Critical Role of l-Arginine in Endothelial Cell Survival During Oxidative Stress
Background— Oxidative damage of vascular endothelium represents an important initiation step in the development of atherosclerosis. Recently, we reported about protection of inducible nitric oxide synthase (iNOS)-derived high-output NO in endothelial cells. Because iNOS activity critically depends on the availability of its substrate l-arginine, the present study aims at elucidating iNOS-mediated effects on H2O2-induced apoptosis of cytokine-activated rat aortic endothelial cells (AECs) subject to medium l-arginine concentrations.
Methods and Results— In cytokine-activated AECs, iNOS activity was found to be half-maximal at 60 μmol/L arginine, which represents the medium serum level in rats but also in humans. Maximal activity is seen at and above 200 μmol/L arginine. Activated cells grown in the absence of arginine with minimal iNOS activity are highly sensitive toward H2O2-induced apoptosis, and increases in medium arginine concentrations result in increased cell survival. Moreover, competition experiments show that iNOS activity is completely dependent on cationic amino acid transporter-mediated arginine uptake. We also find that the arginine-dependent protection includes inhibition of endothelial lipid peroxidation and increases in the expression of vasoprotective stress response genes.
Conclusions— Our data demonstrate that arginine concentrations corresponding to physiological serum levels do not allow for optimal endothelial iNOS activity. Thus, decreases in systemic arginine concentrations, or locally within atherosclerotic plaques, will impair the endothelial iNOS-mediated stress response and will significantly increase the risk of endothelial dysfunction.
Received December 31, 2002; accepted February 20, 2003.
Atherosclerosis is an excessive and inflammatory/proliferative response of the vascular wall to various forms of injury. During inflammation, reactive oxygen species (ROS)-induced endothelial cell damage represent an important event in the process of atherosclerotic lesion formation.1
Recent data provide strong evidence that high-output NO formation by the inducible NO synthase (iNOS) can exert powerful protection from oxidative stress-induced apoptotic or necrotic cell death. In endothelial cells, the basic mechanisms of these protective actions comprise NO-induced increases in expression of the antiapoptotic protein Bcl-2 as well as sustained cell membrane integrity and function as a consequence of NO-mediated inhibition of lipid peroxidation.2 These findings support the observations of increased atheroma incidence in iNOS-deficient mice3 and that overexpression of iNOS has been used successfully to inhibit experimental vascular lesion formation.4 Indeed, iNOS-derived NO has been recognized as an autoregulatory feedback inhibitor of vascular inflammation.5 Interestingly, in contrast to fibroblasts, for instance, even prolonged exposure to higher concentration of NO does not induce apoptosis of endothelial cells.6 Thus, iNOS activity seems to play a protective role for the vasculature, suggesting a therapeutic benefit for iNOS gene transfer.4
Appropriate endogenous iNOS activity during inflammation depends on the availability of its substrate l-arginine as well as expression and activity of the cationic amino acid transporters (CAT), which are responsible for l-arginine influx into the cell.7 In vivo, serum arginine levels range from 60 to 100 μmol/L in healthy human individuals,8 and similar values are described for rats.9 Substrate depletion, limited availability, or disturbed CAT function may lead to impaired endothelial iNOS-derived NO synthesis and impaired protection from atherosclerosis.
In this study, we examine the input of physiological arginine concentrations on the protective role of iNOS activity against H2O2-induced apoptosis of aortic endothelial cells.
We now show for the first time that in cytokine-activated endothelial cells, arginine supply at physiological levels restricts endothelial iNOS activity. Arginine concentrations corresponding to the lower physiological serum levels will decrease stress responses at the level of gene expression and will increase ROS-induced lipid peroxidation, thereby promoting and augmenting ROS-induced endothelial cell dysfunction.
Cell Cultures and Characterization
Rat aorta endothelial cells (AECs) were isolated by outgrowth from rat aortic rings and characterized exactly as described previously.10 Rat AECs showed the antigen phenotype vWFhigh Ox43high eNOShigh exactly as published.10
All measurements were performed with cells from passages 2 through 8. Endothelial cells (2×105) were cultured in 6-well tissue culture plates in 1000 μL RPMI 1640/20% FCS. After adherence, medium was removed and cells were cultured for 48 hours in l-arginine–free RPMI 1640 medium containing 2% FCS. Finally, cells were cultured for 24 hours in 1000 μL RPMI/2% FCS in the absence or presence of the respective additives at the concentrations indicated. ROS-mediated endothelial cell death was induced by an 18-hour incubation with H2O2 at the concentrations indicated. Inhibition of caspases after H2O2 challenge was obtained by incubation with the pan-caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (ZVAD) (30 μmol/L). Expression and activation of the iNOS was achieved by cytokine challenge with a mix of interleukin-1β, tumor necrosis factor-α, and interferon-γ (500 U/mL each) and inhibition of NOS activity by L-NIO (0.5 mmol/L). Arginase activity was inhibited by addition of valine (20 mmol/L). CAT-mediated arginine-influx was suppressed by incubation with lysine plus ornithine (10 mmol/L each).
Determination of Endothelial NO Production
After 24 hours of incubation, nitrite concentration was determined in culture supernatants using the diazotization reaction as described previously10 and using NaNO2 as standard. All values were normalized to the cell number and are given in micromolar nitrite produced from 1×105 cells in 1 mL medium in 24 hours.
Determination of Arginase Activity
Cellular arginase activity was determined indirectly by measuring the concentration of urea nitrogen in culture supernatants and using the urea detection kit (Berthelot Determination) and urea as standard following the manufactures instructions.
Polymerase Chain Reaction
Total cellular RNA (1 μg each) prepared from resting or cytokine-activated cells grown was used for cDNA synthesis using the dT16-oligonucleotide as primer. Reverse transcription was carried out at 37°C for 60 minutes. The cDNA (500 ng each) was used as template for polymerase chain reaction (PCR) primed by using the oligonucleotides and cycle protocols shown in the Table. The PCR conditions used were obtained from a cycle-controlled analysis and ensure that amplification of the gene products mentioned occur within the linear phase. The cycle-controlled analysis was performed by running PCR for each gene within the intervals of 10 to 40 cycles. Then aliquots of amplified products were subjected to electrophoresis on 1.8% agarose gels, and densitometric analysis of the amplification products (visualized by ethidium-bromide staining) was performed by using the KODAK 1D software (KODAK). In the cycle-controlled analysis, values were calculated as percent of the maximal amplification (=100%). Otherwise, product/GAPDH ratios were calculated and values were expressed as relative increases or decreases compared with the controls.
Determination of Growth Rates and Viability and Detection of Cell Death
Cell growth or the number of live cells was determined at different times by neutral red (NR) staining.11 The cultures were washed twice with serum-free medium and were then incubated for 60 minutes in the presence of the growth medium containing NR (50 μg/mL) to allow uptake of the vital dye into the lysosomes of viable uninjured cells. Thereafter, the NR medium was removed and the cells were rinsed with PBS (pH 7.4 at 37°C) to remove unincorporated stain. Cells were dried for 1 hour at room temperature, and the destaining solution (1% 1N HCl in isopropanol), 1.0 mL, was added to each well to remove the NR from the cells. The absorbance of the solution in each well was read at 530 nm in a spectrophotometer using the destaining solution alone as the blank. Additionally, viability of endothelial cells was assessed routinely at the beginning and the end of every experiment using the trypan blue exclusion assay. Apoptosis or necrosis was detected by using the Hoechst dye H33342 and by detecting DNA strand breaks in acetone-fixed cells by the in situ nick-translation method or was determined by incorporation of the red fluorescence dye propidium iodide exactly as described by us recently,12 respectively. In each cell culture sample, a minimum of 500 cells were counted, and apoptotic or necrotic cells were expressed as percent of total cell number.
Determination of Lipid Peroxidation
Resting endothelial cells (2×107) were incubated for 18 hours with H2O2 at concentrations indicated in the absence or presence of the respective additives at the concentrations indicated. Then lipid peroxidation was stopped by addition of butylated hydroxytoluene (BHT, 10 μmol/L). Cells were lysed by repeated freezing and thawing. Lipid peroxidation was measured by determination of thiobarbituric acid reactive substances with high performance liquid chromatography (HPLC) and expressed as malondialdehyde (MDA) equivalents exactly as described previously.13
Values were reported as mean±SD. For statistical analysis, we used ANOVA followed by an appropriate post hoc multiple comparison test (Tukey method). P<0.05 was considered significant.
Endothelial Cells Express iNOS, Arginases, and Arginine Transporters
First, endothelial expression of l-arginine metabolizing or transporting proteins under resting or activated conditions was evaluated by reverse transcriptase-PCR. An iNOS-specific cDNA amplification product is found in cytokine-challenged cells only. In contrast, constitutive expression of mRNA for arginase-1, arginase-2, and the cationic amino acid transporters CAT-1 and CAT-2 is seen and, with the exception of CAT-2, remains unchanged during cytokine-activation (Figure 1A). Additionally, resting cells exhibit a small NOS activity (0.3±0.1 μmol/L nitrite in culture supernatant), whereas cytokine challenge results in a substantial increase in NO formation (17.1±6.2 μmol/L nitrite in culture supernatant). In contrast and irrespective of the presence of cytokines, identical accumulation of urea in supernatants as equivalents for arginase activities was found throughout the experiments (resting cells, 520±120 μmol/L of urea or 480±90 μmol/L of urea in cytokine-activated cells) (Figure 1B).
Cytokine-Induced High-Output NO Formation Protects From Oxidative Stress–Induced Cell Death
Culturing of resting AECs at the high physiological arginine concentration of 200 μmol/L in the presence of increasing hydrogen peroxide concentrations results in concentration-dependent cell death as determined 18 hours after hydrogen peroxide addition. Half-maximal cytotoxicity was observed at 0.6±0.05 mmol/L H2O2 and maximal death of >80% of the cells at and above 0.8 mmol/L H2O2 (Figure 2A). Cytokine activation with concomitant iNOS-mediated high-output NO synthesis fully protects cells from peroxide-induced death, an effect not seen when activation is performed in the presence of the NOS inhibitor NIO (0.5 mmol/L).
Cell death occurs via apoptosis, as confirmed by staining with the Hoechst dye (Figures 2D and 2E) or by detecting DNA strand breaks using in situ nick translation (Figures 2F and 2G). Significant inhibition of cell death by the pan-caspase inhibitor ZVAD (Figure 2B) additionally supports apoptosis as the mode of cell death. Additionally, inhibition of cell death by addition of butylated hydroxytoluene (BHT, 10 μmol/L), an inhibitor of lipid peroxidation (Figure 2B), indicates that lipid peroxidation represents an initiating event. Indeed, with resting cells, H2O2 challenge leads to a marked increase in lipid peroxidation, as monitored by detection of malondialdehyde formation (Figure 2C). This increase was inhibited by the addition of catalase (2000 U/mL) or of BHT (10 μmol/L). In contrast, after cytokine challenge with concomitant iNOS expression and high-output NO synthesis, hydrogen peroxide–induced lipid peroxidation was completely suppressed. Inhibition of NOS activity by adding L-NIO (0.5 mmol/L) again abrogated this protective effect. Despite the presence of both oxygen-derived radicals and NO in these experiments, absolutely no protein nitration, as an indicator for peroxynitrite anion formation, occurs, as verified by staining cell lysates with a specific antibody for nitrotyrosine in dot blots (data not shown).
Exogenous Arginine Concentrations or Transport Regulates iNOS or Arginase Activity Within the Physiological Range
Next we examined the effect of various physiological l-arginine concentration on NOS and arginase enzyme activities. With resting cells, no increase in nitrite formation was found irrespective of the arginine concentration used. After cytokine challenge, the increases in NO formation were strongly dependent on exogenous arginine concentrations. Half-maximal activity was found at 60 μmol/L of arginine and a maximum at and above 200 μmol/L arginine (Figure 3A).
Within the range of exogenous arginine concentrations tested, arginase activities in resting or cytokine-activated cells are similar (Figure 3B). In the presence of valine (20 mmol/L), a potent inhibitor of arginase activity, endothelial arginase activity was suppressed (Figure 3B; 1±1% activity at 80 μmol/L versus 17±4% activity at 1000 μmol/L of arginine), with no effect on iNOS activity (Figure 3A). Minimizing arginine transport via competition by the amino acids lysine and ornithine either applied singly (data not shown) or in combination results in a strong inhibition of both iNOS as well as arginase activity (Figure 3).
Exogenous Arginine Regulates H2O2-Induced Cell Death
Because a strong correlation between different physiological l-arginine concentrations and iNOS-mediated NO production was found in live cells, we now examined the impact of these substrate concentrations on oxidative stress–induced cell death. A close correlation between the available arginine and protection from H2O2-induced (18 hours with 0.8 mmol/L) cell death was observed. Thus, cytokine-activated AECs grown in the absence of arginine are as sensitive to H2O2-induced death as are resting cells at any arginine concentration, whereas cells grown in the presence of 200 μmol/L arginine are fully protected from H2O2-induced cell death (Figure 4A). Additional evidence for the essential role of arginine supply or utilization on cell protection from ROS-mediated apoptosis is shown in Figure 4B. A substantial diminution of arginine transport by lysine/ornithine-mediated competition of CAT-transporter–mediated arginine influx also impairs endothelial survival. The protective effects can be fully restored by addition of the NO donor DETA/NO (1 mmol/L), demonstrating that the arginine-dependent cell survival is attributable to high-output NO synthesis.
Exogenous Arginine Regulates H2O2-Induced Lipid Peroxidation
Arginine concentration-dependent and iNOS activity-mediated protection from cell death parallels the degree of H2O2-induced lipid peroxidation. As shown in Figures 2 and 4⇑C, H2O2 challenge of resting cells strongly augments lipid peroxidation. Cytokine-activated AEC cultures grown at 200 μmol/L arginine with maximal iNOS activity lack this peroxide-induced increase in lipid peroxidation. Furthermore, decreasing medium-arginine concentrations lead to corresponding increases in lipid peroxidation. Complete arginine deprivation allows for lipid peroxidation to a degree equivalent to resting and peroxide-challenged cells. The observed protection at 200 μmol/L of arginine was completely reversed by addition of L-NIO (0.5 mmol/L) but also after competition of CAT transporter-mediated arginine import (lysine plus ornithine, 10 mmol/L each). Again, exogenous NO generation (DETA/NO, 1 mmol/L) leads to protection of these cultures.
Impact of Arginine Concentration on the Expression of Endothelial Stress Response Genes
High-output NO synthesis is also known to modulate cellular gene expression. We therefore examined whether such modified gene usage is altered within these physiological l-arginine levels. As examples for NO-modulated genes, we analyzed the expression levels of bcl-2, HO-1, and vascular endothelial growth factor (VEGF). Protection from H2O2-induced cell death completely parallels the expression of the 3 vasoprotective genes (Figures 5 and 6⇓). Compared with resting AEC cultures, cytokine challenge with resulting high-output NO synthesis in the presence of 200 μmol/L of arginine increases bcl-2-mRNA expression by the factor of 3.5±0.4 and HO-1 or VEGF mRNA expression by 2.0±0.2-fold or 2.3±0.6-fold, respectively. Lowering of medium-arginine concentration or competitive reduction of arginine utilization (0.5 mmol/L L-NIO) or inhibition of arginine import (lysine plus ornithine, 10 mmol/L each) all decreased mRNA expression of these 3 genes toward control values. Again, DETA/NO challenge (1 mmol/L) restored the increases in bcl-2, HO-1, and VEGF mRNA expression levels.
Reactive oxygen species (ROS), oxidized low-density lipoprotein, and inflammatory stress have been shown to induce apoptosis of endothelial cells, which seems to represent an initial step in the development of atherosclerotic lesions.14 Recently, we demonstrated that ROS-induced endothelial cell death can be completely inhibited by endogenously produced high-output NO synthesis or by exogenously applied NO through inhibition of cytochrome c leakage, increases in bcl-2-gene expression, and inhibition of lipid peroxidation.2
We here show for the first time that in live endothelial cells, physiological serum arginine concentrations do not support maximal/optimal iNOS-protective activity during proinflammatory conditions. Indeed, normal arginine concentrations of 60 to 120 μmol/L result in 50% to 70% of the maximal NO production only. At first glance, this is a surprising result, because the Km values determined on isolated NOS enzymes are ≈10 μmol/L of arginine. Thus, arginine levels of 60 μmol/L should be well above the level needed for maximal activity. This so-called l-arginine paradox seems to be the result of additional factors influencing the substrate availability for iNOS-mediated NO synthesis. These factors comprise the presence of the endogenous inhibitors like asymmetric dimethylarginine or Nω-monoethyl-l-arginine as well as transport competition with the other cationic amino acids and substrate competition by arginases and arginine-decarboxylase activities plus protein synthesis.15
However, our data demonstrate a linear increase of iNOS activity with increasing l-arginine concentrations that are within the normal serum levels. Therefore, even small concentration changes within this physiological range will lead to significantly altered iNOS-mediated NO production and thus altered antioxidative effects. For example, arginine levels of 100 μmol/L will allow for suppression of lipid peroxidation twice as effective as arginine levels of 60 μmol/L, with a concomitant decrease of stress-relevant gene expression.
Lipid peroxidation can be limited by inhibiting the generation of or by quenching the initiating radical species by antioxidants. The NO radical reacts rapidly via simple radical-radical combination reactions with species possessing unpaired electrons. However, as shown by us recently,12 experimental data on endothelial cells do not corroborate such a role for NO in H2O2-mediated cell death but rather suggest a radical-chain terminating activity.16
Local tissue-arginine deprivation, for instance attributable to active arginases of infiltrating macrophages, is known to occur at inflammatory sites and in wound healing.17 Such a local arginine depletion or low systemic arginine levels will lead to reduced activity and thus impaired NO formation by NO synthases,18 which in vessels may then contribute to the initiation and progression of atherosclerosis, as hypothesized recently.19 Although arginase as well as iNOS activities are expressed in parallel in activated endothelial cells, they strongly differ in their substrate kinetics from results obtained with macrophages.20 Interestingly, and in contrast to these cells, in endothelial cells arginine utilization by arginases seems not to be rate limiting for iNOS activity.21
The Janus-faced properties of NO has prompted a debate for whether NO plays a deleterious or a protective role in tissue injury. As regards the development of the atherosclerotic plaque, the activation of iNOS may have complex effects. It has been hypothesized that induction of cell death in macrophages and vascular smooth muscle cells may contribute to the development of the necrotic core in lesions. On the other hand, protection of endothelial cell death by attenuating lipid peroxidation may represent the major pathway by which NO limits oxidative injury. In addition, NO-mediated increases in metalloproteinase activities may contribute to fibrous dissolution and reduce matrix deposition.22,23
In summary, we here give evidence that endothelial iNOS-derived high-output NO synthesis does not contribute to destruction but stands for an arginine-dependent cell-protective mechanism. Thus, our data support the notion that strategies to enhance NO synthesis or activity via iNOS will be useful in maintaining cardiovascular health24 and are consistent with the observation that supplementation of dietary arginine may delay or reduce atheroma formation.25
This study was supported by a grant from the Deutsche Forschungsgemeinschaft, SFB 503 A3 and SFB 575 B2 to Dr Kolb-Bachofen, and SFB 503 B1 and SFB 575 B4 to Dr Sies. Dr Sies is a Fellow of the National Foundation for Cancer Research (NFCR), Bethesda, Md. The authors thank Marija Lenzen for technical assistance.
Oxidative damage of vascular endothelium represents an important initiation step in the development of atherosclerosis. We here show that endothelial cytokine-inducible NO synthase–derived high-output NO exerts full protection against reactive oxygen species–induced cell death and increases protective stress gene expression. Endothelial inducible NO synthase activity critically depends on exogenous l-arginine availability. Surprisingly, arginine concentrations corresponding to physiological serum levels may not always allow for maximal endothelial inducible NO synthase activity. Thus, small decreases in systemic or local arginine concentrations will impair the endothelial inducible NO synthase–mediated stress response and will significantly increase the risk of endothelial dysfunction.
von der Leyen HE, Dzau VJ. Therapeutic potential of nitric oxide synthase gene manipulation. Circulation. 2001; 103: 2760–2765.
Peng HB, Spiecker M, Liao JK. Inducible nitric oxide: an autoregulatory feedback inhibitor of vascular inflammation. J Immunol. 1998; 161: 1970–1976.
Closs EI, Scheld JS, Sharafi M, et al. Substrate supply for nitric-oxide synthase in macrophages and endothelial cells: role of cationic amino acid transporters. Mol Pharmacol. 2000; 57: 68–74.
Dimmeler S, Zeiher AM. Endothelial cell apoptosis in angiogenesis and vessel regression. Circ Res. 2000; 87: 434–439.
Gobert AP, Daulouede S, Lepoivre M, et al. l-Arginine availability modulates local nitric oxide production and parasite killing in experimental trypanosomiasis. Infect Immunol. 2000; 68: 4653–4657.
Kolpakov V, Gordon D, Kulik TJ. Nitric oxide-generating compounds inhibit total protein and collagen synthesis in cultured vascular smooth muscle cells. Circ Res. 1995; 76: 305–309.
Cannon RO III. Role of nitric oxide in cardiovascular disease: focus on the endothelium. Clin Chem. 1998; 44: 1809–1819.
Wang BY, Ho HK, Lin PS, et al. Regression of atherosclerosis: role of nitric oxide and apoptosis. Circulation. 1999; 99: 1236–1241.