The Impact of Blunted -Adrenergic Responsiveness on Growth Regulatory Pathways in Hypertension

The effects of vasodilator hormones acting through receptors linked to adenylyl cyclase are impaired in the hypertensive state. This has been ascribed to impaired receptor-G protein coupling. However, these receptors also act via effectors not linked to adenylyl cyclase activation. These “alternate” mechanisms may be especially important in growth regulation and might be unaffected (or enhanced) with G protein-coupled receptor-G protein uncoupling. Therefore, we assessed the effects of -adrenergic activation on 1) regulation of phosphatidylinositol 3-kinase (PI3 kinase) and extracellular signalregulated kinase (ERK) activation—two tyrosine kinasedependent enzymes linked to cell growth—and 2) microarray analysis in vascular smooth muscle cells from spontaneously hypertensive rats (SHR). Isoproterenol-stimulated phosphorylation of ERK1/2 was impaired in SHR. The effect of forskolin was unaltered. In contrast, both vasopressin and angiotensin 2-mediated stimulation of ERK activation was enhanced in SHR. In addition, -adrenergic-mediated inhibition of PI3 kinase activity was attenuated in SHR (whereas the effect of forskolin remained intact). In microarray studies, the effect of isoproterenol to regulate transcription was significantly impaired in SHR (as was the effect of forskolin). Together, these data support the hypothesis that the blunted vasodilator effects of hormones linked to adenylyl cyclase activation are an index of a more generalized impairment in modulating growth regulatory pathways. Furthermore, this study supports the hypothesis that the blunting of -adrenergic responses relating to increased G protein-coupled receptor kinase 2 expression reflects a “generalized uncoupling” of -adrenergic-mediated responses and do not support the concept of “enhanced coupling” of “alternate” pathways of -adrenergic growth regulatory pathways in the hypertensive state. The increase in vascular resistance characteristic of hypertension is mediated by alterations in both structural (hypertrophy/hyperplasia) and functional determinants (e.g., vascular signaling mechanisms). We (and others) have suggested that the functional defect leading to increased vascular resistance reflects an imbalance between vasoconstrictor and vasodilator mechanisms (Feldman and Gros, 1998). We have focused on the hypothesis that impaired receptor-mediated vasodilation may contribute. In support of this hypothesis, we identified both in human hypertension and in experimental models that the action of G proteincoupled receptors (GPCRs) linked to adenylyl cyclase is attenuated. The prototype GPCR examined in these studies was the -adrenoceptor. However, this defect is common for a range of GPCRs commonly linked to adenylyl cyclase activation through G protein (Gs) (Feldman and Gros, 1998). The impairment in GPCR-mediated adenylyl cyclase activation in hypertension has been characterized as an “uncoupling” of the receptor from its Gs (Feldman and Gros, 1998). This has been related to an increased expression of GRK2 (an enzyme that mediates phosphorylation of agonist-occupied GPCRs and consequent uncoupling from its linked G protein). However, it has been appreciated that GPCRs, like the -adrenoceptor, may have multiple G protein linkages not related to adenylyl cyclase activation (e.g., to Gi) and may also be linked to other effector pathways (Daaka et al., 1997). This study was supported by a grant-in-aid from the Canadian Institutes of Health Research (to R.D.F.). Microarray experiments were in part supported by a group grant from the Heart and Stroke Foundation of Ontario. J.G.P. is Career Investigators of the Heart and Stroke Foundation of Ontario. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.105.013953. ABBREVIATIONS: GPCR, G protein-coupled receptor; GRK, G protein-coupled receptor kinase; PI3 kinase, phosphatidylinositol 3-kinase; ERK, extracellular signal-regulated kinase; PKA, protein kinase A; WKY, Wistar Kyoto; SHR, spontaneously hypertensive rat(s); DMEM, Dulbecco’s modified Eagle’s medium; VSMC, vascular smooth muscle cell; ISO, isoproterenol; FSK, forskolin; PDGF, platelet-derived growth factor; PBS, phosphate-buffered saline; EST, expressed sequence tag; SSC, standard saline citrate; CRE, cAMP response element. 0026-895X/06/6901-317–327$20.00 MOLECULAR PHARMACOLOGY Vol. 69, No. 1 Copyright © 2006 The American Society for Pharmacology and Experimental Therapeutics 13953/3071980 Mol Pharmacol 69:317–327, 2006 Printed in U.S.A. 317 at A PE T Jornals on M ay 6, 2017 m oharm .aspeurnals.org D ow nladed from For example, in cardiomyocytes, -adrenergic antiapoptotic effects have been reported to be mediated via both cAMPdependent and Gi-dependent coupling to phosphatidyinositol 3-kinase (PI3 kinase) (Chesley et al., 2000; Leblais et al., 2004). Furthermore, a linkage between GPCRs and -arrestin has been demonstrated, which, in some models, mediates src kinase and ERK activation (Luttrell et al., 1997). Notably several of these linkages (i.e., to Gi and to -arrestin) are enhanced by increased GRK activity (and/or PKA activity) and consequent GPCR phosphorylation (Shenoy and Lefkowitz, 2003). Thus, whether the uncoupling of GPCRs, like the -adrenoceptor, represents a global reduction in responses or a “selective” uncoupling of Gs-linked responses (with potentially enhanced non-Gs-mediated responses) is unknown. The impact of alterations in “coupling” of GPCRs linked to adenylyl cyclase activation has most commonly been studied in the context of short-term vascular re-sponses—primarily by assessing alterations in vasodilatory responses. However, it has also been appreciated that GPCR-mediated adenylyl cyclase activation might have an important role in longer term regulation of vascular growth. Elevations in intracellular cyclic AMP may be either growth inhibitory or growth stimulatory in vascular smooth muscle cells, dependent on the context (Nakaki et al., 1990). The impact of any potential shift in coupling in hypertension from Gs-linked adenylyl cyclase activation to activation of other pathways (e.g., Giand -arrestin-linked pathways) on the “net” growth-modulating effects of -adrenoceptor activation in the hypertensive state is unknown. Therefore, we have examined the effect of hypertension on 1) -adrenergic and 2) adenylyl cyclase-mediated activation of growth regulatory pathways (ERK and PI3 kinase) as well as on the regulation of gene expression as assessed by microarray analysis in vascular smooth muscle cells from normotensive and hypertensive rats. Data demonstrate that the effects of -adrenergic-mediated regulation of both ERK and PI3 kinase are blunted in hypertension. More importantly, there is a blunted effect of -adrenoceptor activation on early expression of genes important in cell growth. In net, we would conclude that in the hypertensive state there is a global attenuation of the growth regulatory effects of -adrenoceptor activation that could be an important determinant in the dysregulation of vascular smooth muscle growth characteristic of the hypertensive state. Furthermore, these data do not support an important role of “enhanced -adrenergic coupling” via “alternate” pathways in the growth regulatory effects of -adrenoceptor activation. Materials and Methods Animal Protocol. Tento 12-week-old male normotensive Wistar Kyoto (WKY) rats and spontaneously hypertensive rats (SHR) (Harlan, Indianapolis, IN) were used. The rats were cared for in accordance with Canadian Council on Animal Care guidelines and housed under a 12-h light/dark cycle with free access to standard laboratory chow and drinking water. Indirect tail-cuff measurements of systolic blood pressure were obtained in lightly anesthetized rats as described previously (Gros et al., 1994, 2000). Mean systolic pressures in SHR were significantly higher compared with WKY rats (SHR: 169 4 mm Hg, n 10; WKY: 113 2 mm Hg, n 11; p 0.001). Vascular Smooth Muscle Cell Primary Cultures. Rat aortic vascular smooth muscle cells primary cultures from SHR and WKY rats were isolated by a modification of the methods of Touyz et al. (1994). In brief, freshly isolated thoracic aortae from both normotensive and hypertensive rats were concurrently digested using collagenase and elastase incubations as described previously (Touyz et al., 1994). After digestion/isolation, vascular smooth muscle cells were resuspended in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, gentamicin, and fungizone. For all the experiments, cells were used from the third to the 12th passages. Experimental Protocol. Cells were serum starved for 24 h (DMEM supplemented with 0.1% bovine serum albumin, gentamicin, and Fungizone) before experimental treatment with drugs. After serum starvation, VSMCs were incubated in the absence or presence of isoproterenol (ISO; 10 M), forskolin (FSK; 10 M), angiotensin II (1 M), vasopressin (1 M), and platelet-derived growth factor BB (20 ng/ml) for the times indicated in the figure legends. Vascular smooth muscle cells were then used for various experimental protocols as described below. Assessment of Adenylyl Cyclase Activity. Adenylyl cyclase activity in response to isoproterenol (100 M) or forskolin (10 M) was determined by the rate of conversion of [ -P]ATP to [P]cAMP as reported previously (Gros et al., 1994, 2000). In brief, digitoninpermeabilized vascular smooth muscle cells were resuspended in a solution of Hanks’ balanced salt solution with 33 mM HEPES, 0.5 mM EDTA, and 1 mM magnesium sulfate, pH 7.4 at 4°C, added in an aliquot of 40 l to give a final incubation volume of 100 l with 1 Ci of [ -P]ATP, 0.3 mM ATP, 2 mM MgSO4, 0.1 mM cAMP (used in lieu of a phosphodiesterase inhibitor), 5 mM phosphoenol pyruvate, 40 g/ml pyruvate kinase, and 20 g/ml myokinase. Incubations were carried out at 37°C for 10 min and terminated by addition of 1 ml of a solution containing 100 g of ATP, 50 g of cAMP, and 15,000 cpm of [H]cAMP. Cells were pelleted by centrifugation at 300g for 5 min. cAMP was isolated from the supernatant by sequential Dowex and alumina chromatography and was corrected for recovery with [H]cAMP as the internal standard. Adenylyl cyclase activity was linear with time and cell number over the ranges used. Isoproterenol-mediated responses were assessed with the addition of GTP (100 M). Forskolin-mediated responses were assessed in the presence of MnCl2 (10 mM). As in our previous studies, the extent of isoproterenol-mediated adenylyl cyclase activity was expressed relatively (i.e., as a fraction of forskolin-stimulated activity), thereby minimizing the coefficient of variation seen in the comparison of absolute levels of adenylyl cyclase activity. Arborization of Vascular Smooth Muscle Cells in Response to Drug Treatment. Short-term -adrenergic effects on contractile function were determined by assessment of the extent of vascular smooth muscle arborization mechanism (Nabika et al., 1985, 1988). The arborization response mediated by elevations of cAMP has been linked to cytoskeletal changes, including reorganization of actin fibers (Westermark and Portor, 1982; Ben-Ze’ev and Amsterdam, 1987) and assembly of microtubules (Nabika et al., 1985) Vascular smooth muscle cells were cultured onto 35-mm dishes. Plates were placed in a temperature-controlled chamber maintained at 37°C (Bionomic controller; 20/20 Technology, Inc., Wilmington, NC) on an inverted microscope (Axiovert S100; Carl Zeiss, Thornwood, NY). Smooth muscle cell arborization was induced by the addition of isoproterenol (1–100 M) or forskolin (10 M). To assess the reversibility of the arborization process, the -adrenergic antagonist propranolol (1 M) was added during isoproterenol-induced arborization (i.e., the reversibility of arborization), or vascular smooth muscle cells were pretreated with propranolol for 30 min before isoproterenol stimulation. Progression of arborization was evaluated using time-lapse video microscopy with a digital recording system. Images were obtained every minute and the extent of arborization was assessed by determining the change in image intensity using the threshold setting within the image analysis software (Northern Eclipse 6.0; Empix Imaging, Toronto, ON, Canada). The change in image intensity was expressed as a percentage of basal intensity (before the addition of drug). The change in image intensity was 318 Gros et al. at A PE T Jornals on M ay 6, 2017 m oharm .aspeurnals.org D ow nladed from plotted against time, and slopes were determined from linear regression analysis using Prism 4.0 (GraphPad Software Inc., San Diego, CA). Vascular Smooth Muscle Cell Proliferation. Vascular smooth muscle cells were cultured in 24-well plates and serum starved for 24 h before experimentation. Vascular smooth muscle cells were incubated in the absence or presence of platelet-derived growth factor-BB (PDGF; 20 ng/ml) or with PDGF in the absence or presence of isoproterenol (100 M). After 21 h, [H]thymidine (1 Ci/ml; Valeant Pharmaceuticals, Costa Mesa, CA) was added for an additional 3 h before harvest. After incubation, the medium was aspirated, and the cells washed three times with ice-cold PBS, 10% trichloroacetic acid and then with distilled water and allowed to air dry. Cells were solubilized with 1 ml of 1% SDS, and radioactivity of each sample was determined by liquid scintillation spectrometry. Immunoblotting. After drug treatment, vascular smooth muscle cells were washed twice with ice-cold PBS and directly lysed in a buffer containing 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl2, 0.3% Nonidet P-40, and 1 mM Na3VO4 with 1 mM phenylmethylsulfonyl fluoride. The cell lysates were centrifuged at 500g for 5 min at 4°C. Twenty micrograms of proteins was resolved on 12% SDS-PAGE and blotted electrophoretically onto Immun-Blot polyvinylidene fluoride membrane (Bio-Rad, Hercules, CA). The membranes were blocked with 5% skim milk and incubated either with anti-phospho ERK1/2, anti-ERK1/2, or anti-PI3 kinase p85 antibody (BD Transduction Laboratories, Lexington, KY, or Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and were detected by chemiluminescence as described by the manufacturer’s protocol (PerkinElmer Life and Analytical Sciences, Boston, MA). PI3 Kinase Assay. After drug treatment, vascular smooth muscle cells were washed twice with ice-cold PBS and lysed in a buffer containing 20 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 137 mM NaCl, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, and 0.1 mM Na3VO4. Two hundred to 400 g of lysates was immunoprecipitated with anti-PY 20 (6 l) for 2 h at 4°C. Immunoprecipitates were washed once with “lysis” buffer; once with 20 mM Tris, pH 7.4, containing 0.5 mM LiCl; once with water; and once with 20 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 0.5 mM NaVO3. Washed immunoprecipitates were resuspended in 50 l of a buffer containing 20 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.5 mM EGTA, and 10 g of phosphatidylinositol. [ -P]ATP (10 Ci/assay) and MgCl2 (final concentration 20 mM) and incubated at room temperature for 30 min. The reaction was stopped by addition of 20 l of 8 M HCl, and 200 l of chloroform/methanol (50:50) was added to separate the phases. Seventy microliters of the organic phase was taken from each reaction and spotted onto a silica gel 60 plate (Whatman, Maidstone, UK), which was developed in chloroform/methanol/28% ammonium hydroxide/water (60:47:2:11.6) for 2 h. The plate was then air-dried and exposed to radiographic film. RNA Isolation for Microarray. After treatment, total RNA was extracted from vascular smooth muscle cells using an RNeasy kit (QIAGEN). The concentration and the quality (A260/A280 ratio) of the RNA were determined by spectrophotometry. Only RNA samples with A260/A280 ratios greater than 1.8 were used. Further assessment of the integrity of the RNA was tested using an Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, CA) at the London Regional Genomics Centre (Robarts Research Institute, London, ON, Canada; www.lrgc.ca). RNA with sharp and defined 28S and 18S ribosomal peaks validated good RNA integrity. RNA samples with poorly defined or missing peaks indicated degraded RNA and were not used. Microarray Experiments. All microarray experiments were performed at the London Regional Genomics Centre using the m15K mouse cDNA microarrays from the University Health Network (University of Toronto, Toronto, ON, Canada). These arrays contain 15,296 sequence-verified mouse ESTs obtained from the National Institute on Aging (Bethesda, MD), each spotted in duplicate, and include 392 duplicate clones, 109 triplicate clones, and 30 quadruplicate clones. For a complete gene list, see http://www.microarrays. ca/support/glists.html. For each treatment and each condition, at least two pools of isolated RNA samples were prepared on separate occasions. Each of these “biological replicates” comprised purified RNA from the aortae of at least five animals. Within each biological replicate, three microarray assays were performed (i.e., “technical replicates”), two of which were hybridized with the treated sample labeled with Cy3-dCTP and the untreated control labeled with Cy5dCTP. This labeling scheme was reversed for the third array (dye flips). Thus, for each condition/animal type studied, six assays were performed, based on RNA isolated from at least 10 animals. The rationale for this strategy reflected several considerations, including 1) use of RNA pools from multiple animals (as opposed to studying RNA sample from individual animals), thereby minimizing biological variability, but also minimizing the opportunity to identify “outliers” of potential interest (Kendziorski et al., 2003; Peng et al., 2003); and 2) the limited impact of the use of larger numbers of replicate chips for each condition to reduce total variability (Han et al., 2004). Labeling of the cDNA for hybridization to microarrays was carried out using the direct labeling protocol supplied by the University Health Network and done in the dark. In brief, the target was fluorescently labeled by combining 8.0 l of 5 first strand reaction buffer (SuperScript II; Invitrogen, Carlsbad, CA), 3.75 M anchored T mRNA primer (5 -TTTTTTTTTTTTTTTTTTTTVN-3 ; V A,C,G; N A,C,G,T), 500 M each of dATP, dGTP, and dTTP, 50 M dCTP, 0.63 M Cy3or Cy5-dCTP (GE Healthcare, Little Chalfont, Buckinghamshire, UK), 10 mM dithiothreitol, and 10 g of total RNA in a volume of 40 l. The labeling reaction was incubated in the dark at 65°C for 5 min and then at 42°C for 5 min. Thereafter, 2 l of reverse transcriptase (SuperScript II; Invitrogen) was added and incubated at 42°C for 2 to 3 h. The reaction was stopped by the addition of 4 l of 50 mM EDTA, pH 8.0, and 2 l of 10 N NaOH was added and incubated at 65°C for 20 min to hydrolyze the RNA. Then, 4 l of 5 M acetic acid was added to neutralize the solution. The control and experimental reactions were combined, 100 l of isopropanol was added, and the DNA was precipitated overnight at 20°C. Labeled cDNA was pelleted by centrifugation at 16,000g in the dark at 4°C, the isopropanol was decanted, and the pellet was rinsed with ice-cold 70% ethanol. The samples were then pulse centrifuged, and all remaining alcohol was removed. The pellet was resuspended in 5 l of RNaseand DNasefree distilled, deionized H2O (Invitrogen). Before hybridization, microarrays were prehybridized in 4 SSC, 0.1% SDS, and 0.2% bovine serum albumin for 20 min at 50°C. Prehybridized slides were washed well in filtered, distilled, deionized H2O and dried in a filtered nitrogen gas stream. Then, 5 l of yeast tRNA (10 mg/ml; Invitrogen) and 5 l of calf thymus DNA (10 mg/ml) (Sigma-Aldrich) were added to 100 l of DIG Easy Hyb solution (Roche Diagnostics, Indianapolis, IN). Eighty microliters of the hybridization solution was then added to each pooled pair of Cy5and Cy3-labeled cDNA, and the solution was mixed and incubated at 65°C for 2 min and cooled to room temperature in the dark. This solution was applied to the slides and placed into a humidified hybridization chamber incubated at 37°C for 18 h. When the incubation was complete, the slides were washed three times with prewarmed 1 SSC and 0.1% SDS at 50°C for 10 min with constant gentle agitation, quickly rinsed in 1 SSC, and dried in a filtered nitrogen gas stream. Microarray slides were scanned using the Virtek Chip Reader (Bio-Rad), and the resulting raw scanned images of Cy3 and Cy5 fluorescence intensities were processed using Arrayvision 6.0 (Imaging Research, St. Catharines, ON, Canada). Background-subtracted spot values were then imported into GeneSpring 6.0 (Agilent Technologies) for further analysis. For all slides, the Cy3 and Cy5 signal intensities were normalized using the intensity dependent Locally weighted linear regression algorithm (LOWESS; Yang et al., 2002) of GeneSpring 6.0, with the smoothing factor set at 20%. Any intensity value of less than 0.01 was set to 0.01. In addition, separate normalization factors were -Adrenergic Growth Signaling in Hypertension 319 at A PE T Jornals on M ay 6, 2017 m oharm .aspeurnals.org D ow nladed from calculated for each subgrid on the slide to control spatial intensity bias across the slide (“Print-tip group normalization”). Differentially expressed genes were determined using the Student’s t test algorithm of GeneSpring 6.0. A gene was said to be differentially expressed if the mean normalized expression ratio from all six arrays was significantly different from 1.0 with a p value of 0.05 set as the minimum level of significance. Determination of significant differences in the effect of isoproterenol (or forskolin) to mediate early alterations in gene expression in SHR versus WKY vascular smooth muscle cells was determined by a differential effects approach. The rationale for the utilization of this analytical approach reflects 1) appreciation of the common finding of variability in respect to which specific genes are regulated between species/strains and 2) the importance placed in determining whether quantitatively the extent of gene regulation mediated by isoproterenol differed between SHR and WKY rats. For each condition and each cell type (WKY versus SHR), the effect of drug on the regulation of gene expression was determined (versus control cells) using the criteria outlined above. The extent of that increase or decrease (each analyzed separately) was compared with the extent of change in the other cell type, and the level of statistical significance determined. Because the genes affected by the same treatment in the two cell types were not uniformly identical, the analysis was repeated but using the extent of change in affected genes in the opposite cell type as the index. The conclusion of a significant difference between cell types was made when the extent of change using one cell type as the index was determined to be significantly different from the extent of change using the other cell type as an index. Differences were determined by analysis of variance with p 0.05 as the minimum level of significance. To determine whether a specific gene contained a cyclic AMP response element, we used the searchable CREB Target Gene Database at http://natural.salk.edu/CREB as described recently by Zhang et al. (2005). Materials. All drugs (unless otherwise specified) were purchased from Sigma-Aldrich. DMEM, Fungizone, and gentamicin were purchased from Invitrogen. Anti-phospho-ERK1/2, anti-ERK1/2, and anti-PI3k p85 were purchased from Upstate Biotechnology (Lake Placid, NY). Anti-PY20 antibody was purchased from BD Transduction Laboratories.