Posttranscriptional Control of Renin Synthesis
Identification of Proteins Interacting With Renin mRNA 3′-Untranslated Region
Stabilization and correct localization of mRNA are important features of renin synthesis. To elucidate the molecular basis of cAMP-mediated posttranscriptional control via mRNA stabilization, we analyzed the interaction of human preprorenin (hREN) mRNA 3′-untranslated region (3′-UTR) with proteins of renin synthesizing Calu-6 cells and investigated their functional impact on messenger integrity. To identify hREN mRNA binding proteins, electrophoretic mobility shift assays, UV cross-linking and RNA-affinity chromatography with subsequent matrix-assisted laser desorption/ionization time-of-flight mass spectrometry were performed. The following six proteins were unambiguously identified as hREN mRNA 3′-UTR binding proteins: hnRNP E1 (synonyms α-CP or PCBP), hnRNP K, dynamin, nucleolin, YB-1, and MINT-homologous protein. All proteins contain various RNA binding motifs, and most have been described in the context of mRNA binding and mRNA stabilization. Four proteins for which antibodies were available were verified by immunological techniques (dynamin, nucleolin, hnRNP E1, and YB-1). Forskolin, an activator of cAMP synthesis, considerably stimulates renin synthesis via inhibition of REN mRNA decay. Functionally, this cAMP-based mRNA stabilization is accompanied by a 3- to 6-fold upregulation of REN mRNA binding proteins. RNase degradation assays confirm that 3′-UTR binding proteins are able to protect and stabilize REN mRNA in vitro.
Plasma renin results from transcriptional control of the prorenin gene, posttranscriptional control at the mRNA level, posttranslational processing of the prorenin protein, uptake of renin into vesicles, and renin release.1–4 Cultures of renin-producing pulmonary, eg, Calu-6, cells2 are commonly used for studying control of expression of the REN gene. Quantitatively, a very important regulatory step of renin synthesis seems to occur after the transcription processes. Remarkably, cAMP, a well-known inductor of renin synthesis, not only activates transcription of the REN gene, but also increases REN mRNA half-life, thereby elevating REN synthesis.1,2,5,6 In Calu-6 cells, hREN mRNA levels increase up to 100-fold in response to forskolin stimulation under conditions of blocked transcription.5 Although posttranscriptional REN mRNA stabilization contributes to developmental or cAMP-based upregulation of renin synthesis, very little is known about the mediators of mRNA stability. Moreover, it remains to be unraveled how REN mRNA interacts with intracellular structures to target REN mRNA in such a way that renin can be efficiently deposited in storing vesicles. Determinants involved in control of functional properties of mRNAs such as translational efficiency, metabolic stability, or intracellular localization reside predominantly in 5′- or 3′-untranslated regions (UTRs) of the mRNA. Their regulatory potential takes place via interaction with regulatory proteins.7,8 Neither cis control elements in REN mRNA nor interacting regulatory trans factors are known.
In this study, proteins are identified that interact with hREN mRNA 3′-UTR. For characterization we used gel-shift analysis, UV cross-linking, RNA-affinity chromatography with subsequent matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) identification, and immunological techniques. Functionally, we can show that the cAMP-based increase of REN mRNA stability is accompanied by an upregulation of REN mRNA binding proteins that are known for their mRNA stabilizing potential.
Materials and Methods
The 196-nt 3′-UTR of human prorenin mRNA was obtained by polymerase chain reaction (PCR; forward primer, 5′-GGCCCTCTGCCACCCAG-3′; backward primer, 5′-GTGAACATGAAGTCTTTATT-3′) from human genomic DNA and cloned into the pCRII-TOPO vector (Invitrogen).
The 283-nt 3′-UTR1 of collagen 1A1 mRNA was amplified by PCR (forward primer, 5′-ACTCCCTCCATCCCAACCT-3′; backward primer, 5′-ACCAAGCTTCCTTTTTTAAAAAC-3′) from human genomic DNA and cloned into the pCRII-TOPO vector (Invitrogen).
LOX-DICE (Lipoxygenase Differentiation Control Element) and LOX-NR (LOX Nonrepeat Part of mRNA 3′-UTR)
Construction of 15-lipoxygenase mRNA 3′-UTR plasmids and fragments therefrom was described earlier.9
Calu-6 cells from the American Type Culture Collection were cultured in MEM with Earle’s salt supplemented with l-glutamine, sodium bicarbonate, nonessential amino acids, sodium pyruvate, penicillin/streptomycin, and 10% FCS (Biochrom KG Seromed). Cells were grown in 15-cm dishes at 37°C in humidified air with 5% CO2. At 80% to 90% confluence, cells were harvested by 0.05% trypsin/0.02% EDTA treatment. Cell pellets were washed twice with PBS, frozen in liquid nitrogen, and stored at −80°C.
Protein Extract Preparation
Cell pellets were mixed with two volumes of ice-cold lysis buffer (containing, in mmol/L, HEPES 10 [pH 7.2], MgCl2 1.5, LiCl 10, and DTT 0.56, and the “complete mini” protease inhibitor cocktail [Roche]) and homogenized with a Polytron-PT300 blender (Kinematica AG). The homogenate was centrifuged twice to spin down cell membranes and organelles (10 minutes, 4°C, 10 000g) The supernatant is designated S10. Protein extracts were aliquoted, frozen in liquid nitrogen, and stored at −80°C.
Preparation of Polysomes, mRNPs, and RNA
Polysomes were obtained from S10 protein extracts by centrifugation for 2 hours at 100 000g and 4°C. The postpolysomal mRNP fraction was sedimented from the S100 supernatant by additional centrifugation for 3 hours at 300 000g and 4°C.
Polysomal and mRNP pellets were dissolved in TKM buffer (containing, in mmol/L, Tris 50, KCl 25, and MgCl2 5). RNA isolations from polysomes and RNPs were performed with RNA Clean (Hybaid).
For amplification of hREN 3′-UTR, 4 μg of total RNA from polysomes, or mRNPs, superscript II reverse transcriptase and oligo(dT) 12 to 18 primers (Gibco BRL Life Technologies) were used. The consecutive PCR (25 μL) contained 1 μL of the reverse transcriptase reaction, 1.5 mmol/L MgCl2, 0.2 mmol/L each dNTP, and 1 μmol/L of each primer (forward, 5′-GGCCCTC-TGCCACCCAG-3′; backward, 5′-GTGAACATGAAGT-CTTTATT-3′, 0.05 U DNA-polymerase/μL [Rapidozym/Genecraft]). Amplification of the samples comprehended 35 cycles (94°C, 58°C, and 72°C). Simultaneous amplification of a dilution series of a plasmid containing the full-length human renin cDNA served as an internal standard for quantification.
In Vitro Transcription
32P-labeled ([α-32P]UTP, ICN, specific activity 800 Ci/mmol) and unlabeled mRNAs for electrophoretic mobility shift assays (EMSA) and UV cross-linking assays were synthesized by in vitro transcription from the linearized plasmids using either T7 , T3, or SP6 polymerase (Promega). Biotin-labeled renin 3′-UTR for affinity chromatography was generated in the same way using a mixture of CTP/Biotin-CTP (Gibco) in a ratio of 1:1.
Electrophoretic Mobility Shift Assays
Calu-6 cell extracts, representing 70 to 80 μg protein, were incubated with 10 000 cpm 32P-labeled transcripts (0.1 to 0.2 ng) in a buffer containing (in mmol/L) HEPES 10 (pH 7.2), MgCl2 3, and DTT 1; 50 ng/μL Escherichia coli rRNA; 0.1 mol/L KCl; 20 U RNasin (Promega); and 5% glycerol for 15 minutes at room temperature (RT). After addition of heparin to a final concentration of 5 mg/mL, followed by an incubation of 10 minutes at RT, the samples were analyzed by electrophoresis on 4% polyacrylamide gels in 0.5× TBE (Tris-boric acid-EDTA) and autoradiography.
Incubation of protein extracts and transcripts (100 000 cpm, 1 to 2 ng) was done as for EMSA. After the addition of heparin, samples were exposed to UV light (255 nm, 15 minutes) on ice. Then, samples were digested with RNase A (30 μg/mL final concentration) and RNase T1 (750 U/mL final concentration) and analyzed by 12% SDS-PAGE and autoradiography.
Streptavidin-coated agarose beads suspension (600 μL; Gibco BRL Life Science) was equilibrated four times with washing buffer (WB; in mmol/L, KCl 150, MgCl2 1.5, Tris 10 [pH 7.5], and DTT 0.5). Washing was followed by 2 minutes of centrifugation (3500 rpm). Packed beads were suspended in 50 μL of WB. Biotinylated hREN 3′-UTR or control LOX mRNA 3′-UTR–NR9 transcript (15 to 20 μg) was added, and the mixture was incubated on a rotating wheel (100 minutes, 4°C). After centrifugation and supernatant removal, 1 mL protein extract of Calu-6 cells (pretreated with 160 U RNasin, Promega) was added and incubated for 45 minutes at 4°C and 15 minutes at RT. After five washing cycles with 1 mL WB each, the bound proteins were eluted in two steps with 100 μL 1 mol/L and 3 mol/L KCl in 0.75 mmol/L MgCl2 and 5 mmol/L Tris (pH 7.5) each. Eluates were dialyzed, analyzed by 12% SDS-PAGE, or used in RNA degradation assays. For MALDI-TOF-MS analysis, bands were cut out from Coomassie-stained gels.
Immunology (Western Analysis and RIA)
Proteins (20 to 40 μg) were subjected to 10% SDS-PAGE and transferred to Immobilon-P membranes (Millipore) by semidry electroblotting. Blocking of nonspecific binding took place by incubating the membranes for 1 hour in TBS-T (20 mmol/L Tris base, 137 mmol/L NaCl, and 0.1% Tween 20) with 5% nonfat dried milk before immunological detection. Incubations with primary and horseradish peroxidase enzyme–conjugated secondary antibodies, respectively, were performed in TBS-T with 5% nonfat dried milk for 1 hour at RT. Primary antibodies were from Santa Cruz Biotechnology and BioGenes GmbH.
Washing after membrane blocking and antibody incubations were performed with TBS-T. The last wash was followed by incubation of membranes with ECL Western Blotting Detection Reagent (Amersham Pharmacia Biotechnology, 5 minutes, RT). Membranes were exposed to x-ray films for 1 to 5 minutes.
Total renin (prorenin/renin, sum of intracellular and secreted renin) was determined by an RIA as described by the manufacturer (Nichols Institute Diagnostics).
Cut gel pieces from Coomassie-stained polyacrylamide gels were prepared for tryptic digestion using standard techniques (0.1 mol/L NH4HCO3 and acetonitrile equilibration, treatment with 10 mmol/L DTT and 55 mmol/L iodine acetamide in 0.1 mol/L NH4HCO3). Trypsinization was performed at 37°C overnight with 12.5 ng/μL trypsin (Boehringer Mannheim, modified, sequencing grade) in 50 mmol/L NH4HCO3. Peptides were extracted with 25 mmol/L NH4HCO3 and acetonitrile was supported by sonication. For mass spectrometric peptide mapping, dried peptides were redissolved in 5 μL 0.1% trifluoroacetic acid and purified on a 200-nL reversed-phase (C18) nanocolumn. Peptides were eluted in 5 μL 70% (vol/vol) acetonitrile and subsequently cocrystallized with α-cyano-4-hydroxycinnamic acid (20 mg/mL) in 70% acetonitrile on a SCOUT 384- to 600-μm anchor target. Mass spectrometry was performed on MALDI-TOF-MS (Reflex III, Bruker Daltonics) in reflector mode with external calibration. Tryptic fragments were annotated using the BioTools 2.0 software (Bruker Daltonics). For the database search the Mascot software (Matrix Science) was used.
In Vitro mRNA Degradation Assay
The in vitro mRNA degradation assay was performed as described earlier.10 Briefly, 10 000 cpm (0.1 to 0.2 ng) in vitro transcribed 32P-labeled REN mRNA was capped and polyadenylated as described in Reference 10. The decay reaction was carried out by adding 50 μg cytosolic protein extract (S300) from Calu-6 cells as a source for degrading RNases and 100 ng affinity-purified RNA binding proteins. Aliquots were removed at the times indicated, and phenol/chloroform was extracted and analyzed by agarose gel electrophoresis and blotting onto membranes. Signals were generated by exposing the membranes to x-ray films and quantified by scanning using the ScanPack 3.0 software of Biometra.
In comparison with the REN mRNA 3′-UTR, the REN 5′-UTR is very short and not highly conserved throughout species (Figure 1). In the well-conserved 3′-UTR, clusters of CU-rich elements were found that resemble the control element LOX-DICE.11
To analyze the potential of REN mRNA 3′-UTR to bind regulatory proteins, we performed binding studies using EMSA, UV cross-linking, and RNA-affinity chromatography. Figure 2 shows that hREN 3′-UTR forms stable protein complexes with proteins from Calu-6 cytoplasm (lanes 1 and 2). As an indication of specific interaction, the complexes can be competed out with unlabeled transcript (lanes 3 and 4). When a cold control transcript that is not related to REN 3′-UTR (rabbit 15-lipoxygenase mRNA 3′-UTR–NR, nt 2262 to 2595, M27214) is used as competitor, no competition takes place up to a 100-fold concentration over labeled REN 3′-UTR transcript (lanes 5 to 8). This indicates a specific interaction of distinct Calu-6 proteins with REN 3′-UTR. REN 3′-UTR is also able to bind purified recombinant hnRNP proteins E1 and K (Figure 3, lanes 5 and 6). The well-known E1 and K binders LOX-3′-UTR–DICE and COL1A1 3′-UTR (lanes 9 and 12; only K is shown) served as positive controls. The nonrepeat motif of LOX 3′-UTR (lane 15)9 served as a negative control. Furthermore, we tested whether the REN 3′-UTR binding proteins are restricted to Calu-6 cytoplasm (S300) or whether they are also components of the two functional forms of mRNA containing mRNP complexes, polysomes, and postpolysomal mRNPs. Lanes 3 and 4 of Figure 3 demonstrate that REN 3′-UTR binding proteins are not only present in particle-free cytoplasm (S300) but are also associated with polysomes and postpolysomal mRNP complexes.
Figure 4 is a representative photo cross-linking experiment. Several proteins were able to bind to REN 3′-UTR. The binding proteins were obtained from the following three Calu-6 cytosolic protein sources: cytoplasm (lane 2), polysomes (lane 3), and postpolysomal mRNPs (lane 4). SDS-electrophoresis of the cross-linked proteins visualizes protein bands in the range of ≈20 to 100 kDa (Figure 4A). The most prominent bands were designated p30, p35, p43, p55 p66, p76, and p97. Interestingly, some bands appear in all three fractions (p30, p35, p43, and p66), and others are prominent only in cytoplasm (p55) or are enriched in polysomes/RNPs (p76) and cytoplasm/polysomes (p97). Gel-shift experiments (Figure 3) indicated that hnRNP proteins E1 and K could be natural binding partners of REN 3′-UTR. In the cross-linking pattern, candidate bands are visible that fit the molecular weights of authentic E1 (43 kDa) and K (66 kDa). This was verified with cross-linking experiments using recombinant E1 and K (Figure 4B). REN 3′-UTR binds K with the same strength as control LOX 3′-UTR (lanes 1 and 3), E1 with somewhat lower efficiency (lanes 2 and 4).
The cross-linking pattern of REN 3′-UTR with Calu-6 cytoplasm indicates that still other proteins might be potential interaction partners. Another approach to identify RNA binding proteins is RNA-affinity chromatography via biotinylated transcripts coupled to streptavidin-agarose11 as shown in Figure 5. This method gives a considerable background of proteins nonspecifically bound to the matrix without transcript (Figure 5A, lane 6). This, however, is superimposed by a pattern of a few additionally bound proteins (lanes 4 and 5, see >) eluted with high salt concentration. Bands were cut from lanes 4 (1 mol/L KCl) and 5 (3 mol/L KCl), combined, and subjected to MALDI-TOF-MS analysis. Six proteins were unambiguously identified and verified by high lot scores (Figure 6). They were identified as Y-box binding protein YB-1 (p35, NP 035862), nucleolin (p55, NP 005372), dynamin (p97, NP 056384; authentic dynamin has a molecular weight of 100 kDa), hnRNP E1 (p43; synonyms α-CP or PCBP, NP 006187), hnRNP K (p66, NP 002131), and a 56-kDa MINT homologue protein (p30, NP 0055816).
To verify the MALDI-TOF-MS results, SDS gels were probed with the available four antibodies. Duplicate strips of affinity-enriched binding proteins separated by SDS-electrophoresis were probed as shown in Figure 5B. The results confirm the MALDI-TOF-MS identification. The immunological analysis also indicated that the lower molecular bands of nucleolin and YB-1 were the result of proteolytic degradation and that the differences with YB-1 were due to unusual electrophoretic migration.
Finally, to analyze the functional relevance of REN mRNA binding proteins, we asked how they respond to cAMP induction, a stimulus known to stabilize REN mRNA.1,5 Furthermore, we asked whether they were able to influence messenger lifetime. For this purpose, Calu-6 cells were preincubated with forskolin, a cyclase activator, for 24 hours and the stability of REN mRNA (Northern blots), REN protein (RIA), and four REN mRNA binding proteins (Western blots) was determined (Figure 7A). As expected, 24-hour forskolin stimulation increased the REN mRNA level ≈10- to 20-fold and REN total protein by a factor of 15. The level of REN mRNA binding proteins also increased, but to a different extent, as follows: nucleolin 3-fold, dynamin 5-fold, hnRNP E1 4-fold, and YB-1 6-fold. Under the same conditions, control β-actin mRNA and protein were unchanged. Actinomycin D inhibition experiments (not shown) proved that the increase in REN mRNA was due to mRNA stabilization and not to transcriptional activation (authors’ unpublished results, 2002, see also Reference 1). To verify the effect of REN mRNA binding proteins on mRNA stability, in vitro RNase degradation assays were performed (Figure 7B). Labeled, in vitro transcribed REN mRNA (capped and polyadenylated) was incubated with and without REN mRNA binding proteins (purified by RNA-affinity chromatography; see Materials and Methods and Figure 5) in the presence of Calu-6 cytosol (S100) as the source of the responsible RNases. As shown in Figure 7B, binding proteins exhibit clearly a protecting effect on REN mRNA integrity. Under the conditions chosen, the half-life of REN mRNA increased ≈3-fold. A control preparation of proteins obtained by RNA-affinity chromatography using LOX mRNA 3′-UTR–NR as a matrix (a sequence element known not to bind hnRNPs E1 and K)9 has only a marginal stabilizing effect.
Several posttranscriptional events are essential for REN mRNA synthesis and intracellular localization. For instance, modulation of REN mRNA stability has been shown,1 which is crucial for cAMP-based upregulation of renin expression.5,6 Furthermore, because REN is secreted from vesicles,12 it is also important for preprorenin to be synthesized at membrane-bound polysomes. We applied techniques to investigate hREN mRNA protein interaction concentrating on hREN 3′-UTR as the most probable target of interaction with control proteins. These proteins could mediate the effects on REN mRNA stability and the association with membrane-bound polysomes. UV cross-linking studies and RNA-affinity chromatography combined with MALDI-TOF-MS analysis and immunological techniques identified the following six proteins with high affinity for hREN mRNA 3′-UTR: hnRNP proteins E1 and K, dynamin, nucleolin, YB-1, and MINT-homologous protein. All six contain RNA- or general nucleotide–recognition motifs. Moreover, immunologically detectable hREN mRNA binding proteins were found not only in free form in the cytoplasm, but also in hREN mRNA containing mRNP complexes (polysomes and postpolysomal free mRNP particles). As discussed below, dynamin and YB-1 are essential for vesicle trafficking. YB-1, nucleolin, and hnRNP E1 and K have importance for mRNA stability.
Posttranscriptional modulation often occurs by protein-nucleotide interactions at the mRNA 5′- or 3′-UTR. Sequence alignments for preprorenin mRNAs showed that the 5′-UTR is not highly conserved between the four cloned species (Figure 1). With only 32 to 46 nucleotides, these short sequences are assumably only important for ribosome binding. Conversely, hREN 3′-UTR consists of 196 nucleotides and is well conserved between mammalian REN mRNAs (Figure 1). The hREN mRNA 3′-UTR contains CU-rich sequence elements, which are very similar to the LOX-DICE, a motif that binds hnRNP proteins E1 and K.11 CU-rich motifs of the DICE type have been found in many mRNA 3′-UTRs.13 However, another basic determinant involved in mRNA stability, the 3′-terminal poly(A), does not seem to play a role in cAMP-dependent stabilization of REN mRNA.5
To elucidate the capability of REN mRNA 3′-UTR to interact with proteins of potential regulatory significance, we performed protein binding studies using in vitro transcribed human REN mRNA 3′-UTR and cytosolic protein extracts of Calu-6 cells. The six identified proteins (except the 56-kDa MINT homologue) contain RNA binding or general nucleotide binding motifs. The 56-kDa MINT homologue, however, is part of a much larger 358-kDa RNA binding protein MINT (Msx2-interacting nuclear target protein, interacting with the homeodomain transcriptional repressor Msx2), which is predicted by MALDI-TOF with the same probability. This large 358-kDa molecule contains three RNA-recognition motifs.14
hnRNP E1 and K
hnRNP E1 and K belong to a group of nuclear RNA binding proteins also operating in the cytosol to modulate mRNA function.7,15 hnRNP E proteins contain a specific RNA-recognition motif, the KH (K homology) domain. The three KH domains of E1/E2 and K interact with pyrimidine-rich target RNAs mediating translational regulation and mRNA stabilization. The most prominent example for translational regulation of by E1/K is inhibition of 15-lipoxygenase synthesis via CU-rich DICE control elements in LOX mRNA 3′-UTR.9,11,15 The paradigm of RNA stabilization is the extremely high metabolic stability of the globin mRNAs caused by assembly of E1/K to pyrimidine-rich sequences in α/β-globin mRNA 3′-UTRs.16–18 Interestingly, hREN mRNA 3′-UTR contains oligo(C) clusters (Figure 1), which meet exactly the requirements for E1 and K binding as a prerequisite for mRNA stabilization.19 Our UV cross-linking experiments with Calu-6 cytoplasm and recombinant E1 and K confirm the theoretical prediction and show that both are effectively bound to hREN mRNA and may substantially contribute to metabolic stabilization. It has been shown, at least for hnRNP K, that the efficiency of mRNA binding is controlled by cAMP-dependent phosphorylation via ERK kinases.20 This might provide the molecular basis for the observed increase in renin mRNA half-life via cAMP.5 Moreover, hnRNP E1 is also a target for phosphorylation.21
Dynamin, a 100-kDa GTPase, is an essential component of formation, recycling, and trafficking of vesicles. In addition to the GTPase domain near the N terminus, which also contains the two nucleotide binding motifs, dynamin contains other domains such as a pleckstrin homology domain (PH) implicated in membrane binding and a C-terminal proline-rich domain (PRD) involved in protein-protein interaction.22 Dynamin is not a classical RNA binding protein. However, it cannot be excluded that the nucleotide binding motifs, which are part of the GTPase domain, have a general affinity to RNA. Secondary protein-protein interaction with one of the RNA binding proteins is also possible. hnRNP K may be a candidate. It has been described as a general RNA-associated docking platform, capable of interacting with a wide variety of other proteins.23 As a secretory protein, preprorenin is synthesized at membrane-bound polysomes. Thus, dynamin could be a part of the system that targets REN polysomes to endoplasmic reticulum and Golgi membranes and helps to coordinate synthesis, intracellular mRNA localization, and storage of renin in vesicles.
Nucleolin is an abundant 76-kDa protein of the nucleolus and is also found in the cytoplasm. The four RNA binding domains in the central portion of the molecule and the carboxyl-terminal RGG domain allow interaction with different RNA sequences and proteins. Nucleolin has been implicated in chromatin structure, rDNA transcription, rRNA maturation, ribosome assembly, nuclear/cytoplasmic RNA transport, mRNA stabilization, and mRNP assembly and masking.24
Nucleolin may act to stabilize mRNA in hREN mRNP complexes. Remarkably, hnRNP C and nucleolin are the main factors controlling posttranscriptional upregulation of amyloid precursor protein (APP) synthesis in Alzheimer disease. This occurs by binding to the 3′-UTR of APP mRNA, thereby increasing its half-life.25
Another prominent example of nucleolin-mediated mRNA stabilization is interleukin-2 mRNA.26 Together with YB-1, nucleolin is part of an interleukin-2 mRNP complex responding to specific mRNA stabilizing signals. This is in parallel to our results showing that REN mRNP complexes also contain a combination of nucleolin and YB-1. Furthermore, the common occurrence of E1/K, YB-1, and nucleolin in hREN mRNPs is in striking resemblance to the mRNA coding for the shuttling, RNA binding KH-domain protein FMRP, which is involved in fragile X mental retardation. 3′-UTR of FMRP mRNA binds its own protein product, FMRP, in conjunction with YB-1 and nucleolin, thereby stabilizing the FMRP message.27,28
Y-box proteins belong to a superfamily of DNA and RNA binding proteins. Whereas many Y-box proteins serve as transcription regulators, others have a predominantly cytoplasmic function serving to stabilize mRNA structure and influence translation.29
In vitro, YB-1/p50 exhibits no protection of mRNA against degradation. Instead, it considerably increases RNase sensitivity. This may result from the capability of YB-1 to melt RNA secondary structures.30 However, in combination with other cis/trans elements/factors, protection can be demonstrated in vivo.26 Binding to granulocyte-macrophage colony-stimulating factor mRNA 3′-UTR YB-1 is the main factor responsible for mRNA stabilization.31
YB-1 has further interesting interacting capabilities with other RNA-regulatory proteins, as follows. hnRNP K, which we also detected in hREN mRNPs, was found to be an interaction partner of YB-1 in 2-hybrid screens.32 In REN-producing cells, YB-1 could also mediate the correct intracellular positioning of hREN mRNA, given that YB-1 interacts with actin microfilaments.33 Fifteen to thirty percent of mRNAs and polysomes of the cell are associated with the cytoskeleton. This can be important in targeting mechanisms and in the transport of mRNA from the nucleus to its correct location in the cytoplasm.34
The 358-kDa MINT contains three N-terminal RNA-recognition motifs that are also present in the physiologically occurring 56-kDa fragment.13 Although MINT and the N-terminal fragments contain three RNA-recognition motifs, no cytoplasmic RNA binding activity has been reported so far. Our result showing the binding of 56-/30-kDa MINT to hREN 3′-UTR is the first indication of such an RNA interaction.
Calu-6 cells respond very well to forskolin stimulation. Sinn and Sigmund5 reported an up to 100-fold stimulation at the steady-state level of REN mRNA. We generally observed a ≈10 to 20-fold induction of the mRNA amount and a ≈3-fold increase in half-life, consistent with data obtained with primary JG cells and the work cited above.1,5 As a first clue on the functional role of REN mRNA binding proteins, we can show that induction of REN mRNA binding proteins is linked to elevation of REN mRNA stability. This leads to subsequent upregulation of renin synthesis. Moreover, this applies also to the REN mRNA binding proteins, which are known for their mRNA stabilizing potential as hnRNP E1, nucleolin, or YB-1. Moreover, we show for the first time that the synthesis of RNA binding proteins is stimulated by cAMP (forskolin). Whether this is also an effect of mRNA stabilization or primarily a transcription effect remains to be determined. It is evident that they can bind to a large variety of mRNAs and may therefore be part of a network of cAMP-mediated posttranscriptional effects targeting on mRNA stability. Our findings that REN mRNA proteins are able to stabilize authentic REN mRNA in vitro and that these proteins are synergistically upregulated by cAMP give a plausible clue for their regulatory significance.
In conclusion, the situation found with hREN mRNA 3′-UTR reveals striking similarities to other eukaryotic mRNP complexes. A combination of a limited number of RNA/DNA binding proteins associates with the 3′-UTR, thereby modulating the functional properties of mRNA. How these proteins act mechanistically, and how the signaling is mediated in the context of regulation of renin synthesis, remains to be investigated. To our knowledge, with our findings we present the first data on the critical role of REN mRNA binding proteins as mediators of cAMP-based posttranscriptional regulation of renin gene expression.
This work was supported by the Deutsche Forschungsgemeinschaft (to B.-J.T. and P.B.P.). We thank Karsten Jürchott (Berlin-Buch) for the YB-1 antibody.
Original received July 16, 2002; resubmission received November 22, 2002; revised resubmission received January 22, 2003; accepted January 22, 2003.
Chen M, Schnermann J, Smart AM, Brosius FC, Killen PD, Briggs JP. Cyclic AMP selectively increases mRNA stability in cultured juxtaglomerular granular cells. J Biol Chem. 1993; 268: 24138–24144.
Lang JA, Ying LH, Morris BJ, Sigmund CD. Transcriptional and posttranscriptional mechanisms regulate human renin gene expression in Calu-6 cells. Am J Physiol. 1996; 271(1 pt 2): F94–F100.
Kurtz A, Wagner C. Cellular control of renin secretion. J Exp Biol. 1999; 202: 219–225.
Pan L, Black TA, Shi Q, Jones CA, Petrovic N, Loudon J, Kane C, Sigmund CD, Gross KW. Critical roles of cyclic AMP responsive element and an E-box in regulation of mouse renin gene expression. J Biol Chem. 2001; 276: 45530–45538.
Sinn PL, Sigmund CD. Human renin mRNA stability is increased in response to cAMP in Calu-6 cells. Hypertension. 1999; 33: 900–905.
Wang JJ, Rose JC. Developmental changes in renal renin mRNA half-life and responses to stimulation in fetal lambs. Am J Physiol. 1999; 277(4 pt 2): R1130–R1135.
Wang Z, Day N, Trifillis P, Kiledjian M. An mRNA stability complex functions with poly(A)-binding protein to stabilize mRNA in vitro. Mol Cell Biol. 1999; 19: 4552–4560.
Friis UG, Jensen BL, Sethi S, Andresen D, Hansen PB, Skott O. Control of renin secretion from rat juxtaglomerular cells by cAMP-specific phosphodiesterases. Circ Res. 2002; 90: 996–1003.
Holcik M, Liebhaber SA. Four highly stable eukaryotic mRNAs assemble 3′ untranslated region RNA-protein complexes sharing cis and trans components. Proc Natl Acad Sci U S A. 1997; 94: 2410–2414.
Chkeidze AN, Lyakhov DL, Makeyev AV, Morales J, Kong J, Liebhaber SA. Assembly of the α-globin mRNA stability complex reflects binary interaction between the pyrimidine-rich 3′ untranslated region determinant and poly(C) binding protein αCP. Mol Cell Biol. 1999; 19: 4572–4581.
Yu J, Russell JE. Structural and functional analysis of an mRNP complex that mediates the high stability of human β-globin mRNA. Mol Cell Biol. 2001; 21: 5879–5888.
Thisted T, Lyakhov D, Liebhaber SA. Optimized targets of two closely related triple KH domain proteins, heterogeneous nuclear ribonucleoprotein K and αCP-2KL, suggest distinct modes of RNA recognition. J Biol Chem. 2001; 276: 17484–17496.
Denisenko ON, O’Neill B, Ostrowski J, Van Seuningen I, Bomszyk K. Zik1, a transcriptional repressor that interacts with the heterogeneous nuclear ribonucleoprotein particle K protein. J Biol Chem. 1996; 271: 27701–27706.
Ginisty H, Sicard H, Roger B, Bouvet P. Structure and functions of nucleolin. J Cell Sci. 1999; 112: 761–772.
Zaidi SHE, Malter JS. Nucleolin and heterogeneous nuclear ribonucleoprotein C proteins specifically interact with the 3′-untranslated region of amyloid protein precursor mRNA. J Biol Chem. 1995; 270: 17292–17298.
Chen CY, Gherzi R, Andersen JS, Gaietta G, Jürchott K, Royer HD, Mann M, Karin M. Nucleolin and YB-1 are required for JNL-mediated interleukin-2 mRNA stabilization during T-cell activation. Genes Dev. 2000; 14: 1236–1248.
Ceman S, Brown V, Warren ST. Isolation of an FMRP-associated messenger ribonucleoprotein particle and identification of nucleolin and the fragile-X related proteins as components of the complex. Mol Cell Biol. 1999; 19: 7925–7932.
Evdokimova VM, Wie CL, Sitikov AS, Simonenko PN, Lazarev OA, Vasilenko KS, Ustinov VA, Hershey JWB, Ovchinnikov LP. The major protein of messenger ribonucleoprotein particles in somatic cells is a member of the Y-box binding transcription factor family. J Biol Chem. 1995; 270: 3186–3192.
Capowski EE, Esnault S, Bhattacharya S, Malter JS. Y box-binding factor promotes eosinophil survival by stabilizing granulocyte-macrophage colony-stimulating factor mRNA. J Immunol. 2001; 167: 5970–5976.
Shnyreva M, Schullery DS, Suzuki H, Higaki Y, Bomstyk K. Interaction of two multifunctional proteins heterogeneous nuclear ribonucleoprotein K and Y-box-binding protein. J Biol Chem. 2000; 275: 15498–15503.