Endoglin Regulates Cyclooxygenase-2 Expression and Activity
http://www.100md.com
Mirjana Jerkic, Juan V. Rivas-Elena, Jua
参见附件。
the Instituto "Reina Sofía" de Investigación Nefrológica (M.J., J.V.R.-E., M.Prieto, A.R.-B., F.P.-B., M.Pericacho, J.M.L.-N.), Departamento de Fisiología & Farmacología, Universidad de Salamanca, Salamanca, Spain
Hospital for Sick Children (M.J., M.L.)
Heart and Stroke Lewar Center of Excellence, University of Toronto, Toronto, Canada, Centro de Investigaciones Biológicas (J.F.S., C.B.), Consejo Superior de Investigaciones Científicas, Madrid, Spain, Laboratorio de Biologia Celular (J.F.S.), Instituto de Nutricion y Tecnologia de los Alimentos, INTA, Universidad de Chile, Santiago, Chile, Departamento de Anatomía e Histología Humanas (M.A.), Universidad de Salamanca, Salamanca, Spain, Center for Molecular Medicine (C.P.H.V.), Maine Medical Center Research Institute, Scarborough, Me.
Abstract
The endoglin heterozygous (Eng+/–) mouse, which serves as a model of hereditary hemorrhagic telangiectasia (HHT), was shown to express reduced levels of endothelial NO synthase (eNOS) with impaired activity. Because of intricate changes in vasomotor function in the Eng+/– mice and the potential interactions between the NO- and prostaglandin-producing pathways, we assessed the expression and function of cyclooxygenase (COX) isoforms. A specific upregulation of COX-2 in the vascular endothelium and increased urinary excretion of prostaglandin E2 were observed in the Eng+/– mice. Specific COX-2 inhibition with parecoxib transiently increased arterial pressure in Eng+/– but not in Eng+/+ mice. Transfection of endoglin in L6E9 myoblasts, shown previously to stimulate eNOS expression, led to downregulation of COX-2 with no change in COX-1. In addition, COX-2 promoter activity and protein levels were inversely correlated with endoglin levels, in doxycyclin-inducible endothelial cells. Chronic NO synthesis inhibition with N-nitro-L-arginine methyl ester induced a marked increase in COX-2 only in the normal Eng+/+ mice. N-nitro-L-arginine methyl ester also increased COX-2 expression and promoter activity in doxycyclin-inducible endoglin expressing endothelial cells, but not in control cells. The level of COX-2 expression following transforming growth factor-1 treatment was less in endoglin than in mock transfected L6E9 myoblasts and was higher in human endothelial cells silenced for endoglin expression. Our results indicate that endoglin is involved in the regulation of COX-2 activity. Furthermore, reduced endoglin levels and associated impaired NO production may be responsible, at least in part, for augmented COX-2 expression and activity in the Eng+/– mice.
Key Words: endoglin COX-2 NO TGF-1
Introduction
Endoglin (CD105) is a homodimeric membrane glycoprotein that, in association with transforming growth factor (TGF)- family receptors, binds TGF-1, TGF-3, activin, bone morphogenetic protein (BMP)-2, and BMP-7.1 Endoglin is constitutively expressed on endothelial cells of capillaries, veins, and arteries2,3 and can also be observed, albeit at lower levels, in contractile cells such as vascular smooth muscle cells,4 and mesangial cells.5 Mutations in the endoglin (ENG) gene cause hereditary hemorrhagic telangiectasia type 1 (HHT1), also known as Rendu–Osler–Weber syndrome.6 HHT is an autosomal dominant vascular dysplasia that affects 1:10 000 individuals. This disorder is associated with epistaxis and telangiectases in the majority of patients and with pulmonary and cerebral arteriovenous malformations that are more frequent, particularly in HHT1 patients. ENG mutations are distributed throughout the gene and lead to haploinsufficiency,7 indicating that endoglin levels are critical in maintaining vascular homeostasis. Endoglin null (Eng–/–) mice die at midgestation of vascular and cardiac defects, demonstrating the critical role of endoglin in cardiovascular development.8,9,10
The vascular endothelium secretes vasodilators including NO, which is produced mostly by endothelial NO synthase (eNOS), and the prostacyclin (PGI2) and prostaglandin E2 (PGE2), which are produced by cyclooxygenases (COXs).11 Endoglin is a regulatory component of the TGF- receptor system in endothelial cells capable of modulating specific responses to this multipotent growth factor.10,12,13 Its reduced expression may result in disruption of the delicate balance in the secretion of endothelium-derived vasodilators and vasoconstrictors, thus inducing changes in vascular tone regulation. We have previously demonstrated that endoglin haploinsufficiency leads to a complex modification in the regulation of vascular resistance associated with a decrease in eNOS expression and impaired activity in Eng+/– mice.14,15
Molecular cross-talk between NOS and COX pathways has been documented in several cell types including endothelial cells.16 However, most studies have focused on inflammatory cells and pharmacological manipulation of both pathways. Therefore, the mechanisms regulating the production, release, and fine balance of NO and prostaglandins (PGs) in endothelial cells are still poorly understood. There are reports of stimulation of PGs by NO in human microvascular and umbilical vein endothelial cells that suggest an additional pathway used by nitrovasodilators to elicit vasodilation.16 However, chronic inhibition of NO synthesis by N-nitro-L-arginine methyl ester (L-NAME) treatment of rats leads to a decreased flow-induced dilatation in resistance mesenteric arteries, compensated by an increase in COX-2 expression and vasodilatory PGs.17,18 NO stimulates PGE2 release in COX-2–deficient cells, but inhibits such release in COX-1 null cells. The inhibition of COX-2 activity is mediated by nitration of this enzyme in tyrosine residues.19 It was also reported that the diverse effects of NO on COX-2 expression can be: (1) mediated directly or via other stimuli20,21; (2) related to the cell type and state of activation22; and (3) dependent on NO concentration, being stimulatory at low levels and inhibitory at high levels.23
In the current study, we measured COX isoforms expression and activity in tissues and endothelial cells isolated from Eng+/– mice and identified an inverse relationship between endoglin and COX-2 levels. We assessed the effect of NOS and COX-2 inhibition on arterial pressure in Eng+/– and Eng+/+ mice. We also tested the effects of L-NAME on COX-2 expression and activity by chronic in vivo administration in the murine model of HHT or by in vitro treatment of cells expressing various levels of endoglin. Our data indicate that Eng+/– mice, shown previously to have impaired eNOS activity, had elevated COX-2, suggesting that endoglin plays a role in the maintenance of vascular homeostasis and the fine balance between eNOS and COX-2 in endothelial cells.
Materials and Methods
An expanded materials and methods section is available in the online data supplement at http://circres.ahajournals.org.
Mice
Generation and genotyping of Eng+/– mice on a C57BL/6 background was previously described.8,24 Mice were kept in ventilated rooms in a germ-free facility. All studies were performed in parallel in Eng+/– and Eng+/+ littermate female mice aged 4 to 6 months (20 to 25 g). All animal procedures were approved by the University of Salamanca Animal Care and Use Committee. To inhibit NO synthesis in vivo, L-NAME (Sigma) was administered at a dose of 10 mg/kg per day in the drinking water for 28 days.
Blood Pressure Measurements
Changes in mean arterial pressure (MAP) in response to vasoactive substances were measured in Eng+/– and Eng+/+ mice as described.14 Mice were treated with the nonselective COX inhibitor indomethacin (Sigma, I-7378; 5 mg/kg b.m. SC) for 1 hour, the selective COX-2 inhibitor parecoxib (Dynastat, Pharmacia EEIG, Buckinghamshire, UK; 40 mg/kg b.m. IV) for 30 minutes, or the NO synthesis inhibitor L-NAME (50 mg/kg body mass IV) for 1 hour. Subsequent combined treatments were given as indicated.
Biological Samples, Tissues, and Aortic Endothelial Cell Preparation
Urine PGE2 was measured using a high-sensitivity immunoassay (R&D Systems), and creatinine was quantified using the Jaffe method. Plasma and urine concentrations of nitrites were determined by a modification of the Griess method.14 Mice were deeply anesthetized using isoflurane (Forane, Abbott). The kidneys, lungs, and femoral arteries were removed, frozen in liquid nitrogen, individually ground into a fine powder, and stored at –80°C until used for protein and total RNA extraction. Mouse aortic endothelial cells (MAECs) from Eng+/+ and Eng+/– mice were isolated and cultured as described.14
Cell Culture and Transfections
The rat myoblast cell line L6E9 and the doxycycline-inducible bovine endothelial GM7372-EL cell line25 were cultured in DMEM, and the human microvascular endothelial (HMEC)-1 cells,26 were grown in eosin/methylene blue medium (Gibco). The generation of doxycyclin-inducible endoglin-expressing GM7372-EL cells and stable rat myoblast transfectants expressing human L-endoglin was previously described.25,27 The myoblast transfectants were usually cultured in the presence of 400 μg/mL of the G418 antibiotic. COX-2 transcriptional activity was measured using luciferase reporter gene assays28 and was expressed as relative luciferase units (RLU). To analyze the effect of TGF-1 on COX-2 expression, L6E9 myoblasts were incubated with different concentrations of recombinant human TGF-1 (0 to 500 pM) for 24 hours before Western blot analysis. For RNA interference studies, the human endothelial cell line HMEC-1 was transiently transfected with the pXP2–COX-2 luc reporter vector28 and pSUPER-Endo/Ex4 plasmid,29 encoding an endoglin-specific sequence of small interference RNA (siRNA), or pSUPER-C plasmid,29 used as negative control. After incubation for 36 hours, 400 pM TGF-1 was added, and the cells were incubated for 24 hours before measurement of the transcriptional activity.
Western and Northern Blot Analyses
Preparation of tissue and cell extracts for Western blot analysis was described.5,30 For Northern blot analysis, total RNA was isolated from tissues using the guanidinium thiocyanate–phenol–chloroform method and processed as described.14 A 4.5-kb fragment of mouse COX-2 cDNA (kindly given by Dr Santiago Lamas, Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain) was used as probe. The 28S ribosomal subunit probe (0.7 kb) served as an internal control.
Results
COX Expression in Endoglin Heterozygous Mice
We first examined the expression of COX-1 and COX-2 in tissues of mice by Western blot analysis. No difference in COX-1 protein expression was observed between the Eng+/+ and Eng+/– kidneys (Figure 1A), isolated femoral arteries (supplemental Figure I), or lung tissues (data not shown). However, expression of COX-2 protein was higher in kidneys (&2-fold; Figure 1A), femoral arteries (&2.6-fold; Figure I), and lungs (&1.8-fold; data not shown) of Eng+/– mice compared with control Eng+/+ mice. The rise in COX-2 expression was associated with increased biological activity as confirmed by elevated urinary excretion of PGE2 in Eng+/– mice (Figure 1B). Expression of COX-2 mRNA was higher in renal and pulmonary tissues of Eng+/– mice when compared with control Eng+/+ mice (&2.2- and 2.1-fold, respectively; Figure 1C).
Immunohistochemical analysis of femoral artery sections revealed that COX-2 is expressed mainly in endothelial cells but also in scattered smooth muscle cells in both Eng+/+ and Eng+/– mice (Figure 2A and 2B). The cellular staining was specific as demonstrated by the lack of reactivity of the nonimmune serum (Figure 2C). The expression of COX-2 was higher in femoral arteries of Eng+/– mice, in agreement with the elevated levels of COX-2 observed in kidneys and lungs.
COX-2 Expression Is Increased in Endothelial Cells Isolated From Endoglin Heterozygous Mice
To confirm the selective increase in COX-2 expression in endothelial cells of Eng+/– mice, we prepared primary cultures of MAECs. As expected for engineered heterozygous mice, Eng+/– MAECs had lower endoglin levels than Eng+/+ cells (Figure 3A). COX-1 expression was unchanged, whereas COX-2 expression was increased by 3-fold in aortic endothelial cells from Eng+/– mice (Figure 3A). These results indicate that endothelial cells are at least in part responsible for the higher COX-2 expression and activity observed in Eng+/– mice.
COX-2 Expression Is Decreased in Endoglin Transfected L6E9 Myoblasts
To determine whether the increase in COX-2 observed in the Eng+/– mice was related to reduced endoglin levels, we assessed COX-2 expression in mock-transfected and human endoglin-transfected L6E9 myoblast cell lines. Figure 3B illustrates the efficiency of transfection in terms of endoglin expression and reveals no changes in COX-1 levels. However, the endoglin-transfected myoblasts showed much reduced COX-2 expression relative to the mock-transfected cells. Thus, the inverse correlation between COX-2 and endoglin levels, seen in endothelial cells from the HHT murine model, was reproduced in endoglin transfected myoblasts.
COX-2 Expression Is Inversely Correlated With Endoglin Expression in GM7372-EL Cells
To substantiate the potential modulation by endoglin of COX-2 expression in endothelial cells, a tetracycline-inducible bovine endothelial cell line, GM7372-EL, was used.25 These cells have been engineered to express human endoglin when treated with doxycycline (Dox). Figure 4A shows that the level of endoglin measured by Western blot was proportional to the dose of Dox used. Interestingly, the increase in endoglin was accompanied by a parallel decrease in COX-2 and increase in eNOS expression levels (Figure 4A). We next analyzed the COX-2 promoter-driven transcription in transiently transfected GM7372-EL endothelial cells. Induction of endoglin expression with increasing concentrations of Dox resulted in a marked inhibition of COX-2 transactivity (Figure 4B).
Chronic NOS Inhibition Leads to Increased COX-2 Levels in Eng+/+ But Not in Eng+/– Mice
To assess whether lower eNOS expression in Eng+/– mice14,15 is responsible for the decreased COX-2 activity, we tested the effect of chronic NOS inhibition by treating mice with L-NAME for 28 days. The effectiveness of chronic NO synthesis inhibition was demonstrated by a decrease in plasma nitrite concentration in Eng+/+ mice (from 2.82±0.48 μmol/L in untreated mice to 0.77±0.20 μmol/L after treatment with L-NAME, P<0.01) and in Eng+/– mice (from 1.67±0.24 μmol/L in untreated mice to 0.90±0.20 μmol/L after L-NAME, P<0.05).
Western blot analysis showed that COX-2 expression in kidneys (Figure 5A) and lungs (Figure 5B) of Eng+/+ mice was higher in the L-NAME–treated group, suggesting that NOS inhibition is associated with an increase in COX-2. In Eng+/– mice, which show higher basal levels of COX-2 than the Eng+/+ mice, no further increase was obtained by chronic treatment with L-NAME. Hence, the COX-2 level in Eng+/+ mice chronically treated with L-NAME was equivalent to the basal level observed in Eng+/– mice. Urinary PGE2 excretion was also measured after chronic L-NAME treatment, as an indicator of COX-2 activity. PGE2 excretion in Eng+/+ mice was higher in the L-NAME–treated (7.2±0.6 ng/mg of creatinine) than untreated (4.2±0.2 ng/mg of creatinine; P<0.005) group. Basal PGE2 excretion in untreated Eng+/– mice (6.4±0.6 ng/ mg of creatinine) was higher than in Eng+/+ mice, as shown in Figure 1B, and these levels were not increased by chronic treatment with L-NAME (7.4±0.6 ng/ mg of creatinine; P>0.1).
Effect of NO on Endoglin-Dependent Regulation of COX-2 in Endothelial Cells
We next analyzed the effects of eNOS inhibition on COX-2 promoter activity in GM7372-EL cells, whose endoglin levels were overinduced at the highest Dox concentration. L-NAME selectively increased COX-2 promoter activity in endoglin-induced cells (Figure 6A), suggesting that reduced NOS activity was associated with increased COX-2 expression. By contrast, treatment with the NO donor sodium nitroprusside (SNP) decreased COX-2 promoter activity in both groups of cells, indicating that NO inhibited COX-2 activity.
Western blot analysis also revealed that COX-2 expression was selectively increased by L-NAME in endoglin overexpressing GM cells as compared with controls (Figure 6B). The NO donor decreased COX-2 protein levels in untreated GM cells but had no effect in endoglin overexpressing cells (Figure 6B). L-NAME also increased COX-2 protein levels in MAECs from Eng+/+, but not in those from Eng+/– mice (Figure 6C).
Induction of COX-2 Expression By TGF-1 Is Not Affected By Endoglin
To test the possibility that TGF-1 is involved in endoglin-dependent COX-2 expression, we assessed COX-2 levels in myoblasts. TGF-1 induced a dose-dependent increase in COX-2 expression in both mock and endoglin transfectants. The basal level of COX-2 was higher in the mock- versus endoglin-transfected cells, and both basal levels steadily increased after stimulation with increasing concentrations of TGF-1 (Figure 7A). The relative rise in COX-2 induced by TGF-1 was similar (1.4- and 1.6-fold at 5 pM TGF-1; 2.1- and 1.8-fold at 50 pM TGF-1), indicating that endoglin affected basal levels, but not the TGF-1–mediated increase in COX-2.
The potential involvement of endoglin in the TGF-–induced COX-2 promoter transactivation was further tested by siRNA experiments in human endothelial cells. Transfection with pSUPER-endoEx4 vector encoding specific sequences of siRNA that suppress endoglin expression30 increased the basal levels of COX-2 but did not alter the stimulation by TGF-1 (2.5-fold versus 3.3-fold in the control cells) (Figure 7B). Nevertheless, taking into account the overall absolute values observed in Figure 7, it appears that the COX-2 expression levels in response to TGF-1 are maximal in the absence of endoglin and decrease in its presence.
Hemodynamic Studies
Because we observed an increase in COX-2 expression in tissues of Eng+/– mice and a higher urinary excretion of PGE2, we next assessed if these alterations modify vascular function. Changes in MAP were measured in Eng+/+ and Eng+/– mice after selective COX-2 blockade with parecoxib. A rapid (5 minutes), transient, and statistically significant increase in blood pressure was observed only in Eng+/– mice (Figure 8A). As NO and COX products are both implicated in maintenance of vascular resistance, we determined changes in blood pressure in Eng+/+ and Eng+/– mice after COX and NOS blockade. Before inhibition of NO synthesis with L-NAME further substantiated the effect of parecoxib on Eng+/– mice. A small but significant and sustained increase in MAP was observed in Eng+/– mice, but not in their littermate controls (Figure 8B), and after 30 minutes no significant differences in MAP were seen between the 2 groups of mice. Administration of the NOS inhibitor L-NAME alone induced a sustained hypertensive response in Eng+/+ animals (&45 mm Hg; Figure 8C), whereas almost no response was observed in Eng+/– mice (Figure 8D), as reported previously.14 To assess the effect of blocking NO synthesis in the absence of COX-2–derived prostanoids, we administered L-NAME to animals pretreated with parecoxib or indomethacin. COX-2 inhibition did not modify the effects of L-NAME in Eng+/+ mice (Figure 8C). In contrast, both COX inhibitors (parecoxib and indomethacin) led to a significant increase in the pressor effect of L-NAME in Eng+/– mice (Figure 8D).
Discussion
The present study demonstrates that COX-2 was overexpressed in several tissues of Eng+/– mice relative to Eng+/+ littermates. This increased COX-2 expression was mainly observed in the vascular endothelium and was not accompanied by changes in COX-1 expression. This finding was supported by the observation that cultured aortic endothelial cells from Eng+/– mice expressed higher COX-2 but similar COX-1 levels when compared with cells from Eng+/+ mice. The inverse correlation between endoglin and COX-2 levels was also seen in the L6E9 rat myoblast cell line, which showed reduced COX-2 expression when transfected with endoglin. Also, in the bovine endothelial GM7372-EL line, Dox-inducible endoglin expression was inversely correlated with COX-2 promoter activity and protein levels.
An additional proof of increased COX-2 activity in the Eng+/– mice is their higher urinary excretion of PGE2, a major product of COX-initiated arachidonic acid metabolism in the kidney.31 Because we did not observe differences in renal COX-1 expression between Eng+/– and Eng+/+ mice but increased renal COX-2 expression in Eng+/– mice, we suggest that the increased urinary excretion of PGE2 is mainly caused by increased COX-2 metabolism.
The increase in COX-2 expression in tissues and the higher urinary excretion of PGE2 in Eng+/– mice led to measurements of arterial pressure in Eng+/+ and Eng+/– mice after COX-2 blockade. The selective inhibitor parecoxib induced a transient blood pressure increase only in Eng+/– mice, which have higher levels of COX-2. Because this enzyme generates predominantly vasodilators,32 the rise in pressure on its inhibition suggests a more significant role for COX-2 in the vasoregulation of Eng+/– than in Eng+/+ mice. As we previously showed that eNOS expression and activity was lower in these mice14,15 and that their hypertensive response to L-NAME was reduced, we measured the effects of combined inhibition of both NOS and COX activities. Pretreatment of Eng+/+ mice with parecoxib did not enhance the increase in MAP induced by L-NAME. In contrast, the same treatment applied to Eng+/– mice restored the hypertensive response to L-NAME, suggesting that COX-2–derived vasodilators contribute substantially to MAP regulation in Eng+/– mice. Furthermore, depletion of NO by pretreatment with L-NAME followed by selective COX-2 inhibition induced an increase in MAP only in Eng+/– mice. These findings suggest a critical role for COX-2 derivatives in the vasoregulation of Eng+/– mice, which may compensate for the decreased NO availability.
To investigate the specific mechanism responsible for increased COX-2 expression in Eng+/– mice, we postulated that decreased NO production in Eng+/– mice,14,15 could stimulate COX-2 expression. Our results corroborate this hypothesis, showing that chronic inhibition of NOS activity in Eng+/+ mice leads to a significant increase in COX-2 expression and PGE2 urinary excretion. However, L-NAME had no effect on the already elevated COX-2 found in Eng+/– mice. These findings suggest that chronically decreased NO synthesis in Eng+/– mice might have led to a maximal increase in COX-2 and represents a mechanism of physiological adaptation to endoglin haploinsufficiency.
Because endoglin is a coreceptor for TGF-, it was of interest to measure the effects of this cytokine on COX-2 expression in cells expressing either no or high endoglin. COX-2 can be stimulated by TGF-1 in several tissues and cell types,33–35 and endoglin can neutralize some cellular effects of TGF-1.27,36,37 Our experiments with cultured cells demonstrated that although TGF-1 increased COX-2 expression, the level of induction was not different in mock- and endoglin-transfected myoblasts, despite higher basal levels in endoglin negative cells. Additionally, suppression of endoglin expression in human endothelial cells resulted in increased basal levels in COX-2 promoter activity but did not lead to further induction by TGF-1. These results suggest that endoglin regulates total levels of COX-2 in basal and TGF-1–treated cells, but does not alter the TGF-1–dependent induction of COX-2 expression.
In conclusion, our results demonstrate upregulation of endothelial COX-2 and a consequent increase in vasodilator prostaglandin production in a mouse model of HHT1. The inverse correlation between endoglin and COX-2 expression was confirmed in endothelial and nonendothelial cells and suggests that endoglin modulates COX-2 expression. Our results also indicate that decreased eNOS activity, likely via reduction of NO production, might mediate the rise in COX-2.
Acknowledgments
Sources of Funding
This work was supported by grants from Ministerio de Educacion y Ciencia (SAF2001/1701 to J.M.L.-N. and SAF2004–01390 to C.B.), Fondo de Investigación Sanitaria (PI020200 to C.B.), HHT Foundation International to C.B., and by the Heart and Stroke Foundation of Canada (T5016) to M.L. M.J. was supported by a Fellowship from Instituto Reina Sofía de Investigación Nefrológica. C.P.H.V. was supported by NIH grant #P2015555 from the National Center for Research Resources.
Disclosures
None.
Footnotes
Original received July 15, 2005; resubmission received April 12, 2006; revised resubmission received June 22, 2006; accepted June 28, 2006.
References
Barbara NP, Wrana JL, Letarte M. Endoglin is an accessory protein that interacts with the signaling receptor complex of multiple members of the transforming growth factor- superfamily. J Biol Chem. 1999; 274: 584–594.
Li D, Chen H, Mehta JL. Angiotensin II via activation of type 1 receptor upregulates expression of endoglin in human coronary artery endothelial cells. Hypertension. 2001; 38: 1062–1067.
Cheifetz S, Bellón T, Calés C, Vera S, Bernabéu C, Massagué J, Letarte M. Endoglin is a component of the transforming growth factor- receptor system in human endothelial cells. J Biol Chem. 1992; 267: 19027–19030.
Conley BA, Smith JD, Guerrero-Esteo M, Bernabeu C, Vary CP. Endoglin, a TGF-beta receptor-associated protein, is expressed by smooth muscle cells in human atherosclerotic plaques. Atherosclerosis. 2000; 153: 323–335. [Order article via Infotrieve]
Rodriguez-Barbero A, Obreo J, Eleno N, Rodríguez-Pea A, Düwell A, Jerki M, Sánchez-Rodríguez A, Bernabéu C, López-Novoa JM. Endoglin expression in human and rat mesangial cells and its up-regulation by TGF-1. Biochem Biophys Res Commun. 2001; 282: 142–147. [Order article via Infotrieve]
MacAllister KA, Grogg KM, Johnson DW, Gallione CJ, Baldwin MA, Jackson CE, Helmbold EA, Markel DS, McKinnnon WC, Murrell J, MCCormick MK, Perick-Vance MA, Heutink P, Oostra BA, Haitjema T, Westerman CJJ, Porteus ME, Guttmacher AE, Letarte M, Marchuk DA. Endoglin, a TGF- binding protein of endothelial cells, is the gene for hereditary haemorrhagic telangiectasia type 1. Nat Genet. 1994; 8: 345–351. [Order article via Infotrieve]
Paquet ME, Pece-Barbara N, Vera S, Cymerman U, Karabegovic A, Shovlin C, Letarte M. Analysis of several endoglin mutants reveals no endogenous mature or secreted protein capable of interfering with normal endoglin function. Hum Mol Genet. 2001; 10: 1347–1357.
Bourdeau A, Dumont DJ, Letarte M. A murine model of hereditary hemorrhagic telangiectasia. J Clin Invest. 1999; 104: 1343–1351.
Li DY, Sorensen LK, Brooke BS, Urness LD, Davis EC, Taylor DG, Boak B, Wendel DP. Defective angiogenesis in mice lacking endoglin. Science. 1999; 284: 1534–1537.
Arthur HM, Ure J, Smith AJ, Renforth G, Wilson DI, Torsney E, Charlton R, Parums DV, Jowett T, Marchuk DA, Burn J, Diamond AG. Endoglin, an ancillary TGFbeta receptor, is required for extraembryonic angiogenesis and plays a key role in heart development. Dev Biol. 2000; 217: 42–53. [Order article via Infotrieve]
Morita I. Distinct functions of COX-1 and COX-2. Prostaglandins Other Lipid Mediat. 2002; 68–69:165–175.
Lebrin F, Goumans MJ, Jonker L, Carvalho R, Valdimarsdottir G, Thorikay M, Mummery C, Arthur HM, ten Dijke P. Endoglin promotes endothelial cell proliferation and TGF-/ALK1 signal transduction. EMBO J. 2004; 23: 4018–4028. [Order article via Infotrieve]
Blanco FJ, Santibanez JF, Guerrero-Esteo M, Langa C, Vary CPH, Bernabeu C. Interaction and functional interplay between endoglin and ALK1, two components of the endothelial transforming growth factor- receptor complex. J Cell Physiol. 2005; 204: 574–584. [Order article via Infotrieve]
Jerkic M, Rivas-Elena JV, Prieto M, Carron R, Sanz-Rodriguez F, Perez-Barriocanal F, Rodriguez-Barbero A, Bernabeu C, Lopez-Novoa JM. Endoglin regulates nitric oxide-dependent vasodilatation. FASEB J. 2004; 18: 609–611.
Toporsian M, Gros R, Kabir MG, Vera S, Govindaraju K, Eidelman DH, Husain M, Letarte M. A Role for endoglin in coupling eNOS activity and regulating vascular tone revealed in hereditary hemorrhagic telangiectasia. Circ Res. 2005; 96: 684–692.
Vassalle C, Domenici C, Lubrano V, L’Abbate A. Interaction between nitric oxide and cyclooxygenase pathways in endothelial cells. J Vasc Res. 2003; 40: 491–499. [Order article via Infotrieve]
Beierwaltes WH. Cyclooxygenase-2 products compensate for inhibition of nitric oxide regulation of renal perfusion. Am J Physiol Renal Physiol. 2002; 283: F68–F72.
Henrion D, Dechaux E, Dowell FJ, Maclour J, Samuel JL, Levy BI, Michel JB. Alteration of flow-induced dilatation in mesenteric resistance arteries of L-NAME treated rats and its partial association with induction of cyclo-oxygenase-2. Br J Pharmacol. 1997; 121: 83–90. [Order article via Infotrieve]
Clancy R, Varenika B, Huang W, Ballou L, Attur M, Amin AR, Abramson SB. Nitric oxide synthase/COX cross-talk: nitric oxide activates COX-1 but inhibits COX-2-derived prostaglandin production. J Immunol. 2000; 165: 1582–1587.
Perkins DJ, Kniss DA. Blockade of nitric oxide formation down-regulates cyclooxygenase-2 and decreases PGE2 biosynthesis in macrophages. J Leukoc Biol. 1999; 65: 792–799.
Cheng HF, Wang JL, Zhang MZ, McKanna JA, Harris RC. Nitric oxide regulates renal cortical cyclooxygenase-2 expression. Am J Physiol Renal Physiol. 2000; 279: F122–F129.
Pérez-Sala D, Lamas S. Regulation of cyclooxygenase-2 expression by nitric oxide in cells. Antiox Redox Signal. 2001; 3: 231–248. [Order article via Infotrieve]
Swierkosz TA, Mitchell JA, Tomlinson A, Botting RM, Warner TD, Vane JR. Relationship between different isoforms of nitric oxide synthase and cyclooxygenase in various cell types. Pol J Pharmacol. 1994; 46: 587–592. [Order article via Infotrieve]
Rodriguez-Pea A, Eleno N, Duwell A, Arevalo M, Perez-Barriocanal F, Flores O, Docherty N, Bernabeu C, Letarte M, Lopez-Novoa JM. Endoglin upregulation during experimental renal interstitial fibrosis in mice. Hypertension. 2002; 40: 713–720.
Conley BA, Koleva R, Smith JD, Kacer D, Zhang D, Bernabeu C, Vary CP. Endoglin controls cell migration and composition of focal adhesions: Function of the cytosolic domain. J Biol Chem. 2004; 279: 27440–27449.
Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, Lawley TJ. HMEC-1: Establishment of an immortalized human microvascular endothelial cell line. J Invest Dermatol. 1992; 99: 683–690. [Order article via Infotrieve]
Letamendía A, Lastres P, Botella LM, Raab U. Langa C, Velasco B, Attisano L, Bernabeu C. Role of endoglin in cellular responses to transforming growth factor-. A comparative study with betaglycan. J Biol Chem. 1998; 273: 33011–33019.
Iiguez MA, Martínez-Martínez S, Punzón C, Redondo JM, Fresno M. An essential role of the nuclear factor of activated T cells in the regulation of the expression of the cyclooxygenase-2 gene in human T lymphocytes. J Biol Chem. 2000; 275: 23627–23635.
Sanz-Rodriguez F, Guerrero-Esteo M, Botella LM, Banville D, Vary CP, Bernabeu C. Endoglin regulates cytoskeletal organization through binding to ZRP-1, a member of the Lim family of proteins. J Biol Chem. 2004; 279: 32858–32868.
Valdivielso JM, Crespo C, Alonso JR, Martínez-Salgado C, Eleno N, Arévalo M, Pérez-Barriocanal F, López-Novoa JM. Renal ischemia in the rat stimulates nitric oxide synthesis. Am J Physiol. 2001; 280: R771–R779.
Breyer MD, Breyer RM. Prostaglandin E receptors and the kidney. Am J Physiol Renal Physiol. 2000; 279: F12–F23.
Qi Z, Hao CM, Langenbach RI, Breyer RM, Redha R, Morrow JD, Breyer MD. Opposite effects of cyclooxygenase-1 and -2 activity on the pressor response to angiotensin II. J Clin Invest. 2002; 110: 61–69.
Sheng H, Shao J, O’Mahony CA, Lamps L, Albo D, Isakson PC, Berger DH, DuBois RN, Beauchamp RD. Transformation of intestinal epithelial cells by chronic TGF-beta1 treatment results in downregulation of the type II TGF-beta receptor and induction of cyclooxygenase-2. Oncogene. 1999; 18: 855–867. [Order article via Infotrieve]
Sheng H, Shao J, Dixon DA, Williams CS, Prescott SM, DuBois RN, Beauchamp RD. Transforming growth factor-beta1 enhances Ha-ras-induced expression of cyclooxygenase-2 in intestinal epithelial cells via stabilization of mRNA. J Biol Chem. 2000; 275: 6628–6635.
Keerthisingam CB, Jenkins RG, Harrison NK, Hernandez-Rodriguez NA, Booth H, Laurent GJ, Hart SL, Foster ML, McAnulty RJ. Cyclooxygenase-2 deficiency results in a loss of the anti-proliferative response to transforming growth factor-beta in human fibrotic lung fibroblasts and promotes bleomycin-induced pulmonary fibrosis in mice. Am J Pathol. 2001; 158: 1411–1422.
Diez-Marques L, Ortega-Velazquez R, Langa C, Rodriguez-Barbero A, Lopez-Novoa JM, Lamas S, Bernabeu C. Expression of endoglin in human mesangial cells: modulation of extracellular matrix synthesis. Biochim Biophys Acta. 2002; 1587: 36–44. [Order article via Infotrieve]
Obreo J, Diez-Marques L, Lamas S, Duwell A, Eleno N, Bernabeu C, Pandiella A, Lopez-Novoa JM, Rodriguez-Barbero A. Endoglin expression regulates basal and TGF-beta1-induced extracellular matrix synthesis in cultured L6E9 myoblasts. Cell Physiol Biochem. 2004; 14: 301–310. [Order article via Infotrieve]
the Instituto "Reina Sofía" de Investigación Nefrológica (M.J., J.V.R.-E., M.Prieto, A.R.-B., F.P.-B., M.Pericacho, J.M.L.-N.), Departamento de Fisiología & Farmacología, Universidad de Salamanca, Salamanca, Spain
Hospital for Sick Children (M.J., M.L.)
Heart and Stroke Lewar Center of Excellence, University of Toronto, Toronto, Canada, Centro de Investigaciones Biológicas (J.F.S., C.B.), Consejo Superior de Investigaciones Científicas, Madrid, Spain, Laboratorio de Biologia Celular (J.F.S.), Instituto de Nutricion y Tecnologia de los Alimentos, INTA, Universidad de Chile, Santiago, Chile, Departamento de Anatomía e Histología Humanas (M.A.), Universidad de Salamanca, Salamanca, Spain, Center for Molecular Medicine (C.P.H.V.), Maine Medical Center Research Institute, Scarborough, Me.
Abstract
The endoglin heterozygous (Eng+/–) mouse, which serves as a model of hereditary hemorrhagic telangiectasia (HHT), was shown to express reduced levels of endothelial NO synthase (eNOS) with impaired activity. Because of intricate changes in vasomotor function in the Eng+/– mice and the potential interactions between the NO- and prostaglandin-producing pathways, we assessed the expression and function of cyclooxygenase (COX) isoforms. A specific upregulation of COX-2 in the vascular endothelium and increased urinary excretion of prostaglandin E2 were observed in the Eng+/– mice. Specific COX-2 inhibition with parecoxib transiently increased arterial pressure in Eng+/– but not in Eng+/+ mice. Transfection of endoglin in L6E9 myoblasts, shown previously to stimulate eNOS expression, led to downregulation of COX-2 with no change in COX-1. In addition, COX-2 promoter activity and protein levels were inversely correlated with endoglin levels, in doxycyclin-inducible endothelial cells. Chronic NO synthesis inhibition with N-nitro-L-arginine methyl ester induced a marked increase in COX-2 only in the normal Eng+/+ mice. N-nitro-L-arginine methyl ester also increased COX-2 expression and promoter activity in doxycyclin-inducible endoglin expressing endothelial cells, but not in control cells. The level of COX-2 expression following transforming growth factor-1 treatment was less in endoglin than in mock transfected L6E9 myoblasts and was higher in human endothelial cells silenced for endoglin expression. Our results indicate that endoglin is involved in the regulation of COX-2 activity. Furthermore, reduced endoglin levels and associated impaired NO production may be responsible, at least in part, for augmented COX-2 expression and activity in the Eng+/– mice.
Key Words: endoglin COX-2 NO TGF-1
Introduction
Endoglin (CD105) is a homodimeric membrane glycoprotein that, in association with transforming growth factor (TGF)- family receptors, binds TGF-1, TGF-3, activin, bone morphogenetic protein (BMP)-2, and BMP-7.1 Endoglin is constitutively expressed on endothelial cells of capillaries, veins, and arteries2,3 and can also be observed, albeit at lower levels, in contractile cells such as vascular smooth muscle cells,4 and mesangial cells.5 Mutations in the endoglin (ENG) gene cause hereditary hemorrhagic telangiectasia type 1 (HHT1), also known as Rendu–Osler–Weber syndrome.6 HHT is an autosomal dominant vascular dysplasia that affects 1:10 000 individuals. This disorder is associated with epistaxis and telangiectases in the majority of patients and with pulmonary and cerebral arteriovenous malformations that are more frequent, particularly in HHT1 patients. ENG mutations are distributed throughout the gene and lead to haploinsufficiency,7 indicating that endoglin levels are critical in maintaining vascular homeostasis. Endoglin null (Eng–/–) mice die at midgestation of vascular and cardiac defects, demonstrating the critical role of endoglin in cardiovascular development.8,9,10
The vascular endothelium secretes vasodilators including NO, which is produced mostly by endothelial NO synthase (eNOS), and the prostacyclin (PGI2) and prostaglandin E2 (PGE2), which are produced by cyclooxygenases (COXs).11 Endoglin is a regulatory component of the TGF- receptor system in endothelial cells capable of modulating specific responses to this multipotent growth factor.10,12,13 Its reduced expression may result in disruption of the delicate balance in the secretion of endothelium-derived vasodilators and vasoconstrictors, thus inducing changes in vascular tone regulation. We have previously demonstrated that endoglin haploinsufficiency leads to a complex modification in the regulation of vascular resistance associated with a decrease in eNOS expression and impaired activity in Eng+/– mice.14,15
Molecular cross-talk between NOS and COX pathways has been documented in several cell types including endothelial cells.16 However, most studies have focused on inflammatory cells and pharmacological manipulation of both pathways. Therefore, the mechanisms regulating the production, release, and fine balance of NO and prostaglandins (PGs) in endothelial cells are still poorly understood. There are reports of stimulation of PGs by NO in human microvascular and umbilical vein endothelial cells that suggest an additional pathway used by nitrovasodilators to elicit vasodilation.16 However, chronic inhibition of NO synthesis by N-nitro-L-arginine methyl ester (L-NAME) treatment of rats leads to a decreased flow-induced dilatation in resistance mesenteric arteries, compensated by an increase in COX-2 expression and vasodilatory PGs.17,18 NO stimulates PGE2 release in COX-2–deficient cells, but inhibits such release in COX-1 null cells. The inhibition of COX-2 activity is mediated by nitration of this enzyme in tyrosine residues.19 It was also reported that the diverse effects of NO on COX-2 expression can be: (1) mediated directly or via other stimuli20,21; (2) related to the cell type and state of activation22; and (3) dependent on NO concentration, being stimulatory at low levels and inhibitory at high levels.23
In the current study, we measured COX isoforms expression and activity in tissues and endothelial cells isolated from Eng+/– mice and identified an inverse relationship between endoglin and COX-2 levels. We assessed the effect of NOS and COX-2 inhibition on arterial pressure in Eng+/– and Eng+/+ mice. We also tested the effects of L-NAME on COX-2 expression and activity by chronic in vivo administration in the murine model of HHT or by in vitro treatment of cells expressing various levels of endoglin. Our data indicate that Eng+/– mice, shown previously to have impaired eNOS activity, had elevated COX-2, suggesting that endoglin plays a role in the maintenance of vascular homeostasis and the fine balance between eNOS and COX-2 in endothelial cells.
Materials and Methods
An expanded materials and methods section is available in the online data supplement at http://circres.ahajournals.org.
Mice
Generation and genotyping of Eng+/– mice on a C57BL/6 background was previously described.8,24 Mice were kept in ventilated rooms in a germ-free facility. All studies were performed in parallel in Eng+/– and Eng+/+ littermate female mice aged 4 to 6 months (20 to 25 g). All animal procedures were approved by the University of Salamanca Animal Care and Use Committee. To inhibit NO synthesis in vivo, L-NAME (Sigma) was administered at a dose of 10 mg/kg per day in the drinking water for 28 days.
Blood Pressure Measurements
Changes in mean arterial pressure (MAP) in response to vasoactive substances were measured in Eng+/– and Eng+/+ mice as described.14 Mice were treated with the nonselective COX inhibitor indomethacin (Sigma, I-7378; 5 mg/kg b.m. SC) for 1 hour, the selective COX-2 inhibitor parecoxib (Dynastat, Pharmacia EEIG, Buckinghamshire, UK; 40 mg/kg b.m. IV) for 30 minutes, or the NO synthesis inhibitor L-NAME (50 mg/kg body mass IV) for 1 hour. Subsequent combined treatments were given as indicated.
Biological Samples, Tissues, and Aortic Endothelial Cell Preparation
Urine PGE2 was measured using a high-sensitivity immunoassay (R&D Systems), and creatinine was quantified using the Jaffe method. Plasma and urine concentrations of nitrites were determined by a modification of the Griess method.14 Mice were deeply anesthetized using isoflurane (Forane, Abbott). The kidneys, lungs, and femoral arteries were removed, frozen in liquid nitrogen, individually ground into a fine powder, and stored at –80°C until used for protein and total RNA extraction. Mouse aortic endothelial cells (MAECs) from Eng+/+ and Eng+/– mice were isolated and cultured as described.14
Cell Culture and Transfections
The rat myoblast cell line L6E9 and the doxycycline-inducible bovine endothelial GM7372-EL cell line25 were cultured in DMEM, and the human microvascular endothelial (HMEC)-1 cells,26 were grown in eosin/methylene blue medium (Gibco). The generation of doxycyclin-inducible endoglin-expressing GM7372-EL cells and stable rat myoblast transfectants expressing human L-endoglin was previously described.25,27 The myoblast transfectants were usually cultured in the presence of 400 μg/mL of the G418 antibiotic. COX-2 transcriptional activity was measured using luciferase reporter gene assays28 and was expressed as relative luciferase units (RLU). To analyze the effect of TGF-1 on COX-2 expression, L6E9 myoblasts were incubated with different concentrations of recombinant human TGF-1 (0 to 500 pM) for 24 hours before Western blot analysis. For RNA interference studies, the human endothelial cell line HMEC-1 was transiently transfected with the pXP2–COX-2 luc reporter vector28 and pSUPER-Endo/Ex4 plasmid,29 encoding an endoglin-specific sequence of small interference RNA (siRNA), or pSUPER-C plasmid,29 used as negative control. After incubation for 36 hours, 400 pM TGF-1 was added, and the cells were incubated for 24 hours before measurement of the transcriptional activity.
Western and Northern Blot Analyses
Preparation of tissue and cell extracts for Western blot analysis was described.5,30 For Northern blot analysis, total RNA was isolated from tissues using the guanidinium thiocyanate–phenol–chloroform method and processed as described.14 A 4.5-kb fragment of mouse COX-2 cDNA (kindly given by Dr Santiago Lamas, Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain) was used as probe. The 28S ribosomal subunit probe (0.7 kb) served as an internal control.
Results
COX Expression in Endoglin Heterozygous Mice
We first examined the expression of COX-1 and COX-2 in tissues of mice by Western blot analysis. No difference in COX-1 protein expression was observed between the Eng+/+ and Eng+/– kidneys (Figure 1A), isolated femoral arteries (supplemental Figure I), or lung tissues (data not shown). However, expression of COX-2 protein was higher in kidneys (&2-fold; Figure 1A), femoral arteries (&2.6-fold; Figure I), and lungs (&1.8-fold; data not shown) of Eng+/– mice compared with control Eng+/+ mice. The rise in COX-2 expression was associated with increased biological activity as confirmed by elevated urinary excretion of PGE2 in Eng+/– mice (Figure 1B). Expression of COX-2 mRNA was higher in renal and pulmonary tissues of Eng+/– mice when compared with control Eng+/+ mice (&2.2- and 2.1-fold, respectively; Figure 1C).
Immunohistochemical analysis of femoral artery sections revealed that COX-2 is expressed mainly in endothelial cells but also in scattered smooth muscle cells in both Eng+/+ and Eng+/– mice (Figure 2A and 2B). The cellular staining was specific as demonstrated by the lack of reactivity of the nonimmune serum (Figure 2C). The expression of COX-2 was higher in femoral arteries of Eng+/– mice, in agreement with the elevated levels of COX-2 observed in kidneys and lungs.
COX-2 Expression Is Increased in Endothelial Cells Isolated From Endoglin Heterozygous Mice
To confirm the selective increase in COX-2 expression in endothelial cells of Eng+/– mice, we prepared primary cultures of MAECs. As expected for engineered heterozygous mice, Eng+/– MAECs had lower endoglin levels than Eng+/+ cells (Figure 3A). COX-1 expression was unchanged, whereas COX-2 expression was increased by 3-fold in aortic endothelial cells from Eng+/– mice (Figure 3A). These results indicate that endothelial cells are at least in part responsible for the higher COX-2 expression and activity observed in Eng+/– mice.
COX-2 Expression Is Decreased in Endoglin Transfected L6E9 Myoblasts
To determine whether the increase in COX-2 observed in the Eng+/– mice was related to reduced endoglin levels, we assessed COX-2 expression in mock-transfected and human endoglin-transfected L6E9 myoblast cell lines. Figure 3B illustrates the efficiency of transfection in terms of endoglin expression and reveals no changes in COX-1 levels. However, the endoglin-transfected myoblasts showed much reduced COX-2 expression relative to the mock-transfected cells. Thus, the inverse correlation between COX-2 and endoglin levels, seen in endothelial cells from the HHT murine model, was reproduced in endoglin transfected myoblasts.
COX-2 Expression Is Inversely Correlated With Endoglin Expression in GM7372-EL Cells
To substantiate the potential modulation by endoglin of COX-2 expression in endothelial cells, a tetracycline-inducible bovine endothelial cell line, GM7372-EL, was used.25 These cells have been engineered to express human endoglin when treated with doxycycline (Dox). Figure 4A shows that the level of endoglin measured by Western blot was proportional to the dose of Dox used. Interestingly, the increase in endoglin was accompanied by a parallel decrease in COX-2 and increase in eNOS expression levels (Figure 4A). We next analyzed the COX-2 promoter-driven transcription in transiently transfected GM7372-EL endothelial cells. Induction of endoglin expression with increasing concentrations of Dox resulted in a marked inhibition of COX-2 transactivity (Figure 4B).
Chronic NOS Inhibition Leads to Increased COX-2 Levels in Eng+/+ But Not in Eng+/– Mice
To assess whether lower eNOS expression in Eng+/– mice14,15 is responsible for the decreased COX-2 activity, we tested the effect of chronic NOS inhibition by treating mice with L-NAME for 28 days. The effectiveness of chronic NO synthesis inhibition was demonstrated by a decrease in plasma nitrite concentration in Eng+/+ mice (from 2.82±0.48 μmol/L in untreated mice to 0.77±0.20 μmol/L after treatment with L-NAME, P<0.01) and in Eng+/– mice (from 1.67±0.24 μmol/L in untreated mice to 0.90±0.20 μmol/L after L-NAME, P<0.05).
Western blot analysis showed that COX-2 expression in kidneys (Figure 5A) and lungs (Figure 5B) of Eng+/+ mice was higher in the L-NAME–treated group, suggesting that NOS inhibition is associated with an increase in COX-2. In Eng+/– mice, which show higher basal levels of COX-2 than the Eng+/+ mice, no further increase was obtained by chronic treatment with L-NAME. Hence, the COX-2 level in Eng+/+ mice chronically treated with L-NAME was equivalent to the basal level observed in Eng+/– mice. Urinary PGE2 excretion was also measured after chronic L-NAME treatment, as an indicator of COX-2 activity. PGE2 excretion in Eng+/+ mice was higher in the L-NAME–treated (7.2±0.6 ng/mg of creatinine) than untreated (4.2±0.2 ng/mg of creatinine; P<0.005) group. Basal PGE2 excretion in untreated Eng+/– mice (6.4±0.6 ng/ mg of creatinine) was higher than in Eng+/+ mice, as shown in Figure 1B, and these levels were not increased by chronic treatment with L-NAME (7.4±0.6 ng/ mg of creatinine; P>0.1).
Effect of NO on Endoglin-Dependent Regulation of COX-2 in Endothelial Cells
We next analyzed the effects of eNOS inhibition on COX-2 promoter activity in GM7372-EL cells, whose endoglin levels were overinduced at the highest Dox concentration. L-NAME selectively increased COX-2 promoter activity in endoglin-induced cells (Figure 6A), suggesting that reduced NOS activity was associated with increased COX-2 expression. By contrast, treatment with the NO donor sodium nitroprusside (SNP) decreased COX-2 promoter activity in both groups of cells, indicating that NO inhibited COX-2 activity.
Western blot analysis also revealed that COX-2 expression was selectively increased by L-NAME in endoglin overexpressing GM cells as compared with controls (Figure 6B). The NO donor decreased COX-2 protein levels in untreated GM cells but had no effect in endoglin overexpressing cells (Figure 6B). L-NAME also increased COX-2 protein levels in MAECs from Eng+/+, but not in those from Eng+/– mice (Figure 6C).
Induction of COX-2 Expression By TGF-1 Is Not Affected By Endoglin
To test the possibility that TGF-1 is involved in endoglin-dependent COX-2 expression, we assessed COX-2 levels in myoblasts. TGF-1 induced a dose-dependent increase in COX-2 expression in both mock and endoglin transfectants. The basal level of COX-2 was higher in the mock- versus endoglin-transfected cells, and both basal levels steadily increased after stimulation with increasing concentrations of TGF-1 (Figure 7A). The relative rise in COX-2 induced by TGF-1 was similar (1.4- and 1.6-fold at 5 pM TGF-1; 2.1- and 1.8-fold at 50 pM TGF-1), indicating that endoglin affected basal levels, but not the TGF-1–mediated increase in COX-2.
The potential involvement of endoglin in the TGF-–induced COX-2 promoter transactivation was further tested by siRNA experiments in human endothelial cells. Transfection with pSUPER-endoEx4 vector encoding specific sequences of siRNA that suppress endoglin expression30 increased the basal levels of COX-2 but did not alter the stimulation by TGF-1 (2.5-fold versus 3.3-fold in the control cells) (Figure 7B). Nevertheless, taking into account the overall absolute values observed in Figure 7, it appears that the COX-2 expression levels in response to TGF-1 are maximal in the absence of endoglin and decrease in its presence.
Hemodynamic Studies
Because we observed an increase in COX-2 expression in tissues of Eng+/– mice and a higher urinary excretion of PGE2, we next assessed if these alterations modify vascular function. Changes in MAP were measured in Eng+/+ and Eng+/– mice after selective COX-2 blockade with parecoxib. A rapid (5 minutes), transient, and statistically significant increase in blood pressure was observed only in Eng+/– mice (Figure 8A). As NO and COX products are both implicated in maintenance of vascular resistance, we determined changes in blood pressure in Eng+/+ and Eng+/– mice after COX and NOS blockade. Before inhibition of NO synthesis with L-NAME further substantiated the effect of parecoxib on Eng+/– mice. A small but significant and sustained increase in MAP was observed in Eng+/– mice, but not in their littermate controls (Figure 8B), and after 30 minutes no significant differences in MAP were seen between the 2 groups of mice. Administration of the NOS inhibitor L-NAME alone induced a sustained hypertensive response in Eng+/+ animals (&45 mm Hg; Figure 8C), whereas almost no response was observed in Eng+/– mice (Figure 8D), as reported previously.14 To assess the effect of blocking NO synthesis in the absence of COX-2–derived prostanoids, we administered L-NAME to animals pretreated with parecoxib or indomethacin. COX-2 inhibition did not modify the effects of L-NAME in Eng+/+ mice (Figure 8C). In contrast, both COX inhibitors (parecoxib and indomethacin) led to a significant increase in the pressor effect of L-NAME in Eng+/– mice (Figure 8D).
Discussion
The present study demonstrates that COX-2 was overexpressed in several tissues of Eng+/– mice relative to Eng+/+ littermates. This increased COX-2 expression was mainly observed in the vascular endothelium and was not accompanied by changes in COX-1 expression. This finding was supported by the observation that cultured aortic endothelial cells from Eng+/– mice expressed higher COX-2 but similar COX-1 levels when compared with cells from Eng+/+ mice. The inverse correlation between endoglin and COX-2 levels was also seen in the L6E9 rat myoblast cell line, which showed reduced COX-2 expression when transfected with endoglin. Also, in the bovine endothelial GM7372-EL line, Dox-inducible endoglin expression was inversely correlated with COX-2 promoter activity and protein levels.
An additional proof of increased COX-2 activity in the Eng+/– mice is their higher urinary excretion of PGE2, a major product of COX-initiated arachidonic acid metabolism in the kidney.31 Because we did not observe differences in renal COX-1 expression between Eng+/– and Eng+/+ mice but increased renal COX-2 expression in Eng+/– mice, we suggest that the increased urinary excretion of PGE2 is mainly caused by increased COX-2 metabolism.
The increase in COX-2 expression in tissues and the higher urinary excretion of PGE2 in Eng+/– mice led to measurements of arterial pressure in Eng+/+ and Eng+/– mice after COX-2 blockade. The selective inhibitor parecoxib induced a transient blood pressure increase only in Eng+/– mice, which have higher levels of COX-2. Because this enzyme generates predominantly vasodilators,32 the rise in pressure on its inhibition suggests a more significant role for COX-2 in the vasoregulation of Eng+/– than in Eng+/+ mice. As we previously showed that eNOS expression and activity was lower in these mice14,15 and that their hypertensive response to L-NAME was reduced, we measured the effects of combined inhibition of both NOS and COX activities. Pretreatment of Eng+/+ mice with parecoxib did not enhance the increase in MAP induced by L-NAME. In contrast, the same treatment applied to Eng+/– mice restored the hypertensive response to L-NAME, suggesting that COX-2–derived vasodilators contribute substantially to MAP regulation in Eng+/– mice. Furthermore, depletion of NO by pretreatment with L-NAME followed by selective COX-2 inhibition induced an increase in MAP only in Eng+/– mice. These findings suggest a critical role for COX-2 derivatives in the vasoregulation of Eng+/– mice, which may compensate for the decreased NO availability.
To investigate the specific mechanism responsible for increased COX-2 expression in Eng+/– mice, we postulated that decreased NO production in Eng+/– mice,14,15 could stimulate COX-2 expression. Our results corroborate this hypothesis, showing that chronic inhibition of NOS activity in Eng+/+ mice leads to a significant increase in COX-2 expression and PGE2 urinary excretion. However, L-NAME had no effect on the already elevated COX-2 found in Eng+/– mice. These findings suggest that chronically decreased NO synthesis in Eng+/– mice might have led to a maximal increase in COX-2 and represents a mechanism of physiological adaptation to endoglin haploinsufficiency.
Because endoglin is a coreceptor for TGF-, it was of interest to measure the effects of this cytokine on COX-2 expression in cells expressing either no or high endoglin. COX-2 can be stimulated by TGF-1 in several tissues and cell types,33–35 and endoglin can neutralize some cellular effects of TGF-1.27,36,37 Our experiments with cultured cells demonstrated that although TGF-1 increased COX-2 expression, the level of induction was not different in mock- and endoglin-transfected myoblasts, despite higher basal levels in endoglin negative cells. Additionally, suppression of endoglin expression in human endothelial cells resulted in increased basal levels in COX-2 promoter activity but did not lead to further induction by TGF-1. These results suggest that endoglin regulates total levels of COX-2 in basal and TGF-1–treated cells, but does not alter the TGF-1–dependent induction of COX-2 expression.
In conclusion, our results demonstrate upregulation of endothelial COX-2 and a consequent increase in vasodilator prostaglandin production in a mouse model of HHT1. The inverse correlation between endoglin and COX-2 expression was confirmed in endothelial and nonendothelial cells and suggests that endoglin modulates COX-2 expression. Our results also indicate that decreased eNOS activity, likely via reduction of NO production, might mediate the rise in COX-2.
Acknowledgments
Sources of Funding
This work was supported by grants from Ministerio de Educacion y Ciencia (SAF2001/1701 to J.M.L.-N. and SAF2004–01390 to C.B.), Fondo de Investigación Sanitaria (PI020200 to C.B.), HHT Foundation International to C.B., and by the Heart and Stroke Foundation of Canada (T5016) to M.L. M.J. was supported by a Fellowship from Instituto Reina Sofía de Investigación Nefrológica. C.P.H.V. was supported by NIH grant #P2015555 from the National Center for Research Resources.
Disclosures
None.
Footnotes
Original received July 15, 2005; resubmission received April 12, 2006; revised resubmission received June 22, 2006; accepted June 28, 2006.
References
Barbara NP, Wrana JL, Letarte M. Endoglin is an accessory protein that interacts with the signaling receptor complex of multiple members of the transforming growth factor- superfamily. J Biol Chem. 1999; 274: 584–594.
Li D, Chen H, Mehta JL. Angiotensin II via activation of type 1 receptor upregulates expression of endoglin in human coronary artery endothelial cells. Hypertension. 2001; 38: 1062–1067.
Cheifetz S, Bellón T, Calés C, Vera S, Bernabéu C, Massagué J, Letarte M. Endoglin is a component of the transforming growth factor- receptor system in human endothelial cells. J Biol Chem. 1992; 267: 19027–19030.
Conley BA, Smith JD, Guerrero-Esteo M, Bernabeu C, Vary CP. Endoglin, a TGF-beta receptor-associated protein, is expressed by smooth muscle cells in human atherosclerotic plaques. Atherosclerosis. 2000; 153: 323–335. [Order article via Infotrieve]
Rodriguez-Barbero A, Obreo J, Eleno N, Rodríguez-Pea A, Düwell A, Jerki M, Sánchez-Rodríguez A, Bernabéu C, López-Novoa JM. Endoglin expression in human and rat mesangial cells and its up-regulation by TGF-1. Biochem Biophys Res Commun. 2001; 282: 142–147. [Order article via Infotrieve]
MacAllister KA, Grogg KM, Johnson DW, Gallione CJ, Baldwin MA, Jackson CE, Helmbold EA, Markel DS, McKinnnon WC, Murrell J, MCCormick MK, Perick-Vance MA, Heutink P, Oostra BA, Haitjema T, Westerman CJJ, Porteus ME, Guttmacher AE, Letarte M, Marchuk DA. Endoglin, a TGF- binding protein of endothelial cells, is the gene for hereditary haemorrhagic telangiectasia type 1. Nat Genet. 1994; 8: 345–351. [Order article via Infotrieve]
Paquet ME, Pece-Barbara N, Vera S, Cymerman U, Karabegovic A, Shovlin C, Letarte M. Analysis of several endoglin mutants reveals no endogenous mature or secreted protein capable of interfering with normal endoglin function. Hum Mol Genet. 2001; 10: 1347–1357.
Bourdeau A, Dumont DJ, Letarte M. A murine model of hereditary hemorrhagic telangiectasia. J Clin Invest. 1999; 104: 1343–1351.
Li DY, Sorensen LK, Brooke BS, Urness LD, Davis EC, Taylor DG, Boak B, Wendel DP. Defective angiogenesis in mice lacking endoglin. Science. 1999; 284: 1534–1537.
Arthur HM, Ure J, Smith AJ, Renforth G, Wilson DI, Torsney E, Charlton R, Parums DV, Jowett T, Marchuk DA, Burn J, Diamond AG. Endoglin, an ancillary TGFbeta receptor, is required for extraembryonic angiogenesis and plays a key role in heart development. Dev Biol. 2000; 217: 42–53. [Order article via Infotrieve]
Morita I. Distinct functions of COX-1 and COX-2. Prostaglandins Other Lipid Mediat. 2002; 68–69:165–175.
Lebrin F, Goumans MJ, Jonker L, Carvalho R, Valdimarsdottir G, Thorikay M, Mummery C, Arthur HM, ten Dijke P. Endoglin promotes endothelial cell proliferation and TGF-/ALK1 signal transduction. EMBO J. 2004; 23: 4018–4028. [Order article via Infotrieve]
Blanco FJ, Santibanez JF, Guerrero-Esteo M, Langa C, Vary CPH, Bernabeu C. Interaction and functional interplay between endoglin and ALK1, two components of the endothelial transforming growth factor- receptor complex. J Cell Physiol. 2005; 204: 574–584. [Order article via Infotrieve]
Jerkic M, Rivas-Elena JV, Prieto M, Carron R, Sanz-Rodriguez F, Perez-Barriocanal F, Rodriguez-Barbero A, Bernabeu C, Lopez-Novoa JM. Endoglin regulates nitric oxide-dependent vasodilatation. FASEB J. 2004; 18: 609–611.
Toporsian M, Gros R, Kabir MG, Vera S, Govindaraju K, Eidelman DH, Husain M, Letarte M. A Role for endoglin in coupling eNOS activity and regulating vascular tone revealed in hereditary hemorrhagic telangiectasia. Circ Res. 2005; 96: 684–692.
Vassalle C, Domenici C, Lubrano V, L’Abbate A. Interaction between nitric oxide and cyclooxygenase pathways in endothelial cells. J Vasc Res. 2003; 40: 491–499. [Order article via Infotrieve]
Beierwaltes WH. Cyclooxygenase-2 products compensate for inhibition of nitric oxide regulation of renal perfusion. Am J Physiol Renal Physiol. 2002; 283: F68–F72.
Henrion D, Dechaux E, Dowell FJ, Maclour J, Samuel JL, Levy BI, Michel JB. Alteration of flow-induced dilatation in mesenteric resistance arteries of L-NAME treated rats and its partial association with induction of cyclo-oxygenase-2. Br J Pharmacol. 1997; 121: 83–90. [Order article via Infotrieve]
Clancy R, Varenika B, Huang W, Ballou L, Attur M, Amin AR, Abramson SB. Nitric oxide synthase/COX cross-talk: nitric oxide activates COX-1 but inhibits COX-2-derived prostaglandin production. J Immunol. 2000; 165: 1582–1587.
Perkins DJ, Kniss DA. Blockade of nitric oxide formation down-regulates cyclooxygenase-2 and decreases PGE2 biosynthesis in macrophages. J Leukoc Biol. 1999; 65: 792–799.
Cheng HF, Wang JL, Zhang MZ, McKanna JA, Harris RC. Nitric oxide regulates renal cortical cyclooxygenase-2 expression. Am J Physiol Renal Physiol. 2000; 279: F122–F129.
Pérez-Sala D, Lamas S. Regulation of cyclooxygenase-2 expression by nitric oxide in cells. Antiox Redox Signal. 2001; 3: 231–248. [Order article via Infotrieve]
Swierkosz TA, Mitchell JA, Tomlinson A, Botting RM, Warner TD, Vane JR. Relationship between different isoforms of nitric oxide synthase and cyclooxygenase in various cell types. Pol J Pharmacol. 1994; 46: 587–592. [Order article via Infotrieve]
Rodriguez-Pea A, Eleno N, Duwell A, Arevalo M, Perez-Barriocanal F, Flores O, Docherty N, Bernabeu C, Letarte M, Lopez-Novoa JM. Endoglin upregulation during experimental renal interstitial fibrosis in mice. Hypertension. 2002; 40: 713–720.
Conley BA, Koleva R, Smith JD, Kacer D, Zhang D, Bernabeu C, Vary CP. Endoglin controls cell migration and composition of focal adhesions: Function of the cytosolic domain. J Biol Chem. 2004; 279: 27440–27449.
Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, Lawley TJ. HMEC-1: Establishment of an immortalized human microvascular endothelial cell line. J Invest Dermatol. 1992; 99: 683–690. [Order article via Infotrieve]
Letamendía A, Lastres P, Botella LM, Raab U. Langa C, Velasco B, Attisano L, Bernabeu C. Role of endoglin in cellular responses to transforming growth factor-. A comparative study with betaglycan. J Biol Chem. 1998; 273: 33011–33019.
Iiguez MA, Martínez-Martínez S, Punzón C, Redondo JM, Fresno M. An essential role of the nuclear factor of activated T cells in the regulation of the expression of the cyclooxygenase-2 gene in human T lymphocytes. J Biol Chem. 2000; 275: 23627–23635.
Sanz-Rodriguez F, Guerrero-Esteo M, Botella LM, Banville D, Vary CP, Bernabeu C. Endoglin regulates cytoskeletal organization through binding to ZRP-1, a member of the Lim family of proteins. J Biol Chem. 2004; 279: 32858–32868.
Valdivielso JM, Crespo C, Alonso JR, Martínez-Salgado C, Eleno N, Arévalo M, Pérez-Barriocanal F, López-Novoa JM. Renal ischemia in the rat stimulates nitric oxide synthesis. Am J Physiol. 2001; 280: R771–R779.
Breyer MD, Breyer RM. Prostaglandin E receptors and the kidney. Am J Physiol Renal Physiol. 2000; 279: F12–F23.
Qi Z, Hao CM, Langenbach RI, Breyer RM, Redha R, Morrow JD, Breyer MD. Opposite effects of cyclooxygenase-1 and -2 activity on the pressor response to angiotensin II. J Clin Invest. 2002; 110: 61–69.
Sheng H, Shao J, O’Mahony CA, Lamps L, Albo D, Isakson PC, Berger DH, DuBois RN, Beauchamp RD. Transformation of intestinal epithelial cells by chronic TGF-beta1 treatment results in downregulation of the type II TGF-beta receptor and induction of cyclooxygenase-2. Oncogene. 1999; 18: 855–867. [Order article via Infotrieve]
Sheng H, Shao J, Dixon DA, Williams CS, Prescott SM, DuBois RN, Beauchamp RD. Transforming growth factor-beta1 enhances Ha-ras-induced expression of cyclooxygenase-2 in intestinal epithelial cells via stabilization of mRNA. J Biol Chem. 2000; 275: 6628–6635.
Keerthisingam CB, Jenkins RG, Harrison NK, Hernandez-Rodriguez NA, Booth H, Laurent GJ, Hart SL, Foster ML, McAnulty RJ. Cyclooxygenase-2 deficiency results in a loss of the anti-proliferative response to transforming growth factor-beta in human fibrotic lung fibroblasts and promotes bleomycin-induced pulmonary fibrosis in mice. Am J Pathol. 2001; 158: 1411–1422.
Diez-Marques L, Ortega-Velazquez R, Langa C, Rodriguez-Barbero A, Lopez-Novoa JM, Lamas S, Bernabeu C. Expression of endoglin in human mesangial cells: modulation of extracellular matrix synthesis. Biochim Biophys Acta. 2002; 1587: 36–44. [Order article via Infotrieve]
Obreo J, Diez-Marques L, Lamas S, Duwell A, Eleno N, Bernabeu C, Pandiella A, Lopez-Novoa JM, Rodriguez-Barbero A. Endoglin expression regulates basal and TGF-beta1-induced extracellular matrix synthesis in cultured L6E9 myoblasts. Cell Physiol Biochem. 2004; 14: 301–310. [Order article via Infotrieve]
您现在查看是摘要介绍页，详见ORG附件。