transient expression system used in this manuscript to express eVLPs and also a named inventor on a filed patent “Production of viral capsids” PBL. US 61/186,970 which describes the production and uses of CPMV eVLPs.

Footnotes

AUTHOR CONTRIBUTIONS

P.S., Y.M., and G.P.L. provided the reagents. P.S., N.H., and Y.C.K. crystallized CPMV eVLPs. N.H., J.E.J., and V.S.R. determined and analyzed the eVLP crystal structure and compared it with other CPMV structures. E.L.M. and N.A.R. determined the cryo-EM structure. L.H., N.H., and V.S.R. performed and analyzed mass spectrometry proteomics experiments. N.H., P.S., Y.M., G.P.L., and V.S.R. wrote and prepared the manuscript.

ACCESSION NUMBERS

The coordinates and structure factor amplitudes of the eVLP crystal structure have been deposited in the PDB (PDB: 5FMO).

SUPPLEMENTAL INFORMATION

Supplemental Information includes three figures and one table and can be found with this article online at http://dx.doi.org/10.1016/j.str.2016.02.011.

References

  1. Aljabali AA, Barclay JE, Butt JN, Lomonossoff GP, Evans DJ. Redox-active ferrocene-modified Cowpea mosaic virus nanoparticles. Dalton Trans. 2010;39:7569–7574. doi: 10.1039/c0dt00495b. [DOI] [PubMed] [Google Scholar]
  2. Aljabali AA, Shah SN, Evans-Gowing R, Lomonossoff GP, Evans DJ. Chemically-coupled-peptide-promoted virus nanoparticle templated mineralization. Integr Biol (Camb) 2011;3:119–125. doi: 10.1039/c0ib00056f. [DOI] [PubMed] [Google Scholar]
  3. Aljabali AA, Barclay JE, Steinmetz NF, Lomonossoff GP, Evans DJ. Controlled immobilisation of active enzymes on the cowpea mosaic virus capsid. Nanoscale. 2012;4:5640–5645. doi: 10.1039/c2nr31485a. [DOI] [PubMed] [Google Scholar]
  4. Aljabali AA, Shukla S, Lomonossoff GP, Steinmetz NF, Evans DJ. CPMV-DOX delivers. Mol Pharm. 2013;10:3–10. doi: 10.1021/mp3002057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brunger AT. X-PLOR, version 3.1: a System for X-ray Crystallography and NMR. Yale University Press; 1992. [Google Scholar]
  6. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, et al. Crystallography & NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr. 1998;54:905–921. doi: 10.1107/s0907444998003254. [DOI] [PubMed] [Google Scholar]
  7. Buchan DW, Minneci F, Nugent TC, Bryson K, Jones DT. Scalable web services for the PSIPRED protein analysis workbench. Nucleic Acids Res. 2013;41:W349–W357. doi: 10.1093/nar/gkt381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Canizares MC, Taylor KM, Lomonossoff GP. Surface-exposed C-terminal amino acids of the small coat protein of Cowpea mosaic virus are required for suppression of silencing. J Gen Virol. 2004;85:3431–3435. doi: 10.1099/vir.0.80454-0. [DOI] [PubMed] [Google Scholar]
  9. CCP4. The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr. 1994;50:760–763. doi: 10.1107/S0907444994003112. [DOI] [PubMed] [Google Scholar]
  10. Cowtan K. Fitting molecular fragments into electron density. Acta Crystallogr D Biol Crystallogr. 2008;64:83–89. doi: 10.1107/S0907444907033938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dalsgaard K, Uttenthal A, Jones TD, Xu F, Merryweather A, Hamilton WD, Langeveld JP, Boshuizen RS, Kamstrup S, Lomonossoff GP, et al. Plant-derived vaccine protects target animals against a viral disease. Nat Biotechnol. 1997;15:248–252. doi: 10.1038/nbt0397-248. [DOI] [PubMed] [Google Scholar]
  12. Destito G, Schneemann A, Manchester M. Biomedical nano-technology using virus-based nanoparticles. Curr Top Microbiol Immunol. 2009;327:95–122. doi: 10.1007/978-3-540-69379-6_5. [DOI] [PubMed] [Google Scholar]
  13. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr. 2010;66:486–501. doi: 10.1107/S0907444910007493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Evans P. Scaling and assessment of data quality. Acta Crystallogr D Biol Crystallogr. 2006;62:72–82. doi: 10.1107/S0907444905036693. [DOI] [PubMed] [Google Scholar]
  15. Franssen H, Goldbach R, Broekhuijsen M, Moerman M, van Kammen A. Expression of middle-component RNA of cowpea mosaic virus: in vitro generation of a precursor to both capsid proteins by a bottom-component RNA-encoded protease from infected cells. J Virol. 1982;41:8–17. doi: 10.1128/jvi.41.1.8-17.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Geelen JL, Rezelman G, van Kammen A. The infectivity of the two electrophoretic forms of cowpea mosaic virus. Virology. 1973;51:279–286. doi: 10.1016/0042-6822(73)90428-5. [DOI] [PubMed] [Google Scholar]
  17. Heinig M, Frishman D. STRIDE: a web server for secondary structure assignment from known atomic coordinates of proteins. Nucleic Acids Res. 2004;32:W500–W502. doi: 10.1093/nar/gkh429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hesketh EL, Meshcheriakova Y, Dent KC, Saxena P, Thompson RF, Cockburn JJ, Lomonossoff GP, Ranson NA. Mechanisms of assembly and genome packaging in an RNA virus revealed by high-resolution cryo-EM. Nat Commun. 2015;6:10113. doi: 10.1038/ncomms10113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jones TA. A set of averaging programs. In: Dodson EJ, Gover S, Wolf W, editors. In Molecular Replacement. SERC Daresbury Laboratory; 1992. [Google Scholar]
  20. Jones TA, Zou JY, Cowan SW, Kjeldgaard M. Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallogr. 1991;47:110–119. doi: 10.1107/s0108767390010224. [DOI] [PubMed] [Google Scholar]
  21. Khayat R, Brunn N, Speir JA, Hardham JM, Ankenbauer RG, Schneemann A, Johnson JE. The 2.3-angstrom structure of porcine circovirus 2. J Virol. 2011;85:7856–7862. doi: 10.1128/JVI.00737-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Larson SB, Day JS, McPherson A. Satellite tobacco mosaic virus refined to 1.4 A resolution. Acta Crystallogr D Biol Crystallogr. 2014;70:2316–2330. doi: 10.1107/S1399004714013789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Laskowski RA, MacArthur MW, Thornton JM. PROCHECK – a program to check the stereochemical quality of protein structures. J Appl Crystallogr. 1993;26:283–291. [Google Scholar]
  24. Lebedev N, Griva I, Dressick WJ, Phelps J, Johnson JE, Meshcheriakova Y, Lomonossoff GP, Soto CM. A virus-based nanoplasmonic structure as a surface-enhanced Raman biosensor. Biosens Bioelectron. 2016;77:306–314. doi: 10.1016/j.bios.2015.09.032. [DOI] [PubMed] [Google Scholar]
  25. Lin T, Chen Z, Usha R, Stauffacher CV, Dai JB, Schmidt T, Johnson JE. The refined crystal structure of cowpea mosaic virus at 2.8 A resolution. Virology. 1999;265:20–34. doi: 10.1006/viro.1999.0038. [DOI] [PubMed] [Google Scholar]
  26. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ. Phaser crystallographic software. J Appl Crystallogr. 2007;40:658–674. doi: 10.1107/S0021889807021206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Montague NP, Thuenemann EC, Saxena P, Saunders K, Lenzi P, Lomonossoff GP. Recent advances of cowpea mosaic virus-based particle technology. Hum Vaccin. 2011;7:383–390. doi: 10.4161/hv.7.3.14989. [DOI] [PubMed] [Google Scholar]
  28. Niblett CL, Semancik JS. Conversion of the electrophoretic forms of cowpea mosaic virus in vivo and in vitro. Virology. 1969;38:685–693. doi: 10.1016/0042-6822(69)90187-1. [DOI] [PubMed] [Google Scholar]
  29. Niblett CL, Semancik JS. The significance of the coat protein in infection by the electrophoretic forms of cowpea mosaic virus. Virology. 1970;41:201–207. doi: 10.1016/0042-6822(70)90072-3. [DOI] [PubMed] [Google Scholar]
  30. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997;276:307–326. doi: 10.1016/S0076-6879(97)76066-X. [DOI] [PubMed] [Google Scholar]
  31. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem. 2004;25:1605–1612. doi: 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]
  32. Porta C, Spall VE, Loveland J, Johnson JE, Barker PJ, Lomonossoff GP. Development of cowpea mosaic virus as a high-yielding system for the presentation of foreign peptides. Virology. 1994;202:949–955. doi: 10.1006/viro.1994.1417. [DOI] [PubMed] [Google Scholar]
  33. Powers ME, Smith PA, Roberts TC, Fowler BJ, King CC, Trauger SA, Siuzdak G, Romesberg FE. Type I signal peptidase and protein secretion in Staphylococcus epidermidis. J Bacteriol. 2011;193:340–348. doi: 10.1128/JB.01052-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rae CS, Khor IW, Wang Q, Destito G, Gonzalez MJ, Singh P, Thomas DM, Estrada MN, Powell E, Finn MG, et al. Systemic trafficking of plant virus nanoparticles in mice via the oral route. Virology. 2005;343:224–235. doi: 10.1016/j.virol.2005.08.017. [DOI] [PubMed] [Google Scholar]
  35. Sainsbury F, Saunders K, Aljabali AA, Evans DJ, Lomonossoff GP. Peptide-controlled access to the interior surface of empty virus nanoparticles. Chembiochem. 2011;12:2435–2440. doi: 10.1002/cbic.201100482. [DOI] [PubMed] [Google Scholar]
  36. Sainsbury F, Saxena P, Aljabali AA, Saunders K, Evans DJ, Lomonossoff GP. Genetic engineering and characterization of Cowpea mosaic virus empty virus-like particles. Methods Mol Biol. 2014;1108:139–153. doi: 10.1007/978-1-62703-751-8_11. [DOI] [PubMed] [Google Scholar]
  37. Saunders K, Sainsbury F, Lomonossoff GP. Efficient generation of cowpea mosaic virus empty virus-like particles by the proteolytic processing of precursors in insect cells and plants. Virology. 2009;393:329–337. doi: 10.1016/j.virol.2009.08.023. [DOI] [PubMed] [Google Scholar]
  38. Shah SN, Steinmetz NF, Aljabali AA, Lomonossoff GP, Evans DJ. Environmentally benign synthesis of virus-templated, monodisperse, iron-platinum nanoparticles. Dalton Trans. 2009:8479–8480. doi: 10.1039/b906847c. [DOI] [PubMed] [Google Scholar]
  39. Stauffacher CV, Usha R, Harrington M, Schmidt T, Hosur M, Johnson JE. The structure of cowpea mosaic virus at 3.5Å resolution. In: Moras D, Drenth J, Strandberg B, Suck D, Wilson L, editors. Crystallography in Molecular Biology. Plenum; 1987. pp. 293–308. [Google Scholar]
  40. Steinmetz NF, Lomonossoff GP, Evans DJ. Cowpea mosaic virus for material fabrication: addressable carboxylate groups on a programmable nanoscaffold. Langmuir. 2006a;22:3488–3490. doi: 10.1021/la060078e. [DOI] [PubMed] [Google Scholar]
  41. Steinmetz NF, Lomonossoff GP, Evans DJ. Decoration of cowpea mosaic virus with multiple, redox-active, organometallic complexes. Small. 2006b;2:530–533. doi: 10.1002/smll.200500453. [DOI] [PubMed] [Google Scholar]
  42. Steinmetz NF, Shah SN, Barclay JE, Rallapalli G, Lomonossoff GP, Evans DJ. Virus-templated silica nanoparticles. Small. 2009;5:813–816. doi: 10.1002/smll.200801348. [DOI] [PubMed] [Google Scholar]
  43. Tars K, Fridborg K, Bundule M, Liljas L. Structure determination of bacteriophage PP7 from Pseudomonas aeruginosa: from poor data to a good map. Acta Crystallogr D Biol Crystallogr. 2000a;56:398–405. doi: 10.1107/s0907444900001232. [DOI] [PubMed] [Google Scholar]
  44. Tars K, Fridborg K, Bundule M, Liljas L. The three-dimensional structure of bacteriophage PP7 from Pseudomonas aeruginosa at 3.7-A resolution. Virology. 2000b;272:331–337. doi: 10.1006/viro.2000.0373. [DOI] [PubMed] [Google Scholar]
  45. Taylor KM, Spall VE, Butler PJ, Lomonossoff GP. The cleavable carboxyl-terminus of the small coat protein of cowpea mosaic virus is involved in RNA encapsidation. Virology. 1999;255:129–137. doi: 10.1006/viro.1998.9567. [DOI] [PubMed] [Google Scholar]
  46. Usha R, Rohll JB, Spall VE, Shanks M, Maule AJ, Johnson JE, Lomonossoff GP. Expression of an animal virus antigenic site on the surface of a plant virus particle. Virology. 1993;197:366–374. doi: 10.1006/viro.1993.1598. [DOI] [PubMed] [Google Scholar]
  47. Wen AM, Shukla S, Saxena P, Aljabali AA, Yildiz I, Dey S, Mealy JE, Yang AC, Evans DJ, Lomonossoff GP, et al. Interior engineering of a viral nanoparticle and its tumor homing properties. Biomacromolecules. 2012;13:3990–4001. doi: 10.1021/bm301278f. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS764299-supplement-1.pdf (2.6MB, pdf)
2
NIHMS764299-supplement-2.pdf (5.5MB, pdf)

RESOURCES