Comparative genomics of Papaya ringspot virus pathotypes P and W from India

Abstract

Papaya ringspot virus (PRSV), a definite member of the family Potyviridae and the genus Potyvirus [5], occurs worldwide and is a major constraint to papaya (Carica papaya) and cucurbits production throughout the tropical and subtropical regions [21]. The virus is readily vectored by aphids in a non-persistent manner [16]. The virus was first recorded from western India in 1958 [4] and since then it has spread to different geographical locations, limiting papaya and cucurbits production [1, 8, 9]. Prevalence of both papaya-infecting (Type P) and non-papaya-infecting (Type W) pathotypes has been recorded. The virus induces variety of symptoms on foliage, stem and fruit in papaya and cucurbits. PRSV has flexuous filamentous particles (760–800 nm 9 12 nm) with a single-stranded, positivesense RNA genome of approximately 10,000 nucleotides. Fifteen PRSV isolates belonging to different geographical locations have been completely sequenced. PRSV isolates from India have been characterized for coat protein only [1, 8, 9], and information on their complete genome sequences is lacking. In this study, complete genome sequences of pathotypes P and W from India and their sequence divergence, phylogenetic relationships and recombination with 15 other PRSV genomes have been determined to investigate their origin and host specificity. Preliminary results on pathotype P have been published [15]. PRSV isolates from symptomatic papaya (Carica papaya) (pathotype P) and spongegourd (Luffa aegyptiaca) (pathotype W) plants were collected from the Indian Agricultural Research Institute experimental farm in New Delhi and propagated in papaya (cv. Pusa Nanha) and pumpkin (Cucurbita pepo), respectively, in a glass house through sap inoculations using 0.1 M phosphate buffer, pH 7.2, containing 0.1% 2-mercaptoethanol and celite as an abrasive. The details of the PRSV genome sequences collected from GenBank (http://www.ncbi.nlm.nih.gov) are shown in Table 1. The analysis utilized the nucleotide and translated amino acid sequences of the whole genome or of individual cistrons. The complete genome sequence of PRSV RNA was generated with nine overlapping fragments using nine pairs of specific primers. Viral RNA was reverse transcribed and amplified using high-fidelity Taq DNA polymerase (New England Biolabs). Two clones of each of the nine amplified products (700–1,900 bp) were sequenced in both (30 and 50) directions with 100% identity in the overlapping regions. The sequences of overlapping cloned fragments that were generated were assembled manually using the BioEdit software version 5.09.04 [7]. Nucleotide and amino acid sequences representing the complete genomes of pathotypes P and W and their cistrons were aligned using CLUSTAL W [19]. A sequence similarity matrix between polyproteins and specific cistrons was calculated using the BioEdit software version 5.09.04 [7]. Recombination sites were identified using RDP [11], GENECONV [14], BOOTSCAN [12], MAXIMUM CHISQUARE [13], CHIMAERA [12] and SISTER SCAN [6], the non-parametric recombination detection methods implemented in RDP2 [12]. A multiple comparison corrected P-value cutoff 0.05 and default settings were used throughout, and only events detectable by two or more different methods were retained for further analysis. S. K. Mangrauthia Plant Pathology Section, Directorate of Rice Research, Hyderabad, India