help button home button Am J Pathol ASIP MEMBERSHIP
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Originally published online as doi:10.2353/ajpath.2008.070439 on February 14, 2008

Published online before print February 14, 2008
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Material
Right arrow All Versions of this Article:
ajpath.2008.070439v1
172/3/830    most recent
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Billon, A.
Right arrow Articles by Arnal, J.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Billon, A.
Right arrow Articles by Arnal, J.-F.
(American Journal of Pathology. 2008;172:830-838.)
© 2008 American Society for Investigative Pathology
DOI: 10.2353/ajpath.2008.070439

The Estrogen Effects on Endothelial Repair and Mitogen-Activated Protein Kinase Activation Are Abolished in Endothelial Nitric-Oxide (NO) Synthase Knockout Mice, but Not by NO Synthase Inhibition by N-Nitro-L-arginine Methyl Ester

Audrey Billon*, Stéphanie Lehoux{dagger}, Laetitia Lam Shang Leen{ddagger}, Henrik Laurell*, Cédric Filipe*, Vincent Benouaich*, Laurent Brouchet*, Chantal Dessy§, Pierre Gourdy*, Alain-Pierre Gadeau{ddagger}, Alain Tedgui{dagger}, Jean-Luc Balligand§ and Jean-François Arnal*

From INSERM U858,*Toulouse Cedex, France; INSERM U689,{dagger}Centre de Recherche Cardiovasculaire Inserm Lariboisière, Paris, France; INSERM U828,{ddagger}Pessac, France; and Unité de Pharmacothérapie,§Université Catholique de Louvain, Brussels, Belgium


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
We have previously shown that estrogen exerts a vasoprotective effect by accelerating reendothelialization after perivascular artery injury through activation of the estrogen receptor {alpha}. Because 17β-estradiol (E2) is known to increase the bioavailability of nitric oxide, in this study, we used the same perivascular model to characterize the role of the endothelial nitric oxide synthase (eNOS) pathway in reendothelialization. Surprisingly, we found that the stimulatory effect of E2 on reendothelialization was not altered following pharmacological inhibition of nitric-oxide synthase enzymatic activity by N-nitro-L-arginine methyl ester, whereas it was abolished in eNOS-deficient (eNOS–/–) mice. This discrepancy between eNOS gene inactivation and the pharmacological inhibition of eNOS was confirmed in a classical model of endovascular injury. When assessing the involvement of eNOS in short-term membrane-associated signaling events induced by E2, we found that E2 stimulated phosphorylation of extracellular signal-regulated kinase 1/2 in isolated perfused carotid arteries from wild-type mice in the absence or presence of N-nitro-L-arginine methyl ester, whereas this stimulation was abolished in carotid arteries from eNOS–/– mice. Similar results were obtained in primary cultures of mouse aortic endothelial cells. These data reveal an original and unexpected role of eNOS, in which its presence but not its enzymatic activity appears to be a determinant for estrogen signaling in the endothelium. The consequences of this novel function of eNOS with respect to vascular diseases should be explored.

The endothelium is a major regulator of vascular homeostasis, influencing vascular tone and exerting anticoagulant, antiplatelet adhesion/aggregation, and fibrinolytic activities. Loss of endothelial monolayer integrity represents an important early step in atherosclerotic plaque development, and likewise, endothelial destruction provoked by endoluminal angioplasty is a major effector of the subsequent restenosis.1

17β-Estradiol (E2) exerts several vasoprotective effects, such as vascular healing. We previously demonstrated that E2 accelerates reendothelialization2 and that E2 was also shown to prevent neointimal hyperplasia in rats3 and medial hyperplasia in mice.4 In addition, E2 promotes vasorelaxation and inhibits platelet aggregation through potentiation of endothelial NO and prostacyclin production. Indeed, E2 stimulates endothelial NO synthase (eNOS) activity acutely through a membrane-derived nongenomic effect.5,6 E2 also increases NO bioactivity chronically through a decreased breakdown of NO, as a consequence of a reduced production of reactive oxygen species.7,8

The effects of E2 can be mediated by estrogen receptor {alpha} (ER{alpha}) or β (ERβ), two members of the nuclear receptor superfamily that are encoded by two distinct genes.9 We and others previously showed that ER{alpha}, but not ERβ, is responsible for important vascular effects of E2 such as NO production10-12 and arterial healing.2,13 Although ERs are classically defined as ligand-activated transcription factors,14 it has become clear that short-term "extragenomic" responses play an important role in cultured endothelial cells, leading to phosphatidylinositol 3-kinase (PI3K)-AKT and eNOS activation.6,10,15 Both eNOS and a fraction of ER{alpha} are present in caveolae. Even though, to date, no direct interaction between the two proteins have been reported, they appear to be organized into a functional signaling complex in caveolae.16

Losordo and colleagues17 have previously shown that the accelerating effect of E2 on vascular reendothelialization is abolished in eNOS–/– mice, demonstrating the implication of eNOS. Concomitantly, we observed a persistence of the E2 effect on reendothelialization in a perivascular injury model using a pharmacological approach where eNOS was inhibited by N-nitro-L-arginine methyl ester (L-NAME). We have pursued this initial and contradictory observation through more thorough studies using both pharmacological inhibition of nitric-oxide synthase (NOS) activity by L-NAME and mice deficient in eNOS in both endovascular and perivascular injury models. In both models, we confirmed that the accelerating effect of E2 was abolished in eNOS–/– mice but also unexpectedly that the effect of E2 was unaltered after pharmacological inhibition of NOS activity. A similar discrepancy was also observed on the phosphorylation status of mitogen-activated protein kinase (MAPK), extracellular signal-regulated kinase 1 (ERK1) and ERK2, after E2 treatment in an ex vivo model of isolated perfused carotid artery and in vitro in primary culture of mouse aortic endothelial cells. Thus, these data argue for a crucial role of eNOS, which is independent of its nitric-oxide synthase activity in the mediation of the estrogen effect in mouse carotid artery.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Animals

The investigation was in agreement with the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health (Bethesda, MD). The mice were housed in stainless steel cages in groups of five and kept in a specific pathogen-free facility. C57Bl/6J wild-type mice were purchased from Charles River (L’Arbresle, France). eNOS–/– mice18 and ER{alpha}–/– mice19 were maintained in our animal facility.

For all surgical procedures, female mice were anesthetized by intraperitoneal injection of a mixture of ketamine and xylasine and allowed to recover on a 37°C heat pack. They were ovariectomized at 4 weeks of age and given or not given 60-day time release E2 pellets (0.1 mg of E2, releasing 80 µg/kg/day) (Innovative Research of America, Sarasota, FL) implanted subcutaneously into the backs of the animals with a sterile trochar. The NOS inhibitor, L-NAME (Sigma, Saint Quentin Fallavier, France) (50 mg/kg/day), was given in the drinking water after ovariectomy. Electric carotid artery injury was performed 2 weeks later in 6-week-old mice.

cGMP Measurements

cGMP (the second messenger of NO) content in the thoracic aorta, the lung, and the cerebellum was measured to assess the level of inhibition of NO synthase activity. Mice given L-NAME (0 or 50 mg/kg/day) (eight mice in each group) were sacrificed with an overdose of ketalar and used for determination of cGMP (the second messenger of NO) content in the thoracic aorta, in the lung, and in the cerebellum. Thoracic aortae, lungs, and cerebella were removed and frozen in liquid nitrogen, and the tissues were stored frozen (–80°C) until measurement of the cGMP level. Frozen samples were thawed in glass potter in the presence of 3-isobutyl-1-methylxanthine (10–4 M) and sonicated 10 times using ten 20-seconds bursts. The homogenates were centrifuged (4000 rpm at 4°C during 15 minutes), and the supernatant fraction was taken for radioimmunoassay of cGMP using a commercial kit (Amersham, Piscataway, NJ). The samples were acetylated to increase the sensitivity of the assay. Data are the average of values from triplicate incubations. cGMP content was expressed as picomole per milligram of protein.

Perivascular Injury Model

The perivascular injury model was performed as described previously.2 In brief, the left common carotid artery was exposed via an anterior incision of the neck. The electric injury was applied to the distal part of the common carotid artery. The optimal conditions were determined as follows: electric power of 2 W applied for 2 seconds to each millimeter of carotid artery over a total length of 4 mm, with the help of a size marker placed parallel to the long axis of the carotid.

Endovascular Injury Model

The endovascular injury model was adapted from Lindner et al.20 The right common carotid artery was exposed from the bifurcation via an anterior incision of the neck. The external carotid was ligated distally and looped proximally with 8-0 silk suture. Silk loops were placed around the internal and common carotids to temporally restrict blood flow in the area of surgical manipulation. The ligature on the common carotid was precisely positioned against the sternocleidomastoid muscle 4 mm from the carotid bifurcation. This distance was very uniform from one animal to another. The external carotid artery was incised. The injury was performed using a 0.3-mm-diameter swab built by loosely winding and sticking an 8-0 silk suture around a 0.16-mm-diameter blunted guide wire. The swab was introduced, advanced through the common carotid artery, and withdrawn three times. A 4-mm-length denudation from the bifurcation of the common carotid artery was performed using the silk ligature on the common carotid as injury limit. The device was removed, and the external carotid was ligated proximally. Blood flow was then restored by loosening the loops around the internal and common carotids. The skin incision was closed, and animals were allowed to wake up under warm conditions.

Quantification of Endothelial Regeneration

Three or 5 days after injury, the endothelial regeneration process was evaluated by staining the denuded areas with Evans blue dye (Sigma) as described previously.2 In brief, the common carotid artery was dissected with an adjacent portion of the aortic arch and carotid bifurcation. The artery was then opened longitudinally and placed between slides with Fluoprep (BioMérieux, Marcy l’Etoile, France). The ratio between the area stained in blue and the total carotid artery area was calculated. The surface of the area that remained deendothelialized was indexed to the total carotid artery area to take into account the changes in vessel area due to both elasticity of the carotid artery and the flattening of the vessel between slides.

Organ Culture

Female mice between 8 and 10 weeks of age were sacrificed by a lethal injection of sodium pentobarbital (50 mg/kg i.p.). Carotid arteries were immersed in an organ culture bath filled with Dulbecco’s modified Eagle’s medium containing antibiotics (100 IU/mL penicillin, 100 mg/L streptomycin, and 10 µg/L fungizone) and supplemented with 5% steroid-free fetal calf serum. Each arterial segment was maintained for 72 hours in a closed perfusion circuit described previously,21 and exposed to an intraluminal hydrostatic pressure of 80 mmHg with continuous renewal of medium within the intraluminal space at minimal shear stress (0.5 dynes/cm2). Organ culture of the carotid segments was performed under sterile conditions in an incubator containing 5% CO2 at 37°C. At the end of the organ culture period, 10 nmol/L E2 (Sigma) or vehicle (dimethyl sulfoxide) was injected in the intraluminal compartment for 15 to 60 minutes in the presence (100 µmol/L) or absence of L-NAME. Thereafter, vessels were quickly removed and processed as described below.

Cultured Endothelial Cells

Endothelial cells were obtained from mouse aortae according to the primary explant procedure adapted from a previously described protocol,22 allowing the isolation of limited but pure cell preparations. Serum-starved endothelial cells were exposed for the indicated periods of time to E2 (10 nmol/L) in the presence or absence of L-NAME (100 µmol/L) or NG-monomethyl-L-arginine acetate (100 and 300 µmol/L; Sigma).

Western Blot

Cultured vessels were ground in ice-cold lysis buffer containing 20 mmol/L Tris-HCl (pH 7.5), 5 mmol/L EGTA, 150 mmol/L NaCl, 20 mmol/L sodium glycerophosphate, 10 mmol/L NaF, 1 mmol/L sodium orthovanadate, 1% Triton X-100, 0.1% Tween 20, and protease inhibitors (Roche, Neuilly-sur-Seine, France). Detergent-soluble fractions were retained, and protein concentrations in samples were measured using a Bradford protein assay (Bio-Rad, Marnes-la-Coquette, France). For Western blot analysis, lysates containing 30 µg of protein were electrophoresed on polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Amersham). Membranes were incubated with anti-phospho-ERK1/2, anti-ERK1/2 (Cell Signaling Technology, Beverly, MA) or anti-ERK2 antibodies (Santa Cruz Biotechnology, Inc., Heidelberg, Germany). An enhanced chemiluminescence system was used as the detection method (ECL Plus; Amersham).

Small Interference (siRNA) Transfection

siRNA duplex oligonucleotides (ON-TARGET Plus pre-designed siRNA) were from Dharmacon, Inc. (Lafayette, CO). EA.hy926 cells (courtesy of C. J. Edgell, Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, NC) were cultured as described previously23 and transfected with siRNA (100 nmol/L) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) following the instruction provided by the supplier. Forty-eight hours later, cells were harvested and subjected to Western blot analysis as described above. The antibodies used were from Santa Cruz Biotechnology, Inc. (anti-eNOS) and Cell Signaling Technology (anti-phospho-ERK1/2 and anti-ERK1/2).

Immunohistochemical Analysis

Arteries were embedded in paraffin, and sections perpendicular to the long axis of the carotid were cut at the level of the reendothelialized area of the injured carotid artery. Sections were subjected to standard hematoxylin/eosin staining and eNOS immunodetection. Detection of eNOS expression was determined with rabbit polyclonal anti-eNOS (1:200, catalog number sc-654; Santa Cruz Biotechnology, Inc.). The specificity of eNOS antibody was tested on carotid artery cross-sections from eNOS–/– mice.

Statistics

Data are expressed as mean ± SEM. For cultured vessels, Western blot band density was quantified using Image Gauge software (Fuji, Saint Quentin en Yvelines, France). One-way analysis of variance (or two-way) was used to compare data for different genotypes or inhibitors. When analyses of variance yielded significant results, comparisons were done using Bonferroni’s t-test.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
eNOS, but Not NO Production, is Required for the Effect of E2 on Reendothelialization

To define the role of NO production in the stimulating effect of E2 on reendothelialization in the perivascular injury model, we first used a pharmacological approach; ovariectomized wild-type C57Bl/6J mice were treated chronically with L-NAME (50 mg/kg/day). As shown in Figure 1A , pharmacological inhibition of eNOS activity did not alter the accelerating effect of E2 on reendothelialization 5 days postinjury. Similar results were obtained at day 3 postinjury (data not shown). Similar results were also obtained at day 5 postinjury in C57Bl/6J mice treated chronically with a twofold higher dose of L-NAME (100 mg/kg/day; data not shown).


Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		Figure 1. Perivascular injury model: the effect of E2 on the reendothelialization process in ovariectomized wild-type (A) and eNOS–/– (B) mice treated or not treated (placebo) with E2 (80 µg/kg/day) and receiving or not receiving L-NAME (50 mg/kg/day) in the drinking water. Results show percentage of de-endothelialization calculated as ratio of remaining de-endothelialized area to total carotid artery area at day 5. *P < 0.05 vs respective placebo group.

 
Given that these results were inconsistent with a previous study by Losordo and colleagues,17 implicating eNOS in the E2 effect on reendothelialization, we decided to repeat the study in the endovascular injury model. As observed in the perivascular injury model, treatment with L-NAME did not influence the capacity of E2 to speed up reendothelialization in the endovascular injury model (Figure 2A) . To verify that the dose of L-NAME used (50 mg/kg/day) efficiently blocked NO production in our experiments, the cGMP content in the thoracic aorta of C57Bl/6J mice treated with L-NAME (0 or 50 mg/kg/day) for 8 days were 1.06 ± 0.05 and 0.35 ± 0.03 pmol/mg protein, respectively (P < 0.0001), demonstrating a threefold drop in the second messenger of NO in L-NAME-treated mice.


Figure 2
View larger version (27K):
[in this window]
[in a new window]

  		Figure 2. Endovascular injury model: the effect of E2 on the reendothelialization process in ovariectomized wild-type (A) and eNOS–/– (B) mice treated or not treated (placebo) with E2 (80 µg/kg/day) and receiving or not receiving L-NAME (50 mg/kg/day) in the drinking water. Results show percentage of de-endothelialization calculated as ratio of remaining de-endothelialized area to total carotid artery area at day 5. *P < 0.05 vs respective placebo group.

 
The cGMP content in the lung of C57Bl/6J mice given placebo or L-NAME 50 mg/kg/day were 0.22 ± 0.04 and 0.11 ± 0.02 pmol/mg protein, respectively (P < 0.001). These decreased levels of vascular cGMP were not further altered by doubling the dose of L-NAME (ie, 100 mg/kg/day for 8 days), strongly suggesting that the dose of 50 mg/kg/day used in this study as well as in previous works24,25 maximally blocks the eNOS activity. Moreover, we observed a 4.5-fold drop in cGMP in cerebellum preparations from L-NAME-treated mice, demonstrating an inhibition of neuronal NOS (data not shown). Taken together, our data show that inhibiting NO production with L-NAME did not influence the effect of E2 on endothelial regrowth in vivo, irrespective of the arterial injury model.

The divergence between our findings and those previously published17 prompted us to reassess the role of eNOS five days after injury using a transgenic approach. Indeed, in agreement with the reported data, the accelerating effect of E2 on reendothelialization was abolished in eNOS–/– mice, both in the perivascular injury model (Figure 1B) and in the endovascular injury model (Figure 2B) . Again, similar results were obtained at day 3 postinjury (data not shown).

As shown in Figures 1B and 2B , the absence of E2 effect on reendothelialization was confirmed in these animals, and no "additional" effect of L-NAME was observed in either vascular injury model. Altogether, these data showed that in vivo, eNOS, but not NO production, is required for the accelerating effect of E2 on reendothelialization.

eNOS Expression is Enhanced in the Reendothelialized Intimal Area

Given the importance of the eNOS protein in the reendothelialization process, we next performed immunohistochemical analyses to compare its expression levels in injured and uninjured carotid arteries. Compared with the endothelium of an intact (uninjured) artery (Figure 3A) , eNOS expression was clearly enhanced in the regenerated endothelium of healing carotid artery, day 5 postinjury in wild-type mice (Figure 3B) , in agreement with previous observations made in regenerated endothelium in rat26 and in proliferating cultured endothelial cells.27 In this electric injury model, the regenerated endothelial area can be easily recognized by the characteristic absence of underlying smooth muscle cells. The expression of eNOS was restricted to the intima in both intact and injured vessels, suggesting that it was mainly expressed at the level of the endothelium even after injury. As anticipated, no eNOS-specific staining was detected in arteries of eNOS–/– mice (Figure 3, C and D) .


Figure 3
View larger version (65K):
[in this window]
[in a new window]

  		Figure 3. Immunostaining of eNOS in noninjured (A, C) and reendothelialized (B, D) carotid arteries of wild-type (A, B) and eNOS–/– mice (C, D) at day 5 after perivascular injury.

 
eNOS, but Not NO Production, is Required for E2-induced ERK1/2 Phosphorylation in Isolated Perfused Carotid Artery and Primary Culture of Mouse Aortic Endothelial Cells

We then sought to explore the respective roles of eNOS deficiency and eNOS enzyme inhibition in E2-induced ERK1/2 activation using an ex vivo model. As shown in Figure 4 , ERK1/2 phosphorylation was enhanced in wild-type carotid arteries exposed to E2, reaching an activation peak at 30 minutes and declining at 60 minutes. A delayed ERK1/2 activation has previously been reported in cultured endothelial cells exposed to E2.28 We also found that ERK1/2 phosphorylation by E2 was preserved in carotid arteries from wild-type mice treated with 100 µmol/L L-NAME, demonstrating that E2 acted independently of NO production. In contrast, E2-induced ERK1/2 activation was abolished in carotid arteries from eNOS–/– mice. Similarly, E2 failed to induce ERK1/2 phosphorylation in carotid arteries from ER{alpha}–/– mice.


Figure 4
View larger version (31K):
[in this window]
[in a new window]

  		Figure 4. Acute phosphorylation of ERK1/2 induced by E2 in carotid arteries from wild-type mice is insensitive to L-NAME but is abolished in eNOS–/– carotid arteries. A: Representative Western blot showing short-term phosphorylation kinetics of ERK1/2 induced by E2 (10 nmol/L), administered in the intraluminal compartment of carotid arteries after 3 days in organ culture. ERK1/2 phosphorylation is equivalent in untreated and L-NAME-treated (100 µmol/L) carotid arteries from wild-type mice but abolished in carotid arteries from eNOS–/– and ER{alpha}–/– mice. B: Bar graph represents the quantification of the phospho-ERK1/2/ERK2 ratio, normalized to the control within each group. *P < 0.05 and ***P < 0.001 vs respective control group.

 
We then reassessed the respective roles of eNOS deficiency and eNOS enzyme inhibition in cultured endothelial cells from mouse aortae. As shown in Figure 5 , we found that E2 stimulated the phosphorylation of ERK1/2 equally in untreated and in L-NAME-treated cells from wild-type mice. NG-Monomethyl-L-arginine acetate, another NOS inhibitor, did not alter the effect of E2-stimulated phosphorylation of ERK1/2 in endothelial cells from wild-type mice (data not shown). Moreover, pretreatment with 0.5 mmol/L dipropylenetriamine NONOate, a NO donor, did not restore the effect of E2 in endothelial cells from eNOS–/– mice (data not shown). On the other hand, activation of ERK1/2 by E2 was abolished in eNOS–/– and ER{alpha}–/– cells. A deficiency in intracellular BH4 levels does not appear to explain these differences, because neither the supplementation of BH4 (3 µmol/L) nor that of the BH4 synthesis substrate sepiapaterin (50 µmol/L) to cultured aortic endothelial cells from C57BL/6 wild-type mice altered the phosphorylation of ERK1/2 in response to E2 (data not shown).


Figure 5
View larger version (31K):
[in this window]
[in a new window]

  		Figure 5. eNOS deficiency prevents E2-induced ERK1/2 activation in MAECs. Isolated mouse endothelial cells were treated with E2 (10 nmol/L) for different times. A: Representative immunoblotting experiments show that phosphorylation of ERK1/2 is enhanced by E2 in endothelial cells from wild-type but not eNOS–/– or ER{alpha}–/– mice. B: Bar graph represents the quantification of the P-ERK/ERK ratio, normalized to the control within each group (n = 2 experiments).

 
Given that gene inactivation is known to favor gene expression remodeling, particularly "gene compensations", we tried to restore eNOS expression in eNOS-deficient endothelial cells through transfection. However, we did not succeed with this approach due to the difficulty of transfecting these fragile cells, which underwent apoptosis soon after transfection.

Therefore, we sought to abolish eNOS expression in eNOS-expressing endothelial cells using siRNA against eNOS. We first performed this experiment in endothelial cells from eNOS+/+ mice, but again these cells did not tolerate transfection. We next tried a more classical cell model of endothelial cells: human umbilical vein endothelial cells. Unfortunately, we were unable to elicit ERK1/2 activation in response to E2 in human umbilical vein endothelial cells, even in freshly harvested cells at very early passages. This was probably due to the lack of ER gene expression (assessed by reverse transcription-polymerase chain reaction and Western blotting) at variance with previous work,29 but in agreement with other.30

Finally, we tried a human umbilical vein endothelial cell-derived cell line (EA.hy926) that has been shown to express ER{alpha}.31 In this model, we were able to elicit MAPK activation in response to E2 and a clear decrease of this signal after eNOS siRNA treatment (Supplemental Figure S1, available at http://ajp.amjpathol.org). These data indicate that the level of eNOS protein expression influences the capacity of E2 to activate ERK1/2 in vitro. However, we were struck by the poor reproducibility of this model, because in two of four experiments, we did not observe an increase in ERK1/2 phosphorylation in response to E2 in control cells.

Altogether, these data suggest that the E2 effect leading to ERK1/2 phosphorylation is mediated by ER{alpha} and appears to be strictly dependent on the presence of eNOS protein, but not its activity, in both ex vivo and in vitro models. It is noteworthy that the reproducibility of the effects observed in vivo or ex vivo was much more robust than the ones observed in cultured endothelial cells.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
Altogether, our results reveal an original and unexpected role for eNOS in endothelial regeneration and in plasma membrane E2 signaling, acting as a signaling protein independently of its NO-producing function. Indeed, our finding that accelerated reendothelialization induced by E2 was abolished in eNOS–/– mice agreed with a previous report;17 however, we were surprised to observe that the pharmacological inhibition of NOS with L-NAME had no influence on the beneficial effect of E2 in two models of vascular injury.

Our finding that the in vivo effects of E2 differ between L-NAME-treated and eNOS–/– mice appears robust for the following reasons. 1) It was established in two different experimental models, endovascular and electric perivascular carotid artery injury. 2) Pharmacological inhibition of eNOS using L-NAME was previously shown to maximally block NO synthase activity,24,25,32 and its efficacy was verified in the present work. 3) Abrogation of the effect of E2 by eNOS gene inactivation has now been demonstrated in two different eNOS gene inactivation models. Indeed, the model used by Losordo and colleagues17 targets exon 12 in the eNOS gene,33 whereas the one used in the present study targets exons 24 to 25.

To better understand the interaction between E2 and eNOS, we explored the respective influences of eNOS gene inactivation and eNOS activity inhibition on the short-term effects of E2 in an ex vivo artery model. This model was used to specifically study endothelium in its native matrix and mechanical (submitted to physiological pressure) environment, in the absence of confounding hormonal or circulating factors encountered in vivo.34 As previously described in cultured human, bovine, and porcine endothelial cells,28,35,36 we found that E2 induced ERK1/2 activation in isolated perfused carotid arteries. Interestingly, and in agreement with our reendothelialization data, the plasma membrane signaling of E2 was strictly dependent on the presence of eNOS protein but not eNOS activity in isolated perfused carotid arteries. The limitation of this model in interpreting ERK1/2 activation is that both endothelial and smooth muscle cells express ERK1/2. However, the exclusive expression of eNOS in the endothelium (Figure 3) on the one hand and the similarity in results between in vivo and ex vivo models on the other hand reinforce the concept that plasma membrane signaling of E2 is strictly dependent on the presence of eNOS protein but not eNOS activity in the endothelial cells of carotid arteries.

eNOS gene disruption and NOS pharmacological inhibition are often viewed as comparable strategies, as they in some aspects generate similar results, such as the induction of hypertension and endothelial dysfunction.18,24,25,32,33 However, discrepancies between these two models have already been suggested, but never directly demonstrated. For instance, eNOS–/–apoE–/– mice develop twofold more fatty streak lesions than apoE–/– mice,37 whereas inhibition of NO production by L-NAME does not influence fatty streak deposit in apoE–/– mice.25,37 Similarly, de Mey and colleagues38 have previously shown an abrogation of shear stress induced vascular remodeling in eNOS–/– mice, but not in L-NAME-treated rats,39 suggesting that eNOS, but not eNOS activity, could play a role in vascular remodeling. This kind of dissociation between protein expression and enzymatic activity was nicely demonstrated for PI3K{gamma},40 which participates in two distinct signaling cascades: a kinase-dependent pathway that controls PKB/AKT activity and a kinase-independent pathway that relies on protein interactions with a phosphodiesterase to negatively modulate cardiac contractility.

Given that eNOS is largely present in caveolae and a fraction of ER{alpha} is also localized in this membrane compartment, these two molecules could associate in a functional molecular complex.16,31 Moreover, both ER{alpha} and eNOS were shown to interact with heat shock protein 9041 and caveolin 1,42,43 which reinforces the probability of a caveolar interaction of these proteins and supports of the hypothesis that eNOS acts in a multiprotein assembly, allowing ERK1/2 activation in response to E2.

Even though L-NAME is not a specific inhibitor of eNOS but inhibits other NOS isozymes, such as neuronal NOS and to some extent inducible NOS, the lack of specificity could hardly be influencing our conclusions concerning the effects on the carotid artery. Indeed, the inhibition of inducible NOS by L-NAME is only partial in vivo at the dose used in the present study.44,45 However, L-NAME did not alter the observed loss of E2 effect in eNOS–/– mice, ruling out an inference of the other NOS isoforms in the abrogation of the E2 effects in eNOS–/– mice. Finally, a significant contribution of superoxide production by eNOS uncoupling is unlikely because 1) BH4 supplementation did not influence the result observed in wild-type cultured cells; 2) L-NAME inhibits superoxide production, as demonstrated in previous in vivo46 and ex vivo47 studies; and 3) L-NAME inhibits superoxide production of purified neuronal NOS.48,49

The in vivo effects of E2 probably involve both genomic and nongenomic mechanisms. Although ERs are classically defined as ligand-activated transcription factors,14 it has become clear that short-term "extragenomic-to-genomic" responses play an important role in cultured endothelial cells,6 but also in osteoblasts and in cortical neurons (as for instance the activation of PI3K-AKT pathways as well as MAPK pathways).50,51 The characterization of the respective roles of these "membrane" and "classical" effects represents an important but difficult challenge for the next few years.15,52,53

Beyond this unexpected role of eNOS in E2 signaling, the clinical implications of our observations could be significant. First, the strong correlation between our in vivo and ex vivo findings support an implication of plasma membrane estrogen signaling in its protective effects. Second, given that deficiency in E2 has been recognized to modulate endothelial function,54 altered eNOS expression or trafficking in the context of endothelial dysfunction could hinder the beneficial effects of E2, thus sustaining a vicious circle. This could for instance help to explain the highly negative relationship reported between smoking and estrogen atheroprotection.55 Third, the proposed role of eNOS, acting within a molecular complex independently of its NO producing activity, may not be restricted to E2 signaling but may also influence endothelial response to other vasoactive compounds. These specific roles of eNOS should be more precisely explored in future work.


   Acknowledgements
 
We thank Alexia Schambourg, Marie-José Fouque, and Hortense Berges for invaluable technical assistance. eNOS-deficient mice were kindly provided by Pr Godecke (Dusseldorf, Germany), and ER{alpha}-deficient mice were kindly provided by Pr Chambon and Dr A. Krust (Strasbourg, France).


   Footnotes
 
Address reprint requests to Jean-François Arnal, INSERM U858, I2MR, CHU Rangueil–BP 84225, 31432 Toulouse Cedex 4, France. E-mail: arnal{at}toulouse.inserm.fr

Supported by INSERM, Université Paul Sabatier and Faculté de Médecine Toulouse-Rangueil, the European Vascular Genomics Network (the European Community’s Sixth Framework Programme for Research, Contract Number LSHM-CT-2003-503254), the Fondation de France, the Fondation de l’Avenir, and the Conseil Régional Midi-Pyrénées and Aquitaine. A.B. was supported by a grant from the Société Française d’Hypertension Artérielle.

Supplemental material for this article can be found on http://ajp.amjpathol.org.

Accepted for publication December 5, 2007.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bennett MR, O'Sullivan M: Mechanisms of angioplasty and stent restenosis: implications for design of rational therapy. Pharmacol Ther 2001, 91:149-166[CrossRef][Medline]
  2. Brouchet L, Krust A, Dupont S, Chambon P, Bayard F, Arnal JF: Estradiol accelerates reendothelialization in mouse carotid artery through estrogen receptor-alpha but not estrogen receptor-beta. Circulation 2001, 103:423-428[Abstract/Free Full Text]
  3. Krasinski K, Spyridopoulos I, Asahara T, van der Zee R, Isner JM, Losordo DW: Estradiol accelerates functional endothelial recovery after arterial injury. Circulation 1997, 95:1768-1772[Abstract/Free Full Text]
  4. Sullivan TR, Jr, Karas RH, Aronovitz M, Faller GT, Ziar JP, Smith JJ, O'Donnell TF, Jr, Mendelsohn ME: Estrogen inhibits the response-to-injury in a mouse carotid artery model. J Clin Invest 1995, 96:2482-2488[Medline]
  5. Lantin-Hermoso RL, Rosenfeld CR, Yuhanna IS, German Z, Chen Z, Shaul PW: Estrogen acutely stimulates nitric oxide synthase activity in fetal pulmonary artery endothelium. Am J Physiol 1997, 273:L119-L126[Medline]
  6. Mendelsohn ME: Nongenomic E: R-mediated activation of endothelial nitric oxide synthase: how does it work? What does it mean? Circ Res 2000, 87:956-960[Abstract/Free Full Text]
  7. Arnal JF, Clamens S, Pechet C, Nègre-Salvayre A, Allera C, Girolami JP, Salvayre R, Bayard F: Ethinylestradiol does not enhance the expression of nitric oxide synthase in bovine aortic endothelial cells but increases the release of bioactive nitric oxide by inhibiting superoxide anion production. Proc Natl Acad Sci USA 1996, 93:4108-4113[Abstract/Free Full Text]
  8. Wagner AH, Schroeter MR, Hecker M: 17beta-Estradiol inhibition of NADPH oxidase expression in human endothelial cells. FASEB J 2001, 15:2121-2130[Abstract/Free Full Text]
  9. Couse JF, Korach KS: Estrogen receptor null mice: what have we learned and where will they lead us? Endocr Rev 1999, 20:358-417[Abstract/Free Full Text]
  10. Chen Z, Yuhanna IS, Galcheva-Gargova Z, Karas RH, Mendelsohn ME, Shaul PW: Estrogen receptor alpha mediates the nongenomic activation of endothelial nitric oxide synthase by estrogen. J Clin Invest 1999, 103:401-406[Medline]
  11. Darblade B, Pendaries C, Krust A, Dupont S, Fouque MJ, Rami J, Chambon P, Bayard F, Arnal JF: Estradiol alters nitric oxide production in the mouse aorta through the alpha-, but not beta-, estrogen receptor. Circ Res 2002, 90:413-419[Abstract/Free Full Text]
  12. Pendaries C, Darblade B, Rochaix P, Krust A, Chambon P, Korach KS, Bayard F, Arnal JF: The AF-1 activation-function of ERalpha may be dispensable to mediate the effect of estradiol on endothelial NO production in mice. Proc Natl Acad Sci USA 2002, 99:2205-2210[Abstract/Free Full Text]
  13. Pare G, Krust A, Karas RH, Dupont S, Aronovitz M, Chambon P, Mendelsohn ME: Estrogen receptor-alpha mediates the protective effects of estrogen against vascular injury. Circ Res 2002, 90:1087-1092[Abstract/Free Full Text]
  14. Reid G, Denger S, Kos M, Gannon F: Human estrogen receptor-alpha: regulation by synthesis, modification and degradation. Cell Mol Life Sci 2002, 59:821-831[CrossRef][Medline]
  15. Haynes MP, Li L, Russell KS, Bender JR: Rapid vascular cell responses to estrogen and membrane receptors. Vascul Pharmacol 2002, 38:99-108[CrossRef][Medline]
  16. Chambliss KL, Yuhanna IS, Mineo C, Liu P, German Z, Sherman TS, Mendelsohn ME, Anderson RG, Shaul PW: Estrogen receptor alpha and endothelial nitric oxide synthase are organized into a functional signaling module in caveolae. Circ Res 2000, 87:E44-E52[Medline]
  17. Iwakura A, Luedemann C, Shastry S, Hanley A, Kearney M, Aikawa R, Isner JM, Asahara T, Losordo DW: Estrogen-mediated, endothelial nitric oxide synthase-dependent mobilization of bone marrow-derived endothelial progenitor cells contributes to reendothelialization after arterial injury. Circulation 2003, 108:3115-3121[Abstract/Free Full Text]
  18. Godecke A, Decking UK, Ding Z, Hirchenhain J, Bidmon HJ, Godecke S, Schrader J: Coronary hemodynamics in endothelial NO synthase knockout mice. Circ Res 1998, 82:186-194[Abstract/Free Full Text]
  19. Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M: Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development 2000, 127:4277-4291[Abstract]
  20. Lindner V, Fingerle J, Reidy MA: Mouse model of arterial injury. Circ Res 1993, 73:792-796[Abstract/Free Full Text]
  21. Lemarie CA, Esposito B, Tedgui A, Lehoux S: Pressure-induced vascular activation of nuclear factor-kappaB: role in cell survival. Circ Res 2003, 93:207-212[Abstract/Free Full Text]
  22. Sonveaux P, Martinive P, DeWever J, Batova Z, Daneau G, Pelat M, Ghisdal P, Gregoire V, Dessy C, Balligand JL, Feron O: Caveolin-1 expression is critical for vascular endothelial growth factor-induced ischemic hindlimb collateralization and nitric oxide-mediated angiogenesis. Circ Res 2004, 95:154-161[Abstract/Free Full Text]
  23. Edgell CJ, McDonald CC, Graham JB: Permanent cell line expressing human factor VIII-related antigen established by hybridization. Proc Natl Acad Sci USA 1983, 80:3734-3737[Abstract/Free Full Text]
  24. Arnal JF, Warin L, Michel JB: Determinants of aortic cyclic guanosine monophosphate in hypertension induced by chronic inhibition of nitric oxide synthase. J Clin Invest 1992, 90:647-652[Medline]
  25. Elhage R, Bayard F, Richard V, Holvoet P, Duverger N, Fievet C, Arnal JF: Prevention of fatty streak formation of 17beta-estradiol is not mediated by the production of nitric oxide in apolipoprotein E-deficient mice. Circulation 1997, 96:3048-3052[Abstract/Free Full Text]
  26. Poppa V, Miyashiro JK, Corson MA, Berk BC: Endothelial NO synthase is increased in regenerating endothelium after denuding injury of the rat aorta. Arterioscler Thromb Vasc Biol 1998, 18:1312-1321[Abstract/Free Full Text]
  27. Arnal JF, Yamin J, Dockery S, Harrison DG: Regulation of endothelial nitric oxide synthase mRNA, protein, and activity during cell growth. Am J Physiol 1994, 267:C1381-C1388[Medline]
  28. Klinge CM, Blankenship KA, Risinger KE, Bhatnagar S, Noisin EL, Sumanasekera WK, Zhao L, Brey DM, Keynton RS: Resveratrol and estradiol rapidly activate MAPK signaling through estrogen receptors alpha and beta in endothelial cells. J Biol Chem 2005, 280:7460-7468[Abstract/Free Full Text]
  29. Li L, Haynes MP, Bender JR: Plasma membrane localization and function of the estrogen receptor alpha variant (ER46) in human endothelial cells. Proc Natl Acad Sci USA 2003, 100:4807-4812[Abstract/Free Full Text]
  30. Baker VL, Chao VA, Murai JT, Zaloudek CJ, Taylor RN: Human umbilical vessels and cultured umbilical vein endothelial and smooth muscle cells lack detectable protein and mRNA encoding estrogen receptors. J Soc Gynecol Investig 1997, 4:316-324[CrossRef][Medline]
  31. Figtree GA, McDonald D, Watkins H, Channon KM: Truncated estrogen receptor alpha 46-kDa isoform in human endothelial cells: relationship to acute activation of nitric oxide synthase. Circulation 2003, 107:120-126[Abstract/Free Full Text]
  32. Chatziantoniou C, Boffa J, Ardaillou R, Dussaule J: Nitric oxide inhibition induces early activation of type I collagen gene in renal resistance vessels and glomeruli in transgenic mice. J Clin Invest 1998, 101:2780-2789[Medline]
  33. Shesely EG, Maeda N, Kim HS, Desai KM, Krege JH, Laubach VE, Sherman PA, Sessa WC, Smithies O: Elevated blood pressures in mice lacking endothelial nitric oxide synthase. Proc Natl Acad Sci USA 1996, 93:13176-13181[Abstract/Free Full Text]
  34. Lehoux S, Esposito B, Merval R, Tedgui A: Differential regulation of vascular focal adhesion kinase by steady stretch and pulsatility. Circulation 2005, 111:643-649[Abstract/Free Full Text]
  35. Geraldes P, Sirois MG, Bernatchez PN, Tanguay JF: Estrogen regulation of endothelial and smooth muscle cell migration and proliferation: role of p38 and p42/44 mitogen-activated protein kinase. Arterioscler Thromb Vasc Biol 2002, 22:1585-1590[Abstract/Free Full Text]
  36. Razandi M, Pedram A, Park ST, Levin ER: Proximal events in signaling by plasma membrane estrogen receptors. J Biol Chem 2003, 278:2701-2712[Abstract/Free Full Text]
  37. Knowles JW, Reddick RL, Jennette JC, Shesely EG, Smithies O, Maeda N: Enhanced atherosclerosis and kidney dysfunction in eNOS(–/–)Apoe(–/–) mice are ameliorated by enalapril treatment. J Clin Invest 2000, 105:451-458[Medline]
  38. van der Heijden OW, Essers YP, Spaanderman ME, De Mey JG, van Eys GJ, Peeters LL: Uterine artery remodeling in pseudopregnancy is comparable to that in early pregnancy. Biol Reprod 2005, 73:1289-1293[Abstract/Free Full Text]
  39. Ceiler DL, De Mey JG: Chronic N(G)-Nitro-L-arginine methyl ester treatment does not prevent flow-induced remodeling in mesenteric feed arteries and arcading arterioles. Arterioscler Thromb Vasc Biol 2000, 20:2057-2063[Abstract/Free Full Text]
  40. Patrucco E, Notte A, Barberis L, Selvetella G, Maffei A, Brancaccio M, Marengo S, Russo G, Azzolino O, Rybalkin SD, Silengo L, Altruda F, Wetzker R, Wymann MP, Lembo G, Hirsch E: PI3Kgamma modulates the cardiac response to chronic pressure overload by distinct kinase-dependent and -independent effects. Cell 2004, 118:375-387[CrossRef][Medline]
  41. Russell KS, Haynes MP, Caulin-Glaser T, Rosneck J, Sessa WC, Bender JR: Estrogen stimulates heat shock protein 90 binding to endothelial nitric oxide synthase in human vascular endothelial cells. Effects on calcium sensitivity and NO release. J Biol Chem 2000, 275:5026-5030[Abstract/Free Full Text]
  42. Garcia-Cardena G, Fan R, Stern DF, Liu J, Sessa WC: Endothelial nitric oxide synthase is regulated by tyrosine phosphorylation and interacts with caveolin-1. J Biol Chem 1996, 271:27237-27240[Abstract/Free Full Text]
  43. Razandi M, Oh P, Pedram A, Schnitzer J, Levin ER: ERs associate with and regulate the production of caveolin: implications for signaling and cellular actions. Mol Endocrinol 2002, 16:100-115[Abstract/Free Full Text]
  44. Darblade B, Batkai S, Causse E, Gourdy P, Fouque MJ, Rami J, Arnal JF: Failure of L-nitroarginine to inhibit the activity of aortic inducible nitric oxide synthase. J Vasc Res 2001, 38:266-275[CrossRef][Medline]
  45. Miller M, Thompson J, Lui X, Eboly-Childress S, Sadowska-Krowicka H, Zhang X, CLark D: Failure of L-NAME to cause inhibition of nitric oxide synthesis: role of inducible nitric oxide synthase. Inflamm Res 1996, 45:272-276[CrossRef][Medline]
  46. van Deel ED, Merkus D, van Haperen R, de Waard MC, de Crom R, Duncker DJ: Vasomotor control in mice overexpressing human endothelial nitric oxide synthase. Am J Physiol Heart Circ Physiol 2007, 293:H1144-H1153[Abstract/Free Full Text]
  47. Bendall JK, Alp NJ, Warrick N, Cai S, Adlam D, Rockett K, Yokoyama M, Kawashima S, Channon KM: Stoichiometric relationships between endothelial tetrahydrobiopterin, endothelial NO synthase (eNOS) activity, and eNOS coupling in vivo: insights from transgenic mice with endothelial-targeted GTP cyclohydrolase 1 and eNOS overexpression. Circ Res 2005, 97:864-871[Abstract/Free Full Text]
  48. Cardounel AJ, Xia Y, Zweier JL: Endogenous methylarginines modulate superoxide as well as nitric oxide generation from neuronal nitric-oxide synthase: differences in the effects of monomethyl- and dimethylarginines in the presence and absence of tetrahydrobiopterin. J Biol Chem 2005, 280:7540-7549[Abstract/Free Full Text]
  49. Pou S, Keaton L, Surichamorn W, Rosen GM: Mechanism of superoxide generation by neuronal nitric-oxide synthase. J Biol Chem 1999, 274:9573-9580[Abstract/Free Full Text]
  50. Kousteni S, Bellido T, Plotkin LI, O'Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC: Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 2001, 104:719-730[Medline]
  51. Mannella P, Brinton RD: Estrogen receptor protein interaction with phosphatidylinositol 3-kinase leads to activation of phosphorylated Akt and extracellular signal-regulated kinase 1/2 in the same population of cortical neurons: a unified mechanism of estrogen action. J Neurosci 2006, 26:9439-9447[Abstract/Free Full Text]
  52. Mendelsohn ME, Karas RH: Molecular and cellular basis of cardiovascular gender differences. Science 2005, 308:1583-1587[Abstract/Free Full Text]
  53. Warner M, Gustafsson JA: Nongenomic effects of estrogen: why all the uncertainty? Steroids 2006, 71:91-95[CrossRef][Medline]
  54. Sader MA, Celermajer DS: Endothelial function, vascular reactivity and gender differences in the cardiovascular system. Cardiovasc Res 2002, 53:597-604[Free Full Text]
  55. Bolego C, Poli A, Paoletti R: Smoking and gender. Cardiovasc Res 2002, 53:568-576[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Physiol. Rev.Home page
J.-L. Balligand, O. Feron, and C. Dessy
eNOS Activation by Physical Forces: From Short-Term Regulation of Contraction to Chronic Remodeling of Cardiovascular Tissues
Physiol Rev, April 1, 2009; 89(2): 481 - 534.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
A. Billon-Gales, C. Fontaine, C. Filipe, V. Douin-Echinard, M.-J. Fouque, G. Flouriot, P. Gourdy, F. Lenfant, H. Laurell, A. Krust, et al.
The transactivating function 1 of estrogen receptor {alpha} is dispensable for the vasculoprotective actions of 17{beta}-estradiol
PNAS, February 10, 2009; 106(6): 2053 - 2058.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
D. G.M. Molin, N. M.S van den Akker, and M. J. Post
Affirmative Action of Osteopontin on Endothelial Progenitors
Arterioscler. Thromb. Vasc. Biol., December 1, 2008; 28(12): 2099 - 2100.
[Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
L. Lam Shang Leen, C. Filipe, A. Billon, B. Garmy-Susini, S. Jalvy, F. Robbesyn, D. Daret, C. Allieres, S. R. Rittling, N. Werner, et al.
Estrogen-Stimulated Endothelial Repair Requires Osteopontin
Arterioscler. Thromb. Vasc. Biol., December 1, 2008; 28(12): 2131 - 2136.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
C. Filipe, L. Lam Shang Leen, L. Brouchet, A. Billon, V. Benouaich, V. Fontaine, P. Gourdy, F. Lenfant, J.-F. Arnal, A.-P. Gadeau, et al.
Estradiol accelerates endothelial healing through the retrograde commitment of uninjured endothelium
Am J Physiol Heart Circ Physiol, June 1, 2008; 294(6): H2822 - H2830.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Material
Right arrow All Versions of this Article:
ajpath.2008.070439v1
172/3/830    most recent
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Billon, A.
Right arrow Articles by Arnal, J.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Billon, A.
Right arrow Articles by Arnal, J.-F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS