Diversity and infectivity of the RNA virome among different cryptic species of an agriculturally important insect vector: whitefly Bemisia tabaci

Abstract

A large number of insect-specific viruses (ISVs) have recently been discovered, mostly from hematophagous insect vectors because of their medical importance, but little attention has been paid to important plant virus vectors such as the whitefly Bemisia tabaci, which exists as a complex of cryptic species. Public SRA datasets of B. tabaci and newly generated transcriptomes of three Chinese populations are here comprehensively investigated to characterize the whitefly viromes of different cryptic species. Twenty novel ISVs were confidently identified, mostly associated with a particular cryptic species while different cryptic species harbored one or more core ISVs. Microinjection experiments showed that some ISVs might cross-infect between the two invasive whitefly cryptic species, Middle East Asia Minor 1 (MEAM1) and Mediterranean (MED), but others appeared to have a more restricted host range, reflecting the possibility of distinct long-term coevolution of these ISVs and whitefly hosts. Moreover, analysis of the profiles of virus-derived small-interfering RNAs indicated that some of the ISVs can successfully replicate in whitefly and the antiviral RNAi pathway of B. tabaci is actively involved in response to ISV infections. Our study provides a comprehensive analysis of the RNA virome, the distinct relationships and cross-cryptic species infectivity of ISVs in an agriculturally important insect vector.

Fig. 1. Genomic structures of novel insect-specific viruses identified in whitefly B. tabaci.

The viruses were taxonomically classified into nine groups as shown in panels a–i. Each panel contains a genome representing a phylogenetically close reference virus on the top (with red font) and the insect-specific viruses discovered from whiteflies in this study. GenBank accession numbers are provided in parentheses after the name of the reference viruses. The name of the whitefly dataset corresponding to the identified novel viruses is also indicated in parentheses below the virus name (details in Table 2). Conserved functional domains are color-coded and the names of the domains are indicated at the bottom of the figure. Abbreviation of the conserved domain names: CP coat protein, FtsJ RNA ribosomal methyltransferase, MP membrane protein, PA polymerase, PArp proline–alanine-rich protein, RdRp RNA-dependent RNA polymerase.

Fig. 2. Phylogeny of novel insect-specific viruses (ISVs) identified in whitefly B. tabaci with other related viruses.

Trees for Arlivirus (a), Nidovirales (b), Flaviviridae (c), Martellivirales (d), Picornavirales (e), Totiviridae (f), and Orthomyxoviridae (g) are based on the maximum likelihood method and inferred from conserved viral RdRp domains. Novel ISVs identified in this study are shown in red font. Nodes with bootstrap values >50% are marked with solid blue circles, and the larger circles indicate higher bootstrap values. In panels c–f, a taxonomic overview of viruses at order or family level are shown on the left, and a close-up view of the viruses of interest in this study are shown in the dotted frames on the right. The viral sequences used in this study were extracted from GenBank: the accession numbers and other related details are listed in Supplementary Table 2.

Fig. 3. Correlation between whitefly cryptic species and insect-specific viruses.

a Phylogeny of whitefly cryptic species based on mtCOI sequences using the maximum likelihood method. The mtCOI of Bemisia afer (MK360160.1) was used to root the tree. Two datasets (CAAS-BQ BJ,CN and QAU-BQ QD,CN) containing mixed cryptic species (MEAM1 and MED) are highlighted by dotted frames colored with black and purple, respectively. Nodes with bootstrap values >50% are marked with solid blue circles. b Composition of insect-specific viruses (ISVs) in each whitefly cryptic species dataset. ISVs are color-coded and the names of viruses are indicated at the bottom of the figure. c Relative abundance of ISVs across the different whitefly cryptic species datasets. The transcripts per million (TPM) of each ISV are displayed by the heat map. Abbreviations of the cities and countries: JRS, IL: Jerusalem, Israel; NY: New York, USA; AJ, IN: Ajitgarh, India; BJ, CN: Beijing, China; NB, CN: Ningbo, China; HZ, CN: Hangzhou, China; QD, CN: Qingdao, China; AH, CN: Anhui, China; HER, GR: Heraklion, Greece; PE: Penryn. Abbreviation and details of the whitefly datasets and newly discovered ISVs are listed in Tables 1 and 2, respectively. Abbreviations of the whitefly cryptic species: MEAM1 Middle East Asia Minor 1, MED Mediterranean, NW1 New World 1, SSA1 sub-Saharan Africa 1.

Fig. 4. Evaluation of the ability of insect-specific viruses to cross-infect the cryptic whitefly species Middle East Asia Minor 1 (MEAM1) and Mediterranean (MED).

a Diagram illustrating the experimental workflow. In the upper panel whitefly individuals of NBU-Q (representing cryptic species MED) were microinjected with homogenate of NBU-B (representing cryptic species MEAM1). The reciprocal injection of NBU-B with homogenate of NBU-Q whiteflies is shown in the lower panel. b Detection of insect-specific viruses (ISVs) 0, 3, 6, and 12 days post injection (DPI), and in the next generation (F1). The presence of ISVs in each whitefly sample was determined by RT-PCR. Untreated NBU-B and NBU-Q whiteflies were used as controls. A pool of 20–30 whiteflies were collected for MEAM1 and MED whiteflies at each time point, and three independent biological replicates were performed. Abbreviation of virus names: BtArV2, Bemisia tabaci arlivirus 2, BtPiLV1 Bemisia tabaci picorna-like virus 1, BtQuV1 Bemisia tabaci quaranjavirus 1, BtViLV1 Bemisia tabaci virga-like virus 1, BtViLV2 Bemisia tabaci virga-like virus 2.

Fig. 5. Profiles of virus-derived small interfering RNAs (vsiRNAs).

vsiRNAs derived from whitefly datasets NBU-B (a), NBU-Q (b), and FY-Q (c, d). The upper panel shows the size distribution of vsiRNAs, while the lower panel shows the distribution of vsiRNAs along the corresponding viral genome. Color coding shows vsiRNAs derived from the sense (black, plus) and antisense (red, minus) genomic strands. All reads in this analysis are redundant. The abbreviation of the virus names is listed in Table 2.