Project description:Nonsense-mediated decay (NMD) is an evolutionarily conserved surveillance mechanism that targets mRNAs undergoing premature translation termination for rapid degradation as well as normal physiological transcripts. From yeast to humans, NMD requires the function of three conserved Up-frameshift (Upf) factors (Upf1, Upf2, Upf3). Interestingly, in humans, mutations in UPF2 and UPF3B are associated with intellectual disability and autism spectrum disorder (ASD). However, the neurobiological mechanism by which deficient NMD leads to neurodevelopmental disorders remains unknown. Consequently, no treatment is currently available. Here we report that mice lacking Upf2 in the murine forebrain (Upf2 fb-KO mice) show endophenotypes associated with ASD and deficits in learning and memory. In addition, Upf2-mediated deficient NMD leads to abnormal long-term potentiation (LTP) in the hippocampus. Like patients with mutations in UPF2, Upf2 fb-KO mice showed neuroanatomical abnormalities including a smaller corpus callosum. Surprisingly, transcriptomic analysis revealed elevated mRNA expression of immune-related genes in the hippocampus of Upf2-fb-KO mice, which was accompanied by increased inflammation and progressive infiltration of peripheral immune cells. More importantly, treatment with an FDA-approved immunosuppressant, cyclophosphamide (CyP), significantly reduces brain inflammation, improves the neuroanatomical defects and the deficits in LTP and memory deficits, and reverses some of the ASD-like behaviors, notably the social deficits. Collectively, our results reveal the biological basis underlying neurodevelopmental disorders associated with dysfunctional NMD. Moreover, these findings suggest that immunosuppressant drugs, like CyP, may provide a new therapeutic approach for the treatment of patients with mutations in core NMD components.
Project description:Nonsense-mediated mRNA decay (NMD) functions to degrade transcripts bearing premature stop codon (PTC) and is a crucial regulator of gene expression. NMD and the UPF3B gene have been implicated as the cause of various forms of intellectual disability (ID) and other neurological symptoms. Here, we reports three patients with global developmental delay carrying hemizygous deletions of the UPF2 gene, another important member of the NMD pathway and direct interacting partner of UPF3B. Using RNA-SEQ on lymphoblastoid cells from UPF2 deletion patients, we identified 1009 differently expressed genes (DEGs). 38% of these DEGs overlapped with DEGs identified in UPF3B patients. More importantly, 95% of all DEGs in either UPF2 or UPF3B patients share the same trend of de-regulation. This demonstrates that the transcriptome deregulation in these two patient groups is similar and that UPF2 should be considered as a new candidate gene for ID in man. We expanded our inq`uiries and performed a comprehensive search for copy number variations (CNVs) encompassing all NMD genes in cohorts of ID patients and controls. We found that UPF2, UPF3A, Y14, SMG6 and EIF4A3 are frequently deleted and/or duplicated in ID patients. These CNVs are likely to be the root of the problems or to act as predisposing factors. Our results suggest that dosage imbalance of NMD factors is associated with ID and further emphasize the importance of NMD in normal learning and memory processes.
Project description:All eukaryotes studied to date have the capacity to detect and degrade mRNAs harboring premature translation termination codons (PTCs) in a process called nonsense-mediated mRNA decay (NMD) (reviewed in Wagner E & Lykke-Andersen J, 2002). This surveillance system allows the cell to prevent the expression of potentially harmful truncated proteins. trans-acting factors required for NMD in Drosophila are the proteins UPF1, UPF2, UPF3, SMG-1, SMG-5 and SMG-6. To identify mRNAs naturally regulated by NMD, we analyzed expression profiles in Drosophila cells RNAi-depleted of known NMD factors using high-density oligonucleotide arrays.
Project description:Nonsense-mediated mRNA decay (NMD) is a major translation-dependent RNA degradation pathway required for embryo development and telomere maintenance. Core NMD factors Upf1, Upf2 and Upf3 are conserved from yeast to mammals but a model of the NMD machinery compatible with all eukaryotes is not yet available. We performed the first large-scale quantitative characterization of yeast NMD complexes through affinity purification and mass-spectrometry with 7 different NMD-related factors, with or without Rnase, in strains deleted or not for NMD genes. This extensive characterization of NMD complexes identified two distinct complexes associated with Upf1: Detector (Upf1/2/3) and Effector. Effector contained, in addition to Upf1, the mRNA decapping enzyme and two potential equivalents of mammalian Smg6/5/7: Nmd4 and Ebs1. Like the Smg proteins, Nmd4 and Ebs1 were required for efficient NMD. Our results suggest that the core eukaryotic NMD machinery is conserved across species and operates through successive Upf1-bound Detector and Effector complexes.
Project description:Nonsense-mediated mRNA decay (NMD) is a conserved co-translational mRNA surveillance and turnover pathway across eukaryotes. NMD has a central role in degrading defective mRNAs and also regulates the stability of a significant portion of the transcriptome. The pathway is organized around UPF1, an RNA helicase that can interact with several NMD-specific factors. In human cells, degradation of the targeted mRNAs begins with a cleavage event that requires the recruitment of the SMG6 endonuclease to UPF1. Previous studies have identified functional links between SMG6 and UPF1, but the underlying molecular mechanisms have remained elusive. In this work, we used mass spectrometry, structural biology and biochemical approaches to identify and characterize a conserved short linear motif in SMG6 that interacts with the cysteine/histidine-rich (CH) domain of UPF1. Unexpectedly, we found that the UPF1-SMG6 interaction is precluded when the UPF1 CH domain is engaged with another NMD factor, UPF2. Based on cryo-EM data, we propose that the formation of distinct SMG6-containing and UPF2-containing NMD complexes may be dictated by the RNA-binding status of UPF1. Our findings rationalize a key event in the metazoan NMD quality control pathway and progress our understanding of mechanisms regulating activity and guiding substrate recognition by the SMG6 endonuclease.
Project description:Knocking down Upf2, which encodes a protein that binds to prematurely terminated mRNAs to activate NMD, has been postulated to induce tumor cell expression of neoantigens to promote tumor recognition by T cells. To investigate whether Upf2 knockdown in breast cancer generates novel mRNA isoforms, we performed bulk RNA sequencing (RNA-seq) to compare an EpCAMhi MDA-MB-231 human breast cancer cell line transfected with noncoding control or UPF2 siRNA for 72 h. Results: 222 examples of differential exon usage (DEU) within 281 genes were identified. The number and diversity of DEUs suggest that UPF2 knockdown could have caused novel alternative splicing. To test this idea, UPF2 knockdown-related transcriptional diversity was deconvoluted to identify and estimate the abundance of transcript isoforms. 42 genes with potential differential isoform usage (DIU) were identified. Collectively, our data suggest that reducing NMD activity by UPF2 knockdown may induce expression of tumor neoantigens.
Project description:The RNA helicase UPF1 interacts with mRNAs, mRNA decay machinery, and the terminating ribosome to promote nonsense-mediated mRNA decay (NMD). Structural and biochemical data have revealed that UPF1 exists in an enzymatically autoinhibited “closed” state. Upon binding the NMD protein UPF2, UPF1 undergoes an extensive conformational change into a more enzymatically active “open” state, which exhibits enhanced ATPase and helicase activity. However, mechanically deficient UPF1 mutants can support efficient NMD, bringing into question the roles of UPF1 enzymatic autoinhibition and activation in NMD. Here, we identify two additional important features of the activated open state: slower nucleic acid binding kinetics and enhanced ATP-stimulated nucleic acid dissociation kinetics. Computational modeling based on empirical measurements of UPF1, UPF2, and RNA interaction kinetics predicts that the majority of UPF1-RNA binding and dissociation events in cells occur independently of UPF2 binding. We find that UPF1 mutants with either reduced or accelerated dissociation from RNA have NMD defects, whereas UPF1 mutants that are more dependent on UPF2 for catalytic activity remain active on well-established NMD targets. These findings support a model in which the kinetics of UPF1-mRNA interactions are important determinants of cellular NMD efficiency.