Identification of an attenuated barley stripe mosaic virus for the virus-induced gene silencing of pathogenesis-related wheat genes

Abstract

Background: Virus-induced gene silencing (VIGS) has become an emerging technology for the rapid, efficient functional genomic screening of monocot and dicot species. The barley stripe mosaic virus (BSMV) has been described as an effective VIGS vehicle for the evaluation of genes involved in wheat and barley phytopathogenesis; however, these studies have been obscured by BSMV-induced phenotypes and defense responses. The utility of BSMV VIGS may be improved using a BSMV genetic background which is more tolerable to the host plant especially upon secondary infection of highly aggressive, necrotrophic pathogens such as Fusarium graminearum.

Results: BSMV-induced VIGS in Triticum aestivum (bread wheat) cv. 'Fielder' was assessed for the study of wheat genes putatively related to Fusarium Head Blight (FHB), the necrotrophism of wheat and other cereals by F. graminearum. Due to the lack of 'Fielder' spike viability and increased accumulation of Fusarium-derived deoxynivalenol contamination upon co-infection of BSMV and FHB, an attenuated BSMV construct was generated by the addition of a glycine-rich, C-terminal peptide to the BSMV γ b protein. This attenuated BSMV effectively silenced target wheat genes while limiting disease severity, deoxynivalenol contamination, and yield loss upon Fusarium co-infection compared to the original BSMV construct. The attenuated BSMV-infected tissue exhibited reduced abscisic, jasmonic, and salicylic acid defense phytohormone accumulation upon secondary Fusarium infection. Finally, the attenuated BSMV was used to investigate the role of the salicylic acid-responsive pathogenesis-related 1 in response to FHB.

Conclusions: The use of an attenuated BSMV may be advantageous in characterizing wheat genes involved in phytopathogenesis, including Fusarium necrotrophism, where minimal viral background effects on defense are required. Additionally, the attenuated BSMV elicits reduced defense hormone accumulation, suggesting that this genotype may have applications for the investigation of phytohormone-related signaling, developmental responses, and pathogen defense.

Keywords: Abscisic acid; Deoxynivalenol; Jasmonic acid; Necrotrophic fungi; Pathogenesis-related 1; Salicylic acid; Vomitoxin; Yield.

Fig. 1

BSMV and Fusarium co-infection result in nonviable ‘Fielder’ spikes, increased DON contamination, and reduced yield. a WT, Fusarium-infected, BSMV:GFP-infected, BSMV:GFP and Fusarium co-infected, BSMV:00-infected, and BSMV:00 and Fusarium co-infected ‘Fielder’ spikes (from left to right) at 10 days post Fusarium inoculation. b Necrotic ‘Fielder’ spikes are averages of three replicate experiments, each composed of a minimum of 25 spikes, with standard error. c From one representative experiment, DON contamination was determined in five biological replicate spikes with standard deviation at 5 day intervals over the course of the infection. d Grains from non-BSMV, BSMV:GFP, and BSMV:00 infected tissue (left to right) without (top) and with (bottom) secondary Fusarium infection. e Yield represents the average of three biological replicate grain samples, sufficient for thousand grain weight calculation, from a representative experiment with standard deviation. ‘Uninfected’ on the x-axis refers to BSMV. ‘Uninfected’ in the inserted legend refers to Fusarium. f From one representative experiment, DON contamination was evaluated in three biological replicate grain samples with standard deviation. ‘Uninfected’ on the x-axis refers to BSMV. Differences in spike necrosis and DON contamination were analyzed with one-way ANOVA with Dunnett post hoc comparisons; while yield differences were evaluated with two-way ANOVA with Sidak post hoc comparisons. Changes between upon BSMV-infection are denoted with ‘#’ while Fusarium infected are denoted with ‘*’ (*p ≤ 0.05; ***p ≤ 0.001; ****p ≤ 0.0001)

Fig. 2

Sequence alignment of BSMV γ b constructs. a Gene and b protein sequence alignments with red nucleic and amino acids demonstrating the 3′ and C-terminal differences with underlined stop codons

Fig. 3

attBSMV:00 exhibits improved Fusarium co-infection tolerance, reduced DON contamination, and increased yield compared to BSMV:GFP-infected ‘Fielder’ tissue. a WT, Fusarium-infected, BSMV:GFP-infected, BSMV:GFP and Fusarium co-infected, attBSMV:00-infected, and attBSMV:00 and Fusarium co-infected ‘Fielder’ spikes (left to right) at 10 days post Fusarium inoculation. b Necrotic ‘Fielder’ spikes are averages of three replicate experiments, each composed of a minimum of 25 spikes, with standard error. c From one representative experiment, DON contamination was determined in five biological replicate spikes with standard deviation at 5 day intervals over the course of the infection. d Grains from non-BSMV, BSMV:GFP, and attBSMV:00 infected tissue (left to right) without (top) and with (bottom) secondary Fusarium infection. e Yield represents the average of three biological replicate grain samples, sufficient for thousand grain weight calculation, from a representative experiment with standard deviation. ‘Uninfected’ on the x-axis refers to BSMV. ‘Uninfected’ in the inserted legend refers to Fusarium. f From one representative experiment, DON contamination was evaluated in three biological replicate grain samples with standard deviation. ‘Uninfected’ on the x-axis refers to BSMV. Differences in spike necrosis and DON contamination were analyzed with one-way ANOVA with Dunnett post hoc comparisons; while yield differences were evaluated with two-way ANOVA with Sidak post hoc comparisons. Changes between upon BSMV-infection are denoted with ‘#’ while Fusarium infected are denoted with ‘*’ (*p ≤ 0.05; ***p ≤ 0.001; ****p ≤ 0.0001)

Fig. 4

attBSMV:00 is capable of VIGS in ‘Fielder’ leaves and spikes. Uninfected, BSMV:GFP-infected, BSMV:PDS-infected, attBSMV:00-infected, and attBSMV:PDS-infected (left to right) ‘Fielder’ a leaves and b spikes at 15 days post BSVM inoculation. c Relative PDS quantitation by semiquantitative RT-PCR represents the average of three biological replicate ‘Fielder’ leaves or spikes at 15 days post BSMV inoculation, from a representative experiment with standard deviation. Differences were evaluated with two-way ANOVA with Sidak post hoc comparisons (*p ≤ 0.05)

Fig. 5

attBSMV:00 may be used to characterize the role of wheat genes in FHB pathogenesis. a Relative PR1 quantitation values represent the average of three biological replicates ‘Fielder’ spikes at 7 days post BSMV inoculation with standard deviation. b FHB-infected spikelets were calculated as the average of three independent experiments, each composed of a minimum of 25 spikes, with standard error. c DON was detection in ten biological replicates from a representative experiment with deviation. Differences between the attBSMV:00 and attBSMV:PDS are evaluated with T-tests (*p ≤ 0.05)