Association Between ACVR2A and Early-Onset Preeclampsia: Replication Study in a Northeastern Brazilian Population

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Placenta 36 (2015) 186e190

Contents lists available at ScienceDirect

Placenta

journal homepage: www.elsevier.com/locate/placenta

Association between ACVR2A and early-onset preeclampsia: Replication study in a Northeastern Brazilian population

L.C. Ferreira a, b, C.E.M. Gomes b, c, A.C.P. Araújo d, P.F. Bezerra e, P. Duggal f, * S.M.B. Jeronimo a, b, g,

a Department of Biochemistry, Federal University of Rio Grande do Norte, Natal, Brazil b Institute of Tropical Medicine of Rio Grande do Norte, Federal University of Rio Grande do Norte, Natal, Brazil c Department of Biophysics and Pharmacology, Federal University of Rio Grande do Norte, Natal, Brazil d Department of Obstetrics and Gynecology, Federal University of Rio Grande do Norte, Natal, Brazil e Maternidade Escola Januario Cicco, Federal University of Rio Grande do Norte, Natal, Brazil f Bloomberg School of Public Health, Johns Hopkins University, Baltimore, USA g Institute of Science and Technology of Tropical Diseases (INCT-DT), Brazil

article info abstract

Article history: Introduction: Preeclampsia is a complex and heterogeneous disease with increased risk of maternal Accepted 10 November 2014 mortality, especially for earlier gestational onset. There is a great inconsistency regarding the genetics of preeclampsia across the literature. The gene Activin A receptor, type IIA (ACVR2A), was reported as Keywords: associated to preeclampsia in Australian/New Zealand and Norwegian populations. The goal of this study Activin was to validate this genetic association in a Brazilian population. ACVR2A Methods: We performed a case-control study using 693 controls and 613 cases (443 preeclampsia, 64 Genetic association eclampsia and 106 HELLP syndrome), from a Northeastern Brazilian population. Five single nucleotide Preeclampsia SNP polymorphisms (SNPs) in ACVR2A were tested for association through multiple logistic regression models. Results: There was no statistical association with preeclampsia (per se), eclampsia or HELLP. However, by grouping preeclampsia in accordance to the gestational age at delivery, SNPs rs1424954 (OR ¼ 1.86; 95% CI, 1.25e2.78; p ¼ 0.002) and rs1014064 (OR ¼ 1.77; 95% CI, 1.21e2.60; p ¼ 0.004) were signiﬁcantly associated with early onset preeclampsia (gestational age 34 weeks). The risk haplotype had a frequency of 0.468 in early preeclampsia compared to 0.316 in controls (p ¼ 0.0008 and permuted p ¼ 0.002). Discussion: Activin A receptors are important in decidualization, trophoblast invasion and placentation processes during pregnancy. The gene ACVR2A was associated with the more severe early onset preeclampsia. This ﬁnding supports the hypothesis of different pathogenic mechanisms contributing to the early- and late-onset preeclampsia. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction gestation. Gestational age of onset can be considered in the severity of preeclampsia [3], with early onset (34.0 weeks gestation) Preeclampsia is a pregnancy-related hypertensive disorder and associated with increased maternal complications and mortality a leading cause of maternal and infant mortality. It complicates while late-onset (>34.0 weeks gestation) are often secondary to 2e8% of pregnancies worldwide with rates of this disease reaching other co-morbidities, like obesity [4,5]. 17% in developing countries [1,2]. Generally, preeclampsia is Preeclampsia is shown to aggregate in families, suggesting deﬁned as de novo proteinuria and hypertension after 20 weeks genes may inﬂuence outcome [6e8], however this mechanism re- mains unclear. The genetic architecture behind preeclampsia is complex, including environmental factors, maternal and fetal * Corresponding author. Department of Biochemistry and Institute of Tropical genes, and their combined effects (i.e. interaction between paternal Medicine of Rio Grande do Norte, Federal University of Rio Grande do Norte, Av. and maternal genes) [9]. Five genome-wide linkage studies þ Senador Salgado Filho, 3000, Brazil. Tel./fax: 55 3342 2286. e E-mail address: [email protected] (S.M.B. Jeronimo). [10 14], two genome-wide association studies [15,16] and several

http://dx.doi.org/10.1016/j.placenta.2014.11.007 0143-4004/© 2014 Elsevier Ltd. All rights reserved. L.C. Ferreira et al. / Placenta 36 (2015) 186e190 187 candidate gene studies [17] identified several loci linked or asso- because of less severity. Preeclampsia cases were grouped into either early PE (34.0 ciated to preeclampsia. weeks' gestation) or late PE (>35.0 weeks' gestation) categories. Preeclamptic women who were greater than 34 weeks but less than or equal to 35 weeks at ACVR2A encodes activin A type II receptor (ActRIIA) which, along gestational delivery were excluded (n ¼ 31), in order to avoid misclassification due with Activin A receptor, type IB (Alk-4), plays an important role in to imprecision when assessing gestational age. In addition, we also excluded 21 the establishment of pregnancy, especially during decidualization, cases of preeclampsia from the analysis of time of onset the disease, because of trophoblast invasion and placentation processes [18]. There is missing information regarding the gestational age; therefore, the sample considered increasing evidence that Activin A and its receptors, including in Table 3 includes 391 cases. Controls were pregnant women who had normal blood pressure and no previous history of hypertension or preeclampsia and were from ActRIIA, are involved in preeclampsia pathophysiology [19,20]. The the same socio-economic background as the cases. Cases and controls with ACVR2A gene has been implicated in susceptibility to developing multiple-birth pregnancies, diabetes, kidney and/or heart disease were excluded preeclampsia in Australian/New Zealand family studies [21,22]. from the study. Additionally, an independent Norwegian population-based study identified an association between single nucleotide poly- 2.4. DNA extraction and genotyping morphisms (SNPs) in ACVR2A and preeclampsia [23]. In this study, At the time of enrollment, 8 ml of anticoagulated whole blood (EDTA) was we aimed to validate the genetic association between ACVR2A and collected by venipuncture. DNA was extracted by erythrocyte lysis in 0.87 mM NH4H2CO3 and 0.13 M NH4Cl followed by lysis of leukocytes in 1% sodium dodecyl preeclampsia in a northeastern Brazilian population in Natal, Brazil. sulfate, 100 mM EDTA, and 200 mM Tris (pH 8.5). Proteins were removed by pre- cipitation with ammonium acetate 5 M. DNA was precipitated from the supernatant 2. Material and methods by isopropanol, followed by wash with 70% ethanol. DNA was solubilized in 500 mL 2.1. Participants and study design of TE (Tris-EDTA) buffer and stored at 20 C until genotyping assay. ® Samples were genotyped by SNaPshot technique and the capillary electro- ® A total of 1306 women (693 controls and 613 cases) were recruited at a referral phoresis performed on ABI PRISM 3100 Avant Genetic Analyzer (Applied Bio- center for high-risk pregnancy (Maternidade Escola Januario Cicco) in Natal, State of systems). This technique standardization was carried out according to Lins et al. [27]. ® Rio Grande do Norte, Brazil, from 2001 to 2012. High-risk pregnant women are sent Genotyping was done in a blind way for case-control status. GeneMapper software to this center from across the region for evaluation and medical treatment. Clinical v3.5 (Applied Biosystems, CA, USA) was used for the genotype calling. data were collated from patients' charts at the time of enrollment. 2.5. Genetic markers selection 2.2. Ethical considerations We selected 5 SNPs in or near the ACVR2A gene. SNP rs1424954, located 50 to the The protocol was reviewed and approved by the Federal University of Rio ACVR2A gene (z1.8 kb from the transcription start site), was associated with pre- Grande do Norte (CEP-UFRN 88) and Brazilian National Ethical Committee (CONEP eclampsia in a family-based study [21]. SNPs rs2161983, rs1014064, rs1424941 and 5059). All participants or their legal guardian provided informed consent. rs3768687, located in intronic regions, were associated with preeclampsia in a population-based candidate gene study [23]. Fig. 1 shows the ACVR2A genetic or- fi 2.3. Phenotype de nition ganization and SNP positions (Assembly, reference; Genome build, 36.3). Preeclampsia was defined as a pregnant woman with de novo hypertension (systolic blood pressure (SBP) 140 mmHg or diastolic blood pressure 2.6. Population stratification assessment (DBP) 90 mmHg) and proteinuria after 20 weeks of gestation. For those women The Brazilian population is composed of Europeans (predominantly Portu- missing proteinuria information, we looked for one or more of the following guese), western Africans and Native Brazilians. Confounding by ethnicity or popu- symptoms: severe epigastric or right upper quadrant pain; hypereflexia with sus- lation stratification may arise when allele frequencies and disease risk vary across tained clonus; severe headache; persistent visual disturbances; pulmonary edema; ancestries. We randomly selected 756 individuals (Supplementary Table 1) and fetal growth restriction; placental abruption [24]. Eclampsia was defined as the genotyped a panel of 27 ancestry informative markers designed to infer genetic occurrence of seizures in preeclamptic women while HELLP syndrome was defined ancestry in a Brazilian population [28]. Genotyping for this ancestry panel was in accordance to the Mississippi-Triple Class System: AST or ALT 40 IU/L; performed by LGC Genomics (UK). We used principal components analysis (PCA) in LDH 600 IU/L and platelets 150 109/L [25]. Based on these criteria, a total of Eigenstrat [29] to assess the populations and no stratification was detected 443 women had preeclampsia, including 61 chronic hypertension with super- (Supplementary Fig. 1). imposed preeclampsia; an additional 64 had eclampsia and 106 had HELLP syndrome (Table 1). Diagnoses were determined by trained physicians at the Maternity 2.7. Statistical analysis Hospital (Maternidade Escola Januario Cicco, Natal, RN, Brazil) and all clinical and laboratorial data were individually reviewed by the study investigators in order to Categorical variables were analyzed using two-sided Fisher's exact test; assure consistent criteria in accordance with the American College of Obstetricians continuous variables were analyzed using Student's t-test. To assess the genetic and Gynecologists [26]. association of ACVR2A markers, odds ratios (OR) were estimated by multinomial The gestational age at delivery was used as a proxy for gestational age of onset logistic regression comparing preeclampsia (early and late gestational age), for early cases since all preeclampsia (PE) deliveries before 34 weeks would also eclampsia and HELLP outcomes with the normal pregnancy outcome (controls). have onset before 34 weeks. There is the potential for some misclassification of the Genotypes were coded as 0, 1 and 2 according to the number of minor alleles under late gestational age cases as a proxy for onset, since some of the late PE gestational an additive genetic model. Maternal age, parity (primigravida vs. others) and chronic age cases may have presented with early onset but were able to delay delivery hypertension affection status were included in the model as covariates. Statistical

Table 1 Demographics and clinical features for the phenotypic groups.

Variables Control Preeclampsia Eclampsia HELLP N ¼ 693 N ¼ 443 N ¼ 64 N ¼ 106

Maternal age, years 24.4 (±6.2) 25.9 (±7.1)a 20.7 (±6.6)a 27.6 (±6.8)a Primigesta, n (%) 300 (43.3) 217 (49.0) 45 (70.3)a 41 (38.7) Chronic hypertension, n (%) 0 (0.0) 61 (13.8) 2 (3.1) 10 (9.4) Systolic blood pressure, mmHg 117 (±12) 158 (±22)a 158 (±18)a 154 (±20)a Diastolic blood pressure, mmHg 75 (±9) 106 (±14)a 108 (±13)a 102 (±12)a Proteinuria (dipstick) Negative, n (%) 693 (100) 0 1 (1.6) 9 (8.5) 1þ, n (%) 0 (0.0) 386 (87.1) 47 (73.4) 90 (84.9) Missing, n (%) 0 (0.0) 57 (12.9) 16 (25.0) 7 (6.6) Gestational age at delivery, week 38.8 (±3.1) 37.0 (±3.2)a 36.1 (±3.7)a 34.1 (±4.2)a Infant's birth weight, grams 3143 (±634) 2735 (±838)a 2568 (±809)a 1949 (±778)a

For continuous variables, mean (±SD) were estimated. Two-sided Fisher exact test was used to compare the distribution of categorical variables and two-sided t-test to compare means. ap value 0.05 for the comparison with control group. 188 L.C. Ferreira et al. / Placenta 36 (2015) 186e190

Table 2 Genetic association of ACVR2A SNPs with preeclampsia per se, eclampsia and HELLP syndrome phenotypes.

SNP Minor allele Control Preeclampsia Eclampsia HELLP n ¼ 693 n ¼ 443 n ¼ 64 n ¼ 106

MAF MAF OR (95% CI) p MAF OR (95% CI) p MAF OR (95% CI) p

rs1424954 A 0.311 0.352 1.18 (0.97e1.45) 0.11 0.352 1.18 (0.78e1.77) 0.44 0.275 0.82 (0.58e1.17) 0.28 rs1014064 G 0.313 0.351 1.19 (0.98e1.44) 0.08 0.336 1.12 (0.75e1.69) 0.57 0.282 0.86 (0.61e1.22) 0.41 rs1424941 T 0.156 0.146 0.90 (0.70e1.15) 0.46 0.133 0.81 (0.47e1.41) 0.46 0.144 0.91 (0.59e1.39) 0.66 rs2161983 A 0.324 0.356 1.13 (0.93e1.38) 0.22 0.328 1.00 (0.67e1.51) 0.97 0.281 0.82 (0.58e1.16) 0.26 rs3768687 T 0.317 0.342 1.10 (0.91e1.34) 0.26 0.328 1.04 (0.69e1.58) 0.82 0.283 0.85 (0.60e1.20) 0.36

Odds ratio (OR) was estimate by multinomial logistic regression model (control as reference group), adjusting for maternal age, parity and chronic hypertension status using an additive genetic model. MAF: minor allele frequency.

Table 3 Genetic association of ACVR2A SNPs with early onset preeclampsia.

SNP Minor allele Control Late preeclampsia Early preeclampsia n ¼ 693 n ¼ 321 n ¼ 70

MAF MAF OR (95% CI) p MAF OR (95% CI) p

rs1424954 A 0.31 0.33 1.08 (0.87e1.36) 0.476 0.46 1.86 (1.25e2.78) 0.002 rs1014064 G 0.31 0.33 1.10 (0.88e1.36) 0.401 0.45 1.77 (1.21e2.60) 0.004 rs1424941 T 0.16 0.14 0.88 (0.67e1.17) 0.376 0.16 1.03 (0.62e1.71) 0.914 rs2161983 A 0.32 0.34 1.05 (0.84e1.30) 0.680 0.46 1.70 (1.15e2.53) 0.008 rs3768687 T 0.32 0.33 1.03 (0.83e1.28) 0.769 0.42 1.52 (1.02e2.26) 0.039

Odds ratio (OR) was estimate by multinomial logistic regression model (control as reference group), adjusting for maternal age, parity and chronic hypertension status. MAF: minor allele frequency. analysis was performed using Stata 11.2 software. The linkage disequilibrium (LD) syndrome cases had the earliest intervention, usually cesarean 2 pattern among markers was assessed by pairwise correlation coefﬁcient (r ) esti- section, in order to decrease infant mortality and maternal mates using Haploview 4.2 [30]. Haplotype frequencies were estimated via EM al- morbidity, but this resulted in lower infant birth weight (Table 1). gorithm and tested for association through logistic regression model implemented in PLINK [31]. The p-values were estimated by max (T) permutation (n ¼ 10,000) in PLINK, which controls for family wise error rate. 3.2. Genetic analysis

3. Results All studied SNPs achieved a genotyping call rate 94% and were in HardyeWeinberg equilibrium (p > 0.05). The schematic of the 3.1. Demographics and clinical features ACVR2A gene and the LD pattern among genotyped SNPs are shown in Fig. 1. The SNPs rs1424954, rs1014064, rs2161983 and rs3768687 The demographic and clinical characteristics among the groups were in strong pairwise LD (D0 > 0.95; r2 > 0.9). However, are shown in Table 1. Preeclamptic women were older than the rs1424941 was poorly correlated with the other four SNPs control women (mean age, 25.9 vs. 24.4) and had the highest (r2 ¼ 0.08). This same LD pattern was observed in both Australian prevalence of chronic hypertension (13.8%). Women with eclampsia and Norwegian populations [22,23]. were younger at delivery (mean age ¼ 20.7) with the highest The ﬁve SNPs showed no association with preeclampsia (per se), proportion of primigravidas (70.3%), whereas women with HELLP eclampsia or HELLP syndrome (Table 2). However, markers syndrome were the oldest at delivery (mean age ¼ 27.6). HELLP rs1014064 (OR ¼ 1.77; p ¼ 0.004), rs1424954 (OR ¼ 1.86; p ¼ 0.002)

Fig. 1. Linkage disequilibrium (LD) pattern and the schematic localization of the ﬁve genotyped ACVR2A SNPs. The Haploview 4.2 software created a LD plot using the r2 color scheme. Black squares represent r2 > 0.8 and the light gray squares r2 < 0.1. The inside-square number represents the correlation coefﬁcient value. The markers rs1424954, rs1014064, rs2161983 and rs3768687 are in strong LD with each other (r2 > 0.9). L.C. Ferreira et al. / Placenta 36 (2015) 186e190 189

Table 4 and this may include environmental factors (e.g. life style, health Haplotype association tests using logistic regression model. care system, social-economic and educational features) that could Haplotypea Haplotype frequency OR p-value Permuted modify the genetic effect of ACVR2A variants. Australian/New Zea- p-value Control Early PE land, Norwegian, Finnish and Brazilian populations have distinct ancestry and, therefore, it is reasonable that there may be differ- A/G/C/A/T 0.316 0.468 1.86 0.0008 0.002 ences underlying the LD pattern within this gene region among G/A/T/G/C 0.161 0.158 0.95 0.8288 0.987 G/A/C/G/C 0.524 0.374 0.50 0.0004 0.001 these populations. In addition, genetic heterogeneity may cause variation in the association strength across studies and it is likely a The alleles composing the haplotypes refer to the respective markers: rs2161983/rs1014064/rs1424941/rs1424954/rs3768687. that other genes, in addition to ACVR2A, could also be responsible for some of these preeclampsia cases. The role of activin receptors and its ligands (e.g. activins and and rs2161983 (OR ¼ 1.70; p ¼ 0.008) were strongly associated with inhibins) on healthy and pathological pregnancies have been well the more severe early onset preeclampsia (Table 3). The underlying documented. Several studies have reported higher levels of activin linkage disequilibrium between these SNPs suggest that these were A in serum of preeclamptic women when compared to normal not independent associations, but represent the same signal. pregnancies [20,33] and, recently, it was shown that preeclamptic Therefore, we estimated the haplotypes in this gene region for placentas expressing high levels of activin A induce trophoblast cell these SNPs. The haplotype including the significant minor alleles apoptosis via Nodal/ALK7 signaling [34]. Manuelpillai and col- increased the risk to early PE whereas the haplotype composed by leagues showed that the amount of activin A and ActRII in placenta the major alleles was protective (Table 4). Since the haplotype as- and choriodecidua, respectively, differed in early- and late-onset sociation was more significant than the individual SNPs, this sug- preeclampsia [19]. In addition, the mRNA expression level of acti- gests that the causal allele is within this haplotype block. These vin A receptor, type I (ACVR1) was elevated in the decidua of early statistical associations are robust to a Bonferonni correction for preeclamptic placentas [35]. Lastly, it was detected a remarkable multiple testing using a threshold of p < 0.01. In addition, by difference between early and late preeclamptic placentas, by both considering chronic hypertension with superimposed preeclamp- global gene expression profile [36e38] and histopathology [39]. sia apart from purely preeclampsia cases, we found that ACVR2A Altogether, these findings support our hypothesis that poly- SNPs remained specifically associated with early onset pre- morphisms in ACVR2A may disrupt placentation leading to the eclampsia (Supplementary Table 2). Finally, by grouping all 613 early-onset preeclampsia. cases (n ¼ 443 preeclampsia; n ¼ 64 eclampsia; and n ¼ 106 HELLP) In conclusion, we showed that the ACVR2A gene is associated no association was found (p > 0.1) (data not shown). Although we with early onset preeclampsia. The association between ACVR2A did not detect population stratification in this studied population, and preeclampsia in three unrelated populations (Australian/New we recognize that Brazilian population is highly admixed and there Zealand, Norwegian and Brazilian) suggests an important role of may be inflated associations due to population confounding. Thus, this gene in the preeclampsia etiology. In addition, our finding also we evaluated the association in a subset of the population with supports the hypothesis of different pathogenic mechanisms un- available ancestry informative markers, and the association derlying the late- and early-onset forms of the disease as well as remained for SNPs rs1424954 and rs1014064, after adjusting for the distinct genetic mechanisms behind preeclampsia, eclampsia and first three principal components (Supplementary Table 3). HELLP syndrome.

4. Discussion Funding

Activin A receptors are important in decidualization, tropho- This work was funded in part by a grant from Conselho Nacional blast invasion and placentation processes during pregnancy. de Desenvolvimento Científico e Tecnologico and CAPES. LCF Studies in European ancestry populations show an association of received a fellowship from CAPES-Fulbright. Activin A receptor, type IIA (ACVR2A) with preeclampsia [21,23].In the present study, we successfully validated the association be- Conflict of interest tween ACVR2A gene and preeclampsia in a population from northeast Brazil. An important caveat to our study is that the ge- There is no conflict of interest from any of the authors. netic association was detected only in early onset preeclampsia. Interestingly, neither eclampsia nor HELLP syndrome were associ- Acknowledgment ated with ACVR2A, suggesting distinct genetic mechanisms underlying hypertensive disorders of pregnancy. We thank the staff of Maternidade Escola Januario Cicco for their The genetic association with ACVR2A and preeclampsia is not help in ascertaining patients' clinical classification and the medical clear with some studies finding genetic associations and others students Luana Andrade and Homero Ferreira. We also would like failing to see an association. Moses et al., were the first to report the to thank Dr Jenefer Blackwell (Western Australia University) for her genetic association of ACVR2A (rs1424954) with preeclampsia in 34 helpful discussion of this data and Dr. Genevieve Wojcik (Stanford Australian/New Zealand families [21], but this finding was not University) for her help in the principal component analysis. replicated in a second study when 40 additional families were included [22]. More recently, Lokki et al. failed to replicate the Appendix A. Supplementary data genetic association between rs1424954 and preeclampsia in a Finnish population [32]. Roten et al. found an association with Supplementary data related to this article can be found at doi 10. ACVR2A in the SNPs rs2161983, rs1014064, rs1424941 and 1016/j.placenta.2014.11.007. rs3768687 in a large Norwegian population-based study [23].Un- like our study, those studies [21e23,32] did not take into account References the time of disease onset. Besides differences in sample size and [1] Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PFA. WHO analysis of case ascertainment scheme, characteristics of each population may causes of maternal death: a systematic review. Lancet 2006 Apr 1;367(9516): contribute to the differences observed among the genetic studies 1066e74. 190 L.C. Ferreira et al. / Placenta 36 (2015) 186e190

[2] Osungbade KO, Ige OK. Public health perspectives of preeclampsia in devel- [22] Fitzpatrick E, Johnson MP, Dyer TD, Forrest S, Elliott K, Blangero J, et al. Ge- oping countries: implication for health system strengthening. J Pregnancy netic association of the activin A receptor gene (ACVR2A) and pre-eclampsia. 2011 Apr 4;2011:e481095. Mol Hum Reprod 2009 Jan 6;15(3):195e204. [3] Von Dadelszen P, Magee LA, Roberts JM. Subclassification of preeclampsia. [23] Roten LT, Johnson MP, Forsmo S, Fitzpatrick E, Dyer TD, Brennecke SP, et al. Hypertens Pregnancy 2003;22(2):143e8. Association between the candidate susceptibility gene ACVR2A on chromo- [4] MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from pre- some 2q22 and pre-eclampsia in a large Norwegian population-based study eclampsia and eclampsia. Obstet Gynecol 2001 Apr;97(4):533e8. (the HUNT study). Eur J Hum Genet 2008 Sep 10;17(2):250e7. [5] Mbah A, Kornosky J, Kristensen S, August E, Alio A, Marty P, et al. Super- [24] Steegers EA, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet obesity and risk for early and late pre-eclampsia: super-obesity and risk of 2010;376(9741):631e44. pre-eclampsia. BJOG Int J Obstet Gynaecol 2010 Jul;117(8):997e1004. [25] Haram K, Svendsen E, Abildgaard U. The HELLP syndrome: clinical issues and [6] Chesley LC, Annitto JE, Cosgrove RA. The familial factor in toxemia of preg- management. A review. BMC Pregnancy Childbirth 2009;9(1):8. nancy. Obstet Gynecol 1968 Sep;32(3):303e11. [26] American College of Obstetricians and Gynecologists. Task force on hyper- [7] Salonen Ros H, Lichtenstein P, Lipworth L, Cnattingius S. Genetic effects on the tension in pregnancy. Hypertension in pregnancy. Report of the American liability of developing pre-eclampsia and gestational hypertension. Am J Med college of obstetricians and gynecologists' task force on hypertension in Genet 2000 Apr 10;91(4):256e60. pregnancy. Obstet Gynecol 2013 Nov;122(5):1122e31. [8] Bezerra PC, Leao~ MD, Queiroz JW, Melo E, Pereira FV, Nobrega MH, et al. [27] Lins TCL, Nogueira LR, Lima RM, Gentil P, Oliveira RJ, Pereira RW. A multiplex Family history of hypertension as an important risk factor for the develop- single-base extension protocol for genotyping Cdx2, FokI, BsmI, ApaI, and TaqI ment of severe preeclampsia. Acta Obstet Gynecol Scand 2010;89(5):612e7. polymorphisms of the vitamin D receptor gene. Genet Mol Res GMR [9] Cnattingius S, Reilly M, Pawitan Y, Lichtenstein P. Maternal and fetal genetic 2007;6(2):316e24. factors account for most of familial aggregation of preeclampsia: a population- [28] Lins TC, Vieira RG, Abreu BS, Grattapaglia D, Pereira RW. Genetic composition based Swedish cohort study. Am J Med Genet A 2004 Nov 1;130A(4):365e71. of Brazilian population samples based on a set of twenty-eight ancestry [10] Arngrímsson R, Siguroard~ ottir S, Frigge ML, Bjarnadottir RI, Jonsson T, informative SNPs. Am J Hum Biol Off J Hum Biol Counc 2010 Apr;22(2): Stefansson H, et al. A genome-wide scan reveals a maternal susceptibility 187e92. locus for pre-eclampsia on chromosome 2p13. Hum Mol Genet 1999 Sep;8(9): [29] Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Prin- 1799e805. cipal components analysis corrects for stratification in genome-wide associ- [11] Harrison GA, Humphrey KE, Jones N, Badenhop R, Guo G, Elakis G, et al. ation studies. Nat Genet 2006 Aug;38(8):904e9. A genomewide linkage study of preeclampsia/eclampsia reveals evidence for [30] Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD a candidate region on 4q. Am J Hum Genet 1997 May;60(5):1158e67. and haplotype maps. Bioinformatics 2005 Jan 15;21(2):263e5. [12] Lachmeijer AM, Arngrímsson R, Bastiaans EJ, Frigge ML, Pals G, [31] Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, et al. Sigurdardottir S, et al. A genome-wide scan for preeclampsia in the PLINK: a tool set for whole-genome association and population-based linkage Netherlands. Eur J Hum Genet EJHG 2001 Oct;9(10):758e64. analyses. Am J Hum Genet 2007 Sep;81(3):559e75. [13] Laivuori H, Lahermo P, Ollikainen V, Widen E, Haiv€ a-M€ allinen€ L, Sundstrom€ H, [32] Lokki AI, Klemetti MM, Heino S, Hiltunen L, Heinonen S, Laivuori H. Associ- et al. Susceptibility loci for preeclampsia on chromosomes 2p25 and 9p13 in ation of the rs1424954 polymorphism of the ACVR2A gene with the risk of Finnish families. Am J Hum Genet 2003 Jan;72(1):168e77. pre-eclampsia is not replicated in a Finnish study population. BMC Res Notes [14] Moses EK, Lade JA, Guo G, Wilton AN, Grehan M, Freed K, et al. A genome scan 2011;4(1):545. in families from Australia and New Zealand confirms the presence of a [33] Petraglia F, Aguzzoli L, Gallinelli A, Florio P, Zonca M, Benedetto C, et al. Hy- maternal susceptibility locus for pre-eclampsia, on chromosome 2. Am J Hum pertension in pregnancy: changes in activin A maternal serum concentration. Genet 2000;67(6):1581e5. Placenta 1995 Jul;16(5):447e54. [15] Johnson MP, Brennecke SP, East CE, Goring€ HHH, Kent JW, Dyer TD, et al. [34] Yu L, Li D, Liao Q, Yang H, Cao B, Fu G, et al. High levels of activin A detected in Genome-wide association scan identifies a risk locus for preeclampsia on preeclamptic placenta induce trophoblast cell apoptosis by promoting nodal 2q14, near the inhibin, beta B gene. Grant SFA, editor PLoS One 2012 Mar signaling. J Clin Endocrinol Metab 2012 Aug;97(8):E1370e9. 14;7(3):e33666. [35] Yong HE, Murthi P, Borg A, Kalionis B, Moses EK, Brennecke SP, et al. Increased [16] Zhao L, Bracken MB, DeWan AT. Genome-wide association study of pre- decidual mRNA expression levels of candidate maternal pre-eclampsia sus- eclampsia detects novel maternal single nucleotide polymorphisms and ceptibility genes are associated with clinical severity. Placenta 2014 copy-number variants in subsets of the hyperglycemia and adverse pregnancy Feb;35(2):117e24. outcome (HAPO) study cohort. Ann Hum Genet 2013 Jul;77(4):277e87. [36] Sitras V, Paulssen RH, Grønaas H, Leirvik J, Hanssen TA, Vårtun A, et al. Dif- [17] Mütze S, Rudnik-Schoneborn€ S, Zerres K, Rath W. Genes and the preeclampsia ferential placental gene expression in severe preeclampsia. Placenta 2009 syndrome. J Perinat Med 2008;36(1):38e58. May;30(5):424e33. [18] Florio P, Gabbanini M, Borges LE, Bonaccorsi L, Pinzauti S, Reis FM, et al. [37] Quinn KH, Lacoursiere DY, Cui L, Bui J, Parast MM. The unique pathophysi- Activins and related proteins in the establishment of pregnancy. Reprod Sci ology of early-onset severe preeclampsia: role of decidual T regulatory cells. Thousand Oaks Calif 2010 Apr;17(4):320e30. J Reprod Immunol 2011 Sep;91(1e2):76e82. [19] Manuelpillai U, Schneider-Kolsky M, Dole A, Wallace EM. Activin A and activin [38] Junus K, Centlow M, Wikstrom A-K, Larsson I, Hansson SR, Olovsson M. Gene receptors in gestational tissue from preeclamptic pregnancies. J Endocrinol expression profiling of placentae from women with early- and late-onset pre- 2001 Oct;171(1):57e64. eclampsia: down-regulation of the angiogenesis-related genes ACVRL1 and [20] Muttukrishna S, Knight PG, Groome NP, Redman CW, Ledger WL. Activin A EGFL7 in early-onset disease. Mol Hum Reprod 2012 Mar 1;18(3):146e55. and inhibin A as possible endocrine markers for pre-eclampsia. Lancet 1997 [39] Van der Merwe JL, Hall DR, Wright C, Schubert P, Grove D. Are early and late May 3;349(9061):1285e8. preeclampsia distinct subclasses of the diseasedwhat does the placenta [21] Moses EK, Fitzpatrick E, Freed KA, Dyer TD, Forrest S, Elliott K, et al. Objective reveal? Hypertens Pregnancy 2010 Nov;29(04):457e67. prioritization of positional candidate genes at a quantitative trait locus for pre-eclampsia on 2q22. Mol Hum Reprod 2006 Aug;12(8):505e12.