UPF1 encodes an ATP-dependent RNA helicase that functions in several mRNA decay pathways, most notably the translation-dependent nonsense mediated mRNA decay (NMD) pathway. Loss-of-function variants genes encoding NMD components UPF2, UPF3B, SMG8 and SMG9 are known to cause intellectual disability (ID). We now identified 24 individuals with ID harboring de novo heterozygous UPF1 missense variants. Variants were predicted to be deleterious to function and were clustered in regions which impact RNA and/or ATP binding.
Previously engineered mutations which disable the UPF1 helicase domain cause increased UPF1 phosphorylation, which we also observed in patient derived cells. While RNA sequencing revealed that 5% of the transcriptome was deregulated in UPF1 variant cells, we were initially surprised to find that NMD was not overtly affected. We investigated an alternative mechanism involving the mRNA binding protein FMR1, which functions as a translational inhibitor and whose loss-of-function causes Fragile X, a leading genetic cause of ID in males. Using immunoprecipitation and proximity-ligation assays, we confirm UPF1 interacts with FMR1 endogenously and in cytoplasmic P-body-like puncta, and that the UPF1-FMR1 interaction was disrupted in UPF1 variant cells. Reciprocally, we show increased UPF1 phosphorylation in Fragile X cells. Remarkably, we discovered the differentially expressed genes (DEGs) identified in UPF1 variant cells significantly overlapped (32%, p=9.8x10-61) and correlated (R=0.95; p=2.2x10-16) with those identified in Fragile X cells. Further investigations revealed downregulated mRNAs had low-GC content, which indicated a possible involvement of defective mRNA processing and storage within P-bodies. Using immunoprecipitation coupled proteomics we subsequently found proteins which were differentially bound to variant UPF1 were 3.2-fold enriched for protein components of P-bodies (p< 7.4x10-07). Finally, we show a >50% reduction in the number of P-bodies in UPF1 variants cells, which were unusually resistant to dissolution using the solvent 1,6-hexanediol.
In summary, we define a novel UPF1-related syndromic ID, with transcriptome impacts shared with Fragile X syndrome, suggesting a convergent pathogenic mechanism involving the co-regulation of mRNAs in P-bodies by UPF1 and FMRP. We speculate that variants in UPF1 which disable its helicase function dissociate its role in NMD from an emerging and unforeseen role in the regulation of mRNA storage and translation in P-bodies.