Alternative isoform regulation in human tissue transcriptomes

Abstract

Through alternative processing of pre-mRNAs, individual mammalian genes often produce multiple mRNA and protein isoforms that may have related, distinct or even opposing functions. Here we report an in-depth analysis of 15 diverse human tissue and cell line transcriptomes based on deep sequencing of cDNA fragments, yielding a digital inventory of gene and mRNA isoform expression. Analysis of mappings of sequence reads to exon-exon junctions indicated that 92-94% of human genes undergo alternative splicing (AS), ∼86% with a minor isoform frequency of 15% or more. Differences in isoform-specific read densities indicated that a majority of AS and of alternative cleavage and polyadenylation (APA) events vary between tissues, while variation between individuals was ∼2- to 3-fold less common. Extreme or ‘switch-like’ regulation of splicing between tissues was associated with increased sequence conservation in regulatory regions and with generation of full-length open reading frames. Patterns of AS and APA were strongly correlated across tissues, suggesting coordinated regulation of these processes, and sequence conservation of a subset of known regulatory motifs in both alternative introns and 3′ UTRs suggested common involvement of specific factors in tissue-level regulation of both splicing and polyadenylation.

The landscape of human mutually exclusive splicing

“…The 519 genes contain 942 differentially expressed MXEs (67% of the total 1,399 MXEs; Fig 3C). This number is in agreement with earlier analyses on small sets of MXEs (66 and 57%) (Wang et al , 2008; Abascal et al , 2015a). Expectedly, the expression of novel MXEs seems to be considerably more tissue specific than the expression of annotated MXEs and cassette exons (Appendix Fig S23).…”

“…This is best demonstrated by genes in arthropods that contain both multiple MXE clusters (“multi‐cluster”) and large clusters with up to 53 MXEs such as in the Drosophila Dscam genes (Graveley et al , 2004; Pillmann et al , 2011). This is in strong contrast to mutually exclusive splicing in vertebrates as there is to date no evidence of multi‐cluster or higher order MXE clusters (Matlin et al , 2005; Pan et al , 2008; Wang et al , 2008; Gerstein et al , 2014; Abascal et al , 2015a,b). …”

“…This modulation suggests crucial spatio‐temporal functional roles for MXEs and can in many cases not be observed at the gene level, as gene counts can remain largely invariant. A well‐known case for similar expression of MXEs in newborn heart but expression of only one MXE variant in adult heart is the ion channel CACNA1C (Diebold et al , 1992), an example for the switch of expression are the MXEs of the SLC25A3 gene (Wang et al , 2008). We surmise that the observed specificity in combination with a generally lower expression could also explain the discovery of 654 (358) novel exons that have so far eluded annotation efforts (Fig 1A, Appendix Fig S23).…”

“…High‐throughput studies suggested that not only 95–100% of all multi‐exon genes in human are affected (Pan et al , 2008; Wang et al , 2008; Gerstein et al , 2014) but also that alternative splicing patterns strongly diverged between vertebrate lineages implying a pronounced role in the evolution of phenotypic complexity (Barbosa‐Morais et al , 2012; Merkin et al , 2012). Five types of alternative splicing have been identified to contribute to most mRNA isoforms, which are differential exon inclusion (exon skipping), intron retention, alternative 5′ and 3′ exon splicing, and mutually exclusive splicing (Blencowe, 2006; Pan et al , 2008; Wang et al , 2008; Nilsen & Graveley, 2010). Mutually exclusive splicing generates alternative isoforms by retaining only one exon of a cluster of neighbouring internal exons in the mature transcript and is a sophisticated way to modulate protein function (Letunic et al , 2002; Meijers et al , 2007; Pohl et al , 2013; Tress et al , 2017a).…”

“…Opposed to arthropods, current evidence suggests that vertebrate MXEs only occur in pairs (Matlin et al , 2005; Gerstein et al , 2014; Abascal et al , 2015a), and genomewide estimates in human range from 118 (Suyama, 2013) to at most 167 cases (Wang et al , 2008). Despite these relatively few reported cases, mutually exclusive splicing might be far more frequent in humans than currently anticipated, as has been recently revealed in the model organism Drosophila melanogaster (Hatje & Kollmar, 2013).…”