Down-regulation of Rac-1 GTPase by Estrogen.

Rac1 GTPase is essential for the activation of the NAD(P)H oxidase complex and, thereby, regulates the release of reactive oxygen species (ROS) in the vessel wall. 17 beta-estradiol (E2) inhibits vascular ROS production. To elucidate the underlying molecular mechanisms we investigated the potential regulation of Rac1 by E2 in vascular smooth muscle cells. Treatment of vascular smooth muscle cells with angiotensin II as well as overexpression of the constitutively active mutant RacL61 increased ROS release as assessed by dichlorofluorescein fluorescence, whereas inhibition of Rac1 by Clostridium sordellii lethal toxin or overexpression of dominant-negative RacN17 inhibited ROS production. Treatment with E2 (100 nm) completely prevented angiotensin II-induced NAD(P)H oxidase activity and ROS production. E2 time and concentration dependently decreased angiotensin II-induced and basal Rac1 mRNA and protein expression as well as Rac1 activity. Down-regulation of Rac1 expression by E2 was mediated by inhibition of gene transcription (nuclear run-on assays), but E2 had no effect on Rac1 mRNA stability. Regulation of Rac1 was mediated by estrogen receptors since co-incubation with ICI 182.780 prevented down-regulation of Rac1. To test these observations in vivo, ovariectomized spontaneously hypertensive rats were treated with E2 or vehicle. Real-time PCR and Western blotting showed reduction of aortic Rac1 mRNA and protein by 32 and 58%, respectively. Furthermore, down-regulation of Rac1 by E2 was observed in human mononuclear cells of women with elevated E2 levels after controlled ovarian hyperstimulation. Rac1 GTPase gene-transcription and activity is regulated by 17 beta-estradiol, which may be an important molecular mechanism contributing to the cardiovascular effects of estrogens.

Rac1 belongs to the small (21 kDa) Rho GTPase family that binds to and hydrolyzes guanosine triphosphate (GTP). Rho proteins have been shown to be central regulators of the actin cytoskeleton. Rho proteins function as transducers between mechanical forces, cell morphology, and gene regulation. In its active GTP-bound state, Rac1 plays an important role in the regulation of cell shape, adhesion, movement, endocytosis, secretion, and growth (1,2). In the cardiovascular system, activation of Rac1 is necessary for the release of reactive oxygen species (ROS) 1 in the vessel wall (3)(4)(5). Oxygen radicals impair endothelial function and accelerate the progression of atherosclerotic lesions by promoting lipid oxidation, the expression of proinflammatory genes, and by oxidative inactivation of endothelial nitric oxide (6 -8). The NAD(P)H oxidase complex in vascular smooth muscle cells is regarded the most important source of the primordial oxygen radical, superoxide, in the vessel wall (9). Rac1 GTPase plays a pivotal role in the assembly and activation of the NAD(P)H enzymatic system, which is composed of several subunits including p22phox, the flavoprotein p91phox (or its homologues, such as nox1 in VSMC), and the cytoplasmic subunits p47phox and p67phox (3,10). Consequently, inhibition of Rac1 activity has been shown to inhibit oxygen radical release in vascular smooth muscle and endothelial cells as well as in phagocytes (3,5). In addition to vascular superoxide production, activation of Rac1 signaling leads to cellular hypertrophy cardiac myocytes (11).
Despite the importance of Rac1 GTPase for vascular ROS release, the regulation of Rac1 in the cardiovascular system is only partially understood. It is thought that the reduced prevalence of cardiovascular disease in women is based on atheroprotective effects of estrogens. The latter are potentially mediated directly through binding to vascular estrogen receptors (12)(13)(14)(15)(16)(17). Although the antioxidative properties of estrogens are among the most prominent vasoprotective functions of sex steroids, the underlying molecular mechanisms are only partially known. Furthermore, it is not known whether small GTPases are regulated by steroid hormones. We hypothesized that 17␤estradiol may regulate Rac1 GTPase expression and activity and, thereby, inhibit the release of ROS from VSMC.

MATERIALS AND METHODS
Materials-Angiotensin II, L-mevalonate and chemicals were purchased from Sigma. [ 32 P]dCTP and Hybond N-nylon membranes were obtained from Amersham Biosciences. [ 35 S]GTP␥S was supplied by PerkinElmer Life Sciences. H 2 DCF-DA was purchased from Molecular Probes (Eugene, OR). Antibiotics, calf serum, and cell culture medium were obtained from Invitrogen. RNA-clean was purchased from AGS (Heidelberg, Germany). Clostridium sordellii lethal toxin was kindly provided by K. Aktories (Freiburg, Germany) (1). RhoN19 and RacN17 were a kind gift from A. Hall (London, UK) (2).
Cell Culture-VSMC were isolated from female rat thoracic aorta (strain, male Sprague-Dawley, 6 -10 weeks old, Charles River Wega GmbH, Sulzfeld, Germany) by enzymatic dispersion and cultured over several passages. Cells were grown in a 5% CO 2 atmosphere at 37°C in Dulbecco's modified Eagles medium without phenol supplemented with 100 units/ml of penicillin, 100 g/ml streptomycin, 1% nonessential amino acids (100ϫ), and 10% fetal calf serum (free of steroid hormones, S-15-M, c.c.pro GmbH). Experiments were performed with cells from passage 5-10. Cells were kept in quiescent medium without fetal calf * This work was supported by the Deutsche Forschungsgemeinschaft (to U. L. and G. N.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ To whom correspondence should be addressed. Animal Treatment-Female, spontaneously hypertensive rats put on a standard chow and were ovariectomized or sham-operated (control group) 16 weeks after birth. For treatment, 17␤-estradiol pellets (containing 1.7 mg of estradiol each, 60-day release, Innovative Research) were implanted subcutaneously. E2 levels were determined by radioimmunoassay (DPC Biermann, Bad Nauheim, Germany). The thoracic aorta was harvested 5 weeks after surgery. All animal experiments were conducted in accordance to the German animal protecting law.
Transfection-Female VSMC were harvested and resuspended in electroporation medium (Optimem 1, Invitrogen) at a concentration of 5 ϫ 10 7 cells/ml. The following constructs were transfected: insertless vector (pcDNA3) as control, pRK5-myc-Rac1-L61 (constitutively active Rac1 mutant), and pRK5-myc-Rac1-N17 (dominant-negative Rac1 mutant) (18,19). 20 g of plasmid DNA and 200 l of cell suspension were placed in a 0.4-cm cuvette, mixed, and incubated for 30 min on ice. After incubation at 37°C for 30 s, the cuvette was pulsed with 300 V and 500 F (Electro Cell Manipulator, BioRad). The pulse length was determined by the electroporator based on capacitance, field strength, and resistance of the medium. Upon electroporation the cuvette was incubated at room temperature for an additional 30 min. The cells were plated at tissue culture plates and cultured for 48 h before treatment with angiotensin II, E2, and vehicle as indicated.
Measurement of Reactive Oxygen Species-Intracellular reactive oxygen species production was measured by 2Ј,7Ј-dichlorofluorescein (DCF) fluorescence using confocal laser scanning microscopy techniques. Dishes of subconfluent cells were washed and incubated in the dark for 30 min in the presence of 10 mmol/liter 2Ј,7Ј-dichloro-dihydrofluorescein-diacetate (H 2 DCF-DA). Culture dishes were transferred to a Zeiss Axiovert 135 inverted microscope (Carl Zeiss, Jena, Germany), equipped with a 25ϫ, numerical aperture 0.8, oil-immersion objective (Plan-Neofluar, Carl Zeiss) and Zeiss LSM 410 confocal attachment, and reactive oxygen species generation was detected as a result of the oxidation of H 2 DCF (excitation, 488 nm; emission longpass LP515-nm filter set). 512 ϫ 512 pixel images were collected by single rapid scans, and identical parameters, such as contrast and brightness, were used for all samples. Five groups of 25 cells for each sample were randomly selected from the image, and fluorescent intensity was taken. The relative fluorescence intensity are average values of all experiments.
For measurement of superoxide release in intact vessel segments, aortas were excised carefully and placed in chilled, aerated Krebs-HEPES buffer as described (20). Chemiluminescence of aortic rings was assessed over 10 min in the presence of 5 mol/liter lucigenin in a scintillation counter (Lumat LB 9501, Berthold, Bad Wildbad, Germany) in 1-min intervals. Superoxide release is expressed as relative chemiluminescence per mg of aortic tissue. NAD(P)H Oxidase Activity Assay-NADH or NADPH oxidase activity was measured by a lucigenin-enhanced chemiluminescence assay in a 50-mmol liter Ϫ1 phosphate buffer (buffer A), pH 7.0, containing 1 mmol liter Ϫ1 EGTA, protease inhibitors (Complete, Roche Molecular Biochemicals), 150 mmol liter Ϫ1 sucrose, 5 mol liter Ϫ1 lucigenin, and either 100 mol liter Ϫ1 NADH or 100 mol liter Ϫ1 NADPH as substrate (21). Cell cultures were treated as indicated, washed twice with ice-cold phosphate-buffered saline, pH 7.4, and scraped from the dishes. After a step of low spin centrifugation, the pellet was resuspended in ice-cold buffer A, lacking lucigenin and substrate. Then, the cells were mechanically lysed by using a glass/Teflon potter on ice. The total protein concentration was determined using the Bradford assay (BioRad) and adjusted to 1 mg ml Ϫ1 . 100-l aliquots of the protein sample were measured over 10 min in quadruplicates using NADH or NADPH as substrate in a scintillation counter (Berthold Lumat LB 9501) in 1min intervals.
Real-time RT-PCR-Real-time quantitative reverse transcriptionpolymerase chain reaction (RT-PCR) was performed with the TaqMan system (Prism 7700 Sequence Detection System, PE Biosystems). For rat Rac1 the primers were 5Ј-GTA AAA CCT GCC TGC TCA TC and 5Ј-GCT TCG TCA AAC ACT GTC TTG. The nox1 primers were 5Ј-CCC GCA ACT GTT CAT ACT C and 5Ј-CAT TGT CCC ACA TTG GTC TC. For 18 S the primers were 5Ј-TTG ATT AAG TCC CTG CCC TTT GT and 5Ј-CGA TCC GAG GGC CTA ACTA. For quantification, Rac1 mRNA expression was normalized to the expressed housekeeping gene 18 S.
Rac1 GST-PAK Pull Down Assay-A glutathione-S-transferase (GST)-PAK-CD (PAK-CRIB domain) fusion protein, containing the Rac1 binding region from human PAK1B (22) was used to determine Rac1 activity as described (23). Escherichia coli transformed with the GST-PAK-CD construct were grown at 37°C to an absorbance of 0.3. The construct was a kind gift of R. C. Roovers and J. G. Collard, The Netherlands Cancer Institute, Amsterdam, The Netherlands. Expression of recombinant protein was induced by addition of 0.1 mmol/liter isopropyl thiogalactoside for 2 h. Cells were harvested, resuspended in lysis buffer (50 mmol/liter Tris-HCl, pH 8, 2 mmol/liter MgCl 2 , 0.2 mmol/liter Na2S2O, 10% glycerol, 20% sucrose, 2 mmol/liter dithiothreitol, 1 g/ml leupeptin, 1 g/ml pepstatin, and 1 g/ml aprotinin), and then sonicated. Cell lysates were centrifuged at 4°C for 20 min at 45,000 ϫ g, and the supernatant was incubated with glutathionecoupled Sepharose 4B beads (Amersham Biosciences) for 30 min at 4°C. Protein bound to the beads was washed three times in lysis buffer, and the amount of bound fusion protein was estimated using Coomassie-stained SDS gels.
Vascular smooth muscle cells were treated as indicated and washed with ice-cold phosphate-buffered saline, incubated 5 min on ice in lysis buffer (50 mmol/liter Tris-HCl, pH 7.4, 2 mmol/liter MgCl 2 , 1% Nonidet P-40, 10% glycerol, 100 mmol/liter NaCl, 1 mmol/liter benzamidine, 1 g/ml leupeptin, 1 g/ml pepstatin, 1 g/ml aprotinin), and then centrifuged for 5 min at 21,000 ϫ g at 4°C. Aliquots were taken from the supernatant to compare protein amounts. Equal amounts of supernatant protein were incubated with the bacterially produced GST-PAK-CD fusion protein bound to glutathione-coupled Sepharose beads at 4°C for 30 min. The beads and proteins bound to the fusion protein were washed three times in an excess of lysis buffer, eluted in Laemmli sample buffer (60 mmol/liter Tris, pH 6.8, 2% SDS, 10% glycerin, 0.1% bromphenol blue), and then analyzed for bound Rac1 molecules by Western blotting.
Nuclear Run-on Assays-Vehicle-and E2-treated VSMC were collected and washed. After lysis for 10 min on ice, nuclei were isolated by centrifugation through 0.6 mol/liter sucrose (4,24). The nuclei (ϳ3-5 ϫ 10 8 /reaction) were used to carry out the in vitro transcription in a reaction mixture containing 40% glycerol, 50 mmol/liter Tris/HCl, 5 mmol/liter MgCl 2 , 0.1 mmol/liter EDTA, 0.5 mmol/liter levels of CTP, GTP, ATP, and UTP at 30°C for 30 min. Reactions were terminated by addition of RNA-clean. Immediately before transcription a sample of each condition was removed. Total RNA before and after transcription was isolated and Rac1 and 18 S mRNA were quantitated using realtime RT-PCR (see above). The extent of Rac1 mRNA transcription was determined by subtracting the amount of Rac1 mRNA standardized to 18 S mRNA prior to transcription from the amounts post transcription. For some experiments, transcription was performed in the presence of 0.2 mol/liter [ 32 P]UTP (Ͼ3000 Ci/mmol). The transcribed radioactive RNA was hybridized with nylon membranes dotted with linearized pKSϩ BlueScript, Rac1, and glyceraldehyde-3-phosphate dehydrogenase cDNA, 5 g of each, as described in detail previously (4,24). Quantification using dot-blots did not differ from quantifications by real-time RT-PCR.
Human Mononuclear Cells-Blood samples of patients from the gynecology outpatient clinic scheduled for planned in vitro fertilization were investigated. Controlled ovarian hyperstimulation following the long-protocol was initiated in all patients with the gonadotropin-releasing hormone analogue triptorelin, 0.1 mg subcutaneously daily, starting in the midluteal phase of the previous cycle until pituitary desensitization was achieved. Then gonadotropin therapy (recombinant folliclestimulating hormone 150 -200 IE subcutaneously daily, Gonal-F; Serono) was given to induce follicular growing. Gonadotropin-releasing hormone analogue injection was continued up to and including the day of ovulation induction (day 10 -12). 30 ml of EDTA plasma were taken before and after 6 -10 days of follicle-stimulating hormone treatment. Estradiol levels were evaluated, and mononuclear cells were separated immediately by standard Ficoll gradient centrifugation.
Data Analysis-Band intensities were analyzed by densitometry. All values are expressed as mean Ϯ S.E. compared with controls. Paired and unpaired Student's t tests and analysis of variance for multiple comparisons were employed. Post-hoc comparisons were performed with the Newman-Keuls test. Differences were considered significant at p Ͻ 0.05.  Fig. 1A, data analysis in B). E2 alone had no significant effect on basal ROS production. In addition, VSMC intracellular superoxide anion formation in the presence of NADPH and NADH was detected by lucigenin assays as described by Griendling et al. (21). Angiotensin II mediated an up-regulation of both NADH (243 Ϯ 109%) and NADPH oxidase activity (307 Ϯ 126%), which was inhibited after pretreatment with E2 (100 nM, 16 h) (n ϭ 3, * p Ͻ 0.05) (Fig. 1C).

Inhibition of Rac1-dependent ROS Release in Vascular
Clostridium sordellii lethal toxin inhibits Rac1 activity by specific glucosylation (1). Treatment with lethal toxin (200 ng/ml, 16 h) completely abolished angiotensin II-stimulated oxygen radical release ( Fig. 2A). Similarly, overexpression of the dominant-negative RacN17 reduced angiotensin-mediated ROS production (Fig. 2B). Transfection with the constitutively active mutant RacL61 increased ROS release by 2-fold (Fig.  2B). E2 completely reversed angiotensin II-mediated ROS release in cells transfected with empty vector but had no significant effect after transfection with RacL61. These data show that E2 inhibits angiotensin II-stimulated free radical release from VSMC and that Rac1 activity is both necessary and sufficient for ROS production. The experiments suggest that Rac1 is involved in E2-induced decrease of oxidative stress.
In contrast to the down-regulation of Rac1 mRNA, the expression the NAD(P)H oxidase subunits p22phox and nox1 was not significantly altered by E2 (1 M, 16 h, n ϭ 5) (Fig. 4, D  and E).
Inhibition of Rac1 Gene Transcription by E2-To elucidate the mechanism of down-regulation of Rac1 mRNA expression by E2, the rate of Rac1 gene transcription and mRNA stability were studied in the presence and absence of E2 (1 M, 16 h). Nuclear run on assays showed a reduction of Rac1 transcription to one-third compared with untreated cells (34.5 Ϯ 10%, p Ͻ 0.05) (Fig. 6A). In contrast, DRB studies showed no significant alteration of Rac1 mRNA half-life in the presence of estrogen (Fig. 6B). Down-regulation of Rac1 expression by E2 is mediated by inhibition of gene transcription.
Down-regulation of Rac1 Mediated by Estrogen Receptor-To study whether the effects of E2 on Rac1 were receptor-mediated, VSMC were treated with E2 (0.01-1 M, 16 h) in the presence of ICI 182.780, 1 M. Co-treatment with ICI showed complete inhibition of E2-induced down-regulation of Rac1 expression, suggesting receptor-mediated signaling (Fig. 7A). To verify the expression of estrogen receptor ␣ (ER␣) and ␤ (ER␤) in vascular smooth muscle cells, Western analysis was performed. Both receptor subtypes were expressed abundantly. Treatment with E2 (0.01-10 M, 16 h) lead to concentrationdependent up-regulation of ER␣ and ER␤ expression (Fig. 7, B  and C).
Inhibition of Rac1 Expression in Mononuclear Cells of Women with Elevated E2 Levels-To assess whether the cell culture and animal studies may have significance in humans, mononuclear cells were collected from women before and during controlled ovarian hyperstimulation prior to in vitro fertilization, leading to significant increase of 17␤-estradiol blood levels (Fig. 9A). Real-time PCR showed down-regulation of Rac1 mRNA levels to 51 Ϯ 36% in the presence of elevated estrogen levels (n ϭ 6, p Ͻ 0.05) (Fig. 9B). DISCUSSION This study shows that 17␤-estradiol inhibits the expression and activity of Rac1 GTPase leading to inhibition of free radical production in vascular smooth muscle cells. Similar effects were observed in the vessel wall in vivo. Down-regulation of Rac1 by E2 was not limited to VSMC but was observed in mononuclear cells of women with elevated E2 levels after controlled ovarian hyperstimulation.
An important step in the pathogenesis of endothelial dysfunction and the progression of atherosclerosis is the activation of NAD(P)H oxidase enzyme complex in VSMC by angiotensin II, the primary source of superoxide production in the vessel wall (25). Rac1 GTPase plays a pivotal role during the assembly of the NAD(P)H system (3, 10, 26). Here we show, using overexpression of dominant-negative and active Rac1 mutants, that Rac1 activity is both necessary for ROS production in vascular smooth muscle cells and sufficient for ROS release. In agreement with previous studies (27), E2 effectively and completely inhibited angiotensin II-mediated ROS release. More specifically, E2 prevents angiotensin II-mediated NADH and NADPH oxidase activity. But E2 did not significantly reduce ROS after transfection with the active RacL61, pointing toward a role of Rac1 for the anti-oxidative effects of E2. Indeed, Western and Northern analyses demonstrated that E2 concentration and time dependently down-regulated Rac1 protein and mRNA expression, both alone and in the presence of angiotensin II. Similarly, E2 inhibited basal and stimulated Rac1 activity. The molecular mechanism is the inhibition of Rac1 gene transcription, whereas E2 had no significant effect on Rac1 mRNA half-life. The estrogen receptors ␣ and ␤ were abundantly expressed in VSMC and up-regulated by treatment with E2. Down-regulation of Rac1 expression by E2 was completely blocked in the presence of the nonselective estrogen receptor antagonist ICI 182.780, demonstrating a receptor-mediated event.
To test the relevance of these findings in vivo, a well characterized animal model of estrogen deficiency by ovariectomy and E2 replacement therapy was studied (20). In the aortas of ovariectomized SHR significant down-regulation of Rac1 mRNA and protein expression by E2 was observed. Depression of Rac1 by estrogen replacement strongly correlates with reduced vascular oxidative stress. The presented cell culture data assign an essential role to Rac1 in NAD(P)H oxidase-mediated radical release. Thus, it may be suggested that estrogen-induced inhibition of Rac1 reduces production of ROS in vitro as well as in vivo.
To further extend these findings to the human situation, mononuclear cells of young women with elevated estrogen levels undergoing controlled ovarian hyperstimulation prior to in vitro fertilization were studied. Elevation of serum 17␤-estradiol correlated with a decrease of Rac1 mRNA expression. These data suggest that estrogen may regulate Rac1 GTPase in humans, but additional studies are needed before conclusions regarding a potential effect of estrogen replacement therapy, especially in combination with progesterone, should be drawn (27).
In the vascular wall, estrogens exert anti-oxidant effects in addition to the inhibition of Rac1 GTPase in VSMC, which are primarily located in the media of the arterial wall. ROS release from the endothelium as well as the adventitia may play an important role in vivo. 3-nitrotyrosine immunoreactivity as well as expression of the NAD(P)H oxidase subunit gp91(phox) have been shown to increase in the endothelium and adventitia of mice treated with angiotensin II (28,29). Importantly, recent work by Wagner et al. shows that E2 decreases the function of the NAD(P)H oxidase in endothelial cells, which is mediated by down-regulation gp91phox (30). The expression of its homologue in vascular smooth muscle cells, nox1, was not significantly altered by E2, suggesting a potential dichotomy between endothelial cells and VSMC, which may help to address the cell-specific function of NAD(P)H oxidases in different cell types in further studies. Down-regulation of gp91phox in the endothelium and Rac1 in the media are likely complementary effects in vivo. In addition, estrogen reduces oxidative stress by down-regulation of the AT1 receptor (7,12,27). The estrogeninduced reduction of ROS is closely connected to another beneficial action of estrogen on vascular cells, namely the upregulation of endothelial nitric-oxide synthase activity (12,30,31). Therefore, the well established increase of NO bioavailability is caused by increased NO production and decreased superoxide release.
It is thought that the putative vasoprotective effects of estrogens are at least in part mediated via reduction of oxidative stress. Decreased Rac1 expression and activity may resemble a novel and important mechanism by which estrogens interfere with free radical production. In addition, recent evidence suggests an important role for Rac1 GTPase for the control of oxygen radical release outside the vascular wall in several cell types, including leukocytes, fibroblasts, and cardiac myocytes (10,32). Inhibition of Rac1 by expression of dominant-negative N17rac1 has been shown to protect from hypoxia/reoxygenation-induced cell death in a variety of cell types including vascular smooth muscle cells, fibroblasts, endothelial cells, and ventricular myocytes (33). In cardiomyocytes, Rac1 has been identified as a mediator of hypertrophy (11,34,35). Inhibition of Rac1 activity in the heart, e.g. by inhibition of Rac1 isoprenylation using HMG-CoA reductase inhibitors, has been shown to prevent the hypertrophic phenotype as well as cardiac ROS production (36,37). Interestingly, estrogen has been reported to prevent cardiac hypertrophy by a mechanism yet unknown (16,38). We speculate that the antihypertrophic effects of estrogen could at least in part be mediated by regulation of Rac1 GTPase.
In summary, Rac1 GTPase gene transcription and activity are regulated by E2, which may be an important molecular mechanism contributing to the cardiovascular effects of estrogens.