The FASEB Journal express article 10.1096/fj.02-0321fje. Published online December 17, 2002.
AT2-Receptor activation regulates myocardial eNOS expression via the calcineurin-NF-AT pathway Oliver Ritter*,§, Kai Schuh*,§, Marc Brede†, Nicola Röthlein*, Natalie Burkard*, Lutz Hein†, and Ludwig Neyses‡ *Department of Medicine, University of Wuerzburg, Germany; †Department of Pharmacology and Toxicology, University of Wuerzburg, Germany; and ‡University Department of Medicine, Manchester Royal Infirmary, Manchester, United Kingdom. §Both authors contributed equally to this work. Corresponding author: Ludwig Neyses, University Department of Medicine, Manchester Heart Center, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, United Kingdom. Email:
[email protected]. Cc:
[email protected] ABSTRACT The role of AT2-receptors has recently been subject of considerable debate. We investigated the influence of AT2-stimulation/inhibition on myocardial endothelial NO-synthase (eNOS, NOSIII) promoter activity and eNOS protein expression. Stimulation of rat cardiomyocytes with angiotensin II (AngII) increased eNOS protein expression 3.3-fold. This was blocked by Cyclosporin A (CsA). Inhibition of the AT1-receptor did not reduce AngII-mediated eNOS protein expression, whereas AT2 stimulation increased it 2.4-fold and AT2 inhibition suppressed it. The modulatory effects of the AT2-receptor on eNOS expression was confirmed in mice with a genetic deletion of the AT2-receptor (AT2-KO). In gel shift assays two putative NF-AT sites in a 1.6 kb eNOS promoter fragment showed NF-AT binding and a supershift by NF-AT2(-c1)-specific antibodies. Stimulation of transfected cells with AngII or specific AT2-receptor agonists resulted in a significant increase in eNOS promoter activity, which was blocked by CsA, MCIP1, and mutation of an upstream NF-AT site. Conclusion: 1) AngII-stimulation of the myocardium, both in vivo and in vitro, is accompanied by increased expression of eNOS. 2) This effect is mediated by the calcineurin pathway and is induced by the AT2-receptor. 3) These results define a calcineurin/NF-AT/eNOS pathway as downstream effector of AT2-receptor activation in the myocardium. Key words: angiotensin • nitric oxide synthase • hypertrophy
T
hree AngII-receptor subtypes, AT1A (1), AT1B (2), and AT2 (3) have been identified. The AT1-receptor has a widespread tissue distribution and mediates most of the cardiovascular AngII actions, such as vasoconstriction and development of some forms of cardiac hypertrophy. The expression pattern of the AT2-receptor is more restricted. It is up-regulated in a variety of models of myocardial growth. It has been hypothesized that activation of the AT2receptor may be antagonistic to AT1-receptor activation, that is, AT2-receptors exert antihypertrophic effects (4). However, models of AT2-receptor deficient mice yielded conflicting results of this concept in cardiac tissue depending on genetic background differences (5–7),
whereas anti-hypertrophic effects of AT2-receptors have been shown in the vasculature in vivo (8). Thus, the AT2-effect may depend on cellular and temporal context (9). Recent studies have demonstrated that AT1-receptor stimulation is sufficient to activate the calcineurin–NF-AT (nuclear factor of activated T cells) pathway in cardiomyocytes, which results in myocardial hypertrophy (10). In this model, elevated cytosolic Ca2+ levels lead to an activation of the Ca2+/calmodulin-dependent phosphatase calcineurin, which in turn dephosphorylates the transcription factors NF-AT thus enabling its nuclear translocation (11) and results in induction of marker genes of cardiac hypertrophy (12). NF-AT consensus sites are present in the promoter region of a number of genes up-regulated in myocyte hypertrophy (e.g., BNP) (12). Here we aimed at defining new NF-AT targets and their upstream activating receptors in cardiomyocytes. The eNOS promoter region (13) harbors four putative NF-AT binding sites showing high similarity with the NF-AT consensus binding sequence (T/AGGAAA) (14). We therefore tested the hypothesis that eNOS is up-regulated in cardiomyocytes upon AngII stimulation, that this upregulation is mediated via the AT2-receptor, and that it involves the calcineurin/NF-AT pathway. MATERIALS AND METHODS Preparation of neonatal rat cardiomyocytes and fibroblasts Neonatal rat cardiomyocytes of Wistar rats were isolated as described previously (15). Briefly, hearts were removed from 1 to 3 days old rats, the ventricular tissue was digested in calcium and bicarbonate-free Hank’s medium with HEPES with 1.5 mg/ml trypsin and 10 µg/ml DNase. To inactivate trypsin, FCS was added to the resulting cell suspension, and cells were pelleted by centrifugation (5 min at 700 g). The cells were resuspended in MEM/5 (minimum essential medium/5% fetal calf serum), and fibroblasts were removed by pre-plating for 1 h and used for cell cultures. After pre-plating, the supernatant (cardiomyocytes) was recovered and cells were plated in MEM/5 on six-well plates at a density of 1 × 106 cells/well. Twenty-four hours after plating, MEM/5 was replaced by MEM/1% FCS (MEM/1). Medium for cardiomyocytes contained 5-Bromo-2`-deoxyuridine (0.1 mM) to suppress fibroblast growth. Fibroblast contamination of cardiomyocyte cultures was between 4–7% as regularly determined by immunohistochemical staining for troponin T. Stimulation of cell cultures Forty-eight hours after preparation cells were stimulated with 10, 100, or 500 nM angiotensin II (AngII), cyclosporin A (CsA; 0.5 µg/ml), a specific AT2-receptor agonist (CPG 42112, 10 µg/ml), a specific AT2-receptor antagonist (PD 123319, 10 µg/ml), a specific AT1-receptor antagonist (Losartan, 10 µg/ml), or a combination of these compounds in MEM/1 medium. Cells were harvested 48 h after stimulation. Jurkat T cells were stimulated for 16 h with PMA (phorbol-12-myristate-13-acetate, 20 ng/ml) and ionomycin (50 µM) alone or in parallel with CsA (0.5 µg/ml) and were harvested after stimulation. The protocol was described previously (15). For specific inhibition of calcineurin, cells were transiently transfected with an MCIP1 expression vector (myocyte-enriched calcineurin interacting protein, MCIP1). The plasmid vector was a gift from S. Williams, Dallas, TX (16).
Western blotting and determination of eNOS protein expression Cells were harvested in 300 µl/well Laemmli’s buffer, and 35 µl of samples were separated in a 8% denaturing SDS-polyacrylamide gel. After separation, proteins were blotted to nitrocellulose membrane (Amersham Pharmacia Biotech) and were detected (ECL system, Amersham Pharmacia Biotech) by using the following antibodies: eNOS: 1:200, mAb, Transduction Laboratories, Cat. #: N30020; actin: 1:1000, goat pAb, St. Cruz, Cat. #: sc-1616. Relative eNOS protein expression was quantified with ScanPack 3.0 image analysis software (Biometra) in relation to actin expression. Construction of human eNOS promoter reporter vector and mutagenesis The human eNOS promotor (positions –1600 up to +20) was amplified from genomic DNA by using the following primers: ATCTGATGCTGCCTGTCACC and ATGTTACTGTGCGTCCACTC. The PCR product was cloned in the T/A cloning vector pTAdv (Clontech), and the insert sequence was verified. The human eNOS promotor was excised with Kpn I and Xho I and cloned in both orientations in pGL3-basic reporter vector (Promega). The reporter plasmid harboring the human eNOS promoter in sense orientation upstream of the luciferase gene was called “NP3-luci,” the eNOS promoter antisense construct was called “NP1luci” and served as control. Mutations 1 (“M1,” –1540CCTAGGAAAA → –1540CCTACTAAAA) -and 2 (“M2,” – TCCAGGAAAT → –640TCCACTAAAT) were introduced by using the QuikChange™ sitedirected mutagenesis system (Stratagene) and the following primers:
640
Mutation 1: 5′-TTAACTGGGACCTACTAAAATGAGTCATCCTTGGTC-3′ and 5′-GACCAAGGATGACTCATTTTAGTAGGTCCCAGTTAA-3′. Mutation 2: 5′-GGGGTGTGGGGGTTCCACTAAATTGGGGCTGGGAGGG-3′ and 5′-CCCTCCCAGCCCCAATTTAGTGGAACCCCCACACCCC-3′. The double mutations (“M1+M2”) were introduced one after the other. Transfection and determination of eNOS promoter activity Neonatal cardiomyocytes were transfected by using the LIPOFECTAMINE™ PLUS™ system (Life Technologies), and the following protocol: 1 µg reporter plasmid was diluted in 100 µl serum-free medium (80 µl OPTI-MEM® I, Life Technologies + 20 µl MEM, Sigma), 2 µl PLUS reagent were added, and the mixture was incubated for 15 min at room temperature. During incubation period, 4 µl LIPOFECTAMINE reagent were diluted in 100 µl serum-free medium. After mixing and additional incubation for 15 min, the mixture was added to cells in 800 µl serum-free medium. Cells were incubated for 3 h at 37°C. After incubation, 1 ml MEM/1 was added and subsequently cells were stimulated as described above. Cells were harvested in 200 µl lysis buffer (Promega), and cellular debris was removed by centrifugation. Luciferase activity was determined according to the manufacturer’s protocol (Promega).
Determination of eNOS activity For quantitative determination of eNOS activity in cultured cardiomyocytes, cyclic GMP (cGMP) formation was investigated by using a cGMP immunoassay (R+D Systems, Cat. No DE0550, Wiesbaden, Germany). Competitive binding of cGMP, subsequent addition of substrate solution and determination of absorbance at 405 nm was performed according to the manufacturer’s protocol. Electrophoretic mobility shift asays (EMSA) and detection of NF-AT factors in nuclear extracts of cardiac myocytes Nuclear protein extracts were prepared according to Neumann et al. (1995) (17). Protein concentrations were determined by a Bradford assay (18). The following (double-strand) oligonucleotides [sequences according to Zhang et al., 1995 (13)] were used for EMSAs: “-1535 NF-AT site”: 5′-GGGACCTAGGAAAATGAGTCA-3′ CCCTGGATCCTTTTACTCAGT “-1535 mutNF-AT site”: 5′-GGGACCTACTAAAATGAGTCA-3′ CCCTGGATGATTTTAGAGAGT “-650 NF-AT site”: 5′-GTGTGGGGGTTCCAGGAAATTGGGG-3′ CACACCCCCAAGGTCCTTTAACCCC “-600 NF-AT site”: 5′-GTCAGACTCAAGGACAAAAAGTCAC-3′ CAGTCTGAGTTCCTGTTTTTCAGTG “-50 NF-AT site”: 5′-GGTCCCCTCCCTCTTCCTAAGGAAAAGGCC-3′ CCAGGGGAGGGAGAAGGATTCCTTTTCCGG For each lane, 5 µg nuclear protein extract of cardiac myocytes and 3 µg nuclear extract of Jurkat T cells were incubated with 50,000 cpm 32P-labeled oligonucleotide probe in the presence of 0.5 µg poly[d(I-C)] (Roche Diagnostics) as nonspecific competitor. To show specificity of binding, 50 ng of unlabeled probe or mutated NF-AT sites (sequences shown above) were used as
competitor. In supershift assays 2 µg antibodies (anti-NF-AT1: St. Cruz, Cat. #: sc-7295 X, antiNF-AT2, clone 7A6: St. Cruz, sc-7294 X, anti-NF-AT3: Oncogene, PC625–100UG, anti-NFAT4: St. Cruz, sc-8405 X, and anti-NF-AT5: Oncogene, PC626–50UG) were added to the binding reactions. Samples were incubated 30 min on ice and were fractionated on a 6% nondenaturing polyacrylamide gel at 200 V/20 cm. The dried gel was exposed to X-ray film. In parallel, presence of different NF-AT factors in nuclear extracts of cardiac myocytes was examined by Western blotting by using 30 µg nuclear protein extract per lane. As positive control served 10 µg nuclear extracts of PMA/ionomycin-stimulated Jurkat T cells, stimulation see above. eNOS Protein expression in vivo The generation and genotyping of mice lacking a functional AT2-receptor was described previously (19). Mice were backcrossed over six generations. Left ventricular myocardium of 12- to 16-week-old Agtr2-/Y (n=6) and WT (n=6) mice was investigated for eNOS protein expression. eNOS mRNA Synthesis Inhibition of RNA synthesis was perfomed with a RNA polymerase II inhibitor (5,6-dichloro-1β-D-ribofuranosylbenzimidazole; DRB; 10 µM, 45 min). Experimental procedures for Nothern blot studies were performed as described earlier (20). PCR-cloned rat eNOS cDNA was used as a probe. Statistics Data are presented as mean ± SEM. Data were compared by using one-way ANOVA test. A value of P
