Alternative splicing of anciently exonized 5S rRNA regulates plant transcription factor TFIIIA

Abstract

Identifying conserved alternative splicing (AS) events among evolutionarily distant species can prioritize AS events for functional characterization and help uncover relevant cis- and trans-regulatory factors. A genome-wide search for conserved cassette exon AS events in higher plants revealed the exonization of 5S ribosomal RNA (5S rRNA) within the gene of its own transcription regulator, TFIIIA (transcription factor for polymerase III A). The 5S rRNA-derived exon in TFIIIA gene exists in all representative land plant species but not in green algae and nonplant species, suggesting it is specific to land plants. TFIIIA is essential for RNA polymerase III-based transcription of 5S rRNA in eukaryotes. Integrating comparative genomics and molecular biology revealed that the conserved cassette exon derived from 5S rRNA is coupled with nonsense-mediated mRNA decay. Utilizing multiple independent Arabidopsis overexpressing TFIIIA transgenic lines under osmotic and salt stress, strong accordance between phenotypic and molecular evidence reveals the biological relevance of AS of the exonized 5S rRNA in quantitative autoregulation of TFIIIA homeostasis. Most significantly, this study provides the first evidence of ancient exaptation of 5S rRNA in plants, suggesting a novel gene regulation model mediated by the AS of an anciently exonized noncoding element.

Figure 1.

Identification of a conserved cassette exon event associated with the TFIIIA gene in Arabidopsis and rice. (A) Exon-inclusion (EI) and exon-skipping (ES) isoforms of TFIIIA transcripts in Arabidopsis and rice (red boxes represent cDNA/EST alignments indicating exon locations; green boxes represent predicted gene models; orange boxes represent predicted coding regions). (B) A conserved ES event associated with AtTFIIIA (At1G72050) and OsTFIIIA (Os05G03020) and the sequence alignment (BLAST parameters: −W 7 –F F) between orthologous exon skipped/included (gray box) and the flanking constitute orthologous exons (green box, the second and third C2H2 zinc finger). The inclusion of the skipped/included exon introduces an in-frame stop codon (red triangle in the top panel and red text in the alignments of the bottom panel). OsTFIIIA Exon5 has higher similarity to AtTFIIIA cassette exon (Exon3) than OsTFIIIA Exon4 (81% vs. 74%).

Figure 2.

Identification of a 5S-rRNA-like cis-element that coincides with the conserved cassette exon AS event. For clarity, only those sequences corresponding to the ECR in seven plant species representing evolutionarily significant nodes from Bryophyta to Magnoliophyta were displayed (for complete alignments see Supplemental Fig. 2). Alignments are based on 49 5S rRNA (A) and 52 5S-rRNA-like element sequences (B). (C) Similarity between the predicted secondary structure of the 5S-rRNA-like element of AtTFIIIA (left) and 5S rRNA (right). Element names are given according to the standard 5S rRNA nomenclature: Helices are designated by roman numbers (I, II, III, IV, and V) in colored boxes; loop regions are designated by letters (A, B, C, D, and E) in open boxes. Helical regions, which belong together, are colored identically in the element line, in the alignment, and in the structure models; similar colors are used to emphasize features similar in 5S rRNA and the 5S-rRNA-like element. In lines labeled “Logo,” the size of nucleotide characters is proportional to the frequency of occurrence of this nucleotide at this position; the height of the character stack signifies the information content (in bits) of the sequences at that position (Beitz 2000). The last line of the alignments shows the predicted consensus structure in bracket-dot notation (Wilm et al. 2008). In the structure model of 5S rRNA, binding sites of TFIIIA and ribosomal protein L5 are shaded in gray. In the structure model of the 5S-rRNA-like element, sequence regions identical in 5S rRNA and the 5S-rRNA-like element are highlighted in gray.

Figure 3.

Phylogenetic relationship of plant 5S rRNAs (dark gray) and 5S-rRNA-like elements (light gray). The tree is based on the program SplitsTree4 (Huson and Bryant 2006), LogDet distance (calculated on basis of the alignment of 5S rRNAs and 5S-rRNA-like elements shown in Supplemental Fig. 2C, but after gap removal) and neighbor joining. Only the leaves with the sequences given in Figure 2 are labeled.

Figure 4.

RT-PCR validation of accumulation of EI mRNA isoform after blocking NMD with cycloheximide (CHX) treatment for 4 h.

Figure 5.

Characterization of the EI mRNA isoform of TFIIIA. (A) In vitro translation of Arabidopsis EI mRNA. Lanes (1–6) 0, 0.5, 1, 2, 4, and 8 μg EI mRNA, respectively, added in a cell-free wheat germ extract. (B) In vivo translation of EI. Western blot using anti-HA antibody (top) and SDS-PAGE analysis (bottom) of the overexpression lines using L5-cDNA (left) and EI-cDNA (right) fused with HA-tag. Electrophoretic mobility shift assay using GST-LeTFIIIA(zf1-9) (C) and GST-LeTFIIIA(zf1-2) (D). 5S DNA was titrated with increasing concentrations of GST-LeTFIIIA(zf1-9) (lanes 1–9: 0, 0.2, 0.4, 0.9, 1.8, 3.5, 7, 14, and 28 nM, respectively) or GST-LeTFIIIA(zf1-2) (lanes 1–8: 0, 14, 28, 55, 110, 220, 440, and 880 nM, respectively).

Figure 6.

Phenotypes of transgenic Arabidopsis lines overexpressing EI and ES isoforms of TFIIIA under osmotic stress. (A) Overexpressing TFIIIA-EI lines (EI-3,6) and TFIIIA-ES lines (ES-3, 6, 9) and Col-0 as wild type (WT) without (top) and with (bottom) treatment of 100 mM mannitol. (B) Relative accumulation levels of TFIIIA transgene mRNA (T) in ES-3, ES-6, and ES-9 using Tubulin 8 (TUB8) as a reference gene (C).

Figure 7.

Strong accordance between molecular and phenotypic evidences in multiple independent transgenic Arabidopsis lines overexpressing TFIIIA under salt stress. (A) mRNA accumulation ratio of endogenous TFIIIA EI to ES isoform in ES-3, ES-6, and ES-9 lines. Quantification (bottom) with three replicates for each line. (B) The high positive correlation (R² = 0.952) between the relative accumulation level of TFIIIA transgene mRNA and mRNA accumulation ratio of endogenous TFIIIA EI to ES isoform in ES-3, ES-6, and ES-9 lines with three replicates for each line. (C) Overexpression lines without (top) and with treatments of 50 mM NaCl (middle) and 100 mM NaCl (bottom).

Figure 8.

Autoregulatory model for homeostatic maintenance of the level of TFIIIA in the plant cell via conserved coupling of the alternative splicing and NMD of TFIIIA gene.