The serine/arginine-rich protein family in rice plays important roles in constitutive and alternative splicing of pre-mRNA

Abstract

Ser/Arg-rich (SR) proteins play important roles in the constitutive and alternative splicing of pre-mRNA. We isolated 20 rice (Oryza sativa) genes encoding SR proteins, of which six contain plant-specific characteristics. To determine whether SR proteins modulate splicing efficiency and alternative splicing of pre-mRNA in rice, we used transient assays in rice protoplasts by cotransformation of SR protein genes with the rice Waxy(b) (Wx(b))-beta-glucuronidase fusion gene. The results showed that plant-specific RSp29 and RSZp23, an SR protein homologous to human 9G8, enhanced splicing and altered the alternative 5' splice sites of Wx(b) intron 1. The resulting splicing pattern was unique to each SR protein; RSp29 stimulated splicing at the distal site, and RSZp23 enhanced splicing at the proximal site. Results of domain-swapping experiments between plant-specific RSp29 and SCL26, which is a homolog of human SC35, showed the importance of RNA recognition motif 1 and the Arg/Ser-rich (RS) domain for the enhancement of splicing efficiencies. Overexpression of plant-specific RSZ36 and SRp33b, a homolog of human ASF/SF2, in transgenic rice changed the alternative splicing patterns of their own pre-mRNAs and those of other SR proteins. These results show that SR proteins play important roles in constitutive and alternative splicing of rice pre-mRNA.

Figure 1.

Putative Structure of Rice SR Proteins. Only major types of domains are shown (not to scale). SR protein families were classified according to the structure of the domains. SR20 may be a truncated protein because of the presence of a premature stop codon. All putative protein sequences are shown in Supplemental Figure 1 online. RRM, purple; RS, blue; ψRBD, pseudo RRM, green; Zn, zinc knuckle, yellow.

Figure 2.

Expression of SR Protein mRNA in Different Tissues of Rice. Total RNA from different tissues of rice cv Kinmaze was analyzed. RT-PCR was performed using specific primers for each SR protein mRNA. R, root; L, leaf; S, immature seeds at 15 d after pollination; C, cell culture.

Figure 3.

Alternative Splicing Patterns of Rice SR Protein Genes. Schematic diagram of rice SR protein genes having alternative splicing form and their mRNA isoforms. Constitutive exons are shown as white boxes, and alternative exons are shown as gray boxes. Asterisks indicate premature stop codons.

Figure 4.

Effects of Various Rice SR Proteins on the Expression of the Wx^b-gus Reporter Gene. The assays were performed with transiently transformed protoplasts derived from the Oc cell culture. The GUS activities of the extracts obtained from the protoplasts were measured. (A) Constructs used to analyze the effects of SR protein genes on the reporter gene expression. The promoter-ATG region of Wx^b was translationally fused to the gus reporter gene (Isshiki et al., 1998). Arrows show the relative positions of primers for RT-PCR. The cDNAs of RSZp23, RSp29, SRp32, SRp33a, SCL25, SCL26, RSp33, RSZ36, and RSZ37a were fused with the cauliflower mosaic virus 35S promoter. (B) to (D) GUS activity in cotransfected protoplasts. The indicated SR plasmids were added to protoplasts together with a Wx^b-gus fusion gene. Ten micrograms each of Wx^b-gus and various amounts of 35S-SR plasmid DNA (1, 2, 5, and 10 μg) were added to 5 × 10⁶ protoplasts. The mean and standard error for three independent experiments are shown. (B) RSp29 and RSZp23, which enhanced GUS activity. (C) SRp32, SRp33a, SCL25, and SCL26, which reduced GUS activity. (D) RSp33, RSZ36, and RSZ37a, which did not affect GUS activity when 1 to 2 μg of SR protein genes was added.

Figure 5.

Analysis of RNA to Examine Splice Site Selection in Transfection Assays. Total RNA isolated from transfected protoplasts shown in Figure 4 was subjected to RT-PCR analysis. PCR conditions were the same for spliced and unspliced transcripts. The RT-PCR products were sequenced to confirm the splice sites. The top loading gels show the spliced transcripts of Wx^b-gus. The spliced products are indicated at right: the top fragment shows the site 1–spliced transcript of Wx^b-gus, and the bottom fragment shows the site 2–spliced transcript. The bottom PCR product is 93 bp shorter than the top one. E1 and E2 show exon 1 and exon 2 of the Wx^b gene, respectively. The bottom loading gels show unspliced transcripts of Wx^b-gus. (A) Effects of 35S-RSp29 on the splicing of Wx^b intron 1. (B) Effects of 35S-RSZp23 on the splicing of Wx^b intron 1. (C) Effects of eight 35S-SR plasmids on splicing patterns. The extracts obtained from transfected protoplasts were used for RNA isolation for RT-PCR analysis. Ten micrograms of Wx^b-gus and 2 μg of 35S-SR DNAs were cotransfected into 5 × 10⁶ protoplasts.

Figure 6.

Splicing Efficiencies of RSp29-SCL26 Chimeric Genes Tested in Rice Protoplasts. (A) Diagram of RSp29 and SCL26 proteins and the chimeric proteins used in the experiments. Domains of RSp29 and SCL26 proteins are shown by white and gray boxes (not to scale), respectively. (B) GUS activity and splicing pattern. The Wx^b-gus plasmid (10 μg) and each 35S-SR plasmid (2 μg) were transfected into 5 × 10⁶ protoplasts. The GUS activities of the extracts obtained from transfected protoplasts are shown. The mean and standard error for three independent experiments are shown. The aliquots of the extracts were used for RNA isolation for RT-PCR analysis. The top loading gels show the spliced transcripts of Wx^b-gus. The bottom loading gels show unspliced transcripts of Wx^b-gus. E1 and E2 show exons 1 and 2 of the Wx^b gene, respectively.

Figure 7.

Effects of the ESE-Like Sequences in Wx^b Exon 1 on Splicing by RSp29 and RSZp23. (A) Sequences of Wx^b exon 1, which does not contain the start codon. It is present in exon 2. Sequences that are similar to the human SC35 and ASF/SF2 binding sequences are shown by a single underline (A) and double underline (B), respectively. The 5′ splice sites 1 and 2 are shown by two open triangles and a closed triangle, respectively. The SC35 binding sequences overlaps with site 2. (B) Mutant vector constructs used for splicing assays. Wx^b(mutA)-gus was changed to ACCAGTA from TGCAGTC. Wx^b(mutB)-gus was changed to a pyrimidine-rich sequence (CCTTCTTG) from a purine-rich sequence (GGAAGAAC). (C) GUS activity and splicing in cotransfected protoplasts. Protoplasts (5 × 10⁶) were transfected by Wx^b-gus or the mutant Wx^b-gus plasmid (10 μg) with 35S-RSp29 or 35S-RSZp23 plasmids (2 μg). The mean and standard error for three independent experiments are shown. RT-PCR analysis of RNA isolated from protoplasts transfected with mutant Wx^b-gus constructs. The top loading gels show spliced transcripts of Wx^b-gus. The bottom loading gels show unspliced transcripts of Wx^b-gus. E1 and E2 show exons 1 and 2 of the Wx^b gene, respectively.

Figure 8.

Overexpression of RSZ36 Affects the Alternative Splicing of Its Own RNAs in Transgenic Rice Plants. (A) Schematic representation of the RSZ36 gene structure and its mRNA isoforms detected in RSZ36-overexpressing plants. Exons are shown as boxes and introns as lines. A gray box indicates an alternative exon in intron 2; a hatched box shows an alternative intron in exon 5. Positions of primers used for RT-PCR analyses are shown with arrowheads. Asterisks indicate premature stop codons. (B) RT-PCR analysis of RSZ36 transcripts in the T0 generation of RSZ36-overexpressing plants. Lanes 3-1, 3-3, and 3-4 were regenerated from the same transformed calli. Lanes 3-1, 4, 12, 13, and 15 were independent transgenic lines. The top panel displays the expression of the endogene and transgene amplified using the F-2 and R primers. This PCR was performed with 25 cycles. The bottom panel shows only the expression of the endogenous RSZ36 gene amplified by F-1 and R primers. This PCR was performed with 30 cycles for the detection of endogenous RSZ36 transcripts. a, b, and c correspond to mRNA-a, -b, and -c, respectively.

Figure 9.

Overexpression of SRp33b Affects the Alternative Splicing of SRp33a and SRp32 in Transgenic Rice Plants. (A) Schematic representations of SRp33a and SRp32 gene structures and their mRNA isoforms detected in SRp33b-overexpressing plants. Gray boxes indicate alternative exons. Asterisks indicate premature stop codons. (B) RT-PCR analysis in the T0 generation of SRp33b-overexpressing plants. The lane number shows independent transgenic lines. SRp33a, SRp32, and SRp33b transcripts were amplified by each specific primer designed from 5′ and 3′ untranslated region sequences. The RT-PCR data of SRp33b display the expression of the endogene and transgene. a and c show transcripts encoding the full-length protein. b, d, and e show alternative splicing products. Total RNA was isolated from the leaf of transgenic rice plants.