Argininosuccinate synthase expression is required to maintain nitric oxide production and cell viability in aortic endothelial cells.

Although cellular levels of arginine greatly exceed the apparent K(m) for endothelial nitric-oxide synthase, current evidence suggests that the bulk of this arginine may not be available for nitric oxide (NO) production. We propose that arginine regeneration, that is the recycling of citrulline back to arginine, defines the essential source of arginine for NO production. To support this proposal, RNA interference analysis was used to selectively reduce the expression of argininosuccinate synthase (AS), because the only known metabolic role for AS in endothelial cells is in the regeneration of l-arginine from l-citrulline. Western blot analysis demonstrated a significant and dose-dependent reduction of AS protein as a result of AS small interfering RNA treatment with a corresponding diminished capacity to produce basal or stimulated levels of NO, despite saturating levels of arginine in the medium. Unanticipated, however, was the finding that the viability of AS small interfering RNA-treated endothelial cells was significantly decreased when compared with control cells. Trypan blue exclusion analysis suggested that the loss of viability was not because of necrosis. Two indicators, reduced expression of Bcl-2 and an increase in caspase activity, which correlated directly with reduced expression of AS, suggested that the loss of viability was because of apoptosis. The exposure of cells to an NO donor prevented apoptosis associated with reduced AS expression. Overall, these results demonstrate the essential role of AS for endothelial NO production and cell viability.

Nitric oxide (NO) 1 is an important modulator for a wide range of functions including vasodilation of blood vessels, immune system function, angiogenesis, inhibition of leukocyte adhesion and platelet aggregation, gene regulation, and apoptosis (1)(2)(3). Moreover, NO has a dual role in cell viability depending on the tissue type and concentration. Either very high or very low concentrations of NO may induce cell death, whereas basal concentrations may inhibit apoptosis (4 -7). Previous work has shown that NO protects against serum starvation- (8), H 2 O 2 -(9), TNF-␣- (10), and oxidized low density lipoprotein-induced apoptosis (11,12) in endothelial cells.
Argininosuccinate synthase (AS), the rate-limiting step (13) in the regeneration of arginine from citrulline, catalyzes the synthesis of argininosuccinate, AMP, and inorganic pyrophosphate from citrulline, ATP, and aspartate. Argininosuccinate is then cleaved by argininosuccinate lyase (AL) to produce Larginine and fumarate. In the liver, AS and AL function together as components of the urea cycle, ultimately to form arginine from citrulline. Although the expression of AS and AL in the liver is high, both enzymes are found in most mammalian tissues, although at significantly lower levels. The discovery of arginine-derived NO, catalyzed by nitric-oxide synthases (NOSs), revealed a second role for AS and AL (1)(2)(3). Together with NOS, they function as part of a citrulline-NO cycle where AS and AL convert citrulline to arginine, which is then oxidized to form NO and citrulline by NOS.
Although both extracellular and intracellular concentrations of arginine are much higher than the reported K m of arginine for endothelial nitric-oxide synthase (eNOS), NO production still appears to be limited by the availability of arginine (13)(14)(15)(16)(17)(18)(19)(20). Recently, a DNA microarray analysis of shear stress-induced NO production demonstrated that up-regulation of AS was coordinated with an increase in NO production (21), supporting an important role for AS in endothelial NO production. If AS is essential for endothelial NO production then even in the presence of excess arginine, depletion of AS should decrease the capacity of endothelial cells to produce NO. To test this hypothesis, RNA interference analysis was carried out using AS-specific RNA duplexes in bovine aortic endothelial cells (BAEC). In this report, we demonstrate a significant and dosedependent reduction of AS protein following siRNA transfection. We also show a concomitant decrease in enzyme activity corresponding to a decrease in stimulated and unstimulated NO production in endothelial cells with reduced AS expression, despite excess arginine in the media. In addition, we demonstrate that the viability of endothelial cells grown in excess arginine and treated with AS siRNA was significantly diminished compared with control cells.

EXPERIMENTAL PROCEDURES
Cell Culture-Bovine aortic endothelial cells were propagated in Dulbecco's modified Eagle's medium (1 g/liter glucose, Mediatech) supplemented with 10% fetal bovine serum (Hyclone Laboratories), 100 units ml Ϫ1 penicillin, and 100 g ml Ϫ1 streptomycin (Mediatech) at 37°C and 5% CO 2 . Twenty-four hours prior to transfection, BAEC were seeded in a 24-well plate at 5 ϫ 10 4 cells/well. Transfection of siRNA was carried out with TransIT-TKO (Mirus) as described by the manufacturer. For each well, 0.1-10 nM siRNA duplex was combined with 3 l of liposome in serum-free Dulbecco's modified Eagle's medium and applied to cells at 50 -70% confluency. Cells were assayed after a 24 -48-h transfection.
RNA Duplex Preparation-A Silencer siRNA construction kit (Ambion) was used to synthesize the 21-nucleotide RNA duplexes. Target sequences were chosen following the guidelines described by Tuschl and colleagues (22). 2 The siRNA sequence specific to AS corresponded to bp 73-93 relative to the first nucleotide of the start codon GGAGCAAG-GCUAUGACGUCtt. A control siRNA, UAGAUGGAGAGGCA-CUCGCtt, was designed by scrambling the bases of the AS siRNA. Both sequences were subjected to a BLAST search to rule out homology to known proteins.
Cell Lysate Preparation and Immunoblotting-BAEC were trypsinized and then washed in ice-cold phosphate-buffered saline and resuspended in radioimmune precipitation assay buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 1ϫ protease inhibitors in phosphate-buffered saline) by vigorous pipetting followed by brief vortexing. The lysate was incubated on ice for 30 min, and the protein concentration was determined by BCA assay (Pierce). Equal amounts (5-10 g) of protein were resolved on 4 -15% polyacrylamide gels (Bio-Rad) and blotted onto Immobilon-P polyvinylidene difluoride membranes. Western blotting was performed as described previously (24). Briefly, membranes were blocked for 1 h in 5% nonfat dry milk in Tris-buffered saline-Tween-20 and subsequently washed. Membranes were incubated with primary antibody (1:2500 anti-AS (BD Transduction Laboratories), 1:1000 Bcl-2 (Santa Cruz), and 1:7500 anti-actin (Sigma)) in 5% nonfat dry milk for 1 h. Following washing, membranes were incubated with a secondary antibody in 5% nonfat dry milk for 1 h. The signal was visualized by chemiluminescence using ECL reagent (Amersham Biosciences) and then exposed to film. Band intensities were quantitated using ImageQuant software (Molecular Dynamics).
RNA Isolation and Quantitative Reverse Transcriptase (RT)-PCR-Total RNA was isolated from transfected BAEC using a commercially available kit according to the manufacturer's protocol (Ambion). RNA was treated with DNase (Ambion, DNase-free) and quantitated prior to reverse transcription, which was performed as described previously (24). Real time quantitative PCR was performed using AS-specific primers ASL228 and ASR278 (24). Results were normalized to 18 S rRNA.
Argininosuccinate Synthase Activity Assay-To assay AS activity in intact cells, BAEC were transfected with AS and control siRNA as described above. After incubation for 24 h, cells were depleted of arginine by incubation in synthetic Dulbecco's modified Eagle's medium containing 500 M citrulline and no arginine. Cells were incubated for 30 min in this medium and then stimulated with sodium orthovanadate (50 M) and A23187 calcium ionophore (0.25 M) (25) for 30 min. Nitric oxide production was measured as described below. Cells were counted by trypan blue exclusion following assay completion, and data were normalized as nitrite produced/h/1 ϫ 10 6 cells. In a separate experiment, in vitro AS activity was assayed by measuring conversion of [ 3 H]aspartate to [ 3 H]argininosuccinate in cell lysates as described by O'Brien (26) with the exception that [ 3 H]aspartate (500 M, 39 Ci/mmol) was used. Whole cell lysates were prepared by lysing the cells in 10 mM Tris-HCl containing protease inhibitors, 0.5 mM citrulline, and 0.5 mM aspartate followed by three cycles of freeze-thaw in a dry ice ethanol bath. Equal amounts of protein (50 g) for AS siRNA and control siRNA samples were added to each reaction, which contained citrulline (5 mM), Tris-HCl (10 mM, pH 7.5), ATP (0.1 mM), MgCl 2 (6 mM), KCl (920 mM), phosphoenolpyruvate (1.5 mM), pyruvate kinase (4.5 units), myokinase (4 units), and pyrophosphatase (0.2 units) in a final volume of 150 l. Reactions were run for 90 min at 37°C. At the end of the reaction period, 50 l of 1 M acetic acid was added, and the tubes were heated to 90°C for 30 min. Samples were brought up to 1 ml with water and applied to AG 1-X8 resin (200 -400 mesh, Bio-Rad) in 0.05 M acetic acid. An additional 2 ϫ 1 ml of 0.05 M acetic acid was applied to the columns. Radioactivity in the 3 ml of column flow-through ([ 3 H]argininosuccinate) was quantified by liquid scintillation counting in CytoScint (MP Biomedical).
Cell Viability-Twenty-four hours after transfection, cells were trypsinized and counted by trypan blue exclusion. Briefly, 20 l of the cell suspension was combined with 180 l of trypan blue (Invitrogen), and cell counts were performed using a hemacytometer. Cell number was determined by averaging the cells in four squares, multiplying by the dilution factor (10), multiplying by 10,000, and adjusting for cell volume to determine cells/well. Viability was also detected using a colorimetric assay that measures tetrazolium dye reduction as follows (27). siRNA transfections were carried out in 96-well plate format. AS knockdown and control BAEC were incubated with 20 l/well (5 mg/ml stock) 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma) for 4 h under normal culture conditions. The medium was removed without disturbing the formazan crystals, and 100 l/well Me 2 SO (Sigma) was used to resuspend the product. The plate was read on a Quant spectrophotometer (Bio-Tek Instruments) at 570 nm. Necrotic cell death was measured using a commercially available cytotoxicity kit (Promega). Lactate dehydrogenase (LDH) activity was measured in cell culture supernatants from AS knockdown and control cells using a FLUOstar Galaxy spectrofluorometer (BMB Labtechnologies) with 544 nm excitation and 590 nm emission. Results were presented as the percent of the maximum LDH release, which was determined by complete lysis of cells.
Apoptosis Detection-Caspase-3/7 activity was measured using Apo-ONE TM homogenous caspase-3/7 assay (Promega). Cells were transfected in 96-well plates with AS and control siRNA as described above. Cells were lysed, and caspase-3/7 activity was measured by cleavage of the caspase-3/7 substrate rhodamine 110 (bis-(N-benzyloxycarbonyl-Laspartyl-L-glutamyl-L-valyl-L-aspartic acid amide)), (Z-DEVD-R110). Samples were measured on a FLUOstar Galaxy spectrofluorometer with 492 nm excitation and 520 nm emission. For analysis of the effect an NO donor had on AS siRNA-induced apoptosis, BAEC were transfected with siRNA as described above. Four hours after transfection, the media were replaced with Dulbecco's modified Eagle's medium containing Glyco-SNAP-2 (Calbiochem) and incubated for 48 h prior to caspase 3/7 detection.
Nitric Oxide Assay-Twenty-four hours after transfection with siRNA, BAEC were stimulated with sodium orthovanadate (50 M) and A23187 calcium ionophore (0.25 M) (25). Nitrite was measured in the medium as an indicator of cellular NO using a fluorometric method (28). Samples were read on a Jasco spectrofluorometer exciting at 365 nm and detecting emission at 409 nm. Following stimulation, cells were counted by trypan blue exclusion, and data are presented as nitrite produced/1 ϫ 10 6 cells.
Statistical Analyses-Experimental data are expressed as the mean Ϯ S.E. Each experiment was performed independently at least three times.

RESULTS
Silencing of AS by siRNA Transfection-AS is known to catalyze the rate-limiting step (13) in arginine synthesis from citrulline and aspartate. Argininosuccinate lyase (AL) is also required for this conversion (29). To further elucidate the function of AS in endothelial cells, particularly with respect to its role in NO production, we utilized the recently developed technique of siRNA to selectively reduce AS expression. A 21nucleotide RNA duplex specific to the AS sequence was identified as outlined by Tuschl and colleagues (30) 2 and compared with known sequences using BLAST search to eliminate any sequences homologous to other genes. The AS siRNA was transcribed in vitro using the Silencer siRNA construction kit. A control siRNA was synthesized containing the same base composition as the AS siRNA but in a scrambled sequence. BAEC were transfected with 10 nM AS or control siRNA (except where indicated in the figures) for 24 h. AS protein levels were monitored by immunoblot analysis, and as shown in Fig. 1A, AS siRNA treatment specifically reduced cellular AS protein levels to less than 50% of control levels.
To determine whether reduction of AS protein correlated to the reduction of AS mRNA, total RNA was isolated, and AS message was quantitated by real time reverse transcriptase-PCR. As shown in Fig. 1B, AS siRNA treatment resulted in a reduction of AS mRNA that correlated with the reduction of AS protein.
Argininosuccinate Synthase Activity Is Reduced in AS siRNA-treated Cells-Previously we have demonstrated that citrulline was sufficient to sustain NO production in the absence of arginine, based on the presence of the recycling enzymes AS and AL. To confirm that the functional activity of AS was decreased in intact cells, corresponding to the observed decrease in protein expression, cells were equilibrated in arginine-free medium containing 500 M citrulline for 30 min and then stimulated with sodium orthovanadate and A23187 to give maximal NO production (25). Nitric oxide production was measured as nitrite using the 2,3-diaminonaphthalene assay (28). In the endothelial cells with reduced AS expression, nitric oxide production/cell was decreased by more than 80%, indicating a significant reduction in activity of the enzyme (Table I). For a direct measure of in vitro AS activity, conversion of [ 3 H]aspartate to [ 3 H]argininosuccinate in lysates from siRNAtreated cells was assayed (26). Activity in control cells was 2.9 pmol/min/mg, whereas activity in AS siRNA-treated cells was only 0.8 pmol/min/mg confirming that expression and activity of AS protein was reduced as a result of AS siRNA treatment- (Table II).
Reduction of Stimulated NO Production in AS-depleted BAEC-If the production of arginine from citrulline, catalyzed by the enzymes AS and AL, provided an essential source of arginine for NO production in endothelial cells, a reduction of AS protein expression should result in a decrease in the amount of NO produced upon stimulation (13-17) despite saturating levels of extracellular arginine. To test whether AS is essential for endothelial NO production, AS siRNA-treated BAEC were stimulated with sodium orthovanadate and cal-cium ionophore in the presence of excess extracellular arginine (ϳ500 M), and NO release was determined. As shown in Table  I, reduction in AS expression resulted in a 56% decrease in stimulated NO produced as compared with control. Because nitrite levels were normalized to the number of cells counted by trypan blue exclusion, the decrease in NO detected could be attributed to a decrease in production for each viable cell as opposed to any effect on cell viability. These results demonstrate that sufficient levels of AS must be expressed to maintain NO production and that arginine regeneration plays an essential role in stimulated NO production in endothelial cells even in the presence of excess arginine.
AS Silencing Results in a Reduced Basal Level of NO Production-The significant effect of AS silencing on stimulated NO production in endothelial cells led us to investigate whether basal (unstimulated) levels of NO production in BAEC were also reduced by AS siRNA treatment. Twenty-four hours after transfection with AS and control siRNAs, the media were replaced with fresh media (containing excess arginine and no phenol red indicator), and samples were collected after 24 h for nitrite determination. Cell number was assessed by trypan blue exclusion analysis, and results are presented as nitrite produced/h/1 ϫ 10 6 cells. As shown in Table I, a significant reduction (70%) in basal production of NO was observed in FIG. 1. Specific reduction of AS protein and mRNA by siRNA transfection. BAEC were transiently transfected with AS and control siRNA using Transit-TKO. A, twenty-four hours after transfection, cell lysates were prepared, 10 g of each sample was loaded onto an SDSpolyacrylamide gel, and standard Western blotting was performed. Anti-AS and anti-actin antibodies were used to detect the amount of protein present. B, twentyfour hours after transfection, total RNA was isolated using RNAqueous total isolation kit and reverse-transcribed with Superscript II (Invitrogen). AS message was detected using real time quantitative PCR. Results were normalized to 18 S rRNA. L, liposome only. BAEC, whereas AS expression had been decreased by siRNA transfection. Thus, AS expression is also required to sustain unstimulated levels of NO produced by endothelial cells even in the presence of excess levels of extracellular arginine.
Decrease in Cell Survival in AS-depleted BAEC-Interestingly, AS siRNA treatment of endothelial cells resulted in a noticeable and unexpected decrease in cell viability over control cells. To confirm this observation, AS siRNA-treated cells were evaluated for viability by trypan blue exclusion. As shown in Fig. 2A, a 46% decrease in cell survival was observed compared with control. To further substantiate this observation, cellular respiration was measured as a marker of viability using the MTT assay 24 h after AS siRNA transfection. MTT assay also showed a 40% decrease in cell survival in AS-depleted endothelial cells (Fig. 2B). A reduction of AS protein beyond the levels achieved in this study was not possible because of extensive cell death associated with further increased concentrations of the AS siRNA.
LDH Release in AS-transfected BAEC-During necrosis, cells become permeable allowing the leakage of proteins into the media. A marker used to detect this phenomenon is the release of LDH into the culture medium (31). After treatment with AS and control siRNA, cell cultures were assayed for LDH activity in the media. Although treatment with AS siRNA resulted in a loss of cell viability, there was no significant difference in LDH activity detected in the media in AS siRNAtransfected cells as compared with the control (Fig. 3). These results suggested that the loss of cell viability resulting from partial AS depletion was not necrosis.
Reduction in Bcl-2 Levels in Cells with Reduced AS Expression-Because trypan blue exclusion analysis, as well as the absence of LDH release, suggested that necrosis was not the pathway directing cell death in cells with reduced AS expression, the possible involvement of apoptosis was investigated. Bcl-2 is known to be an important protein expressed in many cell types, including endothelial cells, that protects against apoptosis (32). To determine whether apoptosis accounted for the loss of viability in AS knockdown cells, lysates from cells transfected with AS and control siRNAs were prepared, and Bcl-2 expression was determined by standard Western blotting analysis. As shown in Fig. 4, a significant decrease in Bcl-2 protein was detected in AS-depleted cells as compared with control consistent with the suggestion that the observed loss of cell viability results from the induction of apoptosis.
Apoptosis-inducing Factor Is Unaffected by Depletion of AS-In a recently described pathway in endothelial cells, apoptosis-inducing factor (AIF) was identified as an important regulator of apoptosis that circumvents the requirement for caspase-3 activation in the induction of apoptosis (33). To examine whether increased levels of AIF result from AS siRNA treatment, cell lysates from AS and control siRNA-treated cells were subjected to Western blot analysis to monitor changes in AIF protein expression. As shown in Fig. 5A, there was no significant effect on AIF levels resulting from AS siRNA treatment that would account for the loss of cell viability induced by an apoptotic response.

TABLE I
Effect of AS siRNA transfection on endothelial NO production BAEC were transfected with AS siRNA or with control siRNA as described under "Experimental Procedures." Transfected BAEC were equilibrated in synthetic DMEM that contained either no arginine and 500 M citrulline or no citrulline and 500 M arginine. Basal NO production was determined over a 24-h period, and stimulated NO production was determined over a 1-h period following stimulation with 0.25 M A23187 and 20 M sodium orthovanadate. NO was measured as nitrite produced/1 ϫ 10 6 cells. (Nitrite is a stable reaction product of NO and molecular oxygen.) Control rates of NO production are not directly comparable because of different cell preparations.  Increase in Caspase-3/7 Activation in BAEC Depleted of AS-Because of the observed degradation of Bcl-2 protein and the lack of effect on AIF protein, the effect of AS silencing on caspase activation was investigated. Caspase-3 is a cysteine protease that on activation of the apoptotic cascade is proteolytically cleaved from an inactive procaspase form to an active caspase. Following partial AS depletion, there was an over 3-fold increase in caspase-3/7 activity compared with control cells (Fig. 5B). This finding strongly supports the proposal that the reduction of AS expression results in induction of apoptosis.
Treatment of AS siRNA-transfected BAEC with an NO Donor Prevents Apoptosis-Basal NO production in endothelial cells has been suggested to prevent the induction of apoptosis (34) first by decreasing the mRNA stability of MAP kinase phosphatase-3 (35) and second, by directly inactivating caspase-3 via S-nitrosylation of its active site thiol (36) (see Fig. 7). To determine whether a reduction in basal NO production caused by AS knockdown correlated with the observed increase in apoptosis, BAEC were treated with Glyco-SNAP-2 (Calbiochem), an NO donor with a half-life of 27.4 h. Glyco-SNAP-2 (25-400 M) was added to the media immediately following the transfection with AS siRNA. Forty-eight hours after transfection, caspase 3/7 activity was determined to quantitate the level of the apoptotic response. As shown in Fig. 6, AS siRNAinduced apoptosis in treated endothelial cells was decreased proportionately by the concentration of exogenous NO donor added. This correlation suggests that induction of apoptosis may be due, in part, to the reduction in basal NO production that results from AS knockdown, although a more direct effect of AS protein on apoptosis cannot be excluded. DISCUSSION NO production appears to be limited by the availability of arginine, despite extracellular and intracellular concentrations that are much higher than the reported K m for eNOS (13)(14)(15)(16)(17)(18)(19); thus, the source of arginine required to sustain NO production in endothelial cells has been investigated and debated. Part of this debate seemed to be resolved with the finding that the CAT1 transporter, the major transporter of arginine for endothelial cells, colocalizes with eNOS in caveolae (37). Thus, it was suggested that endothelial NO production was maintained through extracellular transport of arginine (37). This was a reasonable proposal because serum arginine levels normally vary from 80 to 90 M, well above the K m for eNOS. Nevertheless, other evidence persisted demonstrating that endothelial NO production was limited by the capacity to regenerate arginine from citrulline (13)(14)(15)(16)(17)(18)(19).
For example, our laboratory has shown previously (18) that extracellular citrulline was as effective as arginine in stimu- A, following siRNA transfection in 24-well plates, total protein was prepared and analyzed by SDS-PAGE and immunoblotted with anti-AIF antibody. B, BAEC were transfected with AS and control siRNA in 96-well plates. Twenty-four hours after transfection, caspase-3/7 activity was measured using Apo-ONE on a Fluostar Galaxy spectrofluorometer using an excitation of 490 nm and detecting at 520 nm. Control wells were transfected with liposome only. Results are presented as the increase in caspase activity over control.
lating NO production, even in media containing saturating levels of arginine (ϳ500 M). Because extracellular citrulline levels had no effect on intracellular arginine levels, these results suggested that citrulline enhancement of NO production was mediated through the regeneration of arginine directed to NO production and therefore could be observed despite saturating levels of extracellular arginine. Similarly, Wu et al. (38) showed that the synthesis of arginine from citrulline was stimulated by addition of exogenous citrulline. In this case, the authors speculated that the function of citrulline recycling was to salvage the carbon backbone to maintain sufficient L-arginine to sustain prolonged NO synthesis.
Other lines of evidence that supported the requirement of arginine regeneration for NO production include the overexpression of AS in vascular smooth muscle cells (13). As a result of AS overexpression there was a dramatic enhancement in the ability of the transfected smooth muscle cells to produce NO over that of untransfected cells, again despite saturating levels of extracellular arginine. Thus, Xie et al. (13,14) concluded that the capacity to recycle citrulline back to arginine is "ratelimiting" to NO production. Su et al. (17) arrived at a similar conclusion showing that hypoxia in pulmonary artery endothelial cells inhibited induction of AS by endotoxin. As a consequence, the production of NO, independent of sufficient extracellular arginine levels, was significantly impaired. Although our studies focused on the role of recycling for endothelial NO production by eNOS, previous studies have demonstrated the importance of recycling for NO production by both iNOS and nNOS (39 -49). For example, AS and inducible NOS are coinduced in immunostimulated macrophages (39,40) as well as in stimulated RPE-J cells where the citrulline-NO cycle is shown to be functioning (41). In addition, coinduction of iNOS, CAT-2, and AS in rat microglial cells indicates that both arginine transport by CAT-2 and citrulline-arginine recycling are important in the production of large amounts of NO in activated microglial cells (42). In neurons, colocalization of neuronal NOS, AS, and AL was identified in the canine gastrointestinal tract providing morphological evidence of a citrulline-NO cycle (43). Finally, in the rat gastric fundus, functional evidence of recycling is supported by colocalization of AS, AL, and nNOS (44).
Gene expression studies, using DNA microarray analyses, demonstrated that a significant and coordinate up-regulation of AS gene expression occurred in response to fluid shear stress stimulation of NO production by human umbilical vein endothelial cells (21). Because extracellular arginine was again not limiting in these studies, the authors concluded that a prerequisite for shear stress-induced NO production, in the absence of synthesis of additional eNOS, was an increase in arginine regeneration via increased AS expression (21). In other words, increased expression of AS resulted in the increased capacity to provide the necessary substrate to sustain elevated NO production associated with the shear stress response.
Our results provide further evidence supporting the necessity for the regeneration of arginine for NO production. Specifically, AS expression was demonstrated to be necessary and sufficient to maintain both stimulated and resting levels of NO synthesis in endothelial cells. However, the finding that reduction of AS expression in endothelial cells resulted in a decrease in cell viability was unexpected. Initially, we speculated that cell death may be a consequence of superoxide toxicity, which can result when eNOS is deprived of arginine (45). However, superoxide production was unaffected by partial AS depletion (data not shown). Moreover, because trypan blue exclusion and LDH release did not indicate that necrosis was the mechanism FIG. 6. Prevention of AS siRNA-induced apoptosis with an NO donor. BAEC were transfected with AS siRNA in 96-well plates. Immediately following transfection increasing concentrations of the NO donor, Glyco-SNAP-2, was added to the media. Caspase-3/7 activity was measured as an indicator of apoptosis.
FIG. 7. Mechanism of NO suppression of apoptosis. Activation of MAP kinase phosphatase-3 (MKP-3) induces ERK1/2 activation. ERK1/2 activation results in degradation of the apoptotic protective protein Bcl-2. NO prevents cascade activation by destabilizing MAP kinase phosphatase-3 RNA thus blocking the apoptotic cascade prior to cytochrome c release (35). In addition, NO can also inactivate caspase-3 through S-nitrosylation of the active cysteine (36). of cell death, we investigated the possibility that these ASdepleted endothelial cells were being driven to apoptosis.
NO has a bifunctional role in cell death; it can either stimulate or inhibit cytotoxicity. The level of NO produced and the type of cell involved determines the effect NO has on cell viability (5,6). High concentrations of NO have been shown to induce cell death via apoptosis. In a more complex pathway, NO can switch apoptosis to necrosis (4 -7). In contrast, lower concentrations of NO have been shown to protect cells such as endothelial cells (34), thymocytes (46), and lymphocytes (47) from apoptosis. In endothelial cells, induction of NO by sphingosine 1-phosphate protects endothelial cells from serum-deprived apoptosis (8).
There are several mechanisms that have been elucidated in the anti-apoptotic effects of NO (Fig. 7). Caspase-3 activation is inhibited by S-nitrosylation of the enzyme by NO (36,48). In addition, Bcl-2 cleavage is inhibited by NO as well as the subsequent release of cytochrome c (5,47). Furthermore, MAP kinase phosphatase-3 RNA is down-regulated via an NO-dependent mechanism, maintaining active ERK1/2 and thus protecting the cell from the apoptotic cascade (35). Finally, NO can induce apoptosis by directly inhibiting cytochrome c oxidase thus causing a decrease in the membrane potential of the mitochondria and release of cytochrome c (49). In our study, AS siRNA-induced apoptosis was modulated by the addition of an NO donor. This finding suggests that basal levels of NO in endothelial cells, sustained by the recycling of citrulline back to arginine, may provide protection against apoptosis. We recognize however that AS may be performing another function in the cell that is preventing apoptosis. The recent characterization of AS as a Jak-2 interacting protein (50), the regulation of AS by c-Myc (51), and the repression of AS expression by the antiproliferative drug FAP48 (52) warrant investigation into an alternative role of the enzyme in maintaining cell viability.
The maintenance of endothelial cell viability through basal NO production has important clinical implications. In atherosclerotic lesions there are regions characterized by reduced shear stress, which correlate with reduced NO synthesis (53,54). High endothelial cell turnover because of induction of apoptotic signaling pathways also characterize these regions (55). Importantly, NO production by endothelial cells may play a preventive role in atherosclerosis where cultured endothelial cell apoptosis is inhibited by increased endothelial NO production (56). All of these studies, however, demonstrate a protection from the induction of apoptosis, not simply maintenance of viability as we demonstrate in the current study. Basal levels of NO may provide maintenance of viability by keeping the proteins involved in cell cycle versus cell death in check, thus maintaining viability.
A decrease in endothelial NO production associated with endothelial dysfunction is a common feature in medical conditions such as hypertension, diabetes mellitus, and atherosclerosis. Supplementation with arginine in humans has produced conflicting results. In some studies, dietary arginine supplementation results in improved vascular function by providing substrate for eNOS and improving NO production (57)(58)(59). Other studies have not seen an improvement in condition, and harmful effects may occur (23, 60 -62). We demonstrate in this study that a functional citrulline-NO cycle is essential for endothelial NO production; therefore, targeting an increase in AS activity in endothelial cells may provide a pharmaceutical alternative for improving endothelial dysfunction.
In summary, NO production by endothelial cells plays an important role in the function of the endothelium and modulates cell survival signaling pathways such as Bcl-2 expression and caspase activity. This report supports our hypothesis that endothelial NO synthesis is dependent on the availability of a specific pool of arginine maintained through the conversion of citrulline to arginine by the enzymes AS and AL.