Electron cryomicroscopy and bioinformatics suggest protein fold models for rice dwarf virus
Z. Hong Zhou1, 2, Matthew L. Baker2, 3, Wen Jiang2, 3, Matthew Dougherty3, Joanita Jakana3, Gang Dong4, Guangying Lu4
& Wah Chiu2, 3
1 Department of Pathology and Laboratory Medicine, University of Texas−Houston Medical School, Houston, Texas 77030, USA.
2 Program in Structural and Computational Biology and Molecular Biophysics, Baylor College of Medicine, Houston, Texas 77030, USA.
3 National Center for Macromolecular Imaging, Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, Texas 77030, USA.
4 National Laboratory of Protein Engineering and Plant Genetic Engineering, College of Life Sciences, Peking University, Beijing 100871, China.
Correspondence should be addressed to Wah Chiu wah@bcm.tmc.edu
The three-dimensional structure of rice dwarf virus was determined to 6.8 Å resolution by single particle electron cryomicroscopy. By integrating the structural analysis with bioinformatics, the folds of the proteins in the double-shelled capsid were derived. In the outer shell protein, the uniquely orientated upper and lower domains are composed of similar secondary structure elements but have different relative orientations from that of bluetongue virus in the same Reoviridae family. Differences in both sequence and structure between these proteins may be important in defining virus−host interactions. The inner shell protein adopts a conformation similar to other members of Reoviridae, suggesting a common ancestor that has evolved to infect hosts ranging from plants to animals. Symmetry mismatch between the two shells results in nonequivalent, yet specific, interactions that contribute to the stability of this large macromolecular machine.
Rice dwarf virus (RDV) is a major pathogen of the rice plants in Southeast Asia, where the spread of RDV can have severe economic consequences. RDV, belonging to Reoviridae, is transmitted by Nephotettix cincticeps leafhoppers to susceptible plant species, replicating in both hosts and vectors. Viruses in the Reoviridae, whose hosts include plants, insects and animals, share a common mechanism of replication1,
2 but vary substantially in their structural organization3.
RDV, a double-shelled particle containing 12 segments of dsRNA, has a total protein mass of >26 MDa. Previous efforts in studying the RDV by electron cryomicroscopy at low resolution have illustrated the structural organization4,
5,
6,
7. The outer capsid shell contains 260 trimers of P8 (46 kDa) arranged on a T = 13l icosahedral lattice. The inner capsid shell is composed of 60 dimers of P3 (114 kDa) arranged on a T = 1 icosahedral lattice. Although this capsid organization is the same as that of the double-shelled particles of several other members of the Reoviridae, the sequence similarity to P3 and P8 is insignificant (identity <20%)4.
Overall structure With recent improvements in the data acquisition and processing of electron cryomicroscopy images8, determining the structure for RDV to a significantly higher resolution has become possible. The integration of sequence analysis and computational tools9 has allowed us to identify the secondary structure elements of both P3 and P8 and, consequently, construct a structural model for the double-shelled RDV capsid.
The computer power spectrum of a representative area from a close-to-focus micrograph of RDV (Fig. 1a) contained detectable contrast beyond 6 Å (Fig. 1b). The effective resolution of this reconstruction was assessed to be 6.8 Å (Fig. 1c), from which the structures of the outer and inner capsid shells were computationally segmented (Fig. 1d,e). Although the overall organization of the double-shelled capsid is similar to that of the transcriptionally competent particles of bluetongue virus (BTV)10 and animal reovirus11, which are circular, RDV is distinctly angular.
a, Representative area from a micrograph of RDV embedded in vitreous ice at -170 °C. Images were taken at 50,000 magnification in a JEOL 4000 electron cryomicroscope with an electron dose of 10−13 electrons Å-2. Individual particles were 'boxed' out into an array of 300 300 pixels. b, Power spectrum. The computed power spectrum was obtained by summing the Fourier transforms of 100 particle images from the micrograph shown in (a) with an underfocus of 0.6 m. c, Resolution assessment. The effective resolution consistently improved with increasing numbers of particles in the data set and reached 6.8 Å for the data set with 3,261 particles. d, Shaded surface representation of the 6.8 Å resolution density map of the outer capsid shell (P8) of RDV. The mass densities corresponding to inner shell proteins and RNA were computationally removed. The outer shell is composed of 132 trimers of P8 organized on a T = 13l icosahedral lattice. Each asymmetric unit contains four 1/3 unique trimers4, including one P (red), Q (orange), R (green), S (yellow) and 1/3 of T (blue). e, Surface representation of the inner capsid shell computationally extracted from the 6.8 Å resolution map. The inner shell consists of 60 P3 dimers, each containing one A (light green) and one B (light purple) P3 subunit. The maps are contoured at 2 (standard deviation) above the average density.
Outer capsid and P8 model The outer shell of each RDV capsid is made of five unique trimers (P, Q, R, S and T) (Fig. 1d) of protein P8 organized on a T = 13l icosahedral lattice. Although only the T trimer has icosahedral three-fold symmetry imposed during the reconstruction, all five trimers appeared very similar to each other. Additionally, all of the trimers correlated well with each other and displayed a high degree of three-fold symmetry with a correlation coefficient >0.85. Thus, to improve the signal-to-noise ratio, the P, Q, R and S trimers were aligned and averaged. The averaged trimer is 70 Å in height and 70 Å in diameter, with a monomeric subunit consisting of an upper and lower domain twisted 60° with respect to each other about the three-fold axis (Fig. 2a). The upper domain of each P8 subunit primarily contains thin, flat and contiguous densities, whereas the lower domain is mainly composed of rod-like densities (Fig. 2b).
Figure 2. Fold model of the outer shell protein P8.
a, Side view of an averaged trimer contoured at 4 with a single subunit highlighted in blue. The molecular boundaries between adjacent subunits were determined by interactively examining the continuity of mass densities at a high density threshold. b, Stereo view of the P8 monomer. The wire frame represents a low threshold (4 ), whereas the solid density represents a high threshold (6 ), illustrating the ability to identify structural features at this resolution. The view direction is 180° from (a) about the three-fold axis. The three layers of -sheets are evident in the upper domain of the P8 monomer. c, P8 subunit and helices identified by HELIXHUNTER9. A computationally isolated P8 subunit is shown with a wire frame representation. Assigned helices from HELIXHUNTER, shown as 5 Å diameter cylinders, are superimposed on the monomer. Seven of these helices (green) were identified with high degree of confidence, whereas two relatively short helices (orange) had low HELIXHUNTER correlation values. Helix 5 and 8 (orange) were fairly short, and the density in this region was not well resolved, as judged by a low HELIXHUNTER correlation value. d, P8 subunit and a ribbon diagram of the homologous -sandwich motif from BTV VP7. The location of the homologous fold was determined by FOLDHUNTER9. e, A proposed model for the fold of P8. A black line denotes the boundary of the P8 subunit. Helices are represented using the same criterion as described in (c), whereas strands of the sheets are displayed as blue arrows. The -strands are modeled based on the arrangement of strands from BTV VP7 and, thus, are speculative. Solid red lines represent the connectivity of the helices, whereas the dashed red lines represent the speculative -strand connections. The asterick in (e) refers to the putative hinge region between the upper and lower domain of P8.
Using HELIXHUNTER9, we identified and localized nine helices in the lower domain of P8 (Fig. 2c). To identify the corresponding amino acid sequences to these helices, we correlated their relative spatial locations and lengths of helical segments with those predicted from multiple secondary structure prediction methods (PsiPred12, PHD13 and SSPRO14). Although the locations of the nine consensus helices identified by secondary structure prediction in the sequence of P8 (Table 1) are consistent, the exact length of the predicted helices varies by up to seven residues. The assignment of sequence to the putative helices from the HELIXHUNTER analysis (Table 1) was proposed based on the match of helix length and continuity of the mass density between sequential helices at various contour levels (Fig. 2e). In the N-terminus, five helices (1−5) and their connections could be assigned. However, in the C-terminal helices, the mass densities of only helices 6, 7 and 9 are contiguous. Helix 8 was left unassigned, because it was poorly resolved in sequence and structure as well as not connected to the densities at the C-terminal region. This helix assignment was further supported by the match in the sequential ordering to that of the helices in the BTV outer capsid protein VP7, as assessed by two commonly used methods of comparing protein structures9,
15,
16. For instance, six of the helices in RDV P8 (helices 1, 2, 4, 6, 7 and 9) matched well, with <5 Å root mean square (r.m.s) deviation based on helical centroids, to the corresponding helices in the BTV VP7 (Table 1). The N-terminal helices appeared visually to match the corresponding BTV helices better than those of the C-terminus, suggesting a domain movement in regards to the N- and C-terminal domains.
The secondary structure prediction methods also showed that the middle region of the P8 sequence (128−315) contained -strands. However, the variability of the predictions within this region is considerably higher than the helix-rich terminal regions. Using fold recognition17, this region was predicted to be structurally homologous to the -sandwich domain of the BTV VP7 (ref. 4). Although individual -strands cannot be resolved in our 6.8 Å map, the flat, continuous shapes in the upper domain exhibit features typical of -sheets at this resolution (Fig. 2b). Using FOLDHUNTER9, the homologous -sandwich fold was localized to the upper domain of the RDV P8 subunit (Fig. 2d). Although the overall match appears reasonable, an obvious mismatch in the outermost portion of the P8 upper region can be mapped back to a corresponding difference between the -sheet domain of RDV P8 and BTV VP7 protein sequences4. This structural difference in the outermost region of P8 suggests a possible involvement in specific host cell receptor recognition, attachment and/or entry.
From both sequence and structural analysis, we proposed a fold model for RDV P8 (Fig. 2e). Although the placement of the strands and their connectivity within the upper domain is speculative, the fold of the lower helical domain is substantiated by the well-resolved helices, their connectivities evident at a lower density contour, the agreement with sequence-based secondary structure prediction and the match with the corresponding secondary structure elements from BTV VP7 (Table 1). Although the two-domain architecture of P8 is similar to VP7 of BTV, the relative orientation of these domains is different from both structural isoforms of VP7: REF and HEX, which exist in two different crystal forms18,
19. P8 appears to be in a conformation with less twist (60°) between the upper and lower domains than the REF isoform (120°) but more than the HEX isoform (0°) of VP7. This difference may be influenced by variances in flexible segments within the hinge region (asterisk, Fig. 2e). Additionally, the subunit arrangement results in a funnel-shaped opening at the top of the trimer, which is in contrast to the fully filled density on the top surface of the corresponding trimer of BTV. This unique conformation may facilitate the subunit interactions required for capsid stability while maintaining the requisite surface for host receptor interactions.
Inner capsid and P3 model Underlying the trimers of the outer shell are 60 P3 dimers forming a T = 1 inner shell (Fig. 1e). The structures of the A and B subunits of P3 dimer in RDV were graphically segmented from the inner capsid. Both P3 A and B are essentially planar, 150 Å long and 25−35 Å thick, with an overall crescent shape. P3 can be partitioned into three domains, referred to as the apical (close to the five-fold axis), carapace (plate-like) and dimerization (close to the two-fold axis) domains in BTV10. A gross structural similarity to other reovirus inner capsid proteins was evident.
Using HELIXHUNTER9, 28 helices in each of the subunits in the asymmetric unit of the inner shell were identified (Fig. 3a,b). However, the exact lengths and positions of the corresponding helices in P3A and P3B varied slightly, reflecting the conformational variations between the subunits as also seen in both BTV and animal reovirus. As with P8, the three aforementioned secondary structure prediction methods yielded similar lengths and sequence locations of the predicted helices (Fig. 3g). The sequence-predicted helices were then correlated with those visualized in our structure as described for P8 (Fig. 3a,b). Additionally, four -sheets were visually apparent in the dimerization domain of both subunits (Fig. 3e). One extra -sheet was seen in the carapace domain of the A subunit.
Figure 3. Fold model of the inner shell protein P3.
a,b, P3 and HELIXHUNTER9 results. The identified helices, represented as cylinders of 5 Å diameter, are shown superimposed on the surface views of P3A and P3B structures. Length, resolvability, position, connectivity and relative agreement with consensus secondary structure prediction were used to qualitatively judge the fidelity of helix assignment. The less reliable helices are colored in orange. Helices 13, 14 and 15 were better ordered in P3A than P3B. Helix 20 was observed only in P3A, whereas helix 21 was resolved only in P3B. Helices in the dimerization domain (helices 27, 29 and 30) were in regions dominated by -sheets and loops and, thus, more speculative in assignment. c,d, Proposed folds for P3A and P3B, respectively. The -sheets are represented by light blue parallelograms. The red lines represent the connectivity between the secondary structure elements. Note that some of the connectivities through the -sheets are drawn as dotted lines because the individual strands are not resolved. The numbers on the helices correspond to those identified in (g). e,f, Representative P3A densities. Regions from the -sheet-rich dimerization domain (e) and the predominantly helical carapace domain (f) are enlarged and rendered using two thresholds: 2 for the wire frame and 3 for the solid density. g, P3 secondary structural prediction using SSPRO14. Shown is the assignment of the predicted helices with respect to the identified helices in our 3D map. The bracketed region (red) refers to the insertion sequence compared to the corresponding region in VP3 of BTV.
An overall similarity between the arrangement of secondary structure elements within P3A and P3B subunits to the corresponding subunits in BTV10 and animal reovirus core11 was evident and, thus, facilitated the connection of the observed structural elements (Fig. 3c,d). In addition, the connectivity of the helices is apparent in our map (Fig. 3f). The proposed connectivity (red line, Fig. 3c,d) between helices, based on the BTV and animal reovirus core proteins, is consistent with the observed densities in our map, whereas the connections of the -strands (dotted red line, Fig. 3c,d) are not well resolved and, thus, more speculative.
The observed differences in the detailed spatial arrangement of the secondary structure elements between P3A and P3B may be related to their different functional and structural roles. The apical domains of P3 A and B, although highly similar, exhibit differences in a stretch of three short helices (helices 13−15). These helices in P3A appear to be well ordered and located near the five-fold axis. However, the corresponding helices in P3B are less resolved. The clustering of these P3A variable helical regions about the five-fold axis results in the formation of a pore along this axis (Fig. 1e). Thus, these helices are likely involved in the nascent mRNA transcription and extrusion during viral replication, a process common among dsRNA viruses occurring near the five-fold axis2,
20,
21. The conformational flexibility of these helices could conceivably allow for the expansion and/or contraction of the pore necessary during and after mRNA release.
Like the apical domain, the carapace domain helices are well conserved between P3A and P3B. However, there is an obvious difference in a region of the carapace domain involved in the intersubunit contact. In P3B, a 30 Å long helix (helix 1), also seen in an equivalent position in the animal reovirus11, protrudes from the back of the carapace domain. In P3B, this helix inserts into a pocket formed by helices 22−25 of P3A (Figs 1e, 3c,d) but is not visible in the corresponding location in P3A, similar to both the corresponding inner capsid proteins of BTV and the animal reovirus core.
The dimerization domains have the least amount of similarity between P3A and P3B, and our model suggests that both are composed mainly of loops and -sheets. The packing of the inner capsid subunits in conjunction with the potentially flexible nature of this region might be responsible for the conformational differences.
Symmetry mismatch As shown in other members of Reoviridae3, the primary function of the inner shell is to protect the RNA and provide anchoring sites for enzymes to facilitate the process of endogenous transcription, whereas the outer shell mediates host specificity and plays important roles in capsid assembly and stability. In our structure, neither the P3 dimers nor the P8 trimers have extensive contacts within their respective shells (Fig. 1d,e). However, the interactions between the inner and outer capsid proteins are substantial, because each outer capsid trimer contacts multiple inner capsid proteins. The contact sites between the two shells are mainly localized to helices 2 and 3 of P8 but are more widespread in P3 (Fig. 4). Specifically, the P trimer of P8 makes exclusive contacts with the apical domains of P3A and P3B (red, Fig. 4c), whereas the T trimer makes contacts with the dimerization domains of three neighboring P3B molecules about the three-fold axis (light blue, Fig. 4c). The remaining three trimers maintain multiple contact sites with several neighboring P3A and P3B molecules (Fig. 4b,c). These identified contact sites are conserved across all asymmetric units. The nonequivalent and specific interactions allow the T = 1 inner and T = 13l outer shells to maintain a unique and stable geometrical relationship throughout RNA replication and release.
Figure 4. Molecular interactions between the shells.
a, A zoom-in view of the inner shell and the five unique trimers (color codes specified in Fig. 1) illustrating the interaction surfaces between the inner and outer capsid. P3A is colored in green, and P3B is colored in purple. b, Same as in (a) but the trimers are replaced with contour lines to reveal the contact surfaces of the trimers on the P3 molecules. c, Sites of interaction of P8 on the P3 A−B dimer (gray). Regions of continuous density at relatively high threshold (>2 ) between proteins are colored. Exclusive interactions between P3 and the T and P trimer are colored in red and light blue, respectively, whereas multiple but specific interactions between P3 and the other trimers are colored in dark blue.
Conclusion Our analysis suggests the overall folds of the RDV capsid shell proteins appear similar among different members in the same virus family3, even though they have no significant sequence similarity. This is indicative of the divergent evolution of these viruses from a common ancestor in order to adapt to various hosts and environments. By maintaining a similar capsid architecture and protein structure, the specific interactions necessary for capsid stability can be maintained, whereas the structural differences among the outer shell proteins may play key roles in conferring host specificity during infection.
The structure determination of RDV provides the first demonstration of deriving a high-resolution model for the molecular components in a large assembly without a crystal, through the integration of electron cryomicroscopy with bioinformatic tools. We showed that at 6.8 Å resolution, although only helices and -sheets of proteins are discernable, establishing a high-resolution fold model is feasible. The construction of these models epitomizes the integration of information, spanning de novo structure assignment (P8 lower domain), domain localization through fold recognition (P8 upper domain) and model-aided fold assignment (P3). Such an integrated approach is generally applicable to other large biological machines.
Methods Electron cryomicroscopy. A JEOL4000 electron cryomicroscope with a Gatan cryoholder was used to record focal pairs of purified RDV at 400 keV as described8. Individual micrographs were digitized at 1.35 Å pixel-1 on a SCAI scanner (Z/I Imaging) and subsequently averaged to 2.7 Å pixel-1. From 81 focal pairs, >4,000 pairs of RDV particle images were selected for further analysis. Particle images from the far-from-focus micrographs were used to determine the initial orientation parameters for the corresponding particles in the close-to-focus micrograph8,
22,
23. The orientation and center parameters of each close-to-focus particle were iteratively refined as described23,
24. A contrast transfer function correction with an amplitude contrast of 7% and a Fourier amplitude damping factor of 100 Å2 was applied to each particle image prior to 3D merging in Fourier space25. Only the close-to-focus particle images, from 0.3 to 2.2 m underfocus, were included in the final 3D reconstruction, which was computed by merging 3,261 particles25,
26. The effective resolution was determined with the 0.5 threshold criterion for the Fourier shell correlation coefficient between two independent reconstructions27,
28.
Structure analysis. From the density map, slightly larger than one asymmetric unit was graphically extracted using custom-designed modules built in IRIS Explorer (NAG). Definitions between the capsid proteins, as well as their contacts, were identified visually, allowing the geometrically distinct subunits to be further segmented from the capsid. Each of the P8 trimers was aligned to the T trimer using FOLDHUNTER9. The local three-fold symmetry axis within the trimers was iteratively refined and the degree of three-fold symmetry was assessed about this axis29. The aligned trimers, excluding the T trimer, were then summed to produce an averaged trimer. A P8 monomer was subsequently extracted from the averaged P8 trimer. HELIXHUNTER9 was then run on all capsid subunits and the P3 monomers to identify potential helices. Furthermore, the -sheet domain of BTV VP7 was blurred to 7 Å resolution using PDB2MRC30 and localized to RDV P8 using FOLDHUNTER9. Secondary structure prediction using SSPRO14, PsiPred12 and PHD13 was done for both P3 (SWISS-PROT ID VP8_RDVF) and P8 (SWISS-PROT ID VP3_RDVF). The results obtained from HELIXHUNTER9 and secondary structure predictions were then correlated in order to map the sequence to the observed helices. Assignment of the helices was based on length, location, density continuity and spatial arrangement. In P3, the structures of the corresponding BTV and reovirus proteins were used as references during the model building process. Approximate connections between helices and the -sheet regions were assigned based primarily on visually identified connectivity of the secondary structure elements in our density map and similarities to the BTV and reovirus structures.
Acknowledgments This research has been supported by grants from National Institutes of Health, the Robert Welch Foundation and the National Natural Science Foundation of China. Z.H.Z. is a Pew Scholar in the Biomedical Sciences and a Basil O'Connor Starter Scholar of the March of Dimes Foundation. M.L.B. was supported in part by Baylor Research Advocates for Student. We would like to thank B.V.V. Prasad, M.F. Schmid and M. Baker for their helpful comments on the manuscript. Supplementary animations and 3D VRML models are available at http://ncmi.bcm.tmc.edu/~wjiang/rdv/.
magnification in a JEOL 4000 electron cryomicroscope with an electron dose of 10−13 electrons Å-2. Individual particles were 'boxed' out into an array of 300 
100 particle images from the micrograph shown in (a) with an underfocus of 0.6 
m. c, Resolution assessment. The effective resolution consistently improved with increasing numbers of particles in the data set and reached 6.8 Å for the data set with 3,261 particles. d, Shaded surface representation of the 6.8 Å resolution density map of the outer capsid shell (P8) of RDV. The mass densities corresponding to inner shell proteins and RNA were computationally removed. The outer shell is composed of 132 trimers of P8 organized on a T = 13l icosahedral lattice. Each asymmetric unit contains four 1/3 unique trimers
(standard deviation) above the average density.
-sheets are evident in the upper domain of the P8 monomer. c, P8 subunit and helices identified by HELIXHUNTER
