17β-Estradiol Inhibits Subarachnoid Hemorrhage–Induced Inducible Nitric Oxide Synthase Gene Expression by Interfering With the Nuclear Factor κB Transactivation
Background and Purpose— Previously, we showed that 17β-estradiol (E2) treatment prevented the subarachnoid hemorrhage (SAH)-induced cerebral vasospasm in male rats. The aim of this study was designed to further delineate the molecular mechanisms underlying E2-induced inhibition of inducible nitric oxide synthase (iNOS) upregulation and relief of vasospasm caused by SAH.
Methods— The 2-hemorrhage SAH model was induced by 2 autologous injections of blood into the cisterna magna of adult male rats. The rats were then subcutaneously implanted of a Silastic tube containing corn oil with or without 17β-estradiol benzoate and received daily intraperitoneal injections of various doses of ICI 182,780, a nonselective estrogen receptor (ER) antagonist, for 7 days after the first hemorrhage. Basilar arteries were then removed for protein extraction, RNA isolation, and gel mobility assay. The protein levels of iNOS, p65, and ER were examined by Western blot analysis, and that iNOS mRNA expression was evaluated by reverse-transcription polymerase chain reaction.
Results— E2 prevented the SAH-induced vasospasm and increases of the levels of iNOS protein and mRNA in basilar artery through an ER-dependent mechanism. Treatment of the SAH rat with E2 did not affect the nuclear translocation of p65 subunit of nuclear factor κB, but caused an increase of the association of p65/ER, and reversed the SAH-induced increase of the p65 binding on iNOS promoter.
Conclusions— E2 inhibits the SAH-induced increase of iNOS by increasing the association of p65/ER, which in turn inhibits the binding of p65 to iNOS DNA. Our data suggest the potential applications of E2 in the treatment of SAH patient.
Subarachnoid hemorrhage (SAH) remains one of the most important causes of death and affects ≈10 of every 100 000 persons annually.1 It leads to severe spasm of the cerebral arteries, which develops 4 to 9 days after SAH in 30% to 70% of patients.2 Of those individuals with delayed cerebral vasospasm, 20% to 30% experience delayed ischemic neurological deficits and approximately half of them experience severe permanent neurological dysfunction or death.1 SAH is associated with high mortality, even after the aneurysm has been secured surgically or radiologically, because of the lack of specific treatment modalities. The prognosis of SAH patients depends primarily on 3 factors: (1) the severity of the initial bleed; (2) the endovascular or neurosurgical procedure to occlude the aneurysm; and (3) the occurrence of late sequelae. Thus, the critical care management of the patient with SAH is of utmost importance to maximize the chances of satisfactory recovery.2
We recently showed that physiological serum levels of estrogen could prevent SAH-induced cerebral vasospasm and increase of of inducible nitric oxide synthase (iNOS) in basilar arteries in rats.3 It has been suggested that an increase of iNOS expression might play a critical role in the occurrence and progression of vasospasm after SAH.4 Analysis of the 5′ region of the iNOS gene revealed that it contains redox-sensitive elements including nuclear factor κB (NFκB)-responsive element.5 Recent studies have demonstrated molecular cross-talk between nuclear transcription factors in which the ER mediates inhibition of NFκB activity at several levels.6 We, therefore, proposed that interference of NFκB-induced iNOS expression by ER might contribute to the prevention of SAH-induced cerebral vasospasm by E2.
The findings of this study will provide important insights into the molecular mechanisms of vascular protective effects of E2 in SAH. Only when the mechanism of vascular protective effects of E2 is fully understood can we begin to design a strategy for preventing and treating SAH and its complications.
Materials and Methods
All procedures were approved by the Kaohsiung Medical University Animal Care and Use Committee. Total 275 Sprague-Dawley male rats (350 to 420 grams) used in this study were purchased from the National Animal Center (Nan-Kung, Taiwan). E2 was given by subcutaneous implantation of a 30-mm-long Silastic tube, 2-mm inner diameter, 4-mm outer diameter (Shin-Etsu polymer Co, Ltd) containing 0.1 mL of 17β-estradiol benzoate (0.3 mg/mL) in corn oil (Sigma), which gave rise to levels (56 to 92 pg/mL)3 within the physiological range. Previously, we demonstrated that vehicle (corn oil) injection did not affect the SAH-induced vasospasm and E2-mediated prevention effects on the SAH-induced vasospasm, suggesting that surgery and oil injection did not interfere with the E2 effects.3 Accordingly, naive rats without any treatment were used as control animals in this study. To study the involvement of ER in the E2-mediated vascular effects, some animals were injected daily intraperitoneally with ICI 182,780 (Tocris) for 7 days after the first hemorrhage.7 ICI 182,780 was dissolved in corn oil (Sigma) to make a final concentration of 20% solution. Three basilar arteries were pooled in each sample, and 3 pooled samples in each group were examined for each outcome measurement. In our studies, the mortality rate was ≈7.3% after the second injection of autologous blood and all of the survival rats developed vasospasm at 7 days after SAH. All of the survival rats were included in outcome examinations.
Induction of Experimental SAH
The 2-hemorrhage SAH model was induced by 2 autologous injections of blood into the cisterna magna.8 Male rats were anesthetized by pentobarbital (50 mg/kg, intraperitoneally). The cisterna magna was punctured percutaneously with a 25-gauge needle. Fresh autologous and nonheparinized blood (0.3 mL) was withdrawn from the tail artery, and then injected slowly into the cisterna magna (first hemorrhage). The same procedure was repeated at 48 hours later (second hemorrhage). Seven days after the first SAH, the brain was perfused with either 4% paraformaldehyde for basilar artery morphometric analysis or normal saline for mRNA or protein isolation.
The middle third of basilar artery was dissected, cut in cross-sections (0.5 μm in thickness), mounted on glass slides, and stained with 0.5% Toluidine Blue. Five randomly selected arterial cross-sections from each animal were analyzed, and cross-sectional areas were measured using computer-assisted morphometry (Image 1; Universal Imaging Corp).
Reverse-Transcription Polymerase Chain Reaction Amplification
Samples were homogenized in 1 mL TRIzol reagent (GIBCO BRL, Life Technologies). The first cDNA strand was reversely transcribed from 5 μg of total RNA. The polymerase chain reaction primers for iNOS are forward: 5′-CCAAGAACGTGTTCACCATG-3′, and reverse: 5′-GAATGTCCAGGAAGTAGGTGAGG-3′. The polymerase chain reaction primers for GAPDH are forward primer: 5′-TATG- ATGACATCAAGAAGGTGG-3′, and reverse primer: 5′-CACCACCCTGTTGCTGTA-3′. The amplification profile involved denaturation at 94°C for 1 minute, primer annealing at 60°C for 30 seconds, extension at 72°C for 1 minute, and repeated 30 cycles.
Western Blot Analysis
Nine volumes of dissecting buffer (50 mmol/L Tris acetate, pH 7.4, 10% sucrose, 5 mmol/L EDTA) were added to the sample. After homogenization, the suspension was centrifuged at 16 000g for 30 minutes, and the pellets were resuspended and rehomogenized. Protein (50 μg) was separated by 7.5% SDS-polyacrylamide gel and transferred onto polyvinylidene difluoride membrane by electroblotting for 1 hour (100 V). The membrane was incubated overnight at 4°C with block buffer containing 5% nonfat dry milk, and then with primary antibody for 1 hour at 37°C. Anti-iNOS polyclonal antibody (Santa Cruz Biotechnology), anti-p65 monoclonal antibody (BD Pharmingen), anti-Lamin B monoclonal antibody (BD Pharmingen), and anti-GAPDH polyclonal antibody (Stratagene) were used at a concentration of 1:100, 1:500, 1:200, and 1:1000, respectively. The membrane was washed with phosphate-buffered saline, and then incubated with mouse anti-goat IgG (horseradish peroxidase conjugated; Transduction Laboratories, San Jose, Calif) at 1:5000 dilution for polyclonal primary antibody or goat anti-mouse IgG (horseradish peroxidase conjugated; Transduction Laboratories) at 1:5000 dilution for monoclonal primary antibody. Immunoreactivity was visualized by enhanced chemiluminescence (Amersham).
Preparation of Nuclear and Cytosol Protein
The basilar arteries were homogenized in buffer A (10 mmol/L HEPES, pH 7.9, 10 mmol/L KCl, 1.5 mmol/L MgCl2, 0,5 mmol/L DTT, 1 mmol/L AEBSF, and 10 μg/mL leupeptin). Cytosolic protein extracts were collected by centrifugation at 4000g for 10 minutes. The pellets were resuspended in buffer B (20 mmol/L HEPES, pH 7.9, 0.42 mol/L NaCl, 20% glycerol, 1 mmol/L EDTA, 1 mmol/L DTT, 1 mmol/L AEBSF, and 10 μg/mL leupeptin) for 30 minutes at 4°C. Nuclear protein extracts were collected by centrifugation at 14 000g for 15 minutes.
Electrophoretic Mobility Shift Assays
For determination of protein–DNA interactions, the ds oligonucleotide containing NFκB binging sequence (5′-CCTACTGGGGACT- CTCCC-TTTG-3′) or cAMP response element-binding protein (CREB) consensus binding sequence (5′-AGAGATTGCCTGACG- TCAGACAGCTAG-3′) was used and end-labeled with γ32P-ATP. The NFκB or CREB binding reactions were performed in a final volume of 20 μL mixtures containing buffer (10 mmol/L HEPES, pH 7.5, 1 μg poly [dI-dC], 0.1 mol/L NaCl, 0.8 mmol/L EDTA, 1 mmol/L DTT, 0.05% NP-40, 4% glycerol, 10 μg nuclear proteins, and 100 000 cpm of radiolabeled probe). For competition assay, 50-fold excess amount of cold CREB or NFκB oligonucleotide was pre-incubated with nuclear proteins for 15 minutes before the addition of labeled oligonucleotide probe. For the identification of p65, supershift assays were conducted after incubation of the nuclear extract with 2 μL (1 μg/ μL) anti-p65 antibody (BD Pharmingen) for 1 hour at room temperature. The DNA—protein–antibody complex was resolved on 5% nondenatured polyacrylamide gel in 1× Tris-Borate-EDTA buffer and exposed to x-ray film at −70°C.
Two hundred μg of nuclear protein were mixed with 1 μg of anti-ER monoclonal antibody (Chemicon International), which is a nonselective ER antibody, or anti-p65 monoclonal antibody (BD Pharmingen), and incubated at 4°C overnight. Twenty μL protein A/G-agarose (Calbiochem) were added and shaken for additional 2 hours at 4°C. Precipitated complexes were washed with immunoprecipitation binding buffer (50 mmol/L Tris-base, 150 mmol/L NaCl, 5 mmol/L EDTA, pH 7.0, 0.1% NP-40, 1-fold protease-inhibitor-cocktail [Roche]), and then boiled in 2-fold SDS-sample buffer for 5 minutes. The supernatant were analyzed by SDS-PAGE and immunoblotted using anti-p65 or anti-ER monoclonal antibody.
The data were expressed as means±SD and analyzed by using a 1-way analysis of variance (ANOVA) followed by Scheffe post hoc test. Significance was accepted at P<0.05.
Estrogen Prevents SAH-Induced Vasospasm via an ER-Dependent Pathway
Previously, we reported that E2 prevented the SAH-induced basilar artery vasospasm. Here, we further showed that pretreatment of the rat with ICI 182,780 reversed the E2-mediated prevention on the SAH-induced vasospasm (Figure 1a). ICI 182,780 alone had no vascular effect on control or SAH rats. The ICI 182,780-mediated effect was in a dose-dependent manner with a maximal effect at a dose of 2 mg/kg (Figure 1b). These results suggested that E2 exerted its vascular effects on SAH through an ER-mediated pathway.
SAH-Induced Increases of iNOS mRNA and Protein Are Prevented by E2 Through an ER-Mediated Pathway
As illustrated in Figure 2, the prevention effect of E2 in the SAH-induced upregulations in iNOS mRNA (Figure 2a) and protein (Figure 2b) was reversed by pretreatment the animals with ICI 182,780 at a dose of 2 mg/kg, suggesting that activation of estrogen receptor is involved in these E2-mediated effects.
Effect of E2 on SAH-Induced p65 Nuclear Translocation
The levels of cytosolic p65 protein were significantly lower in the SAH group as compared with the control and E2-treated group, whereas the levels of nuclear p65 protein were significantly higher in the SAH group (Figure 3). Both E2 and E2 plus ICI 182,780 did not affect the SAH-induced nuclear translocation of p65, suggesting that ER activation is not involved in the SAH-induced nuclear translocation of p65. The absence of Lamin B immunoreactivity in the cytosolic fraction and GAPDH in the nuclear fraction ruled out the possibility of contamination from each other.
Alterations in ER/p65 Complex in Nucleus
As shown in Figure 4, the levels of ER/p65 were not changed in the SAH rat and E2-treated normal rat. However, the levels of ER/p65 in the SAH/E2-treated rats were increased. Moreover, the E2-induced increase of ER/p65 complex was reversed by pretreatment of the animal with ICI 182,780.
Effect of SAH and E2 on the DNA Binding Activities of p65 to iNOS Promoter Sequence
As illustrated in Figure 5a, the DNA binding activity of p65 to iNOS promoter sequence was increased in the SAH rat as compared with the control rat. E2 treatment prevented the SAH-induced increase in the DNA binding activity of p65, whereas E2 alone did not affect the basal DNA binding activity of p65. This inhibition induced by E2 was blocked by pretreatment of the animal with ICI 182,780. The specificity of p65 binding was confirmed by competition experiment showing that the DNA binding activity was blocked by a 50-fold of the unlabeled p65 consensus sequence, but not by the nonspecific p65 consensus sequence. The demonstration of the supershift of p65 antibody-p65-iNOS DNA complex further confirmed the presence of p65 protein in the complex. Figure 5b shows the quantitative results of electrophoretic mobility shift assays.
Epidemiological studies indicated that premenopausal women are protected from stroke relative to men.9 We previously showed that E2 at physiological levels prevents the SAH-induced cerebral vasospasm. Here, we demonstrated that E2 prevented the SAH-induced vasospasm and increases of iNOS protein and mRNA levels in basilar artery through an ER-dependent pathway.
Although cerebral vasospasm has been recognized for half a century, the molecular mechanisms underlying are not well understood. Recently, data have implicated that the inflammatory response may represent a critical common pathway in the pathogenesis of cerebral vasospasm.10 It has been suggested that various constituents of the inflammatory response, including inflammatory cytokines11 and iNOS,12 may be critical in the pathogenesis of SAH-induced cerebral vasospasm. iNOS is induced in the brain vascular tissue of the SAH rat. Treatment of the rat with aminoguanidine, a selective inhibitor of iNOS, could relieve the SAH-induced cerebral vasospasm.4 Because the expression of iNOS mRNA was increased after SAH13 and significant delayed vasospasm at day 7 in the 2-hemorrhage rats resembled human vasospasm,8 we studied vascular protective effects of E2 on 7 days after SAH induction using 2-hemorrhage rat model.
It has been demonstrated that iNOS expression can be increased by NFκB activation through a sequence of events.14 The 5′ region of the iNOS gene contains NFκB-responsive element, suggesting that iNOS is one of NFκB’s target genes. It was also reported that intra-cisternal injection of decoy NFκB oligonucleotides could relieve the SAH-induced vasospasm in rabbits.15 Here, we showed that the nuclear translocation of p65, a subunit of NFκB, from cytosol was increased in SAH rats. However, treatment of E2 did not affect the SAH-induced increase of p65 translocation into nucleus (Figure 3), suggesting that E2-induced inhibition of iNOS expression might act through interference of NFκB/iNOS DNA binding activity.
Previous studies showed that estrogen could suppress the induction of inflammatory genes through: (1) interference of NFκB DNA binding by physical association of estrogen receptors with NFκB transcription factors;6 (2) maintenance of steady level of inhibitor κB in the cytoplasm;16 and (3) prevention of cytosolic NFκB from translocation into the nucleus.17 Interestingly, the association of p65/ER was not increased in the SAH rat (Figure 4), although the translocation of p65 was increased (Figure 3). It is most likely that we used male rats in our study and the basal level of E2 in the male rat is very low. Therefore, an increased translocation of p65 alone in the nucleus did not cause an increased association of p65/ER. In the E2-treated SAH animal, however, an increased E2 concentration enhanced the translocation of ER into nucleus18 and hence the association of p65/ER was increased (Figure 4). The reason we used male rats instead of female rats in this study was to avoid the influence of endogenous E2 on the plasma E2 concentrations. In normal male rats, reported physiological concentrations of serum estrogen range from 8 to 40 pg/mL, whereas in normal cycling female rats range between 20 and 500 pg/mL.19 The functional interaction between ER and p65 has been previously demonstrated both in vitro and in vivo.20 Several groups have identified the potential for a reciprocal transcription inhibition between agonist-bound ER and activated p65.21 Including inhibition of DNA binding, several mechanisms have been suggested to be involved in this cross-talk.22 Here, we showed that the binding between p65 and iNOS DNA was increased in the SAH animal and treatment with E2 blocked this increase through an ER-dependent pathway (Figure 5).
It has been previously reported that a prominent nuclear expression of ER-β was observed in the vascular wall of several parts of the vascular tree, whereas ER-α predominantly was expressed in uterine vessels, suggesting that ER-α and ER-β might play different roles in the regulation of vascular functions.23 However, identification of ER subtype in basilar arteries has not been reported yet. It has been indicated that estrogen alters the reactivity of cerebral arteries by modifying production of endothelium-dependent vasodilators. Moreover, regulation of endothelial nitric oxide synthase pathway mediated by ER-α, but not ER-β, appears to contribute to effects of estrogen on cerebral artery reactivity.24 Although we demonstrated the involvement of ER in E2-mediated prevention of the SAH-induced vasospasm, an important issue to be resolved is identification of ER subtype involved in these E2-mediated protective effects.
In this study, we used male rats to demonstrate that exogenous E2 administration can reduce the SAH-induced vasospasm through interference of p65/iNOS DNA binding, which in turn suppresses iNOS expression. These data suggest the potential applications of E2 in the treatment of SAH patient. Whether physiological concentrations of E2 can account for the protective effect of premenopausal women from SAH-induced vasospasm needs further investigation.
Sources of Funding
This work was supported by research grants from the National Science Council of Taiwan (NSC-94-2320-B-037-003 and NSC95-2320-B-037-032-MYZ to Dr Hsu and Topnotch Stroke Research Center Grant, Ministry of Education to Dr Lee).
- Received June 12, 2006.
- Revision received July 28, 2006.
- Accepted August 14, 2006.
Bakir B, Mori T, Durand J, Chen YF, Thompson JA, Oparil S. Estrogen-induced vasoprotection is estrogen receptor dependent: evidence from the balloon-injured rat carotid artery model. Circulation. 2000; 101: 2342–2344.
Aihara Y, Kasuya H, Onda H, Hori T, Takeda J. Quantitative analysis of gene expressions related to inflammation in canine spastic artery after subarachnoid hemorrhage. Stroke. 2001; 32: 212–217.
Kuo PC, Abe K, Schroeder RA. Superoxide enhances interleukin 1beta-mediated transcription of the hepatocyte-inducible nitric oxide synthase gene. Gastroenterology. 2000; 118: 608–618.
Ghisletti S, Meda C, Maggi A, Vegeto E. 17beta-estradiol inhibits inflammatory gene expression by controlling NF-kappaB intracellular localization. Mol Cell Biol. 2005; 25: 2957–2968.
Geary GG, Krause DN, Duckles SP. Gonadal hormones affect diameter of male rat cerebral arteries through endothelium-dependent mechanisms. Am J Physiol Heart Circ Physiol. 2000; 279: H160–H618.
Evans MJ, Eckert A, Lai K, Adelman SJ, Harnish DC. Reciprocal antagonism between estrogen receptor and nf-kappab activity in vivo. Circ Res. 2001; 89: 823–830.
Andersson C, Lydrup ML, Ferno M, Idvall I, Gustafsson J, Nilsson BO. Immunocytochemical demonstration of oestrogen receptor beta in blood vessels of the female rat, J Endocrinol. 2001; 169: 241–247.
Darblade B, Pendaries C, Krust A, Dupont S, Fouque MJ, Rami J, Chambon P, Bayard F, Arnal JF. Estradiol alters nitric oxide production in the mouse aorta through the alpha-, but not beta-, estrogen receptor. Cir Res. 2002; 90: 413–441.