- Email: [email protected]
Placental Mitochondrial DNA and Respiratory Chain Enzymes in the Etiology of Preeclampsia
Placental Mitochondrial DNA and Respiratory Chain Enzymes in the Etiology of Preeclampsia KLAUS VUORINEN, MD, PhD, ANNE REMES, MD, PhD, RAIJA SORMUNEN, PhD, JUHA TAPANAINEN, MD, PhD, AND ILMO E. HASSINEN, MD, PhD Objective: To evaluate the occurrence of the most common mutations and deletions in mitochondrial DNA and deficiencies in the enzyme complexes of the mitochondrial respiratory chain in placentas from preeclamptic women. Methods: Mitochondria were isolated from the placentas of 17 preeclamptic or 25 control women, and the activities of mitochondrial respiratory chain complexes were measured. Deletions and three common point mutations of mitochondrial DNA were searched for by the Southern blot and polymerase chain reaction (PCR) methods from the same placentas. Results: Mean (6 standard deviation) mitochondrial respiratory chain enzyme complex activities in placentas on protein basis (nmol/min/mg of protein) were similar in preeclamptics and controls (nicotinamide adenine dinucleotide, reduced form– ubiquinone oxidoreductase 25.84 6 9.29 versus 31.02 6 7.52; nicotinamide adenine dinucleotide, reduced form– cytochrome-c oxidoreductase 77.88 6 42.24 versus 104.06 6 56.73; succinate– cytochrome-c oxidoreductase 57.90 6 13.83 versus 64.44 6 20.16; cytochrome-c oxidase 106.43 6 35.46 versus 128.37 6 48.64, respectively) and they were similar also when referenced to the mitochondrial marker enzyme citrate synthase. The sample sizes in both patient and control groups were found to be large enough by post hoc test. Large-scale deletions or the common 5-kb and 7.4-kb deletions were not detected, even at the sensitivity level of PCR. The three most common point mutations were not found in either control or preeclamptic placental samples. Conclusion: Common mitochondrial DNA mutations seem to play no major role in the universal etiology of preeclampsia, as assessed by analysis of the mitochondrial genome and respiratory chain enzyme activities in vitro. This does not exclude possible alterations in the energy state of the preeclamptic placenta. (Obstet Gynecol 1998;91:950 –5. © 1998 by The American College of Obstetricians and Gynecologists.)
From the Departments of Obstetrics and Gynecology, Neurology, and Pathology, Oulu University Hospital; and the Department of Medical Biochemistry, University of Oulu, Oulu, Finland. Supported (IEH) by grants from the Health Research Council of the Academy of Finland and the Sigrid Juselius Foundation.
950 0029-7844/98/$19.00 PII S0029-7844(98)00081-7
It has been suggested recently that mitochondrial dysfunction may be involved in the etiology of preeclampsia. Among the first observations were those of Shanklin and Sibai,1 who described ultrastructural changes in the mitochondria of the myometrium in preeclampsia. Barton et al2 found similar ultrastructural changes in endomyocardial biopsy specimens from a woman with preeclampsia. The case of a patient with mitochondrial disease who developed preeclampsia has been reported,3 and a high incidence of preeclampsia, eclampsia, and stillborn infants has been observed in a family with a known mitochondrial disorder.4 Furthermore, mitochondrial cytochrome-c oxidase activity has been found to be decreased in placentas from preeclamptic women, along with levels of the messenger RNA for subunit I of the same enzyme.5 Mitochondria depend on two separate genomes, the nuclear genome and the mitochondrial genome. Mitochondrial DNA, a 16.5-kb circular molecule, encodes 13 proteins of the mitochondrial respiratory chain, so that only succinate dehydrogenase is totally devoid of mitochondrially encoded proteins. It also encodes two ribosomal RNAs and all 22 transfer RNAs necessary for mitochondrial protein synthesis.6 Each mitochondrion has two to ten copies of mitochondrial DNA; therefore, there are more than 1000 copies in a cell. The mutation frequency in mitochondrial DNA is several times higher than that in nuclear genes, and such mutations may lead to impaired mitochondrial transcription or the formation of a nonfunctional enzyme and thereby to a decrease in respiratory chain enzyme activity. Several mitochondrial DNA mutations have been found to be associated with some neuromuscular diseases. Some of the most common are the nucleotide 3243 G3 A mutation in transfer RNALeu(UUR), leading to the mitochondrial encephalopathy with lactic acidosis and strokelike episodes disease; the nucleotide 8344 A3 G mutation in the transfer RNALys, causing the myoclonus-epilepsy and ragged-red fibers syndrome; and the
Obstetrics & Gynecology
nucleotide 8993 T3 G mutation in the adenosine triphosphate synthase subunit 6 gene, resulting in neurogenic muscle weakness, ataxia, and retinitis pigmentosa syndrome. Large-scale deletions, several point mutations in both structural and transfer RNA genes, and duplications and depletion (decrease in total mitochondrial DNA) have been described.7 In mitochondrial diseases, both wild-type and mutant mitochondrial DNA can exist in one cell (heteroplasmy), and in some cases the degree of this heteroplasmy can correlate with clinical symptoms. The placenta is responsible for supplying oxygen and substrates to the fetus. The function of placental cells is dependent on energy supplied by the mitochondria. Mutations in the genes coding for respiratory chain enzymes may lead to a decrease in mitochondrial activity, causing energy depletion and impairment of placental function. A mitochondrial mutation can cause several syndromes, depending on the percentage of mutated DNA in the cell and the presence of background mutations. We hypothesized that mitochondrial DNA mutations leading to impaired function of the respiratory chain may have a role in the etiology of preeclampsia, and we investigated in more detail the activities of mitochondrial DNA-coded respiratory chain enzymes and the three most frequently occurring mitochondrial DNA mutations and deletions in placentas from preeclamptic women.
The study design was approved by the Ethical Committee of the Medical Faculty of Oulu University, and written informed consent was obtained from the subjects. Because of varying admission hours and because mitochondrial isolation must be performed immediately and lasts up to 2 hours, consecutive recruitment of the patients and controls was not possible; therefore, the study consisted of convenience samples of 17 cases of mild or moderate preeclampsia (11 cases), severe preeclampsia (five cases), or eclampsia (one case) and samples of 25 controls with no preeclampsia or any history of mitochondrial disease in the family; all women were admitted to the Oulu University Hospital between August 1995 and July 1996. Eight control cases and one mild preeclampsia case were excluded from the enzyme activity measurements because of autolysis of the placentas; thus, 16 study and 17 control placental samples were analyzed. Mitochondrial DNA analysis was performed on all 42 placentas. The delivery was either spontaneous or induced vaginal delivery or elective cesarean delivery. In the control group, elective cesarean deliveries were performed because of fetal macrosomia, poor pelvic diameter of the mother, or
twin pregnancy complicated by preeclampsia (only one placenta). The number of cesarean deliveries among the controls was high but matched that in the preeclampsia group. The criteria for preeclampsia were edema of the upper body, proteinuria of 300 mg or more per day, and blood pressure (BP) over 140/90 mmHg or an increase in BP of 30/15 mmHg over the first-trimester values. Villous portions of the placentas were acquired immediately after delivery, and mitochondria were isolated immediately by a modification of the method described by Drouin.8 Ten grams of placental tissue was cut into small pieces with scissors, washed three times in ice-cold 0.15 mol/L NaCl, and homogenized in an Ultra-Turrax (Ika Verk, Rheinstetten, Germany) homogenizer in 10 mL of ice-cold buffer. After the isolation procedure,8 the last pellet was resuspended in 1–2 mL of 0.25 mol/L sucrose and 0.1 mmol/L ethylenediaminetetra-acetic acid and kept at –20C until analyzed. The frozen and thawed suspension was sonicated just before the enzyme assays, which were performed within 2 weeks of isolation of the mitochondria. The reduced form of nicotinamide adenine dinucleotide– ubiquinone oxidoreductase was assayed using ubiquinone1 as an electron acceptor,9 and the reduced form of nicotinamide adenine dinucleotide– cytochrome-c reductase and succinate– cytochrome-c reductase activities were measured as described previously.10 Cytochrome-c oxidase activity was assayed using a previously described method11 and expressed as the rate in the presence of 29 mmol/L ferrocytochrome c, which was the initial concentration in the cuvette. Citrate synthase activity was measured as described previously.12 Protein was measured using the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). Electron microscopy of four consecutive control and three consecutive preeclamptic mitochondrial homogenates was performed to evaluate their purity. Immediately after isolation of the mitochondria, 1 mL of homogenate was centrifuged at 13,500 3 g and the pellet was fixed in glutaraldehyde at 4C and studied by electron microscopy. The electron micrographs were analyzed by computerized planimetry (MCID/M2 program, Imaging Research Inc., St. Catharines, Ontario, Canada) to estimate the amount of mitochondria and other cellular organelles in the mitochondrial suspension. The area of mitochondria (pixels) and nonmitochondrial material (pixels) was divided by the total area of the electron micrograph (pixels) to determine the relative areas (unitless). A small portion of placenta was quick-frozen in liquid nitrogen for mitochondrial DNA analysis. Total DNA was isolated by the standard SDS-proteinase K
VOL. 91, NO. 6, JUNE 1998
Vuorinen et al
Materials and Methods
Mitochondrial Function in Preeclampsia
951
method.13 For Southern blotting, 4 mg of total DNA was digested with the restriction enzyme Pvu II14 and 32 P-labeled human mitochondrial DNA was used as the hybridization probe. The most common 5-kb (from nucleotide 8469 to 13,447) and 7.4-kb deletions (from nucleotide 8637 to 16,084) were analyzed by first amplifying the mitochondrial DNA with a polymerase chain reaction (PCR) as described previously.15 The transfer RNALeu(UUR) (nucleotide 3243) mutation was detected by PCR and digestion by the restriction enzyme ApaI. Heteroduplex formation may impede the detection of very small amounts of a point mutation,11 and therefore the data were confirmed by a modified minisequencing method.16 The point mutation at nucleotide 8344 was screened by BglI restriction enzyme digestion of a PCR product.17 The nucleotide 8993 mutation was screened by adding a PCR amplification step to the AvaI restriction enzyme digestion method of Harding et al.18 Statistical analysis was performed using Student t test for independent means without adjustment for multiple comparisons. P # .05 was considered statistically significant. The results are given as mean 6 standard deviation (SD). The sample sizes were tested post hoc and determined to be adequate for the purposes of the present study. The equation used was 2 z @~z 2a z v !/d 0# 2 in which z2a denotes second fractile of the standardized gaussian frequency, which has 1.96 as its numerical value. v denotes the previously estimated SD of the future study. d0 denotes the mean difference expected to be the clinically significant difference between the two groups. Taking the SD (v) of each measured parameter as the d0, the number of experiments in each group must be more than eight.19
Results Medical and obstetric data from the mothers and infants are shown in Table 1. Gestational age was significantly lower in the preeclampsia group than in the control group (265 6 20 versus 279 6 8 days, respectively; P , .05) due to the progressive disease and the obligatory induction of labor, and the weights of infants born to women with preeclampsia tended to be lower than those of infants born to controls when matched for gestational age (3346 6 749 versus 3849 6 571 g), but this difference did not achieve statistical significance. Severe fetal growth restriction (FGR) was present in two patients with severe preeclampsia. One patient in the preeclampsia group developed eclampsia. The mitochondrial respiratory chain enzyme activities (Table 2) tended to be lower on a protein basis in the
952 Vuorinen et al
Mitochondrial Function in Preeclampsia
Table 1. Clinical Characteristics
Gestational age (d) Maternal age (y) Nulliparous Vaginal delivery† Cesarean delivery Birth weight of infant (g) Placental weight (g)
Controls (n 5 17)
Patients (n 5 16)
279 6 8 31.9 6 5.7 6 6 11 3849 6 571
265 6 20* 28.3 6 6.1 11 8 8 3346 6 749‡ (n 5 11) 607 6 141 (n 5 10)
711 6 163
Data are presented as mean 6 standard deviation or n. * P , .05 compared with control group. † Vaginal delivery was either spontaneous or induced by vaginal prostaglandin or intravenous oxytocin. ‡ P 5 .058 compared with control group. The birth weight of the infant and weight of the placenta were included if the gestational age was above 37 weeks but were not included in one full-term twin pregnancy.
preeclamptic placental mitochondria. This was also the case for the mitochondrial marker enzyme citrate synthase, which is nonmitochondrially encoded. However, the differences were not statistically significant. When the respiratory enzyme activities were referenced to the activity of citrate synthase, the tendency disappeared (Table 2). This tendency could be explained by a different degree of purity of mitochondrial preparations in the two groups, and therefore seven mitochondrial suspensions were examined by electron microscopy, which revealed rather crude suspensions containing some cellular debris (data not shown). The relative areas of mitochondria in pellet from patients and controls were 0.0999 6 0.0391 and 0.0618 6 0.0009 and the relative areas of nonmitochondrial material were 0.2298 6 0.0793 and 0.274 6 0.0091, respectively. (Because of the small sample sizes, statistical evaluation was not performed). Therefore, there must have been an increase in mitochondrial protein not involving the soluble mitochondrial enzyme, citrate synthase, in placentas from preeclamptic pregnancies. In Southern blot analysis, only the normal-sized mitochondrial DNA signal and no large-scale deletions could be found in either the control or preeclamptic placenta samples (Figure 1). To validate the method, a sample from a patient with Kearns-Sayre syndrome is included in the blot depicted in Figure 2. The sensitive PCR method revealed no cases of either the most common mitochondrial DNA 5-kb deletion or the 7.4-kb deletion. The negative result is not due to insensitivity of the method, because both types of deletion could be detected in patients with verified mitochondrial disease and also in the skeletal or heart muscle of older people under the same amplification conditions (Figure 3). The point mutations at nucleotide 3243, 8344,
Obstetrics & Gynecology
Table 2. Respiratory Complex Activities of Placental Submitochondrial Particles Specific activity (nmol/min/mg protein)
Activity divided by citrate synthase activity
Enzyme
Controls (n 5 17)
Patients (n 5 16)
Controls (n 5 17)
Patients (n 5 16)
NADH– ubiquinone oxidoreductase NADH– cytochrome-c oxidoreductase Succinate– cytochrome-c oxidoreductase Cytochrome-c oxidase* Citrate synthase
31.02 6 7.52 104.06 6 56.73 64.44 6 20.16 128.37 6 48.64 184.72 6 40.89
25.84 6 9.29 77.88 6 42.24 57.90 6 13.83 106.43 6 35.46 157.76 6 37.06
0.175 6 0.056 0.630 6 0.374 0.353 6 0.099 0.712 6 0.252 1
0.172 6 0.073 0.502 6 0.236 0.376 6 0.101 0.690 6 0.240 1
NADH 5 reduced form of nicotinamide adenine dinucleotide. Data are presented as mean 6 standard deviation. There are no statistically significant differences between the controls and patients. * At substrate concentration of 29 mmol/L.
or 8993 of the Cambridge sequence map6 were not detectable in either the preeclampsia or the control samples (Figure 4).
Discussion The respiratory chain enzymes studied in the present case are coded by both the mitochondrial and nuclear genomes. Seven of the 41 subunits of the reduced form of nicotinamide adenine dinucleotide:ubiquinone oxidoreductase, one of the 11 subunits of ubiquinol– cytochrome-c reductase, and three of the 13 subunits of cytochrome-c oxidase are coded by mitochondrial DNA. In spite of recent claims of an involvement of impaired mitochondrial function and mitochondrial DNA in preeclampsia,4,5 our findings show that this condition is not related to defects of the mitochondrial respiratory chain, and no common mitochondrial DNA mutations or deletions were observed in this study.
In contrast to our findings, Furui et al5 suggested that in preeclampsia, the impaired placental function may lower the mitochondrial energy state, resulting in reduced mitochondrial transcription and protein synthesis. Thus, a vicious circle may ensue, causing progressive de-energization of the placental cells. This deduction nevertheless was based only on measurement of the activity of one respiratory chain enzyme (cytochrome c oxidase) and the level of messenger RNA of one of its mitochondrially coded subunits (subunit I). It has been reported recently that the preeclamptic placenta may be a major source of lipid peroxides20 and therefore a producer of free radicals. These may cause placental mitochondrial DNA damage. Torbergsen et al4 described a high incidence of stillborn infants, preeclampsia, and eclampsia in a family
Figure 1. Southern blot analysis of placental mitochondrial DNA. Lanes 1 and 2: control. Lane 3: patient with mitochondrial myopathy (Kearns-Sayre syndrome). Arrow indicates mitochondrial DNA with large-scale deletion. Lanes 4 –7: patients with preeclampsia.
Figure 2. Electrophoretograms of amplified mitochondrial DNA fragments for specific detection of deletions. Detection of the 5-kb deletion giving a 0.6-kb polymerase chain reaction product (arrow). Lanes 1– 4: patients with preeclampsia. Lane 5: elderly patient with dilated cardiomyopathy (positive control). Lane 6: negative control. Lane 7: DNA molecular size standard ladder.
VOL. 91, NO. 6, JUNE 1998
Vuorinen et al
Mitochondrial Function in Preeclampsia
953
Figure 3. Electrophoretograms of amplified mitochondrial DNA fragments for specific detection of deletions. Detection of the 7.4-kb deletion as a 0.6-kb polymerase chain reaction product (arrow). Lanes 1–3: patients with preeclampsia. Lanes 4 and 5: elderly patients with dilated cardiomyopathy (positive controls). Lane 6: negative control. Lane 7: DNA molecular size standard ladder.
carrying a hereditary mitochondrial disease causing neuromuscular symptoms, and recently the same group demonstrated two separate mitochondrial DNA point mutations in two families having a high incidence of preeclampsia and eclampsia.21 The common mitochondrial encephalopathy with lactic acidosis and strokelike episodes mutation has been found in one of these families, and it is heteroplasmic.4 In the other family, an A3 G mutation at nucleotide 12,308 in transfer RNA for leucine is homoplasmic but is found with a frequency of 1:30 also in the normal population.22 The findings of Torbergsen et al4 indicate that the incidence of preeclampsia and eclampsia increases in the presence of
Figure 4. Autoradiographs of restriction-enzyme digested amplified mitochondrial DNA fragments. Detection of the common nucleotide 3243 mutation. Lanes 1– 4: patients with preeclampsia. Lane 5: patient with nucleotide 3243 mutation (positive control). bp 5 base pairs.
954 Vuorinen et al
Mitochondrial Function in Preeclampsia
familiar mitochondrial disease. Our findings emphasize that mitochondrial mutation does not seem to be a universal cause of the disease. Observations concerning the frequencies of the common 5-kb deletion in placental mitochondrial DNA are at variance: Furui et al5 and Kurauchi et al23 did not find accumulation of the common 5-kb deletion or markedly reduced expression of the mitochondrial genome, but in a recent report by Hashimoto et al,24 appreciable levels of deleted mitochondrial DNA were detected in lateterm placentas. Our findings show no accumulation of common deletions in placentas from preeclamptic women, although it has been thought that preeclampsia results in preterm placental aging. Aging itself is associated with accumulation of mitochondrial DNA deletions.15 A lowered energy state in the placenta, as suggested by Furui et al,5 still may be involved in preeclampsia, although our findings suggest that the common mitochondrial DNA mutations are not its universal cause. The discrepancy between our findings and those of Furui et al5 is difficult to explain. It may be that in their study there were a larger number of severe preeclampsia cases including FGR. However, in our study, FGR was seen in two placentas with normal enzyme activity levels. Direct measurement of the placental energy state becomes necessary for verifying that the placental energy state is low in preeclampsia. 31P nuclear magnetic resonance spectroscopy, a noninvasive method, has been used to evaluate the energy state in perfused rat heart and in vitro human placenta and also could be applied in studies of preeclampsia.25,26 One intriguing finding was that the activities of the mitochondrial enzymes, including citrate synthase, tended to be lower in the preeclampsia group when expressed on protein basis, although the difference did not reach statistical significance. The tendency disappeared when the respiratory enzyme activities were divided by the citrate synthase activity. This may be explained by the reduced numbers of mitochondria in the placentas from preeclamptic women or by structural changes in the preeclamptic placenta affecting the yield and purity of the mitochondrial preparation. The electron micrographs showed that this minor difference in protein contamination probably is not due to particulate material. To decide between these possibilities, it would be necessary to measure enzymes in a total placental homogenate or to perform a mitochondrial count by electron microscopy of placental slices. On the basis of an investigation of a wide selection of known mitochondrial DNA mutations, our findings indicate that these are not a universal cause of preeclampsia. This screening was incomplete, however, and cannot rule out the possibility that other point
Obstetrics & Gynecology
mutations may be associated with mitochondrial disease. Furthermore, the accumulation of several mildly deleterious mutations in the mitochondrial genome may alter the severity of mitochondrial disease or even may be a cause of such disease. Because the whole mitochondrial genome was not sequenced, we cannot rule out the possibility that other or new mitochondrial DNA mutations may be associated with preeclampsia. However, unchanged mitochondrial enzyme activities do not support the role of pathogenic mitochondrial DNA mutations as a general cause of preeclampsia, although hereditary mitochondrial DNA mutation exposes the mother to hypertensive disease or preeclampsia, as shown previously.3,4,21
References 1. Shanklin DR, Sibai BM. Ultrastructural aspects of preeclampsia. II. Mitochondrial changes. Am J Obstet Gynecol 1990;163:943–53. 2. Barton JR, Hiett AK, O’Connor WM, Nissen SE, Greene JW. Endomyocardial ultrastructural findings in preeclampsia. Am J Obstet Gynecol 1991;165:389 –91. 3. Berkowitz K, Monteagudo A, Marks F, Jackson U, Baxi L. Mitochondrial myopathy in preeclampsia associated with pregnancy. Am J Obstet Gynecol 1990;162:146 –7. 4. Torbergsen T, Oian P, Mathiesen E, Borud O. Pre-eclampsia—A mitochondrial disease? Acta Obstet Gynecol Scand 1989;68:145– 8. 5. Furui T, Kurauchi O, Tanaka M, Mizutani S, Ozawa T, Tomoda Y. Decrease in cytochrome c oxidase and cytochrome oxidase subunit I messenger RNA levels in preeclamptic pregnancies. Obstet Gynecol 1994;84:283– 8. 6. Anderson S, Bankier AT, Barrell BG, de Bruijn MHL, Coulson AR, Drouin J, et al. Sequence and organization of human mitochondrial genome. Nature 1981;290:457– 65. 7. Wallace DC, Lott MT, Brown MD, Huoponen K, Torroni A. Report of the Committee on Human Mitochondrial DNA. In: Cuticchia AJ, ed. Human gene mapping 1995: A compendium. Baltimore, Maryland: Johns Hopkins University Press, 1995:910 –54. 8. Drouin J. Cloning of human mitochondrial DNA in Eschericia coli. J Mol Biol 1980;140:15–34. 9. Vuokila PT, Hassinen IE. NN9-dicyclohexylcarbodi-imidesensitivity of bovine heart mitochondrial NADH:ubiquinone oxidoreductase. Biochem J 1988;249:339 – 44. 10. Sottocasa GL, Kuylenstierna BO, Ernster L, Bergstrand A. An electron transport system associated with outer membrane of liver mitochondria: A biochemical and morphological study. J Cell Biol 1967;32:415–38. 11. Cooperstein SJ, Lazarow AA. Microspectrophotometric method for the determination of cytochrome oxidase. J Biol Chem 1951;189: 665–70. 12. Shepherd D, Garland PB. Citrate synthase from rat liver. In: Lowenstein JM, ed. Methods in enzymology. Vol 13. New York: Academic Press, 1969:11– 6. 13. Sambrook J, Maniatis T, Fritsch EF. Molecular cloning: A laboratory manual. New York: Cold Spring Harbor Laboratory, 1982. 14. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975;98:503–17. 15. Remes AM, Hassinen IE, Ika¨heimo MJ, Herva R, Hirvonen J, Peuhkurinen KJ. Mitochondrial DNA deletions in dilated cardiomyopathy: A clinical study employing endomyocardial sampling. J Am Coll Cardiol 1994;23:935– 42.
VOL. 91, NO. 6, JUNE 1998
16. Suomalainen A, Kollmann P, Octave JN, So¨derlund H, Syva¨nen AC. Quantitation of mitochondrial DNA carrying the tRNA8344Lys point mutation in myoclonus epilepsy and ragged-red-fiber disease. Eur J Hum Genet 1993;1:88 –95. 17. Zeviani M, Amati P, Bresolin N, Antozzi C, Piccolo G, Toscano A, et al. Rapid detection of the AG(8344) mutation of mtDNA in Italian families with myoclonus-epilepsy and ragged-red fibers (MERRF). Am J Hum Genet 1991;48:203–11. 18. Harding AE, Holt IJ, Sweeney MG, Brockington M, Davis MB. Prenatal diagnosis of mitochondrial DNA8993 T3 G disease. Am J Hum Genet 1992;50:629 –33. 19. Armitage P, Berry G. Statistical methods in medical research. 2nd ed. Oxford: Blackwell, 1987. 20. Morikawa S, Kurauchi O, Tanaka M, Yoneda M, Uchida K, Itakura A, et al. Increased mitochondrial damage by lipid peroxidation in trophoblast cells of preeclamptic placentas. Biochem Mol Biol Int 1997;41:767–75. 21. Folgerø T, Storbakk N, Torbergsen T, Øian P. Mutations in mitochondrial transfer ribonucleic acid genes in preeclampsia. Am J Obstet Gynecol 1996;174:1626 –30. 22. Houshmand M, Larsson NG, Holme E, Oldfors A, Tulinius MH, Andersen O. Automatic sequencing of mitochondrial tRNA genes in patients with mitochondrial encephalomyopathy. Biochim Biophys Acta 1994;1226:49 –55. 23. Kurauchi O, Furui T, Tanaka M, Mizutani S, Ozawa T, Tomoda Y. The study of mitochondrial gene modifications in human placenta. Placenta 1995;16:461–7. 24. Hashimoto K, Azuma C, Kamiura S, Taniguchi T, Shimoya K, Nobunago T, et al. Detection of DNA deletion in the late-term placenta. Horm Metab Res 1996;28:615–7. 25. Vuorinen K, Ylitalo K, Peuhkurinen K, Raatikainen P, Ala-Ra¨mi A, Hassinen IE. Mechanisms of ischemic preconditioning in rat myocardium. Roles of adenosine, cellular energy state, and mitochondrial F1F0 ATPase. Circulation 1995;91:2810 – 8. 26. Malek A, Miller RK, Mattison DR, Kennedy S, Panigel M, di Sant’ Agnese PA, et al. Energy charge monitoring via magnetic spectroscopy 31P in the perfused human placenta: Effects of cadmium, dinitrophenol and iodoacetate. Placenta 1996;17:496 –506.
Address reprint requests to:
Klaus Vuorinen, MD, PhD Department of Obstetrics and Gynecology Oulu University Hospital University of Oulu Kajaanintie 50 FIN-90220 Oulu Finland E-mail: [email protected]
Received June 3, 1997. Received in revised form January 30, 1998. Accepted February 13, 1998.
Copyright © 1998 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Vuorinen et al
Mitochondrial Function in Preeclampsia
955
From the Departments of Obstetrics and Gynecology, Neurology, and Pathology, Oulu University Hospital; and the Department of Medical Biochemistry, University of Oulu, Oulu, Finland. Supported (IEH) by grants from the Health Research Council of the Academy of Finland and the Sigrid Juselius Foundation.
950 0029-7844/98/$19.00 PII S0029-7844(98)00081-7
It has been suggested recently that mitochondrial dysfunction may be involved in the etiology of preeclampsia. Among the first observations were those of Shanklin and Sibai,1 who described ultrastructural changes in the mitochondria of the myometrium in preeclampsia. Barton et al2 found similar ultrastructural changes in endomyocardial biopsy specimens from a woman with preeclampsia. The case of a patient with mitochondrial disease who developed preeclampsia has been reported,3 and a high incidence of preeclampsia, eclampsia, and stillborn infants has been observed in a family with a known mitochondrial disorder.4 Furthermore, mitochondrial cytochrome-c oxidase activity has been found to be decreased in placentas from preeclamptic women, along with levels of the messenger RNA for subunit I of the same enzyme.5 Mitochondria depend on two separate genomes, the nuclear genome and the mitochondrial genome. Mitochondrial DNA, a 16.5-kb circular molecule, encodes 13 proteins of the mitochondrial respiratory chain, so that only succinate dehydrogenase is totally devoid of mitochondrially encoded proteins. It also encodes two ribosomal RNAs and all 22 transfer RNAs necessary for mitochondrial protein synthesis.6 Each mitochondrion has two to ten copies of mitochondrial DNA; therefore, there are more than 1000 copies in a cell. The mutation frequency in mitochondrial DNA is several times higher than that in nuclear genes, and such mutations may lead to impaired mitochondrial transcription or the formation of a nonfunctional enzyme and thereby to a decrease in respiratory chain enzyme activity. Several mitochondrial DNA mutations have been found to be associated with some neuromuscular diseases. Some of the most common are the nucleotide 3243 G3 A mutation in transfer RNALeu(UUR), leading to the mitochondrial encephalopathy with lactic acidosis and strokelike episodes disease; the nucleotide 8344 A3 G mutation in the transfer RNALys, causing the myoclonus-epilepsy and ragged-red fibers syndrome; and the
Obstetrics & Gynecology
nucleotide 8993 T3 G mutation in the adenosine triphosphate synthase subunit 6 gene, resulting in neurogenic muscle weakness, ataxia, and retinitis pigmentosa syndrome. Large-scale deletions, several point mutations in both structural and transfer RNA genes, and duplications and depletion (decrease in total mitochondrial DNA) have been described.7 In mitochondrial diseases, both wild-type and mutant mitochondrial DNA can exist in one cell (heteroplasmy), and in some cases the degree of this heteroplasmy can correlate with clinical symptoms. The placenta is responsible for supplying oxygen and substrates to the fetus. The function of placental cells is dependent on energy supplied by the mitochondria. Mutations in the genes coding for respiratory chain enzymes may lead to a decrease in mitochondrial activity, causing energy depletion and impairment of placental function. A mitochondrial mutation can cause several syndromes, depending on the percentage of mutated DNA in the cell and the presence of background mutations. We hypothesized that mitochondrial DNA mutations leading to impaired function of the respiratory chain may have a role in the etiology of preeclampsia, and we investigated in more detail the activities of mitochondrial DNA-coded respiratory chain enzymes and the three most frequently occurring mitochondrial DNA mutations and deletions in placentas from preeclamptic women.
The study design was approved by the Ethical Committee of the Medical Faculty of Oulu University, and written informed consent was obtained from the subjects. Because of varying admission hours and because mitochondrial isolation must be performed immediately and lasts up to 2 hours, consecutive recruitment of the patients and controls was not possible; therefore, the study consisted of convenience samples of 17 cases of mild or moderate preeclampsia (11 cases), severe preeclampsia (five cases), or eclampsia (one case) and samples of 25 controls with no preeclampsia or any history of mitochondrial disease in the family; all women were admitted to the Oulu University Hospital between August 1995 and July 1996. Eight control cases and one mild preeclampsia case were excluded from the enzyme activity measurements because of autolysis of the placentas; thus, 16 study and 17 control placental samples were analyzed. Mitochondrial DNA analysis was performed on all 42 placentas. The delivery was either spontaneous or induced vaginal delivery or elective cesarean delivery. In the control group, elective cesarean deliveries were performed because of fetal macrosomia, poor pelvic diameter of the mother, or
twin pregnancy complicated by preeclampsia (only one placenta). The number of cesarean deliveries among the controls was high but matched that in the preeclampsia group. The criteria for preeclampsia were edema of the upper body, proteinuria of 300 mg or more per day, and blood pressure (BP) over 140/90 mmHg or an increase in BP of 30/15 mmHg over the first-trimester values. Villous portions of the placentas were acquired immediately after delivery, and mitochondria were isolated immediately by a modification of the method described by Drouin.8 Ten grams of placental tissue was cut into small pieces with scissors, washed three times in ice-cold 0.15 mol/L NaCl, and homogenized in an Ultra-Turrax (Ika Verk, Rheinstetten, Germany) homogenizer in 10 mL of ice-cold buffer. After the isolation procedure,8 the last pellet was resuspended in 1–2 mL of 0.25 mol/L sucrose and 0.1 mmol/L ethylenediaminetetra-acetic acid and kept at –20C until analyzed. The frozen and thawed suspension was sonicated just before the enzyme assays, which were performed within 2 weeks of isolation of the mitochondria. The reduced form of nicotinamide adenine dinucleotide– ubiquinone oxidoreductase was assayed using ubiquinone1 as an electron acceptor,9 and the reduced form of nicotinamide adenine dinucleotide– cytochrome-c reductase and succinate– cytochrome-c reductase activities were measured as described previously.10 Cytochrome-c oxidase activity was assayed using a previously described method11 and expressed as the rate in the presence of 29 mmol/L ferrocytochrome c, which was the initial concentration in the cuvette. Citrate synthase activity was measured as described previously.12 Protein was measured using the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). Electron microscopy of four consecutive control and three consecutive preeclamptic mitochondrial homogenates was performed to evaluate their purity. Immediately after isolation of the mitochondria, 1 mL of homogenate was centrifuged at 13,500 3 g and the pellet was fixed in glutaraldehyde at 4C and studied by electron microscopy. The electron micrographs were analyzed by computerized planimetry (MCID/M2 program, Imaging Research Inc., St. Catharines, Ontario, Canada) to estimate the amount of mitochondria and other cellular organelles in the mitochondrial suspension. The area of mitochondria (pixels) and nonmitochondrial material (pixels) was divided by the total area of the electron micrograph (pixels) to determine the relative areas (unitless). A small portion of placenta was quick-frozen in liquid nitrogen for mitochondrial DNA analysis. Total DNA was isolated by the standard SDS-proteinase K
VOL. 91, NO. 6, JUNE 1998
Vuorinen et al
Materials and Methods
Mitochondrial Function in Preeclampsia
951
method.13 For Southern blotting, 4 mg of total DNA was digested with the restriction enzyme Pvu II14 and 32 P-labeled human mitochondrial DNA was used as the hybridization probe. The most common 5-kb (from nucleotide 8469 to 13,447) and 7.4-kb deletions (from nucleotide 8637 to 16,084) were analyzed by first amplifying the mitochondrial DNA with a polymerase chain reaction (PCR) as described previously.15 The transfer RNALeu(UUR) (nucleotide 3243) mutation was detected by PCR and digestion by the restriction enzyme ApaI. Heteroduplex formation may impede the detection of very small amounts of a point mutation,11 and therefore the data were confirmed by a modified minisequencing method.16 The point mutation at nucleotide 8344 was screened by BglI restriction enzyme digestion of a PCR product.17 The nucleotide 8993 mutation was screened by adding a PCR amplification step to the AvaI restriction enzyme digestion method of Harding et al.18 Statistical analysis was performed using Student t test for independent means without adjustment for multiple comparisons. P # .05 was considered statistically significant. The results are given as mean 6 standard deviation (SD). The sample sizes were tested post hoc and determined to be adequate for the purposes of the present study. The equation used was 2 z @~z 2a z v !/d 0# 2 in which z2a denotes second fractile of the standardized gaussian frequency, which has 1.96 as its numerical value. v denotes the previously estimated SD of the future study. d0 denotes the mean difference expected to be the clinically significant difference between the two groups. Taking the SD (v) of each measured parameter as the d0, the number of experiments in each group must be more than eight.19
Results Medical and obstetric data from the mothers and infants are shown in Table 1. Gestational age was significantly lower in the preeclampsia group than in the control group (265 6 20 versus 279 6 8 days, respectively; P , .05) due to the progressive disease and the obligatory induction of labor, and the weights of infants born to women with preeclampsia tended to be lower than those of infants born to controls when matched for gestational age (3346 6 749 versus 3849 6 571 g), but this difference did not achieve statistical significance. Severe fetal growth restriction (FGR) was present in two patients with severe preeclampsia. One patient in the preeclampsia group developed eclampsia. The mitochondrial respiratory chain enzyme activities (Table 2) tended to be lower on a protein basis in the
952 Vuorinen et al
Mitochondrial Function in Preeclampsia
Table 1. Clinical Characteristics
Gestational age (d) Maternal age (y) Nulliparous Vaginal delivery† Cesarean delivery Birth weight of infant (g) Placental weight (g)
Controls (n 5 17)
Patients (n 5 16)
279 6 8 31.9 6 5.7 6 6 11 3849 6 571
265 6 20* 28.3 6 6.1 11 8 8 3346 6 749‡ (n 5 11) 607 6 141 (n 5 10)
711 6 163
Data are presented as mean 6 standard deviation or n. * P , .05 compared with control group. † Vaginal delivery was either spontaneous or induced by vaginal prostaglandin or intravenous oxytocin. ‡ P 5 .058 compared with control group. The birth weight of the infant and weight of the placenta were included if the gestational age was above 37 weeks but were not included in one full-term twin pregnancy.
preeclamptic placental mitochondria. This was also the case for the mitochondrial marker enzyme citrate synthase, which is nonmitochondrially encoded. However, the differences were not statistically significant. When the respiratory enzyme activities were referenced to the activity of citrate synthase, the tendency disappeared (Table 2). This tendency could be explained by a different degree of purity of mitochondrial preparations in the two groups, and therefore seven mitochondrial suspensions were examined by electron microscopy, which revealed rather crude suspensions containing some cellular debris (data not shown). The relative areas of mitochondria in pellet from patients and controls were 0.0999 6 0.0391 and 0.0618 6 0.0009 and the relative areas of nonmitochondrial material were 0.2298 6 0.0793 and 0.274 6 0.0091, respectively. (Because of the small sample sizes, statistical evaluation was not performed). Therefore, there must have been an increase in mitochondrial protein not involving the soluble mitochondrial enzyme, citrate synthase, in placentas from preeclamptic pregnancies. In Southern blot analysis, only the normal-sized mitochondrial DNA signal and no large-scale deletions could be found in either the control or preeclamptic placenta samples (Figure 1). To validate the method, a sample from a patient with Kearns-Sayre syndrome is included in the blot depicted in Figure 2. The sensitive PCR method revealed no cases of either the most common mitochondrial DNA 5-kb deletion or the 7.4-kb deletion. The negative result is not due to insensitivity of the method, because both types of deletion could be detected in patients with verified mitochondrial disease and also in the skeletal or heart muscle of older people under the same amplification conditions (Figure 3). The point mutations at nucleotide 3243, 8344,
Obstetrics & Gynecology
Table 2. Respiratory Complex Activities of Placental Submitochondrial Particles Specific activity (nmol/min/mg protein)
Activity divided by citrate synthase activity
Enzyme
Controls (n 5 17)
Patients (n 5 16)
Controls (n 5 17)
Patients (n 5 16)
NADH– ubiquinone oxidoreductase NADH– cytochrome-c oxidoreductase Succinate– cytochrome-c oxidoreductase Cytochrome-c oxidase* Citrate synthase
31.02 6 7.52 104.06 6 56.73 64.44 6 20.16 128.37 6 48.64 184.72 6 40.89
25.84 6 9.29 77.88 6 42.24 57.90 6 13.83 106.43 6 35.46 157.76 6 37.06
0.175 6 0.056 0.630 6 0.374 0.353 6 0.099 0.712 6 0.252 1
0.172 6 0.073 0.502 6 0.236 0.376 6 0.101 0.690 6 0.240 1
NADH 5 reduced form of nicotinamide adenine dinucleotide. Data are presented as mean 6 standard deviation. There are no statistically significant differences between the controls and patients. * At substrate concentration of 29 mmol/L.
or 8993 of the Cambridge sequence map6 were not detectable in either the preeclampsia or the control samples (Figure 4).
Discussion The respiratory chain enzymes studied in the present case are coded by both the mitochondrial and nuclear genomes. Seven of the 41 subunits of the reduced form of nicotinamide adenine dinucleotide:ubiquinone oxidoreductase, one of the 11 subunits of ubiquinol– cytochrome-c reductase, and three of the 13 subunits of cytochrome-c oxidase are coded by mitochondrial DNA. In spite of recent claims of an involvement of impaired mitochondrial function and mitochondrial DNA in preeclampsia,4,5 our findings show that this condition is not related to defects of the mitochondrial respiratory chain, and no common mitochondrial DNA mutations or deletions were observed in this study.
In contrast to our findings, Furui et al5 suggested that in preeclampsia, the impaired placental function may lower the mitochondrial energy state, resulting in reduced mitochondrial transcription and protein synthesis. Thus, a vicious circle may ensue, causing progressive de-energization of the placental cells. This deduction nevertheless was based only on measurement of the activity of one respiratory chain enzyme (cytochrome c oxidase) and the level of messenger RNA of one of its mitochondrially coded subunits (subunit I). It has been reported recently that the preeclamptic placenta may be a major source of lipid peroxides20 and therefore a producer of free radicals. These may cause placental mitochondrial DNA damage. Torbergsen et al4 described a high incidence of stillborn infants, preeclampsia, and eclampsia in a family
Figure 1. Southern blot analysis of placental mitochondrial DNA. Lanes 1 and 2: control. Lane 3: patient with mitochondrial myopathy (Kearns-Sayre syndrome). Arrow indicates mitochondrial DNA with large-scale deletion. Lanes 4 –7: patients with preeclampsia.
Figure 2. Electrophoretograms of amplified mitochondrial DNA fragments for specific detection of deletions. Detection of the 5-kb deletion giving a 0.6-kb polymerase chain reaction product (arrow). Lanes 1– 4: patients with preeclampsia. Lane 5: elderly patient with dilated cardiomyopathy (positive control). Lane 6: negative control. Lane 7: DNA molecular size standard ladder.
VOL. 91, NO. 6, JUNE 1998
Vuorinen et al
Mitochondrial Function in Preeclampsia
953
Figure 3. Electrophoretograms of amplified mitochondrial DNA fragments for specific detection of deletions. Detection of the 7.4-kb deletion as a 0.6-kb polymerase chain reaction product (arrow). Lanes 1–3: patients with preeclampsia. Lanes 4 and 5: elderly patients with dilated cardiomyopathy (positive controls). Lane 6: negative control. Lane 7: DNA molecular size standard ladder.
carrying a hereditary mitochondrial disease causing neuromuscular symptoms, and recently the same group demonstrated two separate mitochondrial DNA point mutations in two families having a high incidence of preeclampsia and eclampsia.21 The common mitochondrial encephalopathy with lactic acidosis and strokelike episodes mutation has been found in one of these families, and it is heteroplasmic.4 In the other family, an A3 G mutation at nucleotide 12,308 in transfer RNA for leucine is homoplasmic but is found with a frequency of 1:30 also in the normal population.22 The findings of Torbergsen et al4 indicate that the incidence of preeclampsia and eclampsia increases in the presence of
Figure 4. Autoradiographs of restriction-enzyme digested amplified mitochondrial DNA fragments. Detection of the common nucleotide 3243 mutation. Lanes 1– 4: patients with preeclampsia. Lane 5: patient with nucleotide 3243 mutation (positive control). bp 5 base pairs.
954 Vuorinen et al
Mitochondrial Function in Preeclampsia
familiar mitochondrial disease. Our findings emphasize that mitochondrial mutation does not seem to be a universal cause of the disease. Observations concerning the frequencies of the common 5-kb deletion in placental mitochondrial DNA are at variance: Furui et al5 and Kurauchi et al23 did not find accumulation of the common 5-kb deletion or markedly reduced expression of the mitochondrial genome, but in a recent report by Hashimoto et al,24 appreciable levels of deleted mitochondrial DNA were detected in lateterm placentas. Our findings show no accumulation of common deletions in placentas from preeclamptic women, although it has been thought that preeclampsia results in preterm placental aging. Aging itself is associated with accumulation of mitochondrial DNA deletions.15 A lowered energy state in the placenta, as suggested by Furui et al,5 still may be involved in preeclampsia, although our findings suggest that the common mitochondrial DNA mutations are not its universal cause. The discrepancy between our findings and those of Furui et al5 is difficult to explain. It may be that in their study there were a larger number of severe preeclampsia cases including FGR. However, in our study, FGR was seen in two placentas with normal enzyme activity levels. Direct measurement of the placental energy state becomes necessary for verifying that the placental energy state is low in preeclampsia. 31P nuclear magnetic resonance spectroscopy, a noninvasive method, has been used to evaluate the energy state in perfused rat heart and in vitro human placenta and also could be applied in studies of preeclampsia.25,26 One intriguing finding was that the activities of the mitochondrial enzymes, including citrate synthase, tended to be lower in the preeclampsia group when expressed on protein basis, although the difference did not reach statistical significance. The tendency disappeared when the respiratory enzyme activities were divided by the citrate synthase activity. This may be explained by the reduced numbers of mitochondria in the placentas from preeclamptic women or by structural changes in the preeclamptic placenta affecting the yield and purity of the mitochondrial preparation. The electron micrographs showed that this minor difference in protein contamination probably is not due to particulate material. To decide between these possibilities, it would be necessary to measure enzymes in a total placental homogenate or to perform a mitochondrial count by electron microscopy of placental slices. On the basis of an investigation of a wide selection of known mitochondrial DNA mutations, our findings indicate that these are not a universal cause of preeclampsia. This screening was incomplete, however, and cannot rule out the possibility that other point
Obstetrics & Gynecology
mutations may be associated with mitochondrial disease. Furthermore, the accumulation of several mildly deleterious mutations in the mitochondrial genome may alter the severity of mitochondrial disease or even may be a cause of such disease. Because the whole mitochondrial genome was not sequenced, we cannot rule out the possibility that other or new mitochondrial DNA mutations may be associated with preeclampsia. However, unchanged mitochondrial enzyme activities do not support the role of pathogenic mitochondrial DNA mutations as a general cause of preeclampsia, although hereditary mitochondrial DNA mutation exposes the mother to hypertensive disease or preeclampsia, as shown previously.3,4,21
References 1. Shanklin DR, Sibai BM. Ultrastructural aspects of preeclampsia. II. Mitochondrial changes. Am J Obstet Gynecol 1990;163:943–53. 2. Barton JR, Hiett AK, O’Connor WM, Nissen SE, Greene JW. Endomyocardial ultrastructural findings in preeclampsia. Am J Obstet Gynecol 1991;165:389 –91. 3. Berkowitz K, Monteagudo A, Marks F, Jackson U, Baxi L. Mitochondrial myopathy in preeclampsia associated with pregnancy. Am J Obstet Gynecol 1990;162:146 –7. 4. Torbergsen T, Oian P, Mathiesen E, Borud O. Pre-eclampsia—A mitochondrial disease? Acta Obstet Gynecol Scand 1989;68:145– 8. 5. Furui T, Kurauchi O, Tanaka M, Mizutani S, Ozawa T, Tomoda Y. Decrease in cytochrome c oxidase and cytochrome oxidase subunit I messenger RNA levels in preeclamptic pregnancies. Obstet Gynecol 1994;84:283– 8. 6. Anderson S, Bankier AT, Barrell BG, de Bruijn MHL, Coulson AR, Drouin J, et al. Sequence and organization of human mitochondrial genome. Nature 1981;290:457– 65. 7. Wallace DC, Lott MT, Brown MD, Huoponen K, Torroni A. Report of the Committee on Human Mitochondrial DNA. In: Cuticchia AJ, ed. Human gene mapping 1995: A compendium. Baltimore, Maryland: Johns Hopkins University Press, 1995:910 –54. 8. Drouin J. Cloning of human mitochondrial DNA in Eschericia coli. J Mol Biol 1980;140:15–34. 9. Vuokila PT, Hassinen IE. NN9-dicyclohexylcarbodi-imidesensitivity of bovine heart mitochondrial NADH:ubiquinone oxidoreductase. Biochem J 1988;249:339 – 44. 10. Sottocasa GL, Kuylenstierna BO, Ernster L, Bergstrand A. An electron transport system associated with outer membrane of liver mitochondria: A biochemical and morphological study. J Cell Biol 1967;32:415–38. 11. Cooperstein SJ, Lazarow AA. Microspectrophotometric method for the determination of cytochrome oxidase. J Biol Chem 1951;189: 665–70. 12. Shepherd D, Garland PB. Citrate synthase from rat liver. In: Lowenstein JM, ed. Methods in enzymology. Vol 13. New York: Academic Press, 1969:11– 6. 13. Sambrook J, Maniatis T, Fritsch EF. Molecular cloning: A laboratory manual. New York: Cold Spring Harbor Laboratory, 1982. 14. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975;98:503–17. 15. Remes AM, Hassinen IE, Ika¨heimo MJ, Herva R, Hirvonen J, Peuhkurinen KJ. Mitochondrial DNA deletions in dilated cardiomyopathy: A clinical study employing endomyocardial sampling. J Am Coll Cardiol 1994;23:935– 42.
VOL. 91, NO. 6, JUNE 1998
16. Suomalainen A, Kollmann P, Octave JN, So¨derlund H, Syva¨nen AC. Quantitation of mitochondrial DNA carrying the tRNA8344Lys point mutation in myoclonus epilepsy and ragged-red-fiber disease. Eur J Hum Genet 1993;1:88 –95. 17. Zeviani M, Amati P, Bresolin N, Antozzi C, Piccolo G, Toscano A, et al. Rapid detection of the AG(8344) mutation of mtDNA in Italian families with myoclonus-epilepsy and ragged-red fibers (MERRF). Am J Hum Genet 1991;48:203–11. 18. Harding AE, Holt IJ, Sweeney MG, Brockington M, Davis MB. Prenatal diagnosis of mitochondrial DNA8993 T3 G disease. Am J Hum Genet 1992;50:629 –33. 19. Armitage P, Berry G. Statistical methods in medical research. 2nd ed. Oxford: Blackwell, 1987. 20. Morikawa S, Kurauchi O, Tanaka M, Yoneda M, Uchida K, Itakura A, et al. Increased mitochondrial damage by lipid peroxidation in trophoblast cells of preeclamptic placentas. Biochem Mol Biol Int 1997;41:767–75. 21. Folgerø T, Storbakk N, Torbergsen T, Øian P. Mutations in mitochondrial transfer ribonucleic acid genes in preeclampsia. Am J Obstet Gynecol 1996;174:1626 –30. 22. Houshmand M, Larsson NG, Holme E, Oldfors A, Tulinius MH, Andersen O. Automatic sequencing of mitochondrial tRNA genes in patients with mitochondrial encephalomyopathy. Biochim Biophys Acta 1994;1226:49 –55. 23. Kurauchi O, Furui T, Tanaka M, Mizutani S, Ozawa T, Tomoda Y. The study of mitochondrial gene modifications in human placenta. Placenta 1995;16:461–7. 24. Hashimoto K, Azuma C, Kamiura S, Taniguchi T, Shimoya K, Nobunago T, et al. Detection of DNA deletion in the late-term placenta. Horm Metab Res 1996;28:615–7. 25. Vuorinen K, Ylitalo K, Peuhkurinen K, Raatikainen P, Ala-Ra¨mi A, Hassinen IE. Mechanisms of ischemic preconditioning in rat myocardium. Roles of adenosine, cellular energy state, and mitochondrial F1F0 ATPase. Circulation 1995;91:2810 – 8. 26. Malek A, Miller RK, Mattison DR, Kennedy S, Panigel M, di Sant’ Agnese PA, et al. Energy charge monitoring via magnetic spectroscopy 31P in the perfused human placenta: Effects of cadmium, dinitrophenol and iodoacetate. Placenta 1996;17:496 –506.
Address reprint requests to:
Klaus Vuorinen, MD, PhD Department of Obstetrics and Gynecology Oulu University Hospital University of Oulu Kajaanintie 50 FIN-90220 Oulu Finland E-mail: [email protected]
Received June 3, 1997. Received in revised form January 30, 1998. Accepted February 13, 1998.
Copyright © 1998 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Vuorinen et al
Mitochondrial Function in Preeclampsia
955