Full-length NF-κB repressing factor contains an XRN2 binding domain.

NF-κB repressing factor (NKRF) was recently identified as an RNA binding protein that together with its associated proteins, the 5'-3' exonuclease XRN2 and the helicase DHX15, is required to process the precursor ribosomal RNA. XRN2 is a multi-functional ribonuclease that is also involved in processing mRNAs, tRNAs and lncRNAs. The activity and stability of XRN2 are controlled by its binding partners, PAXT-1, CDKN2AIP and CDKN2AIPNL. In each case, these proteins interact with XRN2 via an XRN2 binding domain (XTBD), the structure and mode of action of which is highly conserved. Rather surprisingly, although NKRF interacts directly with XRN2, it was not predicted to contain such a domain, and NKRF's interaction with XRN2 was therefore unexplained. We have identified an alternative upstream AUG start codon within the transcript that encodes NKRF and demonstrate that the full-length form of NKRF contains an XTBD that is conserved across species. Our data suggest that NKRF is tethered in the nucleolus by binding directly to rRNA and that the XTBD in the N-terminal extension of NKRF is essential for the retention of XRN2 in this sub-organelle. Thus, we propose NKRF regulates the early steps of pre-rRNA processing during ribosome biogenesis by controlling the spatial distribution of XRN2 and our data provide further support for the XTBD as an XRN2 interacting motif.

Identification of the RNA polymerase I-RNA interactome.

Ribosome biogenesis is a complex process orchestrated by a host of ribosome assembly factors. Although it is known that many of the proteins involved in this process have RNA binding activity, the full repertoire of proteins that interact with the precursor ribosomal RNA is currently unknown. To gain a greater understanding of the extent to which RNA-protein interactions have the potential to control ribosome biogenesis, we used RNA affinity isolation coupled with proteomics to measure the changes in RNA-protein interactions that occur when rRNA transcription is blocked. Our analysis identified 211 out of 457 nuclear RNA binding proteins with a >3-fold decrease in RNA-protein interaction after inhibition of RNA polymerase I (RNAPI). We have designated these 211 RNA binding proteins as the RNAPI RNA interactome. As expected, the RNAPI RNA interactome is highly enriched for nucleolar proteins and proteins associated with ribosome biogenesis. Selected proteins from the interactome were shown to be nucleolar in location and to have RNA binding activity that was dependent on RNAPI activity. Furthermore, our data show that two proteins, which are required for rRNA maturation, AATF and NGDN, and which form part of the RNA interactome, both lack canonical RNA binding domains and yet are novel pre-rRNA binding proteins.

Elaborate uORF/IRES features control expression and localization of human glycyl-tRNA synthetase.

The canonical activity of glycyl-tRNA synthetase (GARS) is to charge glycine onto its cognate tRNAs. However, outside translation, GARS also participates in many other functions. A single gene encodes both the cytosolic and mitochondrial forms of GARS but 2 mRNA isoforms were identified. Using immunolocalization assays, in vitro translation assays and bicistronic constructs we provide experimental evidence that one of these mRNAs tightly controls expression and localization of human GARS. An intricate regulatory domain was found in its 5'-UTR which displays a functional Internal Ribosome Entry Site and an upstream Open Reading Frame. Together, these elements hinder the synthesis of the mitochondrial GARS and target the translation of the cytosolic enzyme to ER-bound ribosomes. This finding reveals a complex picture of GARS translation and localization in mammals. In this context, we discuss how human GARS expression could influence its moonlighting activities and its involvement in diseases.