HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Xu, K. Articles by Murphy, T. J. Search for Related Content PubMed PubMed Citation Articles by Xu, K. Articles by Murphy, T. J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 11, 7604-7611, March 17, 2000

Reconstitution of Angiotensin Receptor mRNA Down-regulation in Vascular Smooth Muscle
POST-TRANSCRIPTIONAL CONTROL BY PROTEIN KINASE A BUT NOT MITOGENIC SIGNALING DIRECTED BY THE 5'-UNTRANSLATED REGION^*

Kaiming Xu and T. J. Murphy

From the Department of Pharmacology, Emory University School of Medicine, Atlanta, Georgia 30322

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell surface receptor activation generally leads to changes in mRNA abundance, which may involve regulatory targets in processes working at the post-transcriptional level. Many types of agonists down-regulate vascular smooth muscle angiotensin receptor (AT₁-R) gene expression, but it is unclear which of these activate post-transcriptional mechanisms. To reconstitute faithfully the normal AT₁-R mRNA regulatory environment, tetracycline-suppressible promoters drive highly accurate recombinant AT₁-R mRNA mimics in vascular smooth muscle cells that co-express an endogenous AT₁-R mRNA. Down-regulation of the latter occurs shortly after stimulating mitogenic receptors or by using forskolin, but only cAMP signaling reduces expression of the recombinant AT₁-R mRNA. Transcription of the recombinant mRNA is unaffected by cAMP signaling. Deletions of the AT₁-R mRNA 3'-untranslated region do not impair cAMP-mediated down-regulation. Both loss of function and gain of function mutants show the response is mediated by the 5'-untranslated region. These observations provide the first direct functional evidence for modulation of vascular AT₁-R gene expression by a mechanism involving a protein kinase A-regulated post-transcriptional process.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The vasoconstrictive and volume sparing hormone angiotensin II acts through a cell surface G_q-coupled AT₁-receptor (AT₁-R).¹ A long-standing observation both in vivo and in culture is that AT₁-R expression is dynamically regulated but by cellular and molecular mechanisms that are not well understood (1, 2). Since modulation of AT₁-R mRNA levels is a dominant mechanism for determining target cell hormonal responsiveness (3), understanding how this essential effector of the renin-angiotensin-aldosterone endocrine axis is regulated might provide new insights into how several cardiovascular diseases develop.

Down-regulation of AT₁-R gene expression occurs within a few hours following activation of vascular smooth muscle cells with many different types of stimuli. These include several growth factors acting through receptor tyrosine kinases (4, 5), hormones, and autacoids acting through either G_s- or G_q-coupled receptors (6-8), and also direct activators of protein kinase C or the cAMP-dependent protein kinase A (PKA) (6, 8-10). We suspect that regulated post-transcriptional mechanisms may account for some or all of the effects of these diverse stimuli. Among the indirect evidence supporting this notion are observations that cAMP-elevating agents do not affect AT₁-R gene transcription (8, 10). Although G_q-coupled and growth factor receptor signaling suppresses AT₁-R gene transcription, the more rapid rate of AT₁-R mRNA decay in response to mitogens relative to transcriptional inhibitors raises the possibility of superimposed post-transcriptional processes (4, 6). Gaining tight and selective experimental control over AT₁-R gene transcription would help to clarify these matters.

A role for post-transcriptional regulation in agonist-mediated control of AT₁-R gene expression has wider implications. Current knowledge about regulated post-transcriptional processes lags considerably the more detailed understanding of how gene transcription rates are modulated by receptor signaling (11). Although there is a growing appreciation that receptor-operated post-transcriptional control processes likely exist, gathering unambiguous functional evidence for this modality of gene expression regulation is not a trivial matter (reviewed in Ref. 12). The typical approach involves measuring how signaling changes mRNA decay rates, using general transcriptional inhibitors like actinomycin D or DRB. The nonspecific effects of these drugs are problematic and raise serious questions about their utility and thus the reliability of many studies that resort to their use. For example, actinomycin D can evoke several stress-like responses in cells, potentially disrupting interfaces between signaling and mRNA metabolism (13-18). Furthermore, the mechanism of action of DRB is through inhibiting phosphorylation of RNA polymerase, but its selectivity for the RNA polymerase kinases versus signaling kinases is unknown (19, 20). Finally, actinomycin D can directly interfere with basal mRNA turnover mechanisms, raising concerns that this might be a more general problem than is currently appreciated (12, 21, 22).

The present study surmounts these difficulties, providing direct functional evidence that post-transcriptional processes regulated by signaling contribute to modulating vascular AT₁-R mRNA abundance. We clarify the role of post-transcriptional regulation among the many different factors that can down-regulate AT₁-R mRNA expression. We begin with the premise that an mRNA will be regulated at the post-transcriptional level irrespective of whether it is transcribed from a recombinant or a native locus. Mimicry by a recombinant surrogate of a response affecting its native transcript is essential direct evidence that a post-transcriptional regulatory process exists. The predictive value of recombinant surrogates then rests largely on how accurately mRNA expression dynamics have been reconstructed. By using tetracycline-suppressible transcription factors (23, 24) to express highly accurate AT₁-R mRNA mimics in vascular smooth muscle cells, we directly test whether post-transcriptional mechanisms are involved in its modulation by specific receptor-coupled signaling systems.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VSMC Cell Culture-- Rat aortic vascular smooth muscle (RASM) cells (passages 8-25) and A7r5 smooth muscle cells were cultured as described (3, 8). After reaching confluence, growth media containing 10% serum was replaced with serum-free media for an additional 16 h before initiating treatments.

Retrovirus Production and Infection-- Infectious retroviral supernatants were generated by a helper virus-free, transient transfection protocol and used to infect smooth muscle cells with >99% efficiency as described elsewhere (25). To ensure stable chromatin integration, cells infected with AT₁-R mRNA expression viruses were treated with 200 µg/ml G418 for 7-10 days. Experiments were then performed on such recombinant cell batches over the succeeding 3-5 passes through subculture, representing studies on cells that have gone through approximately 12-20 population doublings after retroviral infection.

Plasmid Construction-- The plasmid pXF42-eGFP was created by removing the neomycin cassette in pXF40-eGFP (10) using BclI and BstBI and first replacing it with a multicloning site using 5'GATCAGGCCTTGTAGGCCTAGGCTCGAGGCCTACAAGGCCTT and its reverse complement (SfiI restriction sites are underlined). The tetracycline transactivator (tTA) cDNA from pUD15-1 (23) was excised with BamHI and EcoRI, blunted, and adapted with the oligonucleotide pair 5'AGGCCTAG and 5'CTAGGCCTACA before cloning into these SfiI sites. All retroviral AT₁-R expression plasmids were constructed in pTJM12, a derivative of pTJM9 (25) from which the internal CMV promoter was removed using BamHI and HindIII before blunting and self-ligation. To create pTSO33, which is the wild-type AT₁-R mRNA expression vector, eight copies of the tetracycline operator sequence (tetOp) derived from pUD10-3 (23) as a XhoI-StuI fragment were fused upstream of a PstI-KpnI fragment containing the minimal (54 to +104 base pair) rat AT_1a-R gene promoter (26). A KpnI-DraIII fragment was fused downstream of this, derived from the vector pTSO31 (3), providing rat vascular AT₁-R cDNA sequence (27) between base pairs 104 and 849, wherein a hemagglutinin (HA) epitope was inserted at the fourth amino acid of the receptor open reading frame. A 1.8-kb DraIII-StuI fragment derived from a subclone of the rat AT₁-R genomic clone RGL8 was cloned downstream from this (26). This latter fragment encodes the remainder of the third exon of the rat AT₁-R gene, including its open reading frame and 3'-UTR region and ~370 base pairs of additional non-transcribed genomic sequence that encode downstream polyadenylation signals. The vector pTSO41 uses the tetracycline-regulated minimal CMV promoter, as a XhoI-EcoRI fragment from pUD10-3, instead of the minimal AT₁-R promoter, but is otherwise identical to pTSO33 downstream from this. The vector pTSO31 is the same as pTSO41 at the 5' end but uses an SV40 pA+ signal, rather than AT₁-R pA+ signals as transcriptional terminators (3). Plasmids for the various mutant and deletion AT₁-R mRNA were constructed using shuttle vectors and standard oligonucleotide site-directed mutagenesis or polymerase chain reaction protocols. The plasmid pXF40-eGFP (10) served as a base to construct chimera AT₁-R/eGFP mRNAs, and unlike the AT₁-R mRNA constructs, these were infected into cells that express the tTA through infection with the retrovirus pTSO5 (3). All mutations and polymerase chain reaction-generated fragments were verified by sequencing using the Emory University DNA sequencing core facility before subcloning into the retroviral expression vectors. Plasmids and their complete sequences are available on request.

Ribonuclease Protection Assays and Data Analysis-- Samples of total RNA were hybridized with a mixture of gel-purified, [-³²P]UTP-labeled antisense AT₁-R and cyclophilin (Cyp) riboprobes and treated with RNase A and T1 before resolution on sequencing gels as described in detail elsewhere (3). All AT₁-R riboprobes incorporate sequences antisense to the HA epitope. The hybridization signals were quantified from phosphorimages using a volume integration protocol in the ImageQuaNT software. AT₁-R mRNA hybridization signals were divided by those for Cyp in each sample, and the data are normalized to this ratio in untreated control samples, which is taken as 100%. Decay rate constants were calculated after combining all sets of experimental data for a given mRNA construct and protocol and performing an iterative best-fit computation using the single-site exponential decay function in GraphPad Prism.

Nuclear Run-on Assays-- Nuclei from confluent RASM cells were isolated, and in vitro run-on transcription assays using 250 µCi of [³²P]UTP (3000 Ci/mmol) were conducted in a 200-µl volume as described previously (8). Radiolabeled nuclear RNA was isolated and split into 2 equivalent aliquots and each hybridized in solution at 22 °C precisely as described (28), using either 0.21 nmol of biotinylated antisense HA epitope oligonucleotide (5'biotin-(18 carbon spacer)-AACGGCGTAGTCTGGGACGTCGTATGGGTAC) or 0.21 nmol of oligonucleotide antisense to the L32 ribosomal protein mRNA (5'biotin-(18 carbon spacer)-TTCACATATCGGTCCGACTGGTGCCTGATGA) as an internal control. After an overnight hybridization, streptavidin-coated ferrous beads (Promega, Inc. Madison, WI; 300 µg, ~1.25 nmol capacity) were added. After a brief incubation, the samples were clamped against a magnetic bar and washed six times with 1-ml aliquots of RNase-free 0.1× SSC. The samples were eluted in a sequencing gel loading dye by heating to 90 °C for 10 min before resolution by electrophoresis through 5% polyacrylamide sequencing gels.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Expression of recombinant AT₁-R mRNAs in vascular smooth muscle cells from tetracycline-regulated promoters exploits a helper virus-free, non-replicating retroviral vector system (Fig. 1). The gene transfer efficiency is routinely close to 100% using this, allowing for rapid derivation of new recombinant cell populations compared with alternative methods. Stable chromatin integration of the recombinant genes allows for continued study on successive passages of cells following a single infection protocol. Fluorescence-activated cell sorting (10) was used first to create vascular smooth muscle cell lines enriched for expression of the tetracycline transactivator (23) using the self-regulating vector pXF42-eGFP (Fig. 1a). After expansion and storage, these cells were used to provide a uniform genetic background for subsequent transduction with recombinant AT₁-R retroviruses.

View larger version (59K): [in this window] [in a new window]   Fig. 1.   Recombinant AT1-R mRNA expression strategy. a depicts the LTR region in the tTA expression vector pXF42-eGFP. b depicts the LTR region of the wtAT1-R mRNA expression vector pTSO33. c is a phosphorimage from a Northern hybridization experiment (10 µg of total RNA/lane from cells treated for 72 h with (+) or without () Antet) using a double-stranded AT1-R cDNA probe, labeled with [32P]dCTP as described previously (8).

The vector used to express what is termed the wild-type recombinant AT₁-R mRNA (referred herein as wtAT₁-R), which represents the most accurate possible mimic of the native transcript, is shown in Fig. 1b. This anhydrotetracycline (Antet)-suppressible mRNA begins and ends, respectively, with transcriptional initiation (minimal promoter) and termination (polyadenylation) sequences derived from the native AT₁-R gene (26). In using these elements, rather than heterologous promoter and poly(A)⁺ sequences, the composition of the 5' and 3' ends of the recombinant mRNA precisely mimic those of the major native 2.2-kb AT₁-R transcript expressed in VSMC and other cells (27). An in-frame hemagglutinin (HA) epitope within the amino terminus of the AT₁-R open reading frame is the only difference between the recombinant wtAT₁-R and the native transcript, which was necessary to distinguish between them using ribonuclease protection assays.

A7r5 smooth muscle do not to express the AT₁-R gene natively (3), but after sequential infection with tTA and wtAT₁-R mRNA viruses, they display an anhydrotetracycline (Antet)-suppressible 2.2-kb AT₁-R mRNA (labeled "tetOp"). It migrates on agarose gels the same as the more abundant native transcript in a second line of cells termed RASM (Fig. 1c), demonstrating that the expression vector produces the expected AT₁-R mRNA product. Notably, this is slightly smaller than the 2.4-kb transcript produced from the vector TSO31 (see below), which has additional heterologous 5'- and 3'-UTR sequences derived from CMV and SV40 elements (3). The recombinant wtAT₁-R mRNA presumably migrates at 2.2 kb in recombinant RASM cells, inferred from the finding that none of the detectable hybridization signals are suppressed by Antet treatment. Of note, the weaker 3.2-kb transcript observed in RASM mRNA represents a previously cloned alternately spliced product of the native AT₁-R gene, rather than a recombinant molecule (26, 27). The barely detectable hybridization signals migrating at ~6 kb, labeled "LTR," likely represent the transcript driven by the LTR promoter, essentially reflecting a neomycin mRNA with antisense AT₁-R mRNA sequences in the 3'-untranslated region (Fig. 1b).

Our ribonuclease protection assay discriminates recombinant from native transcripts in a single sample, because the riboprobes used incorporate sequence antisense to the HA epitope in addition to the flanking AT₁-R mRNA sequence (Fig. 2). The phosphorimage (Fig. 2a) shows a single large protected fragment representative of the recombinant wtAT₁-R mRNA. The three smaller fragments reflect the native mRNA because synthesis of the riboprobe from a recombinant template crosses over both the heterologous HA epitope and an exon that is alternatively spliced within the native mRNA 5'-UTR (26, 27). Although the recombinant transcript is suppressed after treatment with 1 µg/ml Antet, this does not affect expression of the native AT₁-R mRNA, showing the selectivity of Antet (Fig. 2b). A comparatively delayed decay occurs for the wtAT₁-R in cells treated with actinomycin D (20 µg/ml) relative to the effect of Antet, indicating that actinomycin D strongly interferes with basal AT₁-R decay processes.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Antet selectively causes recombinant AT1-R mRNA decay. a, representative RPA phosphorimage of RASM cells expressing both the recombinant wtAT1-R mRNAs, which were treated for the indicated times with either Antet (1 µg/ml) or actinomycin D (20 µg/ml). Protected fragments representing the recombinant transcript (wtAT1-R) and the endogenous transcript (native AT1-R) are noted. Cyp, cyclophilin mRNA internal control. b, quantified effects of Antet (closed symbols) or actinomycin D (open symbols) on the recombinant wtAT1-R (circles) or the native transcript (triangles) are quantified from three experiments (mean ± S.E.). Control represents Cyp-normalized AT1-R mRNA hybridization volumes in untreated cells.

Previous studies have established that forskolin, a direct activator of adenylyl cyclase, down-regulates the AT₁-R mRNA through activation of PKA, because this response is inhibited in cells expressing a highly selective polypeptide inhibitor of the cAMP-dependent kinase (10). A bolus (5 or 10 µM) of forskolin was used to stimulate PKA in RASM cells, and recombinant wtAT₁-R mRNA levels were measured at various times thereafter (Fig. 3). Treatment with forskolin causes down-regulation of the mRNA, with an apparent half-life of 1.7 h (Table I). Notably, decay rates following Antet treatment or Antet plus forskolin treatment are similar (Table I).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Forskolin reduces steady-state recombinant AT1-R mRNA expression. a, phosphorimage of a representative RPA result on RASM cells expressing the wtAT1-R mRNA, which were treated for the indicated times after adding either forskolin (10 µM), Antet (1 µg/ml), or forskolin and Antet combined. b, data quantified from three experiments (mean ± S.E.). Control represents the ratio of AT1-R to Cyp hybridization volumes in untreated cells.

Because Antet suppresses transcription by a mechanism that relieves transactivation, rather than through direct inhibition (23), nuclear run-on assays were performed to ensure that forskolin does not directly inhibit transcription of the wtAT₁-R gene. Nuclei were collected after treating cells for 45 min with forskolin, reasoning that any transcriptional suppression caused by forskolin by this point should be quite pronounced to account for its affect on steady-state mRNA levels. For comparison, cells were also treated with Antet to assess transcriptional suppression and with Antet plus forskolin to test whether forskolin interferes with the mechanism of action for Antet. The recombinant AT₁-R transcripts were selectively captured by solution hybridization using a biotinylated antisense oligonucleotide directed against its HA epitope (28). Capture of the ribosomal L32 protein mRNA similarly from aliquots of the same samples served as an internal control. Antet treatment for 45 min suppresses the transcripts captured by the wtAT₁-R probe, with transcription to about 10% of control levels consistent with the "leak" that occurs in this system (3) (Fig. 4). Similar results were obtained in other run-on experiments in which the Antet treatment period was 30 min, suggesting Antet-mediated transcriptional suppression occurs rapidly (data not shown). Co-treatment with forskolin and Antet does not interfere with the transcriptional suppression, and forskolin has no effect on transcription compared with vehicle-treated cells. These observations establish that transcriptional suppression cannot account for the forskolin-mediated reduction of the recombinant wtAT₁-R mRNA.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Forskolin does not inhibit recombinant AT1-R gene transcription. RASM cells expressing the wtAT1-R mRNA were treated for 45 min with Antet, Antet plus 10 µM forskolin, or forskolin alone before harvesting nuclei and collecting run-on transcripts using [32P]UTP. wtAT1-R and L32 mRNAs were captured from each sample using biotinylated oligonucleotides and resolved on sequencing gels. Each bar represents an L32-normalized wtAT1-R transcript signal quantified using phosphorimage analysis, expressed as a percent of the signal in vehicle-treated cells. Similar results were obtained for each treatment in 1-3 other experiments.

To determine if mitogenic receptor signaling affects recombinant wtAT₁-R mRNA expression, the cells were stimulated for up to 6 h following a bolus of either angiotensin II or the growth factor PDGF-BB. As shown in Fig. 5a, their effects are not comparable to that of forskolin, wherein angiotensin II has no effect and PDGF reduces the recombinant mRNA only at the latest time point tested. Whereas the recombinant mRNA is largely unaffected by mitogens, the native AT₁-R mRNA is down-regulated in the same samples, which establishes that the agonists nevertheless evoked their regulatory effects on AT₁-R gene expression (Fig. 5a). The modest suppressive effect of PDGF on the recombinant wtAT₁-R mRNA at the latter time point might be due to a transcriptional effect. To test this, the PDGF response was measured in cells expressing a recombinant mRNA from the vector TSO41, which is identical to the wtAT₁-R mRNA except that it is driven from a tetracycline-suppressible minimal CMV promoter. This renders an mRNA with ~60 additional bases of heterologous CMV sequence in the 5'-UTR. As shown in Fig. 5b, the TSO41-derived recombinant AT₁-R mRNA is down-regulated by forskolin but unaffected by PDGF-BB treatment. We conclude that the reduction of the recombinant wtAT₁-R mRNA in response to PDGF-BB is most likely due to a modest degree of transcriptional suppression at the minimal AT₁-R promoter.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   No immediate effect of mitogens on steady-state recombinant wtAT1-R mRNA. a, RASM cells expressing recombinant wtAT1-R mRNA were cultured for the indicated times after adding 100 nM angiotensin II (closed symbols) or 50 ng/ml PDGF-BB (open symbols) before quantifying the recombinant wtAT1-R (triangles) or native AT1-R (circles) mRNAs from phosphorimages. b, the effect of PDGF-BB (open circles) or forskolin (closed circles) on a full-length recombinant mRNA (TSO41) transcribed from a tetracycline-regulated minimal CMV promoter. The control value is the ratio of AT1-R to Cyp hybridization volumes in untreated cells. Each point represents the mean ± S.E. of three experiments.

To test whether the AT₁-R mRNA is regulated in another smooth muscle cell phenotype, levels of the recombinant wtAT₁-R mRNA levels were measured after infection of A7r5 cells with the display system. Like RASM, A7r5 cells are derived from rat thoracic aorta (29) but differ in morphological and growth characteristics (30).² As shown in Fig. 6, 8-h treatments with either Antet or forskolin reduce wtAT₁-R mRNA expression in each cell line. In contrast, angiotensin II has no effect in either cell line, and PDGF-BB treatment only reduces the mRNA when expressed in RASM cells. Thus, both cell lines appear to support PKA-specific down-regulation.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   PKA-specific down-regulation of recombinant AT1-R mRNA in two VSMC lines. Recombinant wtAT1-R mRNA levels were measured by quantitative RPA in either A7r5 cells (a) or RASM cells (b) following an 8-h treatment with either 1 µg/ml Antet, 100 nM angiotensin II, 10 µM forskolin, or 50 ng/ml PDGF-BB. RPA hybridization signals were quantified by phosphorimage analysis, where the control value is the ratio of AT1-R to Cyp hybridization volumes in vehicle-treated cells. The data are from three experiments.

We compared four possible combinations of minimal AT₁-R or minimal CMV promoters and AT₁-R or SV40 polyadenylation signals to determine if heterologous sequences influence forskolin responsiveness or intrinsic decay rates. These nominally full-length mRNAs (wtAT₁-R, KX38, TSO41, and TSO31) show generally equivalent Antet-mediated decay rates, although forskolin-stimulated down-regulation of the TSO31 mRNA is slightly delayed compared with the others (Fig. 7; see also Table I). Most significantly, each is down-regulated by forskolin treatment, and like the wtAT₁-R mRNA, these each respond to co-treatment with forskolin and Antet without an accelerated decay compared with Antet alone. Because a combination of minimal AT₁-R gene promoter and SV40 poly(A)⁺ signals (KX38) share response characteristics with the wtAT₁-R mRNA, a series of 3'-UTR deletion mutants were expressed using SV40 poly(A)⁺ signals for termination. As seen in Fig. 7, these mutants show a tendency for increasingly slower rates of decay as a function of deletion, following Antet treatment. This suggests elements within the 3'-UTR confer basal decay characteristics. Again most notably, forskolin treatment alone causes down-regulation of each construct, indicating the 3'-UTR does not mediate this response to PKA activation. Furthermore, co-treatment with forskolin and Antet does not lead to an accelerated rate of decay for any of these eight mRNAs compared with the effect of Antet alone. These results indicate that the 3'-UTR does not mediate PKA-stimulated down-regulation (Table I).

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Heterologous and 3'-UTR mutants are down-regulated by forskolin. Cells were treated with Antet, forskolin, or Antet plus forskolin for the indicated times before quantifying recombinant mRNA levels. Upper panels, results for the various combinations of promoter and polyadenylation signals were used to express recombinant AT1-R mRNAs. Lower panels, a series of deletions from the 850-base long 3'-UTR. The deletions are as follows: TSO32, 200 bases; TSO28, 300 bases; TSO29, 400 bases; KX39, 850 bases. The SV40 poly(A)+ signals supply a short 3'-UTR in these constructs.

We next tested if the 271-base long AT₁-R mRNA 5'-UTR determines responsiveness to PKA signaling. Deletion of the 5'-UTR from the wtAT₁-R mRNA background making KX30 completely impairs forskolin-stimulated down-regulation, and the down-regulation by Antet shows this is not attributable to intrinsic stabilization of the mRNA (Fig. 8, see also Table I). Furthermore, forskolin clearly evoked the regulatory process in these cells as shown by down-regulation of the internal control native mRNA in the same samples (Fig. 8). Because this mRNA does not have any predictable 5'-UTR sequence, we also deleted it within the context of the KX38 mRNA background to make KX63. This would have an ~60-base heterologous 5'-UTR remnant derived from the CMV minimal promoter region. The KX63 mRNA is also insensitive to forskolin yet is not intrinsically stabilized (Fig. 9, see also Table I)). To control for length effect, the 5'-UTR was inverted to make KX65, and this is also shown to be insensitive to forskolin signaling but also not intrinsically stabilized (Fig. 9).

View larger version (41K): [in this window] [in a new window]   Fig. 8.   The 5'-UTR is necessary for forskolin response in the wtAT1-R background. a, representative result from three experiments showing hybridization signals for both the 5'-UTR deletion mutant (-5') and the native transcript (nat) after treatment with forskolin, Antet, or both drugs combined. b, depiction of construct showing use of minimal AT1-R promoter and poly(A)+ signals and quantified data from phosphorimages.

View larger version (22K): [in this window] [in a new window]   Fig. 9.   5'-UTR orientation dependence and conference of responsiveness to a more stable mRNA. Upper panels, AT1-R mRNAs in which the 5'-UTR is deleted (KX63) or inverted (KX65) are refractory to forskolin (5 µM) treatment compared with their control (TSO41), yet all have essentially similar Antet decay characteristics. Because the reverse 5'-UTR orientation in the wtAT1-R mRNA background did not express a detectable mRNA, the TSO41 background with its minimal CMV promoter was used instead. Lower panels, Antet- and forskolin-initiated decay kinetics for a control eGFP mRNA (XF40) and chimeras bearing the AT1-R 5'-UTR in either the forward (KX41) or reverse (KX41()) orientations. Each point represents data from two to four experiments.

We next asked if the 5'-UTR region could confer PKA-mediated regulatory characteristics upon a heterologous mRNA. It was next cloned upstream of an eGFP coding region in either the forward (KX41) or reverse (KX41()) orientations and compared with a control eGFP mRNA (XF40). The XF40 mRNA is unaffected by forskolin treatment but decays with a half-life of ~5.4 h after Antet treatment, representing the most stable mRNA used in this study (Fig. 9; see also Table I). Forskolin treatment causes down-regulation of the forward orientation 5'-UTR AT₁-R/eGFP mRNA chimera but not that with a reverse orientation 5'-UTR. Intriguingly, the forward 5'-UTR chimera but not the reverse also acquires the more rapid basal decay kinetics associated with AT₁-R mRNA, which we suspect may be attributable to basal levels of PKA activity. Together, these observations indicate that the 5'-UTR region functions as a cis-acting element that directs post-transcriptional down-regulation of AT₁-R mRNA expression in response to PKA activation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Difficulties in deriving direct and unambiguous evidence that specific mRNAs are susceptible to regulated post-transcriptional control has impeded progress in developing an understanding of this potentially important interface between signaling and gene expression. A careful reconstitution of the normal cellular environment for an mRNA ensures that a suspected regulatory circuitry remains intact and provides a stringent approach to test this concept. Current "circular" models of mRNA structure argue that the 5' and 3' ends of an mRNA are co-engaged in a complex of interacting proteins linking the 5'-cap structure with the 3' poly(A)⁺ tail (reviewed in Ref. 31). Reasoning that post-transcriptional regulation of the AT₁-R mRNA might very well depend upon accurate formation or function of these or other context-specific complexes, which may also be cell type-specific, we were cautious to recreate a recombinant mRNA as precisely as possible. Coupled with the comparatively selective tetracycline-regulated system, as opposed to general transcriptional inhibitors, and expressing these recombinants in their native cell system, the most accurate possible reconstitution of the normal AT₁-R mRNA regulatory environment has been achieved. From the behavior of the recombinant mRNAs in such a system, solid inferences can be drawn about how the native AT₁-R mRNA is regulated.

A primary goal of this study was to establish a direct functional test for whether post-transcriptional regulation contributes to vascular smooth muscle AT₁-R mRNA down-regulation. Finding this, a secondary goal was to ascertain which of the receptor signaling systems that down-regulate AT₁-R gene expression do so through control by post-transcriptional mechanisms. Three lines of evidence developed in this study indicate a PKA-regulated post-transcriptional process targets the vascular AT1-R mRNA. First, the experimental design is predicated on recapitulating a reduction in steady-state expression of the recombinant wtAT₁-R mRNA should a stimulus evoke a post-transcriptional regulatory process. We show this not only for the stringently designed wtAT₁-R mRNA but also with great consistency on seven additional independent recombinant AT₁-R transcripts (see Fig. 7). The second line of evidence is that nuclear run-on analysis shows that the cAMP-mediated effect is not attributable to interference with tTA-mediated transactivation of the recombinant gene. We found this important to evaluate because of the formal possibility that cAMP signaling might unexpectedly inhibit the tTA protein or its mechanism of action. Third, both loss of function and gain of function mutants establish that the 5'-UTR is an obligate component of PKA-mediated AT₁-R mRNA down-regulation. Acquired from analysis of five independent recombinant mRNAs, they argue quite strongly that the response to PKA signaling is an intrinsic feature of the mRNA molecule. Taken together, these observations provide the first direct and largely unambiguous functional evidence that this post-transcriptional process exists.

This study significantly extends previous findings that predicted post-transcriptional control by PKA signaling and clarifies the mechanism of AT₁-R mRNA down-regulation caused by mitogenic signaling (4, 6, 7, 10). The post-transcriptional response has features of inherent specificity, in that recombinant mRNA down-regulation occurs following cAMP but not mitogenic signaling. This is consistent with previous nuclear run-on observations, wherein AT₁-R gene transcription is suppressed some 50-80% by mitogenic signaling, but has never been observed in response to cAMP signaling (4, 6, 8). We infer from this that mitogens likely down-regulate the endogenous transcript predominantly, if not entirely, by a mechanism that involves transcriptional suppression rather than post-transcriptional control. The present direct, function-based approach and the acquisition of reliable AT₁-R mRNA half-life estimates made possible by using the tetracycline-regulated system highlights previous uncertainties that have plagued us regarding mechanisms of mitogen-induced down-regulation. Despite the evidence showing mitogen-induced transcriptional suppression, the more rapid rate of AT₁-R mRNA down-regulation caused by mitogens compared with actinomycin D or DRB treatment raised the possibility of post-transcriptional regulation because the mRNA did not seem capable of decaying fast enough. We now show just how inaccurate the toxin chase-based half-life estimates were. By using Antet instead, a <2-h half-life shows that a significant reduction in steady-state mRNA can follow inhibiting AT₁-R gene transcription, even if only partially. Additionally, the failure of angiotensin II treatment to reduce levels of the recombinant wtAT₁-R mRNA raises qestions about the relevance of previously described angiotensin II-stimulated AT₁-R mRNA binding protein activity (7). They now seem unlikely to reflect factors involved in post-transcriptional regulation.

We are cautious not to describe this post-transcriptional response as PKA-regulated AT₁-R mRNA "destabilization." Of the several mRNAs created for this study, none are reduced at a faster rate by combined treatment with forskolin and Antet, compared with transcriptional suppression alone. Destabilization or "regulation of mRNA stability" implies a different kinetic behavior than this. As the evidence shows that a regulated post-transcriptional mechanism must account for down-regulation of the mRNA caused by forskolin, we speculate either of two general mechanisms could explain the behavior of the AT₁-R mRNA. One possibility is that PKA stimulates AT₁-R mRNA degradation but through a mechanism that functions upstream of the mRNA decay processes. For example, PKA may prematurely "uncouple" pre-existing mRNA from its normal state and shuttle it into a degradation process wherein decay per se is rate-limiting. Thus, mRNAs would show equivalent decay kinetics irrespective of whether they enter the degradation process through a default or PKA-regulated process.

A more attractive possibility is that separate pools of the AT₁-R mRNA exhibit differential sensitivity to PKA signaling. By the following logic, our data might be more consistent with this latter speculation. At steady state, the rates of transcript formation and degradation must be equivalent. The similarity between Antet and forskolin-stimulated decay rates are consistent with a common failure to regenerate the cellular pool. The pool is not refilled following Antet obviously because transcription is suppressed. Because we show that forskolin does not affect transcription, we speculate that PKA might activate a process that selectively targets only newly transcribed AT₁-R mRNA molecules. This would have the kinetic appearance of failing to regenerate the cellular pool much like for Antet but through a distinct mechanism. A period might exist in which the mRNA is susceptible to this PKA-regulated process. At latter points in its maturation pathway, it either acquires a defense against it or is no longer associated with the components that PKA presumably regulates. Testing this notion will involve developing schemes that selectively measure nascent mRNAs and distinguish them from more mature mRNAs.

The trans-acting iron-responsive element mRNA-binding proteins connect both iron availability and intracellular oxide concentrations to the stability and translation of specific mRNAs by interacting with specific mRNA cis-acting elements (21, 32-34). To the extent that oxidative metabolites serve as second messengers (35), the iron response element-binding proteins remain one of the few unambiguous interfaces between mRNA stability regulation and signal transduction or changes in the extracellular environment.

Other models systems for the concept that the more traditional receptor-regulated signaling pathways control post-transcriptional processes are not without ambiguity. Interestingly, several of these describe cAMP-regulated processes, including down-regulation of adrenergic receptor mRNAs (36-39) and a plasminogen activator inhibitor mRNA (40, 41) and up-regulation of a lactate dehydrogenase mRNA (42, 43). Concomitant transcriptional regulation by the stimuli that affect steady-state levels of these mRNA and the various toxin-based pulse-mRNA chase paradigms used in these studies are the basis for most uncertainty regarding the role of post-transcriptional processes. Nevertheless, they have compelling biochemical findings that largely correlate mRNA protein binding activity and mRNA structure relationships with responses to stimuli. Curiously, they all implicate 3'-UTR adenylate/uridylate (AU)-rich domains as signaling responsive mRNA elements. On this basis, they seem to differ fundamentally from those likely affecting the AT₁-R mRNA, because its 5'-UTR is devoid of AU-rich elements. Also, because AU-rich elements function as powerful epitopes driving the intrinsic decay of many highly unstable mRNAs (44), it remains unclear how they might function as conditional stability determinants responsive to cAMP signaling or other pathways (45, 46).

In summary, this study has general importance by showing a stringent approach useful for understanding roles of post-transcriptional processes in regulation of mRNA abundance. More specifically, it clarifies the general mechanisms that are involved in agonist-induced down-regulation of vascular smooth muscle AT₁-R gene expression. Future studies will be directed at achieving improved insight into the factors that mediate this process. We also hope to gain some insight into how it contributes to the well known plasticity of smooth muscle cell gene expression and its relevance to the generally suppressive role of PKA in smooth muscle cell proliferation and migration (47).

	ACKNOWLEDGEMENTS

We thank Rick Bright, Tracy Obertone, Xiaofei Wang, and Brian Adams for their preliminary contributions to this study.

	FOOTNOTES

^* This work was supported by NHLBI Grants HL52810 and HL56107 from the National Institutes of Health.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Established Investigator of the American Heart Association. To whom correspondence should be addressed: Dept. of Pharmacology, Emory University, 5031 O. W. Rollins Research Bldg., Atlanta, GA 30322. Tel.: 404-727-2467; Fax: 404-727-0365; E-mail: medtjm@bimcore.emory.edu.

² K. Xu and T. J. Murphy, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: AT₁-R, type I angiotensin II receptor; DRB, 5,6-dichloro-1--D-ribofuranosylbenzimidazole; UTR, untranslated region; Antet, anhydrotetracycline; eGFP, enhanced green fluorescent protein; wtAT₁-R, a recombinant AT₁-R mRNA mimic; PKA, cAMP-dependent protein kinase A; VSMC, vascular smooth muscle cells; RASM, rat aortic vascular smooth muscle; tTA, tetracycline transactivator; kb, kilobase pair; HA, hemagglutinin; CMV, cytomegalovirus; Cyp, cyclophilin; LTR, long terminal repeat; PDGF, platelet-derived growth factor; RPA, ribonuclease protection assay.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. Xu, C. M. Kitchen, H.-K. G. Shu, and T. J. Murphy Platelet-derived Growth Factor-induced Stabilization of Cyclooxygenase 2 mRNA in Rat Smooth Muscle Cells Requires the c-Src Family of Protein-tyrosine Kinases J. Biol. Chem., November 9, 2007; 282(45): 32699 - 32709. [Abstract] [Full Text] [PDF]

				M. M. Martin, J. A. Buckenberger, J. Jiang, G. E. Malana, G. J. Nuovo, M. Chotani, D. S. Feldman, T. D. Schmittgen, and T. S. Elton The Human Angiotensin II Type 1 Receptor +1166 A/C Polymorphism Attenuates MicroRNA-155 Binding J. Biol. Chem., August 17, 2007; 282(33): 24262 - 24269. [Abstract] [Full Text] [PDF]

				T. S. Elton and M. M. Martin Angiotensin II Type 1 Receptor Gene Regulation: Transcriptional and Posttranscriptional Mechanisms Hypertension, May 1, 2007; 49(5): 953 - 961. [Full Text] [PDF]

				M. F. Doobay, L. S. Talman, T. D. Obr, X. Tian, R. L. Davisson, and E. Lazartigues Differential expression of neuronal ACE2 in transgenic mice with overexpression of the brain renin-angiotensin system Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2007; 292(1): R373 - R381. [Abstract] [Full Text] [PDF]