Mechanisms of nonsense-mediated mRNA decay target selection

National Heart, Lung, And Blood Institute

Investigators

Linked publications & trials

Abstract

The nonsense-mediated mRNA decay pathway recognizes and degrades transcripts containing premature translation termination codons, as well as 5-10% of apparently normal cellular mRNAs. Despite intensive interest in the mechanisms of NMD, how the pathway selects its targets remains poorly understood. To gain insight into the process and physiological consequences of NMD target selection, we have pursued biochemical and functional screening approaches to identify cellular factors that promote or inhibit decay. 1. Identification and characterization of RNA-binding proteins that inhibit decay of potential NMD targets A longstanding question in the field is why some transcripts with long 3 untranslated regions (3UTRs) are selected for NMD, while others are immune to the pathway. We reasoned that this could be due to widespread protection of potential NMD targets by RNA-binding proteins. To identify such proteins, we used the RNA Stability Element (RSE) from Rous sarcoma virus, previously identified by the laboratory of Karen Beemon as a strong inhibitor of NMD. Using RNA-based affinity purification, we isolated endogenously assembled complexes containing mRNAs with and without RSE sequences and identified PTBP1 as the major RSE-binding factor (Ge et al., eLIFE, 2016). Together with the Beemon lab, we found that PTBP1 protects RSV and cellular mRNAs with long 3UTRs from NMD by limiting the association of the core NMD factor UPF1 with potential decay targets. In recent work, we have found that the balance of pro- and anti-NMD factors regulates the expression of many mRNAs produced through alternative polyadenylation (APA), including mRNAs previously identified as frequent APA targets in cancer (Kishor et al., NAR, 2020). In the course of our work on PTBP1, we discovered that a related protein, hnRNP L, was also capable of inhibiting decay of specific transcripts (Kishor et al., EMBOJ, 2019). By depleting each protein from human cells alone or together with UPF1, we have developed evidence that PTBP1 and hnRNP L combine to protect hundreds to thousands of human mRNAs from NMD. Among the mRNAs protected by hnRNP L, we found that mRNAs encoding the important antiapoptotic protein BCL2 were highly sensitive to NMD in the absence of the protective mechanism. Remarkably, we further demonstrated that aberrant BCL2:IgH fusion mRNAs that are frequent drivers of B cell lymphoma are protected from NMD by hnRNP L. These studies together show how the transcriptome has evolved to shield genuine long 3UTRs from detection by NMD and illustrate serious potential deleterious consequences of this mechanism in human disease. We are currently working to understand how PTBP1 and hnRNP L prevent UPF1 from associating with potential targets. Traditional models for RBP-mediated mRNA stabilization involve competition for binding sites with decay factors, but this model is unlikely to account for the ability of PTBP1 and hnRNP L to antagonize the sequence-independent, high-affinity RNA binding of UPF1 (Fritz et al., JBC, 2020). To address this question, we established a real-time in vitro assay of UPF1 translocation and unwinding activity and used it to show that PTBP1 can strongly repress UPF1 helicase activity. Surprisingly, PTBP1 does not do this by acting as a physical roadblock to UPF1 translocation, but instead engages in direct interactions with UPF1 to promote its dissociation from RNA or DNA. Sensitivity to PTBP1 is determined by a regulatory loop in UPF1 previously shown to reduce the ability of UPF1 to bind RNA in the presence of ATP. To further investigate the ability of PTBP1 to displace UPF1 from authentic mRNPs, we established a novel assay to observe UPF1 dissociation from immobilized affinity purified mRNPs. In this assay, recruitment of PTBP1 to mRNPs caused enhanced dissociation of UPF1 from endogenously assembled mRNPs, in a manner dependent on ATP hydrolysis by UPF1. We have further developed the biochemical and biophysical methods used to investigate the mechanism of protection by PTBP1 to investigate the fundamental question of the role of UPF1 ATPase and helicase activity in NMD. We have established assays to evaluate several ATPase-dependent biochemical activities, including translocation, unwinding, dissociation, and protein displacement. By combining these assays with cellular studies, we have found that UPF1 mutants with severe mechanical defects are able to conduct efficient NMD. Conversely, UPF1 mutants that do not undergo efficient ATPase-stimulated dissociation from nucleic acids exhibit impaired NMD activity. These data point to a model in which UPF1 need not act as an effective helicase to conduct NMD. 2. Identification of novel NMD factors through whole-genome RNAi screening In collaboration with researchers in NCATS, we have used high-throughput RNAi screening data to identify novel components of the NMD pathway. With this approach, we have found that ICE1, previously characterized as a protein involved in transcription of small nuclear RNAs, is required to efficiently link the splicing machinery to the NMD pathway (Baird et al., eLIFE, 2018). In the absence of ICE1, transcripts undergoing NMD stimulated by the exon junction complex (EJC) are stabilized, due to a reduced association between the NMD protein UPF3B and the EJC. This function is mediated by a putative MIF4G domain in ICE1, which we predict interacts directly with the EJC component eiF4A3. We are currently investigating the mechanism by which ICE affects EJC-UPF3B interactions, as well as how these proteins may be used to accomplish cellular gene expression regulation.