Fig 1.
Sequence similarity and phylogeny of the 30K MPs.
(A) Clustering of 30K MP sequences by pairwise sequence similarity (CLANS P-value ≤ 1 × 10−15). The clusters are colored and named by virus families, while the outline boxes indicate if the virus family is part of superclusters 1–4. The lines represent sequence relationships, darker colors indicate closer sequence similarity. The HSP values used for clustering can be found in S2 Table. (B) Maximum-likelihood phylogenetic tree of 30K MP sequences obtained by IQ-TREE. SC, supercluster. The circles at the nodes indicate bootstrap branch support values ≥90. Superclusters 1–5 are also indicated. The tree in newick format can be found in S1 Data. HSP, high scoring pair; MP, movement protein.
Fig 2.
Structural modeling of 30K MPs.
(A) Structural model of a representative full-length MP of the cabbage leaf curl virus (family Geminiviridae). The structure is colored using the rainbow scheme from blue (N-terminus) to red (C-terminus). The β-strands of the jelly-roll domain are indicated with Roman letters. (B) Structural models of the 30K MPs representing different virus families. The variable terminal ends were trimmed for the convenience of presentation. The structures are colored using the rainbow scheme from blue (N-terminus) to red (C-terminus). The structures are grouped according to established virus taxonomy. In the case of Orthornavirae, the corresponding phyla are indicated. Phylum Kitrinoviricota: Virgaviridae is represented by TMV, Betaflexiviridae by actinidia virus, Mayoviridae by raspberry bushy dwarf virus, Bromoviridae by cucumber mosaic virus, Kitaviridae by citrus leprosis virus C, Tombusviridae by carrot mottle virus; phylum Negarnaviricota: family Rhabdoviridae is represented by lettuce necrotic yellows virus, Phenuiviridae by rice stripe virus, Fimoviridae by rose rosette virus, Tospoviridae by tomato spotted wilt virus, Aspiviviridae is represented by citrus psorosis virus and lepidozia ophiovirus tri (LepOV_tri) associated with hairy liverwort; phylum Pisuviricota: Secoviridae is represented by cherry rasp leaf virus and tomato fern seco-like virus (TfSV); phylum Lenarviricota: family Botourmiaviridae is represented by ourmia melon virus. Family Caulimoviridae (kingdom Pararnaviae) is represented by cauliflower mosaic virus, whereas family Geminiviridae (realm Monodnaviria) is represented by cabbage leaf curl virus. The PDB structure files for the modeled MPs can be found in S2 Data. MP, movement protein; PDB, Protein Data Bank; TMV, tobacco mosaic virus; TfSV, tomato fern seco-like virus.
Fig 3.
Structural similarity between SJR CPs and 30K MPs.
(A) Structures of the SJR CPs homologous to 30K MPs obtained after a DALI search of PDB database, in the upper row highlighted with a blue background. The bottom row shows the jelly-roll region for the selected structures of 30K MP representatives, highlighted with a yellow background. The first structures on the utmost left in the upper and bottom row have the BIDG-CHEF β-strands annotated. The structures are colored using the rainbow scheme from blue (N-terminus) to red (C-terminus). (B) Dendrogram and heatmap of complete linkage clustering of 30K representatives and SJR CPs. The red circles indicated in the top dendrogram, represent bootstrap values ≥90 obtained with R package “pvclust.” The CPs and MPs are indicated in blue and yellow, respectively. Structures of 30K MPs and SJR CPs belong to: BMV, CCMV, FBNSV, STNV, ACMV, AYVV, STMV, SPMV, IPNV, IBDV, BBV, NoV, PrV, NomegaV, BFDV, PCV2, PhMV, TYMV, BYDV, BChV, PVYV, FBPV, RGMoV, RYMV, SBMV, TNV, BPMV, CPMV, FCV, NV, HRV16, SBPV, and CrPV. The newick format of the dendrogram obtained in DALI can be found in S3 Data. ACMV, African cassava mosaic virus; AYVV, ageratum yellow vein virus; BBV, black beetle virus; BChV, beet chlorosis virus; BFDV, beak and feather disease virus; BMV, brome mosaic virus; BPMV, bean pod mottle virus; BYDV, barley yellow dwarf virus; CCMV, cowpea chlorotic mottle virus; CP, capsid protein; CPMV, cowpea mosaic virus; CrPV, cricket paralysis virus; FBNSV, faba bean necrotic stunt virus; FBPV, faba bean polerovirus 1; FCV, feline calicivirus; HRV16, human rhinovirus; IBDV, infectious bursal disease virus; IPNV, infectious pancreatic necrosis virus; MP, movement protein; NomegaV, nudaurelia capensis omega virus; NoV, nodamura virus; NV, Norwalk virus; PCV2, porcine circovirus 2; PDB, Protein Data Bank; PhMV, physalis mottle virus; PrV, providence virus; PVYV, pepper vein yellows virus; RGMoV, ryegrass mottle virus; RYMV, rice yellow mottle virus; SBMV, southern bean mosaic virus; SBPV, slow bee paralysis virus; SJR, single jelly-roll; SPMV, satellite panicum mosaic virus; STMV, satellite tobacco mosaic virus; STNV, satellite tobacco necrosis virus; TNV, tobacco necrosis virus; TYMV, turnip yellow mosaic virus.
Fig 4.
Validation of the homology between SJR CPs and 30K MPs by sensitive sequence analysis.
(A) Homologous regions between the CP of PLRV (PDB ID: 6SCO) and Camellia oleifera geminivirus (CaOV) 30K MP (accession number: QIE08114) obtained with HHsearch analysis against the PDB database. Secondary structure prediction is indicated by arrows for beta strands in yellow. (B) The structural model of CaOV 30K MP and PLRV CP. The homology region between the 2 proteins found in HHsearch against the PDB database is shown in red. The superposition of the conserved jelly-roll regions of CaOV 30K MP and PLRV CP is shown in the middle. The PLRV CP is colored light purple, and the CaOV 30K MP is colored light gray. CP, capsid protein; MP, movement protein; PDB, Protein Data Bank; PLRV, potato leaf roll virus; SJR, single jelly-roll.
Fig 5.
Length variation of the terminal regions of the 30K MPs, D motif conservation and charge distribution in the 30K MPs and SJR CPs.
(A) Boxplot of the lengths of the N and C-terminal regions of 30K MPs. Orange boxes indicate values for N-terminal sizes and the green boxes indicate the C-terminal sizes. The x-axis denotes virus families and the y-axis the size of terminal ends by the number of amino acids. All the values are ordered by size from the smallest to the largest. The numeric values corresponding to the lengths of the N- and C-termini used for the boxplot can be found in S4 Table. (B) Top: the D motif region in the alignment of representative 30K MPs and SJR CPs. Note that in SPMV, the aspartate (D) is conservatively substituted with an asparagine (N). Bottom: the position of the D motif mapped on the MP and CP protein structures. The D motif is marked with a red circle. (C) Local charge distribution for CaLCuV 30K MP and CCMV SJR CP (PDB: 1ZA7) sequence by amino acid residue position (window size 21). The jelly-roll region is represented by a light green box. The height of the line above the gray threshold (0.0) indicates the value of the positive charges. The numerical values used to plot the charge distributions can be found in S5 Table. CCMV, cowpea chlorotic mottle virus; CP, capsid protein; MP, movement protein; SJR, single jelly-roll; SPMV, satellite panicum mosaic virus.
Fig 6.
An evolutionary scenario for the origin of the 30K MP superfamily from SJR CPs.
The ancestral virus is predicted to have an RNA genome (green wavy line) and encode an SJR CP, which was responsible for capsid formation and promoted intercellular movement through developing plasmodesmata. Duplication and neofunctionalization of the cp gene (yellow wavy line) led to the emergence of a dedicated mp gene (orange wavy line). Subsequently, the mp gene was horizontally transfered to other RNA viruses and viruses with DNA genomes (red wavy lines). Abbreviations: CP, capsid protein; (pre-)PD, (developing) plasmodesmata; MP, 30K movement protein; dupl., gene duplication; SJR, single jelly-roll.