Trans-species synthetic gene design allows resistance pyramiding and broad-spectrum engineering of virus resistance in plants

Abstract

To infect plants, viruses rely heavily on their host's machinery. Plant genetic resistances based on host factor modifications can be found among existing natural variability and are widely used for some but not all crops. While biotechnology can supply for the lack of natural resistance alleles, new strategies need to be developed to increase resistance spectra and durability without impairing plant development. Here, we assess how the targeted allele modification of the Arabidopsis thaliana translation initiation factor eIF4E1 can lead to broad and efficient resistance to the major group of potyviruses. A synthetic Arabidopsis thaliana eIF4E1 allele was designed by introducing multiple amino acid changes associated with resistance to potyvirus in naturally occurring Pisum sativum alleles. This new allele encodes a functional protein while maintaining plant resistance to a potyvirus isolate that usually hijacks eIF4E1. Due to its biological functionality, this synthetic allele allows, at no developmental cost, the pyramiding of resistances to potyviruses that selectively use the two major translation initiation factors, eIF4E1 or its isoform eIFiso4E. Moreover, this combination extends the resistance spectrum to potyvirus isolates for which no efficient resistance has so far been found, including resistance-breaking isolates and an unrelated virus belonging to the Luteoviridae family. This study is a proof-of-concept for the efficiency of gene engineering combined with knowledge of natural variation to generate trans-species virus resistance at no developmental cost to the plant. This has implications for breeding of crops with broad-spectrum and high durability resistance using recent genome editing techniques.

Keywords: Arabidopsis thaliana; eIF4E; potyvirus; resistance; synthetic allele; translational research.

Figure 1

Design of a synthetic Arabidopsis resistance allele eIF4E1 ^R . (a) Alignment of eIF4E protein sequences from pea susceptible (JI2009) and resistant (JI1405 and PI269818) accessions with the Arabidopsis thaliana eIF4E1 (AteIF4E1). The position of the amino acids that differ between susceptible and resistance pea accessions is highlighted in grey. Black triangles above the alignment indicate the mutations introduced in Arabidopsis eIF4E1 to create the synthetic eIF4E1^R. (b and c) Three‐dimensional predicted structure of AteIF4E1 (b) and of eIF4E1^R (c) based on homology modelling using the pea eIF4E 3D structure as a template (PDB ID: 2WMC‐C). The modified amino acids between AteIF4E1 and eIF4E1^R are coloured in red with side chains shown.

Figure 2

Functional complementation of eif4e1 ^KO plants by eIF4E1 ^R allele. (a) Bolting of eif4e1 ^KO and complemented plants, 4 weeks after sowing. Columbia Wild‐type plants (Col WT) are used as control. Results are shown for three independent transgenic lines (1 to 3) expressing the eIF4E1^R construct in a eif4e1 ^KO background. Table legend shown under (b) applies also for (a). (b) Boxplot representation of the bolting time (in days after sowing) for the same genotypes as in (a). Results are averaged from 16 individual plants per genotype. (a) and (b) represent significantly different groups (P < 0.05). (c) In planta cap‐binding purification of eIF4E1 proteins. Total soluble protein extract from control and transgenic plants was purified on m7GTP‐agarose beads. After purification, the output fraction was analysed by Western blot using anti‐eIF4E1 antibody while equal loading control was checked on total protein (input) by Western blot for actin detection and by Ponceau staining for Rubisco protein detection.

Figure 3

eIF4E1 ^R is a resistance allele to ClYVV in Arabidopsis. ClYVV accumulation in eif4e1 ^KO plants transformed with the wild‐type eIF4E1 genomic construct under its own promoter (eIF4E1), the GUS gene under the control of the 35S promoter (GUS) or with the genomic eIF4E1 ^R construct under its endogenous promoter (three independent lines are shown, eIF4E1^R 1–3). Wild‐type Columbia and eif4e1 ^KO plants are used as positive and negative controls, respectively. Accumulation of ClYVV was assessed by DAS‐ELISA 30 days postinoculation. (a) and (b) represent significantly different groups, P < 0.05.

Figure 4

eIF4E1 ^R complements the eif4e1 ^KO eifiso4e ^KO lethality phenotype at no developmental cost. (a) Phenotype of two independent eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R lines, four weeks after sowing. (b) Dry weight analyses of 4‐week‐old controls and eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plants. Results were averaged from 20 plants for each genotype. (c) Fertility rate is the weight of seeds produced by plants of controls lines and eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plants. Results were averaged from 10 plants for each genotype. Kruskal–Wallis statistical tests were performed to identify statistically significant differences (a) and (b) represent significantly different groups (P < 0.05, standard error bars are represented on the graph).

Figure 5

eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plants cumulate the respective resistances to TuMV and ClYVV. ClYVV and TuMV viral accumulation measured by DAS‐ELISA in control lines and in two eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R lines. DAS‐ELISA was performed 30 days after inoculation with ClYVV (a) or 21 days after inoculation with TuMV (b). a, b and c represent significantly different groups, P < 0.05.

Figure 6

eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plants resistance spectrum extends to WMV. WMV viral accumulation was measured by DAS‐ELISA in control lines and in two eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R lines. DAS‐ELISA was performed at 21 days after inoculation with WMV. a and b represent significantly different groups, P < 0.05.

Figure 7

eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plant resistance spectrum extends to resistance‐breaking TuMV isolates. Controls and eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R lines were challenged with either GFP‐tagged TuMV, GFP‐TuMV‐E116Q (RB) or GFP‐TuMV‐N163Y (RB). Viral accumulation was assessed using GFPcam (PSI) camera‐imaging (a). Fluorescence intensity is shown in false colour from blue (low) to red (high intensity) with the plant pot reflecting light, allowing visualization of the outline of the plant position. (b) Detection of TuMV mRNA by RT‐PCR amplification of the VPg coding sequence for both TuMV‐GFP and TuMV‐N163Y‐GFP 21 dpi in the noninoculated leaves. ADENINE PHOSPHORIBOSYL TRANSFERASE 1 (APT1) gene was used as a control for RNA extraction.

Figure 8

eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R plants resistance spectrum extends to polerovirus species BWYV‐USA but not to TuYV and BMYV. BWYV‐USA, TuYV and BMYV viral accumulation was measured by DAS‐ELISA in control lines and in two eif4e1 ^KO eifiso4e ^KO eIF4E1 ^R lines. DAS‐ELISA was performed 21 days after inoculation with TuYV (a), BMYV (b) or BWYV‐USA (c). 12 plants per genotype were tested, and experiments were repeated twice. a, b and c represent significantly different groups, P < 0.05.

Figure 9

Update on the selective recruitment of translation initiation factors 4E by ssRNA+ viruses in Arabidopsis. Arrows indicate that the virus isolate relies on the 4E factor to infect the plant. WMV, Watermelon mosaic virus; RB‐TuMV, resistance‐breaking Turnip mosaic virus; ClYVV, Clover yellow vein virus; PPV, Plum pox virus; TEV, Tobacco etch virus; LMV, Lettuce mosaic virus; BWYV‐USA, Beet western yellows virus‐USA. Potyviruses are coloured in black and poleroviruses in orange.