| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2004;43:614.)
© 2004 American Heart Association, Inc.
Scientific Contributions |
From Mario Negri Institute for Pharmacological Research (M.N., M.T., P.C., S.B., D.M., R.M., F.Po., A.B., G.R.) Bergamo, Italy; Unit of Obstetrics and Gynecology (F.P., A.C., C.P.), Ospedale San Gerardo, Milano, Italy; and Unit of Nephrology and Dialysis (G.R.), Azienda Ospedaliera, Ospedali Riuniti di Bergamo, Bergamo, Italy.
Correspondence to Dr Marina Noris, Mario Negri Institute for Pharmacological Research, Via Gavazzeni, 11, 24125 Bergamo, Italy. E-mail noris{at}marionegri.it
| Abstract |
|---|
|
|
|---|
Key Words: arginine nitric oxide synthase nitrites oxidative stress preeclampsia
| Introduction |
|---|
|
|
|---|
Unlike in normal pregnancy, in preeclampsia the cytotrophoblast fails to adopt a vascular adhesion phenotype,10 which compromises the supply of blood flow to the maternal-fetal interface. It has been hypothesized that a reduced formation of NO could account for abnormal placental perfusion in preeclampsia.1,11 Most data indicate that preeclamptic placenta has a normal capacity to synthesize ecNOS as compared with normal placenta and the formation of NO is also comparable in preeclamptic and in normal placenta.1,2 However, one study12 found that inhibiting NO synthesis raised the perfusion pressure of isolated human cotyledon preparations from normal pregnant women, but not from preeclamptic women. Moreover, in preeclampsia the concentration of the NO second messenger, cGMP, in the placental circulation was lower than normal.1,13 Thus, preeclampsia is a condition of normal placental expression of ecNOS and normal generation of NO, whose activity, however, is abnormally reduced. This apparent inconsistency can be reconciled considering that the relative activity of NO in a given organ/tissue depends on the rates of synthesis and degradation. Beside NO, NOS enzymes also generate superoxide anion (O2-) and the rate of NO versus O2- formation and disposal is closely regulated by intracellular levels of L-arginine, tetrahydrobiopterin, and superoxide dismutase (SOD).1416 NO, by interacting with O2-, forms peroxynitrite, a cytotoxic anion that inhibits mitochondrial electron transport, oxidizes proteins, initiates lipid peroxidation, and nitrates aromatic amino acids.17 Markers of oxidative stress are increased in placenta18 of women with preeclampsia and nitrotyrosine staining, a marker of peroxynitrite, has been found in the preeclamptic placenta.18,19
The present study was designed: (1) to test whether lower placental NO bioavailability of preeclampsia is caused more by increased NO degradation to peroxynitrite than to low ecNOS expression or activity and (2) to investigate whether excessive peroxynitrite formation is related to a low bioavailability of the NO precursor L-arginine and to find out the biochemical determinants of that.
| Methods |
|---|
|
|
|---|
140 mm Hg, and no significant proteinuria. Four women underwent premature delivery for ruptured membranes (n=2), oligohydramnios (n=1), and fetal distress (n=1). According to previously published criteria,20 preeclampsia was diagnosed as increase in diastolic BP of 15 mm Hg and systolic BP of 30 mm Hg at two measurements at least 4 hours apart compared with BP obtained before 20 weeks of gestation, proteinuria >0.3 g/24 hours in the absence of urinary tract infection, return to normal BP, resolution of proteinuria by 12 weeks postpartum, and edema. One preeclamptic pregnancy was complicated by the HELLP syndrome and two were complicated by intrauterine growth restriction (IUGR). The patients clinical characteristics are summarized in Table 1.
|
Tissue Collection and Histology
For placenta tissue collection, see the online data supplement available at http://www.hypertensionaha.org.
Placenta pathology score was graded according to the presence (in at least 30% of the tissue) of each of the following abnormalities: intervillous thrombosis, subchorionic thrombosis, infarcts, chronic hypoxia, and intima hyperplasia, according to the guidelines of the College of American Pathologists (Table 1).21
NADPH-Diaphorase, Immunoperoxidase, and Western Blotting
NADPH-diaphorase reaction was performed on 7-µm thick frozen sections, as described.22 Immunoperoxidase for ecNOS, nitrotyrosine, HNE-lysine, and arginase was performed on 3-µm paraffin sections.22 Western blot analysis for arginase II was performed in placenta homogenates. An expanded Methods section can be found in online data supplement available at http://www.hypertensionaha.org.
NOS Activity and L-arginine Levels in Placenta Tissue
Homogenates of total placentas or of villous and decidua tissues from NP (n=9) and preeclamptic women (n=6) were used to evaluate NOS activity and L-arginine levels, respectively (see online data supplement available at http://www.hypertensionaha.org).
Conjugated Dienes
Total lipids extracted from frozen tissues were resuspended in cyclohexane and the absorbance read at 233 nm. The extinction coefficient of 2.52x104 mol/L-1cm-1 was used to calculate the diene concentration.23
Plasma L-arginine and Nitrites/Nitrates
(NO2-/NO3-)
Plasma was obtained from maternal (antecubital vein) and fetal blood (umbilical vein) collected on heparin and was stored at -80°C until assayed. Women fasted at least 8 hours before blood collection. In preliminary experiments, we found that in this condition, the interference of dietary intake to NO2-/NO3- plasma levels is negligible (not shown). Plasma L-arginine was measured by HPLC.24 NO2-/NO3- plasma levels were measured as reported in online data supplement http://hypertensionaha.org.
Real-Time Quantitative RT-PCR
To analyze gene expression of the cationic amino acid transporters CAT-1,25 4F2hc,25 LAT-1,25 and arginase II,26 and of the housekeeping gene ß-actin, RNA was treated with DNase and reverse transcribed to cDNA. Quantitative real-time polymerase chain reaction (PCR) was performed on a TaqMan ABI PRISM 5700 Sequence Detection System (PE Applied Biosystems, Monza, Italy) with SYBR Green PCR Core Reagents (Applied Biosystems), in combination with optimal primer concentrations. Detailed methods are available in the online data supplement http://www.hypertensionaha.org.
Statistical Analysis
Results are means±SEM. Groups were compared by 1-way ANOVA using the StatView 4.01 software. Linear regression analysis was used to correlate immunoperoxidase and mRNA expression data with clinical and experimental parameters. The
Spearman Rank test was used to correlate immunoperoxidase data with the placenta pathology score and Apgar score. The statistical level of significance of the two-tailed test was defined as P<0.05.
| Results |
|---|
|
|
|---|
Comparable diaphorase staining, NOS activity, and ecNOS expression were seen in placentas from NP and preeclampsia (online Figure I and Table 2). Detailed results and online Figure I are in online data supplement available at http://www.hypertensionaha.org.
|
|
Exuberant Peroxynitrite Formation in Preeclampsia
Nitrotyrosine staining was absent or very faint in villi and decidua vessels of placentas from NP (Figure 1A, B), whereas it was intense in preeclamptic villous tissue, mainly localized in the syncytiotrophoblast (Figure 1C and Table 2). The endothelium of villous vessels also showed moderate nitrotyrosine staining. A faint nitrotyrosine staining was also seen in the vessels of the decidua in preeclamptic placenta (Figure 1D and Table 2), whereas no staining was found in the cells of the stroma. Nitrotyrosine and ecNOS staining colocalized in preeclamptic villi (Figure 1E, F), indicating that peroxynitrite was formed within the very cells that synthesized NO. Moderate specific ecNOS staining was also found in villous vascular endothelium (Figure 1G). No signal was found in negative control (Figure 1H).
The mean gestational age of NP was significantly greater than in the preeclamptic group. However, we noted no differences in nitrotyrosine staining in placental tissue with differing gestational age in either group. The intensity of nitrotyrosine staining in pre-term NP (n=4) was not different from NP at term (n=9) (syncytiotrophoblast: 0.55±0.29 versus 0.28±0.10; endothelium of villi: 0.38±0.38 versus 0.16±0.10), but was significantly weaker (P<0.05) than in preeclamptic villi. An inverse correlation was found between nitrotyrosine staining in the syncytiotrophoblast and birth weight (r=-0.63, P=0.01) and Apgar score (
=-0.47, P<0.05), whereas a positive correlation was found between nitrotyrosine and placenta pathology score (
=0.48, P<0.05).
Excess Lipid Peroxidation in the Preeclamptic Placenta
As compared with NP, a larger amount of conjugated dienes was extracted from preeclamptic villi (NP: 24±1; preeclampsia: 47±3 pmol/µg phospholipid; P<0.0001) and umbilical cord tissue (NP: 263±27; preeclampsia: 497±54 pmol/µg phospholipid; P<0.001). Dienes concentration in preeclamptic deciduas was also increased, although to a lesser degree (NP: 22±2; preeclampsia: 36±4 pmol/µg phospholipid; P<0.05). Within the NP group, the amount of conjugated dienes was comparable for preterm and for term pregnancies, indicating that placental lipid peroxidation does not depend on the week of gestation.
4-HNE-lysine staining, a marker of cell oxidative stress,27 was mostly faint in the syncytiotrophoblast and villous vessels (Figure 2A) as well as in the endothelium of NP decidua vessels (Figure 2B). A higher 4-HNE-lysine staining intensity was found in the syncytiotrophoblast and in endothelial cells of preeclamptic villous vessels than in NP villi (Figure 2C, Table 2). Moderate staining was also focally found on preeclamptic decidua vessels (Figure 2D and Table 2). 4-HNE-lysine and nitrotyrosine staining showed the same cell localization (Figure 2E and F), and the intensity of 4-HNE-lysine staining in the syncytiotrophoblast (r=0.79, P<0.0001) and in villous endothelium (r=0.64, P<0.005) correlated with the intensity of nitrotyrosine staining, confirming the association between lipid peroxidation and peroxynitrite formation. No signal was found in negative control (Figure 2G). The intensity of 4-HNE-lysine staining in villous tissue was inversely correlated with birth weight (syncytiotrophoblast: r=-0.69, P<0.001, endothelium: r=-0.58, P<0.01) and Apgar score (syncytiotrophoblast:
=-0.61, P<0.01, endothelium:
=-0.62, P<0.01), whereas a positive correlation was found between 4-HNE-lysine staining and placenta pathology score (syncytiotrophoblast:
=0.54, P<0.05).
|
Immunostaining for CuZn-SOD showed no differences in intensity and localization in either villous tissue or decidua vessels between NP and preeclamptic placentas (Table 2).
Fetal-Maternal Gradient of L-arginine and
NO2-/NO3- Is Lost in Preeclampsia
In line with previous data,28 the mean concentration of L-arginine in maternal blood from NP was significantly lower than in nonpregnant women (Figure 3A). The same was true in preeclampsia (Figure 3A). No significant differences were observed in maternal plasma L-arginine in NP and preeclampsia (Figure 3A). In NP, L-arginine was significantly higher in fetal than in maternal blood (Figure 3A), which is in line with the existence of an active placental L-arginine transport from the mother to the fetus.25 By contrast, in preeclampsia, fetal L-arginine concentrations were almost identical to those in maternal blood (Figure 3A) and significantly lower than NP (Figure 3A). As previously reported,28 we found no correlation between maternal (r=0.03, P=0.93) or fetal (r=0.16, P=0.6) L-arginine concentrations and gestational age, thus excluding that the lower fetal L-arginine in preeclampsia depends on shorter gestation.
|
Consistent with L-arginine data, in NP the concentration of the NO metabolites, NO2-/NO3-, in the fetal blood was significantly higher than in the maternal blood (Figure 3B). In preeclampsia, NO2-/NO3- concentration in maternal blood was normal whereas levels in fetal blood were significantly (P<0.05) lower than in NP (Figure 3B). An inverse correlation was found between fetal blood L-arginine concentration and the intensity of nitrotyrosine staining in the syncytiotrophoblast (framed panel at upper right of Figure 3A) and in the endothelium of villous vessels (r=-0.6, P<0.05).
Expression of Cationic Amino Acid Transporters Is Not Reduced in
Preeclamptic Placenta
To search for the possible causes of loss of fetal-maternal L-arginine and NO2-/NO3- gradients in preeclampsia, we evaluated gene expression of two cationic amino acid transporters highly expressed in human placenta, CAT-1 (y+system transporter)25 and 4F2hc and LAT-1 (the 2 subunits of the y+L system transporter).25 As shown in Figure 4A, CAT-1 expression in villous tissue was higher in preeclampsia than in NP. Expression of CAT-1 in decidua and of 4F2hc and LAT-1 (Figure 4A, 4B) in villous tissue and decidua was comparable in NP and in preeclampsia.
|
Excess Expression of Arginase II in Preeclamptic Placenta
We then looked for placental expression of arginase IIthe extrahepatic isoform of arginase that degrades arginine into ornithine and urea.26 Real-time RT-PCR analysis of villous tissue specimens showed a higher expression of arginase II in preeclampsia than in NP, whereas arginase II expression in decidua tissue was comparable in the two groups (Figure 5A). The expression of arginase II in preterm NP villous tissue (3.08±1.32) was not different from arginase II in villous tissue from NP at term (4.08±1.05), indicating that placental arginase II expression does not depend on the week of gestation. The levels of arginase II mRNA in villous tissue were inversely correlated with fetal L-arginine concentrations (r=-0.8, P<0.005), indicating that arginase II expressed in placenta may modulate L-arginine availability to the fetus. Immunostaining and Western blot for arginase II was performed to evaluate tissue localization and protein expression. Arginase II staining was mostly faint in chorionic villi and in decidua from NP (Figure 5B, Table 2). An intense arginase II staining was instead found in the syncytiotrophoblast and in endothelial cells of villous vessels of preeclamptic placenta (Figure 5C, Table 2). A moderate staining was also focally found in preeclamptic decidua vessels (Table 2); however, mean values were not significantly different from those recorded in NP. No signal was found in negative control (Figure 5D). Consistent with immunostaining results, Western blot analysis of placenta homogenates showed a faint immunoreactive band at 39 kDa in NP whereas a strong band was found in preeclampsia (Figure 5E).
|
To verify whether increased arginase II in the preeclamptic placenta limits L-arginine availability, L-arginine tissue levels were evaluated. As shown in Figure 5F, L-arginine levels in preeclamptic villi were lower (P<0.05) than levels in villi from NP, whereas L-arginine levels in the decidua were comparable in the two groups.
| Discussion |
|---|
|
|
|---|
The relevance of this biochemical pathway to the pathogenesis of preeclampsia rests on present and previous28 findings that in NP the L-arginine concentration is higher in fetal than in maternal blood and that the fetal-maternal L-arginine gradient was completely lost in preeclampsia. This abnormality was not because of a defect in L-arginine transporters within the syncytiotrophoblast25 because expression of y+ and y+L system cationic transporters was normal or even increased in preeclamptic placenta. Based on the observation that L-arginine levels in preeclamptic villi were lower than in NP villi, we then addressed the possibility that L-arginine deficiency was caused by consumption of the amino acid in the preeclamptic villous tissue by arginase II, the extrahepatic isoform of arginase expressed in human placenta.26 Both arginase and NOS use arginine as common substrate and arginase inhibits NO synthesis by reducing arginine bioavailability. Transfection of endothelial cells with arginase II increased L-arginine consumption and reduced NO synthesis26 and in vivo administration of arginase to experimental animals significantly depleted plasma L-arginine.32 Arginase II mRNA expression was more than 4-fold higher in preeclamptic than in NP villous tissue; also, protein expression was increased, as documented by immunostaining and Western blot data. In addition, levels of arginase II mRNA in villous tissue inversely correlated with fetal L-arginine concentration. We hypothesize that in preeclamptic placenta, higher than normal expression of arginase II causes less L-arginine available for ecNOS in trophoblast cells and in the villous endothelium, which can lead to the aberrant catalytic activity described, giving a rapid NO degradation by reactive oxygen species. This possibility is also supported by finding in our patients that the intensity of nitrotyrosine and 4-HNE-lysine staining in villous tissue and fetal L-arginine concentration were inversely correlated. Deficiency in tetrahydrobiopterin (BH4) can also poise ecNOS toward producing superoxide anion;33 however, levels of BH4 were found to be normal in preeclamptic placentas.33
Further investigation is required to clarify the mechanism responsible for arginase II upregulation in preeclampsia. Testosterone could be a possible candidate mediator of the aforementioned phenomenon. Indeed in studies in female rats and mice, testosterone stimulates arginase activity and patients with preeclampsia have been shown to have higher levels of testosterone than NP women.11
Previous reports1 of higher resistance in the fetal-placental circulation in the preeclamptic placenta are consistent with the reduced placental NO availability we observed, because NO is a potent vasodilator in the vascular district.1 In addition, there is evidence of a major role for NO as an angiogenic and vascular remodelling factor.69 Exposure of human endothelial cells to NO donors leads to a dose-dependent increase in endothelial cell migration and differentiation,34 and NO is a mediator of growth factor-induced angiogenesis.6,34,35 That NO may be instrumental to endovascular invasion and vessel remodeling of developing placenta is suggested by a number of experimental findings. First, NO release is coupled to VEGF and hepatocyte growth factor (HGF)-induced trophoblast invasion and motility.36,37 Second, in trophoblast cells, NO upregulates the expression and the activity of the matrix-degrading proteases MMP-2 and MMP-9, which are required for invasion during embryo implantation.38 Finally, NO causes dilation of the uteroplacental arteries, which is another prerequisite for trophoblast invasion and remodeling of the endothelium.1 It is tempting to speculate that low NO availability could contribute to impaired cytotrophoblast invasion in preeclampsia, a hypothesis that should be formally tested in developing placenta obtained before 20 weeks of gestation. However, the present study, as most previously published reports on preeclamptic placentas10,18,19 have been necessarily performed at the time of disease onset, usually in the third trimester of pregnancy, represents a limitation that cannot be easily overcome.
In conclusion, our findings indicate that normal placenta with enough tissue L-arginine sustains adequate generation of NO by ecNOS. By contrast, in preeclampsia, when the placental L-arginine concentration is low because of excessive arginase II expression, activation of ecNOS leads to excessively high generation of superoxide anion, which reduces NO half-life by forming peroxynitrite. This may promote microvascular oxidative damage and favor abnormal placenta perfusion.
Perspectives
The present findings provide a rationale for clinical trials with L-arginine or antioxidant supplementation39 that, either by providing more substrate to ecNOS or by lowering the rate of NO degradation to peroxynitrite, could help prevent this disease, which remains one of the leading causes of maternal and fetal morbidity and mortality.
| Acknowledgments |
|---|
Received August 25, 2003; first decision September 17, 2003; accepted December 24, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Lopez-Jaramillo, W. D. Arenas, R. G. Garcia, M. Y. Rincon, and M. Lopez Review: The role of the L-arginine-nitric oxide pathway in preeclampsia Therapeutic Advances in Cardiovascular Disease, August 1, 2008; 2(4): 261 - 275. [Abstract] [PDF] |
||||
![]() |
B. D. LaMarca, J. Gilbert, and J. P. Granger Recent Progress Toward the Understanding of the Pathophysiology of Hypertension During Preeclampsia Hypertension, April 1, 2008; 51(4): 982 - 988. [Full Text] [PDF] |
||||
![]() |
J. Belik, D. Shehnaz, J. Pan, and H. Grasemann Developmental changes in arginase expression and activity in the lung Am J Physiol Lung Cell Mol Physiol, March 1, 2008; 294(3): L498 - L504. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Santhanam, H. K. Lim, H. K. Lim, V. Miriel, T. Brown, M. Patel, S. Balanson, S. Ryoo, M. Anderson, K. Irani, et al. Inducible NO Synthase Dependent S-Nitrosylation and Activation of Arginase1 Contribute to Age-Related Endothelial Dysfunction Circ. Res., September 28, 2007; 101(7): 692 - 702. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ckless, A. van der Vliet, and Y. Janssen-Heininger Arginase Modulates NF-{kappa}B Activity via a Nitric Oxide-Dependent Mechanism Am. J. Respir. Cell Mol. Biol., June 1, 2007; 36(6): 645 - 653. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Belkacemi, S. A. Bainbridge, M. A. Dickinson, G. N. Smith, and C. H. Graham Glyceryl Trinitrate Inhibits Hypoxia/Reoxygenation-Induced Apoptosis in the Syncytiotrophoblast of the Human Placenta: Therapeutic Implications for Preeclampsia Am. J. Pathol., March 1, 2007; 170(3): 909 - 920. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. F. Schwartz, T. Chernichovsky, D. Hagin, M. Ingbir, R. Reshef, G. Chernin, Y. Levo, and D. Schwartz Differential regulation of L-arginine transporters (cationic amino acid transporter-1 and -2) by peroxynitrite in rat mesangial cells Nephrol. Dial. Transplant., December 1, 2006; 21(12): 3409 - 3414. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Topal, A. Brunet, L. Walch, J.-L. Boucher, and M. David-Dufilho Mitochondrial Arginase II Modulates Nitric-Oxide Synthesis through Nonfreely Exchangeable L-Arginine Pools in Human Endothelial Cells J. Pharmacol. Exp. Ther., September 1, 2006; 318(3): 1368 - 1374. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. R. White, S. Ryoo, D. Li, H. C. Champion, J. Steppan, D. Wang, D. Nyhan, A. A. Shoukas, J. M. Hare, and D. E. Berkowitz Knockdown of Arginase I Restores NO Signaling in the Vasculature of Old Rats Hypertension, February 1, 2006; 47(2): 245 - 251. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Roxstrom-Lindquist, E. Ringqvist, D. Palm, and S. Svard Giardia lamblia-Induced Changes in Gene Expression in Differentiated Caco-2 Human Intestinal Epithelial Cells Infect. Immun., December 1, 2005; 73(12): 8204 - 8208. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Racasan, B. Braam, H. A. Koomans, and J. A. Joles Programming blood pressure in adult SHR by shifting perinatal balance of NO and reactive oxygen species toward NO: the inverted Barker phenomenon Am J Physiol Renal Physiol, April 1, 2005; 288(4): F626 - F636. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Cherla and E. A. Jaimes Role of L-Arginine in the Pathogenesis and Treatment of Renal Disease J. Nutr., October 1, 2004; 134(10): 2801S - 2806S. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. T. M. Raijmakers, R. Dechend, and L. Poston Oxidative Stress and Preeclampsia: Rationale for Antioxidant Clinical Trials Hypertension, October 1, 2004; 44(4): 374 - 380. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Davison, V. Homuth, A. Jeyabalan, K. P. Conrad, S. A. Karumanchi, S. Quaggin, R. Dechend, and F. C. Luft New Aspects in the Pathophysiology of Preeclampsia J. Am. Soc. Nephrol., September 1, 2004; 15(9): 2440 - 2448. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Germain, G. Valdes, M. C. Romanik, and M. S. Reyes Letter to the Editor: Evidence Supporting a Beneficial Role for Long-Term L-Arginine Supplementation in High-Risk Pregnancies Hypertension, July 1, 2004; 44(1): e1 - e1. [Full Text] [PDF] |
||||