N-Acetylcysteine Decreases Angiotensin II Receptor Binding in Vascular Smooth Muscle Cells

Michael E. Ullian, Andrew K. Gelasco, Wayne R. Fitzgibbon, C. Nicole Beck, and Thomas A. Morinelli Medical University of South Carolina, Ralph H. Johnson Veterans Administration Hospital, Charleston, South Carolina

Antioxidants seem to inhibit angiotensin II (Ang II) actions by consuming stimulated reactive oxygen species. An alternative hypothesis was investigated: Antioxidants that are also strong reducers of disulfide bonds inhibit the binding of Ang II to its surface receptors with consequent attenuation of signal transduction and cell action. Incubation of cultured vascular smooth muscle cells, which possess Ang II type 1a receptors, with the reducing agent n-acetylcysteine (NAC) for1hat37°C resulted in decreased Ang II radioligand binding in a concentration-dependent pattern. NAC removal restored Ang II binding within 30 min. Incubation with n-acetylserine, a nonreducing analogue of NAC, did not lower Ang II binding, and oxidized NAC was less effective than reduced NAC in lowering Ang II binding. NAC did not decrease Ang II type 1a receptor protein content. Other antioxidants regulated Ang II receptors differently: ␣-Lipoic acid lowered Ang II binding after 24 h, and vitamin E did not lower Ang II binding at all. NAC inhibited Ang II binding in cell membranes at 21 or 37 but not 4°C. Dihydrolipoic acid (the reduced form of ␣-lipoic acid), which contains free sulfhydryl groups as NAC does, decreased Ang II receptor binding in cell membranes, whereas ␣-lipoic acid, which does not contain free sulfhydryl groups, did not. Ang II–stimulated inositol phosphate formation was decreased by preincubation with NAC (1 h) or ␣-lipoic acid (24 h) but not vitamin E. In conclusion, certain antioxidants that are reducing agents lower Ang II receptor binding, and Ang II–stimulated signal transduction is decreased in proportion to decreased receptor binding. J Am Soc Nephrol 16: 2346–2353, 2005. doi: 10.1681/ASN.2004060458

he octapeptide angiotensin II (Ang II) contributes to be altered if antioxidants regulate Ang II surface binding, be- maintenance of blood pressure (BP) and, in excess, cause second messenger responses to Ang II are tightly coupled T causes hypertension. Binding of Ang II to specific cell to Ang II surface receptor density (21–25). surface receptors initiates a signal transduction cascade that The literature regarding regulation of Ang II receptor bind- culminates in contraction, hypertrophy, and sodium reabsorp- ing by antioxidants is limited and conflicting. Because the tion. Most Ang II actions are transduced through the Ang II chemical reducing agent dithiothreitol lowers binding capabil- type 1 (AT1) receptor. Recent studies have demonstrated that ity of AT1 receptors (26,27), presumably via interaction of its reactive oxygen species (ROS) are important mediators of Ang sulfhydryl groups with disulfide bonds in the ligand-binding II action. Ang II activates enzyme systems that create ROS (1–3), moiety of the receptor, the antioxidant/reducing agent n-ace- and ROS are formed in response to Ang II in a time- and tylcysteine (NAC) might act similarly. However, in cultured concentration-dependent pattern (4–8). ROS are intermediates vascular smooth muscle cells (VSMC), incubation with NAC at in the Ang II signal transduction cascade (9–12), antioxidants 4°C did not alter Ang II radioligand binding (28), and the block Ang II–stimulated signal transduction (13–15), and ROS antioxidant vitamin E, which is not a reducing agent, decreased mediate Ang II–related vascular diseases (16–20). AT1 receptor mRNA expression in peritoneal macrophages The mechanisms by which antioxidants interfere with Ang II (29). Alternatively, ROS have been shown to decrease Ang II signal transduction have been only partially elucidated. The receptor mRNA levels and Ang II receptor density in VSMC prevailing hypothesis is that antioxidants consume Ang II– (28,30). Antioxidants might reverse this mechanism, resulting stimulated ROS and thus prevent distal Ang II signaling. In this in increased Ang II receptor mRNA levels and receptor density. study, we addressed antioxidant interaction with the initial Because of this conflicting literature, we performed studies in segment of the Ang II signal transduction cascade, the Ang II cultured VSMC to determine whether antioxidant compounds surface receptor. Ang II–stimulated signal transduction would decrease Ang II surface receptor binding. We used antioxidants that are potent chemical reducing agents (NAC), that can be converted to reducing agents (␣-lipoic acid), and that are not Received June 9, 2004. Accepted April 24, 2005. reducing agents (vitamin E).

Published online ahead of print. Publication date available at www.jasn.org. Materials and Methods Address correspondence to: Dr. Michael E. Ullian, Medical University of South Carolina, Division of Nephrology, CSB 829, 96 Jonathan Lucas Street, P.O. Box Cultured Cells 250623, Charleston, SC 29425. Phone: 843-792-4123; Fax: 843-792-8399; E-mail: Most experiments were performed on VSMC, which contain Ang II [email protected] AT1a receptors. Under pentobarbital anesthesia, aortas from 150-g male

Sprague-Dawley rats were partially digested, adhered to culture flasks, Immunoblotting and covered with minimal essential medium that contained 10% new- VSMC proteins were solubilized and subjected to SDS-PAGE under born calf serum, 1% nonessential amino acids, 100 U/ml penicillin, and reducing conditions. After semidry transfer to polyvinylidene difluo- ␮ 100 g/ml streptomycin. Cells were incubated in humidified 5% CO2/ ride membranes, blocking buffer (5% defatted dried milk in 10 mM 95% air atmosphere until confluent. Medium was changed every 5 d, Tris, 150 mM NaCl, 1% Tween-20 [pH 8.0]) was added. Membranes and cells were passaged every 7 to 10 d. Studies were performed on were incubated overnight with primary antibody in blocking buffer. cells in passages 3 to 12. Cells exhibited characteristic stellate morphol- Membranes were washed, exposed to secondary antibody for 1 h, and ogy and stained for ␣-smooth muscle actin (31). Endothelial cell con- then washed again. Immunoreactive bands were visualized by the CDP tamination was minimal (25). For this and other animal studies, we Star chemiluminescent method (New England Biolabs, Beverly, MA). followed National Institutes of Health guidelines, and they were ap- Bands were scanned and quantitated with ScanAnalysis densitometry proved by our Institutional Animal Care and Use Committee. software (Elsevier/BIOSOFT, Ferguson, MD).

Human embryonic kidney (HEK) cells were transfected with DNA Immunoblotting of AT1a receptors was performed with 1:1000 dilu- coding for an AT1a receptor–green fluorescence protein (GFP) fusion tion of a polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz,

protein (AT1aR-GFP). DNA was transfected with Lipofectamine (In- CA). Specificity of the antibody was confirmed by us in preliminary ␮ 2 vitrogen, Carlsbad, CA), using 9 g of plasmid DNA per 100-cm flask studies: Rat VSMC, which contain AT1a receptors, displayed specific ␮ 125 or 1 g per cell culture well in 80% confluent cells. DNA for AT1aR-GFP I–Ang II binding, inositol phosphate responses to Ang II, and repro- was generated as described (32). The plasmid pCa18b, containing the ducible bands on immunoblot, whereas A7r5 cells, which do not con- 125 rat AT1a receptor cDNA, was amplified by PCR and cloned into pCR tain AT1a receptors, displayed no specific I–Ang II binding, no ino- II-Topo (Invitrogen). The insert was excised by Xho1/BamH1 and sub- sitol phosphate responses to Ang II, and no bands on immunoblot. cloned into the expression vector pEGFP-N3 (Clontech, Palo Alto, CA) Secondary antibody was goat anti-rabbit IgG conjugated to alkaline

to form the AT1aR-GFP construct. Fidelity of expression was verified by phosphatase at 1:1000 dilution. Immunoblotting of AT1aR-GFP was sequencing. Transfected cells were selected with F12 medium that performed with 1:500 dilution of a mAb against GFP (BD Biosciences, contained 400 ␮g/ml G418. Palo Alto, CA), and the secondary antibody was goat anti-mouse IgG conjugated to alkaline phosphatase at 1:1000 dilution.

Radioligand Binding Oxidation of NAC Binding studies were performed on confluent VSMC in duplicate NAC was oxidized by bubbling the reduced NAC stock solution with wells of 24-well plates. Binding buffer consisted of 50 mM Tris, 100 mM airfor1hatroom temperature, and analysis by atmospheric pressure NaCl, 5 mM KCl, 5 mM MgCl2, 0.25% BSA, and 0.5 mg/ml bacitracin chemical ionization mass spectrometry was performed. A single peak (pH 7.4). Incubation volume was 0.3 ml. Single concentration Ang II with a molecular weight of 163.9 was observed in the reduced stock 125 receptor binding studies were performed by adding 50 fmol of I–Ang sample, whereas the 1-h oxidation period resulted in Ͼ90% loss of the ␮ II (New England Nuclear, Boston, MA) to all wells and 1 M unlabeled 163.9 peak and the appearance of multiple new peaks of larger and Ͻ Ang II to certain wells (to determine nonspecific binding, 15% of smaller sizes. total). Competition binding studies were performed by adding 125I– Ang II (50 fmol) to all wells and 6 concentrations of unlabeled Ang II Protein Content (0.5 nM to 10 ␮M) to various wells. Studies were performed at 4°C for Cellular protein was measured by the Lowry method (34). 90 min to obtain binding equilibrium without receptor internalization (33). Free hormone was removed by washing three times with cold saline. Cells were solubilized with 0.1% SDS/0.1 N NaOH, and gamma Statistical Analyses Data were expressed as mean Ϯ SEM. Comparisons of two group radioactivity was counted. Receptor density and binding affinity were means were performed by unpaired t test, and comparisons of three or determined by Scatchard analysis. Ang II binding studies were also more group means were performed by ANOVA. Differences in slopes performed on VSMC membranes. Membranes were prepared from were assessed by trend analysis. Significance was assigned to the 0.05 VSMC with a polytron and aliquotted into duplicate test tubes (3 to 5 level. mg of cell protein per test tube), and bound hormone was isolated from unbound hormone with a suction manifold. Methods for arginine vasopressin radioligand binding were the same as those described Results above for Ang II. Ang II Binding in Intact VSMC Incubation with NAC for 1 h resulted in concentration-dependent decreases in 125I–Ang II surface binding (Figure 1)ina Inositol Phosphate Formation protocol that consisted of exposure to NAC at 37°C followed by VSMC in six-well plates were incubated with 5 to 10 ␮Ci of [3H]i- radioligand binding at 4°C. When NAC exposure was extended nositol/well in inositol-deficient growth medium for 24 h. Steady-state beyond 1 h, decreases in binding persisted (Figure 2). NAC at 3 uptake of [ H]inositol after 24 h was 300,000 to 500,000 cpm/well. After 1 mM, a concentration achievable in blood after intravenous exposure to effectors, reactions were terminated with 20% TCA. Protein administration (35), lowered Ang II binding modestly (12%; precipitates were discarded, and supernatants were extracted three Figure 2). Removal of NAC allowed recovery of Ang II binding times with diethyl ether. The upper ether phase was discarded. Sam- within 30 min (Figure 2). Inclusion of 10 mM NAC in binding ples were adjusted to pH 7 and transferred to AG1-X8 anion exchange resin columns (Bio-Rad, Hercules, CA). Radioactivity elutable with buffer at 4°C (rather than a 37°C preincubation) failed to lower Ϯ water and 5 mM sodium borate–60 mM sodium formate was dis- Ang II binding (102 4% of binding from VSMC unexposed to ϭ carded. Total inositol phosphates were eluted with 1.0 M ammonium NAC; n 6), as others have reported (28). To determine formate in 0.1 M formic acid. Fractions were counted in a scintillation whether other antioxidants decrease Ang II binding, we ex- counter. posed VSMC to vitamin E (nonreducing) or ␣-lipoic acid (be- 2348 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2346–2353, 2005

II binding (Table 1), demonstrating the importance of the sulfur atom rather than the general structure of NAC in decreasing the

level of functional AT1a receptors. That oxidized NAC was much less effective than reduced NAC in lowering Ang II binding (Table 1) highlights the reducing properties of NAC. We also studied the ability of NAC to alter binding of Ang II to a fusion protein that consisted of the coding region (but not the

regulatory region) of the AT1a receptor and GFP (AT1aR-GFP), shown previously to bind Ang II and transduce its signal (32). Treatment of HEK cells that were transfected with DNA for

AT1aR-GFP with NAC for1hat37°C resulted in concentration- dependent decreases in Ang II radioligand binding (99 Ϯ 4% of control [1 mM NAC], 89 Ϯ 5% of control [5 mM NAC], 70 Ϯ 9% of control [10 mM NAC]; n ϭ 6; P Ͻ 0.05, for 5 and 10 mM NAC compared with control). NAC did not lower Ang II receptor

protein content as demonstrated by immunoblotting for AT1a

receptors in VSMC and for AT1aR-GFP in transfected HEK cells Figure 1. Effects of N-acetylcysteine (NAC) and vitamin E on (Figure 5). angiotensin II (Ang II) radioligand surface binding in vascular smooth muscle cells (VMSC). Serum-deprived VSMC were exposed to vehicle, NAC, or vitamin E for various periods at 37°C, followed by specific Ang II radioligand binding for 90 min at 4°C. Binding in cells that were treated with vehicle is defined as 100% (control) and is represented by the dashed line. *Differences from control at P Ͻ 0.01. This experiment was performed four to five times.

comes a reducing agent when modified intracellularly) before 125I–Ang II binding. Vitamin E did not alter Ang II binding after brief or prolonged incubations (Figure 1). In contrast, ␣-lipoic acid or its reduced derivative dihydrolipoic acid that contains sulfhydryl groups elicited concentration-dependent decreases in Ang II surface receptor binding; longer incubations were required for ␣-lipoic acid, whereas dihydrolipoic acid was effective after shorter exposures (Figure 3). To determine the specificity of NAC targets, we also studied arginine vasopressin radioligand binding. In contrast to Ang II binding, arginine vasopressin binding was not decreased by1hofNACat37°C (89 Ϯ 5% of control [1 mM NAC], 90 Ϯ 6% of control [2 mM NAC], 107 Ϯ 6% of control [5 mM NAC], 120 Ϯ 7% of control [10 mM NAC]; n ϭ 5).

Mechanisms of NAC Effect on Ang II Binding Because NAC is acidic (it must be neutralized before use), acidity of the medium could have attenuated ligand–receptor interaction and lowered 125I–Ang II binding. After1hofincu- bation with vehicle or 10 mM NAC at 37°C, neither color (pH indicator) nor pH of the medium was different between treat- ments. NAC did not induce dehiscence of VSMC from the Figure 2. Time-dependent regulation of Ang II radioligand surface binding by NAC in VMSC. Serum-deprived VSMC were exposed culture plate, because cell protein content (␮g/well) was not to vehicle or various concentrations of NAC for various periods at different after NAC (222 Ϯ 16) or vehicle (204 Ϯ 9; n ϭ 6 wells). 37°C. In some cells, NAC was removed and replaced by culture Scatchard plots of binding data after 10 mM NAC for1hat medium lacking NAC. Then specific Ang II radioligand binding 37°C revealed no change in slope with decrease in x intercept for 90 min at 4°C was performed. Binding in cells that were treated (Figure 4), suggesting that NAC lowered surface receptor den- with vehicle is defined as 100% (control) and is represented by the sity rather than binding affinity. N-acetylserine, identical to dashed line. *P Ͻ 0.01 and **P Ͻ 0.05 versus control. This experi- NAC except for oxygen in place of sulfur, did not decrease Ang ment was performed four to eight times. J Am Soc Nephrol 16: 2346–2353, 2005 NAC Lowers Angiotensin Receptor Binding 2349

Figure 4. Scatchard plot of Ang II radioligand binding data from VSMC that were treated with NAC. Serum-deprived VSMC were exposed to vehicle or 10 mM NAC for1hat37°C. Then 125I–Ang II and increasing concentrations of unlabeled Ang II were added for 90 min at 4°C so that Scatchard plots of competition binding could be generated. Data represent mean val- ues from nine individual studies. Slopes of lines do not differ, whereas x intercepts were different at P Ͻ 0.05.

was added for 1 min to stimulate inositol phosphates. Ang II stimulated inositol phosphates by 92% in the absence of antioxidants. Antioxidant conditions that resulted in decreased Ang II binding (NAC for1hor␣-lipoic acid for 24 h) resulted in less Ang II–stimulated inositol phosphate formation, Figure 3. Effects of lipoic acid compounds on Ang II radioligand whereas a condition that did not alter Ang II binding (24 h of surface binding in VMSC. Serum-deprived VSMC were ex- exposure to vitamin E) did not lower Ang II–stimulated inositol posed to ␣-lipoic acid (A) or dihydrolipoic acid (B) for various phosphate formation (Figure 7). periods at 37°C, followed by specific Ang II radioligand binding for 90 min at 4°C. Binding in cells that were treated with vehicle is defined as 100% (control) and is represented by the Discussion dashed line. *Difference of line from control at P Ͻ 0.05. This Because ROS are second messengers for Ang II action (1–15), experiment was performed four to five times. inactivation of ROS by antioxidants within the cell should interrupt the Ang II signal transduction cascade and attenuate Ang II action. However, we considered that antioxidants might Ang II Binding in Membranes also regulate Ang II action by another mechanism: Altering When NAC was added to VSMC membranes, Ang II binding surface receptor binding for Ang II, the initial step in Ang II was decreased when experiments were performed at warm signaling. temperatures (markedly at 37°C, less pronounced at 21°C) but Certain antioxidants that are also potent chemical reducers of not at all at 4°C (Figure 6A). In similar experiments with VSMC disulfide bonds via their free sulfhydryl groups might interact membranes at 37°C, dihydrolipoic acid, which contains sulfhy- with the disulfide bonds of the Ang II receptor, alter its tertiary dryl groups, effected concentration-dependent decreases in structure, and inhibit Ang II–AT1a receptor interactions. The

Ang II binding, but ␣-lipoic acid, which does not contain sulf- AT1a receptor contains two disulfide bonds in the extracellular hydryl groups, did not alter Ang II binding (Figure 6B). region of the receptor: C18-C274 and C101-C180 (36–39). Two of the antioxidants used in our study, NAC and dihydrolipoic Effects on Signal Transduction acid, can reduce disulfide bonds. Vitamin E and ␣-lipoic acid Cells were exposed to antioxidants under conditions used do not contain sulfhydryl groups, although the latter can be earlier in this study: 10 mM NAC for 1 h, 1 mM ␣-lipoic acid for reduced to dihydrolipoic acid after cellular uptake. This may 24 h, and 100 ␮g/ml vitamin E for 24 h. Then 100 nM Ang II explain why ␣-lipoic acid was delayed in its ability to decrease 2350 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2346–2353, 2005

Table 1. Mechanisms of NAC effects on Ang II bindinga

Ang II Binding P Experimental Maneuver (% of Control) Comparison Value N

Control 100 — 8 10 mM n-acetylserine 98 Ϯ 4 With control NS 8 10 mM reduced NAC 32 Ϯ 2 With control Ͻ0.01 4 10 mM oxidized NAC 66 Ϯ 8 With reduced NAC Ͻ0.05 4

aAntioxidant incubations were performed for1hat37°C in vascular smooth muscle cells. NAC, N-acetylcysteine; Ang II, angiotensin II.

Figure 5. Ang II receptor immunoblotting after treatment with NAC. VSMC (A and B) and HEK cells that were transfected with

DNA for Ang II type 1a receptor–green fluorescence protein fusion protein (AT1aR-GFP; C and D) were exposed to vehicle or 10 mMNACfor1hat37°C. Immunoblotting with antibodies against the AT1a receptor was performed in lysates from VSMC, and immunoblotting with antibodies against GFP was performed in membrane lysates from transfected HEK cells. (A and C)

Representative blots. (B and D) Quantification of band densities. In A, the band for the AT1a receptor (and the AT1a receptor control) is located at approximately 43 kD. In B, the arrow marks the band corresponding to AT1aR-GFP, and the band at 32 kD corresponds to recombinant GFP. These studies were performed in samples from five control wells and six NAC-treated wells. C, control; N, NAC.

Ang II binding compared with dihydrolipoic acid in intact dized NAC, whose free sulfhydryl groups have been modified, VSMC (Figure 3) and why vitamin E did not decrease Ang II weakly lowered Ang II binding, whereas reduced NAC, whose binding in any time frame (Figure 1). Consistently, dihydroli- sulfhydryl groups are intact, was more potent. poic acid but not ␣-lipoic acid inhibited Ang II receptor binding If NAC alters Ang II receptor tertiary structure, then binding in isolated membranes, presumably because broken cell prep- affinity rather than apparent receptor density might be ex- arations lack the biochemistry to convert ␣-lipoic acid to dihy- pected to be decreased. However, our results demonstrate the drolipoic acid. NAC or dihydrolipoic acid may mimic dithio- opposite (Figure 4). This result remains unexplained. It is in- threitol, another potent reducing agent, in regulating Ang II teresting that at 4°C, NAC did not alter Ang II binding in intact binding; it is well documented that dithiothreitol inhibits bind- cells (data in text) or in membranes (Figure 6). The effect of cold ing of Ang II to AT1 receptors (26,27). The inability of N- temperature on reduction of disulfide bonds by sulfhydryl acetylserine, which is structurally similar to NAC but contains groups is unclear. The temperature effect might also be consis- a hydroxyl in the place of the sulfhydryl group, to decrease Ang tent with an effect of NAC on membrane fluidity and alter- II binding (Table 1) strongly suggests that the sulfhydryl group ations of the physical state of Ang II receptors. It is most is important in inhibiting Ang II binding. Furthermore, oxi- interesting that the Ang II AT1a receptor, which has four extra- J Am Soc Nephrol 16: 2346–2353, 2005 NAC Lowers Angiotensin Receptor Binding 2351

Figure 7. Effect of NAC on Ang II–stimulated inositol phosphate formation in VSMC. After serum-deprived VSMC were loaded with [3H]inositol for 24 h, vehicle or antioxidants were added at 37°C (10 mM NAC for 1 h, 1 mM ␣-lipoic acid for 24 h, or 100 ␮g/ml vitamin E for 24 h). Then total inositol phosphates were measured in response to 100 nM Ang II for 60 s. Basal (un- stimulated) levels of inositol phosphates were defined as 100%. This experiment was performed four to six times.

ing do so by reducing AT1a receptor mRNA transcription, although we must consider this possibility, because ROS have been shown to downregulate Ang II receptors by this mechanism (28,30). The time course for the NAC effect on Ang II binding is very rapid (Figure 2) and thus unlikely to involve

regulation of AT1a receptor mRNA levels. Similarly, NAC de- Figure 6. Effects of NAC on Ang II radioligand binding to creased Ang II binding in HEK cells bearing AT1aR-GFP, even VMSC membranes. Serum-deprived VSMC were scraped from though its DNA lacks a 5Ј regulatory region (32). In addition, dishes and converted to membranes. Some membranes (A) immunoblotting data (Figure 5), demonstrating an unchanged were exposed to 0 or 10 mM NAC for 1 h at 4, 21, or 37°C. Then amount of Ang II receptor protein, are also not consistent with specific Ang II radioligand binding was measured. Preincuba- tions with NAC and radioligand binding were performed at the a gene regulation mechanism. It is inconceivable that NAC or same temperature. Other membranes (B) were exposed to var- dihydrolipoic acid can regulate Ang II receptors via modulation ious concentrations of ␣-lipoic acid or dihydrolipoic acid for 1 h of AT1a receptor mRNA levels in a broken cell preparation at 37°C, and then specific Ang II radioligand binding was (Figure 6). measured at 37°C. Binding in cells that were treated with We dissociated antioxidant effects of these compounds from vehicle is defined as 100% (control) and is represented by the their regulatory effects on Ang II receptor binding in a prelim- Ͻ dashed line. *Difference from control at P 0.05. These exper- inary way. Both NAC (5) and vitamin E (41,42) have been iments were performed four to eight times. shown to consume Ang II–stimulated ROS, yet these compounds regulate Ang II binding in distinctly different ways: NAC downregulated Ang II binding robustly, whereas vitamin cellular cysteine residues and two extracellular disulfide bonds, E did not regulate Ang II binding at all (Figure 1). was inactivated by NAC as evidenced by decreased Ang II Our study confirms that changes in Ang II receptor binding radioligand binding, whereas the arginine vasopressin V1a re- correlate well with changes in Ang II–stimulated signal trans- ceptor, which has only one extracellular cysteine residue and duction. In intact VSMC, NAC for 1 h decreased Ang II receptor thus no extracellular disulfide bonds (40), was unaffected by binding (Figure 1) more than ␣-lipoic acid for 24 h (Figure 3), NAC with regard to arginine vasopressin radioligand binding. and incubation with vitamin E did not reduce Ang II binding It is unlikely that the antioxidants that decrease Ang II bind- (Figure 1). Consistently, NAC for 1 h reduced Ang II–stimu- 2352 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2346–2353, 2005 lated inositol phosphate formation more than ␣-lipoic acid for the NAD(P)H oxidase in vascular smooth muscle cells. Free 24 h, and vitamin E did not inhibit Ang II–stimulated inositol Radic Biol Med 30: 603–612, 2001 phosphate formation (Figure 7). These data are consistent with 9. Hannken T, Schroeder R, Zahner G, Stahl RAK, Wolf G: previous data in demonstrating that Ang II surface receptor Reactive oxygen species stimulate p44/42 mitogen-acti- density is coupled closely with Ang II–stimulated cell re- vated protein kinase and induce p27Kip1: Role in angio- sponses and argue against the existence of spare Ang II recep- tensin II-mediated hypertrophy of proximal tubular cells. J Am Soc Nephrol 11: 1387–1397, 2000 tors on the surface of VSMC (21–25). 10. Puri PL, Avantaggiati L, Burgio VL, Chirillo P, Collepardo NAC and ␣-lipoic acid both are in clinical use at present, the D, Natoli G, Balsano C, Levrero M: Reactive oxygen inter- former to treat acetaminophen toxicity (43,44) and to provide mediates mediate angiotensin II-induced c-Jun/c-Fos het- prophylaxis against contrast-induced acute renal failure (45) erodimer DNA binding activity and proliferative hypertro- and the latter for diabetic neuropathy (46,47). Our results raise phic responses in myogenic cells. J Biol Chem 270: 22129– the possibility that these agents may ameliorate renal and vas- 22134, 1995 cular diseases mediated by Ang II, especially if they mimic 11. Ushio-Fukai M, Griendling KK, Becker PL, Hilenski L, dithiothreitol in achieving the favorable combination of lower Halleran S, Alexander Q: Epidermal growth factor receptor binding to AT1 receptors (our study) and increased binding to transactivation by angiotensin II requires reactive oxygen

AT2 receptors (26). Plasma concentrations of NAC after intra- species in vascular smooth muscle cells. Arterioscler Thromb venous treatment of paracetamol overdose reached 3 to 5 mM Vasc Biol 21: 489–495, 2001 (35), similar to concentrations that decreased Ang II binding in 12. Du J, Peng T, Scheidegge KJ, Delafontaine P: Angiotensin II this study, whereas plasma levels achieved after conventional activation of insulin-like growth factor 1 receptor tran- oral dosing (1200 mg) are at least 10-fold less (48). scription is mediated by a tyrosine kinase-dependent re- dox-sensitive mechanism. Arterioscler Thromb Vasc Biol 19: 2119–2126, 1999 Acknowledgments 13. Studer RK, Craven PA, Derubertis FR: Antioxidant inhibi- This work was supported by funds from Dialysis Clinic Incorpo- tion of protein kinase C-signaled increases in transforming rated. growth factor-beta in mesangial cells. Metabolism 46: 918– A portion of this work was presented in abstract form at the annual 925, 1997 meeting of the American Society of Nephrology, Philadelphia, PA, 14. Takahashi T, Taniguchi T, Okuda M, Takahashi A, Ka- October 30 to November 1, 2002. wasaki S, Domoto K, Taguchi M, Ishikawa Y, Yokoyama We thank Jana J. Fine for technical assistance. M: Participation of reactive oxygen intermediates in the angiotensin II-activated signaling pathways in vascular References smooth muscle cells. Ann N Y Acad Sci 902: 283–287, 2000 15. Wang D, Yu X, Brecher P: Nitric oxide and N-acetylcyste- 1. Griendling KK, Minieri CA, Ollerenshaw JD, Alexander ine inhibit the activation of mitogen-activated protein ki- RW: Angiotensin II stimulates NADH and NADPH oxi- nases by angiotensin II in rat cardiac fibroblasts. J Biol Chem dase activity in cultured vascular smooth muscle cells. Circ 273: 33027–33034, 1998 Res 74: 1141–1148, 1994 16. Rajagopalan S, Kurz S, Munzel T, Tarpey M, Freeman BA, 2. Ushio-Fukai M, Zafari AM, Jukai T, Ishizaka N, Griendling Griendling KK, Harrison DG: Angiotensin II-mediated hy- KK: P22phox is a critical component of the superoxide- pertension in the rat increases vascular superoxide produc- generating NADH/NADPH oxidase system and regulates tion via membrane NADH/NADH oxidase activation. angiotensin II-induced hypertrophy in vascular smooth muscle cells. J Biol Chem 271: 23317–23321, 1996 J Clin Invest 97: 1916–1923, 1996 3. Pagano PJ, Chanock SJ, Siwik DA, Colucci WS, Clark JK: 17. Landmesser U, Cai H, Dikalox S, McCann L, Hwang J, Jo Angiotensin II induces p67phox mRNA expression and H, Holland SM, Harrison DG: Role of p47phox in vascular NADPH oxidase superoxide generation in rabbit aortic oxidative stress and hypertension caused by angiotensin II. adventitial fibroblasts. Hypertension 2: 331–337, 1998 Hypertension 40: 511–515, 2002 4. Jamies EA, Galceran JM, Raij L: Angiotensin II induces 18. Tojo A, Onozato ML, Kobayashi N, Goto A, Matsuoka H, superoxide anion production by mesangial cells. Kidney Int Fujita T: Angiotensin II and oxidative stress in Dahl salt- 54: 775–784, 1998 sensitive rat with heart failure. Hypertension 40: 834–839, 5. Hannken T, Schroeder R, Stahl RAK, Wolf G: Angiotensin 2002 II-mediated expression of p27Kip1 and induction of cellu- 19. Heitzer T, Wenzel U, Hink U, Krollner D, Skatchkov M, lar hypertrophy in renal tubular cells depend on the gen- Stahl RAK, Macharzina R, Basen JH, Meinertz T, Munzel T: eration of oxygen radicals. Kidney Int 54: 1923–1933, 1998 Increased NAD(P)H oxidase-mediated superoxide produc- 6. Zafari AM, Ushio-Fukai M, Akers M, Yin Q, Shah A, tion in renovascular hypertension: Evidence for an in- Harrison DG, Taylor WR, Griendling KK: Role of NADH/ volvement of protein kinase C. Kidney Int 33: 252–260, 1998 NADPH oxidase-derived H2O2 in angiotensin II-induced 20. Kawada N, Imai E, Karber A, Welch WJ, Wilcox CS: A vascular hypertrophy. Hypertension 32: 488–495, 1998 mouse model of angiotensin II slow pressor response: Role 7. Touyz RM, Schiffrin EL: AngII-stimulated superoxide pro- of oxidative stress. J Am Soc Nephrol 13: 2860–2868, 2002 duction is mediated via phospholipase D in human vascu- 21. Ullian ME, Linas SL: Angiotensin II surface receptor cou- lar smooth muscle cells. Hypertension 34: 976–982, 1999 pling to inositol trisphosphate formation in vascular 8. Sorescu D, Somers MJ, Lassegue B, Grant S, Harrison DG, smooth muscle cells. J Biol Chem 265: 195–200, 1990 Griendling KK: Electron spin resonance characterization of 22. Ullian ME, Schelling JR, Linas SL: Aldosterone enhances J Am Soc Nephrol 16: 2346–2353, 2005 NAC Lowers Angiotensin Receptor Binding 2353

angiotensin II receptor binding and inositol phosphate re- 36. Murphy TJ, Alexander RW, Griendling KK, Runge MS, sponses. Hypertension 20: 67–73, 1992 Bernstein KE: Isolation of a cDNA encoding the vascular 23. Ullian ME, Fine JJ: Mechanisms of enhanced angiotensin type-1 angiotensin II receptor. Nature 351: 233–236, 1991 II-stimulated signal transduction in vascular smooth mus- 37. Yamano Y, Ohyama K, Chaki S, Guo DF, Inagami T: Iden- cle by aldosterone. J Cell Physiol 161: 201–208, 1994 tification of amino acid residues of rat angiotensin II re- 24. Sato A, Suzuki H, Murakami M, Nakazato Y, Iwaita Y, ceptor for ligand binding by site directed mutagenesis. Saruta T: Glucocorticoid increases angiotensin II type 1 Biochem Biophys Res Commun 187: 1426–1431, 1992 receptor and its gene expression. Hypertension 23: 25–30, 38. Hjorth SA, Schambye HT, Greenlee WJ, Schwartz TW: 1994 Identification of peptide binding residues in the extracel- 25. Ullian ME, Raymond JR, Willingham MC, Paul RV: Regu- lular domains of the AT1 receptor. J Biol Chem 269: 30953– lation of vascular angiotensin II receptors by EGF. Am J 30959, 1994 Physiol 273: C1241–C1249, 1997 39. Ohyama K, Yamano Y, Sano T, Nakagomi Y, Hamakubo T, 26. Chiu AT, McCall DE, Nguyen TT, Carini DJ, Duncia JV, Morishima I, Inagami T: Disulfide bridges in extracellular Herblin WF, Uyeda RT, Wong PC, Wexler RR, Johnson AL, domains of angiotensin II receptor type 1a. Regul Pept 57: Timmermans PMBWM: Discrimination of angiotensin II 141–147, 1995 receptor subtypes by dithiothreitol. Eur J Pharmacol 170: 40. Cotte N, Balestre MN, Aumelas A, Mahe E, Phalipou S, 117–118, 1989 Morin D, Hibert M, Manning M, Durroux T, Barberis C, 27. Heerding JN, Hines J, Fluharty SJ, Yee DK: Identification Mouillac B: Conserved aromatic residues in the transmem- and function of disulfide bridges in the extracellular do- brane region VI of the V1a vasopressin receptor differen- mains of the angiotensin type 2 receptor. Biochemistry 40: tiate agonist vs antagonist ligand binding. Eur J Biochem 8369–8377, 2001 267: 4253–4263, 2000 28. Ichiki T, Takeda K, Tokunou T, Funakoshi Y, Ito K, Iino N, 41. Ortiz MC, Manriquez MC, Romero JC, Juncos LA: Antioxi- Takeshita A: Reactive oxygen species-mediated homolo- dants block angiotensin II-induced increases in blood pres- gous downregulation of angiotensin II type 1 receptor sure and endothelin. Hypertension 38: 655–659, 2001 mRNA by angiotensin II. Hypertension 37: 535–540, 2001 42. Bapat S, Post JA, Braam B, Goldschmeding R, Koosmans 29. Keidar S, Heinrich R, Kaplan M, Aviram M: Oxidative HA, Verkleij AJ, Joles JA: Visualizing tubular lipid peroxi- stress increases the expression of the angiotensin-II recep- dation in intact renal tissue in hypertensive rats. JAmSoc tor type 1 in mouse peritoneal macrophages. J Renin An- Nephrol 12: 2990–2996, 2002 giotensin Aldosterone Syst 3: 24–30, 2002 43. Buckley NA, Whyte IM, O’Connell DL, Dawson AH: Oral 30. Nickenig G, Strehlow K, Baumer AT, Baudler S, Wabmann or intravenous N-acetylcysteine: Which is the treatment of S, Sauer H, Bohm M: Negative feedback regulation of choice for acetaminophen poisoning? J Toxicol Clin Toxicol reactive oxygen species on AT1 receptor gene expression. 37: 759–767, 1999 Br J Pharmacol 131: 795–803, 2000 44. Amirzadeh A, McCotter C: The intravenous use of oral 31. Ullian ME, Hutchison FN, Hazen-Martin DJ, Morinelli TA: acetylcysteine for the treatment of acetaminophen over- Angiotensin II-aldosterone interactions on protein synthe- dose. Arch Intern Med 162: 96–97, 2002 sis in vascular smooth muscle cells. Am J Physiol 264: 45. Tepel M, Van Der Giet M, Schwarzfeld C, Laufer U, Lier- C1525–C1531, 1993 mann D, Zidek W: Prevention of radiographic-contrast- 32. Chen R, Mukhin YV, Garnovskaya MN, Thielen TE, Iijima agent-induced reductions in renal function by acetylcys- Y, Huang C, Raymond JR, Ullian ME, Paul RV: A func- teine. N Engl J Med 343: 180–184, 2000 tional angiotensin II receptor-GFP fusion protein: Evidence 46. Haak E, Usadel KH, Kusterer K, Amini P, Frommeyer R, for agonist-dependent nuclear translocation. Am J Physiol Titschler HJ, Haak T: Effects of alpha-lipoic acid on micro- 279: F440–F448, 2000 circulation in patients with peripheral diabetic neuropathy. 33. Ullian ME, Linas SL: Role of receptor cycling in the regu- Exp Clin Endocrinol Diabetes 108: 168–174, 2000 lation of angiotensin II surface receptor number and an- 47. Ametov AS, Barinov A, Dyck PJ, Hermann R, Kozlova N, giotensin II uptake in rat vascular smooth muscle cells. Litchy WJ, Low PA, Nehrdich D, Novosadova M, O’Brien J Clin Invest 84: 840–846, 1989 PC, Reljanovic M, Samigullin R, Schuette K, Strokov I, 34. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein Tritschler HJ, Wessel K, Yakhno N, Ziegler D: The sensory measurement with the Folin phenol reagent. J Biol Chem symptoms of diabetic polyneuropathy are improved with 193: 265–275, 1951 alpha-lipoic acid: The SYDNEY trial. Diabetes Care 26: 770– 35. Prescott LF, Donovan JW, Jarvie DR, Proudfoot AT: The 776, 2003 disposition and kinetics of intravenous N-acetylcysteine in 48. Borgstrom L, Kagedal B: Dose dependent pharmacokinet- patients with paracetamol overdosage. Eur J Clin Pharmacol ics of N-acetylcysteine after oral dosing to man. Biopharm 37: 501–506, 1989 Drug Dispos 11: 131–136, 1990