Molecular Characterization of Potato Virus Y (PVY) Using High-Throughput Sequencing: Constraints on Full Genome Reconstructions Imposed by Mixed Infection Involving Recombinant PVY Strains

Abstract

In recent years, high throughput sequencing (HTS) has brought new possibilities to the study of the diversity and complexity of plant viromes. Mixed infection of a single plant with several viruses is frequently observed in such studies. We analyzed the virome of 10 tomato and sweet pepper samples from Slovakia, all showing the presence of potato virus Y (PVY) infection. Most datasets allow the determination of the nearly complete sequence of a single-variant PVY genome, belonging to one of the PVY recombinant strains (N-Wi, NTNa, or NTNb). However, in three to-mato samples (T1, T40, and T62) the presence of N-type and O-type sequences spanning the same genome region was documented, indicative of mixed infections involving different PVY strains variants, hampering the automated assembly of PVY genomes present in the sample. The N- and O-type in silico data were further confirmed by specific RT-PCR assays targeting UTR-P1 and NIa genomic parts. Although full genomes could not be de novo assembled directly in this situation, their deep coverage by relatively long paired reads allowed their manual re-assembly using very stringent mapping parameters. These results highlight the complexity of PVY infection of some host plants and the challenges that can be met when trying to precisely identify the PVY isolates involved in mixed infection.

Keywords: PVY; Solanaceae; genome; next generation sequencing; potyvirus.

Figure 1

Analysis of samples showing single variant infections by PVY. (A) A maximum likelihood (ML) phylogenetic tree showing the relationship among PVY isolates. Complete genomes of PVY isolates determined in this work (highlighted by an arrow), together with sequences of the selected representing isolate belonging to different molecular groups, were used for phylogenetic analysis. The database isolates are identified by their names, GenBank accession number, country of origin, and strain relationship. Strain affiliation is indicated based on [20]. The phylogenetic analysis was inferred using maximum likelihood (ML) based on the General Time Reversible (GTR + G) model selected as the best-fit model of nucleotide substitution based on Bayesian information criterion (BIC) as implemented in MEGA 7. The divergent PVY isolate AJ439544 was used as an outgroup. Scale bar represents genetic distance and the numbers at the nodes indicate the bootstrap values (1000 replicates) >70%. (B) Schematic representation of the PVY genome showing the nucleotide positions delimiting the respective potyviral functional products (based on the complete genome of SL16 isolate (KX713170, PVY-N-Wi strain). (C) Schematic representation of recombinant PVY genomes isolates characterized in this work, showing the position of parental genome portions. PVY-O-type (azure), PVY-N-type (dark blue). * counts also for the T24, T31, PAP, SL50V, and PAR-P2 genomes.

Figure 2

Analysis of samples showing multiple infections by PVY strains. (A) Schematic representation of the positions of informative regions (red stripes) along the PVY genome used for PVY-O- and N-specific mapping of HTS reads (B) Graphical representation of obtained sequences from samples T1, T40, and T62, where positions with more than one variant detected are shown (azure/dark blue) (C) Possible recombination patterns of PVY variants present in T1, T40, and T62 samples.

Figure 3

Phylogenetic analysis of PVY variants identified in mixed infections. The phylogenetic tree reconstructed from nearly complete genomes of T1, T40, and T62 variants and selected reference genomes using the neighbor-joining algorithm implemented in MEGA v.7.