Regulation and Function of Alternative mRNA Isoform Expression in Mammals

Summary Most human genes contain introns, and presence of introns often increases the expression of the host gene, a phenomenon known as intron-mediated enhancement (IME). IME has been observed in diverse genes in animals, plants and fungi and often varies in magnitude across introns. However, little is known about how

introns impact expression or what intron features modulate IME activity. Recently, we have described a novel phenomenon that we call exon-mediated activation of transcription starts (EMATS), in which the splicing of internal exons impacts the spectrum of promoters used and expression level of the gene. EMATS acts at a

distance of up to a few kb, can alter gene expression by at least severalfold, and appears more active at certain promoters – especially intrinsically weak promoters. The detailed sequence requirements and mode of action of EMATS are not yet known. This proposal is seeks to understand the rules that govern IME and

EMATS, to improve the prediction of gene expression and to enable methods to modulate gene expression by altering splicing. It is organized around the following aims. SA1. Determine the sequence dependence of intron-mediated enhancement. SA2. Explore the scope and rules for EMATS regulation. In Aim 1, we will generate a library of many thousands of distinct random sequences inserted into an intron in a

dual fluorescent reporter system that is chromosomally integrated into human cells. This design will enable high-throughput measurement of the effects of each intron on nascent RNA, mature RNA and protein levels, and these data will be used to identify motifs that enhance or silence expression in a splicing-dependent

manner from an intronic location. In aim 2, we will systematically derive and test rules for how EMATS regulation depends on the location and sequence of the internal exon and on properties of the involved promoter. Finally, we will use the information learned about IME and EMATS to improve predictions of gene

expression from primary sequence. Together, the research described in these aims will establish rules governing how splicing impacts gene expression in mammalian genomes. Identification of motifs that function as splicing-dependent activators or silencers of expression can be used to improve prediction of expression

from genome sequence and may enable detection of intronic variants that alter expression. Understanding how splicing impacts expression may also enable new approaches for gene expression modulation.