PAF Levels and PAF–AH Activities in Placentas from Normal and Preeclamptic Pregnancies

Placenta (2006), 27, 744e749 doi:10.1016/j.placenta.2005.06.005

PAF Levels and PAF–AH Activities in Placentas from Normal and Preeclamptic Pregnancies Y. Gua, S. A. Burlisona and Y. Wanga,b,* a

Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center, Shreveport, LA 71130, USA; b Department of Molecular and Cellular Physiology, Louisiana State University Health Sciences Center, Shreveport, LA 71130, USA Paper accepted 10 June 2005

Objective: The aim of this study was to determine: (1) platelet-activating factor (PAF) levels and PAFeacetylhydrolase (PAFeAH) activities in normal and preeclamptic placentas; (2) lipid peroxide production by placental tissues stimulated with PAF. Methods: Placentas were obtained immediately after delivery from normal and preeclamptic pregnancies. Tissue pieces were snap frozen in liquid nitrogen and stored at ÿ70 (C. One gram of tissue from each placenta was used for PAF extraction and for total RNA isolation. PAF was measured by PAF [3H] scintillation proximity assay (SPA) system. Trophoblast PAFeAH activity was determined by enzyme-linked immunosorbent assay (ELISA). mRNA expression for PAF receptor was assessed by RNase protection assay (RPA). Normal placental explants were incubated with PAF at concentrations of 1 and 10 mg/ml for 48 h. Lipid peroxide productions of 8-isoprostane and malondialdehyde (MDA) were measured by ELISA and thiobarbituric acid reaction, respectively. Data were presented as mean G SE and analyzed by nonparametric ManneWhitney U test and ANOVA. A p level less than 0.05 was considered statistically significant. Results: (1) The mean tissue level for PAF was elevated, but not statistically different, in preeclamptic (n Z 7) than in normal (n Z 8) placentas, 6.45 G 1.05 versus 4.47 G 0.60 ng/g, p Z 0.42. (2) PAFeAH activity was higher in trophoblasts from preeclamptic (n Z 7) placentas than that in trophoblasts from normal (n Z 8) placentas, 0.69 G 0.16 versus 0.38 G 0.03 mmol/ min/mg protein, p ! 0.05. (3) The relative mRNA expression for PAF receptor was not different between normal (0.70 G 0.08) and preeclamptic (0.76 G 0.13) placental tissues, p Z 0.60. (4) Productions of 8-isoprostane and MDA were not increased in tissues with PAF in culture, 8-isoprostane: 0.37 G 0.09 ng/mg (control) versus 0.32 G 0.09 ng/mg (1 mg/ml) and 0.37 G 0.07 ng/mg (10 mg/ml), p O 0.5; MDA: 0.62 G 0.05 nmol/mg (control) versus 0.68 G 0.04 nmol/mg (1 mg/ml) and 0.69 G 0.03 nmol/mg (10 mg/ml), p O 0.5. Conclusions: Increased PAFeAH activity in trophoblasts may be a compensatory effect to control PAF level in the preeclamptic placenta. The co-existence of PAFeAH and PAF receptor in trophoblasts suggests an autocrine regulation of PAF in these cells to limit PAF and its metabolites within the placenta. Placenta (2006), 27, 744e749 Ó 2005 Elsevier Ltd. All rights reserved. Keywords: PAF; PAFeAH; Placenta; Preeclampsia

INTRODUCTION Platelet-activating factor (PAF) is a phospholipid mediator and has a wide spectrum of biological properties. PAF is produced by a variety of cells including platelets, endothelial cells, and macrophages. In the vascular system, it increases the permeability of monolayer endothelial cells, activates monocytes/ * Corresponding author. Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center, PO Box 33932, Shreveport, LA 71130, USA. Tel.: C1 318 675 5379; fax: C1 318 675 4671. E-mail address: [email protected] (Y. Wang). 0143e4004/$esee front matter

macrophages and polymorphonuclear neutrophils (PMN), and stimulates the contraction of smooth muscle cells [1]. Several lines of evidence suggest that PAF may contribute to the vascular dysfunction in preeclampsia, a hypertensive disorder during human pregnancy. It has been reported that PAF concentrations were increased in the maternal blood in women with preeclampsia compared to those in normal pregnancies [2]. Whereas, PAFeacetylhydrolase (PAFeAH) activity, which degrades PAF, was found to be higher in plasma in women with preeclampsia than those in gestational age-matched normal controls [3]. A study by Ohshige et al. [4] showed that normotensive pregnant women with a reduced PAFeAH activity in the maternal plasma could develop Ó 2005 Elsevier Ltd. All rights reserved.

Gu et al.: PAF Levels and PAFeAH Activities in Placentas

pregnancy-induced hypertension during pregnancy. Therefore, altered PAF metabolite may contribute to the pathophysiology of preeclampsia. Abnormal placental function has been considered as a critical factor for the development of preeclampsia, such as improper trophoblast invasion, increased productions of vasoconstrictor thromboxane (TX) and cytokine tumor necrosis factor a (TNFa), which are correlated with increased oxidative stress in the placenta. However, PAF and PAFeAH levels in placentas from women with preeclampsia have not been investigated. In this study, we measured PAF levels in placental tissues and PAFeAH activities in trophoblast cells isolated from normal and preeclamptic pregnancies. mRNA expression for PAF receptor was also determined. PAF is a bioactive oxidized phospholipid [5]. Increased oxidative stress induced by ischemia and perfusion injury is accompanied by increased PAF generation [6]. PAF can also induce oxidative burst in macrophages [7]. Since increased PAF is associated with increased oxidative stress [1], we further determined if PAF could stimulate lipid peroxide production in the human placenta. We found that PAFeAH activities, but not PAF levels, were higher in placentas from preeclamptic pregnancies than those from normal pregnancies. However, normal placental tissue stimulated with PAF did not increase the production of lipid peroxides in vitro. MATERIALS AND METHODS Human placentas were collected immediately after delivery from normal and preeclamptic pregnancies. Tissue pieces without chorionic plate and decidual membrane were snap frozen in liquid nitrogen and stored at ÿ70 (C or immediately processed for isolation of trophoblast cells or villous tissue culture. Normal pregnancy is defined as a pregnancy with maternal blood pressure !140/90 mmHg, without proteinuria and absence of medical and obstetrical complications. Mild preeclampsia is defined as a maternal blood pressure of 140/ 90 mmHg or higher with presence of proteinuria (300 mg/24 h or O1C) on two separate readings at least 6 h apart and severe preeclampsia is defined as a maternal blood pressure of 160/ 110 mmHg or higher and with positive proteinuria O3C. In this study, all patients in the preeclampsia group were clinically diagnosed as severe preeclampsia except two patients who were diagnosed as mild preeclampsia. Table 1 shows the demographic characteristics in normal and preeclamptic pregnancies, from which the placentas were used in this study. This study was approved by Institutional Review Board (IRB) for Human Research at Louisiana State University Health Sciences Center e Shreveport (LSUHSC-Sh). Measurement of PAF PAF measurement was conducted by scintillation proximity assay (SPA). The PAF [3H] SPA system kit was purchased from Amersham (Amersham International plc., Amersham, Piscataway, NJ). This system combined the use of a high

745

Table 1. Demographic characteristics in normal and preeclamptic pregnancies Normal (n Z 18)

Preeclamptic (n Z 17)

p value

Maternal age

23 G 4

23 G 5

0.91

Racial status White Black Others

5 12 1

4 12 1

e e e

Parity Nulliparous Multiparous Gestational age

1 17 39 G 2

11 6 31 G 4

e e !0.0001

Blood pressure Systolic Diastolic

126 G 9 71 G 11

176 G 15 109 G 11

13 5

5 12

Mode of delivery Vaginal Cesarean section

0.0002 0.0002 e e

Data presented as mean G SD.

specific activity tritiated PAF tracer with an antibody specific for PAF and a PAF standard. One gram of placental tissue from each placenta was used in the assay. PAF levels were presented as ng/g tissue. Measurement of PAF–acetylhydrolase activity Trophoblast PAFeacetylhydrolase (PAFeAH) activity was measured by PAFeAH assay kit (Cayman Chemical, Ann Arbor, MI). This assay uses 2-thio PAF as a substrate for PAFeAH. Upon hydrolysis of the acetyl thioester bond at the sn-2 position by PAFeAH, free thiols are released and detected using 5,5#-dithiobis-2-nitrobenzoic acid (DTNB) by UV spectrophotometry at 414 nm. Placental trophoblasts were isolated as previously described [8,9]. Fresh isolated trophoblasts were lysed and the total cell lysate was used to measure the trophoblast cytosolic PAFeAH activity. Each sample was assayed in triplicate. Trophoblast lysate protein concentration was determined and PAFeAH activity was expressed as mmol/min/mg protein. Total RNA extraction and PAF receptor expression mRNA expression for PAF receptor was determined by the RNase protection assay (RPA). Total RNA was isolated from placental tissues by acid guanidine thiocyanate phenol chloroform extraction as previously described [10]. Isolated total RNA was checked by denaturing agarose gel electrophoresis and ethidium bromide staining before running the RNase protection assay (RPA). No RNA degradation was noticed. RiboQuantÔ multi-probe RNase protection assay system was purchased from PharMingen (San Diego, CA). The RPA system includes a multi-probe template set, in vitro transcription kit,

Placenta (2006), Vol. 27

746 10

Placental Tissue PAF levels (ng/gram tissue)

and RPA kit. Total RNA of 10 mg from each sample and [a-32P]UTP (3000 Ci/mmol, 10 mCi/ml) labeled probes were used in each reaction. Samples were allowed to hybridize overnight at 37 (C and then treated for 30 min with ribonuclease solution followed by inactivation with proteinase K. Protected RNA fragments were precipitated and separated on a 5% acrylamide UreaeTBE sequencing gel. mRNA expression of housekeeping gene GAPDH was also determined in each sample. Autoradiograph from the gel was scanned and analyzed by NIH Image 1.16. Placental explant culture

Measurement of lipid peroxides 8-Isoprostane and malondialdehyde (MDA) were used as markers of lipid peroxides. 8-Isoprostane was measured by the BIOXYTECH 8-isoprostane ELISA kit purchased from OXIS International, Inc. (Portland, OR). The concentration of standard curve ranged from 0.05 to 50 ng/ml. An aliquot of 25 ml of culture supernatant from each sample was assayed in duplicate. MDA was measured by thiobarbituric acid (TBA) reaction (TBARS). 1,1,3,3-Tetramethoxypropane (TMP) was used to generate a standard curve and 100 ml of culture supernatant per sample was measured for MDA in duplicate. Concentrations of 8-isoprostane and MDA were normalized by the wet tissue volume in each well. Data were presented as mean G SE per mg wet tissue. Statistical analysis Data are presented as mean G SE. Statistical analysis was performed with nonparametric ManneWhitney U test and ANOVA by a computer software program StatView. Fisher’s PLSD and Bonferroni/Dunn tests were used as post hoc tests. A probability level less than 0.05 was considered statistically significant. Power analysis was performed by a computer software program, Power and Precision (Biostat, Inc. Englewood, NJ). RESULTS

6

4

2

0

1.0

Trophoblast PAF-AH Activities (µmol/min/µg protein)

Normal placental explants of 500 mg wet tissue/well were incubated with 7 ml serum-free Dulbecco’s Modified Eagle Medium (DMEM) containing PAF at concentrations of 0, 1, and 10 mg/ml for 48 h. All cultures were performed in duplicate. The culture supernatant was collected at the end of incubation and stored at ÿ20 (C. Supernatants were used for measuring lipid peroxide production.

8

*

0.8

0.6

0.4

0.2

0 Normal

PE

Figure 1. PAF levels and PAFeAH activities in placentas from normal (n Z 8) and preeclamptic (n Z 7) pregnancies. Upper panel: PAF levels and lower panel: PAFeAH activities. *p ! 0.05.

study. It would reach statistical difference if the subject number increased to 22 per group (alpha Z 0.05, tails Z 2, and power Z 0.80). The mean PAFeAH activity was significantly higher in trophoblasts isolated from preeclamptic placentas (n Z 7), 0.69 G 0.16 mmol/min/mg protein, than that from normal pregnancies (n Z 8), 0.38 G 0.03 mmol/ min/mg protein, p ! 0.05 (Figure 1, lower panel). PAF receptor expression Figure 2 shows the mRNA expression of PAF receptor determined in five normal and five preeclamptic placentas. The expression of GAPDH was used as a housekeeping gene for each sample. The relative mRNA expression was not different in preeclamptic tissues compared to that in normal placental tissues, 0.76 G 0.13 versus 0.70 G 0.08 (ratio of PAF/GAPDH).

Placental tissue PAF levels and PAF–AH activities

Effects of PAF on lipid peroxide production

The mean PAF level was 6.45 G 1.05 ng/g tissue in preeclamptic placentas (n Z 7) and 4.47 G 0.60 ng/g tissue in normal placentas (n Z 8). However, the elevated PAF level was not statistically different in preeclamptic compared to that in normal placental tissue, p Z 0.42 (Figure 1, upper panel). This could be due to the small study population in the present

Increased lipid peroxide production in preeclamptic placentas has been demonstrated by numerous studies [8,11,12]. To determine if PAF could stimulate placental production of lipid peroxides, normal placental tissues were incubated with PAF at concentrations of 1 and 10 mg/ml in serum-free DMEM for 48 h. Placentas delivered from five normal pregnancies were studied in

Gu et al.: PAF Levels and PAFeAH Activities in Placentas

747

PAFr

GAPDH 1

2

3

4

5

6

7

Normal Placentas

8

9

10

PE Placentas

Relative mRNA expression (Density PAFr/GAPDH)

1.2

1.0

0.8

0.6

0.4

0.2

0 Normal

PE

Figure 2. mRNA expression for PAF receptor in placental tissues from normal (n Z 5) and preeclamptic (n Z 5) pregnancies. Upper panel: mRNA expressions of PAF receptor and GAPDH. Total RNA of 10 mg per sample was used in the assay. GAPDH was used as a housekeeping gene for each sample. Lower panel: Relative mRNA expression of PAF receptor in normal and preeclamptic placental tissues.

this experiment. Lipid peroxides were measured by 8-isoprostane and MDA in the culture supernatant. As shown in Figure 3, the 8-isoprostane production was 0.37 G 0.09 ng/mg tissue in controls, 0.32 G 0.09 ng/mg tissue with PAF at 1 mg/ml, and 0.37 G 0.07 ng/mg tissue with PAF at 10 mg/ml in culture, p O 0.5. Similarly, there was also no difference for MDA production in placental tissues stimulated with PAF compared to the control: 0.68 G 0.04 nmol/mg (PAF 1 mg/ml) and 0.69 G 0.03 nmol/mg (PAF 10 mg/ml) versus 0.62 G 0.05 nmol/mg (control), p O 0.5. The productions of 8-isoprostane and MDA by placental tissues were not affected by addition of PAF in the culture.

PAF is a potent biological regulator in the vascular system. It regulates endothelial permeability, vascular smooth muscle contraction, and neutrophil and platelet adherence to vascular endothelial cells. However, the function of PAF in the placenta or placental trophoblasts is largely unknown. Several investigators have studied PAF and PAFeAH in the maternale fetal decidual interface. For example, Narahara et al. [13] found that human decidual macrophages secreted PAFeAH.

1.0

DISCUSSION In this study, we reported PAF levels and PAFeAH activities in placental tissue samples from normal and preeclamptic pregnancies. We found that PAF levels were elevated but not statistically different in placental tissues from preeclamptic pregnancies than those from normal pregnancies. Whereas, PAFeAH activities were increased in trophoblasts isolated from preeclamptic placentas compared to those from normal placentas. It is known that the actions of PAF are regulated at the level of its synthesis, degraded by PAFeAH, and neutralized by its specific receptor, PAF receptor. Therefore, it is very likely that the higher PAFeAH activity detected in trophoblasts may represent a compensatory mechanism to degrade PAF levels in the placenta from preeclamptic pregnancies.

Lipid Peroxide Production

8-isoprostane (ng/mg) 0.8

MDA (nmol/mg)

0.6

0.4

0.2

0 Control

PAF 1µg/ml

PAF10 µg/ml

Figure 3. The production of 8-isoprostane and MDA by the normal placental explants. Placental explants were incubated with PAF at concentrations of 0, 1 or 10 mg/ml in serum-free DMEM for 48 h. The results were means from five independent experiments. The productions of 8-isoprostane and MDA by placental tissues were not affected by addition of PAF in the culture.

748

Kawano et al. [14] reported that inhibition of macrophage secretion of PAFeAH by cytokine IL-8 could result in an increase in PAF levels in the decidua. Furthermore, studies also showed that PAF mediated the angiogenic activities induced by TNFa and hepatocyte growth factor (HGF) [15,16], and that placentas from preeclampsia produced more TNFa and HGF [17e19]. PAF could also enhance VEGFinduced endothelial cell motility and neoangiogenesis in an in vivo matrigel model [20]. These results suggest that cytokines may play an important role in the regulation of PAF and PAFeAH metabolism or vice versa. Elevated circulating PAF levels were previously reported in women with preeclampsia compared with those in normal pregnancies [2]. Since molecules produced by the placenta account for their levels in the maternal circulation such as soluble VEGF receptor-1 (sFlt-1), one may raise the question of whether PAF from the placenta could be released into the maternal circulation and contribute to the increased maternal PAF levels in preeclampsia. However, data from the present study and from others do not support this position. First, PAF levels were not significantly increased in placental tissue from women with preeclampsia. Second, an in vitro study of transferring PAF using 3H-PAF showed that less than 2% of PAF could cross fetal membranes and most of the 3H-PAF were metabolized within the chorio-decidua [21]. These results suggest that PAF might be limited and degraded by PAFeAH in situ within the placental tissue. Lastly, we observed that PAF receptor messenger was strongly expressed in both normal and preeclamptic placental tissues. Thus, our data support the idea that PAF receptor and cellular PAFeAH neutralize and metabolize PAF in situ. It is well known that placental oxidative stress plays a significant role in the pathophysiology of preeclampsia. Increased superoxide generation and lipid peroxide production and decreased superoxide dismutase activity have been found in placental trophoblasts from preeclampsia [8,9]. Increased oxidative stress also contributes to increased placental vasoconstriction [22]. PAF is a biologically active and oxidized phospholipid [5]. PAFeAH has been demonstrated as an

Placenta (2006), Vol. 27

oxidized phospholipid hydrolase of high-density lipoprotein particles [23]. Oxidative mechanisms associated with vascular damage have been reported in studies of PAF-mediated cellular injury both in vivo and in vitro [1,7,24e26]. PAF could induce oxidative burst in macrophages [7]. In an ex vivo animal model, superoxide dismutase could attenuate PAFinduced changes in vascular resistance and enhanced neutrophil adhesion to the microvasculature [25]. The protective effects of superoxide dismutase suggest that these responses to PAF involve the generation of oxygen-derived free radicals. To further investigate whether PAF could, in turn, stimulate placental oxidative stress, normal placental tissues were incubated with PAF and production of lipid peroxides was determined. Unexpectedly, our results show that lipid peroxide production by normal placental tissues was not affected with PAF in culture. This could be due to the compensatory effects of degradation of PAF by PAFeAH and the neutralization of PAF by PAF receptor in the tissue. An alternative explanation is that elevated PAF levels and increased PAFeAH activities could be consequences of increased oxidative stress in the placental tissue in women with preeclampsia. Our data further suggest that placental tissue or trophoblast cells have the ability to metabolize/ degrade PAF within the placenta to protect PAF-induced damage extended to the maternal circulation. In summary, we found that PAFeAH activities are increased in placental trophoblasts from preeclamptic pregnancies compared to those from normal pregnancies. Our results also show that PAF could not stimulate lipid peroxide production in the normal placental tissue. These data suggest that an increase in PAFeAH activity may be a compensatory effect to control PAF levels in the preeclamptic placenta. And the co-existence of PAFeAH and PAF receptor in trophoblasts is likely the mechanism of autocrine regulation of PAF in the placenta to ensure that PAF and its metabolites could be degraded in situ to limit their harmful effects extended to the maternal circulation. Our data also suggest that maternal, other than the placenta, sources may be responsible for the increase in PAF levels in women with preeclampsia.

ACKNOWLEDGEMENT This study was supported in part by grants from National Institute of Health, National Institute of Child Health Development (NICHD) (HD36822) and National Heart Blood Lung Institute (NHBLI) (HL65997).

REFERENCES [1] Montrucchio G, Alloatti G, Camussi G. Role of platelet-activating factor in cardiovascular pathophysiology. Physiol Rev 2000;80:1669e99. [2] Rowland BL, Vermillion ST, Roundebush WE. Elevated circulating concentrations of platelet activating factor in preeclampsia. Am J Obstet Gynecol 2000;183:930e2. [3] Maki N, Magness RR, Miyaura S, Gant NF, Johnston JM. Plateletactivating factoreacetylhydrolase activity in normotensive and hypertensive pregnancies. Am J Obstet Gynecol 1993;168:50e4. [4] Ohshige A, Yoshimura T, Matsui K, Koyama H, Ito M, Okamura H. Reduced platelet-activating factoreacetylhydrolase activity in maternal

plasma obtained from normotensive pregnant women destined to develop pregnancy-induced hypertension. Hypertens Pregnancy 1998;17: 297e306. [5] McIntyre TM, Zimmerman GA, Prescott SM. Biologically active oxidized phospholipids. J Biol Chem 1999;274:25189e92. [6] Lloberas N, Torras J, Herrero-Fresneda I, Cruzado JM, Riera M, Hurtado I, et al. Postischemic renal oxidative stress induces inflammatory response through PAF and oxidized phospholipids. Prevention by antioxidant treatment. FASEB J 2003;16:908e10. [7] Hartung HP, Parnham MJ, Winkelmann J, Englberger W, Hadding U. Platelet activating factor (PAF) induces the oxidative burst in macrophages. Int J Immunopharmacol 1983;5:115e21.

Gu et al.: PAF Levels and PAFeAH Activities in Placentas [8] Walsh SW, Wang Y. Trophoblast and placental villous core production of lipid peroxides, thromboxane, and prostacyclin in preeclampsia. J Clin Endocrinol Metab 1995;80:1888e93. [9] Wang Y, Walsh SW. Increased superoxide generation is associated with decreased superoxide dismutase activity and mRNA expression in placental trophoblast cells in preeclampsia. Placenta 2001;22:206e12. [10] Wang Y, Walsh SW. Antioxidant activities and mRNA expression of superoxide dismutase, catalase, and glutathione peroxidase in normal and preeclamptic placentas. J Soc Gynecol Investig 1996;3:179e84. [11] Wang Y, Walsh SW, Kay HH. Placental lipid peroxides and thromboxane are increased and prostacyclin is decreased in women with preeclampsia. Am J Obstet Gynecol 1992;167:946e9. [12] Poston L, Raijmakers MT. Trophoblast oxidative stress, antioxidants and pregnancy outcome e a review. Placenta 2004;(Suppl A):S72e8. [13] Narahara H, Nishioka Y, Johnston JM. Secretion of platelet-activating factor acetylhydrolase by human decidual macrophages. J Clin Endocrinol Metab 1993;77:1258e62. [14] Kawano Y, Narahara H, Johnston JM. Inhibitory effect of interleukin-8 on the secretion of platelet-activating factor acetylhydrolase by human decidual macrophages. J Soc Gynecol Investig 1999;6:328e32. [15] Montrucchio G, Lupia E, Battaglia E, Passerini G, Bussolino F, Emanuelli G, et al. Tumor necrosis factor alpha-induced angiogenesis depends on in situ platelet-activating factor biosynthesis. J Exp Med 1994;180:377e82. [16] Camussi G, Montrucchio G, Lupia E, Soldi R, Comoglio PM, Bussolino F. Angiogenesis induced in vivo by hepatocyte growth factor is mediated by platelet-activating factor synthesis from macrophages. J Immunol 1997;158:1302e9. [17] Wang Y, Walsh SW. TNFa concentrations and mRNA expression are increased in preeclamptic placentas. J Reprod Immunol 1996;32:157e69. [18] Vince GS, Starkey PM, Austgulen R, Kwiatkowski D, Redman CWG. Interleukin-6, tumour necrosis factor and soluble tumour necrosis factor

749

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

receptors in women with pre-eclampsia. Br J Obstet Gynaecol 1995;102: 20e5. Kauma SW, Bae-Jump V, Walsh SW. Hepatocyte growth factor stimulates trophoblast invasion: a potential mechanism for abnormal placentation in preeclampsia. J Clin Endocrinol Metab 1999; 84:4092e6. Montrucchio G, Lupia E, Battaglia E, Del Sorbo L, Boccellino M, Biancone L, et al. Platelet-activating factor enhances vascular endothelial growth factor-induced endothelial cell motility and neoangiogenesis in a murine matrigel model. Arterioscler Thromb Vasc Biol 2000;20:80e8. Morris C, Sullivan MH, Elder MG. Transfer and metabolism of plateletactivating factor by fetal membranes, amnion and chorio-decidua. Br J Obstet Gynaecol 1992;99:895e8. Wang Y, Walsh SW. The ratio of thromboxane to prostacyclin is increased by peroxide in a dose-dependent manner, along with increased vasoconstriction in the human placenta. Hypertens Pregnancy 1998;17: 1e11. Marathe GK, Zimmerman GA, McIntyre TM. Platelet-activating factor acetylhydrolase, and not paraoxonase-1, is the oxidized phospholipid hydrolase of high density lipoprotein particles. J Biol Chem 2003;278: 3937e47. Droy-Lefaix MT, Drouet Y, Geraud G, Hosford D, Braquet P. Superoxide dismutase (SOD) and the PAF-antagonist (BN 52021) reduce small intestinal damage induced by ischemia-reperfusion. Free Radic Res Commun 1991;12e13(Pt 2):725e35. Schwappach JR, Wood JG, Cheung LY. Superoxide dismutase attenuates effects of platelet-activating factor on gastric microcirculation. J Surg Res 1995;59:733e88. Huang YC, Nozik ES, Piantadosi CA. Superoxide dismutase potentiates platelet-activating factor-induced injury in perfused lung. Am J Physiol 1994;266:L246e54.