| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1998;115:343-350
© 1998 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
From the Department of Surgery, Division of Thoracic Surgery,a and the Department of Medicine, Division of Cardiology,b Duke University Medical Center, Durham, N.C.
Read at the Twenty-third Annual Meeting of The Western ThoracicSurgical Association, Napa, Calif., June 25-28, 1997.
Received for publication July 8, 1997; revisions requested August 29, 1997; revisions received Oct. 17, 1997; accepted for publication Oct. 17, 1997. Address for reprints: Ross M. Ungerleider, MD, Professor ofSurgery, Chief, Pediatric Cardiac Surgery, Duke University MedicalCenter, DUMC 3178, Durham, NC 27710.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Methods: Two groups of 4-week-old piglets were studied. In one group, the piglets were raised in an environment of 10% oxygen from 2 days of age (cyanotic, n = 6), and in the other group the piglets were raised at room air (control, n = 5). Pulmonary hemodynamics were measured in vivo for each animal, and peripheral lung biopsy specimens were obtained for Western blot analysis with the use of antiendothelial nitric oxide synthase antibody and for activity analysis with the use of the tritiated L-arginine assay.
Results: The piglets in the chronically hypoxic group had significant increases in mean pulmonary arterial pressure (44.0 ± 3.8 mm Hg vs 14.8 ± 1.2 mm Hg in controls, p = 0.0007) and pulmonary vascular resistance (7272.0 ± 871.1 dyne · cm · sec-5 vs 1844.5 ± 271.2 dyne · cm · sec-5in controls, p = 0.002). These changes in the pulmonary hemodynamics of the hypoxic piglets were accompanied by a twofold increase in the expression of pulmonary endothelial nitric oxide synthase (p = 0.0043) but no corresponding increase in nitric oxide synthase activity.
Conclusions: Raising infant piglets in an environment of 10% oxygen for 4 weeks results in significant pulmonary arterial hypertension accompanied by increased expression of nitric oxide synthase within the lung endothelium. Furthermore, the increased levels of nitric oxide synthase within the lungs of the hypoxic swine were not accompanied by a proportional increase in enzyme activity. These findings suggest that the development of pulmonary hypertension in infants with congenital cyanotic disease is not due to decreased expression of endothelial nitric oxide synthase, but instead may be related to a decreased ability of the enzyme to produce sufficient nitric oxide.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
1 in 1980. 2,3
NO is produced within endothelial cells by the enzyme endothelial nitric oxide synthase (eNOS). 4 The reaction catalyzed by eNOS is dependent on the intracellular presence of calcium and calmodulin and also requires the presence of certain cofactors including tetrahydrobiopterin and reduced nicotinamide adenine dinucleotide phosphate (NADPH). 5 The NO thus produced diffuses to nearby vascular smooth muscle cells, where it causes a G-protein–mediated increase in cyclic guanosine monophosphate, which in turn leads to smooth muscle cell relaxation and thus vasodilatation. 6 Because of its role as a potent vasodilator and subsequent findings that inhibition of its production enhances the development of hypoxia-induced pulmonary vasoconstriction, a deficiency in the production of NO has been hypothesized to play a role in the development of pulmonary hypertension in the setting of chronic hypoxia. 7-9
Whether this decrease in NO production is due to decreased expression of eNOS has been studied in both human beings and rats, but the results of these investigations have been generally inconclusive. 10-13 Furthermore, most of these studies have involved adult models. Our laboratory is specifically interested in the relationship between eNOS expression and pulmonary hypertension as it applies to infants with congenital cyanotic heart disease. Our aim in this study was to develop a large-animal infant model of hypoxia-induced pulmonary hypertension and to determine the levels of pulmonary eNOS protein and activity within this model. We hypothesized that hypoxia-induced pulmonary hypertension in infants is due to a decrease in the production of NO, either through a reduction in the expression of pulmonary eNOS or through a decrease in this enzyme's activity.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Data acquisition.
At 4 weeks of age, anesthesia was induced in each animal with an intramuscular injection of ketamine (50 mg/kg) and acepromazine (15 µg/kg). Intravenous methylprednisolone (30 mg/kg) was administered via an 18-gauge cannula in the marginal vein of the pinna. Orotracheal intubation was performed and mechanical ventilation (Infant Ventilator; Sechrist Industries, Anaheim, Calif.) was commenced to achieve an arterial oxygen tension of 150 to 250 mm Hg and an arterial carbon dioxide tension of 35 to 45 mm Hg. The animals were paralyzed with intravenous pancuronium (300 µg/kg) and anesthetized with fentanyl (100 µg/kg). Thereafter, anesthesia was maintained with a continuous infusion of fentanyl (25 µg/kg per hour). An 18-gauge cannula was placed in the left femoral artery for blood pressure monitoring and arterial blood sampling. The heart was exposed through a median sternotomy. Cardiac instrumentation consisted of micromanometers (Millar Instruments, Inc., Houston, Tex.) inserted into the left atrium and the main pulmonary artery and a 10 mm ultrasonic flow probe (Transonic Systems, Ithaca, N.Y.) around the proximal pulmonary artery.
eNOS protein analysis.
A peripheral pulmonary biopsy specimen was obtained from each animal and snap frozen in liquid nitrogen after storage at –80° C. A portion of the tissue from each hypoxic and control animal was later thawed, weighed, and homogenized in 150 µl ice cold tromethamine (Tris, 50 mmol/L) containing phenylmethyl-sulfonylfluoride (0.2 mg/ml), leupeptin (0.5 µg/ml), and 0.1% octyphenoxy polyethoxyethanol (Triton X-100, Union Carbide Corp., Canbury, Conn.). The homogenate was then lysed by sonication (Brinkman Instruments, Inc., Westbury, N.Y.), and pelleted by centrifugation (Brinkman Instruments). Total protein concentration of the supernatant was then determined by the Bradford assay (Bio-Rad Laboratories, Richmond, Calif.) with bovine serum albumin as the standard. 14
Equal samples of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis on 9% polyacrylamide gels and then electrophoretically transferred to nitrocellulose membranes for Western blot analysis. The membrane was then blocked in 7.5% nonfat milk in tromethamine-buffered saline solution/polysorbate (Tween) and then incubated with primary monoclonal anti-eNOS antibody at the manufacturer's specified concentration (Transduction Laboratories, Lexington, Ky.) for 1 hour at room temperature. eNOS protein bands were then visualized with antimouse immunoglobulin G horseradish peroxidase-conjugated antibody followed by chemiluminescence (DuPont, Boston, Mass.). The relative quantities of eNOS within the visualized protein bands were then quantified by laser densitometry (LKB Ultroscan XL, LKB Instruments, Inc., Rockville, Md.). Relative protein quantity was expressed as the percentage abundance in the control animals.
NOS activity.
So that NOS activity in the lung tissue could be determined, samples from each biopsy were lysed by sonication in ice-cold, freshly prepared assay buffer containing tromethamine (Tris, 20 mmol/L), pH 7.4, ethylenediaminetetraacetic acid (EDTA, 1 mmol/L), and protease inhibitors soy trypsin inhibitor (100 µg/ml), N
-p-tosyl-L-lysine chloro-methyl keatone (TLCK, 50 µg/ml), phenylmethylsulfonyl fluoride (200 µg/ml), leupeptin (0.5 µg/ml), pepstatin (0.7 µg/ml), aprotinin (1 µg/ml), and antipain (50 µg/ml). So that NOS activity would be preserved, the buffer also contained NOS cofactors flavin adenine dinucleotide (5 µmol/L), flavin mononucleotide (5 µmol/L), biopterin (7 µmol/L), and diphtheria tetanus toxoid (DTT, l mmol/L). The cellular debris were then pelleted by centrifugation and 10 µl of the remaining lysate was added to 100 µl of a pre-prepared assay buffer mix containing calmodulin (150 nmol/L), calcium dichloride (1.75 mmol/L), biopterin (7.3 µmol/L)/DTT (1.33 mmol/L), 3[H]-arginine (1 µCi; 2 x 106 dpm [disintegrations per minute]), and NADPH (1 mmol/L), which was added last. The assay buffer and lysate sample were incubated at 30° C for 10 minutes, and the reaction was then stopped with the addition of 2 ml of stop solution (HEPES, 20 mmol/L; pH 5.5; EDTA, 2 mmol/L). The reaction wells were then washed with 2 ml of stop solution and the total mixture was then poured over a 1 ml silver 50W-X8 (sodium form) resin column (Bio-Rad 143-1451, Bio-Rad Laboratories), which had been pre-washed with 2 ml of stop solution. The flow through was collected and then the column was washed with an additional 2 ml of stop solution. Eluate (500 µl) was placed in scintillation vials quantified by liquid scintillation spectroscopy. The sample used for background calculation was processed in exactly the same manner, whereas the sample used for total counts was added directly to stop solution without column purification. The proportion of arginine that was convened to citrulline (and therefore the production of NO) was calculated as follows: (Disintegrations per minute in sample – Background disintegrations per minute)/(Total disintegrations per minute). To convert this proportion to moles of NO, we calculated the total quantity of arginine in the assay from the total disintegrations per minute (with a factor of 8 since only 500 µl was counted) and the known specific activity of 3[H]-arginine. NOS activity was then expressed as picomoles NO produced per milligrams of tissue per minute of incubation.
Statistical analysis.
The pulmonary hemodynamic variables, eNOS protein levels, and NOS activity between the control and hypoxic groups were compared by nonpaired Student's t tests. All results are expressed as mean ± standard error of the mean. eNOS protein levels are expressed as a percentage of abundance in normoxic control lung.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Pulmonary hemodynamics and ventricular weights.
While undergoing mechanical ventilation at normoxic conditions, the hypoxic animals had significant increases in both mean pulmonary vascular pressure and pulmonary vascular resistance when compared with the control animals (Table I). The right ventricular masses were significantly greater in the hypoxic animals than in the control animals, but the left ventricular masses did not differ. This resulted in a threefold increase in the right ventricle/left ventricle weight ratio in the hypoxic animals (Table I).
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Western blot analysis on lung biopsy tissue from all study animals using specific anti-eNOS antibody revealed that the hypoxic piglets had twofold more pulmonary eNOS than the control piglets, suggesting that neonatal pulmonary eNOS expression is up-regulated in the setting of chronic hypoxia. We then performed in vitro assays on the lung samples for the conversion of 3[H]-arginine to 3[H]-citrulline to determine the activity of the NOS present within the lungs. The assay was performed in room air with an excess of L-arginine and all the cofactors used by the enzyme. We found no significant difference in the activity of NOS per milligram of lung protein between the hypoxic and control groups. Inasmuch as the enzyme concentration was 101% higher in the hypoxic animals, a corresponding increase in the activity of NOS per milligram of lung protein would be expected. The absence of such a proportional increase suggests that the intrinsic ability of the enzyme to produce NO is impaired in the hypoxic animals. One possible explanation for this decrease in NOS activity is that the hypoxic piglets had reduced levels of molecular oxygen, a crucial substrate in the production of NO by the synthase enzyme. 15 Similarly, a reduction in the availability of L-arginine, the other substrate in the NOS reaction, may also account for the decreased production of NO. 16 However, such deficiencies do not explain the in vitro reduction in NOS activity that we demonstrate because this assay was performed in room air with an excess of L-arginine. This indicates that the eNOS enzyme itself may be functionally impaired in chronic hypoxia.
Our finding of increased pulmonary expression of eNOS production in the chronically hypoxic infant swine contrasts with results from anoxic cultured human vascular endothelial cells 17 and in adult patients with pulmonary hypertension of various causes, 13 both of which have shown a reduction in the expression of eNOS messenger ribonucleic acid. These results would suggest that chronic hypoxia compromises the ability of pulmonary vascular endothelium to produce NOS. However, studies in adult rats support our findings of overexpression of pulmonary eNOS in chronic hypoxia. For example, exposing adult rats to hypoxia for extended period of time has repeatedly been shown to result in increased levels of both eNOS messenger ribonucleic acid and protein. 10-12 Other investigations have shown that the pulmonary hypertension in adult rats can be significantly reduced if the rats are exposed to low levels of inhaled NO during hypoxia, indicating that the production of NO is deficient despite an overexpression of the enzyme that produces it. 18 Although neonatal piglets are clearly different from human infants, swine were chosen in this model because of their anatomic and physiologic similarities to human beings.
Although our model shares many of the advantages of the adult rodent model, especially the ability to expose the experimental subjects to low oxygen levels in a standardized laboratory setting, an important difference is that our model attempts to explain the relationship between hypoxia-induced pulmonary hypertension and eNOS expression in an infant model. Because the endothelium-dependent vasodilatory capabilities of pulmonary vessels are likely to show maturational variations, 19,20 it is important to use an infant model when examining the relationship between eNOS and pulmonary hypertension caused by congenital cyanotic heart disease. Recent investigations of isolated lungs from chronically hypoxic neonatal swine have shown a dampened response of these lungs to the endothelium-dependent vasodilatory agent acetylcholine when compared with age-matched normoxic control animals. 21 When exposed to N-nitro-L-arginine methyl ester (L-NAME), a known inhibitor of NOS, 22 the lungs from the control animals showed significant vasoconstriction whereas the hypoxic lungs showed no response. These results suggest that the production of NO in the lungs of the hypoxic swine is reduced, thus supporting our finding of decreased eNOS activity in chronically hypoxic newborn piglets.
Importantly, our study addresses the effects of chronic hypoxia on the expression of pulmonary eNOS in newborn infants. We demonstrate that the development of hypoxia-induced pulmonary hypertension in neonates is associated with an overexpression of pulmonary eNOS, indicating that the ability of the pulmonary vascular endothelium to support the production of this enzyme is not limited by chronic hypoxia, and may in fact be stimulated by chronic hypoxia. Furthermore, we have shown that the ability of this enzyme to produce NO may be reduced in the setting of chronic hypoxia and that this decrease in enzyme function may contribute to the development of pulmonary hypertension. Therefore, efforts in treating infants with pulmonary hypertension associated with cyanotic heart disease should be directed toward understanding the mechanisms of decreased eNOS activity and increasing the production of NO.
|
Appendix: Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
I have three separate areas for further discussion. The first relates to interpretation of your data. You have shown that eNOS doubles, but interestingly, when you take the enzyme and challenge it in vitro to determine how much NO it actually makes, it does not appear to make any more than the samples from the control population. Have you evaluated this any further? Is the actual enzyme itself altered or is an associated inhibitor in your lysate competing with the enzyme. Along those lines, are you implying any kind of cause and effect with the pulmonary hypertension? Pulmonary vascular tone has multiple regulators. Is it possible that your elevated eNOS is simply a marker of other mechanisms and this elevation is simply a reaction to increased vasoactive tone caused by multiple other factors?
Mr. Scarborough. Those are very good points. The activity assay that we used has a couple of limitations that should be noted. First, it is an in vitro assay, which takes the process of NO production away from the body and the in vivo circumstances. Second, the assay is not specific for the endothelial isoform, so we cannot say that eNOS per se is not significantly increased. However, we do not think a reduction in another isoform's activity is masking an increase in the enzyme's activity. We are currently doing some follow-up studies to investigate why the activity of NOS may be decreased. A couple of possibilities that have been suggested in prior literature are possible deficiencies in the substrates L-arginine and molecular oxygen. Because our in vitro assay was performed with an excess of these substrates, we believe that deficiencies in these substrates are not a likely cause. Other explanations include the cofactors; one of particular interest is tetrahydrobiopterin. Although we have not examined this relationship further, some of the literature on coronary atherosclerosis has speculated that a deficiency in tetrahydrobiopterin may reduce the enzyme's activity and that these reductions may not be correctable merely by adding the cofactor in its preformed form into an in vitro assay. We would like to investigate that further. Also, we have undertaken a similar study in 2-week-old hypoxic piglets in which the vasodilatory response to acetylcholine in 2-week-old hypoxic animals was much less than in 2-week-old control animals, an observation that does seem to back up our findings in the in vitro assay that we used for the 4-week-old animals. It is, however, possible that there is an enzyme in the lysate that somehow interferes with the enzyme's ability to produce NO.
Dr. Hanley. The second point for discussion relates to methods. You have shown that the enzyme is doubled in the hypoxic tissue from your biopsy. Have you considered the following potential problem? These are neonatal piglets that have hyperplastic potential in their lung tissue and lung vascular tissue. Have you done any morphometric studies or assays of other endothelial protein to rule out the possibility that you are stimulating more vasculature here, and therefore, when you get 1 gm of lung tissue, you may actually have twice as much vessel and twice as much endothelium so when you look at the samples, you find you have twice as much eNOS but not twice as much expression?
The third point relates more to the appropriateness of the model. You emphasized that you were trying to mimic cyanotic congenital heart disease with this model. I would argue that you have created a very nice model for mountain sickness, but breathing thin air is not a model of cyanotic congenital heart disease. These animals are breathing lower oxygen concentrations, but they have no intracardiac mixing and they have absolutely normal pulmonary blood flow. What their vascular endothelium is seeing in your model is very different from that in the example of cyanotic congenital heart disease clinically. Cyanotic animals are hypoxemic because of mixing and sometimes reduced pulmonary blood flow, but the blood that does go through the microvasculature of the lung is no different than in you or me. A normal individual breathes in room air and oxygenates it; a cyanotic individual does the same thing. Your model is very different. This is still a very important study, but to say that it is an example of what happens in cyanotic congenital heart may well not be correct. I would like your input on this observation.
Mr. Scarborough. The alternative to our method of producing pulmonary hypertension would be to try to construct a shunt in a newborn piglet; we avoided that because we thought it might introduce too many other variables of injury that would cloud the issue. In developing our model, we used as our primary basis the work of another Duke group headed by Dr. James Lowe, who had constructed this hypoxic model to look at myocardial contractility in chronic hypoxia. Our method of producing chronic hypoxia is consistent with those of other investigators studying the effects of hypoxia on eNOS production.
Dr. Hanley. I think you are doing an excellent job in a complex area that has great potential for clinical application, and I would encourage you to continue this work.
Dr. David A. Fullerton (Winnetka, Ill.). I would like to echo some of the comments made earlier. The model you have used is a standard model to develop pulmonary hypertension. Work published in the early 1990s nicely demonstrated in sheep that throughout the newborn period for approximately 2 weeks after birth the amount of eNOS is diminished. It is part of the maturation problem, and it is thought that this may contribute to higher pulmonary arterial pressures in newborn infants. Work has shown that through augmented expression of vascular endothelial growth factor, there is a proliferation of capillaries including endothelial cells. Do you have any histologic studies looking at neovascularization as part of your explanation for some of your findings?
Mr. Scarborough. We do have specimens for histologic studies. However, we have not looked at those and we have not done morphometric analysis at this time.
Dr. Fullerton. It is paradoxical that you find elevated levels of eNOS in the presence of very high pulmonary arterial pressure. I realize that is the whole crux of your issue. It turns out that there is a strain of rats that spontaneously become pulmonary hypertensive, and they too share this paradox; they have high levels of eNOS, yet very high pulmonary arterial pressures. Could you speculate on how this happens? Is it possible that the eNOS is a compensatory response to the hypoxia-induced pulmonary hypertension?
Mr. Scarborough. That would make sense, given the role of eNOS in producing NO, which then vasodilates the vessels. There could be an adaptational up-regulation. There may also be some circumstances surrounding chronic hypoxia that would prevent the enzyme from doing its job, although the same could be said theoretically about a decrease in eNOS enzyme itself causing pulmonary hypertension. That, I believe, has been the contrast in the past, trying to decide which one of those is the case. I think enough studies have been conducted to suggest that eNOS expression probably is increased. Studies in brown humpback rats also lend support to that. The goal now is to evaluate at the molecular level the reason that the enzyme is not producing NO.
Dr. Hanley. Just one last comment: We are trying to get around this cause-and-effect thing. If your house is not on fire, there usually is not much water on it; if your house is on fire, there is an increase in the amount of water on your house, but that is not a paradox. In the same way, if a lung is chronically hypoxic and vasoconstricted, the increase in eNOS is just a reaction, even though eNOS is a vasodilator. I do not find this paradox very difficult to understand myself.
Mr. Scarborough. That is a good point. In fact, we believe that we have shown that the pulmonary vascular endothelium has retained its ability to produce NO despite the hypoxic environment.
Dr. Fullerton. Does endothelin have an effect? Have you measured any of that in the model?
Mr. Scarborough. We have not measured endothelin in this model. That would be a good idea for future studies.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. D. Fike, J. L. Aschner, Y. Zhang, and M. R. Kaplowitz Impaired NO signaling in small pulmonary arteries of chronically hypoxic newborn piglets Am J Physiol Lung Cell Mol Physiol, June 1, 2004; 286(6): L1244 - L1254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. G. Chicoine, J. W. Avitia, C. Deen, L. D. Nelin, S. Earley, and B. R. Walker Developmental differences in pulmonary eNOS expression in response to chronic hypoxia in the rat J Appl Physiol, July 1, 2002; 93(1): 311 - 318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hampl and J. Herget Role of Nitric Oxide in the Pathogenesis of Chronic Pulmonary Hypertension Physiol Rev, October 1, 2000; 80(4): 1337 - 1372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nagashima, U. Stock, G. Nollert, J. Sperling, D. Shum-Tim, S. Hatsuoka, and J. E. Mayer Jr Effects of cyanosis and hypothermic circulatory arrest on lung function in neonatal lambs Ann. Thorac. Surg., August 1, 1999; 68(2): 499 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Martinez-Lemus, R. K. Hester, E. J. Becker, J. S. Jeffrey, and T. W. Odom Pulmonary artery endothelium-dependent vasodilation is impaired in a chicken model of pulmonary hypertension Am J Physiol Regulatory Integrative Comp Physiol, July 1, 1999; 277(1): R190 - R197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
