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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Searles, C. D. Articles by Ramasamy, S. Search for Related Content PubMed PubMed Citation Articles by Searles, C. D. Articles by Ramasamy, S. Related Collections Endothelium/vascular type/nitric oxide Angiogenesis Gene regulation

(Circulation Research. 1999;85:588-595.)
© 1999 American Heart Association, Inc.

Molecular Medicine

Posttranscriptional Regulation of Endothelial Nitric Oxide Synthase During Cell Growth

Charles D. Searles, Yoichi Miwa, David G. Harrison, Santhini Ramasamy

From the Division of Cardiology, Department of Medicine, Emory University School of Medicine, Atlanta, Ga.

Correspondence to Santhini Ramasamy, PhD, and reprint requests to Charles D. Searles, MD, Division of Cardiology, Department of Medicine, Emory University, 1639 Pierce Dr, Atlanta, GA 30322. E-mail sramasamy@issiinc.com (S.R.) or csearle{at}emory.edu (C.D.S.)

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract—The expression of the endothelial NO synthase (eNOS) is dramatically influenced by the state of cell growth. In proliferating cells, mRNA levels are increased 4-fold compared with postconfluent, nonproliferating cells. Nuclear run-on analysis indicated that there is no difference in the transcriptional rate of eNOS in proliferating versus postconfluent cells. The half-life of eNOS mRNA, measured after actinomycin D transcriptional arrest, was 3-fold greater in preconfluent compared with confluent endothelial cells. Using UV–cross-linking analysis, a cytoplasmic protein with an apparent molecular mass of 51 kDa was found to bind to terminal 545-nt eNOS mRNA 3-fold more in confluent cells than in proliferating cells. Further characterization of the eNOS mRNA indicated that a 43-nt sequence at the origin of the 3'-untranslated region (UTR) is critical in binding of this protein. Endothelial cells were stably transfected with a chimeric cDNA plasmid containing chloramphenicol acetyl transferase (CAT) ligated to the eNOS coding region and either the wild-type 3'-UTR (pcDNACAT/eNOS_(wtUTR)) or a mutant 3'-UTR lacking the 43 nt found to bind the 51-kDa protein (pcDNACAT/eNOS_(UTR)). The CAT/eNOS mRNA half-life was dramatically stabilized in these latter cells as compared with cells transfected with pcDNACAT/eNOS_(wtUTR)). Thus, this 43-nt region plays a critical role in destabilizing eNOS mRNA. These studies demonstrate a mechanism for modulation of eNOS expression during cell growth via interactions between the proximal 3'-UTR and a novel 51-kDa cytosolic protein.

Key Words: mRNA stability • proliferation • RNA binding protein • 3'-untranslated region • gene regulation

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Nitric oxide produced by endothelial NO synthase (eNOS) promotes vasodilation,¹ inhibits smooth muscle cell growth,² diminishes platelet aggregation,³ and decreases leukocyte adhesion.⁴ Although eNOS is constitutively expressed, it has become clear that its expression is subject to modest, but likely important, degrees of regulation. In cultured cells, eNOS expression is increased by shear stress,⁵ cyclic strain,⁶ lysophosphatidylcholine,⁷ low concentrations of oxidized LDL,⁸ oxidized linoleic acid,⁹ and 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors.¹⁰ In vivo, exercise training in dogs increases eNOS expression and the ability of blood vessels to dilate in response to agonists that release endogenous nitric oxide (NO·).¹¹ In contrast, exposure of cultured endothelial cells to tumor necrosis factor (TNF)-,¹² hypoxia,¹³ and high concentrations of oxidized LDL^{8 14} decreases eNOS levels. In these latter conditions, posttranscriptional changes in mRNA half-life play an important role in downregulation of eNOS.^{12 13 14}

Earlier, our laboratory found that eNOS expression is rather dramatically influenced by the state of endothelial cell proliferation. The mRNA for eNOS was 4- to 6-fold greater in proliferating cells as compared with cells several days after confluence.¹⁵ In addition, protein expression (by Western analysis) and enzyme activity (by arginine-to-citrulline conversion) were similarly increased in proliferating versus confluent cells.¹⁵ In this study, the mechanisms responsible for the increase in eNOS mRNA expression during cell growth were not defined. Thus, the purpose of the present study was to examine factors responsible for changes in eNOS levels during endothelial cell proliferation. Nuclear run-on assays and studies of mRNA stability were used to examine both transcriptional and posttranscriptional mechanisms, respectively. On the basis of these results, we further examined the role of protein/RNA interactions and the role of specific portions of the 3'-untranslated region (UTR) of the eNOS mRNA in regulation of eNOS expression during cell growth. Our findings demonstrate a novel mechanism for control of eNOS levels in endothelial cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cell Culture
Bovine aorta endothelial cells (BAECs) were harvested and cultured in medium (M199; Gibco Laboratories) containing 10% FCS (Hyclone Laboratories) as described earlier.⁹ Experiments were conducted on cells between passages 3 and 9. Cells were split at a ratio of 1:6. Preconfluent cells were studied 24 hours after splitting. Cells generally reached confluency 3 to 4 days after splitting and were studied at least 3 days after the cells made uniform cell-to-cell contact as judged by visual analysis (6 to 7 days after splitting).

Northern Blotting
Northern analysis was performed as previously described.⁹ For identification of endogenous eNOS mRNA, a ³²P-labeled full-length eNOS cDNA was used. For studies of cells transfected with the chimeric constructs pcDNACAT/eNOS_(wtUTR) and pcDNACAT/eNOS_(UTR), a ³²P-labeled 270-bp probe directed to the 5' region of the chloramphenicol acetyl transferase (CAT) was used. In studies in which mRNA half-life was measured, cells were treated with actinomycin D (2.5 µg/mL) and harvested at the times indicated.

Nuclear Run-On Assay
Nuclear run-on assays were performed using a method described by Greenberg¹⁶ with modifications. Identical numbers of nuclei from preconfluent and confluent cells were used for preparation of nascent transcripts. Nuclei (5x10⁷) were incubated in a reaction buffer containing, in mmol/L, Tris HCl (pH 8.0) 5; MgCl₂ 2.5; KCl 150; and ATP, GTP, and CTP 2 each, as well as 100 µCi of [-³²P]UTP (DuPont NEN) for 30 minutes at 30°C. RNA was extracted using TRI reagent (Molecular Research Center Inc) and hybridized with Zeta-Probe GT membrane (Bio-Rad) previously slot-blotted with eNOS, human ß-actin cDNA, and 2.9-kb pBluescript+. Blots were imaged as described for Northern analysis.

Preparation of RNA probes
The 3'-UTR of eNOS was excised from the full-length cDNA using AccI and EcoRI and inserted into pBluescript 3' to the T7 promoter. The plasmid was linearized with EcoRI, and a 545-nt riboprobe was generated using a T7 polymerase that represented the entire 3'-UTR and 96 nt of the coding region. To obtain a 35-nt probe corresponding to the AU-rich region of the eNOS 3'-UTR (3968 to 4002), oligonucleotides (sense, 5'-CATTCAAATATTTATTATTGAAG -ATTTACCATAAG-3', and antisense, 5'-CTTATGGTAAATCT-TCAATAATAAATATTTGAATG-3') were synthesized and used to produce a double-stranded DNA sequence with Kpn and EcoRI restriction sequences added to allow directional insertion into pBluescript. The ligated plasmid was linearized with EcoRI, and T7 polymerase was used to create this AU-rich element-containing riboprobe. To generate other riboprobes from both the 3'-UTR and the coding region of eNOS, polymerase chain reaction–generated templates were used as shown in Tables 1 and 2, respectively. The template for the 43-nt sense probe was obtained by synthesizing an oligonucleotide corresponding to eNOS 3648 to 3690 with T7 promoter linked at the 5' end. In vitro transcription reactions were performed using the T7 polymerase in the presence of [³²P]UTP with a commercially available kit (MAXIscript, Ambion Inc). In general, 1 µg of linearized plasmid or polymerase chain reaction–generated template was incubated with 50 µCi [³²P]UTP; 0.5 mmol/L each of ATP, GTP, and CTP; and 2.5 µmol/L of cold UTP, in 1x Ambion transcription buffer, with 10 units of the T7 polymerase. Reactions were allowed to proceed at 37°C for 30 to 60 minutes. Specific unlabeled RNA probes for use as competitors were synthesized by replacing labeled UTP with 0.5 mmol/L of cold UTP in the transcription reaction. After in vitro transcription, the probes were purified by size separation in 5% acrylamide. Riboprobes were either gel purified or extracted with phenol:chloroform.

View this table: [in this window] [in a new window]   Table 1. Primers Used for 3'-UTR Riboprobe Construction

View this table: [in this window] [in a new window]   Table 2. Primers Used for Coding Region Riboprobe Construction

UV–Cross-Linking Assay
Nuclear and cytosolic protein fractions were prepared according to the protocol by Dignam et al.¹⁷ Radiolabeled riboprobes (10⁵ to 10⁶ cpm) were incubated with 10 to 20 µg of either cytosolic or nuclear proteins for 10 minutes at 4°C in a binding buffer of the following composition (in mmol/L): HEPES (pH 7.9) 10, KCl 40, MgCl₂ 3, and DTT 2; 7.5% glycerol; 5 mg/mL heparin; and 200 µg yeast tRNA in a total volume of 24 µL. Subsequently, the reaction mixture was irradiated for 7 minutes in a UV–cross-linker, model 2400 (Stratagene). RNase A (20 µg) was then added to the reaction mixture for 30 minutes at 37°C. The samples were boiled for 3 minutes in Laemmli buffer and separated by electrophoresis on a 12% SDS-polyacrylamide gel. The gels were dried and exposed to a phosphor imager. In some experiments, either excess unlabeled probe or proteinase K was added to the binding reaction 15 minutes before adding the labeled probe.

Western Blotting
Twenty µg of cellular protein was electrophoresed using 7.5% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane. AUF-1 protein was detected using a 1:5000 dilution of immune serum against AUF-1 (a gift from Dr Gary Brewer, Department of Microbiology and Immunology, Wake Forest School of Medicine, Winston-Salem, NC) and secondary goat anti-rabbit antibody. Immunoreactive bands were detected using the enhanced chemiluminescence detection system (Amersham Corp) and quantified using densitometry.

Stable Transfection of Endothelial Cells
BAECs (passage 1) in 100-mm plates were transfected with 10 to 20 µg of either the plasmid pcDNACAT/eNOS_(wtUTR) or the plasmid pcDNACAT/eNOS_(UTR) using Lipofectin (Gibco) according to the manufacturer's directions. Stably transfected cells were selected by addition of 500 µg/mL G418 to the culture medium and expanded in the presence of G418.

Statistical Analysis
Statistical significance was assessed by unpaired t test. Data are expressed as mean±SEM. The mean data between different treatments were compared. Differences between the groups were considered significant when P<0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
eNOS mRNA Steady-State, Transcription Rate, and Half-Life in Proliferating Versus Confluent Endothelial Cells
As reported previously, steady-state eNOS mRNA levels were 4-fold higher in proliferating as compared with confluent endothelial cells (Figure 1A and 1B). To determine whether this was due to a change in eNOS transcription, nuclear run-on analyses were performed (Figure 2A). After correcting for background, the intensity of bands for eNOS nascent transcripts was identical between proliferating and confluent endothelial cells (Figure 2B). In contrast, eNOS mRNA half-life was substantially different in the 2 growth states. In confluent cells, eNOS mRNA half-life was 9 hours, whereas in proliferating cells it was 3-fold longer (Figure 3A and 3B).

View larger version (27K): [in this window] [in a new window]   Figure 1. Northern blot showing eNOS mRNA expression in preconfluent and confluent cells (A). Proliferating BAECs were harvested 24 hours after splitting at 50% confluence. Confluent cells were harvested 3 days after uniform cell-to-cell contact was made (6 to 7 days after splitting). Twenty micrograms of total RNA was size-fractionated in 1% agarose gel and transferred to nitrocellulose. The blots were hybridized with a 32P-labeled 4.1-kb eNOS cDNA. After washing, specific eNOS bands were visualized using a phosphor imager. The 28S rRNA bands were visualized using ethidium bromide staining and quantified using scanning densitometry. B, Average results from 3 separate Northern blots comparing eNOS expression in proliferating and confluent endothelial cells. Data (mean±SEM) are presented as ratio of eNOS to 28S rRNA. *P<0.05 vs confluent cells.

View larger version (25K): [in this window] [in a new window]   Figure 2. Nuclear run-on analysis comparing the rate of transcription of eNOS mRNA in preconfluent and confluent BAECs (A). Nuclei were harvested at different states of confluency. Identical numbers of nuclei from each growth state were used to elongate nascent transcripts using [32P]UTP and NTPs. The elongated transcripts were extracted and hybridized with nylon filters previously slot-blotted with 2 µg of the respective cDNAs. Specific bands were visualized using a phosphor imager. Nonspecific binding was negligible as assessed by binding to vector DNA. It should be noted that the apparent decrease in hybridization of nascent transcripts to eNOS cDNA in the blot representing confluent cells is caused by reduced background. B, Mean values for percentage changes of nascent eNOS mRNA in preconfluent cells as compared with confluent BAECs. Percentage nascent eNOS mRNA refers to relative amounts of eNOS nascent mRNA in preconfluent cells as percentage of that obtained from the paired confluent cells, after adjusting for background counts.

View larger version (24K): [in this window] [in a new window]   Figure 3. Stability of eNOS mRNA in proliferating and confluent BAECs. Endothelial cells under the 2 growth conditions were treated with actinomycin D (2.5 µg/mL) and were harvested at 0, 6, 12, and 24 hours thereafter. Endogenous eNOS mRNA levels were determined using Northern blot analysis as described for Figure 1. The amount of total RNA loaded from confluent cells was twice that used for preconfluent cells to allow visualization of eNOS bands adequately in confluent cells at later time points after actinomycin D treatment. A, Representative blots showing decay of eNOS mRNA after actinomycin D treatment. B, Combined results from 3 different experiments. Data are mean±SEM; ecNOS indicates eNOS. *P<0.05 vs confluent.

eNOS mRNA/Protein Interactions in Preconfluent and Confluent Cells
Because mRNA stability is regulated by protein/mRNA interactions, UV–cross-linking analysis was used to study RNA-protein interactions from endothelial cells of different growth states. A ³²P-labeled 545-nt riboprobe corresponding to the full-length eNOS 3'-UTR and 96 bases of the terminal eNOS coding region (bases 3552 to 4096, Table 1) was used to study binding with either nuclear or cytosolic proteins between different growth states. Two proteins were identified to bind to the 545-nt probe. The larger of these, a 75-kDa protein, was predominantly nuclear in location, and binding of this protein to 545-nt riboprobe did not differ between confluent and proliferating cells (Figure 4A). Interestingly, a protein with an apparent molecular mass of 51 kDa, present predominantly in the cytoplasm, bound to the 545-nt riboprobe more in confluent cells than in proliferating cells. Densitometric analysis of repeated experiments showed that the binding of this 51-kDa protein to the 545-nt probe was 3-fold higher in confluent as compared with proliferating cells (Figure 4B). Binding of the 75 and 51 kDa was inhibited by preincubation of cell extracts with 100x unlabeled probe (Figure 4C).

View larger version (33K): [in this window] [in a new window]   Figure 4. UV–cross-linking experiments showing the interaction between the 3'-UTR of eNOS with protein extracts from endothelial cells during different growth states. Cytoplasmic and nuclear protein extracts were incubated with 32P-labeled 545-nt probe and UV-irradiated and digested with RNase A, and the binding complex was run on a 12.5% SDS-polyacrylamide gel. A, Binding of 32P-labeled 545-nt probe with nuclear (Nuc) and cytoplasmic (Cyto) protein extracts from confluent (Con) and preconfluent (Precon) cells. B, Quantitative data from 3 separate experiments on the RNA/51-kDa cytoplasmic protein complex formation between different growth states. Data are mean±SEM. *P<0.05 vs preconfluent cells. C, Results from UV–cross-linking experiments showing the specific competition using 100x unlabeled 545-nt riboprobe for the nuclear and cytoplasmic protein binding to eNOS 3'-UTR.

In other experiments, 4 large riboprobes corresponding to the coding region of eNOS were used in UV–cross-linking assays but were not observed to specifically bind to proteins from either confluent or proliferating endothelial cells (data not shown).

Characterization of the cis Element in the eNOS 3'-UTR Involved in Binding of the 51-kDa Protein
A series of riboprobes corresponding to truncations of the eNOS 3'-UTR were created and used in UV–cross-linking experiments (Table 1 and Figure 5). Binding to the 51-kDa protein was observed for riboprobes corresponding to 3552 to 3900 (349 nt), 3552 to 3854 (303 nt), and 3552 to 3780 (229 nt) of eNOS and not for riboprobe corresponding to 3781 to 4096 (316 nt) (Figure 5). In each of these cases of interaction with the 51-kDa protein, binding was specific, as it was competed by preincubation of the cytoplasmic extract with excess respective unlabeled cold probes. Binding of the last of these, a 229-base riboprobe, to the 51-kDa protein, was easily demonstrated (Figure 6A). Further division of this 229-nt region into a proximal 96-nt probe (corresponding to the terminal coding region) revealed no binding of the 51-kDa protein, although binding to the larger 75-kDa protein present in the nucleus persisted (Figure 6A). A riboprobe corresponding to the proximal portion of the 3'-UTR (a 76-nt riboprobe representing nucleotides 3648 to 3723 of eNOS) also demonstrated binding (Figure 6B) that could be competed with unlabeled probe and prevented by preincubation of the cell extract with proteinase K (data not shown). A 139-nt riboprobe representing the terminal 96 nt of the coding region and the first 43 nt of the 3'-UTR also demonstrated specific binding of the 51-kDa protein (Figure 6B).

View larger version (11K): [in this window] [in a new window]   Figure 5. Size and location as well as results from UV–cross-linking experiments showing interaction between either full-length eNOS 3'-UTR riboprobe or several other truncated riboprobes and protein extracts obtained from confluent endothelial cells. The experimental details are as described in Figure 4. Specific competition was done either with the full-length eNOS 3'-UTR or with the respective unlabeled riboprobes. Experiments were performed at least 3 times for each riboprobe. ND indicates not determined.

View larger version (64K): [in this window] [in a new window]   Figure 6. UV–cross-linking experiments demonstrating the interaction between the protein extracts from confluent endothelial cells and 32P-labeled 229-nt or 96-nt riboprobe (A) and 32P-labeled 76-nt or 139-nt riboprobe (B). Experiments were performed at least 3 times for each riboprobe as described in Figure 4. Abbreviations as in Figure 4.

The above experiments suggest that the proximal 43 nucleotides of the eNOS 3'-UTR, corresponding to the sequence AAC CCC UCU UGC UUC CCA CUG CAG UUC CCG GAG AGA GGG GCU G, are critical in binding of the cytoplasmic 51-kDa protein. To confirm this, additional experiments were performed in which various excess amounts of unlabeled truncated riboprobes were used to compete for binding of the 51-kDa protein to the labeled 545-nt riboprobe. Binding to the 545-nt riboprobe was effectively competed by the unlabeled 229-nt riboprobe, the 76-nt riboprobe, and a 43-nt riboprobe, all of which included structures within the proximal eNOS 3'-UTR. In contrast, the 96-nt riboprobe representing the terminal coding region did not compete for binding of the 51-kDa protein (Figure 7).

View larger version (16K): [in this window] [in a new window]   Figure 7. Competition UV–cross-linking assays in which the 32P-labeled 545-nt probe was incubated with cytoplasmic extract from confluent endothelial cells in the presence of various unlabeled riboprobes representing various portions of either the terminal coding region of eNOS (96 nt) or the proximal 3'-UTR (229, 76, and 43 nt). UV–cross-linking assays were performed as described in Figure 4. Experiments were performed at least 3 times, and mean data are shown. Abbreviations as in Figure 4. O.D. indicates optical density.

Binding of the Cytoplasmic 51-kDa Protein to eNOS 3'-UTR in Confluent Cells Is Not Related to AUF-1
AU-rich regions in the 3'-UTR have been shown to modulate mRNA stability in the case of several cytokines and proto-oncogenes via an interaction with AUF-1. In the 3'-UTR of eNOS, there exist 2 AUUUA repeats that could serve as a target for such regulation. However, this does not seem to be the case, as a 35-nt riboprobe encompassing these AU-rich elements did not bind to proteins in extracts from confluent cells (Figure 5). In addition, 25 ng of recombinant AUF-1 (a gift from Dr Gary Brewer) failed to produce a binding complex with the 545-nt eNOS 3'-UTR (Figure 8A). Finally, a Western analysis of proteins from proliferating and confluent BAECs using immune serum raised against AUF-1 (also supplied by Dr Gary Brewer) demonstrated 2 proteins of 44 to 45 kDa (Figure 8B). The expression of these was predominantly nuclear and was greater in the preconfluent cells than in the confluent cells. This protein doublet corresponding to a 45-kDa protein does not relate, either in size or location, to the proteins we observed binding to the eNOS 3'-UTR.

View larger version (22K): [in this window] [in a new window]   Figure 8. UV–cross-linking experiments showing the interaction between the 3'-UTR of eNOS and either protein extracts from confluent endothelial cells or recombinant AUF-1 protein. Cell extracts from confluent endothelial cells, as well as 25 ng AUF-1 protein, were incubated with 32P-labeled 545-nt probe and UV-irradiated and digested with RNase A, and the binding complex was run on a 12.5% SDS-polyacrylamide gel. A, Results from a representative experiment. B, Western blot of proteins obtained from the preconfluent and confluent cells using the immune serum raised against AUF-1. Abbreviations as in Figure 4.

Examination of Proximal eNOS 3'-UTR in Modulation of Chimeric CAT/eNOS mRNA Stability
To firmly establish a role of the proximal 43 nt of the eNOS 3'-UTR in regulation of mRNA stability, BAECs were stably transfected with 2 chimeric plasmid constructs. One, termed pcDNACAT/eNOS_(wtUTR), consisted of the full-length eNOS coding region and 3'-UTR cloned 3' to a cytomegalovirus-driven 270-bp portion of the CAT gene, in pcDNA plasmid. A second, termed pcDNACAT/eNOS_(UTR), was identical, except that the proximal 43 nt of the 3'-UTR had been deleted using splicing-by-overlap extension. Successful deletion of this region was confirmed by sequencing in both directions. Confluent endothelial cells were treated with actinomycin D, and the CAT/eNOS mRNA was followed by Northern analysis using a ³²P-labeled 270 bp CAT probe. As shown in Figure 9, the half-life of the mRNA transcribed from pcDNACAT/eNOS_(wtUTR) was 2 hours. In striking contrast, deletion of the proximal 43 nt of the eNOS 3'-UTR resulted in marked stabilization of the CAT/eNOS message.

View larger version (18K): [in this window] [in a new window]   Figure 9. Measurement of chimeric CAT/eNOS mRNA half-life in stably transfected BAECs. Endothelial cells were transfected with pcDNACAT/eNOS(wtUTR), which consisted of the full-length eNOS coding region and 3'-UTR downstream from the cytomegalovirus-driven CAT gene. A second construct, pcDNACAT/eNOS(UTR), was identical, except that the proximal 43 nt of the 3'-UTR had been deleted using splicing-by-overlap extension. Stable cells were selected using G418. Confluent transfec-tants were treated with actinomycin D, and the CAT/eNOS mRNA was measured by Northern analysis using a 270-bp 32P-labeled CAT probe. Left, Representative Northern blot. Right, Average ratio of CAT/eNOS mRNA to 28S rRNA in wild-type and mutant transfectants, from 3 separate experiments. Data are mean±SEM. *P<0.05 vs confluent cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
In previous studies, we demonstrated that steady-state levels of eNOS mRNA, as well as protein and enzymatic activity are markedly increased in proliferating versus confluent cells.¹⁵ In the present experiments, we demonstrated that this regulation of eNOS expression is entirely modulated by posttranscriptional mechanisms. Using UV–cross-linking studies, we found a 51-kDa cytoplasmic protein that interacted with the eNOS 3'-UTR to a greater extent in extracts from confluent cells as compared with proliferating cells. Increased binding of this 51-kDa cytoplasmic protein corresponded with a decrease in eNOS mRNA levels in confluent cells. Given that RNA/protein interactions are important in regulating mRNA stability, it is likely that the differential binding of this 51-kDa protein in the 2 growth states is critical in modulation of eNOS gene expression. In addition, we identified the 43-nt cis element in the proximal portion of eNOS 3'-UTR, responsible for binding of this protein. A transfected chimeric CAT/eNOS mRNA was dramatically stabilized by deletion of the proximal 43 bp of eNOS 3'-UTR.

It has recently become apparent that mRNA decay is a highly regulated process involving interaction of cis-acting sequences and trans-acting factors. A well-defined example of posttranscriptional regulation of eukaryotic genes is that of AU element–mediated mRNA decay. In the case of c-fos,¹⁸ c-myc,¹⁹ granulocyte-macrophage colony-stimulating factor,²⁰ interleukin (IL)-1,²¹ IL-2,²² and IL-3,²³ AUUUA repeats in the 3'-UTR have been shown to be critical in mRNA destabilization. The trans-acting factor binding to this motif, AUF-1, has been cloned.²⁴ Interestingly, 2 AUUUA repeats are present in the eNOS 3'-UTR; however, they are unlikely to participate in modulation of eNOS mRNA half-life for the following reasons. First, a riboprobe representing these regions failed to bind proteins derived from BAECs. Second, AUF-1, the protein that recognizes this region, was negligibly expressed in the cytoplasm of confluent endothelial cells. Finally, the recombinant AUF-1 did not bind to the eNOS 3'-UTR (Figure 7). In addition to AUUUA, other 3'-UTR sequences of genes have been shown to influence gene expression. For example, binding of a yet-to-be-identified protein(s) to the pyrimidine-rich motifs has been shown to alter the half-life of mRNAs for -globin,²⁵ tyrosine hydroxylase,^{26 27} and GAP-43.^{28 29} Further, proteins interacting with cis elements in the 3'-UTR are responsible for regulating expression mRNA stability of ribonucleotide reductases r1 and r2.^{30 31}

In addition to the cytosolic 51-kDa protein, a 75-kDa protein also bound to the 545- and 229-nt riboprobes, which were predominantly nuclear in location. Because RNA decay occurs in the cytoplasm, and this protein was not differentially expressed in the 2 growth states, it is unlikely to be involved in modulation of eNOS mRNA stability in the 2 growth conditions. For this reason, we proceeded to characterize the cis element involved in binding the cytosolic 51-kDa protein.

Using UV–cross-linking studies with truncated riboprobes corresponding to various portions of the eNOS mRNA, we were able to demonstrate that the proximal 43-nt portion of the 3'-UTR was critical for binding of the 51-kDa protein. This portion of the 3'-UTR is schematically illustrated in Figure 10, with the critical 43 nt shown in bold. This sequence is unlike previously reported cis-acting sequences involved in mRNA decay. It is quite likely that secondary structure is critical in binding the 51-kDa protein. As shown in Figure 10, the predicted structure of the proximal 3'-UTR of eNOS contains a stem loop with adjacent secondary structures (bold region in Figure 10), which could form a pocket into which the protein is bound. Of note, the 76- and 139-nt riboprobes seemed to exhibit weaker binding than the 229-nt riboprobe or longer riboprobes encompassing this region, again suggesting that flanking structure may enhance protein recognition. Definitive proof that the proximal 43-nt region of eNOS 3'-UTR modulates mRNA stability was derived from studies using stable transfections of CAT/eNOS constructs. In these studies, deletion of the 43-nt region resulted in an mRNA that was dramatically stabilized as compared with the wild-type CAT/eNOS mRNA.

View larger version (19K): [in this window] [in a new window]   Figure 10. Predicted 2-dimensional configuration of the terminal coding region and full-length 3'-UTR of eNOS as determined by the program SQUIGGLE, which estimates the most stable structure on the basis of physical properties of the sequence provided. Bold region denotes the 43 nt involved in binding the 51-kDa protein present in the cytoplasm of confluent endothelial cells.

It has previously been shown that eNOS mRNA is destabilized by the cytokine TNF-. This likely involves a completely different mechanism as compared with growth modulation of eNOS mRNA stability. Recently, TNF- treatment has been shown to increase protein binding to a UC-rich region in the midportion of the eNOS 3'-UTR.³² This sequence is quite different from that identified in our present study.

RNA binding proteins may serve dual roles. For example, the iron-regulatory protein that recognizes transferrin and ferritin mRNAs is identical to aconitase, a Krebs cycle enzyme.³³ GAPDH has been shown to bind to AU-rich regions of mRNAs and modulate RNA turnover in addition to its well-described role in glycolysis.³⁴ Likewise, heat shock protein-70 (HSP-70) has been shown to modulate turnover of erythropoietin mRNA in addition to its role in stress responses.³⁵ At present, the 51-kDa protein we have observed binding to the proximal 3'-UTR of eNOS has not been characterized in terms its structure or sequence; however, it is conceivable that it may also be a protein with other functions.

Of note, the half-life of eNOS mRNA in confluent cells (9 hours) in this study was substantially lower than that reported previously.^{12 13 14} The explanation for this remains unclear; however, it may relate to differences in the state of growth, the origin of the cells, or the cell culture conditions. Of interest, the half-life of eNOS mRNA in proliferating cells in this present study was similar to that previously reported. The use of actinomycin D to study mRNA half-life may be criticized, because the drug may paradoxically induce some genes^{36 37} and may reduce expression of proteins involved in RNA decay. In additional studies, we pulse-chased endothelial cells with [³H]uridine and followed eNOS mRNA levels for the ensuing 24 hours, after hybridizing to cold eNOS cDNA previously slot-blotted on nitrocellulose membranes. Using this approach, we also found that eNOS mRNA half-life was markedly longer in proliferating as compared with confluent cells (data not shown). Given this finding, it seems that the use of actinomycin D allowed a valid comparison of eNOS mRNA half-lives in the 2 growth states.

Previous studies have shown that eNOS expression is influenced by several pathophysiological conditions via posttranscriptional mechanisms. These include exposure to hypoxia, oxidized LDL, 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors,¹⁰ and cytokines.^{8 12 13 14} The precise mechanisms involved in modulation of eNOS mRNA stability in these studies have not been determined. The present study indicates that eNOS mRNA expression is also regulated via posttranscriptional mechanisms during cell growth, likely via interactions with a 51-kDa cytosolic protein. Whether these various conditions share similar mechanisms of posttranscriptional control is unknown. It is interesting to speculate that increased eNOS expression may reflect a beneficial response after endothelial cell damage, such that endothelial cells growing over the denuded region would produce large quantities of nitric oxide to compensate for the loss of adjacent endothelial cells. This would minimize platelet adhesion, smooth muscle cell proliferation, and adhesion molecule expression that occurs in response to vascular injury. Indeed, increased expression of eNOS has been demonstrated in proliferating cells in vivo in rat aortas.³⁸ There has been substantial interest in the role of nitric oxide in modulating vessel growth and development. For example, it has been shown that tube development by growing endothelial cells in 3-dimensional gels in response to transforming growth factor ß is dependent on nitric oxide and inhibited by antagonists of NO synthase.³⁹ Although this role of nitric oxide is controversial,^{40 41} it is possible that increased production of nitric oxide by endothelial cells during the angiogenesis process plays an important role in vessel development. Future studies of interactions of the eNOS mRNA with the 51-kDa protein identified in the present study and potentially other trans-acting factors may provide mechanistic insight into regulation of eNOS in these various physiological and pathophysiological states.

	Acknowledgments

 
This work was supported by NIH grants HL390006 and HL59248 and program project grant HL58000, as well as an American Heart Association Scientist Development Grant award (9630119N; to S.R.). We thank Dr Gary Brewer for helpful suggestions and for the gift of the AUF-1 protein and its antibody.

Received November 12, 1998; accepted July 13, 1999.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Palmer RMJ, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature (Lond). 1987;327:524–526.[Medline] [Order article via Infotrieve]
2. Garg UC, Hassid A. Nitric oxide-generating vasodilators and 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells. J Clin Invest. 1989;83:1774–1777.
3. Alheid U, Frölich JC, Förstermann U. Endothelium-derived relaxing factor from cultured human endothelial cells inhibits aggregation of human platelets. Thromb Res. 1987;47:561–571.[Medline] [Order article via Infotrieve]
4. Kubes P, Suzuki M, Granger DN. Nitric oxide: an endogenous modulator of leukocyte adhesion. Proc Natl Acad Sci U S A. 1991;88:4651–4655.[Abstract/Free Full Text]
5. Nishida K, Harrison DG, Navas JP, Fisher AA, Dockery SP, Uematsu M, Nerem RM, Alexander RW, Murphy TJ. Molecular cloning and characterization of the constitutive bovine aortic endothelial cell nitric oxide synthase. J Clin Invest. 1992;90:2092–2096.
6. Awolesi MA, Sessa WC, Sumpio BE. Cyclic strain upregulates nitric oxide synthase in cultured bovine aortic endothelial cells. J Clin Invest. 1995;96:1449–1454.
7. Zembowicz A, Tang J, Wu KK. Transcriptional induction of endothelial nitric oxide synthase type III by lysophosphatidylcholine. J Biol Chem. 1995;270:17006–17010.[Abstract/Free Full Text]
8. Hirata K, Miki N, Kuroda Y, Sakoda T, Kawashima S, Yokoyama M. Low concentration of oxidized low-density lipoprotein and lysophosphatidylcholine upregulate constitutive nitric oxide synthase mRNA expression in bovine aortic endothelial cells. Circ Res. 1995;76:958–962.[Abstract/Free Full Text]
9. Ramasamy S, Parthasarathy S, Harrison DG. Regulation of endothelial nitric oxide synthase gene expression by oxidized linoleic acid. J Lipid Res. 1998;39:268–276.[Abstract/Free Full Text]
10. Laufs U, La Fata V, Plutzky J, Liao JK. Upregulation of endothelial nitric oxide synthase by HMG CoA reductase inhibitors. Circulation. 1998;97:1129–1135.[Abstract/Free Full Text]
11. Sessa WC, Pritchard K, Seyedi N, Wang J, Hintze TH. Chronic exercise in dogs increases coronary vascular nitric oxide production and endothelial cell nitric oxide synthase gene expression. Circ Res. 1994;74:349–353.[Abstract/Free Full Text]
12. Yoshizumi M, Perrella MA, Burnett JC, Lee M-E. Tumor necrosis factor downregulates an endothelial nitric oxide synthase mRNA by shortening its half-life. Circ Res. 1993;73:205–209.[Abstract]
13. McQuillan LP, Leung GK, Marsden PA, Kostyk SK, Kourembanas S. Hypoxia inhibits expression of eNOS via transcriptional and posttranscriptional mechanisms. Am J Physiol. 1994;267:H1921–H1927.[Abstract/Free Full Text]
14. Liao JK, Shin WS, Lee WY, Clark SL. Oxidized low-density lipoprotein decreases the expression of endothelial nitric oxide synthase. J Biol Chem. 1995;270:319–324.[Abstract/Free Full Text]
15. Arnal J-F, Yamin J, Dockery S, Harrison DG. Regulation of endothelial nitric oxide synthase mRNA, protein and activity during cell growth. Am J Physiol. 1994;267:C1381–C1388.[Abstract/Free Full Text]
16. Greenberg M. Identification of newly transcribed RNA. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA, Seidman JG, Struhl K, eds. Current Protocols in Molecular Biology. New York: John Wiley and Sons; 1987:4.10.1–4.10.11.
17. Dignam JD, Lebovitz RM, Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 1983;11:1475–1489.[Abstract/Free Full Text]
18. Zubiaga AM, Belasco JG, Greenberg ME. The nonamer UUAUUUAUU is the key AU-rich sequence motif that mediates mRNA degradation. Mol Cell Biol. 1995;15:2219–2230.[Abstract]
19. Brewer G. An A+U-rich element RNA-binding factor regulates c-myc mRNA stability in vitro. Mol Cell Biol. 1991;11:2460–2466.[Abstract/Free Full Text]
20. Rajagopalan LE, Malter JS. Modulation of granulocyte-macrophage colony-stimulating factor mRNA stability in vitro by the adenosine-uridine binding factor. J Biol Chem. 1994;269:23882–23888.[Abstract/Free Full Text]
21. Gorospe M, Baglioni C. Degradation of unstable interleukin-1 alpha mRNA in a rabbit reticulocyte cell-free system: localization of an instability determinant to a cluster of AUUUA motifs. J Biol Chem. 1994;269:11845–11851.[Abstract/Free Full Text]
22. Henics T, Sanfridson A, Hamilton BJ, Nagy E, Rigby WF. Enhanced stability of interleukin-2 mRNA in MLA 144 cells. Possible role of cytoplasmic AU-rich sequence-binding proteins. J Biol Chem. 1994;269:5377–5383.[Abstract/Free Full Text]
23. Stoecklin G, Hahn S, Moroni C. Functional hierarchy of AUUUA motifs in mediating rapid interleukin-3 mRNA decay. J Biol Chem. 1994;269:28591–28597.[Abstract/Free Full Text]
24. Zhang W, Wagner BJ, Ehrenman K, Schaefer AW, DeMaria CT, Crater D, DeHaven K, Long L, Brewer G. Purification, characterization, and cDNA cloning of an AU-rich element RNA-binding protein, AUF1. Mol Cell Biol. 1993;13:7652–7665.[Abstract/Free Full Text]
25. Wang X, Kiledjian M, Weiss IM, Liebhaber SA. Detection and characterization of a 3' untranslated region ribonucleoprotein complex associated with human alpha-globin mRNA stability. Mol Cell Biol. 1995;15:1769–1777.[Abstract]
26. Czyzyk-Krzeska MF, Dominski Z, Kole R, Millhorn DE. Hypoxia stimulates binding of a cytoplasmic protein to a pyrimidine-rich sequence in the 3'-untranslated region of rat tyrosine hydroxylase mRNA. J Biol Chem. 1994;269:9940–9945.[Abstract/Free Full Text]
27. Czyzyk-Krzeska MF, Paulding WR, Beresh JE, Kroll SL. Post-transcriptional regulation of tyrosine hydroxylase gene expression by oxygen in PC12 cells. Kidney Int. 1997;51:585–590.[Medline] [Order article via Infotrieve]
28. Kohn DT, Tsai KC, Cansino VV, Neve RL, Perrone-Bizzozero NI. Role of highly conserved pyrimidine-rich sequences in the 3' untranslated region of the GAP-43 mRNA in mRNA stability and RNA-protein interactions. Brain Res Mol Brain Res. 1996;36:240–250.[Medline] [Order article via Infotrieve]
29. Irwin N, Baekelandt V, Goritchenko L, Benowitz LI. Identification of two proteins that bind to a pyrimidine-rich sequence in the 3'-untranslated region of GAP-43 mRNA. Nucleic Acids Res. 1997;25:1281–1288.[Abstract/Free Full Text]
30. Chen FY, Amara FM, Wright JA. Defining a novel ribonucleotide reductase r1 mRNA cis element that binds to an unique cytoplasmic trans-acting protein. Nucleic Acids Res. 1994;22:4796–4797.[Abstract/Free Full Text]
31. Amara FM, Chen FY, Wright JA. Defining a novel cis element in the 3'-untranslated region of mammalian ribonucleotide reductase component R2 mRNA: role in transforming growth factor-beta 1 induced mRNA stabilization. Nucleic Acids Res. 1995;23:1461–1467.[Abstract/Free Full Text]
32. Alonso J, De Miguel LS, Monton M, Casado S, Lopez-Farre A. Endothelial cytosolic proteins bind to the 3' untranslated region of endothelial nitric oxide synthase mRNA: regulation by tumor necrosis factor alpha. Mol Cell Biol. 1997;17:5719–5726.[Abstract]
33. Kaptain S, Downey WE, Tang C, Philpott C, Haile D, Orloff DG, Harford JB, Rouault TA, Klausner RD. A regulated RNA binding protein also possesses aconitase activity. Proc Natl Acad Sci U S A. 1991;88:10109–10113.[Abstract/Free Full Text]
34. Nagy E, Rigby WF. Glyceraldehyde-3-phosphate dehydrogenase selectively binds AU-rich RNA in the NAD(+)-binding region (Rossmann fold). J Biol Chem. 1995;270:2755–2763.[Abstract/Free Full Text]
35. Scandurro AB, Rondon IJ, Wilson RB, Tenenbaum SA, Garry RF, Beckman BS. Interaction of erythropoietin RNA binding protein with erythropoietin RNA requires an association with heat shock protein 70. Kidney Int. 1997;51:579–584.[Medline] [Order article via Infotrieve]
36. Stefanelli C, Flamigni F, Ferrari F, Rossoni C, Caldarera CM. Induction of ornithine decarboxylase by transcriptional inhibitors in quiescent thymocytes. Biochem Int. 1992;26:995–1001.[Medline] [Order article via Infotrieve]
37. Aloj S, Grieco D, Kohn A, Nikodem V, Kohn L. Thyrotropin regulation of malic enzyme in rat FRTL-5 thyroid cells. Mol Endocrinol. 1990;4:611–622.[Abstract]
38. Poppa V, Miyashiro JK, Corson MA, Berk BC. Endothelial NO synthase is increased in regenerating endothelium after denuding injury of the rat aorta. Arterioscler Thromb Vasc Biol. 1998;18:1312–1321.[Abstract/Free Full Text]
39. Papapetropoulos A, Desai KM, Rudic RD, Mayer B, Zhang R, Ruiz-Torres MP, Garcia-Cardena G, Madri JA, Sessa WC. Nitric oxide synthase inhibitors attenuate transforming-growth-factor-beta 1-stimulated capillary organization in vitro. Am J Pathol. 1997;150:1835–1844.[Abstract]
40. Pipili-Synetos E, Sakkoula E, Haralabopoulos G, Andriopoulou P, Peristeris P, Maragoudakis ME. Evidence that nitric oxide is an endogenous antiangiogenic mediator. Br J Pharmacol. 1994;111:894–902.[Medline] [Order article via Infotrieve]
41. Pipili-Synetos E, Sakkoula E, Maragoudakis ME. Nitric oxide is involved in the regulation of angiogenesis. Br J Pharmacol. 1993;108:855–857.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				H. Cai, D. Liu, and J. G.N. Garcia CaM Kinase II-dependent pathophysiological signalling in endothelial cells Cardiovasc Res, January 1, 2008; 77(1): 30 - 34. [Abstract] [Full Text] [PDF]

				I. Kosmidou, J. P. Moore, M. Weber, and C. D. Searles Statin Treatment and 3' Polyadenylation of eNOS mRNA Arterioscler. Thromb. Vasc. Biol., December 1, 2007; 27(12): 2642 - 2649. [Abstract] [Full Text] [PDF]

				D. M. Dudzinski and T. Michel Life history of eNOS: Partners and pathways Cardiovasc Res, July 15, 2007; 75(2): 247 - 260. [Abstract] [Full Text] [PDF]

				J. E. Fish, C. C. Matouk, E. Yeboah, S. C. Bevan, M. Khan, K. Patil, M. Ohh, and P. A. Marsden Hypoxia-inducible Expression of a Natural cis-Antisense Transcript Inhibits Endothelial Nitric-oxide Synthase J. Biol. Chem., May 25, 2007; 282(21): 15652 - 15666. [Abstract] [Full Text] [PDF]

				C. D. Searles Transcriptional and posttranscriptional regulation of endothelial nitric oxide synthase expression Am J Physiol Cell Physiol, November 1, 2006; 291(5): C803 - C816. [Abstract] [Full Text] [PDF]

				I. A. Mawji and P. A. Marsden RNA Transfection Is a Versatile Tool to Investigate Endothelin-1 Posttranscriptional Regulation. Experimental Biology and Medicine, June 1, 2006; 231(6): 704 - 708. [Abstract] [Full Text] [PDF]

				A. V. Andreeva, R. Vaiskunaite, M. A. Kutuzov, J. Profirovic, R. A. Skidgel, and T. Voyno-Yasenetskaya Novel Mechanisms of G Protein-Dependent Regulation of Endothelial Nitric-Oxide Synthase Mol. Pharmacol., March 1, 2006; 69(3): 975 - 982. [Abstract] [Full Text] [PDF]

				M. Weber, C. H. Hagedorn, D. G. Harrison, and C. D. Searles Laminar Shear Stress and 3' Polyadenylation of eNOS mRNA Circ. Res., June 10, 2005; 96(11): 1161 - 1168. [Abstract] [Full Text] [PDF]

				J. M. Cale, S. C. Tsoi, M. Toppe, M. A. Grummer, M. Ochiai, R. R. Magness, and I. M. Bird Molecular Cloning of Ovine Endothelial Nitric Oxide Synthase and Expression in COS-7 Cells Reproductive Sciences, April 1, 2005; 12(3): 156 - 168. [Abstract] [PDF]

				C. D. Searles, L. Ide, M. E. Davis, H. Cai, and M. Weber Actin Cytoskeleton Organization and Posttranscriptional Regulation of Endothelial Nitric Oxide Synthase During Cell Growth Circ. Res., September 3, 2004; 95(5): 488 - 495. [Abstract] [Full Text] [PDF]

				S. C. Tai, G. B. Robb, and P. A. Marsden Endothelial Nitric Oxide Synthase: A New Paradigm for Gene Regulation in the Injured Blood Vessel Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 405 - 412. [Abstract] [Full Text]

				V. STANGL, M. LORENZ, S. MEINERS, A. LUDWIG, C. BARTSCH, M. MOOBED, A. VIETZKE, H.-T. KINKEL, G. BAUMANN, and K. STANGL Long-term up-regulation of eNOS and improvement of endothelial function by inhibition of the ubiquitin-proteasome pathway FASEB J, February 1, 2004; 18(2): 272 - 279. [Abstract] [Full Text] [PDF]

F. Coulet, S. Nadaud, M. Agrapart, and F. Soubrier Identification of Hypoxia-response Element in the Human Endothelial Nitric-oxide Synthase Gene Promoter J. Biol. Chem., November 21, 2003; 278(47): 46230 - 46240.