Co-Transcriptional RNA Processing in Plants: Exploring from the Perspective of Polyadenylation

Abstract

Most protein-coding genes in eukaryotes possess at least two poly(A) sites, and alternative polyadenylation is considered a contributing factor to transcriptomic and proteomic diversity. Following transcription, a nascent RNA usually undergoes capping, splicing, cleavage, and polyadenylation, resulting in a mature messenger RNA (mRNA); however, increasing evidence suggests that transcription and RNA processing are coupled. Plants, which must produce rapid responses to environmental changes because of their limited mobility, exhibit such coupling. In this review, we summarize recent advances in our understanding of the coupling of transcription with RNA processing in plants, and we describe the possible spatial environment and important proteins involved. Moreover, we describe how liquid-liquid phase separation, mediated by the C-terminal domain of RNA polymerase II and RNA processing factors with intrinsically disordered regions, enables efficient co-transcriptional mRNA processing in plants.

Keywords: RNA processing; coupling regulation; gene expression; liquid–liquid phase separation (LLPS); plant; polyadenylation; transcription.

Figure 1

Schematic diagram of the coupling of transcription and RNA processing in plants. (A) Coupling between polyadenylation and the transcription of delay of germination 1 (DOG1) in Arabidopsis. Increasing usage of the DOG1 proximal poly(A) site promotes the expression of asDOG1, which subsequently suppresses DOG1 expression [55]. (B) Coupling between the polyadenylation and splicing of flowering locus C (FLC) transcripts in Arabidopsis. PRP8 facilitates the efficient splicing of the proximal intron in COOLAIR and promotes usage of the proximal poly(A) site in COOLAIR [31]. (C) Coupling between polyadenylation and m⁶A modification in plants. The m⁶A modification blocks binding of the polyadenylation factors CPSF30 and FY to the poly(A) signal, resulting in reduced usage of the proximal poly(A) site [56].

Figure 2

Disorder tendency predictions for proteins involved in the coupling of transcription and RNA processing in plants. All protein sequences were downloaded from TAIR (https://www.arabidopsis.org/ (accessed on 6 November 2020)). (For each figure, top panel) Protein disorder tendency curve. The disorder tendency score of each amino acid was predicted using IUPred2A [134] and subsequently fit to a smooth curve using the R ggplot2 package. The predicted scores were between 0 and 1, with a score above 0.5 (dashed line) indicating disorder. (For each figure, middle panel) Prediction of disordered regions using D²P² [135]. (For each figure, bottom panel) A domain map of the proteins in Arabidopsis based on previous studies [81,88] and the protein databases UniPro [136], Pfam [137], and SMART [138]. The disorder predictions for full-length proteins, except for AtRPB1 (C-terminal domain), are displayed. Abbreviations: WD40, tryptophan-aspartic acid motif repeats; RRM, RNA recognition motif; UHM, U2AF homology motif; NGN, NusG N-terminal domain; KOW, Kyrpides–Ouzounis–Woese motif; CTR, C-terminal repeat region.

Figure 3

Schematic diagram of the coupling machinery generated through liquid–liquid phase separation (LLPS) in plants using different processing factors. The various regulatory processes are carried out efficiently inside the phase-separated condensate, which may be driven by the carboxy-terminal domain (CTD) of RNAPII or RNA processing factors with intrinsically disordered regions (IDRs).