Scott W. Walsh

Antioxidant activities and mRNA expression of superoxide dismutase, catalase, and glutathione peroxidase in normal and preeclamptic placentas.

Objective: Placental production of lipid peroxides is abnormally increased in preeclampsia. The reason for this is not known, but if placental antioxidant enzymes were deficient, lipid peroxides would increase unchecked. In this study, we measured 1) enzyme activities of superoxide dismutase (CuZn-SOD), catalase, and glutathione peroxidase (GSH-Px) and tissue levels of vitamin E, and 2) mRNA expression of CuZn-SOD, catalase, and GSH-Px in normal and preeclamptic placentas. Methods: Placental tissues were obtained from normal (n = 16) preeclamptic (n = 12) pregnancies immediately after delivery. Tissue pieces were frozen in liquid nitrogen and stored at -80C until assayed. The enzyme activities of CuZn-SOD, catalase, nd GSH-Px and the levels of vitamin E were determined by spectrophotmetric assays. Messenger RNA expression of CuZn-SOD, catalase, and GSH-Px was determined by Northern blot analysis. Results: The activities of CuZn-SOD and GSH-Px and the tissue levels of vitamin E were significantly lower in preeclamptic placentas than in normal placentas (CuZn-SOD: 1.13 ± 0.49 versus 3.71 ± 0.71 U/mg protein, P < .01; GSH-Px: 0.18 ± 0.01 versus 0.26 ± 0.02 U/mg, P < .01; vitamin E: 0.08 ± 0.01 versus 0.18 ± 0.01 μg/mg, mean ± standard error, P < .001), whereas the activity of catalase was significantly higher (93 ± 3 verses 83 ± 2 U/mg, P < .05). Relative mRNA expression of CuZn-SOD and GSH-Px was significantly lower in preeclamptic than normal placentas (P < 0.05), but there was no significant difference for catalase (P > .4). Conclusions: 1) Activities of CuZn-SOD and GSH-Px and the tissue levels of vitamin E are significantly lower in preeclamptic than in normal placentas. 2) Activity of catalase is significantly higher in preeclamptic than in normal placentas. 3) Messenger RNA expression for Cu-Zn-SOD and GSH-Px is lower in preeclamptic placentas. We speculate that decreased antioxidant activity may result in increased lipid peroxide levels in preeclamptic placentas.

Placental lipid peroxides and thromboxane are increased and prostacyclin is decreased in women with preeclampsia

OBJECTIVE There is an imbalance of increased thromboxane and decreased prostacyclin in placentas of women with preeclampsia, but this may not be the only imbalance. There is also an abnormal increase in serum lipid peroxides in preeclamptic women. Lipid peroxides are toxic compounds that damage cells and inhibit prostacyclin synthesis. The following study examined lipid peroxides to determine if they were also increased in placentas of preeclamptic women. STUDY DESIGN Placental tissue for nine normal and eight preeclamptic women were frozen in liquid nitrogen immediately after delivery. Frozen tissue samples (1 gm) were homogenized and analyzed for lipid peroxides by malondialdehyde and hydrogen peroxide equivalents and for thromboxane and prostacyclin by radioimmunoassay of their stable metabolites, thromboxane B2 and 6-keto prostaglandin F1 alpha. RESULTS Lipid peroxides were significantly higher in preeclamptic placentas than in normal placentas by both analytic methods (49 +/- 5 vs 31 +/- 1 nmol/gm for malondialdehyde and 5.3 +/- 0.3 vs. 3.2 +/- 0.3 mumol/gm for hydrogen peroxide equivalent; mean +/- SE; p < 0.01, respectively). Thromboxane was significantly higher and prostacyclin significantly lower in preeclamptic placentas than in normal placentas (213 +/- 23 vs 158 +/- 14 ng/gm for thromboxane and 24 +/- 3 vs 53 +/- 7 ng/gm for prostacyclin, p < 0.05). The thromboxane/prostacyclin and lipid peroxides/prostacyclin ratios were threefold higher in preeclamptic placentas than in normal placentas. CONCLUSION Placental levels of both lipid peroxides and thromboxane are increased and prostacyclin decreased in preeclampsia. We speculate that abnormally increased levels of lipid peroxides in preeclamptic placentas may be a cause of decreased prostacyclin.

Placental mitochondria as a source of oxidative stress in pre-eclampsia

Pre-eclampsia is a hypertensive disorder of human pregnancy that is a leading cause of premature delivery and fetal growth retardation. It is characterized by hypertension, reduced uteroplacental blood flow, proteinuria and oedema. Pre-eclampsia is associated with increased lipid peroxidation in the maternal circulation and in the placenta. Mitochondria are sources of oxygen radicals and are enriched with polyunsaturated fatty acids that are susceptible to peroxidation. Therefore, the mitochondria could be an important source of oxidative stress and lipid peroxidation. To study this, the level of lipid peroxidation in the mitochondrial fraction of placentae obtained from normally pregnant women (n=8) and women with pre-eclampsia (n=8) was examined. Placental tissues were homogenized and the mitochondrial fraction was isolated by ultracentrifugation. Mitochondrial lipid peroxides were estimated by malondialdehyde (MDA). NADPH and Fe++ were used to stimulate lipid peroxidation. Superoxide dismutase (SOD) was used to inhibit superoxide radicals and mannitol to inhibit hydroxyl radicals. The following results were found: (1) MDA levels were significantly greater in the mitochondrial fraction isolated from pre-eclamptic placentae than from normal placentae (27.4+/-3.0 versus 17.0+/-1.8 nmol/g tissue, mean+/-s.e., P<0.05); (2) the oxidative potential of the pre-eclamptic mitochondrial fraction was also higher than normal as evidenced by the significantly greater stimulation of lipid peroxidation by NADPH and Fe+ + (248+/-25 versus 164+/-35 nmol/g, P<0.05); (3) superoxide dismutase, but not mannitol, attenuated the lipid peroxidation induced by NADPH and Fe+ + demonstrating that superoxide is the radical responsible for mitochondrial lipid peroxidation in this system; and (4) the amount of mitochondrial protein was 47 per cent greater and the activity of the mitochondrial enzyme, citrate synthase, was 56 per cent greater in the pre-eclamptic placentae indicating an increase in the amount of mitochondria in the pre-eclamptic placentae. It is concluded that: (1) mitochondrial lipid peroxidation is increased in pre-eclampsia; (2) the amount of placental mitochondria is increased in pre-eclampsia; (3) placental mitochondria contribute to the abnormal increase in lipid peroxidation that occurs in pre-eclamptic placentae by both an increase in their amount and an increase in their susceptibility to oxidation; and (4) mitochondrial generation of superoxide could be an important source of oxidative stress in pre-eclampsia.

TNFα concentrations and mRNA expression are increased in preeclamptic placentas

Abstract Tumor necrosis factor-α (TNFα), a cytokine produced mainly by macrophages, is involved in immunoregulation, the modulation of cell growth and differentiation, as well as in the induction of oxygen free radicals. In preeclamptic placentas, lipid peroxides are increased as compared to normal placentas. If TNFα was abnormally increased in preeclamptic placentas, it might contribute to the increased oxidative stress and the formation of lipid peroxides. We hypothesized that TNFα levels would be higher in preeclamptic than in normal placentas. We determined: (1) the levels of TNFα protein in whole placental tissue; (2) the concentrations of TNFα protein and lipid peroxides in the medium after 48 h of incubation of whole placental villi; (3) mRNA expression of TNFα. Placental TNFα protein levels were measured by ELISA, lipid peroxides by a spectrophotometric method specific for peroxides, and TNFα mRNA expression by RT-PCR. Results: (1) TNFα tissue levels were significantly higher in preeclamptic than in normal placentas, P P . (3) TNFα concentrations in the incubation medium were positively correlated with lipid peroxide concentrations, r = 0.608. (4) TNFα mRNA was expressed in preeclamptic placentas, but not in normal placentas. Conclusions: TNFα protein concentrations and mRNA expression are higher in preeclamptic than normal placentas and this is associated with increased lipid peroxidation.

LIPID PEROXIDATION IN PREGNANCY

Lipid peroxides and oxygen radicals are highly reactive and very damaging compounds. In normal pregnancy lipid peroxides increase, but antioxidants also increase to offset their toxic actions. However, this is not the case in preeclampsia. In women with preeclampsia, circulating levels of lipid peroxides are increased, but net antioxidant activity is decreased as compared to normally pregnant women. A source of circulating levels of lipid peroxides in pregnancy is the placenta because the placenta produces and secretes lipid peroxides, and lipid peroxide levels decrease after delivery of the placenta. Other sources are activated neutrophils, radical-initiated propagation, and self-propagation. Although some lipid peroxides are unstable, those present in oxidized low-density lipoproteins have a half-life of 3 h in die circulation and so function as circulating compounds. In preeclampsia, placental production of lipid peroxides is significantly increased, and this is correlated with significantly increased ...

Deficient glutathione peroxidase activity in preeclampsia is associated with increased placental production of thromboxane and lipid peroxides

OBJECTIVE Thromboxane and lipid peroxide levels are abnormally increased in preeclamptic placentas. The cause of this increase is not known. Peroxides stimulate prostaglandin H2 synthase to increase thromboxane and oxygen radicals, which increase lipid peroxides. Glutathione peroxidase inactivates peroxides, thereby decreasing peroxide stimulation of prostaglandin H synthase. If glutathione peroxidase activity were deficient, then peroxides could increase, leading to increased stimulation of prostaglandin H synthase, resulting in increased production of thromboxane and lipid peroxides. The following study tested this hypothesis. STUDY DESIGN Placental tissues from 11 normal and 11 preeclamptic women were immediately frozen in liquid nitrogen after delivery. One gram of tissue from each placenta was homogenized for analysis. Placental tissues were also obtained from six normal pregnancies for incubation with a glutathione peroxidase inhibitor, N-ethylmaleimide. Samples were analyzed for glutathione peroxidase activity, lipid peroxides by hydrogen peroxide equivalents, thromboxane by thromboxane B2, and prostacyclin by 6-keto-prostaglandin F1 alpha. RESULTS Glutathione peroxidase activity was significantly lower in preeclamptic than in normal placentas (9.41 +/- 0.05 vs 13.41 +/- 0.63 units/gm, p < 0.001, mean +/- SE). Lipid peroxides and thromboxane were significantly higher in preeclamptic than in normal placentas (hydrogen peroxide equivalents 4.23 +/- 0.32 vs 2.84 +/- 0.27 mumol/gm, p < 0.01; thromboxane B2 215 +/- 31 vs 138 +/- 15 ng/gm, p < 0.05), whereas prostacyclin was significantly lower (6-keto-prostaglandin F1 alpha 23.3 +/- 2.2 vs 41.6 +/- 6.0 ng/gm, p < 0.01). Inhibition of glutathione peroxidase activity in normal placentas resulted in a dose-response increase in placental production of both lipid peroxides and thromboxane without affecting prostacyclin, so the ratio of thromboxane to prostacyclin increased progressively with inhibition of glutathione peroxidase. CONCLUSIONS Glutathione peroxidase activity is significantly lower, and lipid peroxides and thromboxane are significantly higher, in preeclamptic placentas compared with normal placentas. Inhibition of glutathione peroxidase activity in normal placentas results in significantly increased production of lipid peroxides and thromboxane and an increase in the thromboxane to prostacyclin ratio. We speculate that in normal placentas, glutathione peroxidase limits prostaglandin H synthase activity by reducing the amount of peroxide present, thus reducing peroxide stimulation of prostaglandin H synthase. In preeclampsia deficient glutathione peroxidase activity results in in increased peroxide levels leading to increased stimulation of prostaglandin H synthase, which results in increased production of lipid peroxides and thromboxane.

Secretion of lipid peroxides by the human placenta

OBJECTIVE We attempted to determine whether the human placenta secretes lipid peroxides. If it does, then it could be a source of lipid peroxides in maternal blood. STUDY DESIGN In study 1 isolated human placental cotyledons (n = 7) were perfused serially for 20-minute intervals with control Krebs-Ringer-bicarbonate buffer gassed with 95% oxygen and 5% carbon dioxide and Krebs-Ringer-bicarbonate buffer with progressively increasing concentrations of t-butyl hydroperoxide added (10, 25, 50, and 100 mumol/L) to stimulate endogenous lipid peroxide production. In study 2 placental cotyledons (n = 6) were perfused serially for 20-minute intervals with control Krebs-Ringer-bicarbonate buffer, t-butyl hydroperoxide (100 mumol/L), low-dose aspirin (5 x 10(-5) mol/L), and low-dose aspirin plus t-butyl hydroperoxide. Maternal and fetal effluent samples were analyzed for lipid peroxides by hydrogen peroxide equivalents. RESULTS In study 1, compared with control Krebs-Ringer-bicarbonate perfusion, peroxide perfusion significantly increased, in a dose-response manner, placental lipid peroxide secretion. In study 2, aspirin completely blocked the ability of peroxide to increase the secretion of lipid peroxides. In both studies placental secretion of lipid peroxides was significantly greater toward the maternal side of the placenta than toward the fetal side. CONCLUSIONS (1) The human placenta secretes lipid peroxides primarily into the maternal effluent. (2) Exogenous peroxide stimulates endogenous lipid peroxide production, which is blocked by aspirin, suggesting cyclooxygenase is involved in lipid peroxide production. (3) The placenta could be a source of circulating lipid peroxides in pregnant women.

Maternal levels of prostacyclin, thromboxane, vitamin E, and lipid peroxides throughout normal pregnancy

In normal pregnancy the vasodilating actions of prostacyclin and the antioxidant activity of vitamin E are important for normal physiologic function. Thromboxane and lipid peroxides oppose these actions by promoting vasoconstriction and peroxidation reactions, respectively. An imbalance between thromboxane and prostacyclin and between lipid peroxides and antioxidant activity is implicated in pathologic states such as preeclampsia. We hypothesized that in normal pregnancy there would be a balance in the ratios of prostacyclin to thromboxane and of vitamin E to lipid peroxides that would favor prostacyclin and vitamin E. Blood samples were collected from normally pregnant women throughout gestation and analyzed for prostacyclin, thromboxane, vitamin E, and lipid peroxides. Serum levels of lipid peroxides remained relatively stable throughout gestation, but the levels of vitamin E progressively increased. Plasma levels of prostacyclin progressively increased with advancing gestation, whereas levels of thromboxane progressively decreased. Therefore the ratios of both prostacyclin/thromboxane, and vitamin E/lipid peroxides progressively increased during pregnancy. The increase in the ratios was highly correlated, r = 0.94. We conclude that the changes in the maternal concentrations of these compounds and the progressive increase in the ratios of prostacyclin/thromboxane and vitamin E/lipid peroxides suggest that the vasodilating actions of prostacyclin and the antioxidant activity of vitamin E are progressively favored with advancing gestation in normally pregnant women.

Obesity: a risk factor for preeclampsia.

Obesity is becoming an epidemic worldwide. Among young women, obesity is especially important because maternal obesity confers an increased risk of preeclampsia, a hypertensive disorder of pregnancy and a leading cause of maternal and fetal morbidity and mortality. It is not known why obesity is a risk factor for preeclampsia, but these conditions might be related through common features related to oxidative stress, inflammation and altered vascular function. Recently, extensive vascular infiltration of neutrophils and vascular inflammation has been reported in both preeclamptic women and obese women. Therefore, if the vasculature of obese women is inflamed, they could be at increased risk of developing preeclampsia when they become pregnant and are exposed to the additional burdens of pregnancy.

Neutrophils Infiltrate Resistance-Sized Vessels of Subcutaneous Fat in Women With Preeclampsia

We examined if there is systemic vascular inflammation and neutrophil infiltration in women with preeclampsia. Resistance-sized vessels (10 to 200 μm) of subcutaneous fat were evaluated from normal nonpregnant women, normal pregnant women, and preeclamptic women. Immunohistochemical staining was performed for: (1) interleukin-8 (IL-8), a potent neutrophil chemokine; (2) intercellular adhesion molecule-1 (ICAM-1; CD54), an endothelial cell adhesion molecule; and (3) CD66b, a neutrophil antigen. Vessels of preeclamptic patients had intense IL-8 staining in the endothelium and vascular smooth muscle, as compared with little or no staining for normal pregnant and normal nonpregnant patients. ICAM-1 was expressed on the endothelium of all patient groups. In preeclamptic patients, ICAM-1 was also expressed on vascular smooth muscle. Vessels of preeclamptic patients had significantly more CD66b staining of neutrophils than did normal pregnant or normal nonpregnant patients. There were significantly more vessels stained, more vessels with neutrophils flattened and adhered to endothelium, more vessels with neutrophils infiltrated into the intima, and more neutrophils per vessel. In conclusion, in women with preeclampsia, there was significant infiltration of neutrophils into maternal systemic vasculature associated with inflammation of the vascular smooth muscle indicated by increased expression of IL-8 and ICAM-1. Neutrophil infiltration provides a reasonable explanation for endothelial and vascular smooth muscle dysfunction in preeclampsia because neutrophils produce toxic substances, which may explain clinical symptoms.