Global 3′ UTR shortening has a limited effect on protein abundance in proliferating T cells

Abstract

Alternative polyadenylation is a cellular mechanism that generates mRNA isoforms differing in their 3 0 untranslated regions (3 0 UTRs). Changes in polyadenylation site usage have been described upon induction of proliferation in resting cells, but the underlying mechanism and functional significance of this phenomenon remain largely unknown. To understand the functional consequences of shortened 3 0 UTR isoforms in a physiological setting, we used 3 0 end sequencing and quantitative mass spectrometry to determine polyadenylation site usage, mRNA and protein levels in murine and human naive and activated T cells. Although 3 0 UTR shortening in proliferating cells is conserved between human and mouse, orthologous genes do not exhibit similar expression of alternative 3 0 UTR isoforms. We generally find that 3 0 UTR shortening is not accompanied by a corresponding change in mRNA and protein levels. This suggests that although 3 0 UTR shortening may lead to changes in the RNA-binding protein interactome, it has limited effects on protein output.

Subcellular RNA profiling links splicing and nuclear DICER1 to alternative cleavage and polyadenylation

“…We show that the vast majority of APA isoforms are not differentially exported from the nucleus nor do they differ in mRNA stability. In addition, although aUTRs have been shown to control membrane localization of their resulting proteins at an individual gene level (Berkovits and Mayr 2015), the global impact of aUTRs on translation efficiency appears limited (Gruber et al 2014). This raises the possibility that many APA events in resting cells may be inconsequential and not specifically controlled events.…”

The differential expression of alternatively polyadenylated transcripts is a common stress-induced response mechanism that modulates mammalian mRNA expression in a quantitative and qualitative fashion

“…′ UTR length to influence expression to a minor extent (Spies et al 2013;Gruber et al 2014;Gupta et al 2014). Differences in 3 ′ UTR length provoked by 3 ′ UTR-APA can, however, influence important functions of the mRNA without necessarily changing mRNA abundance, as has recently been highlighted by the finding that cellular proteins can bind to the 3 ′ UTR and thereby determine protein localization (Berkovits and Mayr 2015).…”

“…Early work on alternative polyadenylation in proliferating and cancer cells suggested that APA-provoked changes in 3 ′ UTR length inversely correlated with the expression levels of the respective mRNAs and proteins, likely caused by modified miRNA and RBP binding (Sandberg et al 2008;Mayr and Bartel 2009). This hypothesis was challenged by other studies that reported 3 ′ UTR length to have a limited effect on mRNA and protein expression levels in yeast and mice (Spies et al 2013;Gruber et al 2014;Gupta et al 2014). As a most interesting new dimension, a recent report implicated 3 ′ UTR-APA to influence the localization of newly translated proteins and thus described an essential cellular role of APA beyond modulating mRNA abundance (Berkovits and Mayr 2015).…”

A comprehensive analysis of 3′ end sequencing data sets reveals novel polyadenylation signals and the repressive role of heterogeneous ribonucleoprotein C on cleavage and polyadenylation

“…However, genome-wide measurements of mRNA and protein levels in dividing and resting cells revealed that systematic 3 ′ UTR shortening has a relatively minor impact on mRNA stability, translation, and protein output (Spies et al 2013;Gruber et al 2014). Instead, evidence has started to emerge that 3 ′ UTR shortening results in the loss of interaction with various RBPs, whose effects are not limited to mRNA stability and translation (Gupta et al 2014) but reach as far as the transport of transmembrane proteins to the plasma membrane (Berkovits and Mayr 2015).…”

“…We then clustered all putative sites with sufficient read support, associating lower-ranked sites with higher-ranking sites that were located within at most 12 nt upstream or downstream, as described above. Because in this set of clusters we found cases where the pre-mRNA cleavage site appeared located in an A-rich region upstream of another putative cleavage site, we specifically reviewed clusters in which a putative cleavage site was very close to a poly(A) signal, as these likely reflect internal priming events (Shepard et al 2011;Derti et al 2012;Gruber et al 2014 ′ end processing sites that were within 25 nt of each other were merged into single clusters, the median cluster span was very small, 7 and 3 nt for mouse and human, respectively (Supplemental Fig. 3).…”