- Email: [email protected]
Glutathione S-transferase isoenzymes in decidua and placenta of preeclamptic pregnancies
Glutathione S-Transferase Isoenzymes in Decidua and Placenta of Preeclamptic Pregnancies PETRA L. M. ZUSTERZEEL, MD, WILBERT H. M. PETERS, PhD, MARION A. H. DE BRUYN, MSc, MAARTEN F. C. M. KNAPEN, MD, PhD, HANS M. W. M. MERKUS, MD, PhD, AND ERIC A. P. STEEGERS, MD, PhD
Objective: To investigate a possible involvement of glutathione S-transferases, major detoxificating enzymes, in the pathophysiology of preeclampsia. Methods: Levels of glutathione S-transferase isoforms and enzyme activity were assessed in placental and decidual tissues in 22 preeclamptic and 21 normotensive women. Measured values were analyzed statistically using the Mann-Whitney U test for comparison between groups, and the signed-rank test for comparison within groups. Results: Glutathione S-transferase pi is the main glutathione S-transferase isoform in normal placental and decidual tissue. Lower median placental and decidual glutathione S-transferase pi levels were found in preeclamptic women compared with controls: 1268 (range: 524 –3925) and 2185 (range: 503– 6578), P ⴝ .05, for placenta; 1543 (range: 681– 2967) and 2169 (range: 893–3929), P ⴝ .02, for decidua. The total amount of glutathione S-transferases in control and preeclamptic pregnancies was higher in decidua than in placenta. Conclusion: Reduced levels of glutathione S-transferase class pi in preeclampsia might indicate a decreased capacity of the glutathione/glutathione S-transferase detoxification system. A higher total amount of glutathione S-transferases in decidual tissue might point to a more pronounced protective role of decidua compared with placenta. (Obstet Gynecol 1999;94:1033– 8. © 1999 by The American College of Obstetricians and Gynecologists.)
The etiology of preeclampsia is still largely unknown. Increasing evidence suggests that a central factor in the pathogenesis of preeclampsia appears to be placental ischemia that in some way triggers maternal endothelial From the Departments of Obstetrics and Gynecology and Gastroenterology, University Hospital St Radboud, Nijmegen, The Netherlands. The Dutch ‘Praeventiefonds’ (Grant no. 28-2801) supported this work.
VOL. 94, NO. 6, DECEMBER 1999
cell dysfunction.1,2 Placental ischemia or an increased activity of decidual lymphoid tissue, possibly resulting from immunologic maladaptation, may lead to excessive oxygen free radical production. This may enhance the release of placental lipid peroxidation products into the circulation. 1 Lipid peroxides activate cyclooxygenase and lead to selective prostacyclin synthetase inhibition.3 A variety of antioxidant mechanisms serve to control peroxidative processes.4 Cells and tissues are protected against toxic lipid peroxides by antioxidants, recruited either from endogenous systems or from the diet.5 Antioxidants can be classified according to working mechanisms: preventive antioxidants act to inhibit the initiation of the peroxidation process, and chainbreaking antioxidants act to trap or decompose radicals or peroxides already present. 5 Important chainbreaking antioxidants are alpha-tocopherol, asorbic acid, cartenoids, glutathione, and uric acid. Important antioxidant enzymes are superoxide dismutase, catalase, glucose-6-phosphate dehydrogenase, glutathione peroxidase, and glutathione S-transferase. One of the most important systems involved in the metabolism and detoxification of reactive oxygen, xenobiotics as well as carcinogens, is that of glutathione/ glutathione S-transferase.6 Glutathione S-transferases (EC 2.5.1.18) catalyze the nucleophilic addition of glutathione to electrophilic centers of a wide range of substrates. They are involved also in the cellular transport of a broad range of lipophilic compounds.7 Human cytosolic glutathione S-transferases are dimeric proteins, which can be subdivided into four major groups: alpha, mu, pi, and theta.7 Placental and decidual glutathione S-transferases may contribute greatly to fetal and maternal detoxification capacity.8 Increasing evidence
0029-7844/99/$20.00 PII S0029-7844(99)00433-0
1033
suggests an important role for glutathione and glutathione-related enzyme systems in the development of preeclampsia.9 –14 Although variable results have been published, most studies show a reduction in antioxidant levels and corresponding enzyme activities in women with preeclampsia.15 We determined the isoenzyme levels of glutathione S-transferases in placenta and decidua of preeclamptic and normotensive women. Because glutathione S-transferase pi is the main isoform in placental tissue,16 we hypothesized that placental glutathione S-transferase pi levels are decreased in preeclampsia. Decidual tissue has high levels of glutathione and glutathione S-transferase enzyme activity,14 suggesting an important function in the detoxification process.
Materials and Methods Two groups of women were investigated. A control group of 21 uncomplicated normotensive pregnant women without any obvious medical disorder and 22 preeclamptic patients were studied, nine of whom had hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome. Sixteen of these preeclamptic patients also contributed to another study concerning glutathione levels in whole blood.14 All preeclamptic and control subjects were selected randomly for study between April 1996 and March 1997. The institutional review board of the University Hospital Nijmegen approved the protocol. All women were delivered by elective cesarean and were not in labor. Elective cesarean was performed in the control group because of cephalopelvic disproportion, repeat cesarean, or breech presentation. Cesarean was performed in women with preeclampsia because of deteriorating fetal or maternal conditions. Preeclampsia was defined as pregnancyinduced hypertension (diastolic blood pressure greater than 90 mmHg on two or more consecutive occasions, each more than 4 hours apart, or diastolic blood pressure greater than 110 mmHg on one occasion), and concordant proteinuria (urinary protein greater than 0.3 g/L).17 HELLP syndrome was defined as hemolysis (lactic dehydrogenase [LDH] greater than 600 IU/L); elevated liver enzymes (serum aspartate aminotransferase greater than 70 IU/L and serum alanine aminotransferase greater than 70 IU/L); and low platelets (platelet count less than 100 ⫻ 109/L).18 Blood pressure was measured with a sphygmomanometer while the patient was seated. Diastole was recorded at phase IV Korotkoff sound. Urinary protein was determined as protein-creatinine ratio. Table 1 summarizes the population characteristics. Fragments of placenta and decidua were excised during cesarean and frozen in liquid nitrogen within 20
1034 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
Table 1. Population Characteristics of Preeclamptic and Control Women
Maternal age (y) Gestational age (wk) Parity Diastolic blood pressure (mmHg) Hemoglobin (g/dL) Hematocrit (L/L) Platelet count (109/L)† Serum creatinine (mol/L) Serum uric acid (mmol/L) Serum aspartate aminotransferase (IU/L) Serum alanine aminotransferase (IU/L) Serum lactic dehydrogenase (IU/L) Proteinuria (g/L)
Control (n ⫽ 21)
Preeclamptic (n ⫽ 22)
32 (22– 42) 38⫹5 (37⫹3– 41⫹3) 1 (0 –3) 75 (58 – 85)
29 (22–36) 34⫹1 (26⫹3–38⫹2)* 0 (0 –2)† 100 (95–120)†
11.9 (10.8 –13.5) 0.36 (0.32– 0.41) 217 (150 –301) 60 (54 – 83) 0.28 (0.17– 0.36) 7 (4 –11)
12.9 (10.1–14.5) 0.37 (0.30 – 0.40) 61 (31–305)† 71 (52–108)† 0.26 (0.22– 0.55) 397 (7– 666)† 353 (5–562)†
10 (5–14) 208 (182–302)
1189 (206 –1873)† 0.97 (0.3–27.6)
Values are given as median (range). * P ⬍ .005. † P ⬍ .001 preeclamptic compared with control pregnancy.
minutes. Tissue was stored at ⫺20C. Upon use, tissue fragments (20 –100 mg) were thawed and homogenized in 10 volumes ice-cold homogenizing buffer (250 mM sucrose, 20 mM Tris-HCl, 1 mM dithriothreitol, pH 7.4) by 10 strokes in a small glass-glass homogenizer. The homogenates were centrifuged at 150,000 g at 4C for 1 hour. Supernatants were frozen in liquid nitrogen and stored at ⫺20C in small portions. Protein contents were assayed in triplicate by the method of Lowry et al, using bovine serum albumin as the standard.19 Total glutathione S-transferase isoenzyme activity was determined in triplicate according to Habig et al,20 using 1-chloro-2,4-dinitrobenzene as substrate. Specific glutathione S-transferase isoenzyme levels— classes alpha, mu, pi, and theta (T1-1 and T2-2)—were determined on Western blots essentially, as described by Peters et al.21 In short, cytosolic fractions (glutathione S-transferase alpha, mu, pi: 100 g protein for both placenta and decidua; glutathione S-transferase T1-1: 100 g cytosolic protein for placenta and 25 g for decidua; glutathione S-transferase T2-2: 100 g cytosolic protein for placenta and 50 g for decidua) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (11% acrylamide, w/v), and subsequently to Western blotting, using a semidry blotting system (Novablot II; Pharmacia, Uppsala, Sweden). Control and experimental samples were run on the same blots. Western blots were incubated with monoclonal antibodies against human glutathione Stransferase classes alpha, mu, pi, and T1-1, respectively, and a polyclonal antibody against human glutathione
Obstetrics & Gynecology
Table 2. Glutathione S-Transferase Alpha, Mu, Pi, and Theta Isoenzymes and Enzyme Activity in Decidua and Placenta of Preeclamptic and Control Patients Decidua GST isoenzymes GST GST GST GST GST GST
alpha mu pi T1-1 T2-2 activity
Placenta
Control pregnancy (n ⫽ 21)
Preeclamptic pregnancy (n ⫽ 22)
Control pregnancy (n ⫽ 21)
Preeclamptic pregnancy (n ⫽ 22)
0 (0 –202) 799 (0 –2972)† 2185 (503– 6578) 779 (0 –2437)* 899 (187–1729)* 172 (54 – 624)
0 (0 –141)* 873 (0 –2136)* 1268 (524 –3925)‡ 864 (404 –2926)§ 829 (164 –2278)§ 229 (121– 460)
86 (0 –909) 247 (0 –1180) 2169 (893–3929) 183 (0 –363) 537 (58 –958) 169 (74 – 472)
37 (0 – 601) 1412 (0 –1436) 1543 (681–2967)‡ 911 (30 –944) 315 (35–1133) 224 (94 –337)
GST ⫽ glutathione S-transferase. Results are presented as median (range). Glutathione S-transferase quantities are expressed in ng/mg protein, and glutathione S-transferase enzyme activity in nanomol per minute per milligram protein (nmol/min/mg protein). * P ⬍ .001. † P ⬍ .005. ‡ P ⬍ .005 (signed-rank test), preeclamptic compared with control pregnancy (decidual or placental tissue). § P ⬍ .0001 (Mann-Whitney U test), decidual compared with placental tissue within each group.
S-transferase T2-2. Antibodies against classes alpha (A1-1, A1-2, A2-2), mu (M1-1), and pi (P1-1) were developed in our own laboratory.21–23 Antibodies against glutathione S-transferase classes T1-1 and T2-2 were obtained from Prof. Dr. E. I. Juronen, Tartu, Estonia, and Prof. Dr. P. G. Board, Canberra, Australia, respectively. The specific binding of the primary antibodies to the isoenzymes was quantified after incubation with secondary antibodies: peroxidase-conjugated rabbit antimouse immunoglobulin (Dakopatts, Glostrup, Denmark) for glutathione S-transferase classes alpha, mu, pi, and T1-1, or peroxidaseconjugated swine antirabbit immunoglobulin (Dakopatts) for glutathione S-transferase T2-2, respectively. Subsequently, development of the peroxidase label took place with 4-chloro-1-naphthol (Sigma-Alderich, Zwijndrecht, the Netherlands) and hydrogen peroxide (Merck, Darmstadt, Germany) for glutathione Stransferase alpha and pi. Staining for the other glutathione S-transferase isoenzymes was performed using 0.1% 3, 3-diaminobenzidine (Sigma-Alderich, Zwijndrecht, the Netherlands) in phosphate-buffered saline containing 0.01% hydrogen peroxide as peroxidase substrate, and imidazole (0.34 g/L) and cobalt-chloride 6H2O (0.26 g/L) for enhancement of staining intensity. Staining intensity on immunoblots was quantified using a laser densitometer (Ultroscan XL; LKB, Bromma, Sweden). For calculating the relative amounts of glutathione S-transferase isoenzymes, known amounts of purified cytosolic glutathione S-transferases from human liver (alpha, mu, T1-1, T2-2) or placenta (pi) were run in parallel with the experimental samples and served as standards for the calculation of isoenzyme levels in the cytosolic fractions (values in ng/mg protein). Interassay variability of the immunoblot procedure was ⫾5%.
VOL. 94, NO. 6, DECEMBER 1999
The Mann-Whitney U test with Bonferroni correction for multiple comparison was used to assess statistical significance of differences between groups. The signedrank test with Bonferroni correction was used to assess statistical significance between placental and decidual tissue within groups. P ⱕ .004 (.05/12 tests) was considered significant.
Results Glutathione S-transferase enzyme activities and isoenzyme contents in placental and decidual tissues of both preeclamptic and control pregnancies are shown in Table 2. Placental tissue of controls contained the highest levels of glutathione S-transferase pi. Glutathione S-transferase T2-2 content was approximately 17% of the total amount of placental glutathione S-transferases determined here. The levels of other glutathione Stransferase isoforms were low. The glutathione Stransferase isoenzyme pattern in decidual tissue of controls differed from that in placental tissue. Glutathione S-transferase pi also was expressed at highest levels; however, classes T2-2, mu, and T1-1 also were present in considerable amounts (19, 17, and 17% of the total amount of glutathione S-transferases, respectively). Class alpha glutathione S-transferase was undetectable in the decidua of most women. Decidual glutathione S-transferases mu, T1-1, and T2-2 levels were significantly higher compared with placental levels in both preeclamptic and control groups. In contrast, decidual levels of glutathione S-transferase alpha were lower than placental levels. No differences were found between placental and decidual glutathione Stransferase pi levels in preeclamptic or control groups. Preeclamptic women showed statistically lower glutathione S-transferase pi levels in placental and decid-
Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
1035
Figure 1. Glutathione S-transferase pi levels (ng/mg protein) in control women (solid circles) and preeclamptic women (open boxes). Horizontal line indicates median.
ual tissue compared with women in the control group (P ⫽ .004 and P ⫽ .002, respectively) (Figure 1). No significant differences in median tissue level in decidua and placenta between the preeclamptic and the control group could be demonstrated for classes alpha, mu, T1-1, and T2-2. The results are summarized in Table 2. Median placental and decidual glutathione Stransferase enzyme activity was not significantly different between study groups (Table 2).
Discussion In this study, we compared the main human glutathione S-transferase isoenzyme levels in placental and decidual tissue from controls and preeclamptic patients. We also included nine women in the preeclamptic group who concurrently had HELLP syndrome because this syndrome may be considered a severe variant or complication of preeclampsia.24 An imbalance between lipid peroxides, oxygen free
1036 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
radicals, and other toxins on the one site, and radical scavengers and detoxificating systems on the other, might play an important role in the etiology and pathophysiology of preeclampsia.1,13 In preeclampsia, the oxidant/antioxidant balance is tipped in favor of oxidants at the expense of antioxidants, because in plasma, elevated levels of maternal lipid peroxides were measured.15 Serum-free fatty acids and triglycerides also were elevated in preeclampsia and contribute to the pool of lipid peroxides as they are oxidized.25 Several important antioxidants are decreased significantly in women with preeclampsia, which would contribute to the increased oxidative damage.15 In normal pregnancies, total glutathione S-transferase isoenzymes appeared to be higher in decidua than in placenta. High decidual levels of these isoenzymes may indicate a pronounced role of decidual tissue in protecting the mother or the fetus against damage by free radical peroxides and other toxic compounds produced by placental or fetomaternal interface.
Obstetrics & Gynecology
Glutathione S-transferase pi is the main glutathione S-transferase isoenzyme in placental16 and decidual tissue. Therefore, it is likely that glutathione Stransferase pi contributes greatly to the prevention of fetal damage.8 Both placental and decidual glutathione S-transferase pi levels were statistically lower in preeclamptic pregnancy compared with glutathione Stransferase pi levels in normotensive pregnancy. Decreased placental and decidual glutathione Stransferase pi levels might be related to a decreased effective capacity of the glutathione/glutathione Stransferase system in preeclamptic women. Placental and decidual glutathione S-transferase enzyme activity in preeclamptic and normotensive pregnancies, however, was not different, which is in agreement with an earlier report on placental glutathione S-transferase activity by Poranen et al.26 Numerous studies, recently reviewed by Knapen et al,27 have shown that the expression of glutathione S-transferase may be affected by factors such as diet. Nijhoff et al28 found that diets rich in cruciferous vegetables increased plasma glutathione S-transferase pi and alpha levels. Although dietary differences might have influenced our results, no studies are available concerning possible dietary effects on placental and decidual glutathione S-transferase expression. Other factors that might have influenced our results, given the difference between our study and control groups, are parity and gestational age. Di Ilio et al29 found that human placental glutathione S-transferase activity declined significantly as pregnancy advanced; however, whether or not qualitative and quantitative changes occur in glutathione S-transferase isoenzymes during pregnancy has been studied only in mice.30 In the last 5 days of pregnancy in the mouse, an increment in placental alpha and pi class glutathione S-transferase occurred, whereas class mu glutathione S-transferase decreased.30 Unfortunately, the effects of parity on glutathione S-transferase levels and activity have never been studied.
References 1. Hubel CA, Roberts JM, Taylor RN, Musci TJ, Rogers GM, Mc Laughlin M. Lipid peroxidation in pregnancy: New perspectives on preeclampsia. Am J Obstet Gynecol 1989;160:1025–34. 2. Williams DJ, de Swiet M. The pathophysiology of preeclampsia; review article. Intensive Care Med 1997;23:620 –9. 3. Higgs GA, Vane JR. Inhibition of cyclo-oxygenase and lipoxygenase. Br Med Bull 1983;39:235–70. 4. Sies H. Oxidative stress: Introductionary remarks. In: Sies H, ed. Oxidative stress. London: Academic Press, 1985:1– 8. 5. Halliwell B. Free radicals, antioxidants, and human disease: Curiosity, cause or consequence? Lancet 1994;344:721– 4. 6. Beckett GJ, Hayes JD. Glutathione S-transferases: Biomedical applications. Adv Clin Chem 1993;30:281–380.
VOL. 94, NO. 6, DECEMBER 1999
7. Hayes JD, Pulford DJ. The glutathione S-transferase supergene family: Regulation of GST and the contribution of isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 1995;30:445– 600. 8. Pacifici GM, Franchi M, Colizzi C, Guiliani L, Rane A. Glutathione S-transferase in humans: Development and tissue distribution. Arch Toxicol 1988;61:265–9. 9. Uotila JT, Tuimala RJ, Aarnio TM, Pyykko¨ KA, Uhotopa MO. Findings on lipid peroxidation and antioxidant function in hypertensive complications of pregnancy. Br J Obstet Gynaecol 1993;100: 270 – 6. 10. Walsh SW, Wang Y. Deficient glutathione peroxidase activity in preeclampsia is associated with increased placental production of thromboxane and lipid peroxides. Am J Obstet Gynecol 1993;169: 1456 – 61. 11. Kabi BC, Goel N, Rao YN, Tripathy R, Tempe A, Thakur AS. Levels of erythrocyte malonyldialdehyde, vitamin E, reduced glutathione, G6PD activity and plasma urate in patients of pregnancy induced hypertension. Indian J Med Res 1994;100:23–5. 12. Wang Y, Walsh SW. Antioxidant activities and mRNA expression of superoxide dismutase, catalase and glutathione peroxidase in normal and preeclamptic pregnancies. J Soc Gynecol Invest 1996; 3:179 – 84. 13. Loverro G, Greco P, Capuano F, Carone D, Cormio G, Selvaggi L. Lipid peroxidation and antioxidant enzyme activity in pregnancy complicated with hypertension. Eur J Obstet Gynecol Reprod Biol 1996;70:123–7. 14. Knapen MCFM, Mulder TPJ, Rooij van IALM, Peters WHM, Steegers EAP. Low whole blood glutathione levels in pregnancies complicated by preeclampsia or the hemolysis, elevated liver enzymes, low platelets syndrome. Obstet Gynecol 1998;92:1012–5. 15. Walsh SW. Maternal-placental interactions of oxidative stress and antioxidants in preeclampsia. Semin Reprod Endocrinol 1998;16: 93–104. 16. Awasthi YC, Sharma R, Singhal SS. Human glutathione Stransferases; minireview. Int J Biochem 1994;26:295–308. 17. Davey DA, MacGillivray I. The classification and definition of hypertensive disorders of pregnancy. Am J Obstet Gynecol 1988; 158:892– 8. 18. Sibai BM. The HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets): Much ado about nothing? Am J Obstet Gynecol 1990;162:311– 6. 19. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurements with the folin phenol reagent. J Biol Chem 1951;193:265–75. 20. Habig WH, Pabst MJ, Jacoby WB. Glutathione S-transferases, the first enzymatic step in mercapturic acid formation. J Biol Chem 1974;249:7130 –9. 21. Peters WHM, Boon CEW, Roelofs HMJ, Wobbes T, Nagengast FM, Kremers PG. Expression of drug metabolizing enzymes and P-170 glycoprotein in colorectal carcinoma and normal mucosa. Gastroenterology 1992;103:448 –55. 22. Peters WHM, Nagengast FM, Wobbes T. Glutathione Stransferases in normal and cancerous human colon tissue. Carcinogenesis 1989;10:2371– 4. 23. Peters WHM, Kock L, Nagengast FM, Roelofs HJM. Immunodetection with a monoclonal antibody of glutathione S-transferase Mu in patients with and without carcinomas. Biochem Pharmacol 1990;39:591–7. 24. Geary M. The HELLP syndrome. Br J Obstet Gynaecol 1997;104: 887–91. 25. Hubel CA, McLaughlin MK, Evans RW, Hauth BA, Sims CJ, Roberts JM. Fasting serum triglycerides, free fatty acids, and malondialdehyde are increased in preeclampsia, are positively
Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
1037
26.
27.
28.
29.
30.
correlated, and decrease within 48 hours post partum. Am J Obstet Gynecol 1996;174:975– 82. Poranen AK, Ekblad U, Uotila P, Ahotupa M. Lipid peroxidation and antioxidants in normal and preeclamptic pregnancies. Placenta 1996;17:401–5. Knapen MCFM, Zusterzeel PLM, Peters WHM, Steegers EAP. Glutathione and glutathione-related enzymes in reproduction: A review. Eur J Obstet Gynecol Reprod Biol 1999;82:171– 84. Nijhoff WA, Mulder TPJ, Verhagen H, van Poppel G, Peters WHM. Effects of consumption of Brussels sprouts on plasma and urinary glutathione s-transferase class-␣ and - in humans. Carcinogenesis 1995;16:955–7. Di Ilio C, Polidoro G, Arduini A, Muccini M, Federici G. Glutathione peroxidase, glutathione reductase, glutathione S-transferase and y-glutamyl-transpeptidase activities in human early pregnancy placenta. Biochem Med 1983;29:143– 8. Di Ilio C, Tiboni GM, Sacchetta P, Angelucci S, Bucciarelli T, Bellati U, et al. Time-dependant and tissue-specific variations of glutathione S-transferase activity during gestation in the mouse. Mech Ageing Dev 1995;78:47– 62.
1038 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
Address reprint requests to:
Eric A. P. Steegers, MD, PhD Department of Obstetrics & Gynecology University Hospital Nijmegen PO Box 9101 Nijmegen, 6500 HB The Netherlands E-mail: [email protected]
Received December 28, 1998. Received in revised form May 3, 1999. Accepted May 13, 1999.
Copyright © 1999 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Obstetrics & Gynecology
Objective: To investigate a possible involvement of glutathione S-transferases, major detoxificating enzymes, in the pathophysiology of preeclampsia. Methods: Levels of glutathione S-transferase isoforms and enzyme activity were assessed in placental and decidual tissues in 22 preeclamptic and 21 normotensive women. Measured values were analyzed statistically using the Mann-Whitney U test for comparison between groups, and the signed-rank test for comparison within groups. Results: Glutathione S-transferase pi is the main glutathione S-transferase isoform in normal placental and decidual tissue. Lower median placental and decidual glutathione S-transferase pi levels were found in preeclamptic women compared with controls: 1268 (range: 524 –3925) and 2185 (range: 503– 6578), P ⴝ .05, for placenta; 1543 (range: 681– 2967) and 2169 (range: 893–3929), P ⴝ .02, for decidua. The total amount of glutathione S-transferases in control and preeclamptic pregnancies was higher in decidua than in placenta. Conclusion: Reduced levels of glutathione S-transferase class pi in preeclampsia might indicate a decreased capacity of the glutathione/glutathione S-transferase detoxification system. A higher total amount of glutathione S-transferases in decidual tissue might point to a more pronounced protective role of decidua compared with placenta. (Obstet Gynecol 1999;94:1033– 8. © 1999 by The American College of Obstetricians and Gynecologists.)
The etiology of preeclampsia is still largely unknown. Increasing evidence suggests that a central factor in the pathogenesis of preeclampsia appears to be placental ischemia that in some way triggers maternal endothelial From the Departments of Obstetrics and Gynecology and Gastroenterology, University Hospital St Radboud, Nijmegen, The Netherlands. The Dutch ‘Praeventiefonds’ (Grant no. 28-2801) supported this work.
VOL. 94, NO. 6, DECEMBER 1999
cell dysfunction.1,2 Placental ischemia or an increased activity of decidual lymphoid tissue, possibly resulting from immunologic maladaptation, may lead to excessive oxygen free radical production. This may enhance the release of placental lipid peroxidation products into the circulation. 1 Lipid peroxides activate cyclooxygenase and lead to selective prostacyclin synthetase inhibition.3 A variety of antioxidant mechanisms serve to control peroxidative processes.4 Cells and tissues are protected against toxic lipid peroxides by antioxidants, recruited either from endogenous systems or from the diet.5 Antioxidants can be classified according to working mechanisms: preventive antioxidants act to inhibit the initiation of the peroxidation process, and chainbreaking antioxidants act to trap or decompose radicals or peroxides already present. 5 Important chainbreaking antioxidants are alpha-tocopherol, asorbic acid, cartenoids, glutathione, and uric acid. Important antioxidant enzymes are superoxide dismutase, catalase, glucose-6-phosphate dehydrogenase, glutathione peroxidase, and glutathione S-transferase. One of the most important systems involved in the metabolism and detoxification of reactive oxygen, xenobiotics as well as carcinogens, is that of glutathione/ glutathione S-transferase.6 Glutathione S-transferases (EC 2.5.1.18) catalyze the nucleophilic addition of glutathione to electrophilic centers of a wide range of substrates. They are involved also in the cellular transport of a broad range of lipophilic compounds.7 Human cytosolic glutathione S-transferases are dimeric proteins, which can be subdivided into four major groups: alpha, mu, pi, and theta.7 Placental and decidual glutathione S-transferases may contribute greatly to fetal and maternal detoxification capacity.8 Increasing evidence
0029-7844/99/$20.00 PII S0029-7844(99)00433-0
1033
suggests an important role for glutathione and glutathione-related enzyme systems in the development of preeclampsia.9 –14 Although variable results have been published, most studies show a reduction in antioxidant levels and corresponding enzyme activities in women with preeclampsia.15 We determined the isoenzyme levels of glutathione S-transferases in placenta and decidua of preeclamptic and normotensive women. Because glutathione S-transferase pi is the main isoform in placental tissue,16 we hypothesized that placental glutathione S-transferase pi levels are decreased in preeclampsia. Decidual tissue has high levels of glutathione and glutathione S-transferase enzyme activity,14 suggesting an important function in the detoxification process.
Materials and Methods Two groups of women were investigated. A control group of 21 uncomplicated normotensive pregnant women without any obvious medical disorder and 22 preeclamptic patients were studied, nine of whom had hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome. Sixteen of these preeclamptic patients also contributed to another study concerning glutathione levels in whole blood.14 All preeclamptic and control subjects were selected randomly for study between April 1996 and March 1997. The institutional review board of the University Hospital Nijmegen approved the protocol. All women were delivered by elective cesarean and were not in labor. Elective cesarean was performed in the control group because of cephalopelvic disproportion, repeat cesarean, or breech presentation. Cesarean was performed in women with preeclampsia because of deteriorating fetal or maternal conditions. Preeclampsia was defined as pregnancyinduced hypertension (diastolic blood pressure greater than 90 mmHg on two or more consecutive occasions, each more than 4 hours apart, or diastolic blood pressure greater than 110 mmHg on one occasion), and concordant proteinuria (urinary protein greater than 0.3 g/L).17 HELLP syndrome was defined as hemolysis (lactic dehydrogenase [LDH] greater than 600 IU/L); elevated liver enzymes (serum aspartate aminotransferase greater than 70 IU/L and serum alanine aminotransferase greater than 70 IU/L); and low platelets (platelet count less than 100 ⫻ 109/L).18 Blood pressure was measured with a sphygmomanometer while the patient was seated. Diastole was recorded at phase IV Korotkoff sound. Urinary protein was determined as protein-creatinine ratio. Table 1 summarizes the population characteristics. Fragments of placenta and decidua were excised during cesarean and frozen in liquid nitrogen within 20
1034 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
Table 1. Population Characteristics of Preeclamptic and Control Women
Maternal age (y) Gestational age (wk) Parity Diastolic blood pressure (mmHg) Hemoglobin (g/dL) Hematocrit (L/L) Platelet count (109/L)† Serum creatinine (mol/L) Serum uric acid (mmol/L) Serum aspartate aminotransferase (IU/L) Serum alanine aminotransferase (IU/L) Serum lactic dehydrogenase (IU/L) Proteinuria (g/L)
Control (n ⫽ 21)
Preeclamptic (n ⫽ 22)
32 (22– 42) 38⫹5 (37⫹3– 41⫹3) 1 (0 –3) 75 (58 – 85)
29 (22–36) 34⫹1 (26⫹3–38⫹2)* 0 (0 –2)† 100 (95–120)†
11.9 (10.8 –13.5) 0.36 (0.32– 0.41) 217 (150 –301) 60 (54 – 83) 0.28 (0.17– 0.36) 7 (4 –11)
12.9 (10.1–14.5) 0.37 (0.30 – 0.40) 61 (31–305)† 71 (52–108)† 0.26 (0.22– 0.55) 397 (7– 666)† 353 (5–562)†
10 (5–14) 208 (182–302)
1189 (206 –1873)† 0.97 (0.3–27.6)
Values are given as median (range). * P ⬍ .005. † P ⬍ .001 preeclamptic compared with control pregnancy.
minutes. Tissue was stored at ⫺20C. Upon use, tissue fragments (20 –100 mg) were thawed and homogenized in 10 volumes ice-cold homogenizing buffer (250 mM sucrose, 20 mM Tris-HCl, 1 mM dithriothreitol, pH 7.4) by 10 strokes in a small glass-glass homogenizer. The homogenates were centrifuged at 150,000 g at 4C for 1 hour. Supernatants were frozen in liquid nitrogen and stored at ⫺20C in small portions. Protein contents were assayed in triplicate by the method of Lowry et al, using bovine serum albumin as the standard.19 Total glutathione S-transferase isoenzyme activity was determined in triplicate according to Habig et al,20 using 1-chloro-2,4-dinitrobenzene as substrate. Specific glutathione S-transferase isoenzyme levels— classes alpha, mu, pi, and theta (T1-1 and T2-2)—were determined on Western blots essentially, as described by Peters et al.21 In short, cytosolic fractions (glutathione S-transferase alpha, mu, pi: 100 g protein for both placenta and decidua; glutathione S-transferase T1-1: 100 g cytosolic protein for placenta and 25 g for decidua; glutathione S-transferase T2-2: 100 g cytosolic protein for placenta and 50 g for decidua) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (11% acrylamide, w/v), and subsequently to Western blotting, using a semidry blotting system (Novablot II; Pharmacia, Uppsala, Sweden). Control and experimental samples were run on the same blots. Western blots were incubated with monoclonal antibodies against human glutathione Stransferase classes alpha, mu, pi, and T1-1, respectively, and a polyclonal antibody against human glutathione
Obstetrics & Gynecology
Table 2. Glutathione S-Transferase Alpha, Mu, Pi, and Theta Isoenzymes and Enzyme Activity in Decidua and Placenta of Preeclamptic and Control Patients Decidua GST isoenzymes GST GST GST GST GST GST
alpha mu pi T1-1 T2-2 activity
Placenta
Control pregnancy (n ⫽ 21)
Preeclamptic pregnancy (n ⫽ 22)
Control pregnancy (n ⫽ 21)
Preeclamptic pregnancy (n ⫽ 22)
0 (0 –202) 799 (0 –2972)† 2185 (503– 6578) 779 (0 –2437)* 899 (187–1729)* 172 (54 – 624)
0 (0 –141)* 873 (0 –2136)* 1268 (524 –3925)‡ 864 (404 –2926)§ 829 (164 –2278)§ 229 (121– 460)
86 (0 –909) 247 (0 –1180) 2169 (893–3929) 183 (0 –363) 537 (58 –958) 169 (74 – 472)
37 (0 – 601) 1412 (0 –1436) 1543 (681–2967)‡ 911 (30 –944) 315 (35–1133) 224 (94 –337)
GST ⫽ glutathione S-transferase. Results are presented as median (range). Glutathione S-transferase quantities are expressed in ng/mg protein, and glutathione S-transferase enzyme activity in nanomol per minute per milligram protein (nmol/min/mg protein). * P ⬍ .001. † P ⬍ .005. ‡ P ⬍ .005 (signed-rank test), preeclamptic compared with control pregnancy (decidual or placental tissue). § P ⬍ .0001 (Mann-Whitney U test), decidual compared with placental tissue within each group.
S-transferase T2-2. Antibodies against classes alpha (A1-1, A1-2, A2-2), mu (M1-1), and pi (P1-1) were developed in our own laboratory.21–23 Antibodies against glutathione S-transferase classes T1-1 and T2-2 were obtained from Prof. Dr. E. I. Juronen, Tartu, Estonia, and Prof. Dr. P. G. Board, Canberra, Australia, respectively. The specific binding of the primary antibodies to the isoenzymes was quantified after incubation with secondary antibodies: peroxidase-conjugated rabbit antimouse immunoglobulin (Dakopatts, Glostrup, Denmark) for glutathione S-transferase classes alpha, mu, pi, and T1-1, or peroxidaseconjugated swine antirabbit immunoglobulin (Dakopatts) for glutathione S-transferase T2-2, respectively. Subsequently, development of the peroxidase label took place with 4-chloro-1-naphthol (Sigma-Alderich, Zwijndrecht, the Netherlands) and hydrogen peroxide (Merck, Darmstadt, Germany) for glutathione Stransferase alpha and pi. Staining for the other glutathione S-transferase isoenzymes was performed using 0.1% 3, 3-diaminobenzidine (Sigma-Alderich, Zwijndrecht, the Netherlands) in phosphate-buffered saline containing 0.01% hydrogen peroxide as peroxidase substrate, and imidazole (0.34 g/L) and cobalt-chloride 6H2O (0.26 g/L) for enhancement of staining intensity. Staining intensity on immunoblots was quantified using a laser densitometer (Ultroscan XL; LKB, Bromma, Sweden). For calculating the relative amounts of glutathione S-transferase isoenzymes, known amounts of purified cytosolic glutathione S-transferases from human liver (alpha, mu, T1-1, T2-2) or placenta (pi) were run in parallel with the experimental samples and served as standards for the calculation of isoenzyme levels in the cytosolic fractions (values in ng/mg protein). Interassay variability of the immunoblot procedure was ⫾5%.
VOL. 94, NO. 6, DECEMBER 1999
The Mann-Whitney U test with Bonferroni correction for multiple comparison was used to assess statistical significance of differences between groups. The signedrank test with Bonferroni correction was used to assess statistical significance between placental and decidual tissue within groups. P ⱕ .004 (.05/12 tests) was considered significant.
Results Glutathione S-transferase enzyme activities and isoenzyme contents in placental and decidual tissues of both preeclamptic and control pregnancies are shown in Table 2. Placental tissue of controls contained the highest levels of glutathione S-transferase pi. Glutathione S-transferase T2-2 content was approximately 17% of the total amount of placental glutathione S-transferases determined here. The levels of other glutathione Stransferase isoforms were low. The glutathione Stransferase isoenzyme pattern in decidual tissue of controls differed from that in placental tissue. Glutathione S-transferase pi also was expressed at highest levels; however, classes T2-2, mu, and T1-1 also were present in considerable amounts (19, 17, and 17% of the total amount of glutathione S-transferases, respectively). Class alpha glutathione S-transferase was undetectable in the decidua of most women. Decidual glutathione S-transferases mu, T1-1, and T2-2 levels were significantly higher compared with placental levels in both preeclamptic and control groups. In contrast, decidual levels of glutathione S-transferase alpha were lower than placental levels. No differences were found between placental and decidual glutathione Stransferase pi levels in preeclamptic or control groups. Preeclamptic women showed statistically lower glutathione S-transferase pi levels in placental and decid-
Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
1035
Figure 1. Glutathione S-transferase pi levels (ng/mg protein) in control women (solid circles) and preeclamptic women (open boxes). Horizontal line indicates median.
ual tissue compared with women in the control group (P ⫽ .004 and P ⫽ .002, respectively) (Figure 1). No significant differences in median tissue level in decidua and placenta between the preeclamptic and the control group could be demonstrated for classes alpha, mu, T1-1, and T2-2. The results are summarized in Table 2. Median placental and decidual glutathione Stransferase enzyme activity was not significantly different between study groups (Table 2).
Discussion In this study, we compared the main human glutathione S-transferase isoenzyme levels in placental and decidual tissue from controls and preeclamptic patients. We also included nine women in the preeclamptic group who concurrently had HELLP syndrome because this syndrome may be considered a severe variant or complication of preeclampsia.24 An imbalance between lipid peroxides, oxygen free
1036 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
radicals, and other toxins on the one site, and radical scavengers and detoxificating systems on the other, might play an important role in the etiology and pathophysiology of preeclampsia.1,13 In preeclampsia, the oxidant/antioxidant balance is tipped in favor of oxidants at the expense of antioxidants, because in plasma, elevated levels of maternal lipid peroxides were measured.15 Serum-free fatty acids and triglycerides also were elevated in preeclampsia and contribute to the pool of lipid peroxides as they are oxidized.25 Several important antioxidants are decreased significantly in women with preeclampsia, which would contribute to the increased oxidative damage.15 In normal pregnancies, total glutathione S-transferase isoenzymes appeared to be higher in decidua than in placenta. High decidual levels of these isoenzymes may indicate a pronounced role of decidual tissue in protecting the mother or the fetus against damage by free radical peroxides and other toxic compounds produced by placental or fetomaternal interface.
Obstetrics & Gynecology
Glutathione S-transferase pi is the main glutathione S-transferase isoenzyme in placental16 and decidual tissue. Therefore, it is likely that glutathione Stransferase pi contributes greatly to the prevention of fetal damage.8 Both placental and decidual glutathione S-transferase pi levels were statistically lower in preeclamptic pregnancy compared with glutathione Stransferase pi levels in normotensive pregnancy. Decreased placental and decidual glutathione Stransferase pi levels might be related to a decreased effective capacity of the glutathione/glutathione Stransferase system in preeclamptic women. Placental and decidual glutathione S-transferase enzyme activity in preeclamptic and normotensive pregnancies, however, was not different, which is in agreement with an earlier report on placental glutathione S-transferase activity by Poranen et al.26 Numerous studies, recently reviewed by Knapen et al,27 have shown that the expression of glutathione S-transferase may be affected by factors such as diet. Nijhoff et al28 found that diets rich in cruciferous vegetables increased plasma glutathione S-transferase pi and alpha levels. Although dietary differences might have influenced our results, no studies are available concerning possible dietary effects on placental and decidual glutathione S-transferase expression. Other factors that might have influenced our results, given the difference between our study and control groups, are parity and gestational age. Di Ilio et al29 found that human placental glutathione S-transferase activity declined significantly as pregnancy advanced; however, whether or not qualitative and quantitative changes occur in glutathione S-transferase isoenzymes during pregnancy has been studied only in mice.30 In the last 5 days of pregnancy in the mouse, an increment in placental alpha and pi class glutathione S-transferase occurred, whereas class mu glutathione S-transferase decreased.30 Unfortunately, the effects of parity on glutathione S-transferase levels and activity have never been studied.
References 1. Hubel CA, Roberts JM, Taylor RN, Musci TJ, Rogers GM, Mc Laughlin M. Lipid peroxidation in pregnancy: New perspectives on preeclampsia. Am J Obstet Gynecol 1989;160:1025–34. 2. Williams DJ, de Swiet M. The pathophysiology of preeclampsia; review article. Intensive Care Med 1997;23:620 –9. 3. Higgs GA, Vane JR. Inhibition of cyclo-oxygenase and lipoxygenase. Br Med Bull 1983;39:235–70. 4. Sies H. Oxidative stress: Introductionary remarks. In: Sies H, ed. Oxidative stress. London: Academic Press, 1985:1– 8. 5. Halliwell B. Free radicals, antioxidants, and human disease: Curiosity, cause or consequence? Lancet 1994;344:721– 4. 6. Beckett GJ, Hayes JD. Glutathione S-transferases: Biomedical applications. Adv Clin Chem 1993;30:281–380.
VOL. 94, NO. 6, DECEMBER 1999
7. Hayes JD, Pulford DJ. The glutathione S-transferase supergene family: Regulation of GST and the contribution of isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 1995;30:445– 600. 8. Pacifici GM, Franchi M, Colizzi C, Guiliani L, Rane A. Glutathione S-transferase in humans: Development and tissue distribution. Arch Toxicol 1988;61:265–9. 9. Uotila JT, Tuimala RJ, Aarnio TM, Pyykko¨ KA, Uhotopa MO. Findings on lipid peroxidation and antioxidant function in hypertensive complications of pregnancy. Br J Obstet Gynaecol 1993;100: 270 – 6. 10. Walsh SW, Wang Y. Deficient glutathione peroxidase activity in preeclampsia is associated with increased placental production of thromboxane and lipid peroxides. Am J Obstet Gynecol 1993;169: 1456 – 61. 11. Kabi BC, Goel N, Rao YN, Tripathy R, Tempe A, Thakur AS. Levels of erythrocyte malonyldialdehyde, vitamin E, reduced glutathione, G6PD activity and plasma urate in patients of pregnancy induced hypertension. Indian J Med Res 1994;100:23–5. 12. Wang Y, Walsh SW. Antioxidant activities and mRNA expression of superoxide dismutase, catalase and glutathione peroxidase in normal and preeclamptic pregnancies. J Soc Gynecol Invest 1996; 3:179 – 84. 13. Loverro G, Greco P, Capuano F, Carone D, Cormio G, Selvaggi L. Lipid peroxidation and antioxidant enzyme activity in pregnancy complicated with hypertension. Eur J Obstet Gynecol Reprod Biol 1996;70:123–7. 14. Knapen MCFM, Mulder TPJ, Rooij van IALM, Peters WHM, Steegers EAP. Low whole blood glutathione levels in pregnancies complicated by preeclampsia or the hemolysis, elevated liver enzymes, low platelets syndrome. Obstet Gynecol 1998;92:1012–5. 15. Walsh SW. Maternal-placental interactions of oxidative stress and antioxidants in preeclampsia. Semin Reprod Endocrinol 1998;16: 93–104. 16. Awasthi YC, Sharma R, Singhal SS. Human glutathione Stransferases; minireview. Int J Biochem 1994;26:295–308. 17. Davey DA, MacGillivray I. The classification and definition of hypertensive disorders of pregnancy. Am J Obstet Gynecol 1988; 158:892– 8. 18. Sibai BM. The HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets): Much ado about nothing? Am J Obstet Gynecol 1990;162:311– 6. 19. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurements with the folin phenol reagent. J Biol Chem 1951;193:265–75. 20. Habig WH, Pabst MJ, Jacoby WB. Glutathione S-transferases, the first enzymatic step in mercapturic acid formation. J Biol Chem 1974;249:7130 –9. 21. Peters WHM, Boon CEW, Roelofs HMJ, Wobbes T, Nagengast FM, Kremers PG. Expression of drug metabolizing enzymes and P-170 glycoprotein in colorectal carcinoma and normal mucosa. Gastroenterology 1992;103:448 –55. 22. Peters WHM, Nagengast FM, Wobbes T. Glutathione Stransferases in normal and cancerous human colon tissue. Carcinogenesis 1989;10:2371– 4. 23. Peters WHM, Kock L, Nagengast FM, Roelofs HJM. Immunodetection with a monoclonal antibody of glutathione S-transferase Mu in patients with and without carcinomas. Biochem Pharmacol 1990;39:591–7. 24. Geary M. The HELLP syndrome. Br J Obstet Gynaecol 1997;104: 887–91. 25. Hubel CA, McLaughlin MK, Evans RW, Hauth BA, Sims CJ, Roberts JM. Fasting serum triglycerides, free fatty acids, and malondialdehyde are increased in preeclampsia, are positively
Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
1037
26.
27.
28.
29.
30.
correlated, and decrease within 48 hours post partum. Am J Obstet Gynecol 1996;174:975– 82. Poranen AK, Ekblad U, Uotila P, Ahotupa M. Lipid peroxidation and antioxidants in normal and preeclamptic pregnancies. Placenta 1996;17:401–5. Knapen MCFM, Zusterzeel PLM, Peters WHM, Steegers EAP. Glutathione and glutathione-related enzymes in reproduction: A review. Eur J Obstet Gynecol Reprod Biol 1999;82:171– 84. Nijhoff WA, Mulder TPJ, Verhagen H, van Poppel G, Peters WHM. Effects of consumption of Brussels sprouts on plasma and urinary glutathione s-transferase class-␣ and - in humans. Carcinogenesis 1995;16:955–7. Di Ilio C, Polidoro G, Arduini A, Muccini M, Federici G. Glutathione peroxidase, glutathione reductase, glutathione S-transferase and y-glutamyl-transpeptidase activities in human early pregnancy placenta. Biochem Med 1983;29:143– 8. Di Ilio C, Tiboni GM, Sacchetta P, Angelucci S, Bucciarelli T, Bellati U, et al. Time-dependant and tissue-specific variations of glutathione S-transferase activity during gestation in the mouse. Mech Ageing Dev 1995;78:47– 62.
1038 Zusterzeel et al
Glutathione S-Transferase in Preeclampsia
Address reprint requests to:
Eric A. P. Steegers, MD, PhD Department of Obstetrics & Gynecology University Hospital Nijmegen PO Box 9101 Nijmegen, 6500 HB The Netherlands E-mail: [email protected]
Received December 28, 1998. Received in revised form May 3, 1999. Accepted May 13, 1999.
Copyright © 1999 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Obstetrics & Gynecology