(Circulation. 1996;94:1948-1953.)
© 1996 American Heart Association, Inc.
Articles |
the Departments of Pediatrics of Louisiana State University Medical Center and Tulane University Medical Center (M.R.P), New Orleans, La.
Correspondence to Mark J.S. Miller, PhD, Department of Pediatrics, LSU Medical Center, 1542 Tulane Ave, New Orleans, LA 70112.
| Abstract |
|---|
|
|
|---|
Methods and Results Positive evidence for iNOS gene expression was noted in heart central vessels and placenta of untreated rat fetuses. Rats in the last week of pregnancy were treated for 5 days with L-NG-(1-Iminoethyl)lysine (L-NIL), a selective inhibitor of iNOS, at 1, 10, and 100 µg/mL in the drinking water. To raise NO levels, lipopolysaccharide (LPS) 30 µg/kg was given by intraperitoneal injection, and sodium nitroprusside (SNP) was placed in mini-osmotic pumps to deliver 10 µg/kg per minute. Control animals were undisturbed. On day 21 of gestation, dams were anesthetized and fetuses were delivered by cesarean section and rapidly frozen in isopentane chilled in liquid nitrogen. Frozen sections (10 µm) were used to reconstruct a computer-generated three-dimensional image of the great vessels and ductus arteriosus. Significant constriction of the great vessels and ductus arteriosus was observed with L-NIL, whereas both LPS and SNP dilated these vessels. The vasorelaxant effect of LPS was blocked by L-NIL. NO release from placental explants was 633±41 nmol/L under basal conditions, increasing to 4.0±0.4 µmol/L with LPS administration, although placental iNOS message and protein levels were unchanged.
Conclusions We suggest that nitric oxide, generated by iNOS, plays a significant role in control of major vessel and ductus arteriosus caliber in the rat fetus. In regard to the nitrergic regulation of the circulation, the fetus is clearly different from the adult.
Key Words: ductus arteriosus, patent free radicals pharmacology polymerase chain reaction vasculature
| Introduction |
|---|
|
|
|---|
It is known that NO synthesis is increased during pregnancy.15 In guinea pigs, increased NO synthesis has been attributed to a greater expression of cNOS.16 However, iNOS also has been implicated as a source of NO in the utero-placental unit.17 The present study was designed to obtain data on the role of iNOS-derived NO in the great vessels and DA of the rat fetus near term. We first determined that iNOS is present in fetal vascular tissues, then used LPS as an inducer of NO release and SNP as an NO donor to compare their effects with L-NIL, a selective inhibitor of iNOS,18 and untreated controls.
| Methods |
|---|
|
|
|---|
3 Prime, Inc, Boulder CP). First-stand complementary DNAs were synthesized from 1 µg mRNA with the use of random primers (Mannheim Boehringer Biochemicals) and SuperScript II RNase H Reverse Transcriptase (GIBCO Laboratories). The first-strand complementary DNA templates were amplified for glyceraldehyde-3-phosphate dehydrogenase and iNOS by PCR. The 22 base-pair primers for iNOS represent a conserved region of rat, mouse, and human iNOS (made available from Searle). The sense primer was TCG AAA CAA CAG GAA CCT AAA A at location 529-550. The antisense primer was ACR GGG GTG ATG CTC CCA GAC A at location 1435-1414 (Genbank D14051). The primers for glyceraldehyde-3-phosphate dehydrogenase, used as an internal standard, are the following rat sequences: sense, ATT CA CCC ACG GCA AGT GAC CC (Genbank M17701). PCR cycle was an initial step of 95°C for 3 minutes, followed by 94°C for 30 seconds; 60° for 45 seconds, 72°C for 1 minute of 30 cycles, and a final cycle of 72°C for 4 minutes. The negative control used was a mixture of all reagents except the complementary DNA template. The positive control was a linearized murine complementary DNA template. The positive control was linearized murine complementary DNA clone provided by Dr James M. Cunningham (Brigham and Women's Hospital, Boston, Mass).19
Detection of iNOS by Western Blotting
Rat placenta was homogenized in boiling lysis buffer (10 mmol/L Tris-HCl, pH 7.4; 1% sodium dodecyl sulfate) and centrifuged at 12 000g at 4°C for 15 minutes. The supernatant (50 mg of protein) was separated on 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and the proteins were blotted onto polyvinylidene difluoride membrane. The membrane was blocked with buffer A (10% nonfat milk; 20 mmol/L Tris-HCL, pH 7.5; 500 mmol/L NaCl; 0.1% Tween 20) and subsequently incubated with rabbit polyclonal antibody against mouse iNOS (1:200) diluted in buffer B (25% nonfat milk; 20 mmol/L Tris-HCL, pH 7.5; 50 mmol/L NaCl; 0.1% Tween 20). The specific protein was detected by using goat anti-rabbit immunoglobulin G conjugated with the horseradish peroxidase enzyme (1:1000) (Bio-Rad Laboratories). Protein determinations were made with Bio-Rad protein assay with bovine serum albumin as a standard. Prestained molecular-weight standards (Bio-Rad) were used as markers.
Electrochemical Detection of NO
Placenta was removed from anesthetized rats and transferred to a tissue culture dish containing culture media (RPMI 1640). The tissue was pinned flat at the bottom of the culture dish, followed by insertion of the reference and auxiliary electrodes in the culture media containing the tissue. The NO-sensitive electrode was then placed under the surface of the culture media immediately above the tissue (
5 mm). Once the current had stabilized, the electrode was lowered to the tissue surface as determined microscopically, with the use of a micromanipulator. A steady-state reading of NO release (current flow) was recorded, followed by elevation of the NO-sensitive electrode above the tissue, and the baseline was reestablished. This entire procedure takes about 5 minutes, and NO fluxes were determined within 10 minutes of removing the tissue from the rat. Calibration of the NO concentration was achieved with a pure NO solution with serial dilutions. NO-sensitive microelectrodes were manufactured on site.
Imaging and Measurements of Rat Fetus Vessels
Experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee of Louisiana State University Medical Center. Timed pregnant rats (Holtzman Harlan Sprague-Dawley) weighing 370 to 450 g were obtained from a commercial supplier on the 14th day of gestation, allowed 2 days to acclimate, and started treatments on the 16th day of pregnancy as we have done in previous experiments.11 Fourteen pregnant rats were used in the experiments randomly assigned to one of seven treatments: control, L-NIL1, L-NIL10, L-NIL100, LPS, SNP, and L-NIL100-LPS. L-NIL was given in the drinking water at concentrations of 1 µg/mL (L-NIL1) (
0.2 mg/kg per day), 10 µg/mL (L-NIL10) (
2 mg/kg per day), and 100 µg/mL (L-NIL100) (
20.0 mg/kg per day). LPS was given by intraperitoneal injection 30 µg/kg per day for 5 days. SNP was administered with the use of a miniosmotic pump (Alzet model 2002, 0.5 µL/hr; Alza Corp) implanted under general anesthesia with ketamine 40 mg/kg per minute. Control animals were undisturbed.
On gestational day 21, dams were anesthetized with an intramuscular injection of ketamine 40 mg/kgxylazine 4 mg/kg; fetuses were rapidly delivered by cesarean section and frozen in isopenthane chilled to <-150°C in liquid nitrogen. Tissue blocks of fetal chests were cut in a cryostat set at -20°C, starting from the left upper shoulder toward the lower hepatic area. Frozen sections 10 µm thick were mounted in microscope slides, stained with hematoxylin and eosin, and inspected for sequential images of the AO, PA, and DA. Best sequential images were selected (seven in each group) for computerized reconstruction as previously described.11 Computer reconstructions of DA, AO, and PA procedures involve three steps: capturing, contouring, and assembling. Every section and a measuring bar were captured in a digital image with the use of a JVC 1024x1024 CCD camera mounted on a Nikon stereo zoom microscope. Contours of the internal wall of the vessels were stored as x,y coordinates by means of a digitizing pad. The x and y coordinates and the z spacing (10 µm) measured in any plane were recorded for each image.20 A standard 1-mm measuring reference is included with each image.
Drugs
The following pharmacological agents were used: L-NIL (G.D. Searle Co), LPS (endotoxin from Escherichia coli, Sigma Chemical Co); and SNP (Fisher Scientific Co).
Statistical Analysis
The results are expressed as mean±SEM. Statistical evaluation of the data was performed by ANOVA with Student-Newman-Keuls subsequent test for differences between groups with the use of a commercially available statistics program (SigmaStat, Jandel Scientific). A value of P<.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
Caliber measurements of the great vessels and DA were made in the fetuses of 14 dams with an average weight of 402 g consuming an average 367 mL of water in 5 days. In contrast with previous studies using L-NAME,21 an NOS inhibitor with weak (10-fold) selectivity for iNOS, no congenital anomalies were detected by gross inspection in any of the rat pups treated with L-NIL. The average litter of 9.1 pups produced 128 pups. While more than 7 pups were obtained in each group, the first 7 to succeed adequate sequence of cryostat sectioning were used to proceed with the computer-generated three-dimensional image. In Fig 3
, a graphic representation of the descending aorta caliber is presented. Ascending aorta calibers were equally affected by the treatments, and statistical analysis was identical to that seen in the descending aorta. Figs 4 and 5![]()
are graphic representations of the calibers of the DA and PA. Note that all three concentrations of L-NIL evoked a significant constriction of the DA and reduced caliber of the great vessels (P<.001). Both the NO donor SNP and LPS significantly increased the caliber of the DA and great vessels (P<.05). L-NIL100 given at the same time as LPS caused a significant constriction of all vessels and prevented the vasorelaxant effect of LPS (P<.001). The concentration-dependent response to L-NIL is statistically significant for the great vessels (P<.001). The trend of DA constriction, significantly different from control in all three concentrations, is notably related to dose but without statistical significance between each concentration.
|
|
|
Figs 6 through 8![]()
![]()
show examples of the computer-generated images. Fig 6
is from the control group of rat fetuses. Fig 7
is from the group of rat fetuses born to a dam treated with L-NIL100. Fig 8
is from a rat fetus born to a dam treated with LPS to illustrate the contrasting changes in the caliber of the AO, PA, and DA evoked by treatments.
|
|
|
| Discussion |
|---|
|
|
|---|
NO has important vasodilatory actions in the fetal circulation as it does in the adult. Inhaled NO is currently used for neonatal primary pulmonary hypertension.22 23 24 Although there is excellent evidence that NO is an important contributor to the maintenance of vasodilatory tone in the fetal circulation,25 26 there has been little attention to the enzyme source of NO in the fetus. Since there was one report that the inducible form of NOS was present in the utero-placental unit,17 we addressed the possibility that iNOS was present in the fetus and contributed to the physiological regulation of the circulation as opposed to pathological states as described for the adult.27
The results were surprising and provocative. It appears that iNOS is expressed in a sustained manner in the fetus and that NO derived from iNOS is important in the maintenance of the caliber of the fetal great vessels. Inhibition of iNOS with a selective inhibitor of iNOS, L-NIL, resulted in a decreased vessel diameter, whereas maternal administration of LPS or SNP to further increase iNOS or directly raise NO levels resulted in an increased vessel diameter. While it may be predictable that NO influences the tone of the fetal circulation, the enzyme source is remarkable. The expression of iNOS in the circulation is normally associated with pathological states, eg, shock or atherosclerosis.28 29 In other states, iNOS expression is associated with cell death or cytostasis, as high concentrations of NO, generated by iNOS, are utilized to kill or halt the growth of tumors or invading microorganisms.30 Thus, iNOS is only expressed in states of immune activation requiring a substantive marshaling of host defense forces. That current opinion is challenged by the present results. In the fetus, the same enzyme used to promote cell death is used to maintain the vasculature in a dilated state, apparently without any pathological consequences. While this study does not resolve this enigma, we speculate that the answer to this riddle may lie in the oxidation of NO to alternative products with cytotoxicity. Peroxynitrite (ONOO-) is formed from the interaction of NO with superoxide (O2-) and is a powerful oxidant. While peroxynitrite is ephemeral, it can be traced in vivo immunohistochemically by its ability to nitrate tyrosine residues.31 In inflammation and atherosclerosis, iNOS expression colocalizes with nitrotyrosine formation,29 and we have demonstrated that NOS inhibition eliminates nitrotyrosine immunoreactivity, indicating that nitrotyrosine was derived from NO even though NO cannot directly form nitrotyrosine (it can nitrosate but not nitrate tyrosine residues19 27 31 ). In the placenta, where iNOS is expressed, we have not found any evidence for immunoreactive nitrotyrosine (data not shown). Thus, in the fetus, iNOS can be expressed without the formation of peroxynitrite.
The observation that LPS causes an increase in the diameter of the ductus and great vessels does not prove that it mediates by increasing NO production, although we did observe increased NO release from placental explants of LPS-treated dams. However, the effects of SNP indicate that additional NO can cause further relaxation of these vessels. Endotoxin will also stimulate the expression of cyclo-oxygenase 2 (COX2) and increase prostaglandin production.32 33 Prostaglandins are well known to cause relaxation of the DA; indeed, they are used clinically for this purpose.34 While we are not aware of reports that COX2 can be expressed in the fetal vasculature it is possible, particularly as iNOS-derived NO appears to directly stimulate the expression of COX2 and iNOS inhibitors have been postulated to work by the inhibition of NO and prostaglandin synthesis in models of inflammation.35 However, it should not be misconstrued that prostaglandins mediate all the responses reported in this study. The ability of L-NIL to completely negate the effects of LPS and the direct vasodilatory effects of SNP suggests a direct involvement of NO.
A paradoxical aspect of this study was the observation that LPS treatment resulted in increased NO release from the placenta without increasing iNOS gene expression or protein levels, suggesting that posttranslational mechanisms may account for the LPS-dependent increase in NO release. However, we have no direct evidence for changes in cofactor levels at this time. Previously, LPS was believed to act at transcriptional levels, promoting iNOS gene expression. However, in these states the baseline was zero, ie, iNOS was absent in the basal state. Pregnancy poses a different background, with significant iNOS expression under basal conditions. Further experimentation into how LPS resulted in an increase in placental NO release independent of iNOS transcription or translation may provide important clues to other pathophysiological states.
Previously we noted that the constrictor effects of L-NAME and indomethacin on the caliber of the DA and great vessels were additive.11 Great vessel and DA constriction was greater with L-NIL than L-NAME; comparable effects were not with doses of L-NIL that were 100-fold lower than L-NAME. This may suggest that iNOS plays a greater role than cNOS in the regulation of fetal vascular tone. Alternatively, L-NIL may have been more effective because of combined actions on NO and prostaglandins.35 We have also noted that neonatal brachial or femoral arteries have a higher basal (but not stimulated) efflux of NO when compared with adult arteries.36 The comparable vasorelaxant responses to stimuli (eg, acetylcholine) suggest that cNOS activity was similar in neonatal and adult rats. The high NO-dependent tone under basal conditions is less readily explained in terms of cNOS expression, especially because the method used eliminated the influence of shear stress.36 We believe that the current results suggest that the high basal NO-dependent vasorelaxant responses in the neonate are the result of iNOS-dependent NO production. We have previously reported that L-NAME administration results in limb reduction defects in fetal rats.21 This effect was blocked by NO donors or supplemental L-arginine.36 In contrast to L-NAME, no limb reduction defects were noted with L-NIL treatment. Similarly, aminoguanidine, another selective iNOS inhibitor, does not cause limb reduction defects.36 This suggests that the effect was specific to L-NAME. Factors in addition to NO may be involved, as we have previously noted that L-NAME but not aminoguanidine stimulates endothelin release.37
There are obvious clinical implications to these results. Currently, prostaglandins of the E-series are used to maintain the patency of the DA in neonates with congenital heart defects before corrective surgery. It is possible that a combination of NO donors and prostaglandin E will have improved efficacy and tolerance. The congenital lack of NOS in primary pulmonary hypertension is already being treated with inhaled NO. However, a direct comparison of iNOS versus cNOS in this condition has not been evaluated. Although specific iNOS inhibitors are not clinically available, when they are tested in human subjects these results suggest that pregnant women should be excluded because they may compromise the fetal circulation. This is the first report identifying iNOS in the fetal circulation and demonstrating that L-NIL, a selective inhibitor of iNOS, may compromise the fetal circulation.
Conclusions
We interpret these results to indicate that the fetus utilizes both constitutive and inducible NOS isoforms to produce NO in the circulation. In contrast to the adult, the fetus can express iNOS in a sustained manner for the relaxation of vascular smooth muscle without any untoward effects. A continued evaluation of NO production and effects in the fetus may not only lead to new therapeutic approaches but shed light on neonatal and adult pathophysiology.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received March 4, 1996; revision received April 9, 1996; accepted May 6, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Richard, J. Gao, B. LaFleur, B. W. Christman, J. Anderson, N. Brown, and J. Reese Patency of the preterm fetal ductus arteriosus is regulated by endothelial nitric oxide synthase and is independent of vasa vasorum in the mouse Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2004; 287(3): R652 - R660. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Morin, S. M. Karr, R. A. Faris, and P. A. Gruppuso Developmental variability in expression and regulation of inducible nitric oxide synthase in rat intestine Am J Physiol Gastrointest Liver Physiol, August 1, 2001; 281(2): G552 - G559. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Coceani, Y.-A. Liu, E. Seidlitz, L. Kelsey, T. Kuwaki, C. Ackerley, and M. Yanagisawa Endothelin A receptor is necessary for O2 constriction but not closure of ductus arteriosus Am J Physiol Heart Circ Physiol, October 1, 1999; 277(4): H1521 - H1531. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. S. Miller and M. Sandoval III. A molecular prelude to intestinal inflammation Am J Physiol Gastrointest Liver Physiol, April 1, 1999; 276(4): G795 - G799. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Anwar, M. Schwab, L. Poston, and P. W. Nathanielsz Betamethasone-mediated vascular dysfunction and changes in hematological profile in the ovine fetus Am J Physiol Heart Circ Physiol, April 1, 1999; 276(4): H1137 - H1143. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Rairigh, L. Storme, T. A. Parker, T. D. le Cras, J. P. Kinsella, M. Jakkula, and S. H. Abman Inducible NO synthase inhibition attenuates shear stress-induced pulmonary vasodilation in the ovine fetus Am J Physiol Lung Cell Mol Physiol, March 1, 1999; 276(3): L513 - L521. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. Smolich NO modulates fetoplacental blood flow distribution and whole body oxygen extraction in fetal sheep Am J Physiol Regulatory Integrative Comp Physiol, May 1, 1998; 274(5): R1331 - R1337. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. C. S. Smith The Pharmacology of the Ductus Arteriosus Pharmacol. Rev., March 1, 1998; 50(1): 35 - 58. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Challah, S. Nadaud, M. Philippe, T. Battle, F. Soubrier, B. Corman, and J. B. Michel Circulating and cellular markers of endothelial dysfunction with aging in rats Am J Physiol Heart Circ Physiol, October 1, 1997; 273(4): H1941 - H1948. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Takizawa, H. Yoshikawa, M. Yamada, and H. Morita Expression of nitric oxide synthase isoforms and detection of nitric oxide in rat placenta Am J Physiol Cell Physiol, April 1, 2002; 282(4): C762 - C767. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Rairigh, T. A. Parker, D. D. Ivy, J. P. Kinsella, I-D. Fan, and S. H. Abman Role of Inducible Nitric Oxide Synthase in the Pulmonary Vascular Response to Birth-Related Stimuli in the Ovine Fetus Circ. Res., April 13, 2001; 88(7): 721 - 726. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |