CA repeats in the 3'-untranslated region of bcl-2 mRNA mediate constitutive decay of bcl-2 mRNA

An AU-rich element (ARE) in the 3'-untranslated region (UTR) of bcl-2 mRNA has previously been shown to be responsible for destabilizing bcl-2 mRNA during apoptosis through increasing AUF1 binding. In the present study, we investigated the effect of the region upstream of the ARE on bcl-2 mRNA stability using serial deletion constructs of the 3'-UTR of bcl-2. Deletion of 30 nucleotides mostly consisting of the CA repeats, located upstream of the ARE, resulted in the stabilization of bcl-2 mRNA abundance, in the absence or presence of the ARE. The specificity of the CA repeats in terms of destabilizing bcl-2 mRNA was proven by the substituting the CA repeats with other alternative repeats of purine/pyrimidine, but this had no effect on the stability of bcl-2 mRNA. CA repeats alone, however, failed to confer instability to bcl-2 or gfp reporter mRNAs, indicating a requirement for additional sequences in the upstream region of the 3'-UTR. Serial deletion and replacement of a part of the region upstream of the CA repeats revealed that the entire 131-nucleotide upstream region is an essential prerequisite for the CA repeat-dependent destabilization of bcl-2 mRNA. Unlike the ARE, CA repeat-mediated degradation of bcl-2 mRNA was not accelerated upon apoptotic stimulus. Moreover, the upstream sequences and CA repeats are conserved among mammals. Collectively, CA repeats contribute to the constitutive decay of bcl-2 mRNA in the steady states, thereby maintaining appropriate bcl-2 levels in mammalian cells.     

Crystal Structures and RNA-binding Properties of the RNA Recognition Motifs of Heterogeneous Nuclear Ribonucleoprotein L: INSIGHTS INTO ITS ROLES IN ALTERNATIVE SPLICING REGULATION

Heterogeneous nuclear ribonucleoprotein L (hnRNP L) is an abundant RNA-binding protein implicated in many bioprocesses, including pre-mRNA processing, mRNA export of intronless genes, internal ribosomal entry site-mediated translation, and chromatin modification. It contains four RNA recognition motifs (RRMs) that bind with CA repeats or CA-rich elements. In this study, surface plasmon resonance spectroscopy assays revealed that all four RRM domains contribute to RNA binding. Furthermore, we elucidated the crystal structures of hnRNP L RRM1 and RRM34 at 2.0 and 1.8 Å, respectively. These RRMs all adopt the typical β1α1β2β3α2β4 topology, except for an unusual fifth β-strand in RRM3. RRM3 and RRM4 interact intimately with each other mainly through helical surfaces, leading the two β-sheets to face opposite directions. Structure-based mutations and surface plasmon resonance assay results suggested that the β-sheets of RRM1 and RRM34 are accessible for RNA binding. FRET-based gel shift assays (FRET-EMSA) and steady-state FRET assays, together with cross-linking and dynamic light scattering assays, demonstrated that hnRNP L RRM34 facilitates RNA looping when binding to two appropriately separated binding sites within the same target pre-mRNA. EMSA and isothermal titration calorimetry binding studies with in vivo target RNA suggested that hnRNP L-mediated RNA looping may occur in vivo. Our study provides a mechanistic explanation for the dual functions of hnRNP L in alternative splicing regulation as an activator or repressor.     

Novel functional role of CA repeats and hnRNP L in RNA stability

CA dinucleotide repeat sequences are very common in the human genome. We have recently demonstrated that the polymorphic CA repeats in intron 13 of the human endothelial nitric oxide synthase (eNOS) gene function as an unusual, length-dependent splicing enhancer. The CA repeat enhancer requires for its activity specific binding of hnRNP L. Here we show that in the absence of bound hnRNP L, the pre-mRNA is cleaved directly upstream of the CA repeats. The addition of recombinant hnRNP L restores RNA stability. CA repeats are both necessary and sufficient for this specific cleavage in the 5' adjacent RNA sequence. We conclude that-in addition to its role as a splicing activator-hnRNP L can act in vitro as a sequence-specific RNA protection factor. Based on the wide abundance of CA repetitive sequences in the human genome, this may represent a novel, generally important role of this abundant hnRNP protein.     

PTB Binds to the 3’ Untranslated Region of the Human Astrovirus Type 8: A Possible Role in Viral Replication

The 3′ untranslated region (3′UTR) of human astroviruses (HAstV) consists of two hairpin structures (helix I and II) joined by a linker harboring a conserved PTB/hnRNP1 binding site. The identification and characterization of cellular proteins that interact with the 3′UTR of HAstV-8 virus will help to uncover cellular requirements for viral functions. To this end, mobility shift assays and UV cross-linking were performed with uninfected and HAstV-8-infected cell extracts and HAstV-8 3′UTR probes. Two RNA-protein complexes (CI and CII) were recruited into the 3′UTR. Complex CII formation was compromised with cold homologous RNA, and seven proteins of 35, 40, 45, 50, 52, 57/60 and 75 kDa were cross-linked to the 3′UTR. Supermobility shift assays indicated that PTB/hnRNP1 is part of this complex, and 3′UTR-crosslinked PTB/hnRNP1 was immunoprecipitated from HAstV-8 infected cell-membrane extracts. Also, immunofluorescence analyses revealed that PTB/hnRNP1 is distributed in the nucleus and cytoplasm of uninfected cells, but it is mainly localized perinuclearly in the cytoplasm of HAstV-8 infected cells. Furthermore, the minimal 3′UTR sequences recognized by recombinant PTB are those conforming helix I, and an intact PTB/hnRNP1-binding site. Finally, small interfering RNA-mediated PTB/hnRNP1 silencing reduced synthesis viral genome and virus yield in CaCo2 cells, suggesting that PTB/hnRNP1 is required for HAstV replication. In conclusion, PTB/hnRNP1 binds to the 3′UTR HAstV-8 and is required or participates in viral replication.     

Interplay between hnRNP A1 and a cis-acting element in the 3' UTR of CYP2A5 mRNA is central for high expression of the gene

Our previous evidence suggests that heterogeneous nuclear ribonucleoprotein (hnRNP) A1 plays a part in the regulation of the Cyp2a5 gene by interacting with the 3' untranslated region (UTR) of the CYP2A5 mRNA. However, the exact role of this interaction is not clear. The aim of the present work was to gain further insight into the regulation process of Cyp2a5. For this purpose the 3' UTR of CYP2A5 was fused to the coding region of luciferase mRNA. Luciferase recombinants containing either the full length 3' UTR, or the 3' UTR lacking a previously described 71 nucleotide (nt) region (the hnRNP A1 primary binding site), were transiently expressed in cells expressing or lacking hnRNP A1. The expression of the luciferase recombinants was examined both at mRNA and enzyme activity levels. The results disclosed that the presence of hnRNP A1 was required for the high expression of the recombinant carrying the full length 3' UTR of CYP2A5. Deletion of the hnRNP A1 primary binding site dramatically modified the expression pattern: the mRNA levels and luciferase activities of the deletion mutant were independent from hnRNP A1. These results conclusively demonstrate that the 71 nt region in the 3' UTR of CYP2A5 mRNA can confer hnRNP A1-dependent regulation to a gene. In addition, comparison of RNA levels and luciferase activities suggested that regions flanking the hnRNP A1 binding site could regulate translation of the CYP2A5 mRNA. These results are consistent with a model in which the binding of hnRNP A1 to the 71 nt putative hairpin-loop region in the CYP2A5 mRNA 3' UTR upregulates mRNA levels possibly by protecting the mRNA from degradation.     

HnRNP L stimulates splicing of the eNOS gene by binding to variable-length CA repeats

In the human genome, dinucleotide repeats are common sequence elements of unknown functional significance. Here we demonstrate that CA repeats in intron 13 of the human endothelial nitric oxide synthase (eNOS) gene function as an unusual intronic splicing enhancer, whose activity depends on the CA repeat number. We identify the 65 kDa heterogenous nuclear ribonucleoprotein (hnRNP) L as the major factor that binds specifically and in a length-dependent manner to the CA-repeat enhancer. In addition, we show that hnRNP L functions as a specific activator of eNOS splicing, providing the first evidence for a role of hnRNP L in the regulation of mRNA splicing. We hypothesize that hnRNP L may be involved in the regulation of many other genes containing CA repeats or A/C-rich enhancers.     

The Bin3 RNA methyltransferase is required for repression of caudal translation in the Drosophila embryo

Bin3 was first identified as a Bicoid-interacting protein in a yeast two-hybrid screen. In human cells, a Bin3 ortholog (BCDIN3) methylates the 5′ end of 7SK RNA, but its role in vivo is unknown. Here, we show that in Drosophila, Bin3 is important for dorso-ventral patterning in oogenesis and for anterior–posterior pattern formation during embryogenesis. Embryos that lack Bin3 fail to repress the translation of caudal mRNA and exhibit head involution defects. bin3 mutants also show (1) a severe reduction in the level of 7SK RNA, (2) reduced binding of Bicoid to the caudal 3′ UTR, and (3) genetic interactions with bicoid, and with genes encoding eIF4E, Larp1, polyA binding protein (PABP), and Ago2. 7SK RNA coimmunoprecipitated with Bin3 and is present in Bicoid complexes. These data suggest a model in which Bicoid recruits Bin3 to the caudal 3′ UTR. Bin3's role is to bind and stabilize 7SK RNA, thereby promoting formation of a repressive RNA–protein complex that includes the RNA-binding proteins Larp1, PABP, and Ago2. This complex would prevent translation by blocking eIF4E interactions required for initiation. Our results, together with prior network analysis in human cells, suggest that Bin3 interacts with multiple partner proteins, methylates small non-coding RNAs, and plays diverse roles in development.     

Paralogs hnRNP L and hnRNP LL Exhibit Overlapping but Distinct RNA Binding Constraints

HnRNP (heterogeneous nuclear ribonucleoprotein) proteins are a large family of RNA-binding proteins that regulate numerous aspects of RNA processing. Interestingly, several paralogous pairs of hnRNPs exist that exhibit similar RNA-binding specificity to one another, yet have non-redundant functional targets in vivo. In this study we systematically investigate the possibility that the paralogs hnRNP L and hnRNP LL have distinct RNA binding determinants that may underlie their lack of functional redundancy. Using a combination of RNAcompete and native gel analysis we find that while both hnRNP L and hnRNP LL preferentially bind sequences that contain repeated CA dinucleotides, these proteins differ in their requirement for the spacing of the CAs. Specifically, hnRNP LL has a more stringent requirement for a two nucleotide space between CA repeats than does hnRNP L, resulting in hnRNP L binding more promiscuously than does hnRNP LL. Importantly, this differential requirement for the spacing of CA dinucleotides explains the previously observed differences in the sensitivity of hnRNP L and LL to mutations within the CD45 gene. We suggest that overlapping but divergent RNA-binding preferences, as we show here for hnRNP L and hnRNP LL, may be commonplace among other hnRNP paralogs.     

A 5′UTR-Spliced mRNA Isoform Is Specialized for Enhanced HIV-2 gag Translation

Full-length unspliced genomic RNA plays critical roles in HIV replication, serving both as mRNA for the synthesis of the key viral polyproteins Gag and Gag-Pol and as genomic RNA for encapsidation into assembling viral particles. We show that a second gag mRNA species that differs from the genomic RNA molecule by the absence of an intron in the 5′ untranslated region (5′UTR) is produced during HIV-2 replication in cell culture and in infected patients. We developed a cotransfection system in which epitopically tagged Gag proteins can be traced back to their mRNA origins in the translation pool. We show that a disproportionate amount of Gag is translated from 5′UTR intron-spliced mRNAs, demonstrating a role for the 5′UTR intron in the regulation of gag translation. To further characterize the effects of the HIV-2 5′UTR on translation, we fused wild-type, spliced, or mutant leader RNA constructs to a luciferase reporter gene and assayed their translation in reticulocyte lysates. These assays confirmed that leaders lacking the 5′UTR intron increased translational efficiency compared to that of the unspliced leader. In addition, we found that removal or mutagenesis of the C-box, a pyrimidine-rich sequence located in the 5′UTR intron and previously shown to affect RNA dimerization, also strongly influenced translational efficiency. These results suggest that the splicing of both the 5′UTR intron and the C-box element have key roles in regulation of HIV-2 gag translation in vitro and in vivo.     

Conservative sequences in 3'UTR of TCRzeta mRNA regulate TCRzeta in SLE T cells

We have demonstrated that T-cell receptor ζ (ζ) mRNA with a 562-bp deleted alternatively spliced 3′-untranslated region (3′UTR) observed in T cells of patients with systemic lupus erythematosus (SLE) can lead to a reduction in ζ and TCR/CD3 (J. Immunol., 2003 & 2005). To determine the region in ζ mRNA 3′UTR for the regulation of ζ, ζ mRNA with 3′UTR truncations ligated into pDON-AI was used to infect murine T-cell hybridoma MA5.8 cells, which do not contain ζ. As a Western blot analysis demonstrated the importance of the regions from +871 to +950, containing conservative sequence 1 (CS1), and +1070 to +1136, containing CS2, for the production of ζ, we constructed MA5.8 mutants carrying ζ mRNA 3′UTR with deletions of these regions (ΔCS1 and ΔCS2 mutants). Western blot and FACS analyses showed significant reduction in the cell surface ζ and TCR/CD3 in both these mutants, and IL-2 production was decreased, compared with MA5.8 cells transfected with wild-type ζ mRNA. Furthermore, real-time PCR demonstrated the instability of ζ mRNA with 3′UTR deletions in these MA5.8 mutants. In conclusion, CS1 and CS2 may be responsible for the regulation of ζ and TCR/CD3 through the stability of ζ mRNA in SLE T cells.     

JKTBP1 is involved in stabilization and IRES-dependent translation of NRF mRNAs by binding to 5' and 3' untranslated regions

Heterogeneous nuclear ribonucleoprotein D-like protein (JKTBP) 1 was implicated in cap-independent translation by binding to the internal ribosome entry site in the 5′ untranslated region (UTR) of NF-κB-repressing factor (NRF). Two different NRF mRNAs have been identified so far, both sharing the common 5′ internal ribosome entry site but having different length of 3′ UTRs. Here, we used a series of DNA and RNA luciferase reporter constructs comprising 5′, 3′ or both NRF UTRs to study the effect of JKTBP1 on translation of NRF mRNA variants. The results indicate that JKTBP1 regulates the level of NRF protein expression by binding to both NRF 5′ and 3′ UTRs. Using successive deletion and point mutations as well as RNA binding studies, we define two distinct JKTBP1 binding elements in NRF 5′ and 3′ UTRs. Furthermore, JKTBP1 requires two distinct RNA binding domains to interact with NRF UTRs and a short C-terminal region for its effect on NRF expression. Together, our study shows that JKTBP1 contributes to NRF protein expression via two disparate mechanisms: mRNA stabilization and cap-independent translation. By binding to 5′ UTR, JKTBP1 increases the internal translation initiation in both NRF mRNA variants, whereas its binding to 3′ UTR elevated primarily the stability of the major NRF mRNA. Thus, JKTBP1 is a key regulatory factor linking two pivotal control mechanisms of NRF gene expression: the cap-independent translation initiation and mRNA stabilization.     

The HumanHNRPDLocus Maps to 4q21 and Encodes a Highly Conserved Protein

The hnRNP D protein interacts with nucleic acids bothin vivoandin vitro.Like many other proteins that interact with RNA, it contains RBD (or “RRM”) domains and arg-gly-gly (RGG) motifs. We have examined the organization and localization of the human and murine genes that encode the hnRNP D protein. Comparison of the predicted sequences of the hnRNP D proteins in human and mouse shows that they are 96.9% identical (98.9% similar). This very high level of conservation suggests a critical function for hnRNP D. Sequence analysis of the humanHNRPDgene shows that the protein is encoded by eight exons and that two additional exons specify sequences in the 3′ UTR. Use of two of the coding exons is determined by alternative splicing of theHNRPDmRNA. The humanHNRPDgene maps to 4q21. The mouseHnrpdgene maps to the F region of chromosome 3, which is syntenic with the human 4q21 region.