Perinuclear mRNA localisation by vimentin 3'-untranslated region requires a 100 nucleotide sequence and intermediate filaments

eEF1Bγ binds the Che-1 and TP53 gene promoters and their transcripts

Background

We have previously shown that the eukaryotic elongation factor subunit 1B gamma (eEF1Bγ) interacts with the RNA polymerase II (pol II) alpha-like subunit “C” (POLR2C), alone or complexed, in the pol II enzyme. Moreover, we demonstrated that eEF1Bγ binds the promoter region and the 3’ UTR mRNA of the vimentin gene. These events contribute to localize the vimentin transcript and consequentially its translation, promoting a proper mitochondrial network.

Methods

With the intent of identifying additional transcripts that complex with the eEF1Bγ protein, we performed a series of ribonucleoprotein immunoprecipitation (RIP) assays using a mitochondria-enriched heavy membrane (HM) fraction.

Results

Among the eEF1Bγ complexed transcripts, we found the mRNA encoding the Che-1/AATF multifunctional protein. As reported by other research groups, we found the tumor suppressor p53 transcript complexed with the eEF1Bγ protein. Here, we show for the first time that eEF1Bγ binds not only Che-1 and p53 transcripts but also their promoters. Remarkably, we demonstrate that both the Che-1 transcript and its translated product localize also to the mitochondria and that eEF1Bγ depletion strongly perturbs the mitochondrial network and the correct localization of Che-1. In a doxorubicin (Dox)-induced DNA damage assay we show that eEF1Bγ depletion significantly decreases p53 protein accumulation and slightly impacts on Che-1 accumulation. Importantly, Che-1 and p53 proteins are components of the DNA damage response machinery that maintains genome integrity and prevents tumorigenesis.

Conclusions

Our data support the notion that eEF1Bγ, besides its canonical role in translation, is an RNA-binding protein and a key player in cellular stress responses. We suggest for eEF1Bγ a role as primordial transcription/translation factor that links fundamental steps from transcription control to local translation.

Electronic supplementary material

The online version of this article (doi:10.1186/s13046-016-0424-x) contains supplementary material, which is available to authorized users.

Protein targeting via mRNA in bacteria

Proteins of all living organisms must reach their subcellular destination to sustain the cell structure and function. The proteins are transported to one of the cellular compartments, inserted into the membrane, or secreted across the membrane to the extracellular milieu. Cells have developed various mechanisms to transport proteins across membranes, among them localized translation. Evidence for targeting of Messenger RNA for the sake of translation of their respective protein products at specific subcellular sites in many eukaryotic model organisms have been accumulating in recent years. Cis-acting RNA localizing elements, termed RNA zip-codes, which are embedded within the mRNA sequence, are recognized by RNA-binding proteins, which in turn interact with motor proteins, thus coordinating the intracellular transport of the mRNA transcripts. Despite the rareness of conventional organelles, first and foremost a nucleus, pieces of evidence for mRNA localization to specific subcellular domains, where their protein products function, have also been obtained for prokaryotes. Although the underlying mechanisms for transcript localization in bacteria are yet to be unraveled, it is now obvious that intracellular localization of mRNA is a common mechanism to spatially localize proteins in both eukaryotes and prokaryotes. This article is part of a Special Issue entitled: Protein trafficking and secretion in bacteria. Guest Editors: Anastassios Economou and Ross Dalbey.

In silico estimation of translation efficiency in human cell lines: potential evidence for widespread translational control

Recently large scale transcriptome and proteome datasets for human cells have become available. A striking finding from these studies is that the level of an mRNA typically predicts no more than 40% of the abundance of protein. This correlation represents the overall figure for all genes. We present here a bioinformatic analysis of translation efficiency – the rate at which mRNA is translated into protein. We have analysed those human datasets that include genome wide mRNA and protein levels determined in the same study. The analysis comprises five distinct human cell lines that together provide comparable data for 8,170 genes. For each gene we have used levels of mRNA and protein combined with protein stability data from the HeLa cell line to estimate translation efficiency. This was possible for 3,990 genes in one or more cell lines and 1,807 genes in all five cell lines. Interestingly, our analysis and modelling shows that for many genes this estimated translation efficiency has considerable consistency between cell lines. Some deviations from this consistency likely result from the regulation of protein degradation. Others are likely due to known translational control mechanisms. These findings suggest it will be possible to build improved models for the interpretation of mRNA expression data. The results we present here provide a view of translation efficiency for many genes. We provide an online resource allowing the exploration of translation efficiency in genes of interest within different cell lines (http://bioanalysis.otago.ac.nz/TranslationEfficiency).

How old are RNA networks?

Some major classes of RNAs (such as mRNA, rRNA, tRNA and RNase P) are ubiquitous in all living systems so are inferred to have arisen early during the origin of life. However, the situation is not so clear for the system of RNA regulatory networks that continue to be uncovered, especially in eukaryotes. It is increasingly being recognised that networks of small RNAs are important for regulation in all cells, but it is not certain whether the origin of these networks are as old as rRNAs and tRNA. Another group of ncRNAs, including snoRNAs, occurs mainly in archaea and eukaryotes and their ultimate origin is less certain, although perhaps the simplest hypothesis is that they were present in earlier stages of life and were lost from bacteria. Some RNA networks may trace back to an early stage when there was just RNA and proteins, the RNP-world; before DNA.

The eEF1γ subunit contacts RNA polymerase II and binds vimentin promoter region

Here, we show that the eukaryotic translation elongation factor 1 gamma (eEF1γ) physically interacts with the RNA polymerase II (pol II) core subunit 3 (RPB3), both in isolation and in the context of the holo-enzyme. Importantly, eEF1γ has been recently shown to bind Vimentin mRNA. By chromatin immunoprecipitation experiments, we demonstrate, for the first time, that eEF1γ is also physically present on the genomic locus corresponding to the promoter region of human Vimentin gene. The eEF1γ depletion causes the Vimentin protein to be incorrectly compartmentalised and to severely compromise cellular shape and mitochondria localisation. We demonstrate that eEF1γ partially colocalises with the mitochondrial marker Tom20 and that eEF1γ depletion increases mitochondrial superoxide generation as well as the total levels of carbonylated proteins. Finally, we hypothesise that eEF1γ, in addition to its role in translation elongation complex, is involved in regulating Vimentin gene by contacting both pol II and the Vimentin promoter region and then shuttling/nursing the Vimentin mRNA from its gene locus to its appropriate cellular compartment for translation.

MicroRNA-17-3p is a prostate tumor suppressor in vitro and in vivo, and is decreased in high grade prostate tumors analyzed by laser capture microdissection

MicroRNAs (miRs) are a novel class of RNAs with important roles in regulating gene expression. To identify miRs controlling prostate tumor progression, we utilized unique human prostate sublines derived from the parental P69 cell line, which differ in their tumorigenic properties in vivo. Grown embedded in laminin-rich extracellular matrix (lrECM) gels these genetically-related sublines displayed drastically different morphologies correlating with their behaviour in vivo. The non-tumorigenic P69 subline grew as multicellular acini with a defined lumen and basal/polar expression of relevant marker proteins. M12, a highly tumorigenic, metastatic derivative, grew as a disorganized mass of cells with no polarization, whereas the F6 subline, a weakly tumorigenic, non-metastatic M12 variant, reverted to acini formation akin to the P69 cell line. These sublines also differed in expression of vimentin, which was high in M12, but low in F6 and P69 sublines. Analysis of vimentin's conserved 3'-UTR suggested several miRs that could regulate vimentin expression. The lack of miR-17-3p expression correlated with an increase in vimentin synthesis and tumorigenicity. Stable expression of miR-17-3p in the M12 subline reduced vimentin levels 85% and reverted growth to organized, polarized acini in lrECM gels. In vitro motility and invasion assays suggested a decrease in tumorigenic behaviour, confirmed by reduced tumor growth in male athymic, nude mice dependent on miR-17-3p expression. Analysis of LCM-purified clinical human prostatectomy specimens confirmed that miR-17-3p levels were reduced in tumor cells. These results suggest that miR-17-3p functions as a tumor suppressor, representing a novel target to block prostate tumor progression.

The motility and dynamic properties of intermediate filaments and their constituent proteins

Intermediate filament (IF) proteins exist in multiple structural forms within cells including mature IF, short filaments or 'squiggles', and non-filamentous precursors called particles. These forms are interconvertible and their relative abundance is IF type, cell type- and cell cycle stage-dependent. These structures are often associated with molecular motors, such as kinesin and dynein, and are therefore capable of translocating through the cytoplasm along microtubules. The assembly of mature IF from their precursor particles is also coupled to translation. These dynamic properties of IF provide mechanisms for regulating their reorganization and assembly in response to the functional requirements of cells. The recent findings that IF and their precursors are frequently associated with signaling molecules have revealed new functions for IF beyond their more traditional roles as mechanical integrators of cells and tissues.

Annexin A2 recognises a specific region in the 3'-UTR of its cognate messenger RNA

Annexin A2 is a multifunctional Ca(2+)- and lipid-binding protein. We previously showed that a distinct pool of cellular Annexin A2 associates with mRNP complexes or polysomes associated with the cytoskeleton. Here we report in vitro and in vivo experiments showing that Annexin A2 present in this subset of mRNP complexes interacts with its cognate mRNA and c-myc mRNA, but not with beta(2)-microglobulin mRNA translated on membrane-bound polysomes. The protein recognises sequence elements within the untranslated regions, but not within the coding region, of its cognate mRNA. Alignment of the Annexin A2-binding 3'-untranslated regions of annexin A2 mRNA from several species reveals a five nucleotide consensus sequence 5'-AA(C/G)(A/U)G. The Annexin A2-interacting region of the 3'-untranslated region can be mapped to a sequence of about 100 nucleotides containing two repeats of the consensus sequence. The binding elements appear to involve both single and double stranded regions, indicating that a specific higher order mRNA structure is required for binding to Annexin A2. We suggest that this type of interaction is representative for a group of mRNAs translated on cytoskeleton-bound polysomes.

The 3'-untranslated region directs ribosomal protein-encoding mRNAs to specific cytoplasmic regions

mRNA localization is a conserved post-transcriptional process crucial for a variety of systems. We have analyzed the subcellular distribution of mRNAs encoding human cytosolic and mitochondrial ribosomal proteins. Biochemical fractionation experiments showed that the transcripts for cytosolic ribosomal proteins associate preferentially with the cytoskeleton via actin microfilaments. Transfection in HeLa cells of a GFP reporter construct containing the cytosolic ribosomal protein L4 3'-UTR showed that the 3'-UTR is necessary for the association of the transcript to the cytoskeleton. Using confocal analysis we demonstrate that the chimeric transcript is specifically associated with the perinuclear cytoskeleton. We also show that mRNA for mitochondrial ribosomal protein S12 is asymmetrically distributed in the cytoplasm. In fact, this transcript was localized mainly in the proximity of mitochondria, and the localization was 3'-UTR-dependent. In summary, ribosomal protein mRNAs constitute a new class of localized transcripts that share a common localization mechanism.

An AU-rich stem-loop structure is a critical feature of the perinuclear localization signal of c-myc mRNA

In eukaryotic cells, several mRNAs including those of c-myc and c-fos are localized to the perinuclear cytoplasm and associated with the cytoskeleton. The localization element of c-myc mRNA is present within its 3'UTR (3'-untranslated region) but the precise nature of this signal has remained unidentified. Chemical/enzymatic cleavage with RNases (ribonucleases) and lead have identified single-stranded and double-stranded regions in RNA transcripts of nucleotides 194-280 of the c-myc 3'UTR. Combined with computer predicted structure these results indicate that this region folds so that part of it forms a stem-loop structure. A mutation, that has been previously shown to prevent localization, leads to a different secondary RNA structure in this region as indicated by altered cleavage patterns. Competitive gel-retardation assays, using labelled transcripts corresponding to nucleotides 205-280 of c-myc 3'UTR, and fibroblast extracts revealed that the stem-loop region was sufficient for RNA-protein complex formation. In situ hybridization studies in cells transfected with reporter constructs, in which all or parts of the region corresponding to this stem-loop were linked to beta-globin, indicated that this region was sufficient for localization and that deletion of the nucleotides corresponding to the proposed upper-stem or terminal loop prevented localization. Our hypothesis is that an AU-rich stem-loop structure within nt 222-267 in the c-myc 3'UTR forms the perinuclear localization signal. Bioinformatic analysis suggests that this signal shares features with 3'UTRs of other localized mRNAs and that these features may represent a conserved form of signal in mRNA localization mechanisms.

Ectopic expression of a polyalanine expansion mutant of poly(A)-binding protein N1 in muscle cells in culture inhibits myogenesis

Oculopharyngeal muscular dystrophy (OPMD) is an adult-onset dominant genetic disease caused by the expansion of a GCG trinucleotide repeat that encodes the polyalanine tract at the N-terminus of the nuclear poly(A)-binding protein (PABPN1). Presence of intranuclear inclusions (INIs) containing PABPN1 aggregates in the skeletal muscles is the hallmark of OPMD. Here, we show that ectopic expression of the mutant PABPN1 produced INIs in a muscle cell culture model and reduced expression of several muscle-specific proteins including alpha-actin, slow troponin C, muscle creatine kinase, and two myogenic transcription factors, myogenin and MyoD. However, the levels of two upstream regulators of the MyoD gene, the Myf-5 and Pax3/7, were not affected, but both proteins co-localized with the PABPN1 aggregates in the mutant PABPN1 overexpressing cells. In these cells, although myogenin and MyoD levels were reduced, these two transcription factors did not co-localize with the mutant PABPN1 aggregates. Therefore, sequestration of Myf5 and Pax3/7 by the mutant PABPN1 aggregates was a specific effect on these factors. Our results suggest that trapping of these two important myogenic determinants may interfere with an early step in myogenesis.

An eleven nucleotide section of the 3'-untranslated region is required for perinuclear localization of rat metallothionein-1 mRNA

Localization of mRNAs provides a novel mechanism for synthesis of proteins close to their site of function. MT1 (metallothionein-1) is a small, metal-binding protein that is largely cytoplasmic but which can be found in the nucleus. The localization of rat MT1 requires the perinuclear localization of its mRNA by a mechanism dependent on the 3'-UTR (3'-untranslated region). The present study investigates the nature of this mRNA localization signal using Chinese-hamster ovary cells transfected with gene constructs in which either MT1 or the globin coding region is linked to different sequences from the MT1 3'-UTR. Deletion, mutagenesis and antisense oligonucleotide approaches indicate that nt 45-76 of the 3'-UTR, in particular nt 66-76, are required for the localization of either MT1 mRNA or chimaeric transcripts in which a beta-globin coding region is linked to sequences from the MT1 3'-UTR. This section of the 3'-UTR contains a CACC repeat. Two mutations that are predicted to alter the secondary structure of this region also impair localization. Our hypothesis is that the perinuclear localization signal in MT1 mRNA is formed by a combination of the CACC repeat and its structural context.

Induction of HSP70 expression and recruitment of HSC70 and HSP70 in the nucleus reduce aggregation of a polyalanine expansion mutant of PABPN1 in HeLa cells

Nuclear inclusions formed by the aggregation of a polyalanine expansion mutant of the nuclear poly(A)-binding protein (PABPN1) is a hallmark of oculopharyngeal muscular dystrophy (OPMD). OPMD is a dominant autosomal disease in which patients exhibit progressive difficulty of swallowing and eyelid elevation, starting around the age of 50. At present, there is no specific treatment to reduce the aggregate burden in patients. However, in cell culture models of OPMD, reduction of protein aggregation can be achieved by ectopic expression of HSP70. As gene transfer may not be the most effective means to elevate HSP70 levels, we tested four pharmacological agents for their ability to induce HSP70, recruit both HSP70 and HSC70 into the cell nucleus and reduce mutant PABPN1 aggregation in a HeLa cell culture model. We show here that exposure to moderate levels of ZnSO4, 8-hydroxyquinoline, ibuprofen and indomethacin produced a robust stress response resulting in the induction of HSP70 in HeLa cells expressing the mutant PABPN1 as a green fluorescent protein (GFP) fusion protein. Both HSP70 and the constitutive chaperone HSC70 localized in the nucleus of cells treated with any one of the four agents. This stress response was similar to what was observed following hyperthermia. All four agents also caused a significant reduction in the cellular burden of protein aggregates, as was judged by confocal microscopy and solubility changes of the aggregates. A concomitant reduction of cell death in drug-treated mutant PABPN1 expressing cells was also observed.

Perinuclear localization of slow troponin C m RNA in muscle cells is controlled by a cis-element located at its 3' untranslated region

The process of mRNA localization within a specific cytoplasmic region is an integral aspect of the regulation of gene expression. Furthermore, colocalization of mRNAs and their respective translation products may facilitate the proper assembly of multi-subunit complexes like the thick and thin filaments of muscle. This postulate was tested by investigating the cytoplasmic localization of three mRNAs-the alpha-actin, slow troponin C (sTnC), and slow troponin I (sTnI), which encode different poly-peptide partners of the thin filament. Using in situ hybridization we showed that all three thin filament mRNAs are localized in the perinuclear cytoplasm of cultured C2C12 muscle cells. Their localization differs from that of the nonmuscle beta-actin mRNA, which is localized in the peripheral region of both proliferating nondifferentiated myoblasts and the differentiated myocytes. Analysis of the localization signal of the sTnC mRNA showed that a 40-nucleotide-long region of the sTnC mRNA 3' UTR is sufficient to confer the perinuclear localization on a heterologous reporter beta-Gal mRNA. This localization signal showed tissue specificity and worked only in the differentiated myocytes, but not in the proliferating myoblasts or in HeLa cells. The predicted secondary structure of the localization signal suggests the presence of multiple stem and loop structures in this region of the 3' UTR. Mutations within the stem region of the localization signal, which abolish the base pairing in this region, significantly reduced its perinuclear mRNA localization activity. Using UV-induced photo-cross-linking of RNA and proteins we found that a myotube-specific 42-kDa polypeptide binds to the localization signal.

Annexin A2 binds to the localization signal in the 3' untranslated region of c-myc mRNA

Messenger RNA trafficking, which provides a mechanism for local protein synthesis, is dependent on cis-acting sequences in the 3' untranslated regions (3'UTRs) of the mRNAs concerned acting together with trans-acting proteins. The C-MYC transcription factor is a proto-oncogene product involved in cell proliferation, differentiation and apoptosis. Localization of c-myc mRNA to the perinuclear cytoplasm and its association with the cytoskeleton is determined by a signal in the 3'UTR. Here we show the specific binding of a trans-acting factor to the perinuclear localization element in the 3'UTR of c-myc mRNA and identify this protein as annexin A2. Gel retardation and UV cross-linking experiments showed that proteins in fibroblast extracts formed complexes with the region of c-myc 3'UTR implicated in localization; a protein of approximately 36 kDa exhibited specific, Ca(2+)-dependent binding. Binding was reduced by introduction of a mutation that abrogates localization. Using RNA-affinity columns followed by gel electrophoresis and mass spectrometry this protein was identified as annexin A2. The RNA-protein complex formed by cell extracts was further retarded by anti-(annexin A2). Purified annexin A2 bound to the same region of the c-myc 3'UTR but binding was reduced by introduction of a mutation, as with cell extracts. It is proposed that binding of annexin A2 to the localization signal in the c-myc mRNA leads to association with the cytoskeleton and perinuclear localization. The data indicate a novel functional role for the RNA-binding properties of annexin A2 in perinuclear localization of mRNA and the association with the cytoskeleton.

PKA-dependent binding of mRNA to the mitochondrial AKAP121 protein

Protein kinase A (PKA) anchoring proteins (AKAPs) tether PKA to various subcellular locations. AKAP121, which tethers PKAII to the outer mitochondrial membrane, includes a K homology (KH) RNA-binding motif. Purified AKAP121 KH domain binds the 3' untranslated regions (3'UTRs) of transcripts encoding the Fo-f subunit of mitochondrial ATP synthase and manganese superoxide dismutase (MnSOD). Binding requires a structural motif in the 3'UTR and is stimulated by PKA phosphorylation of the domain or a mutation that mimics this phosphorylation. AKAP121 expressed in HeLa cells promotes the translocation of MnSOD mRNA from cytosol to mitochondria and an increase in mitochondrial MnSOD. Both reactions are stimulated by cAMP. Thus, by focusing translation at the mitochondrial membrane, AKAP121 may facilitate import of mitochondrial proteins in response to cAMP stimulation.

The 3' untranslated region of human vimentin mRNA interacts with protein complexes containing eEF-1gamma and HAX-1

Previously, we have shown that the vimentin 3' untranslated region (3'UTR) contains a highly conserved region, which is sufficient for the perinuclear localization of a reporter mRNA. This region was shown to specifically bind protein(s) by band shift analyses. UV-cross-linking studies suggest these proteins are 46- and 35-kDa in mass. Here, we have used this sequence as 'bait' to isolate RNA binding proteins using the yeast three-hybrid method. This technique relies on a functional assay detecting bona fide RNA-protein interaction in vivo. Three cDNA isolates, HAX-1, eEF-1gamma and hRIP, code for proteins of a size consistent with in vitro cross- linking studies. In all cases, recombinant proteins were capable of binding RNA in vitro. Although hRIP is thought to be a general mRNA binding protein, this represents an unreported activity for eEF-1gamma and HAX-1. Moreover, HAX-1 binding appears to be specific to vimentin's 3'UTR. Both in vivo synthesized eEF-1gamma and HAX-1 proteins were 'pulled out' of HeLa whole cell extracts by binding to a RNA affinity column comprised of vimentin's 3'UTR. Moreover, size-fractionation of extracts results in the separation of large complexes containing either eEF-1gamma or HAX-1. Thus, in addition to their known functions, both eEF-1gamma and HAX-1 are RNA binding proteins, which suggests new roles in mRNA translation and/or perinuclear localization.