Investigation of the effects of P1 on HC-pro-mediated gene silencing suppression through genetics and omics approaches

Abstract

Background: Posttranscriptional gene silencing (PTGS) is one of the most important mechanisms for plants during viral infection. However, viruses have also developed viral suppressors to negatively control PTGS by inhibiting microRNA (miRNA) and short-interfering RNA (siRNA) regulation in plants. The first identified viral suppressor, P1/HC-Pro, is a fusion protein that was translated from potyviral RNA. Upon infecting plants, the P1 protein itself is released from HC-Pro by the self-cleaving activity of P1. P1 has an unknown function in enhancing HC-Pro-mediated PTGS suppression. We performed proteomics to identify P1-interacting proteins. We also performed transcriptomics that were generated from Col-0 and various P1/HC-Pro-related transgenic plants to identify novel genes. The results showed several novel genes were identified through the comparative network analysis that might be involved in P1/HC-Pro-mediated PTGS suppression.

Results: First, we demonstrated that P1 enhances HC-Pro function and that the mechanism might work through P1 binding to VERNALIZATION INDEPENDENCE 3/SUPERKILLER 8 (VIP3/SKI8), a subunit of the exosome, to interfere with the 5'-fragment of the PTGS-cleaved RNA degradation product. Second, the AGO1 was specifically posttranslationally degraded in transgenic Arabidopsis expressing P1/HC-Pro of turnip mosaic virus (TuMV) (P1/HC^Tu plant). Third, the comparative network highlighted potentially critical genes in PTGS, including miRNA targets, calcium signaling, hormone (JA, ET, and ABA) signaling, and defense response.

Conclusion: Through these genetic and omics approaches, we revealed an overall perspective to identify many critical genes involved in PTGS. These new findings significantly impact in our understanding of P1/HC-Pro-mediated PTGS suppression.

Keywords: Comparative network; MicroRNA; Omics; P1/HC-Pro; Posttranscriptional gene silencing; Viral suppressor.

Fig. 1

P1 enhances the HC^Tu-mediated phenotype and HC^Tu suppression in miRNA-mediated regulation. a Schematic binary plasmids containing the various constructs that were used in this study. b Phenotypes of the different transgenic plants. The photographs were taken of 3-week-old seedlings. Bar, 1 cm. c Detection of P1 and HC-Pro of TuMV in various transgenic plants by western blotting. The asterisk (tubulin) is an internal control

Fig. 2

Variation in potyvirus P1/HC-Pro and recombined P1/HC-Pro plants. a Schematic diagram of P1 and HC-Pro amino acid sequence similarity among fifty-seven Potyviruses. The black boxes indicate conserved sequence regions. b The morphologic phenotype of P1/HC^Zy and P1/HC^Te plants. Bar, 1 cm. c. Detection of P1 and HC-Pro in P1/HC^Zy, P1/HC^Te, P1^Zy/HC^Tu, and P1^Te/HC^Tu plants. The @ symbol indicates CDC2 as an internal control. Two-asterisk (**) indicate cross-reaction of the α-HC^Zy antibodies. d Schematic of the binary plasmids containing the various recombined P1/HC-Pro constructs used in this study. e Phenotypes of various recombinant P1/HC-Pro transgenic plants. The photographs were taken of 3-week-old seedlings. Bar, 1 cm. f Heatmap of abnormal miRNA and miRNA* accumulation in various recombinant P1/HC-Pro plants. The values of heatmap were convert from miRNA northern blot. Significant upregulation (Student’s t test; P value < 0.05) is labeled in red. Gray indicates differential expression that is not significant. The values on the right indicate the results of the log₂ (each sample/Col-0) formula

Fig. 3

Abnormal accumulation of miRNA/miRNA*s and target mRNAs. a Heatmaps of miRNA and miRNA* and b miRNA target gene expression in various transgenic plants. The values of heatmap for miRNA were convert from miRNA northern blot. The values of heatmap for target RNAs were convert from transcriptome profiles. Significant upregulation (Student’s t test; P value < 0.05) is labeled in red. Gray indicates differential expression that is not significant. The values on the right indicate the results of the log2 (each sample/Col-0) formula

Fig. 4

The protein and transcript levels of critical genes in various transgenic plants. (a–r) Genes that had a significant difference in protein levels (panel ii) between Col-0 and P1/HC^Tu plants were identified and their transcript levels were observed (panel i) in Col-0, P1^Tu, HC^Tu, and P1/HC^Tu plants. The fragments per kilobase of transcript per million (FPKM) were used to represent the normalized transcript expression. The bars represent standard deviations (n = 3). Normalized abundances were used to represent the protein amounts

Fig. 5

AGO1 protein detection in different viral P1/HC-Pro transgenic Arabidopsis plants. One asterisk (*) indicates the AGO1 isoform. Two-asterisk (**) indicates cross-reaction of the α-HC^Zy or α-HC^Te antibodies. Three-asterisk (***) indicates common bands. The @ symbol indicates RUBISCO as an internal control

Fig. 6

The gene-to-gene network of Col-0 vs. P1/HC^Tu plants. The gene profiles of twofold DEGs between Col-0 and P1/HC^Tu plants were used to generate the Pearson correlation network. The different circle sizes indicate the numbers of correlated genes. A positive correlation (> 0.95) between the two genes is indicated by a red line, whereas a green line indicates a negative correlation (< −0.9). The red circles indicate the genes involved in calcium signaling and are grouped with a red background. The blue circles indicate the genes involved in the defense response and are grouped with a blue background. The green circles indicate the genes involved in the PTGS pathway and are grouped with a green background. The yellow circles indicate the genes that are the miRNA targets and are grouped with a yellow background. The gray circles indicate the genes involved in the JA, ABA, and ethylene biosynthesis pathways and are grouped with a gray background

Fig. 7

The comparative networks for Col-0 vs. P1^Tu and Col-0 vs. HC^Tu plants. a The gene profiles of twofold DEGs between Col-0 and HC^Tu plants were used to generate the Pearson correlation networks. b The gene profiles of twofold DEGs between Col-0 and P1^Tu plants were used to generate the network. The different circle sizes indicate the number of correlated genes. A positive correlation (> 0.95) between the two genes is indicated by a red line, whereas a green line indicates a negative correlation (< −0.9). The unpresented genes and correlation lines are indicated in gray

Fig. 8

Transcript expression comparisons of critical genes in the networks. (a-ai) Genes that showed a significant connection, function, or position in the network were selected to demonstrate their transcript expression. The fragments per kilobase of transcript per million (FPKM) were used to represent the normalized transcript expression. The bars represent standard deviations (n = 3)

Fig. 9

Time courses of endogenous ethylene detection in Col-0 and P1/HC^Tu plants. The bars represent standard deviations (n = 3)