Genetic Diversity among Rose Rosette Virus Isolates: A Roadmap towards Studies of Gene Function and Pathogenicity

Abstract

The phylogenetic relationships of ninety-five rose rosette virus (RRV) isolates with full-length genomic sequences were analyzed. These isolates were recovered mostly from commercial roses that are vegetatively propagated rather than grown from seed. First, the genome segments were concatenated, and the maximum likelihood (ML) tree shows that the branches arrange independent of their geographic origination. There were six major groups of isolates, with 54 isolates in group 6 and distributed in two subgroups. An analysis of nucleotide diversity across the concatenated isolates showed lower genetic differences among RNAs encoding the core proteins required for encapsidation than the latter genome segments. Recombination breakpoints were identified near the junctions of several genome segments, suggesting that the genetic exchange of segments contributes to differences among isolates. The ML analysis of individual RNA segments revealed different relationship patterns among isolates, which supports the notion of genome reassortment. We tracked the branch positions of two newly sequenced isolates to highlight how genome segments relate to segments of other isolates. RNA6 has an interesting pattern of single-nucleotide mutations that appear to influence amino acid changes in the protein products derived from ORF6a and ORF6b. The P6a proteins were typically 61 residues, although three isolates encoded P6a proteins truncated to 29 residues, and four proteins extended 76-94 residues. Homologous P5 and P7 proteins appear to be evolving independently. These results suggest greater diversity among RRV isolates than previously recognized.

Keywords: emaravirus; negative-strand RNA virus; plant bunyavirus; rose; rose rosette virus.

Figure 1

RRV isolates and nucleotide diversity. (A) RRV isolates were obtained from 16 states, and one isolate is from Washington DC. The colored circles serve as a color legend for panel C to easily compare the geographical location in the map and color-coded isolates in the phylogenic tree. The number in each colored circle indicates the number of isolates from each state location. Notably, the Delaware, Washington DC, and Maryland isolates are featured on the right side of the map. (B) Maximum likelihood (ML) analysis of concatenated genome segments with four emaravirus species as an outgroup. The branch length scale bar is presented at the bottom and branches with values >0.5 are labelled. Tip labels are colored according to the circles in panel (A). Groups were assigned to assist the reader. (C) Top illustration of the concatenated RRV genome segments. Open boxes represent the open reading frame (ORF), and the amino acid lengths of the encoded proteins are presented. The viral RNA-dependent RNA polymerase (RdRp), glycoprotein (GP), nucleocapsid protein (NP), and movement protein (MP) are identified above the box. Question marks identify genes of unknown functions. The scale below the boxes provides a reference for the RNA segment lengths that are concatenated and identified by the thick gray bar. The nucleotide lengths of each RNA are provided. Dotted lines represent the termini separating each segment. The graph features the nucleotide diversity along the length of the concatenated sequence.

Figure 2

Breakpoint distribution plot presented below the illustrated concatenated genome. The dotted lines across the top of the plot identify the 95% and 99% confidence thresholds. Diagrammatic representation of the concatenated sequences, p values, nucleotide lengths, and termini of each segment as in Figure 1, but used here as a reference to understand the breakpoint locations in the outputs.

Figure 3

Phylogenetic relationships of RRV RNAs 1, 2, and 3 segments (left to right). The phylogenetic relationships were inferred using the maximum likelihood method (see Section 2). Bootstrap values >50% are shown (10,000 replicates) next to the branches. Branches leading to significant clusters of isolates were assigned a group number to assist in the explanation of the results. The colored diamonds at the branch tips identify the locations as in Figure 1. The trees are drawn to scale. The yellow box surrounds the clusters of isolates that include the two new isolates, SW_DE and RF_DC.

Figure 4

Phylogenetic relationships of RRV RNAs 4, 5, and 6 segments. The phylogenetic relationships were inferred using the maximum likelihood method (see Section 2). Bootstrap values >50 are shown (10,000 replicates) next to the branches. Branches leading to significant clusters of isolates were assigned a group number for reader orientation. The trees are drawn to scale. The yellow boxes surround the clusters of isolates that include the two new isolates, SW_DE and RF_DC. Trees were rooted as before except for RNA6, where there are only two species that can serve as outgroups. The branch patterns of these trees reflect different evolutionary pressures acting on these genome segments.

Figure 5

Analysis of RNA6 indel mutations. (A) Illustrative outcomes featuring highly variable sequences of RNA6 among non-recombinant genomes. The consensus bar identifies conserved sequences in green and the goldenrod color indicates the locations of substitution or indel mutations in the sequence. The legend represents the colors ascribed to different nucleotide substitutions in the alignment. These changes are also shown in the 1, 2, 3, and 4 boxes above the consensus bar. (B) The indel changes in regions 1 through 4 are elaborated. Each box contains the identity of the (non)recombinant genomes associated with each mutation. Some sequences have more than one change in the same region. Sequences with mutations in more than one region are highlighted in orange.

Figure 6

Analysis of twenty-five P6a proteins. Alignment demonstrates that most P6a proteins are 61 amino acids in length and identifies truncated and extended sequences. Amino acid letters are colored to identify similar residues. The * identifies the translation stop codons.

Figure 7

ML trees of sequences encoded by RNA1 through RNA4 with parsimony-informative amino acid changes. Open bars at the top of each panel provide a legend for each tree, the name of the protein, and its amino acid length. (A) RdRp, (B) GPP, (C) NP, and (D) MP. The approximate locations of amino acid changes are represented along the bar as colored lines or colored circles. The consensus amino acid and its position are listed followed by the substituted amino acid. The numbers in parentheses identify the number of isolates with this amino acid change from the consensus. The colored branches in the phylogenic trees correspond to the colored lines in the legends. Colored dots along the ML trees also identify amino acid changes represented by the same dots in the legends. Black branches without associating dots represent the isolates that conform to the consensus sequences.

Figure 8

ML trees of sequences encoded by RNA5 and RNA6 with parsimony-informative amino acid changes. Open bars at the top of each panel provide a legend for each tree, the name of the protein, and the amino acid length of the protein. These open bars are divided with grey boxes to represent subdivision along the sequence of 100 amino acids. (A) P5, (B) P6b, (C) P6a. The amino acid changes are listed in order below each subdivision to feature their locations along the sequence. The estimated locations of amino acid changes are represented along the bar as colored lines or colored circles. The numbers in parentheses identify the number of isolates with this amino acid change from the consensus. The colored branches in the phylogenic trees correspond to the colored lines in the legends. Some isolates are represented by two color branches, indicating that two mutations define that cluster. Colored dots along the ML trees also identify amino acid changes represented by the same dots in the legends. Black branches without associating dots represent the isolates that conform to the consensus sequences.

Figure 9

ML trees of P7 sequences with parsimony-informative amino acid changes. Open bars at the top provide a legend, as shown in the prior Figure 7 and Figure 8, the protein’s name, and amino acid length. Some isolates are represented by two color branches, indicating that two mutations define that cluster.