Unusual intron conservation near tissue-regulated exons found by splicing microarrays

Abstract

Alternative splicing contributes to both gene regulation and protein diversity. To discover broad relationships between regulation of alternative splicing and sequence conservation, we applied a systems approach, using oligonucleotide microarrays designed to capture splicing information across the mouse genome. In a set of 22 adult tissues, we observe differential expression of RNA containing at least two alternative splice junctions for about 40% of the 6,216 alternative events we could detect. Statistical comparisons identify 171 cassette exons whose inclusion or skipping is different in brain relative to other tissues and another 28 exons whose splicing is different in muscle. A subset of these exons is associated with unusual blocks of intron sequence whose conservation in vertebrates rivals that of protein-coding exons. By focusing on sets of exons with similar regulatory patterns, we have identified new sequence motifs implicated in brain and muscle splicing regulation. Of note is a motif that is strikingly similar to the branchpoint consensus but is located downstream of the 5' splice site of exons included in muscle. Analysis of three paralogous membrane-associated guanylate kinase genes reveals that each contains a paralogous tissue-regulated exon with a similar tissue inclusion pattern. While the intron sequences flanking these exons remain highly conserved among mammalian orthologs, the paralogous flanking intron sequences have diverged considerably, suggesting unusually complex evolution of the regulation of alternative splicing in multigene families.

Figure 1. Array Design and Analysis of Splicing-Sensitive Microarray Data

(A) Probe design and expression counts of alternative event-specific probe sets. Probe sets were designed to both the skipping splice junctions and include splice junctions, as well as the alternative exons when possible, ensuring that probe sets are specific to the exon skipped or included spliced isoforms. Probe sets for constitutive portions of the gene were used to measure overall expression of the locus. (B) Scatterplot of skip probe set intensity versus include probe set intensity for Biaip exon 15 in brain (squares) and nonbrain (circles) tissue samples. Each data point is derived from one RNA sample and represents the skip-to-include ratio for that sample. The lines represent the robust regression coefficient (constrained to go through the origin) for each tissue group. The log₂ difference in the slopes is −2.53, indicating 5.7-fold inclusion in brain relative to nonbrain tissues.

Figure 2. Conservation of Cassette Exons Preferentially Included in Brain

(A) Extreme conservation in flanking intronic sequences of Baiap1 cassette exon seen in University of California Santa Cruz genome browser. (B) Median conservation probability at each base 100 base pairs upstream (left) and downstream (right) of the exon for 36 brain-included exons (gray circles), 36 brain-skipped exons (hollow gray squares), about 1,000 skipped mouse exons conserved and alternatively spliced in both human and mouse (gray triangles), and about 47,000 constitutive mouse exons (black circles). These last two sets of exons are from an EST/mRNA-based study [24]. (C) Histograms of the median probability of conservation per 100 base pairs upstream and downstream of the brain preferentially included (light gray), constitutive exons (black), and overlapping regions (dark gray).

Figure 3. Counts of RNA Binding Motifs in Intron Sequences adjacent to Brain-Regulated Exons

The 100 base pairs upstream and downstream regions for constitutive exons (control), preferentially brain-included (brain include), and preferentially brain-skipped exons (brain skip) were evaluated for the presence of sequences related to binding sites for known splicing regulators. Sequences used as the consensus binding sites were Nova: UCAUU or UCAUC; Fox-1: GCAUG; PTB/nPTB: CUCUCU; hnRNP-H/F: GGGGG; and hnRNP A1: UAGGG.

Figure 4. New Motifs in the Tissue-Regulated Exons

(A) A new motif UGYUUUC found upstream of brain-included exons. The logo is shown above, and the graph of the frequency of this motif in different regions of the brain-enriched exons is below. (B) A sequence similar to the recognition sites for SF1 and QK proteins is enriched near the 5′ splice site of the muscle-included exons. The logo is shown above, and the graph of the frequency of the core of this motif in different regions of the muscle-enriched exons is below. (C) Locations of multiple copies of conserved sequences matching the motif in (B) found in a conserved “bump” downstream of a muscle-included (and brain-skipped) exon 16 in the Vldlr gene.

Figure 5. Tissue-Regulated Splicing Controls the C-Terminal Sequences of Mouse MAGI Proteins

(A) Microarray intensity (top) and RT-PCR results (bottom) for the alternative exon in Baiap1. (B) Microarray intensity (top) and RT-PCR results (bottom) for the alternative exon in 653047C02Rik. (C) RT-PCR results (bottom) for the alternative exon in Avricp1 (this gene was not present on the array). (D) Diagram of the alternative splicing and coding patterns at the 3′ end of the mouse MAGI genes.

Figure 6. Multiple Alignment of the Flanking Intron Sequences Downstream of the Alternative Exons from Baiap1 (MAGI-1), Acvrinp1 (MAGI-2), and 4732496O19Rik (MAGI-3) and the Orthologous Sequences from Rat, Human, Dog, and Chicken

While the orthologous sequences have high conservation between them (A), the paralogous sequences have diverged considerably (B). The 5′ splice sites are at the left. Genome sequences similar to TACTAAC are between gray bars. In the region downstream of the 5′ splice site these may act as regulatory binding sites for SF1, quaking (QK), or some other factor. Fox-1 sites are shown between black bars.