Role of Phytochromes in Red Light-Regulated Alternative Splicing in Arabidopsis thaliana: Impactful but Not Indispensable

Abstract

Light is both the main source of energy and a key environmental signal for plants. It regulates not only gene expression but also the tightly related processes of splicing and alternative splicing (AS). Two main pathways have been proposed to link light sensing with the splicing machinery. One occurs through a photosynthesis-related signal, and the other is mediated by photosensory proteins, such as red light-sensing phytochromes. Here, we evaluated the relative contribution of each of these pathways by performing a transcriptome-wide analysis of light regulation of AS in plants that do not express any functional phytochrome (phyQ). We found that an acute 2-h red-light pulse in the middle of the night induces changes in the splicing patterns of 483 genes in wild-type plants. Approximately 30% of these genes also showed strong light regulation of splicing patterns in phyQ mutant plants, revealing that phytochromes are important but not essential for the regulation of AS by R light. We then performed a meta-analysis of related transcriptomic datasets and found that different light regulatory pathways can have overlapping targets in terms of AS regulation. All the evidence suggests that AS is regulated simultaneously by various light signaling pathways, and the relative contribution of each pathway is highly dependent on the plant developmental stage.

Keywords: Arabidopsis; alternative splicing; light signaling; phytochrome.

Figure 1

Phytochromes are needed to fine-tune the AS landscape. (a) Experimental setup. Plants were grown under a 12:12 photoperiod for 10 days. Then, they were either kept in darkness or treated with a 2-h R-light pulse in the middle of the night. The arrows indicate the harvesting time. (b) Differentially spliced events (DSEs) and (c) differentially spliced genes (DSGs) upon R-light treatment in WT and phyQ seedlings. (d) Amount of each type of AS event in the DSEs between WT dark vs. WT R, and phyQ dark vs. phyQ R. Alt, alternative (5′ and/or 3′) splicing site; ES: exon skipping; IR, intron retention; ASCE, AS affecting a consensus exon; CSP, complex splicing pattern.

Figure 2

Representative light-regulated but phytochrome-independent AS events. Coverage plot and corresponding gene models of (a) U2AF⁶⁴ (AT4G36690), (b) SR30 (AT1G09140), and (c) KH29 (AT5G56140). The coverage plots were generated by merging data from three distinct biological replicates. (d–f) Mean percentage spliced-in (PSI) values for each gene across all experimental conditions. (g,i,k) Gene models of the different isoforms annotated for each gene. Boxes represent exons, lines represent introns, and arrows represent the positions of the primers used for validation. (h,j,l) Validation of differentially spliced events (DSEs) through agarose gel electrophoresis of RT–PCR products.

Figure 3

Representative light and phytochrome-regulated splicing events. Coverage plot and corresponding gene models of (a) TTM1 (AT1G73980), (b) SMD1a (AT3G07590), and (c) ANK1 (AT5G02620). The coverage plots were generated by merging data from three independent biological replicates. (d–f) Mean percentage spliced-in (PSI) or mean percentage intron retention (PIR) values for each gene across all experimental conditions. (g,i,k) Gene models of the different isoforms annotated for each gene. Boxes represent exons, lines represent introns, and arrows represent the positions of the primers used for validation. (h,j,l) Validation of differentially spliced events (DSEs) through agarose gel electrophoresis of RT–PCR products.

Figure 4

The effect of phytochromes on gene expression. (a) Overlap between differentially expressed genes (DEGs) upon an acute R-light pulse in WT and phyQ. (b) Overlap between DEGs and DSGs in WT. (c–f) Normalized read count values of (c) HB-2 (AT4G16780), (d) PRR7 (AT5G02810), (f) RSZ22A (AT2G24590), and (e) RFC3 (AT3G17170). Statistical analysis was performed using ASpli. For HB-2 and PRR7, the statistical significance of the effect of genotype is shown. For RFC3 and RSZ22a, the statistical significance of the contrast between WT R and phyQ R is shown. The mean value and standard deviation of three independent biological replicates are shown. *: FDR < 0.05, ***: FDR < 0.001.

Figure 5

Analysis of gene ontology (GO) terms. Selected GO terms enriched among differentially spliced genes (DSGs) and differentially expressed genes (DEGs) identified in WT plants and DSGs and DEGs identified in WT but not in phyQ mutant seedlings are shown. The color gradient represents p values, and the differences in bubble size correlate with the enrichment factor.

Figure 6

Comparison of splicing patterns in multiple datasets. (a) Overlap between differentially spliced genes (DSGs) upon red (R) light treatment in seeds [48], etiolated seedlings [6], and light-grown plants (this work). Both the seed and etiolated seedlings datasets were reanalyzed with the same filters used in this work. (b) Overlap between the union of DSGs in (a) and splicing regulatory factors. The list of splicing regulatory factors was taken from [3]. (c) Overlap between DSGs in (a) and DSGs regulated by COP1 (WT vs. cop1-4) or the splicing factors SPFS-RRC1-SWAP1 (WT vs. spfs rrc1 swap1) upon R-light treatment. The lists of DSGs were taken from [22,50]. (d) Overlap between genes regulated at the splicing level by cryptochromes, phytochromes, and chloroplasts (PHY-independent). CRY-regulated genes were obtained from the union of DSGs in WT but not in the cry1 cry2 double mutant upon blue-light treatment in the dataset by Wang et al. (2016) [51] and Zhao et al. (2022) [5]. PHY-regulated genes were obtained from the union of DSGs between WT and the phyA phyB double mutant reported by [3], and the DSGs between WT and phyB and WT and phyA phyB were obtained from the reanalysis of the dataset of Wu et al. (2019) [47]. The DSGs in the phyQ mutant (Figure 1c) were labeled PHY-independent.