Transposon-Associated Small RNAs Involved in Plant Defense in Poplar

Abstract

Utilizing high-throughput Illumina sequencing, we examined how small RNA (sRNA) profiles vary in Chinese white poplar (Populus tomentosa) across two pivotal infection stages by the rust fungus Melampsora larici-populina: the biotrophic growth phase (T02; 48 h post infection) and the urediniospore development and dispersal phase (T03; 168 h), both essential for plant colonization and prolonged biotrophic engagement. Far exceeding random expectations, siRNA clusters predominantly arose from transposon regions, with pseudogenes also contributing significantly, and infection-stage-specific variations were notably tied to these transposon-derived siRNAs. As the infection advanced, clusters of 24 nt siRNAs in transposon and intergenic regions exhibited pronounced abundance shifts. An analysis of targets indicated that Populus sRNAs potentially regulate 95% of Melampsora larici-populina genes, with pathogen effector genes showing heightened targeting by sRNAs during the biotrophic and urediniospore phases compared to controls, pointing to selective sRNA-target interactions. In contrast to conserved miRNAs across plant species, Populus-specific miRNAs displayed a markedly greater tendency to target NB-LRR genes. These observations collectively highlight the innovative roles of sRNAs in plant defense, their evolutionary roots, and their dynamic interplay with pathogen coevolution.

Keywords: Melampsora larici-populina; defense response; infection; poplar; small RNAs.

Figure 1

The stage differences in siRNA abundance primarily originate from transposons in response to infection. Shown are the 21 and 24 nt siRNA clusters with at least 5 tpm (transcripts per million mapped reads). Clusters are arranged in ascending order of fold change. (A) The common and stage-specific siRNA clusters. (B) The relative abundance of the siRNA clusters in the various types of physical locations in the Populus genome. (C–F) Degree of stage-specific differences for the siRNA clusters in the different locations. The log₁₀ values of their abundance in T01 (control group; 48 h) divided by their abundance in (C,D) T02 (biotrophic growth phase; 48 h post infection) or (E,F) T03 (urediniospore formation and release phase; 168 h post infection). Up indicates the 2 kb upstream region of genes and down indicates the 2 kb downstream region of genes. The 22 nt and 24 nt siRNA clusters with at least 5 tpm in the three samples are shown. The clusters below the horizontal line were negatively regulated, while those above the horizontal line were positively regulated. The clusters are arranged in ascending order of fold change.

Figure 2

The difference in transposon siRNA activity are driven by 24 nt sRNAs. (A) The abundance of siRNAs in the three samples (T01; T02; and T03). The total abundance column displays the relative abundance of 21, 22, and 24 nt siRNAs. (B) The number of pathogen effectors targeted by the siRNAs in the three infection samples. The effectors were predicted by EffectorP and the target prediction were performed by using psRNATarget. T01: control group; 48 h. T02: biotrophic growth phase; 48 h post infection. T03: urediniospore formation and release phase; 168 h post infection. (C) A schematic model illustrating the role of poplar-derived siRNAs in regulating pathogenicity genes of Melampsora larici-populina. Error bars, bootstrap-based 95% confidence intervals on the mean estimates. Statistical inference is conducted with a permutation test on the mean (perm.test in the R package exact RankTests).

Figure 3

The Populus NB-LRR genes are targeted by Populus-specific miRNAs. (A) The species-specific miRNAs of nine plant species. ptc: Populus trichocarpa; mtr: Medicago truncatula; bra: Brassica rapa; sbi: Sorghum bicolor; osa: Oryza sativa; vvi: Vitis vinifera; ath: Arabidopsis thaliana; bdi: Brachypodium distachyon; gma: Glycine max. (B) Comparison of the number of predicted disease-resistance target genes of the conserved and Populus-specific miRNAs. (C) Conservation profile of miR482, miR2118, and miR2111 in diverse plant species. The mature miRNA sequences were retrieved from miRBase. (D) Comparison of the number of predicted NB-LRR genes targeted by miR482, miR2111, miR6579, and miR6590. (E) The genome locations of two Populus-specific miRNAs. (F) The consensus coding sequences of the targets of four Populus miRNAs.