Real-time imaging of cotranscriptional splicing reveals a kinetic model that reduces noise: implications for alternative splicing regulation

Abstract

Splicing is a key process that expands the coding capacity of genomes. Its kinetics remain poorly characterized, and the distribution of splicing time caused by the stochasticity of single splicing events is expected to affect regulation efficiency. We conducted a small-scale survey on 40 introns in human cells and observed that most were spliced cotranscriptionally. Consequently, we constructed a reporter system that splices cotranscriptionally and can be monitored in live cells and in real time through the use of MS2-GFP. All small nuclear ribonucleoproteins (snRNPs) are loaded on nascent pre-mRNAs, and spliceostatin A inhibits splicing but not snRNP recruitment. Intron removal occurs in minutes and is best described by a model where several successive steps are rate limiting. Each pre-mRNA molecule is predicted to require a similar time to splice, reducing kinetic noise and improving the regulation of alternative splicing. This model is relevant to other kinetically controlled processes acting on few molecules.

Figure 1.

Co-transcriptional splicing and characterization of the MINX reporter cell lines. (A) RT-PCR assay to determine whether introns splice cotranscriptionally or not. A hypothetical gene is schematized with the position of the primers used for RT and the competitive PCR. (B) Scheme of the MINX constructs. SAmut_MS2in contains a single AG to GG mutation at the splice acceptor site. The green, red, and blue bars represent the positions of the FISH probes, and the distance is indicated in nucleotides. (C) RNA polymerase II accumulates at the reporter transcription site. WT_MS2in stable cells were transfected with Tat and processed for in situ hybridization with an MS2 probe (red), and for immunofluorescence against the large subunit of RNA polymerase II (RPB1, green). Insets show enlarged images. Bar, 10 µm. (D) WT_MS2in is spliced cotranscriptionally. WT_MS2in or SAmut_MS2in cells were treated with SSA when indicated, and processed for in situ hybridization with probes recognizing the intron (MS2, red), exon 2 (LacZ, blue), or the spliced junction (spliced, green). Bar, 10 µm. Right panels show intensity line scans of the images, using the lines defined by the arrowheads. (E) Quantification of cotranscriptional splicing by RT-PCR. The indicated cell lines were treated as in D, and total RNAs were extracted and RT-PCR amplified with the indicated primers (see scheme, the scissors depict the 3′ end cleavage site). The position of the primers used for reverse transcription are also indicated. pdT, oligo dT primer; 3′ uncleaved primer, primer located downstream the polyadenylation site. Spliced RNAs: 317 bp; unspliced RNAs: 671 bp.

Figure 2.

Recruitment of spliceosomal components to the MINX transcription site. (A–C) Recruitment of snRNPs. WT_MS2in and SAmut_MS2in cells were transfected with Tat vector and processed for multicolor FISH experiments with probes against LacZ in exon 2 (red), and each of the snRNAs (green). In B, cells were treated with SSA. Insets show enlarged views. (D) SF3b155 recruitment in SSA-treated cells. WT_MS2in cells were transfected with vectors expressing Tat and MS2–GFP (green), treated with SSA for 3 h, and processed for immunofluorescence against SF3B155 (red). Bars, 10 µm.

Figure 3.

Analysis of splicing kinetics by FRAP. (A) Time series of FRAP experiments. WT_MS2ex2 (top) or WT_MS2in (bottom) cells were transfected with vectors expressing Tat and MS2–GFPnsl_nes. Transcription sites were bleached and 3D images were captured for 9–11 min. The time after the bleach is indicated in seconds. pre, prebleach image; post, post-bleach image. The arrow points to the transcription site. Bar, 10 µm. (B) FRAP recovery plots. Normalized fluorescence intensity at the transcription site is plotted as a function of time (±SEM). Time is in seconds, and the zero time point corresponds to the bleach. (C) Scheme describing the kinetic models. (top) A single step is rate limiting (k_spl; red). (bottom) Multiple steps are rate-limiting, all with the same rate (k_spl; red). (D) Determination of the number of limiting steps. Plot of the AIC values for the optimal choice of the rate of the limiting steps (α) as a function of the number of limiting steps. L is the linear model that was included as a null hypothesis. (E) Optimal fit of the model to the experimental curve. The FRAP curve of WT_MS2in cells was fitted with either a single-step model (red), a three-step model (green), or a nine-step model (blue).

Figure 4.

A three-step model improves the regulation of alternative splicing. (A) A three-step model improves the regulation of cotranscriptional splicing events. (A, top) Scheme of a cotranscriptional alternative splicing event regulated by the elongation rate of RNA polymerase II. Splicing occurs at splice acceptor SA1 only when the splicing reaction is completed before splice acceptor SA2 is synthesized. When both SA2 and SA1 are present on the pre-mRNA, splicing always occurs at SA2. When the polymerase has a fast elongation rate, 25% of the splicing events take place at site SA1, whereas when the polymerase elongates slowly, 75% of the pre-mRNAs are spliced at SA1. (A, bottom) Calculated curves of the appearance of the spliced product as a function of time after transcription (seconds), for models having a single (left) or three limiting steps (right), and optimized to fit the MINX kinetics. The vertical lines point to the elongation time required to obtain 25% and 75% of splicing at SA1. (B) A three-step model improves the regulation of posttranscriptional splicing events. (B, top) Schematics of a posttranscriptional alternative splicing event. Product A is produced constitutively with n rate-limiting steps, whereas product B is produced with a single rate-limiting step that is regulated. (B, bottom) Calculated curves of the appearance of the spliced product A as a function of the rate of the competing splicing event B, for models having a single (left) or three limiting steps (right), and optimized to fit the MINX kinetics. The vertical lines point to the rates δ required to obtain 25% and 75% of product A.