AJP - Heart Calcium Transients and Cell-Sarcomere
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 282: H219-H231, 2002;
0363-6135/02 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (23)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Piech, A.
Right arrow Articles by Balligand, J.-L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Piech, A.
Right arrow Articles by Balligand, J.-L.
Vol. 282, Issue 1, H219-H231, January 2002

Decreased expression of myocardial eNOS and caveolin in dogs with hypertrophic cardiomyopathy

A. Piech1,*, P. E. Massart2,*, C. Dessy1, O. Feron1, X. Havaux2, N. Morel3, J.-L. Vanoverschelde2, J. Donckier4, and J.-L. Balligand1

1 Unit of Pharmacology and Therapeutics, 2 Division of Cardiology, and Departments of 3 Physiology and 4 Internal Medicine, Université Catholique de Louvain, B-1200 Brussels, Belgium


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Because nitric oxide (NO) regulates cardiac and vessel contraction, we compared the expression and activity of the endothelial NO synthase (eNOS) and caveolin, which tonically inhibits eNOS in normal and hypertrophic cardiomyopathic hearts. NOS activity (L-[3H]citrulline formation), eNOS immunostaining, and caveolin abundance were measured in heart tissue of 23 mongrel dogs before and at 3 and 7 wk of perinephritic hypertension (PHT). Hemodynamic parameters in vivo and endothelial NO-dependent relaxation of macro- and coronary microvessels in vitro were assessed in the same animals. eNOS immunostaining and total calcium-dependent NOS activity decreased at 7 wk in all four heart cavities (in left ventricle, from 17.0 ± 1.3 to 0.2 ± 0.2 fmol · min-1 · mg protein-1, P < 0.001). Caveolin-1 and -3 also decreased in PHT dog hearts. Accordingly, basal vascular tone was preserved, but maximal endothelial NO-dependent relaxation was impaired in all vessels from 7-wk PHT dogs. The latter had preserved systolic function but impaired diastolic relaxation [relaxation time constant (T1), 25.1 ± 0.9 vs. 22.0 ± 1 ms in controls; P < 0.05]. Peripheral infusion of the NOS inhibitor NG-nitro-L-arginine methyl ester increased mean aortic pressure in both groups and reduced diastolic (T1, 31.9 ± 1.4 ms) and systolic function in PHT dogs (DP40, 47.5 ± 2.5 vs. 59.4 ± 3.8 s-1 in control animals). In conclusion, both eNOS and caveolin proteins are decreased in the hypertrophic hearts of PHT dogs. This is associated with altered maximal (but not basal) vascular relaxation and impaired diastolic function. Further degradation of cardiac function after NOS inhibition suggests a critical role of residual NOS activity, probably supported by the concurrent downregulation of caveolin.

cardiac hypertrophy; diastolic function; perinephritic hypertension; endothelial nitric oxide synthase


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

NITRIC OXIDE (NO) produced by the endothelial isoform of NO synthase (eNOS) in endothelial cells is a key regulator of blood vessel tone, as exemplified by the hypertensive phenotype of mice deficient for the eNOS gene (29). In addition, accumulating evidence suggests a role for eNOS in regulating cardiac function independently from changes in peripheral or coronary vasodilatation, e.g., myocardial NO directly regulates beta -adrenergic inotropic responsiveness (23, 31), diastolic relaxation (22), and oxygen consumption (45). Indeed, eNOS was found to be expressed in microvascular endothelial cells lining intramyocardial capillary vessels and in cardiomyocytes themselves from several animal species and in humans (for a review, see Ref. 3), thereby supporting a paracrine or autocrine regulatory role for intramyocardial NO. Insights into NO's physiological regulation of cardiac contraction stemmed mainly from observations of the effect of NO donors or NOS inhibitors in several cardiac preparations and, more recently, of the cardiac phenotype of eNOS genetically deficient mice (23). These two approaches only partly reconciled previously conflicting observations on the effect of NO on cardiac contractility, which may be due to critical differences in the species and experimental preparations. In addition, NO's effects may be altered in specific cardiomyopathies, either because of changes in cardiomyocyte sensitivity to NO or as a result of expressional changes in the isoform(s) of NOS in the myocardium. For example, the decrease in contractility and drop in intracellular pH observed in response to sodium nitroprusside, an NO donor, in normal rat cardiomyocytes is lost in hypertrophic myocytes isolated from rats with aortic banding (30). In addition, previous studies have described changes in the expression of myocardial eNOS with the development of heart failure from various etiologies (for a review, see Ref. 4). Only a few studies have specifically examined expressional changes of eNOS in nonfailing, moderately hypertrophic cardiomyopathy. In a model of genetically hypertensive rats, Bayraktutan et al. (7) found no change in eNOS abundance in endothelial cells but a downregulation of this protein in cardiomyocytes that contrasted, however, with increased total eNOS mRNA. The same group later found no change in eNOS protein abundance in hypertrophic hearts from aortic-banded rats (34) despite an impairment of endothelial NO-dependent diastolic relaxation.

Aside from absolute changes in eNOS abundance, NOS activity can be profoundly influenced by the availability of substrates or cofactors such as L-arginine and tetrahydrobiopterin (THB4) or by other posttranslational regulators such as the chaperone heat shock protein (HSP)90 (17). Changes in the abundance of these regulators with the development of cardiomyopathies, however, are unknown. This issue is of potential importance, because it could explain alterations in the physiological regulation of cardiac function by myocardial NO unexpected from unchanged abundance of eNOS (or other NOS isoforms). Several groups have shown that the caveolar coat protein caveolin binds and inhibits eNOS in a process that could be competitively reversed by calcium-activated calmodulin (15, 18, 21). Importantly, increased flow and vascular pressure in situ can reduce this inhibitory binding of caveolin concomitant with increased eNOS-calmodulin association and increased eNOS activity, including in acute settings (e.g., in seconds) (41). Accordingly, we previously showed that a two- to threefold increase in caveolin-1 abundance was sufficient to inhibit NO production by eNOS in intact endothelial cells (13). A similar interaction between eNOS and caveolin-3 was also shown to regulate NO-dependent contractile function in intact cardiomyocytes (14). Therefore, in the present study, we examined concurrent changes in the abundance of eNOS and caveolin-1 and -3, the isoforms expressed mainly in endothelial cells and cardiomyocytes, respectively, in a dog model of moderate, nonfailing, hypertrophic cardiomyopathy. We confronted these changes with the vascular reactivity ex vivo and sensitivity of these dogs to pharmacological NOS inhibition in vivo after infusion of a NOS inhibitor. We show that hypertensive dogs exhibit profound hemodynamic responses to NOS inhibition accompanied by a concurrent downregulation of the abundance of eNOS and both caveolin isoforms in the hypertrophic myocardium.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Experimental Design

Twenty-three mongrel dogs (21-33 kg) were separated into three different experimental groups. Hypertension was produced in 14 animals by wrapping the left kidney with silk tissue using protocols as described previously (27). In 9 dogs, four consecutive hemodynamic and echocardiographic measurements were performed 1) in the control, normotensive state, 2) control state after 48 h of NG-nitro-L-arginine methyl ester (L-NAME), 3) hypertensive state at 7 wk after kidney wrapping [7-wk periephritic hypertension (PHT)], and 4) 7-wk PHT after 48 h of L-NAME. L-NAME solubilized in water was infused at a rate of 10 µg · kg-1 · min-1 with a minipump stowed solidly to the back and connected to a catheter implanted in the jugular vein. Another group of five dogs was studied at the early stage of PHT (3-wk PHT).

All hypertensive dogs were killed for tissue collection after completion of hemodynamic studies. A third group of nine normotensive dogs was killed, and tissues were collected for comparison with the hypertensive dogs.

Hemodynamic and Echocardiographic Measurements

Conscious animals were instrumented under local anesthesia with lidocaine (2%) while lying quietly on their left side. Hemodynamic parameters were acquired with a 7-F microtipped Millar catheter introduced via the left femoral artery into the left ventricle (LV) and a balloon-tipped 7-F thermodilution catheter (Edwards-Swan-Ganz) positioned in the pulmonary artery. LV echocardiography was performed with a Toshiba SSH 160 instrument equipped with a 3.5-MHz transducer. Acquisition and analysis of hemodynamic parameters were performed as previously described (43). In particular, LV relaxation rate was assessed by a time constant during the first 40 ms after the minimal first derivative of LV pressure (dP/dt) (42). The relaxation time constant (T1) was derived from the regression of dP/dt vs. LV pressure as follows: dP/dt = (-1/T1) × (P - PB), where P is LV pressure and PB is the variable pressure asymptote. This time constant was previously shown to be unaffected by changes in the LV systolic pressure or end-diastolic pressure (LVEDP), stroke volume, or velocity of shortening in dogs (16, 47).

Induction of Hypertension

Hypertension was induced in 14 animals by wrapping the left kidney with silk tissue (two kidneys, one wrapped hypertension). This model, a variant of cellophane wrapping initially described by Page et al. (39), is characterized by a progressive and persistent hypertension associated with perinephritis. General anesthesia was induced by 30 mg/kg pentobarbital intravenous injection followed by 1.5-3% enflurane. After surgery, no signs of infection were noted, and renal function remained within the normal range.

Measurement of Contractile Responses in Coronary and Mesenteric Macrovessels

Rings from the fourth branch of the superior mesenteric artery and proximal branches of the circumflex coronary artery were connected to an isometric lever connected to a force transducer (UC 2 Gould) and set at a tension equivalent to that generated at 0.9 times the diameter of the vessel at 100 mmHg, determined from the diameter-tension curve constructed as described (37). Concentration-dependent relaxation to acetylcholine (<= 1 µM) was measured in rings stably precontracted with 100 mM KCl.

Video-Motion Analysis of Microvessel Contraction

Microdissected coronary arteries branching from the left anterior descending (LAD) coronary artery varying in size from 70 to 170 µm of internal diameter and from 1 to 2 mm in length were cannulated with dual glass micropipettes (30-60 µm diameter) in a Plexiglas isolated organ chamber (Medical Instruments, University of Iowa) and pressurized to 40 mmHg in a no-flow state with a burette manometer filled with oxygenated physiological salt solution (PSS) at 37°C, as previously described (9). Care was taken to avoid any damage to the endothelium. Aerated PSS was continuously circulated externally through the organ chamber. The microvessel chamber was placed on an inverted fluorescence microscope [Axiovert S100, Zeiss, Germany, equipped with Zeiss Fluar ×20 (numerical aperture 0.75) objective lenses] connected to a charge-coupled device camera (Myocam, IONOPTIX, Milton, MA) connected through an interface (Fluorescence system interface, IONOPTIX) to data acquisition software (Ionwizard 4.4, IONOPTIX) allowing on-line monitoring of the external diameter. The viability of the vessel was assessed by stimulation with a high KCl-PSS (50 mM KCl). When the contractile response reached a steady state, carbachol or sodium nitroprusside was added in the perfusion solution.

Measurement of NOS Activity in Atrial and Ventricular Homogenates

NOS activity was quantified by measuring the conversion of L-[3H]arginine to L-[3H]citrulline in the presence of saturating concentrations of the cofactors of the enzyme with or without calcium, as previously described (5). Finely ground frozen atrial and ventricular heart samples obtained from normotensive and PHT dogs (3 and 7 wk after surgery) were suspended in 2 ml/g ice-cold buffer (50 mM Tris, pH 7.4, containing 0.1 mM EGTA, 0.1 mM EDTA, 2 mM beta -mercaptoethanol, and the protease inhibitors 5.25 µM leupeptin and 3.65 µM pepstatin). The suspension was homogenized in three passes of 10 s at 22,000 rpm with an Ultra-Turrax (Labortechnik, Staufen, Germany) and then sonicated on ice in three passes of 10 s at 10 W (Sonifer B-12, Branson Sonic Power, Danbury, CT). The homogenates were centrifuged at 6,000 g for 5 min at 4°C, and protein content was determined using the Bradford method (Bio-Rad) with albumin as a standard. The supernatant was diluted in this same ice-cold buffer containing the protease inhibitors and with 20 mM [3-(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate (CHAPS) to obtain ~600 µg of protein per sample. Twenty-five microliters of supernatant were added to 200 µl of buffer containing 50 mM Tris · HCl (pH 7.4), 10 mM dithiothreitol, 10 µM THB4, 10 µg/ml calmodulin, 1 mM NADPH, 4 µM flavin adenine dinucleotide, 4 µM flavin mononucleotide, 2 µM L-arginine, and 10-3 mCi/ml L-[3H]arginine and incubated for 30 min at 37°C in the presence (1 mM CaCl2) or absence of calcium (i.e., nominally calcium free, with 2 mM EGTA and 2 mM EDTA). Parallel reactions were analyzed in the absence of NADPH to determine background signal. Assays were performed in duplicate and were terminated with 2 ml of ice-cold "stop buffer" (20 mM CH3COONa, pH 5.5, containing 1 mM L-citrulline, 2 mM EDTA, and 0.2 mM EGTA). The total volume was applied to a Dowex AG-50 W-X8 column (Bio-Rad) that had been preequilibrated with 20 mM stop buffer (pH 5.5). L-[3H]citrulline was eluted with 2 ml of deionized water, and radioactivity was quantified by liquid scintillation counting. Total NOS and Ca-independent activity was determined by subtracting counts in the absence of NADPH from counts obtained in the presence or absence of calcium, respectively. Ca-dependent activity was calculated as total NOS activity minus Ca-independent activity, and the results are expressed in femtomoles of L-[3H]arginine converted per milligram of protein per minute.

L-citrulline assay reagents were obtained from Sigma (St. Louis, MO) except for THB4 (Dr. B. Schircks Laboratories, Jona, Switzerland), L-[3H]arginine (Amersham, Chalfont, England) and the liquid scintillation (Lumasafe Plus, Lumac, Groningen, The Netherlands).

Immunohistochemical Analysis of eNOS Staining

At the time of death, the beating heart was removed, and transverse sections (each ~5 mm thick) of the LV wall were performed at the base of the heart. Three sections were embedded in Tissue Tek OCT compound (Miles, Elkhart, IN), snap-frozen in precooled isopentane, and stored at -80°C. Five-micrometer-thick sections were obtained with a cryotome (MTC, Slee, Mainz, Germany) and then directly applied onto Superfrost Plus slides (Menzel-Gläser, Braunschwerg, Germany). Sections were dried and stored at -80°C.

Cryosections were first dried and then fixed in 3.5% formaldehyde [10 min, room temperature (RT)] and washed under tap water. Endogenous peroxidase activity was inhibited by 0.3% H2O2 in PBS (pH 7.4, 10 min, RT). Sections were rinsed under tap water and washed in PBS (10 min, RT). To avoid cross-reactivity of the second antibody, sections were incubated with normal goat serum (1:20 in PBS supplemented with 1% BSA, 30 min, RT). Sections were then incubated (1 h, RT) with a rabbit polyclonal anti-eNOS (raised against a peptide containing residues 599-613 of eNOS, no. 482726, Calbiochem) diluted 1:400 in PBS supplemented with 1% BSA and rinsed with PBS-0.1% BSA (3 × 5 min). Fixed rabbit antibodies were detected with the Envision peroxidase system (Envision + peroxidase rabbit, K4002, Dako, Carpinteria, CA) according to the manufacturer's instructions. Peroxidase activity was revealed (5 min in the dark) by a commercial solution of 3-amino-9-ethylcarbazol (AEC substrate system K0697, Dako). After being washed under tap water, sections were counterstained with Mayer's hematoxylin, rinsed in water (5 min), and mounted with an aqueous mounting medium (Faramount S3025, Dako).

Western Blotting Experiments

Extracts were prepared from whole LV or selectively from epicardium, midmyocardium, and subendocardium as well as from the dog aorta, lung, brain, and kidney. Denatured proteins were size fractionated by SDS-PAGE (8 or 12.5% gels) and transferred to polyvinylidene difluoride membranes (PVDF; NEN) as previously described (13). Naphtol blue-black staining was used to determine equal protein loading. The membranes were blocked overnight in Tris-buffered saline with Tween 20 (0.1%) (TBS-T) containing either 5% milk proteins or a mixture of 3% (50:50) milk proteins and bovine albumin, incubated for 2 h with the primary antibody, i.e., monoclonal anti-caveolin-1, anti-caveolin-3, anti-HSP90 (raised against a COOH-terminal peptide comprising amino acids 586-732 of human HSP90), anti-iNOS, anti-nNOS antibodies (all from Transduction Laboratories), or rabbit polyclonal anti-eNOS antibodies (Affinity Bioreagents), raised against a synthetic peptide (amino acids 559-613) of bovine eNOS. For eNOS immunodetection, numerous attempts with other reagents, [i.e., 2 different mouse monoclonal (from Biomol and Transduction Laboratories) or a rabbit polyclonal antibody raised against a C-terminal peptide (amino acids 1179-1194) of human eNOS (from Affinity Bioreagents)] all failed to identify specific and/or quantifiable immunoblotting signals of a size compatible with the canonical (135 kDa) eNOS, even after affinity purification of whole dog tissue extracts with ADP-Sepharose. Membranes were then washed four times (15 min each) in TBS-T with either 1% milk proteins (with monoclonal antibodies) or 1% albumin (with the polyclonal antibody) and then processed for autoradiography. The density of the bands was quantified by scanning densitometry and expressed as a percentage of the mean density measured on the same blot from extracts of the control group.

Immunoprecipitation and Immunoblotting

Coimmunoprecipitation analysis to study the interaction of caveolin-1 with eNOS was performed on whole LV extracts. Briefly, LV tissue was homogenized in lysis buffer [50 mM Tris · HCl, pH 7.5, 0.1 mM EGTA, 0.1 mM EDTA, protease and phosphatase inhibitor cocktail (Sigma), 0.5% (vol/vol) Igepal (Sigma), 0.1% sodium dodecyl sulfate, 0.1% deoxycholate, and 20 mM sodium molybdate] and extracts were sonicated on ice three times for 5 s and centrifuged at 12,000 rpm for 10 min. The protein concentration was determined using the bicinchoninic acid assay (Pierce). The supernatant (500 µg protein) was incubated for 1 h with excess anti-caveolin-1 polyclonal antibodies (Transduction Laboratories). Immune complexes were precipitated for an additional 1 h by the addition of protein G-Sepharose (Zymed Laboratories), centrifuged, washed three times, and then boiled in Laemli loading buffer. Both the immunoprecipitate and supernatant fractions were separated by discontinuous SDS-PAGE (8 and 12.5% gels) and electroblotted onto PVDF membranes for 180 min. The membranes were blocked in a 3% milk and BSA mixture and incubated with either anti-caveolin-1 monoclonal antibodies (Transduction) or polyclonal anti-bovine eNOS antibodies (Affinity Bioreagents).

Statistics

Data are expressed as means ± SE. LV function parameters were compared using one-way ANOVA for repeated measures. All other data were analyzed by one-way ANOVA followed by the Bonferroni correction for multiple comparisons or unpaired Student's t-test where appropriate. P value of less than 0.05 was considered significant.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

NOS activity and Protein Expression in Dog Hearts

Ca-dependent and Ca-independent NOS activity was measured in whole muscle extracts from all four cavities of normotensive and 3-wk and 7-wk PHT dogs, as illustrated in Fig. 1A. A background Ca-independent activity was detected in control dog hearts but was unchanged in either 3- or 7-wk PHT dogs [LV: 27.1 ± 2.9 vs. 28.2 ± 4.3 fmol · min-1 · mg-1 in control and 7-wk PHT dogs, respectively; P = not significant (NS)]. However, no signal for inducible NOS (iNOS) was detectable in whole extracts from the same hearts by immunoblotting (compared with strong positive control, Fig. 1B, bottom). In all dogs, Ca-sensitive NOS activity tended to be higher in atrial compared with ventricular tissues. In normotensive dogs, this difference was significant (P < 0.01) between the right cavities only but reached significance (P < 0.05 and P < 0.001, respectively) in both sides of the heart in 3-wk PHT dog tissues. By contrast, Ca-dependent NOS strikingly decreased in all cavities from 7-wk PHT dogs, with a near abolition of the signal in ventricular cavities (Fig. 1A, right). No signal for neuronal NOS (nNOS) was detected in extracts of LV from the same hearts (compared with positive control in the dog brain, Fig. 1B, top), nor was any eNOS-specific signal detectable by Western blot using two different mouse monoclonal antibodies or a rabbit polyclonal antibody against a COOH-terminal peptide from human eNOS (data not shown). When LV extracts were probed with rabbit polyclonal antibodies raised against a bovine eNOS peptide, we detected only a 150-kDa protein but no band corresponding to the canonical eNOS of ~135 kDa, although proteins of both sizes were clearly detected in control extracts from eNOS-expressing bovine aortic endothelial cells (Fig. 1B, middle). The abundance of this larger, 150-kDa eNOS did not vary with the development of hypertension (100 ± 5.1% in normotensive vs. 125 ± 17.6% in 7-wk PHT dogs; P = NS; n = 5-6 hearts).


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 1.   Nitric oxide synthase (NOS) activity and isoforms in normotensive (NT) and hypertensive (HT) dog hearts. A: calcium-dependent NOS is selectively decreased in 7-wk HT dogs. NOS activity, reflected as the rate of conversion of L-[3H]arginine to L-[3H]citrulline, is expressed in femtomoles of L-[3H]arginine converted per milligram of protein per minute. Calcium-independent (filled portion of bars) and calcium-dependent activity (open portion) was measured in extracts of the 4 heart chambers from NT (n = 6), 3-wk HT (n = 5) and 7-wk HT (n = 5) dogs. Data are means ± SE of samples from 5-6 experimental animals. #P < 0.05, ##P < 0.01, and ###P < 0.001, calcium-dependent activity in atria vs. same-side ventricle; *P < 0.05 and **P < 0.001, 7-wk HT vs. NT. B: immunoblotting (IB) of NOS isoforms in the dog left ventricle (LV). Western blotting signals for the three isoforms of NOS in LV extracts from NT and 7-wk HT dogs compared with control tissues; top, neuronal NOS (nNOS) detection in dog brain [control (C)] but not in dog LV; middle, 150- and 135-kDa immunoblotted signal for endothelial NOS (eNOS) in extracts of bovine aortic endothelial cells (C) compared with single 150-kDa band in dog LV; bottom, inducible NOS (iNOS) detection in extracts of activated rat macrophages (C) but not in dog LV.

Immunohistochemical Staining for eNOS in Dog Hearts

Figure 2 illustrates the results for eNOS immunodetection with polyclonal anti-eNOS-specific antibodies as well as the negative controls with nonrelevant IgG (Fig. 2, B, D, and F) in sections of LV from normotensive (Fig. 2, A and B), and 3-wk (C and D) and 7-wk PHT dog hearts (E and F). No specific signal was detected above background with normal rabbit serum in any experimental group. The specificity of the immunostaining with the primary antibody was ascertained in at least two samples from each experimental group by competition with the specific immunogenic peptide, which resulted in a complete abrogation of the signals. With the specific antibody, positive staining was observed in both cardiomyocytes and endothelial cells in normotensive hearts. The latter seemed to be more intense in microvascular endothelial cells from 3-wk PHT dogs. By contrast, a striking reduction in immunostaining was seen in endothelial cells from 7-wk PHT hearts. When the number of positively stained endothelial cells was computed from randomly chosen microscopic fields and normalized to the total number of optically identifiable endothelial cells in the same areas, we observed a significant reduction in eNOS-positive cells in 7-wk PHT hearts compared with samples from normotensive animals (from 43.8 ± 10.7 to 4.6 ± 1.3%, respectively; P < 0.05; n = 6-7 hearts).


View larger version (161K):
[in this window]
[in a new window]
  		Fig. 2.   eNOS immunostaining in NT and HT dogs. A, C, and E: positive staining with anti-eNOS-specific antibody; B, D, and F: negative controls with nonrelevant IgG. A and B: NT dogs; C and D: 3-wk HT dogs; long arrows, endothelial cells; short arrows, cardiomyocytes. E and F: 7-wk HT dogs.

Expression of NOS Isoforms in Other Dog Tissues

To assess the specificity of the expressional regulation of eNOS with hypertension, we analyzed the variations of the abundance of other isoforms in several dog tissues. iNOS was undetectable in dog tissues, as mentioned earlier. We observed no significant difference between the expression of nNOS in brains of normotensive and 7-wk PHT dogs (100 ± 15.6 vs. 82.5 ± 19.1%, respectively; P = NS; n = 3-5 brains in each group). nNOS was undetectable in whole dog kidneys and aortic tissue. However, we observed a 10-fold elevation in nNOS protein levels in lungs from 7-wk PHT dogs compared with control animals (Fig. 3, P < 0.0001; n = 4-6 lungs).


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3.   Differential regulation of nNOS in lung tissue from HT dogs. Representative Western blotting (top) and densitometric data (bottom) of immunoblotting of nNOS in lung extracts from NT and 7-wk HT dogs are shown. Contrary to eNOS in the heart, nNOS is upregulated in lungs of HT animals. ***P < 0.0001, n = 4-6 dogs.

Changes in Caveolin-1/3 and HSP90 Abundance in Dog Hearts

Because the activity of eNOS was shown to be regulated posttranslationally through its inhibitory interaction with the caveolar coat protein caveolin and stimulatory interaction with the chaperone protein HSP90, we assessed potential changes in the abundance of the caveolin-1 and -3 isoforms expressed in endothelial cells and cardiomyocytes, respectively, and of HSP90 by Western blot analysis. We analyzed samples dissected from LV subendocardium, midmyocardium, or epicardium from normotensive and 7-wk PHT dogs and observed a significant reduction of both caveolin isoforms in PHT dog hearts that was most prominent in the subendocardium (Fig. 4, A and B; P < 0.001 for caveolin-1; P < 0.05 for caveolin-3; n = 4-5 hearts).


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4.   Downregulation of caveolin (Cav)-1 and -3, but not heat shock protein (HSP)90, in HT dogs. Representative Western blots (insets) and densitometric data reflecting caveolin-1 (A) and caveolin-3 (B) protein abundance in LV subendocardium (Endo), midmyocardium (Meso), and epicardium (Epi) from NT (n = 5) and 7-wk HT (n = 4) dogs are shown. In addition, some filters were reprobed with anti-HSP90 antibodies, as illustrated in the lower parts of the insets (representative of 3 experiments with similar results). *P < 0.05, **P < 0.005, and ***P < 0.001 vs. same layer in NT animals.

Conversely, the abundance of HSP90, measured on the same blots, did not change with the development of hypertension in the midmyocardium or epicardium (P = NS) but even augmented in the subendocardium (214.7 ± 21.6% in 7-wk PHT dogs vs. 100 ± 22.3% in normotensive dogs; P < 0.05; n = 3 hearts in each group).

Caveolin Expression in Other Dog Tissues

To determine whether the decrease of caveolin is specific to heart tissue or generalized in others tissues, we assessed changes in the expression of caveolin in several other organs of normotensive and 7-wk PHT dogs by Western blot analysis. In aortic segments, protein expression of caveolin-1 was not significantly different from control and 7-wk PHT dogs (100 ± 6.6 and 106.1 ± 6.0%, P = NS). In right kidneys (unaffected by perinephritic inflammation in operated dogs), caveolin-1 protein expression was significantly reduced in 7-wk-PHT dogs (26 ± 15.5 vs. 100 ± 27.5% in control dogs; P < 0.05; n = 5 kidneys for each group); caveolin-3 was undetectable in dog kidneys. In contrast, there was a marked increase in caveolin-1 in lung tissue of 7-wk PHT dogs compared with normotensive dogs (201.3 ± 41.3 vs. 99.4 ± 20.3%; P < 0.05; n = 5-6 lungs).

Immunoprecipitation Assay for eNOS and Caveolin-1 Interaction in Dog Hearts

To assess the potential for the association of eNOS with the inhibitory protein caveolin-1 in dog hearts in situ, eNOS was immunoprecipitated with anti-caveolin-1 antibodies from whole LV extracts of normotensive and 7-wk PHT dogs. Western blot analysis of the immunoprecipitates is shown in Fig. 5. The recovery of caveolin-1 in the immunoprecipitate was nearly quantitative (estimated as ~95%; see bottom bands). Immunoblotting with polyclonal anti-eNOS antibodies revealed one band at 150 kDa, corresponding to a similar, larger eNOS protein in extracts of bovine aortic endothelial cells (positive control, lane 5); again, no signal for the canonical 135-kDa eNOS protein was immunoprecipitated in dog LV tissue. The immunoprecipitation was specific, as assessed from the absence of any signal in the absence of immunoprecipitating antibody (Fig. 5, A and B, lane 2 from left). The amount of the 150-kDa eNOS protein associated with caveolin (determined as the ratio of the amount of eNOS detected in the caveolin immunoprecipitate divided by the sum of eNOS in this immunoprecipitate and the remaining supernatant, after correction for dilution) was estimated at 21 ± 1.4% (n = 3); in two similar experiments using LV tissue from 7-wk PHT dogs, the proportion of 150-kDa eNOS associated with caveolin amounted to 16 and 12%.


View larger version (48K):
[in this window]
[in a new window]
  		Fig. 5.   Coimmunoprecipitation (Co-IP) of eNOS and caveolin-1 from whole LV extracts of NT and HT dogs. Whole dog LV extracts were processed for immunoprecipitation with polyclonal anti-caveolin-1 antibodies, as described in METHODS. The immunoprecipitates (IP) and supernatant fractions (S) were resolved on discontinuous SDS-PAGE (8 and 12.5% gels) and eNOS (top) and caveolin-1 (bottom) revealed by immunoblotting with respective specific antibodies. Control IP and S fractions processed in the absence of immunoprecipitating antibody (Ab) were run in parallel (lanes 2 and 4 from left), as well as extracts of bovine aortic (BAEC) and human umbilical vein endothelial cells (HUVEC) as positive controls (rightmost 2 lanes). A: Co-IP of eNOS and caveolin-1 from NT dog LV (LV NT). B: Co-IP of eNOS and caveolin-1 from 7-wk HT dog LV (LV HT 7 wk).

Endothelium-Dependent Relaxation Of Micro- And Macrovessels

Mesenteric and epicardial coronary arteries. Maximum contraction in a 100 mM KCl depolarizing solution was not significantly different between vessels from normotensive, 3-wk, or 7-wk PHT dogs. Relaxation of preconstricted vessels to muscarinic cholinergic agonists amounted to ~25-30% of the maximum contractile tone and was similar in arteries from normotensive and 3-wk PHT dogs. However, the relaxation was significantly reduced in arteries from 7-wk PHT dogs (P < 0.05, Fig. 6).


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 6.   Endothelium-dependent relaxation of micro- and macrovessels ex vivo. Maximum relaxations to acetylcholine (macrovessels) or carbachol (microvessels), expressed as a percentage of the plateau of contraction in KCl-depolarizing solution, is compared in vessels from NT (n = 5) and 3- and 7-wk HT (n = 3-5) dogs. *P < 0.05 compared with NT.

Coronary microvessels. Coronary microarteries dissected from the LV of the three dog groups exhibited similar internal (141 ± 7, 132 ± 14, and 148 ± 11 µm; n = 5) and external (212 ± 6, 196 ± 7, and 196 ± 12 µm; n = 5) diameters when perfused at the same intraluminal pressure, suggesting the absence of a significant hypertrophy of the media. Accordingly, their maximum contraction to 50 mM KCl was not significantly different between the three groups. Arteries from 3- and 7-wk PHT dogs exhibited a significant reduction of their relaxation to 10 µM carbachol compared with vessels from normotensive dogs (P < 0.05; Fig. 6). This impaired maximal relaxation occurred despite unchanged potency of the muscarinic cholinergic agonist, since -log IC50 determined from dose-response curves amounted to 6.7 ± 0.27, 5.7 ± 0.36, and 5.8 ± 0.87 for normotensive, 3-wk, and 7-wk PHT dogs, respectively (n = 5 for each group; P = NS). Moreover, relaxation to sodium nitroprusside, an endothelium-independent vasodilator, was similar between the three groups of vessels (n = 3-5; P > 0.05).

Parameters of LV Function

The hemodynamic and echographic data from normotensive and PHT dogs before and after 48-h L-NAME infusion are summarized in Table 1 and Fig. 7.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Hemodynamic and echographic data of dogs studied in the normotensive and hypertensive states



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 7.   Echocardiographic assessment of LV function in NT and HT dogs. Contractility was assessed with a load-independent plot of corrected mean velocity of circumferential fiber shortening (VCF) against end-systolic wall stress (ESWS) in NT and 7-wk HT dogs with or without 48 h NG-nitro-L-arginine methyl ester (L-NAME) infusion. Systolic function was impaired only in HT dogs treated with L-NAME.

Effect of PHT on LV Parameters

After 7 wk, kidney wrapping resulted in the development of hypertension, as reflected by increases in mean aortic pressure (MAoP) by 23 ± 8% (P < 0.05). Of interest, these pressure increases occurred despite unchanged systemic vascular resistance (SVR). LV mass was significantly increased at 7 wk in the PHT dogs compared with control, normotensive dogs (P < 0.05), as expected from the previous characterization of this hypertension model. Similarly, LV systolic pressure (LVSP) was augmented by 30 ± 10% (P < 0.05), as well as indexes of LV systolic function [peak +dP/dt, first derivative of LV pressure calculated at a developed LV pressure of 40 mmHg (DP40), and fractional shortening (FS)]. Because of increases in indexes of wall thickness [anterior (AWD) and posterior end-diastolic wall thickness (PWD); P < 0.05] secondary to LV hypertrophy, end-systolic wall stress (ESWS) remained unchanged.

Because the first derivative of the LV pressure pulse is dependent on arterial pressure (20), blood pressure increases may confound the interpretation of peak dP/dt in PHT dogs. Therefore, contractility was further assessed with a load-independent plot of corrected mean velocity of circumferential fiber shortening (VCF) against ESWS, as measured echocardiographically and illustrated in Fig. 7. The results were consistent with an unchanged inotropic state. However, the relatively load-insensitive LV relaxation time constant, T1, was significantly prolonged, suggesting an impairment of LV diastolic relaxation (Table 1).

Effect of 48-h L-NAME Infusion in Normotensive Dogs

To gain insight into the influence of NOS activity on hemodynamic parameters, nine dogs were treated with peripheral infusions of the NOS inhibitor L-NAME in the normotensive and hypertensive states. Infusion of the NOS inhibitor resulted in a 80 ± 13% increase in SVR (P < 0.005), reflecting a marked peripheral vasoconstriction. This resulted in significant increases in both LVSP (P < 0.05) and MAoP (P < 0.05), a decrease in cardiac output (CO; -32 ± 6%; P < 0.05) and, to a lesser extent, of heart rate (HR; P = NS). Indexes of inotropic state [peak +dP/dt, DP40, and FS] remained unchanged. Concomitantly with higher ventricular and aortic pressures, a slight but significant decrease of myocardial wall thickness was observed [-4 ± 1% in AWD (P < 0.05)], which resulted in an increased ESWS (P < 0.05). Consistently, load-insensitive assessment of contractility by echocardiography in normotensive dogs treated with L-NAME yielded results within the normal range as observed in control, untreated dogs (Fig. 7). There was a trend toward an impairment of LV relaxation (increase in T1, Table 1), albeit not significant.

Effect of 48-h L-NAME Infusion in Hypertensive Dogs

As in the normotensive dogs, infusion of L-NAME for 48 h resulted in increases in SVR (+113 ± 18%, P < 0.005), LVSP, and MAoP, with a drop in CO (-43 ± 6%, P < 0.005) and HR. Contrary to the effect of L-NAME observed in normotensive dogs, indexes of LV systolic function (peak +dP/dt, DP40, and FS) all decreased (P < 0.05), indicating a depressed inotropic state. Similarly, AWD decreased (P < 0.05), resulting in enhanced ESWS (P < 0.005). Echocardiographic assessment of contractility confirmed an impairment of contractility after 48-h L-NAME in PHT dogs (Fig. 7). Likewise, 48-h inhibition of NOS in PHT dogs resulted in a significant prolongation of the LV relaxation time constant (T1), indicating an impairment of diastolic relaxation (Table 1).


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The main finding of our study is that the expression of eNOS and caveolin in several cell types within cardiac muscle varies with the development of hypertension in the canine PHT model and that resultant changes in NOS activity may participate in the regulation of systolic and diastolic function in the hypertrophic heart.

Changes in NOS Expression with Hypertension

Our analysis of NOS activity in whole muscle extracts with saturating concentrations of cofactors and substrates reflects the abundance of the enzyme(s) in a mixture of different cell types. Given the poor expression of the other calcium-sensitive, constitutively expressed nNOS within the myocardium relative to eNOS (46), as confirmed from our Western blot analysis (Fig. 1B, top), the calcium-dependent signal in our experiments can reasonnably be attributed mainly to eNOS. In addition, an apparently calcium-insensitive enzymatic activity was observed in all extracts. Of note, contrary to the calcium-sensitive signal, this background activity was unchanged despite surgery in dogs that developed hypertension. Therefore, it is unlikely due to the expression of iNOS, which remained undetectable by Western analysis (Fig. 1B, bottom). Surprisingly, the only immunoblotted NOS in dog heart was a larger protein of 150 kDa that coincided with proteins of similar size and immunoaffinity for anti-eNOS antibodies in extracts of bovine aortic endothelial cells. This larger 150-kDa eNOS had previously been detected in extracts of dog hearts in a study by Khadour et al. (33) and in adult rat cardiomyocytes (8). The latter study had identified this larger isoform as a nonpalmitoylated eNOS associated mainly with cytosolic and intracellular enriched membrane fractions and subsequently processed to the palmitoylated and plasma membrane-bound canonical (i.e., of 135 kDa) eNOS. Of interest, this preprocessed eNOS also exhibited calcium-dependent activity in conditions where lysates had been processed in the presence of high detergent concentrations (1% Triton X-100). Given the use of less detergent conditions in our assays, it is possible that the apparent calcium-insensitive signal in our study corresponds to enzymatic activity supported by this larger 150-kDa eNOS sequestrated in intracellular organelles. Our observation of unchanged calcium-insensitive activity, paralleled by unchanged immunoblotted signals at 150 kDa in hypertensive compared with normotensive dogs, would support this interpretation. Therefore, the remaining calcium-sensitive signal likely represents the "mature" eNOS that produces NO at the cell membrane. The inability of all available anti-eNOS antibodies to reveal specific and reliable immunoblotting signals for the 135-kDa eNOS in dog tissues, in our hands, precludes any direct measurement of subcellular distribution of the two "isoforms," e.g., by cellular fractionation. Likewise, we could not directly assess changes in the abundance of the 135-kDa eNOS by Western analysis in hypertensive dogs. However, a clear immunostaining for eNOS was observed in both endothelial cells and cardiomyocytes (Fig. 2). Interestingly, the development of hypertension was associated with a decrease in immunostaining that was more prominent in endothelial cells, whereas a residual immunostaining was still observed in cardiomyocytes (Fig. 2E), where the 150-kDa eNOS was first described (8). To overcome the difficulty in quantitating the data with this approach, we devised a protocol of analysis where the number of positively stained endothelial cells (as a fraction of total identifiable endothelial cells in the same microscopic field) were compared between PHT dogs and controls. Using this binary type of analysis, a clear difference was observed between the two groups at 7 wk of hypertension, which is entirely consistent with the data on calcium-sensitive NOS activity. This further supports our interpretation that the decrease in calcium-dependent activity reflects the decrease in endothelial eNOS.

In animal models of heart failure, as well as in patients, eNOS abundance was found to either increase or decrease in the LV. In a model of spontaneously hypertensive rats, eNOS was shown to be specifically downregulated in cardiomyocytes (7). In the chronically paced dog model, however, Khadour et al. (33) observed an increase in Ca-dependent NOS and eNOS protein specifically in left atria. Accordingly, when we examined levels of NOS activity across the four heart cavities, we found consistently higher Ca-dependent NOS activity in the atria. In our model, however, the abundance of calcium-sensitive activity varied with time, with NOS activity being unchanged (or even slightly increased) at 3 wk of hypertension, but strikingly decreased in all parts of the heart at 7 wk. An important difference between our and previous studies, however, is that hypertensive dogs in the present study do not develop overt LV failure. This may explain some of the difference between our results regarding eNOS expression and those of other studies of decompensated heart failure in the same and other species (for a review, see Ref. 4). Finally, our assessment of nNOS expression in other tissues of normotensive and hypertensive dogs (Fig. 3) showed that these changes in expressional regulation seem to be isoform and tissue specific.

In addition to changes in eNOS abundance, the production of NO in situ can be influenced through posttranslational regulation of eNOS activity. Previous studies have demonstrated that eNOS activity is influenced by the reciprocal interaction of the enzyme with either activated Ca-calmodulin or caveolin, the structural protein of plasmalemmal caveolae. The ability of caveolin-1 or -3 to form a heteroduplex with eNOS supports its ability to inhibit the enzyme's activity (35) that is proportional to caveolin's abundance. In addition, increased flow and vascular pressure in situ can reduce this interaction of caveolin with eNOS concomitant with increased eNOS-calmodulin interaction and increased eNOS activity over very short time periods (seconds) (41). Recently, we showed that, at a fixed level of eNOS, even relatively small increases in caveolin-1 expression could result in substantial inhibition of NO production in intact endothelial cells (13). Conversely, decreases in caveolin-1 potentiated basal and agonist-stimulated eNOS activity (12). Accordingly, we examined whether the development of hypertension was accompanied by changes in caveolin abundance that could alter eNOS's ability to produce NO. Measurements of protein abundance of caveolin-1 and caveolin-3, the isoforms predominantly expressed in endothelial cells and myocytes, respectively, showed a significant reduction at 7 wk of hypertension that was most marked in the subendocardium (Fig. 4). Importantly, this contrasted with unchanged levels of HSP90, a chaperone protein known to promote eNOS activation, which even increased in the subendocardium (see RESULTS and Fig. 4). A similar downregulation of caveolin-1 and -3 has been observed in mice with moderate cardiac hypertrophy induced by chronic infusion of isoproterenol (38). Of note, these changes in caveolin abundance cannot be correlated directly with Ca-dependent NOS activity, as measured in our in vitro assays, because measurements of enzymatic activity are performed in a buffer containing excess calcium and calmodulin sufficient to dissociate the eNOS/caveolin heteroduplex, i.e., the results mostly reflect the maximal velocity proportional to eNOS abundance. This does not preclude alterations of eNOS activity by decreased caveolin in situ, which would be expected to result in enhanced NO production by the residual enzyme. That the two proteins interact in dog LV tissue was demonstrated by coimmunoprecipitation of eNOS with anti-caveolin antibodies. Again, our antibody detected only the 150-kDa eNOS (8). The decrease in total caveolin abundance was also paralleled by decreased interaction with eNOS, as measured by the fraction of eNOS coimmunoprecipitated with caveolin in extracts of hypertensive dog LV (Fig. 5). In addition, the interaction of eNOS with caveolin was previously shown to be reduced acutely by increases in vascular pressure, concomitant with increased eNOS activity (41). Therefore, both acute and chronic effects of high blood pressure on caveolin-eNOS interaction, as well as eNOS regulation by HSP90, may account, in part, for the functional results with NOS inhibitors (see below). Finally, our analysis of caveolin-1 abundance in other tissues than the heart in normotensive and 7-wk PHT dogs showed the tissue-specific regulation of its expression.

NO-Dependent Endothelial Function

Despite reduced eNOS abundance, 7-wk PHT dogs had SVR comparable to control levels, suggesting a preserved endothelial control of vascular tone, at least at baseline (i.e., in the absence of agonist stimulation) (Table 1). Consistent with this observation, the maximum contraction of vascular rings or microcoronary arteries to KCl was unchanged from controls vessels (see RESULTS). Although endothelial factors such as hyperpolarizing factor(s) may compensate for decreased eNOS abundance in hypertensive vessels (for a review, see Ref. 36), this is unlikely the case in our contractility experiments using high KCl solutions (where endothelium-dependent hyperpolarization factor(s) is/are inoperable). Moreover, the involvement of NO was more directly assessed by in vivo infusions of L-NAME. Surprisingly, as in normotensive animals, the NOS inhibitor produced significant increases in MAoP (and SVR) in PHT dogs. This suggests a significant control of basal vascular tone by NO despite decreased eNOS abundance. One explanation may be that the parallel decrease in caveolin-1 (but not of HSP90) results in a potentiation of basal eNOS activity. The sensitivity of PHT dogs to NOS inhibition also argues against a significant role of oxidant stress (and decreased bioavailability of NO) in our preparations, as observed in other models of endothelial dysfunction associated with hypertension (25).

Under agonist stimulation, however, both macro- and microvessels exhibited a deficit in maximal relaxation at 7 wk of hypertension (Fig. 6). This likely results from the decrease in maximally activatable NOS activity that parallels the decrease in endothelial eNOS expression (Fig. 2) and our measurements of calcium-activated NOS in vitro (Fig. 1A). We cannot exclude, however, that some of the impairment of relaxation may also be due to alterations in receptor coupling, as commonly observed in other pathophysiological states with endothelial dysfunction (25). Likewise, the alteration of carbachol-induced relaxation of coronary microvessels despite unchanged endothelial eNOS staining at 3 wk of hypertension suggests that some of these alternative mechanisms (including increased production of oxidant radicals) may operate instead of (or in addition to) absolute changes in eNOS abundance.

eNOS and LV Systolic and Diastolic Function

The aforementioned changes in vascular reactivity were accompanied by alterations in several parameters of LV function, i.e., baseline LV relaxation was altered in 7-wk PHT dogs despite preserved systolic function, as observed in the hypertrophic LV (28). Likewise, in normotensive animals, NOS inhibition did not alter systolic function, as assessed by load-independent echocardiographic parameters (Fig. 7) and tended to impair relaxation. In hypertensive dogs, L-NAME induced more prominent degradations in systolic and diastolic function. Again, these effects contrasted with decreased eNOS expression in PHT dogs, which would be anticipated to decrease their sensitivity to NOS inhibition but were paralleled by decreased expression of caveolin-1 and -3 (but not of HSP90) that may potentiate residual eNOS activity (see above).

Recently, several groups provided evidence for a physiological role of NO as a direct modulator of cardiac systolic and diastolic parameters, independently of its regulation of coronary vasodilating capacity. Specifically, NO produced by eNOS within cardiomyocytes from rodent species and humans was shown to regulate the inotropic response to catecholamines (for a review, see Ref. 2) and also to directly enhance diastolic relaxation (for a review, see Ref. 40), perhaps through decreasing the sensitivity of contractile myofilaments to calcium. This relaxation-promoting effect of endogenous NO was subsequently reproduced in isolated hearts (22) and in patients (6) independently of changes in coronary perfusion. Likewise, loss of total eNOS in PHT dogs at 7 wk may have contributed directly to the impairment of diastolic parameters. The effect of NO on basal systolic function is more controversial, with some studies reporting a small positive inotropic effect that may vary according to the experimental model studied (for a review, see Ref. 32). Of interest, a recent study showed that, although the effect of NOS inhibitors on beta -adrenergic responsiveness was potentiated in dogs with pacing-induced heart failure, the sensitivity of basal (i.e., in the absence of catecholamines) LV function to NOS inhibition was depressed, despite unchanged total eNOS abundance (24). Unlike our observation in nonfailing hypertrophic cardiomyopathy, this was correlated with increased abundance of caveolin-3 in the failing myocardium, which, in agreement with our hypothesis, would decrease basal NOS activity and the effect of NOS inhibition. Conversely, our demonstration of decreased caveolin-3 in the hypertrophic, nonfailing ventricle provides one explanation for the persistent effect of NOS inhibitors despite eNOS downregulation. Therefore, our study emphasizes the differential regulation of caveolin expression according to the type of cardiomyopathy. Because caveolin-3 was shown to be downregulated by chronic beta -adrenergic receptor stimulation (38), the differential regulation may be related to changes in intracellular cAMP content secondary to alterations in beta -adrenergic responsiveness in hypertrophic vs. failing myocardium.

Potential Limitations of the Experimental Model

Causes other than altered production of NO may explain the observed changes in LV function in the hypertrophic heart, such as structural remodeling. In this short-term hypertension model, however, it has been previously reported that myocardial fiber diameter increases by ~35% but that fibrosis is not significantly increased up to 12 wk of hypertension, suggesting that interstitial remodeling and subsequent increased stiffness does not play a major role in the altered diastolic relaxation, at least in our model (11). Endothelial dysfunction such as that observed in this and other studies (1, 44) may account for a decreased coronary vasodilatory reserve, causing myocardial demand ischemia and depressed systolic function in conditions of hemodynamic stress, such as obtained after L-NAME infusion in PHT dogs. Under basal conditions (i.e., in the absence of L-NAME or agonist infusion), however, previous studies in the same PHT dog model have shown that myocardial blood flow, as measured using radioactive microspheres, was unchanged in these hypertensive animals compared with the values observed in controls (27). Infusions of L-NAME resulted in significant increases in LV loading conditions, and despite our assessment of LV function with less load-sensitive echocardiographic (Fig. 7) and hemodynamic (16, 47) parameters, we cannot exclude that increased afterload may affect LV function independently of direct myocardial effects of NO (or lack thereof). However, previous studies in nonobstructive, moderately hypertrophic cardiomyopathy emphasized the existence of an imbalance between external load and intrinsic myocardial inactivation processes, resulting in a paradoxical improved LV relaxation and early diastolic filling with increased afterload (26). Nevertheless, it is likely that mediators other than NO, such as atrial natriuretic factor (10), act to modulate at least basal diastolic function in the moderately hypertrophic heart, as recently demonstrated in mice deficient for the eNOS gene (23), and that decreased NO production may not explain all changes in LV parameters.

In conclusion, we have shown that the development of moderate hypertrophic cardiomyopathy is associated with a downregulation of eNOS and of the isoforms of caveolin that are known to tonically inhibit basal eNOS activity in endothelial cells and cardiomyocytes. These concurrent decreases, contrasting with stable expression of HSP90, which even increases in the subendocardium, may explain, in part, the observed alterations in vascular and cardiac contractility and its sensitivity to NOS inhibition despite decreased eNOS abundance. Finally, these observations emphasize the need to integrate total eNOS abundance with that of its posttranslational regulators, such as caveolin and HSP90, to predict the enzyme's activity and sensitivity to pharmacological inhibitors.


    ACKNOWLEDGEMENTS

We thank Drs. D. Fulton and W. C. Sessa (Yale University, New Haven, CT) for helpful advice.


    FOOTNOTES

* These authors contributed equally to this work.

This work was supported by grants from the Communauté Française de Belgique (Action de Recherche Concertée no. 96/01-199), and from the Belgian Fonds National de la Recherche Scientifique (Fondation pour la Recherche Scientifique Médicale nos. 3.4572.96 and 3.4598.97; to J. L. Balligand).

Address for reprint requests and other correspondence: J.-L. Balligand, Unit of Pharmacology and Therapeutics, Dept. of Medicine, FATH 5349, Université Catholique de Louvain, 53, Ave. E. Mounier, B-1200 Brussels, Belgium (E-mail: balligand{at}mint.ucl.ac.be).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 17 July 2001; accepted in final form 14 September 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Alyono, D, Anderson RW, Parrish DG, Dai XZ, and Bache RJ. Alterations of myocardial blood flow associated with experimental canine left ventricular hypertrophy secondary to valvular aortic stenosis. Circ Res 58: 47-57, 1986[Abstract/Free Full Text].

2.   Balligand, JL. Regulation of cardiac beta-adrenergic response by nitric oxide. Cardiovasc Res 43: 607-620, 1999[Free Full Text].

3.   Balligand, JL, and Cannon PJ. Nitric oxide synthases and cardiac muscle. Autocrine and paracrine influences. Arterioscler Thromb Vasc Biol 17: 1846-1858, 1997[Abstract/Free Full Text].

4.   Balligand, JL, Kobzik L, Han X, Kaye DM, Belhassen L, O'Hara DS, Kelly RA, Smith TW, and Michel T. Nitric oxide-dependent parasympathetic signaling is due to activation of constitutive endothelial (type III) nitric oxide synthase in cardiac myocytes. J Biol Chem 270: 14582-14586, 1995[Abstract/Free Full Text].

5.   Balligand, JL, and Smith TW. Molecular regulation of NO synthase in the heart. In: Endothelial Modulation of Cardiac Contraction, edited by Shah AM, and Lewis MJ.. Boston, MA: Harwood Academic, 1997, p. 53-71.

6.   Bartunek, J, Shah AM, Vanderheyden M, and Paulus WJ. Dobutamine enhances cardiodepressant effects of receptor-mediated coronary endothelial stimulation. Circulation 95: 90-96, 1997[Abstract/Free Full Text].

7.   Bayraktutan, U, Yang ZK, and Shah AM. Selective dysregulation of nitric oxide synthase type 3 in cardiac myocytes but not coronary microvascular endothelial cells of spontaneously hypertensive rat. Cardiovasc Res 38: 719-726, 1998[Abstract/Free Full Text].

8.   Belhassen, L, Feron O, Kaye DM, Michel T, and Kelly RA. Regulation by cAMP of post-translational processing and subcellular targeting of endothelial nitric-oxide synthase (type 3) in cardiac myocytes. J Biol Chem 272: 11198-11204, 1997[Abstract/Free Full Text].

9.   Dessy, C, Matsuda N, Hulvershorn J, Sougnez CL, Sellke F, and Morgan KG. Evidence for involvement of the PKC-alpha isoform in myogenic tone of the coronary microcirculation. Am J Physiol Heart Circ Physiol 279: H916-H923, 2000[Abstract/Free Full Text].

10.   Donckier, J, Hanet C, Galanti L, Stoleru L, Van Mechelen H, Robert A, Ketelslegers JM, and Pouleur H. Low-dose endothelin-1 potentiates volume-induced secretion of atrial natriuretic factor. Am J Physiol Heart Circ Physiol 263: H939-H944, 1992[Abstract/Free Full Text].

11.   Douglas, PS, and Tallant B. Hypertrophy, fibrosis and diastolic dysfunction in early canine experimental hypertension. J Am Coll Cardiol 17: 530-536, 1991[Abstract].

12.   Feron, O, Dessy C, Desager JP, and Balligand JL. Hydroxy-methylglutaryl-coenzyme A reductase inhibition promotes endothelial nitric oxide synthase activation through a decrease in caveolin abundance. Circulation 103: 113-118, 2001[Abstract/Free Full Text].

13.   Feron, O, Dessy C, Moniotte S, Desager JP, and Balligand JL. Hypercholesterolemia decreases nitric oxide production by promoting the interaction of caveolin and endothelial nitric oxide synthase. J Clin Invest 103: 897-905, 1999[ISI][Medline].

14.   Feron, O, Dessy C, Opel DJ, Arstall MA, Kelly RA, and Michel T. Modulation of the endothelial nitric-oxide synthase-caveolin interaction in cardiac myocytes. Implications for the autonomic regulation of heart rate. J Biol Chem 273: 30249-30254, 1998[Abstract/Free Full Text].

15.   Feron, O, Saldana F, Michel JB, and Michel T. The endothelial nitric-oxide synthase-caveolin regulatory cycle. J Biol Chem 273: 3125-3128, 1998[Abstract/Free Full Text].

16.   Frederiksen, JW, Weiss JL, and Weisfeldt ML. Time constant of isovolumic pressure fall: determinants in the working left ventricle. Am J Physiol Heart Circ Physiol 235: H701-H706, 1978.

17.   Garcia-Cardena, G, Fan R, Shah V, Sorrentino R, Cirino G, Papapetropoulos A, and Sessa WC. Dynamic activation of endothelial nitric oxide synthase by Hsp90. Nature 392: 821-824, 1998[Medline].

18.   Garcia-Cardena, G, Oh P, Liu J, Schnitzer JE, and Sessa WC. Targeting of nitric oxide synthase to endothelial cell caveolae via palmitoylation: implications for nitric oxide signaling. Proc Natl Acad Sci USA 93: 6448-6453, 1996[Abstract/Free Full Text].

19.   Gelpi, RJ, Hittinger L, Fujii AM, Crocker VM, Mirsky I, and Vatner SF. Sympathetic augmentation of cardiac function in developing hypertension in conscious dogs. Am J Physiol Heart Circ Physiol 255: H1525-H1534, 1988[Abstract/Free Full Text].

20.   Gleason, WL, and Braunwald E. Studies on the first derivative of the ventricular pressure pulse in man. J Clin Invest 41: 80-91, 1962.

21.   Gratton, JP, Fontana J, O'Connor DS, Garcia-Cardena G, McCabe TJ, and Sessa WC. Reconstitution of an endothelial nitric-oxide synthase (eNOS), hsp90, and caveolin-1 complex in vitro. Evidence that hsp90 facilitates calmodulin stimulated displacement of eNOS from caveolin-1. J Biol Chem 275: 22268-22272, 2000[Abstract/Free Full Text].

22.   Grocott-Mason, R, Anning P, Evans H, Lewis MJ, and Shah AM. Modulation of left ventricular relaxation in isolated ejecting heart by endogenous nitric oxide. Am J Physiol Heart Circ Physiol 267: H1804-H1813, 1994[Abstract/Free Full Text].

23.   Gyurko, R, Kuhlencordt P, Fishman MC, and Huang PL. Modulation of mouse cardiac function in vivo by eNOS and ANP. Am J Physiol Heart Circ Physiol 278: H971-H981, 2000[Abstract/Free Full Text].

24.   Hare, JM, Lofthouse RA, Juang GJ, Colman L, Ricker KM, Kim B, Senzaki H, Cao S, Tunin RS, and Kass DA. Contribution of caveolin protein abundance to augmented nitric oxide signaling in conscious dogs with pacing-induced heart failure. Circ Res 86: 1085-1092, 2000[Abstract/Free Full Text].

25.   Harrison, DG. Cellular and molecular mechanisms of endothelial cell dysfunction. J Clin Invest 100: 2153-2157, 1997[ISI][Medline].

26.   Hausdorf, G, Siglow V, and Nienaber CA. Effects of increasing afterload on early diastolic dysfunction in hypertrophic non-obstructive cardiomyopathy. Br Heart J 60: 240-246, 1988[Abstract/Free Full Text].

27.   Hayashida, W, Donckier J, Van Mechelen H, Charlier AA, and Pouleur H. Diastolic properties in canine hypertensive left ventricular hypertrophy: effects of angiotensin converting enzyme inhibition and angiotensin II type-1 receptor blockade. Cardiovasc Res 33: 54-62, 1997[Abstract/Free Full Text].

28.   Hayashida, W, Donckier J, Van Mechelen H, Charlier AA, and Pouleur H. Load-sensitive diastolic relaxation in hypertrophied left ventricles. Am J Physiol Heart Circ Physiol 274: H609-H615, 1998[Abstract/Free Full Text].

29.   Huang, PL, Huang Z, Mashimo H, Bloch KD, Moskowitz MA, Bevan JA, and Fishman MC. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature 377: 239-242, 1995[Medline].

30.   Ito, N, Bartunek J, Spitzer KW, and Lorell BH. Effects of the nitric oxide donor sodium nitroprusside on intracellular pH and contraction in hypertrophied myocytes. Circulation 95: 2303-2311, 1997[Abstract/Free Full Text].

31.   Keaney, JF, Jr, Hare JM, Balligand JL, Loscalzo J, Smith TW, and Colucci WS. Inhibition of nitric oxide synthase augments myocardial contractile responses to beta -adrenergic stimulation. Am J Physiol Heart Circ Physiol 271: H2646-H2652, 1996[Abstract/Free Full Text].

32.   Kelly, RA, Balligand JL, and Smith TW. Nitric oxide and cardiac function. Circ Res 79: 363-380, 1996[Free Full Text].

33.   Khadour, FH, O'Brien DW, Fu Y, Armstrong PW, and Schulz R. Endothelial nitric oxide synthase increases in left atria of dogs with pacing-induced heart failure. Am J Physiol Heart Circ Physiol 275: H1971-H1978, 1998[Abstract/Free Full Text].

34.   MacCarthy, PA, and Shah AM. Impaired endothelium-dependent regulation of ventricular relaxation in pressure-overload cardiac hypertrophy. Circulation 101: 1854-1860, 2000[Abstract/Free Full Text].

35.   Michel, JB, Feron O, Sacks D, and Michel T. Reciprocal regulation of endothelial nitric-oxide synthase by Ca2+-calmodulin and caveolin. J Biol Chem 272: 15583-15586, 1997[Abstract/Free Full Text].

36.   Mombouli, JV, and Vanhoutte PM. Endothelium-derived hyperpolarizing factor(s): updating the unknown. Trends Pharmacol Sci 18: 252-256, 1997[Medline].

37.   Mulvany, MJ, and Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res 41: 19-26, 1977[Free Full Text].

38.   Oka, N, Asai K, Kudej RK, Edwards JG, Toya Y, Schwencke C, Vatner DE, Vatner SF, and Ishikawa Y. Downregulation of caveolin by chronic beta -adrenergic receptor stimulation in mice. Am J Physiol Cell Physiol 273: C1957-C1962, 1997[Abstract/Free Full Text].

39.   Page, IH. The production of persistant arterial hypertension by cellophane perinephritis. JAMA 113: 2046-2048, 1939.

40.   Paulus, WJ, and Shah AM. NO and cardiac diastolic function. Cardiovasc Res 43: 595-606, 1999[Free Full Text].

41.   Rizzo, V, McIntosh DP, Oh P, and Schnitzer JE. In situ flow activates endothelial nitric oxide synthase in luminal caveolae of endothelium with rapid caveolin dissociation and calmodulin association. J Biol Chem 273: 34724-34729, 1998[Abstract/Free Full Text].

42.   Rousseau, MF, Veriter C, Detry JM, Brasseur L, and Pouleur H. Impaired early left ventricular relaxation in coronary artery disease: effects of intracoronary nifedipine. Circulation 62: 764-772, 1980[Abstract/Free Full Text].

43.   Sansdrap, J, Van Mechelen H, Pouleur H, and Charlier A. CROPA: a research-oriented pro