A massively parallel 3' UTR reporter assay reveals relationships between nucleotide content, sequence conservation, and mRNA destabilization.

Compared to coding sequences, untranslated regions of the transcriptome are not well conserved, and functional annotation of these sequences is challenging. Global relationships between nucleotide composition of 3' UTR sequences and their sequence conservation have been appreciated since mammalian genomes were first sequenced, but the functional relevance of these patterns remain unknown. We systematically measured the effect on gene expression of the sequences of more than 25,000 RNA-binding protein (RBP) binding sites in primary mouse T cells using a massively parallel reporter assay. GC-rich sequences were destabilizing of reporter mRNAs and come from more rapidly evolving regions of the genome. These sequences were more likely to be folded in vivo and contain a number of structural motifs that reduced accumulation of a heterologous reporter protein. Comparison of full-length 3' UTR sequences across vertebrate phylogeny revealed that strictly conserved 3' UTRs were GC-poor and enriched in genes associated with organismal development. In contrast, rapidly evolving 3' UTRs tended to be GC-rich and derived from genes involved in metabolism and immune responses. Cell-essential genes had lower GC content in their 3' UTRs, suggesting a connection between unstructured mRNA noncoding sequences and optimal protein production. By reducing gene expression, GC-rich RBP-occupied sequences act as a rapidly evolving substrate for gene regulatory interactions.

Massively parallel analysis of human 3' UTRs reveals that AU-rich element length and registration predict mRNA destabilization.

AU-rich elements (AREs) are 3' UTR cis-regulatory elements that regulate the stability of mRNAs. Consensus ARE motifs have been determined, but little is known about how differences in 3' UTR sequences that conform to these motifs affect their function. Here, we use functional annotation of sequences from 3' UTRs (fast-UTR), a massively parallel reporter assay (MPRA), to investigate the effects of 41,288 3' UTR sequence fragments from 4653 transcripts on gene expression and mRNA stability in Jurkat and Beas2B cells. Our analyses demonstrate that the length of an ARE and its registration (the first and last nucleotides of the repeating ARE motif) have significant effects on gene expression and stability. Based on this finding, we propose improved ARE classification and concomitant methods to categorize and predict the effect of AREs on gene expression and stability. Finally, to investigate the advantages of our general experimental design we examine other motifs including constitutive decay elements (CDEs), where we show that the length of the CDE stem-loop has a significant impact on steady-state expression and mRNA stability. We conclude that fast-UTR, in conjunction with our analytical approach, can produce improved yet simple sequence-based rules for predicting the activity of human 3' UTRs.

Ribosome changes reprogram translation for chemosurvival in G0 leukemic cells.

Quiescent leukemic cells survive chemotherapy, with translation changes. Our data reveal that FXR1, a protein amplified in several aggressive cancers, is elevated in quiescent and chemo-treated leukemic cells and promotes chemosurvival. This suggests undiscovered roles for this RNA- and ribosome-associated protein in chemosurvival. We find that FXR1 depletion reduces translation, with altered rRNAs, snoRNAs, and ribosomal proteins (RPs). FXR1 regulates factors that promote transcription and processing of ribosomal genes and snoRNAs. Ribosome changes in FXR1-overexpressing cells, including RPLP0/uL10 levels, activate eIF2α kinases. Accordingly, phospho-eIF2α increases, enabling selective translation of survival and immune regulators in FXR1-overexpressing cells. Overriding these genes or phospho-eIF2α with inhibitors reduces chemosurvival. Thus, elevated FXR1 in quiescent or chemo-treated leukemic cells alters ribosomes that trigger stress signals to redirect translation for chemosurvival.

Epithelial miR-141 regulates IL-13-induced airway mucus production.

IL-13-induced goblet cell metaplasia contributes to airway remodeling and pathological mucus hypersecretion in asthma. miRNAs are potent modulators of cellular responses, but their role in mucus regulation is largely unexplored. We hypothesized that airway epithelial miRNAs play roles in IL-13-induced mucus regulation. miR-141 is highly expressed in human and mouse airway epithelium, is altered in bronchial brushings from asthmatic subjects at baseline, and is induced shortly after airway allergen exposure. We established a CRISPR/Cas9-based protocol to target miR-141 in primary human bronchial epithelial cells that were differentiated at air-liquid-interface, and goblet cell hyperplasia was induced by IL-13 stimulation. miR-141 disruption resulted in decreased goblet cell frequency, intracellular MUC5AC, and total secreted mucus. These effects correlated with a reduction in a goblet cell gene expression signature and enrichment of a basal cell gene expression signature defined by single cell RNA sequencing. Furthermore, intranasal administration of a sequence-specific mmu-miR-141-3p inhibitor in mice decreased Aspergillus-induced secreted mucus and mucus-producing cells in the lung and reduced airway hyperresponsiveness without affecting cellular inflammation. In conclusion, we have identified a miRNA that regulates pathological airway mucus production and is amenable to therapeutic manipulation through an inhaled route.

Chromosome wide analysis of CUGBP1 binding sites identifies the tetraspanin CD9 mRNA as a target for CUGBP1-mediated down-regulation.

CUGBP1 is an RNA-binding protein controlling alternative splicing, mRNA translation and stability. In this work we used a motif scoring approach to identify putative CUGBP1 binding sites for genes located on the human chromosome 12. This allowed us to identify the gene CD9 as a presumptive target for CUGBP1-mediated regulation. In a number of cancers, the tetraspanin CD9 is down-regulated, an event correlated with a bad prognostic. Using a combination of biochemical approaches and CUGBP1 knockdown, we showed that CUGBP1 directly controls CD9 expression.

Identification of CUG-BP1/EDEN-BP target mRNAs in Xenopus tropicalis.

The early development of many animals relies on the posttranscriptional regulations of maternally stored mRNAs. In particular, the translation of maternal mRNAs is tightly controlled during oocyte maturation and early mitotic cycles in Xenopus. The Embryonic Deadenylation ElemeNt (EDEN) and its associated protein EDEN-BP are known to trigger deadenylation and translational silencing to several mRNAs bearing an EDEN. This Xenopus RNA-binding protein is an ortholog of the human protein CUG-BP1/CELF1. Five mRNAs, encoding cell cycle regulators and a protein involved in the notch pathway, have been identified as being deadenylated by EDEN/EDEN-BP. To identify new EDEN-BP targets, we immunoprecipitated EDEN-BP/mRNA complexes from Xenopus tropicalis egg extracts. We identified 153 mRNAs as new binding targets for EDEN-BP using microarrays. Sequence analyses of the 3' untranslated regions of the newly identified EDEN-BP targets reveal an enrichment in putative EDEN sequences. EDEN-BP binding to a subset of the targets was confirmed both in vitro and in vivo. Among the newly identified targets, Cdk1, a key player of oocyte maturation and cell cycle progression, is specifically targeted by its 3' UTR for an EDEN-BP-dependent deadenylation after fertilization.

Identification of CELF1 RNA targets by CLIP-seq in human HeLa cells.

The specific interactions between RNA-binding proteins and their target RNAs are an essential level to control gene expression. By combining ultra-violet cross-linking and immunoprecipitation (CLIP) and massive SoliD sequencing we identified the RNAs bound by the RNA-binding protein CELF1, in human HeLa cells. The CELF1 binding sites deduced from the sequence data allow characterizing specific features of CELF1-RNA association. We present therefore the first map of CELF1 binding sites in human cells.

CLIP-seq of eIF4AIII reveals transcriptome-wide mapping of the human exon junction complex.

The exon junction complex (EJC) is a central effector of the fate of mRNAs, linking nuclear processing to mRNA transport, translation and surveillance. However, little is known about its transcriptome-wide targets. We used cross-linking and immunoprecipitation methods coupled to high-throughput sequencing (CLIP-seq) in human cells to identify the binding sites of the DEAD-box helicase eIF4AIII, an EJC core component. CLIP reads form peaks that are located mainly in spliced mRNAs. Most expressed exons harbor peaks either in the canonical EJC region, located ~24 nucleotides upstream of exonic junctions, or in other noncanonical regions. Notably, both of these types of peaks are preferentially associated with unstructured and purine-rich sequences containing the motif GAAGA, which is a potential binding site for EJC-associated factors. Therefore, EJC positions vary spatially and quantitatively between exons. This transcriptome-wide mapping of human eIF4AIII reveals unanticipated aspects of the EJC and broadens its potential impact on post-transcriptional regulation.