Sequencing, Array Studies Find Nearly All Human Genes Undergo Alternative Splicing
NEW YORK (GenomeWeb News) – Alternative splicing of human genes occurs more frequently than previously documented, a trio of new papers suggest.
“A decade ago, alternative splicing of a gene was considered unusual, exotic,” MIT biologist Christopher Burge, senior author on one of the papers, said in a statement. “But it turns out that’s not true at all — it’s a nearly universal feature of human genes.”
Burge led a team of researchers from the Massachusetts Institute of Technology, Illumina, and elsewhere who used mRNA sequencing to investigate alternative splicing in more than a dozen human tissue and cell lines. That work, which appeared online yesterday in Nature, indicates that more than 90 percent of human genes undergo alternative splicing. The team also found that alternative splicing events vary from one tissue to another and — to a lesser extent — between individuals.
In a second paper, Canadian researchers used a combination of mRNA-Seq and EST-cDNA sequence data to investigate splice junctions in six human tissues. That research, appearing online yesterday in Nature Genetics, uncovered new splice junctions in roughly 20 percent of multi-exon genes. Based on these results, the team estimated that roughly 95 percent of gene transcripts undergo alternative splicing.
Meanwhile, in another paper in Nature Genetics appearing online yesterday, researchers at Seattle’s Rosetta Inpharmatics and elsewhere used whole-transcript custom microarrays to look at genome-wide expression of alternative splicing events in nearly 50 human tissue and cell line samples, focusing on the cis-regulation of alternative splicing events. That effort led to the discovery of more than 140 motifs predicted to regulate alternative splicing.
In the past, microarray and EST research suggested that roughly 70 percent of human genes underwent alternative splicing. But while informative, those approaches didn’t reveal everything that’s happening in the cell — and, researchers noted, they can’t always distinguish between similar mRNA isoforms.
Burge and his colleagues used mRNA-Seq using an Illumina Genome Analyzer to assess alternative splicing in 15 different human tissue and cell lines. First, they isolated RNA from the cells or tissue and selected for mRNA, which they then converted to double-stranded cDNA and, eventually, applied to an Illumina flow cell. Next, they mapped to both exons and splice junctions to find new splicing events.
"The [mRNA-Seq] technology gives you a really high-resolution view of the transcriptome," lead author Eric Wang, a graduate student in Burge’s lab, told GenomeWeb Daily News. "I think it will accelerate progress in the field now that we actually have the tools to more accurately profile splicing genome-wide".
Based on their data, the researchers estimated that some 92 percent to 94 percent of human genes undergo alternative splicing. And about 86 percent of all genes had minor isoforms that made up at least 15 percent of the transcripts for that gene in one or more samples. The researchers also found 1,413 new exons and thousands of new splice junctions.
When they compared samples from the same tissue taken from six different individuals, the researchers found that between about ten and 30 percent of the isoforms varied from one individual to the next. But they uncovered even more variation from one tissue to the next — about two to three times that observed between individuals.
“[M]ost alternative splicing events are regulated between tissues, providing an important element of support for the hypothesis that alternative splicing is a principal contributor to the evolution of phenotypic complexity in mammals,” the authors wrote.
Next, the researchers looked specifically at eight different types of splicing events — such as skipped exons, alternative 5’ splice sites, tandem 3’ UTRs and so on. Their results indicate that these different types of splicing events are differentially regulated depending on the tissue tested.
The team is currently interested in looking at how alternative splicing varies in different tissues, under different conditions, and in response to different types of biological stimuli, Wang noted. In addition to mRNA-Seq, he added, the team is using CLIP-Seq — UV cross-linking and immunoprecipitation-combined deep sequencing — to identify the biophysical targets of these splicing factors.
For his part, Wang is focusing on myotonic dystrophy, a disease in which splicing is mis-regulated — looking at the function of specific splicing factors in normal cells and disease cells by perturbing splicing factors and then using sequence data to identify their global targets.
Meanwhile, Benjamin Blencowe, a molecular geneticist at the University of Toronto, led a team of researchers using a combination of mRNA-Seq with an Illumina Genome Analyzer and EST-cDNA data to look at new and known splice junctions in 15,702 multi-exon gene clusters from six normal human tissues: whole brain, cerebral cortex, heart, skeletal muscle, lung, and liver.
They uncovered new splice junctions in about 20 percent of the multi-exon genes tested — or between 4,294 and 11,099 new splice junctions overall. Based on their results, Blencowe and his team estimated that the genes in major human tissues undergo roughly 100,000 intermediate to high-abundance alternative splicing events.
They also began deciphering how different splice junctions related to the tissue in which they were detected, noting that “many of the new splice junctions reflect the physiological origin of the tissues analyzed, and therefore likely represent examples of new tissue-specific alternative splicing events, in addition to new alternative splicing events in transcripts with tissue-restricted expression patterns.”
But researchers are also continuing to learn about the transcriptome using microarrays. Rosetta Inpharmatics researcher Jason Johnson led a team of researchers who designed microarrays to assess 203,672 exons and 178,351 exon-exon junctions in 17,939 human genes.
By looking at 24,426 alternative splicing events in 48 human samples, they found differential expression of 9,516 splicing events in one or more tissues as well as threefold or greater changes in the expression of 11,700 of the 18,000 genes tested.
Johnson and his team also found 143 new cis-regulatory motifs that fell into five main groups. Subsequent experiments suggested that some of these appear to be enriched in all of the tissues and cell lines tested, while others exhibited tissue specificity.
“To further our understanding of alternative splicing expression and regulation, we generated the first human genome-wide alternative splicing-event compendium ... and undertook an unbiased, systematic search to identify and describe putative regulatory motifs,” Johnson and his colleagues wrote.