Original Article Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
Received: April 29, 2013 Accepted after revision: May 27, 2014 Published online: July 22, 2014
TNF-α G308A Gene Polymorphism Has an Impact on Renal Function, Microvascular Permeability, Organ Involvement and Severity of Preeclampsia Pavol Zubor a Karol Dokus a Imrich Zigo a Maria Skerenova b Rudolf Pullmann b Jan Danko a Departments of a Obstetrics and Gynecology and b Medical Biochemistry, Jessenius Faculty of Medicine, Comenius University, Martin, Slovak Republic
Key Words Preeclampsia · Tumor necrosis factor-α · Polymorphism · Complications
Abstract Background/Aims: Preeclampsia (PE) is a life-threatening complication of pregnancy that is associated with a high rate of maternal and perinatal morbidity and/or mortality worldwide. If untreated, it can progress to eclampsia, which can result in the death of the mother, the fetus or both. The etiology of PE is still uncertain; however, recently the role of the immune system has gained in importance. The role of tumor necrosis factor-α (TNF-α), a cytokine involved in inflammation processes, has been widely investigated in obstetric disorders. The aims of the present study were to investigate the effect of TNF-α gene G308A (rs1800629) polymorphism on disease risk, renal function, microvascular permeability, endothelial cell dysfunction and organ involvement in women with PE. Methods: Initially, 102 3rdtrimester pregnant women (preeclamptic cases and healthy controls) with singleton pregnancy were invited for participation, of which 76 were genotyped for TNF-α G308A polymorphism and evaluated for plasma levels of soluble vascular cell adhesion molecule-1 (sVCAM-1), fibronectin and TNF-α, which were tested for correlations with the profile of PE. The odds ratio (OR) and 95% confidence intervals ob-
© 2014 S. Karger AG, Basel 0378–7346/14/0783–0150$39.50/0 E-Mail [email protected] www.karger.com/goi
tained from unconditional logistic regression were used to test the association between the TNF-α polymorphism and PE risk. For continuous variables, we applied Student’s t test and, for categorical variables, the Pearson χ2 or Fisher’s exact test. The two-way ANOVA test with Bonferroni correction was used in multivariate analyses. Results: The A allele was more frequent in cases than controls (22.4 vs. 13.2%), which increased disease risk (OR = 2.73). Maternal serum levels of TNF-α, sVCAM-1 and fibronectin were significantly increased in cases (855.8 ± 385.1 pg/ml, 1,243 ± 671 ng/ml, 0.308 ± 0.231 g/l, respectively) compared to controls (301.1 ± 156.1 pg/ml, 651 ± 250 ng/ml, 0.218 ± 0.101 g/l, respectively; p < 0.0001, p < 0.0001 and p = 0.031, respectively), and these levels showed an increasing trend with the mutant allele genotype. Moderate and severe proteinuria was higher in rs1800629 allele A subjects compared to G/G carriers (53.8 vs. 14.3% (p < 0.05) and 13.0 vs. 4.7% (p < 0.01), respectively). The adverse effect of rs1800629 allele A on renal function was confirmed by increased plasma creatine levels, urinary protein excretion and lower tubular resorption rate in preeclamptic patients. Moreover, rs1800629 allele A preeclamptic carriers showed higher serum levels of fibronectin and sVCAM-1 compared to G/G homozygotes. Conclusion: This study reveals a possible association between clinical and laboratory manifestations of PE and the TNF-α gene G308A (rs1800629) polymorphism. © 2014 S. Karger AG, Basel
Pavol Zubor Department of Obstetrics and Gynecology Jessenius Faculty of Medicine, Comenius University Kollarova 2, SK–036 01 Martin (Slovak Republic) E-Mail zubor @ jfmed.uniba.sk
Introduction
Preeclampsia (PE) is a life-threatening complication of pregnancy that is associated with a high rate of maternal and perinatal morbidity and/or mortality worldwide. If untreated, it can progress to eclampsia, which can result in the death of the mother, the fetus or both. PE is a multisystem disorder that clinically manifests itself during the second half of pregnancy, but its etiology is believed to be related to the preclinical stages of the disease, i.e. the very early stages of pregnancy. The morphological changes in placentas of preeclamptic women (insufficient trophoblast invasion with increased apoptosis and syncytial knot formation, poor remodeling of the uterine spiral arteries, local fibrin deposits) [1] are still mainly explained by the multistep theory, which describes the role of several pathomechanisms in the development of PE, e.g. endothelial cell dysfunction [2], systemic vasospasm, poor angiogenesis due to disorders in the network of angiogenic/antiangiogenic factors [3, 4], placental hypoxia-reoxygenation injury by reactive oxygen species [5], increased placental expression level of Hsp27 [6], genetic susceptibility [7] and excessive maternal inflammation [8]. However, the present hypothesis regarding the etiology of PE is focused mainly on endothelial cell dysfunction and maladaptation of the immune response. This indicates that PE is a systemic disease of maternal endothelial cell dysfunction, resulting from a deficiency of endovascular invasion of fetal extravillous cytotrophoblasts into the maternal spiral arterioles. This process is facilitated by uterine/decidual natural killer (NK) cells that produce cytokines and angiogenic factors and impeded by the more cytotoxic peripheral NK cells of the innate immune system. Therefore, fundamentally, the disease process appears to be a unique variant of immune maladaptation at the level of the maternal-fetal interface in susceptible patients. The primary cause is considered to be a mixture of both insufficient induction of maternal tolerance to the paternal antigens, and unfavorable combinations of genetically determined maternal leukocyte receptors and fetal antigens [9]. The immune background of PE is supported by a newly formed hypothesis: PE is a pregnancy-induced autoimmune condition characterized by the presence of disease-causing angiotensin receptor-activating autoantibodies [10]. The theory of excessive innate activity with a shift towards an inflammatory cytokine profile in the pathogenesis of PE has triggered immense research activity. Preeclamptic women exhibit increased placental mRNA expression and serum or amniotic fluid levels of various TNF-α Polymorphism and Preeclampsia
cytokines, including the principal immune mediator, tumor necrosis factor-α (TNF-α) [11]. This cytokine is a potent and multifunctional proinflammatory Th1 type cytokine that is involved in the pathogenesis of a large number of autoimmune and inflammatory diseases. Its role has previously been associated with pregnancy pathologies such as PE [12], intrauterine growth restriction [13], gestational diabetes mellitus [14], spontaneous preterm delivery [15] and recurrent abortions [16]. Polymorphisms of the cytokine genes have been identified and can significantly influence cytokine production levels. These polymorphisms, which occur frequently within the regulatory region of cytokine genes, correlate with individual allelic variations. In humans, the TNF-α gene is located within the highly polymorphic major histocompatibility complex region on chromosome 6p21.3. The TNF gene cluster contains many polymorphisms including microsatellites and single-nucleotide polymorphisms (SNPs) [17]. For example, the promoter region of the TNF-α gene contains at least 10 polymorphic sites [18], including the –308 (guanine to adenine transition, rs1800629) polymorphism, which influences TNF-α production and activity in vitro and in vivo. Thus, this SNP has been the subject of numerous studies linked to the clinical consequences for the carrier of the mutant allele. A Medline search using the keywords ‘TNF-α polymorphism’ and ‘preeclampsia’ yielded a total of 19 papers. The majority evaluated the genotype and allele frequencies with/without disease risk estimation [19–32] and showed, after three meta-analyses covering over 2,300 cases and 1,900 controls [33–35], that there is no significant impact of this SNP on PE risk. However, the A allele was shown to be functionally important, which may play some role in clinically directed management of preeclamptic women. Carriers of this allele, compared with the groups homozygous for the G allele, may exhibit higher values of maternal blood pressure and proteinuria in the third trimester [36, 37] or IUGR-complicated PE [35]. Unfortunately, there is only one follow-up study, with contradictory results, which aimed to assess the link between the TNF-α G308A gene polymorphism and incidence of fetal growth retardation, preterm delivery and PE [31]. Thus, the scarce and unclear data evaluating the role of the TNF-α gene polymorphism allelic variations at position –308 with clinical consequences in pregnancy complicated by PE needs to be verified. Moreover, there is accumulating evidence suggesting that the inflammatory cytokine TNF-α plays a pivotal role in the disruption of macrovascular and microvascular circulation both in vivo and in vitro through the production of reactive oxygen species [38]. The findings of increased serum concentrations of a number of molecules Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
151
that mediate leukocyte-endothelial adhesion – e.g. vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), E-selectin [39, 40] and fibronectin [41, 42] – in preeclamptic women along with TNFα-induced expression of cell adhesion molecules in human umbilical vein endothelial cells [43] confirmed that the vasculatory inflammatory response contributes to the syndrome of endothelial dysfunction as well as thrombotic and metabolic disturbances seen in PE. We performed a clinically focused study, where the purpose of the investigation was to determine the correlations between the TNF-α rs1800629 gene polymorphism and renal function, microvascular permeability, endothelial cell dysfunction, the level of organ involvement, the severity of PE and preterm delivery in preeclamptic women. In addition, as there is an ethnic-linked prevalence of this polymorphism, we also determined the potential susceptibility to the development of PE in the Slovak population.
Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, 2000 [44]. PE was defined as the development of hypertension (≥140 mm Hg systolic and ≥90 mm Hg diastolic blood pressure) in previously normotensive subjects after 20 weeks of pregnancy measured on two consecutive measurements at least 6 h apart and proteinuria (>300 mg protein excretion in a 24-hour urine collection) in women with no history of proteinuria at baseline. The control group had no signs of obstetric or systemic disease, and all pregnancy tests, blood pressure and urine measurements (dipstick proteinuria score in random urine collection) were normal. The preeclamptic women were considered to have severe PE if they developed hypertension of 160 mm Hg systolic and/or 110 mm Hg diastolic pressures with more than 3 g/24 h proteinuria. Furthermore, preterm severe PE was defined as the onset of severe PE before 34 weeks of gestation. Biological Samples On admission, 10 ml of antecubital venous blood for genotyping and serum protein levels was drawn from each subject in the study. The samples were collected in ethylenediamine tetraacetic acid-coated tubes, which were immediately stored at 4 ° C and further processed within 6 h. Samples were centrifuged at 4,000 g for 10 min to separate the serum plasma and buffy coat containing leukocytes and then frozen at –20 ° C for DNA extraction and serum protein level evaluation.
Materials and Methods Study Patients This case-control study included 3rd-trimester (≥28 weeks’ gestation) pregnant women referred to the Department of Obstetrics and Gynecology, Jessenius Faculty of Medicine, for antenatal control. All participants were Caucasian and resided in the same geographic area. Inclusion criteria were singleton pregnancy, PE and normal physiological fetal status. Exclusion criteria included: multiple gestations; diabetes mellitus; renal, autoimmune or liver disease; hypertension before 20th week of pregnancy; preexisting hypertension or proteinuria; pregnancy-induced hypertension; fetal disorder (IUGR, congenital/chromosomal pathology); in vitro fertilization; infection; placental abruption, and HELLP syndrome. Subjects filled out an entrance questionnaire and underwent biological sample collection (venous blood and urine sampling) followed by measurements of the mother’s blood pressure and fetal heart rate monitoring by cardiotocography. None of the women were in active labor when blood and urine samples were taken. Initially, 102 women were invited for participation. Of these, 5 rejected blood testing, 8 were excluded due to incomplete data in the protocol, 6 because of a failure in specimen collection, processing, dilution and storage, and 7 subjects were excluded due to termination of pregnancy because of signs of fetal hypoxia on cardiotocography. Thus, 76 women were included in the final analyses (38 subjects in each of the preeclamptic and control groups). The Regional Ethical Committee at the Jessenius Faculty of Medicine registered under IRB00005636 at the Office for Human Research Protection, US Department of Health and Human Services, approved the study protocol, and biological samples were obtained after receiving written informed consent from each subject. The study was carried out in accordance with the Declaration of Helsinki for experiments involving humans. The women were classified according to the clinical guidelines for assessment of hypertension disorders in pregnancy using the
152
Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
Genotyping Genomic DNA was extracted from lymphocytes in peripheral venous blood using the SiMax Genomic DNA Extraction Kit (SBS Genetech) and standard salting-out method. The rs1800629 TNF-α genotypes were determined by the restriction fragment-length polymorphism method, performed by PCR followed by restriction endonuclease digestion. PCR amplification was carried out in a 20μl reaction containing 100 ng genomic DNA. The PCR reaction mix consisted of reaction buffer, MgCl2, dNTPs, primers and Taq polymerase (Genekraft). Each PCR amplification was as follows: initial denaturation for 10 min at 95 ° C; 35 cycles of denaturation for 50 s at 95 ° C, annealing for 50 s at 61 ° C and extension for 50 s at 72 ° C; terminal extension at 72 ° C for 6 min. We used a forward primer 5′-AGGCAATAGGTTTTGAGGGCCAT-3′ and reverse primer 5′-TCCTCCCTGCTCCGATTCCG-3′ for amplification of the 107bp PCR products, which were digested by restriction enzyme NcoI (MBI Fermentas) for 6 h at 37 ° C. The sense primer was modified to incorporate the polymorphic site into an NcoI restriction site, as previously described by other groups [45]. Outcomes of analyses were evaluated on 4% agarose gels (Metaphor Agarose, Lonza). PCR products and restriction fragments were visualized after GoldView (SBS Genetech) staining with the use of an ultraviolet transilluminator (fig. 1). The electrophoretic pattern of the TNF-α genotypes showed three bands (107, 87 and 20 bp) for the heterozygous G/A genotype, two bands for the homozygous G/G genotype (87 and 20 bp) and one band (107 bp) for the homozygous A/A genotype. To confirm the genotyping results, selected PCR-amplified samples (n = 6 for each genotype) were reexamined and showed 100% unity.
Analysis of Serum Molecules TNF-α Assay The serum levels of TNF-α in each subject in the study were assessed by quantitative sandwich ELISA assay (Human TNF-α
Zubor/Dokus/Zigo/Skerenova/Pullmann/ Danko
G/G
Fig. 1. Agarose gel electrophoresis of genotypes for analyzed polymorphisms of TNF-α. Column 1: homozygote for the G allele (87 and 20 bp) TNF-α gene. Column 2: heterozygote for the G and A allele (107, 87 and 20 bp) TNF-α gene. Column 3: homozygote for the A allele (107 bp) TNF-α gene. Column 4: marker (ΦX174 DNA BsuRI (HaeIII) Marker 9/Fermentas) with 118- and 72-bp bands among others. Column 5: marker (pBR322 DNA AluI Marker 20/Fermentas) with 100- and 90-bp bands among others.
Marker
Marker
118 bp 100 bp 90 bp 72 bp 1
Soluble Endothelial Cell Adhesion Molecule VCAM-1 Assay Serum levels of soluble VCAM-1 (sVCAM-1) were measured using a commercial ELISA assay (Immunotech EIA sVCAM-1, Beckman Coulter Co., France) according to the manufacturer’s instructions using a two-step immunological sandwich type assay (streptavidin-peroxidase binding the biotinylated antibody). Soluble Plasma Fibronectin Assay The assessment of fibronectin levels in subjects’ plasma was based on the immunoturbidimetric method (ADVIA 1200 Siemens Health Care), where turbidity is caused by the formation of antigen-antibody insoluble immunocomplexes, and this formation of the complexes is accelerated and enhanced by PEG6 buffer at an assay temperature of 18–37 ° C with a wavelength of 340 nm (Fibronectin Liquid Reagents, Dialab, Austria). The interassay precision was determined by measurements of fibronectin in two samples at regular time intervals over a 3-week period after calibration. All serum experiments were performed by an investigator blinded to the study groups and in duplicate to assure the quality of the results.
Statistical Analysis We used descriptive statistics expressed as mean ± standard deviation (±SD) or as number (percentage) to produce a summary of the data for continuous and categorical variables, respectively. For statistical analysis of the allele and genotype distribution in the studied population and to determine the deviation from the Hardy-Weinberg equilibrium according to the formula (p2 + 2pq + q2 = 1), the χ2 test was used based on Pearson’s distribution. The odds ratio (OR) and 95% confidence intervals (95% CIs) obtained
TNF-α Polymorphism and Preeclampsia
A/A
107 bp 87 bp
kit, Anogen, Canada) using microtiter plates precoated with a monoclonal antibody specific to TNF-α and a polyclonal biotinconjugated antibody. In order to quantify the amount of TNF-α present in the sample, avidin-conjugated horseradish peroxidase was added, and the cooler change was measured spectrophotometrically at a wavelength of 450 nm.
G/A
2
3
4
5
from unconditional logistic regression were used to test the association between the TNF-α polymorphism and PE risk. The normality of continuous variables was assessed using the KolmogorovSmirnov test. In cases of normal data distribution for comparisons between two parameters, the parametric statistical method was used. For continuous variables, we applied Student’s t test and, for categorical variables, the Pearson χ2 or Fisher’s exact test. The twoway ANOVA test with Bonferroni correction was used for evaluation of differences in the impact of two qualitative factors on one dependent variable in the multivariate analysis. The statistical level of significance was set at p ≤ 0.05. All statistical calculations were performed by the MedCalc 11.1 (MedCalc Software Inc., Mariakerke, Belgium) software for Windows.
Results
Maternal Demographic Data The women with PE had significantly higher prepregnancy and pregnancy body mass index, systolic and diastolic blood pressure and proportion of surgical pregnancy terminations. Furthermore, they showed a significant increase in negative neonatal outcomes, e.g. lower 1-min Apgar score, birthweight and gestational age of newborns. On average, deliveries occurred 2 weeks earlier in the women with PE than in controls (table 1). There were no cases of eclampsia or HELLP syndrome following delivery. TNF-α G308A Polymorphism and Risk of PE The genotype and allele frequencies of the G308A polymorphism in the promoter region of the TNF-α gene are presented in table 2. A Hardy-Weinberg equilibrium test showed that the distribution of the observed genotypes (p = 0.7765, q = 0.2235; p2 = 0.6029, 2pq = 0.347, Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
153
Table 1. Clinical and demographic characteristics of preeclamptic and unaffected pregnant women
Characteristic
PE group (n = 38)
Controls (n = 38)
Age, years Prepregnancy BMI Pregnancy BMI Systolic BP Diastolic BP Gestational age at delivery, weeks Gestational age at PE diagnosis, weeks Education below college level, % Parity Neonatal birthweight, g Apgar score (1 min) CS rate, % VOD rate, %
29.3±4.7 (18–41) 26.9±6.3 31.9±6.6 159±12.9 102±7.8 37.0 (28–42) 35.9±3.5 83.8 1.3±0.6 2,665±1,105.8 7.7±2.1 62.2 2.7
27.9±5.5 (16–41) 21.9±3.4 26.8±3.5 118±9.6 77±5.5 39.2 (33–42) 97.3 1.8±1.4 3,384±713.7 8.7±0.9 13.5 0
p value 0.2339
Received: April 29, 2013 Accepted after revision: May 27, 2014 Published online: July 22, 2014
TNF-α G308A Gene Polymorphism Has an Impact on Renal Function, Microvascular Permeability, Organ Involvement and Severity of Preeclampsia Pavol Zubor a Karol Dokus a Imrich Zigo a Maria Skerenova b Rudolf Pullmann b Jan Danko a Departments of a Obstetrics and Gynecology and b Medical Biochemistry, Jessenius Faculty of Medicine, Comenius University, Martin, Slovak Republic
Key Words Preeclampsia · Tumor necrosis factor-α · Polymorphism · Complications
Abstract Background/Aims: Preeclampsia (PE) is a life-threatening complication of pregnancy that is associated with a high rate of maternal and perinatal morbidity and/or mortality worldwide. If untreated, it can progress to eclampsia, which can result in the death of the mother, the fetus or both. The etiology of PE is still uncertain; however, recently the role of the immune system has gained in importance. The role of tumor necrosis factor-α (TNF-α), a cytokine involved in inflammation processes, has been widely investigated in obstetric disorders. The aims of the present study were to investigate the effect of TNF-α gene G308A (rs1800629) polymorphism on disease risk, renal function, microvascular permeability, endothelial cell dysfunction and organ involvement in women with PE. Methods: Initially, 102 3rdtrimester pregnant women (preeclamptic cases and healthy controls) with singleton pregnancy were invited for participation, of which 76 were genotyped for TNF-α G308A polymorphism and evaluated for plasma levels of soluble vascular cell adhesion molecule-1 (sVCAM-1), fibronectin and TNF-α, which were tested for correlations with the profile of PE. The odds ratio (OR) and 95% confidence intervals ob-
© 2014 S. Karger AG, Basel 0378–7346/14/0783–0150$39.50/0 E-Mail [email protected] www.karger.com/goi
tained from unconditional logistic regression were used to test the association between the TNF-α polymorphism and PE risk. For continuous variables, we applied Student’s t test and, for categorical variables, the Pearson χ2 or Fisher’s exact test. The two-way ANOVA test with Bonferroni correction was used in multivariate analyses. Results: The A allele was more frequent in cases than controls (22.4 vs. 13.2%), which increased disease risk (OR = 2.73). Maternal serum levels of TNF-α, sVCAM-1 and fibronectin were significantly increased in cases (855.8 ± 385.1 pg/ml, 1,243 ± 671 ng/ml, 0.308 ± 0.231 g/l, respectively) compared to controls (301.1 ± 156.1 pg/ml, 651 ± 250 ng/ml, 0.218 ± 0.101 g/l, respectively; p < 0.0001, p < 0.0001 and p = 0.031, respectively), and these levels showed an increasing trend with the mutant allele genotype. Moderate and severe proteinuria was higher in rs1800629 allele A subjects compared to G/G carriers (53.8 vs. 14.3% (p < 0.05) and 13.0 vs. 4.7% (p < 0.01), respectively). The adverse effect of rs1800629 allele A on renal function was confirmed by increased plasma creatine levels, urinary protein excretion and lower tubular resorption rate in preeclamptic patients. Moreover, rs1800629 allele A preeclamptic carriers showed higher serum levels of fibronectin and sVCAM-1 compared to G/G homozygotes. Conclusion: This study reveals a possible association between clinical and laboratory manifestations of PE and the TNF-α gene G308A (rs1800629) polymorphism. © 2014 S. Karger AG, Basel
Pavol Zubor Department of Obstetrics and Gynecology Jessenius Faculty of Medicine, Comenius University Kollarova 2, SK–036 01 Martin (Slovak Republic) E-Mail zubor @ jfmed.uniba.sk
Introduction
Preeclampsia (PE) is a life-threatening complication of pregnancy that is associated with a high rate of maternal and perinatal morbidity and/or mortality worldwide. If untreated, it can progress to eclampsia, which can result in the death of the mother, the fetus or both. PE is a multisystem disorder that clinically manifests itself during the second half of pregnancy, but its etiology is believed to be related to the preclinical stages of the disease, i.e. the very early stages of pregnancy. The morphological changes in placentas of preeclamptic women (insufficient trophoblast invasion with increased apoptosis and syncytial knot formation, poor remodeling of the uterine spiral arteries, local fibrin deposits) [1] are still mainly explained by the multistep theory, which describes the role of several pathomechanisms in the development of PE, e.g. endothelial cell dysfunction [2], systemic vasospasm, poor angiogenesis due to disorders in the network of angiogenic/antiangiogenic factors [3, 4], placental hypoxia-reoxygenation injury by reactive oxygen species [5], increased placental expression level of Hsp27 [6], genetic susceptibility [7] and excessive maternal inflammation [8]. However, the present hypothesis regarding the etiology of PE is focused mainly on endothelial cell dysfunction and maladaptation of the immune response. This indicates that PE is a systemic disease of maternal endothelial cell dysfunction, resulting from a deficiency of endovascular invasion of fetal extravillous cytotrophoblasts into the maternal spiral arterioles. This process is facilitated by uterine/decidual natural killer (NK) cells that produce cytokines and angiogenic factors and impeded by the more cytotoxic peripheral NK cells of the innate immune system. Therefore, fundamentally, the disease process appears to be a unique variant of immune maladaptation at the level of the maternal-fetal interface in susceptible patients. The primary cause is considered to be a mixture of both insufficient induction of maternal tolerance to the paternal antigens, and unfavorable combinations of genetically determined maternal leukocyte receptors and fetal antigens [9]. The immune background of PE is supported by a newly formed hypothesis: PE is a pregnancy-induced autoimmune condition characterized by the presence of disease-causing angiotensin receptor-activating autoantibodies [10]. The theory of excessive innate activity with a shift towards an inflammatory cytokine profile in the pathogenesis of PE has triggered immense research activity. Preeclamptic women exhibit increased placental mRNA expression and serum or amniotic fluid levels of various TNF-α Polymorphism and Preeclampsia
cytokines, including the principal immune mediator, tumor necrosis factor-α (TNF-α) [11]. This cytokine is a potent and multifunctional proinflammatory Th1 type cytokine that is involved in the pathogenesis of a large number of autoimmune and inflammatory diseases. Its role has previously been associated with pregnancy pathologies such as PE [12], intrauterine growth restriction [13], gestational diabetes mellitus [14], spontaneous preterm delivery [15] and recurrent abortions [16]. Polymorphisms of the cytokine genes have been identified and can significantly influence cytokine production levels. These polymorphisms, which occur frequently within the regulatory region of cytokine genes, correlate with individual allelic variations. In humans, the TNF-α gene is located within the highly polymorphic major histocompatibility complex region on chromosome 6p21.3. The TNF gene cluster contains many polymorphisms including microsatellites and single-nucleotide polymorphisms (SNPs) [17]. For example, the promoter region of the TNF-α gene contains at least 10 polymorphic sites [18], including the –308 (guanine to adenine transition, rs1800629) polymorphism, which influences TNF-α production and activity in vitro and in vivo. Thus, this SNP has been the subject of numerous studies linked to the clinical consequences for the carrier of the mutant allele. A Medline search using the keywords ‘TNF-α polymorphism’ and ‘preeclampsia’ yielded a total of 19 papers. The majority evaluated the genotype and allele frequencies with/without disease risk estimation [19–32] and showed, after three meta-analyses covering over 2,300 cases and 1,900 controls [33–35], that there is no significant impact of this SNP on PE risk. However, the A allele was shown to be functionally important, which may play some role in clinically directed management of preeclamptic women. Carriers of this allele, compared with the groups homozygous for the G allele, may exhibit higher values of maternal blood pressure and proteinuria in the third trimester [36, 37] or IUGR-complicated PE [35]. Unfortunately, there is only one follow-up study, with contradictory results, which aimed to assess the link between the TNF-α G308A gene polymorphism and incidence of fetal growth retardation, preterm delivery and PE [31]. Thus, the scarce and unclear data evaluating the role of the TNF-α gene polymorphism allelic variations at position –308 with clinical consequences in pregnancy complicated by PE needs to be verified. Moreover, there is accumulating evidence suggesting that the inflammatory cytokine TNF-α plays a pivotal role in the disruption of macrovascular and microvascular circulation both in vivo and in vitro through the production of reactive oxygen species [38]. The findings of increased serum concentrations of a number of molecules Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
151
that mediate leukocyte-endothelial adhesion – e.g. vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), E-selectin [39, 40] and fibronectin [41, 42] – in preeclamptic women along with TNFα-induced expression of cell adhesion molecules in human umbilical vein endothelial cells [43] confirmed that the vasculatory inflammatory response contributes to the syndrome of endothelial dysfunction as well as thrombotic and metabolic disturbances seen in PE. We performed a clinically focused study, where the purpose of the investigation was to determine the correlations between the TNF-α rs1800629 gene polymorphism and renal function, microvascular permeability, endothelial cell dysfunction, the level of organ involvement, the severity of PE and preterm delivery in preeclamptic women. In addition, as there is an ethnic-linked prevalence of this polymorphism, we also determined the potential susceptibility to the development of PE in the Slovak population.
Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, 2000 [44]. PE was defined as the development of hypertension (≥140 mm Hg systolic and ≥90 mm Hg diastolic blood pressure) in previously normotensive subjects after 20 weeks of pregnancy measured on two consecutive measurements at least 6 h apart and proteinuria (>300 mg protein excretion in a 24-hour urine collection) in women with no history of proteinuria at baseline. The control group had no signs of obstetric or systemic disease, and all pregnancy tests, blood pressure and urine measurements (dipstick proteinuria score in random urine collection) were normal. The preeclamptic women were considered to have severe PE if they developed hypertension of 160 mm Hg systolic and/or 110 mm Hg diastolic pressures with more than 3 g/24 h proteinuria. Furthermore, preterm severe PE was defined as the onset of severe PE before 34 weeks of gestation. Biological Samples On admission, 10 ml of antecubital venous blood for genotyping and serum protein levels was drawn from each subject in the study. The samples were collected in ethylenediamine tetraacetic acid-coated tubes, which were immediately stored at 4 ° C and further processed within 6 h. Samples were centrifuged at 4,000 g for 10 min to separate the serum plasma and buffy coat containing leukocytes and then frozen at –20 ° C for DNA extraction and serum protein level evaluation.
Materials and Methods Study Patients This case-control study included 3rd-trimester (≥28 weeks’ gestation) pregnant women referred to the Department of Obstetrics and Gynecology, Jessenius Faculty of Medicine, for antenatal control. All participants were Caucasian and resided in the same geographic area. Inclusion criteria were singleton pregnancy, PE and normal physiological fetal status. Exclusion criteria included: multiple gestations; diabetes mellitus; renal, autoimmune or liver disease; hypertension before 20th week of pregnancy; preexisting hypertension or proteinuria; pregnancy-induced hypertension; fetal disorder (IUGR, congenital/chromosomal pathology); in vitro fertilization; infection; placental abruption, and HELLP syndrome. Subjects filled out an entrance questionnaire and underwent biological sample collection (venous blood and urine sampling) followed by measurements of the mother’s blood pressure and fetal heart rate monitoring by cardiotocography. None of the women were in active labor when blood and urine samples were taken. Initially, 102 women were invited for participation. Of these, 5 rejected blood testing, 8 were excluded due to incomplete data in the protocol, 6 because of a failure in specimen collection, processing, dilution and storage, and 7 subjects were excluded due to termination of pregnancy because of signs of fetal hypoxia on cardiotocography. Thus, 76 women were included in the final analyses (38 subjects in each of the preeclamptic and control groups). The Regional Ethical Committee at the Jessenius Faculty of Medicine registered under IRB00005636 at the Office for Human Research Protection, US Department of Health and Human Services, approved the study protocol, and biological samples were obtained after receiving written informed consent from each subject. The study was carried out in accordance with the Declaration of Helsinki for experiments involving humans. The women were classified according to the clinical guidelines for assessment of hypertension disorders in pregnancy using the
152
Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
Genotyping Genomic DNA was extracted from lymphocytes in peripheral venous blood using the SiMax Genomic DNA Extraction Kit (SBS Genetech) and standard salting-out method. The rs1800629 TNF-α genotypes were determined by the restriction fragment-length polymorphism method, performed by PCR followed by restriction endonuclease digestion. PCR amplification was carried out in a 20μl reaction containing 100 ng genomic DNA. The PCR reaction mix consisted of reaction buffer, MgCl2, dNTPs, primers and Taq polymerase (Genekraft). Each PCR amplification was as follows: initial denaturation for 10 min at 95 ° C; 35 cycles of denaturation for 50 s at 95 ° C, annealing for 50 s at 61 ° C and extension for 50 s at 72 ° C; terminal extension at 72 ° C for 6 min. We used a forward primer 5′-AGGCAATAGGTTTTGAGGGCCAT-3′ and reverse primer 5′-TCCTCCCTGCTCCGATTCCG-3′ for amplification of the 107bp PCR products, which were digested by restriction enzyme NcoI (MBI Fermentas) for 6 h at 37 ° C. The sense primer was modified to incorporate the polymorphic site into an NcoI restriction site, as previously described by other groups [45]. Outcomes of analyses were evaluated on 4% agarose gels (Metaphor Agarose, Lonza). PCR products and restriction fragments were visualized after GoldView (SBS Genetech) staining with the use of an ultraviolet transilluminator (fig. 1). The electrophoretic pattern of the TNF-α genotypes showed three bands (107, 87 and 20 bp) for the heterozygous G/A genotype, two bands for the homozygous G/G genotype (87 and 20 bp) and one band (107 bp) for the homozygous A/A genotype. To confirm the genotyping results, selected PCR-amplified samples (n = 6 for each genotype) were reexamined and showed 100% unity.
Analysis of Serum Molecules TNF-α Assay The serum levels of TNF-α in each subject in the study were assessed by quantitative sandwich ELISA assay (Human TNF-α
Zubor/Dokus/Zigo/Skerenova/Pullmann/ Danko
G/G
Fig. 1. Agarose gel electrophoresis of genotypes for analyzed polymorphisms of TNF-α. Column 1: homozygote for the G allele (87 and 20 bp) TNF-α gene. Column 2: heterozygote for the G and A allele (107, 87 and 20 bp) TNF-α gene. Column 3: homozygote for the A allele (107 bp) TNF-α gene. Column 4: marker (ΦX174 DNA BsuRI (HaeIII) Marker 9/Fermentas) with 118- and 72-bp bands among others. Column 5: marker (pBR322 DNA AluI Marker 20/Fermentas) with 100- and 90-bp bands among others.
Marker
Marker
118 bp 100 bp 90 bp 72 bp 1
Soluble Endothelial Cell Adhesion Molecule VCAM-1 Assay Serum levels of soluble VCAM-1 (sVCAM-1) were measured using a commercial ELISA assay (Immunotech EIA sVCAM-1, Beckman Coulter Co., France) according to the manufacturer’s instructions using a two-step immunological sandwich type assay (streptavidin-peroxidase binding the biotinylated antibody). Soluble Plasma Fibronectin Assay The assessment of fibronectin levels in subjects’ plasma was based on the immunoturbidimetric method (ADVIA 1200 Siemens Health Care), where turbidity is caused by the formation of antigen-antibody insoluble immunocomplexes, and this formation of the complexes is accelerated and enhanced by PEG6 buffer at an assay temperature of 18–37 ° C with a wavelength of 340 nm (Fibronectin Liquid Reagents, Dialab, Austria). The interassay precision was determined by measurements of fibronectin in two samples at regular time intervals over a 3-week period after calibration. All serum experiments were performed by an investigator blinded to the study groups and in duplicate to assure the quality of the results.
Statistical Analysis We used descriptive statistics expressed as mean ± standard deviation (±SD) or as number (percentage) to produce a summary of the data for continuous and categorical variables, respectively. For statistical analysis of the allele and genotype distribution in the studied population and to determine the deviation from the Hardy-Weinberg equilibrium according to the formula (p2 + 2pq + q2 = 1), the χ2 test was used based on Pearson’s distribution. The odds ratio (OR) and 95% confidence intervals (95% CIs) obtained
TNF-α Polymorphism and Preeclampsia
A/A
107 bp 87 bp
kit, Anogen, Canada) using microtiter plates precoated with a monoclonal antibody specific to TNF-α and a polyclonal biotinconjugated antibody. In order to quantify the amount of TNF-α present in the sample, avidin-conjugated horseradish peroxidase was added, and the cooler change was measured spectrophotometrically at a wavelength of 450 nm.
G/A
2
3
4
5
from unconditional logistic regression were used to test the association between the TNF-α polymorphism and PE risk. The normality of continuous variables was assessed using the KolmogorovSmirnov test. In cases of normal data distribution for comparisons between two parameters, the parametric statistical method was used. For continuous variables, we applied Student’s t test and, for categorical variables, the Pearson χ2 or Fisher’s exact test. The twoway ANOVA test with Bonferroni correction was used for evaluation of differences in the impact of two qualitative factors on one dependent variable in the multivariate analysis. The statistical level of significance was set at p ≤ 0.05. All statistical calculations were performed by the MedCalc 11.1 (MedCalc Software Inc., Mariakerke, Belgium) software for Windows.
Results
Maternal Demographic Data The women with PE had significantly higher prepregnancy and pregnancy body mass index, systolic and diastolic blood pressure and proportion of surgical pregnancy terminations. Furthermore, they showed a significant increase in negative neonatal outcomes, e.g. lower 1-min Apgar score, birthweight and gestational age of newborns. On average, deliveries occurred 2 weeks earlier in the women with PE than in controls (table 1). There were no cases of eclampsia or HELLP syndrome following delivery. TNF-α G308A Polymorphism and Risk of PE The genotype and allele frequencies of the G308A polymorphism in the promoter region of the TNF-α gene are presented in table 2. A Hardy-Weinberg equilibrium test showed that the distribution of the observed genotypes (p = 0.7765, q = 0.2235; p2 = 0.6029, 2pq = 0.347, Gynecol Obstet Invest 2014;78:150–161 DOI: 10.1159/000364865
153
Table 1. Clinical and demographic characteristics of preeclamptic and unaffected pregnant women
Characteristic
PE group (n = 38)
Controls (n = 38)
Age, years Prepregnancy BMI Pregnancy BMI Systolic BP Diastolic BP Gestational age at delivery, weeks Gestational age at PE diagnosis, weeks Education below college level, % Parity Neonatal birthweight, g Apgar score (1 min) CS rate, % VOD rate, %
29.3±4.7 (18–41) 26.9±6.3 31.9±6.6 159±12.9 102±7.8 37.0 (28–42) 35.9±3.5 83.8 1.3±0.6 2,665±1,105.8 7.7±2.1 62.2 2.7
27.9±5.5 (16–41) 21.9±3.4 26.8±3.5 118±9.6 77±5.5 39.2 (33–42) 97.3 1.8±1.4 3,384±713.7 8.7±0.9 13.5 0
p value 0.2339
