Article Figures & Data
Figures
- Figure 1. Recursive splicing (RS) and the impact of splice donor (SD) competition on cryptic exon exclusion.
(A) Features of intronic RS in Drosophila. The ratchet point (RP) consists of a tandem splice acceptor and splice donor pair. There are hundreds of well-conserved RPs in Drosophila, which predominantly reside within long intronic contexts and exhibit the nucleotide preferences shown. The RP encompasses a cryptic RS-exon, which is short but of variable length (∼50 nt), and flanked by a downstream SD. In general, it is conceived that the RP SD suppresses the usage of the RS-exon SD by a competition mechanism because mutation of the RP SD results in inclusion of the cryptic RS-exon. (B) Proposed path for RS, with rt-PCR primers indicated that can monitor the recursive intermediate and mature mRNA product. (C) Transgenic CRISPR/Cas9 approach for efficient generation of RP mutants. Bx gene models displaying isoforms that use different transcription start sites. The RP is located within the longer isoform in the ∼31 kb intron 2. (D) CRISPR mutagenesis generated specific RP SD mutations as shown. Black nucleotides indicate matches to wild type, whereas red nucleotides designate changes relative in the Bx-RP SD. The allele ID is left of the sequence and changes to RP SD score on the right. The RS-exon SD score is also included; it is unchanged in these alleles. (E) Wildtype and RP SD mutants yield RS intermediate amplicons. However, unlike wild type, all weakened RP mutants include the cryptic RS-exon.
- Figure S1. Isolation of ratchet point (RP) alleles at Beadex and kuzbanian.
We used transgenic CRISPR-Cas9 to induce mutations near RPs of Beadex (Bx) and kuzbanian (kuz). Shown is a larger set of mutants obtained that preserve the core RP (juxtaposed splice acceptor and splice donor AG|GT dinucleotides) but bear downstream indels. The alignments on left are aligned to the reference sequence, whereas the RP-Donor columns at right depict the splice donor sequences that the spliceosome would encounter, along with their splice site scores as determined by NNSPLICE. All of the Bx-RP alleles are homozygous-viable (as the RP resides in the 5′ UTR), whereas some of the kuz-RP alleles are lethal, depending on whether they cause inclusion of frameshifting cryptic exon. Alleles highlighted in blue were saved and subjected to further molecular analysis; the others were since discarded owing to high similarity or identity of mutation patterns.
- Figure 2. Weakened kuz RP1 splice donor (SD) causes recursive splicing (RS)-exon inclusion and enables analysis of downstream RS.
(A) kuz gene models displaying two evenly spaced ratchet points (RPs) within its ∼50 kb intron 3. CRISPR/Cas9 was used to recover the RP1 SD mutations shown. Black nucleotides indicate matches to wild type, whereas red nucleotides designate changes relative in the kuz-RP SD. The allele ID is left of the sequence and changes to RP1 SD score on the right. The unaltered RS-exon SD score is also included for reference. (B) A model for kuz sequential RS. PCR amplicons are displayed using dotted boxes and primers as arrows. (C) Wild type and RP1 SD mutants yield similar RP1 intermediate amplicons. However, differences can be observed for RP2 intermediate and mRNA amplicons. Conversion of the high scoring RP1 SD to a medium or low scoring SD results in cryptic exon inclusion in RP2 intermediates and mRNA. Interestingly, whereas RP2 intermediates exhibit a steady conversion from cryptic exon skipping to fully cryptic exon inclusion as the RP1 SD weakens, mRNA amplicons always yield a minor level of cryptic exon skipped products (i.e., mature mRNAs). As kuz RS appears to be constitutive, the data suggest that weakened kuz RP1 SD can become activated to produce exon-skipped products (see Fig S2). (D) Multiple choices of SD to the downstream coding exon during kuz RS.
- Figure S2. Proposed intron removal trajectories for kuz RP2 intermediate and mRNA products from ratchet point (RP) splice donor (SD) mutants.
The schematic depicts the utilization of SD sequences that yield the observed pre-mRNA intermediates and mRNA in rt-PCR tests of the panel of kuz-RP mutants. Usage of RP SD is indicated with dark blue edges and arrows, whereas usage of the recursive splicing (RS)-exon SD is indicated in red. When kuz RP1 SD is mutated to a poor splice site, the RS-exon SD is activated (red) and the RP2 intermediate includes the cryptic RS-exon 1 (dotted orange box). However, in the next step (conversion to mRNA), one of the two remaining RP SD is used to generate canonical mRNA or mRNA with cryptic RS-exon 1 retention. Surprisingly, weak and poor RP1 SD can become selected for usage at the RP2 intermediate stage, despite the presence of two other strong SD.
- Figure 3. Systematic analysis of Drosophila recursive splicing (RS)-exon splicing properties.
(A) Above: Test backbone for RS splicing minigene reporters. We cloned ∼3 kb centered on the ratchet point (RP) and RS-exon from each test locus (in red) into a splicing minigene bearing the flanking exonic/intronic context of kuz-RP1. Common rt-PCR primers are used to evaluate the inclusion or exclusion of the RS-exon. Below: Comparison of RP and RS-exon splice donor (SD) scores using NNSPLICE. Selected recursively spliced loci whose inclusion/exclusion patterns are not well-explained by SD competition are indicated. (B) rt-PCR of splicing reporters containing cryptic RS-exons. For most substrates, the expected exon skipped amplicon was the major product. (C) Strategy to validate RS in minigene splicing reporters. Schematic of the RS pathway after RP SD disruption. Critically, the skipped cryptic RS-exon will be converted to constitutively included after this mutation. (D) RP SD mutations in cryptic RS-exon substrates lead to complete inclusion of the RS-exon in mRNA. (E) rt-PCR of splicing reporters containing expressed RS-exons. A range of RS-exon inclusion levels can be observed for these RS substrates. Notably, some do not match expectations based on SD scores (see panel 3A). For instance, msi and Ubx-m1 are dominantly included, despite having weaker RS-exon SD than their respective RP SD. (F) Validation that expressed RS-exons undergo RS because mutation of their RP-SDs yields constitutive exon inclusion.
- Figure S3. High conservation of expressed recursive splicing (RS)-exons.
Shown are UCSC genome browser screenshots for expressed RS-exons from smooth (sm) and Ultrabithorax (Ubx). Note that Ubx contains two expressed RS-exons (m1 and m2) and one “0-nt” ratchet point that contains a cryptic non-translated exon.
- Figure 4. Recursive splicing (RS)-exon sequences can autonomously determine their inclusion.
(A) Sequences of the different wildtype, swapped and mutant RS-exons tested. Wildtype sequences are in black, RS-exons swaps are in blue and mutations are in red. (B) Schematics of RS-exon variants built on the Ubx-m1 reporter. Ubx-m1 specific intronic sequence in red. Only the 51 nt Ubx-m1 RS-exon portion of the reporter was swapped with the RS-exons of Ubx-m2, Ubx-RP, or chinmo; Ubx-m1-RS-mut bears mutations internal to the RS-exon. (C) RS-exons contain information to regulate alternative splicing of RS-exons. The Ubx-m1 RS-exon reporter is dominantly included. Swapping the Ubx-m1 RS-exon with others mimics their inclusion or skipping behaviors. Moreover, the Ubx-m1-RS-mut reporter exhibits substantial skipping indicating that it is no longer appropriately included. (D) Schematics of RS-exon variants built on the Ubx-RP RS-exon reporter. Ubx-RP–specific intronic sequence in blue. Only the Ubx-RP RS-exon portion of the reporter was swapped with the RS-exons of Ubx-m2, Ubx-m1 or chinmo. The Ubx-RP-FP variant converts this RS-exon to a frame preserving (FP) length. (E) The Ubx-RP RS-exon reporter is predominantly skipped. Swapping the Ubx-RP RS-exon with others mimics their inclusion or skipping properties. The Ubx-RP-FP reporter is largely exon-skipped, indicating that mRNA stability is not a major confounding factor. (F) Sequences of mutant variants of Ubx-m1 and Ubx-m2 RS-exons. (G) Mutation tests of the Ubx-m1 RS-exon, which is dominantly included. As a control, swapping of its RS-exon with the Ubx-RP exon results in skipping. Mutation of the left half of Ubx-m1 (mutL) did not affect processing, but mutation of its right half (mutR) resulted in substantial RS-exon skipping. (H) Mutation tests of the Ubx-m2 RS-exon, which is dominantly included, even when inserted into the Ubx-RP backbone. The mutL variant was now substantially skipped, whereas the mutR variant exhibited normal inclusion.
- Figure 5. Intron pre-removal, as a proxy of exon junction complex (EJC) loss, induces recursive splicing (RS)-exon skipping.
(A) Left: model of RS-exon splicing, including the deposition of the EJC after removal of intron segment 1. Right: schematic of ∆intron RS reporters. These will not recruit EJC prior to removal of intron segment 2. (B) ∆intron reporters for expressed RS-exons exhibit higher levels of exon skipping. (C) Cryptic RS-exon reporter from Ubx-RP that is normally skipped is unaffected by pre-removal of intron segment 1.
In this Issue
Subjects
Related Articles
Cited By...
- No citing articles found.