Differential Effects of RNA-Dependent RNA Polymerase 6 (RDR6) Silencing on New and Old World Begomoviruses in Nicotiana benthamiana

Abstract

RNA-dependent RNA polymerases (RDRs) are key players in the antiviral defence mediated by RNA silencing in plants. RDR6 is one of the major components of the process, regulating the infection of certain RNA viruses. To better clarify its function against DNA viruses, we analyzed the effect of RDR6 inactivation (RDR6i) in N. benthamiana plants on two phloem-limited begomoviruses, the bipartite Abutilon mosaic virus (AbMV) and the monopartite tomato yellow leaf curl Sardinia virus (TYLCSV). We observed exacerbated symptoms and DNA accumulation for the New World virus AbMV in RDR6i plants, varying with the plant growth temperature (ranging from 16 °C to 33 °C). However, for the TYLCSV of Old World origin, RDR6 depletion only affected symptom expression at elevated temperatures and to a minor extent; it did not affect the viral titre. The accumulation of viral siRNA differed between the two begomoviruses, being increased in RDR6i plants infected by AbMV but decreased in those infected by TYLCSV compared to wild-type plants. In situ hybridization revealed a 6.5-fold increase in the number of AbMV-infected nuclei in RDR6i plants but without egress from the phloem tissues. These results support the concept that begomoviruses adopt different strategies to counteract plant defences and that TYLCSV evades the functions exerted by RDR6 in this host.

Keywords: AbMV; Nicotiana benthamiana; RDR6; RNA silencing; TYLCSV; geminivirus; plant–virus interaction; siRNAs; tissue tropism.

Figure 1

Impact of RDR6 depletion on AbMV or TYLCSV infection. (a) Comparison of symptom expression on wild-type (wt) and RDR6i Nicotiana benthamiana plants following mock inoculation or inoculation with either AbMV or TYLCSV. Photographs were taken 28 days post-inoculation (dpi). The inoculation assay was repeated eight times using more than 300 plants in total. (b) Second and third subapical leaves from mock-inoculated or AbMV/TYLCSV-infected plants, photographed at 49 dpi. Leaves are representative examples of 12–20 samples per infection type. (c) Length of leaves of mock-inoculated or AbMV/TYLCSV-infected plants, measured at 49 dpi. Measurements were conducted on 12–20 leaves per treatment for each genotype. Asterisks indicate significant differences among treatments according to the Student’s t-test (p < 0.01). (d) AbMV and TYLCSV genomic DNA accumulation in wt and RDR6i plants, evaluated by Southern blot on total DNA extracted from plants at 21 or 28 dpi, as indicated. gDNA, total genomic DNA, is shown as the loading control; oc, lin, ccc, and ss represent the open circular, linear, covalently closed circular supercoiled dsDNA, and single-stranded circular ssDNA forms, respectively. M: DNA molecular weight marker.

Figure 2

Impact of plant cultivation temperature and RDR6 depletion genotype (RDR6i) on AbMV and TYLCSV infection. (a,b) Symptom expression on wild-type (wt) and RDR6i Nicotiana benthamiana plants infected by either (a) AbMV or (b) TYLCSV, grown at constant temperatures of 16, 23, and 33 °C. Mock-inoculated plants served as controls. Photographs were taken at 49 dpi. Inoculation assays were repeated twice, using a total of 10 plants per genotype and infection type. (c,d) The height of wt and RDR6i plants mock-inoculated or infected with either (c) AbMV or (d) TYLCSV, measured at 49 dpi. The height of mock-inoculated plants (10 plants per treatment) was set to 100%. Asterisks denote statistically significant differences according to Student’s t-test (p < 0.01), comparing infected vs. mock-inoculated plants for each genotype and growth condition. (e) AbMV and TYLCSV genomic DNA accumulation in wt and RDR6i plants was evaluated using a Southern blot on total nucleic acids (TNAs) extracted from individual representative plants at 49 dpi. The symbols “+” and “−“ indicate the presence or absence of the virus, respectively. gDNA, total genomic DNA, is shown as the loading control; ccc and ss represent covalently closed circular supercoiled dsDNA and single-stranded circular ssDNA forms, respectively.

Figure 3

Localization of AbMV or TYLCSV DNA in leaves of wild-type (wt) and RDR6i plants. (a) FISH analysis of stem cross sections (100 µm thick) fixed at 49 dpi; virus detection using fluorescently labelled AbMV- or TYLCSV-specific oligonucleotide probes. Arrows point at signals representing exemplary virus-infected nuclei inside the two (internal and external) phloem domains adjacent to the auto-fluorescent xylem elements. X: xylem, P_i/e: phloem (internal/external), scale bars: 100 µm. (b) Average numbers of virus-infected nuclei in 100 µm thick stem sections of similar diameters were measured in plants infected by either AbMV or TYLCSV. Data represent 673 signals for AbMV (six or eight sections per plant genotype), and 1154 signals for TYLCSV (three or six sections, respectively). Asterisks indicate significant differences, as indicated above. (c) AbMV DNA in vascular tissues around the apical meristem (for wt plants) or flower bud (for RDR6i plants) of whole-mount specimens. M: meristem, V: vascular bundle (exemplary labels), scale bars: 100 µm.

Figure 4

Northern blot analysis of virus-derived siRNAs in wild type (wt) or RDR6i Nicotiana benthamiana plants, either mock-inoculated (control) or infected by either AbMV or TYLCSV. vsiRNAs were detected using PCR-amplified, digoxigenin-labelled, full-length virus DNA probes via chemiluminescence. Small RNA (≈70 to ≈250 nt apparent lengths) loads of the corresponding denaturing polyacrylamide gels (above panel, Sybr^® Gold stain) are shown as loading control. HS, an AbMV-derived DNA oligonucleotide (from ORF C1), was used as a hybridisation specificity standard.