Archives of Gynecology and Obstetrics https://doi.org/10.1007/s00404-018-4824-3

GYNECOLOGIC ENDOCRINOLOGY AND REPRODUCTIVE MEDICINE

High‑density lipoproteins (HDL) composition and function in preeclampsia Yael Einbinder1,2 · Tal Biron‑Shental2,3 · Moran Agassi‑Zaitler3 · Keren Tzadikevitch‑Geffen2,3 · Jacob Vaya4 · Soliman Khatib4 · Meital Ohana1 · Sydney Benchetrit1,2 · Tali Zitman‑Gal1,2  Received: 27 March 2018 / Accepted: 13 June 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Purpose  To evaluate (a) the properties of high-density lipoproteins (HDL)/cholesterol, which include apolipoprotein A-1 (ApoA1) and paraoxonase1 (PON1), both are negative predictors of cardiovascular risk and (b) HDL function, among women with preeclampsia (PE). PE is a multi-system disorder, characterized by onset of hypertension and proteinuria or other endorgan dysfunction in the second half of pregnancy. Preeclampsia is associated with increased risk for later cardiovascular disease. The inverse association between HDL, cholesterol levels and the risk of developing atherosclerotic cardiovascular disease is well-established. Methods  Twenty-five pregnant women [19 with PE and 6 with normal pregnancy (NP)] were recruited during admission for delivery. HDL was isolated from blood samples. PON1 activity and HDL were analyzed. An in vitro model of endothelial cells was used to evaluate the effect of HDL on the transcription response of vascular cell adhesion molecule-1 (VCAM-1) and endothelial nitric oxide synthase (eNOS) mRNA expression. Results  PON1 activity (units/ml serum) was lower in the PE group compared to normal pregnancy (NP) (6.51 ± 0.73 vs. 9.98 ± 0.54; P = 0.015). Increased ApoA1 was released from PE-HDL as compared to NP-HDL (3.54 ± 0.72 vs. 0.89 ± 0.35; P = 0.01). PE-HDL exhibited increased VCAM-1 mRNA expression and decreased eNOS mRNA expression on TNF-α stimulated endothelial cells as compared to NP-HDL. Conclusions  HDL from women with PE reduced PON1 activity and increased ApoA1 release from HDL particles. This process was associated with increased HDL diameter, suggesting impaired HDL anti-oxidant activity. These changes might contribute to higher long-term cardiovascular risks among women with PE. Keywords  Preeclampsia · HDL · ApoA1 · PON1 · eNOS · VCAM-1

Introduction

* Tali Zitman‑Gal [email protected] 1



Department of Nephrology and Hypertension, Meir Medical Center, 44281 Kfar Saba, Israel

2



Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

3

Department of Obstetrics and Gynecology, Meir Medical Center, Kfar Saba, Israel

4

Laboratory of Oxidative Stress and Human Diseases, Migdal-Galilee Technology Center, Tel Hai College, Kiryat Shmona, Israel



Preeclampsia (PE) is a syndrome characterized by the onset of hypertension and proteinuria or other end-organ dysfunction. It develops after 20 weeks of gestation and can cause significant maternal and fetal morbidity, and mortality [1]. The pathophysiology of PE is thought to be related to developmental abnormalities of placental vasculature leading to placental under-perfusion or ischemia. This ischemia is believed to mediate the release of anti-angiogenic factors into the maternal circulation, causing endothelial dysfunction, hypertension and end-organ injury [1]. The elemental factor leading to the development of placental vasculature abnormalities is still to be elucidated. However, histological changes in the placenta include atherosclerosis (lipid laden cells in the arteriole walls), fibrinoid necrosis, thrombosis,

13

Vol.:(0123456789)



sclerotic narrowing of arterioles, and placental infarction, all of which are similar to findings that appear in atherosclerotic plaque [1–3]. A large epidemiological study demonstrated a strong correlation between PE and the development of cardiovascular disease later in life [4]. Women with PE are more likely than those with a normal pregnancy (NP) to develop chronic hypertension, cardiovascular disease, stroke, diabetes and end-stage renal disease later in life [2, 4]. The association of high-density lipoproteins (HDL) remodeling with a higher risk of cardiovascular events and mortality has been established in patients with end-stage renal disease [5]. The cardio-protective properties of HDL are related to several mechanisms, the most recognized of which is the promotion of cholesterol efflux from cells in the arterial wall. However, anti-inflammatory, anti-oxidative and anti-thrombotic effects, including regulation of endothelial functions that enhance NO production and maintain endothelial integrity were found to have crucial roles in HDL function [6, 7]. Clinical studies that achieved significant increases in plasma HDL could not demonstrate improved cardiovascular outcomes. These studies suggest that the quality of HDL was more important than its quantity [8–10]. HDL was found to lose its potential anti-atherosclerotic properties in several chronic inflammatory diseases, such as anti-phospholipid syndrome [11], systemic lupus erythematosus rheumatoid arthritis [12], scleroderma [13], metabolic syndrome [14], diabetes [15, 16], coronary disease [17–19], and chronic kidney disease [20]. Limited improvement was seen after kidney transplantation [20]. The main protein in HDL is apolipoprotein A-1 (ApoA1), which accounts for approximately 70% of the total protein mass of HDL. ApoA1 binds to and removes lipid hydroperoxides of low-density lipids (LDL) in vitro and in vivo. A high concentration of ApoA-1 was an independent, negative predictor of cardiovascular risk. ApoA1 contributes to preserve enzyme activity, stability, and function of paraoxonase [8, 9]. Animal experiments with apoA-1 deficient mice demonstrated an accelerated atherosclerotic process that was mechanistically attributed to impaired reverse cholesterol transport, reduced paraoxonase activity, and augmented inflammation [10]. Human serum paraoxonase/arylesterase (PON1) has been suggested as an important regulator of the potential anti-atherogenic capacity of HDL [19]. PON1 has a number of enzymatic activities including arylesterase, lactonase, and peroxidase, as well as phospholipase A2-like functions [21]. Various studies have suggested that the direct anti-oxidant effect of HDL on LDL-oxidation, measured as a decrease in lipid peroxides, is mainly mediated by PON1 [19]. In human studies, higher PON1 activity is associated with a lower incidence of major cardiovascular events and conversely, reduced PON1 activity is associated with chronic diseases

13

Archives of Gynecology and Obstetrics

such as rheumatoid arthritis, renal failure, and Alzheimer’s disease [19]. Information regarding the composition and function of serum lipids and lipoproteins in pregnancies complicated by preeclampsia as compared to normal pregnancies is scarce. Early pregnancy dyslipidemia, particularly hypertriglyceridemia appears to be associated with increased risk of preeclampsia [22]. Based on the involvement of oxidative stress and endothelial dysfunction observed in PE, we hypothesized that changes in HDL function and composition might also occur in PE. Thus, we compared HDL from women with normal pregnancies and those with PE.

Materials and methods Study population Pregnant women who were admitted to the Department of Obstetrics and Gynecology, at Meir Medical Center from March 2016 to January 2017 were eligible for inclusion in the study. All participants provided informed consent to participate in the study, which was approved by the Meir Medical Center Ethics Committee (protocol number 012215-MMC). Women with multiple fetuses, known diabetes, chronic hypertension or chronic kidney disease, or who were unable to provide informed consent were excluded. Nineteen women were diagnosed with PE based on the criteria of new onset hypertension above 140/90 mmHg and proteinuria above 300 mg in 24 h, or other end-organ involvement. Six healthy women with uncomplicated NP were enrolled and served as the control group. Blood pressure, BMI, kidney function, lipid profile, and the presence of edema or proteinuria, were assessed among all participants.

Blood samples Blood samples for HDL extraction were taken by venipuncture, into two tubes without EDTA. Plasma was harvested by low speed centrifugation (3000g for 10 min at 4 °C). All plasma samples were stored at − 80 °C before analysis. Blood samples for PON1 activity were taken by venipuncture, into one tube containing EDTA. Serum was harvested by low speed centrifugation (3000g for 3 min at 4 °C) and stored at − 80 °C  before analysis.

HDL assays HDL were isolated from human plasma samples by discontinuous density gradient ultracentrifugation (d = 1.006–1.25 g/ml), as previously described [23]. In brief, solid potassium bromide was added to the plasma samples

Archives of Gynecology and Obstetrics

for density adjustment. Four ml of this plasma were placed into ultraclear centrifuge tube (Beckman 14 × 89 mm) and 4 ml of NaCl solution (d = 1.084 g/ml, 1 mM EDTA, pH 8.6) was layered over the plasma. An additional 4 ml of NaCl solution (d = 1.006 g/ml, 1 mM EDTA, pH 8) was added. Ultracentrifugation was performed in a swinging bucket rotor SW 41 at 4 °C, 37,000 RPM for 48 h. Then, the HDL fraction was carefully aspirated with a syringe and dialyzed against NaCl (d = 1.006 g/ml, 1 mM EDTA, pH 8) to protect the lipoproteins against oxidation. HDL was quantified using Pierce™ Modified Lowry Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL, USA).

Lactonase activity Blood samples for PON1 activity were placed into one tube containing EDTA. Serum was harvested using low speed centrifugation (3000g for 3 min at 4 °C). Ten microlitre of diluted serum (1:20 in PBS buffer) was taken, for a total reaction volume of 200 μl. Lactonase activity was measured using dihydrocoumarine as the substrate. Initial rates of hydrolysis were determined by spectrophotometry at 270 nm. The assay mixture included 2 mM dihydrocoumarine and 1 mM ­CaCl2 in 50 mM Tris–HCl, pH 7.5. Nonenzymatic hydrolysis of dihydrocoumarine was subtracted from the total rate of hydrolysis. One unit of lactonase activity was equal to 1 μmol of dihydrocoumarine hydrolyzed/min/ ml, as reported previously [24].

Cell culture and incubation Human umbilical vein endothelial cells (HUVEC) were isolated from umbilical cords obtained from the Maternity Unit at Meir Medical Center, Kfar Saba, Israel [25]. Only umbilical cords from women who had a normal pregnancy and delivery were used. Informed consent was obtained. HUVEC were grown in M-199 medium supplemented with 20% FCS, 100 U/ml penicillin, 100 µ/ml streptomycin (Biological Industries, Beit Haemek, Israel), 5 U/ml heparin, and 25 µ/ml endothelial mitogen (Biomedical Technologies, Inc., Stoughton, MA, USA). HUVEC were pre-incubated with HDL (50 µg/ml, 60 min) and stimulated with TNF-a (0.1 ng/ ml, 4 h), as previously described [20].

HDL gel electrophoresis HDLs were separated according to hydrodynamic diameter by nondenaturing, 4–12% gradient polyacrylamide gel electrophoresis (PAGE) (4–12% Tris–Glycine gel; Bio-rad Laboratories, Carlsbad, CA, USA), as described previously [26], with modifications. Briefly, HDL samples were diluted 1–1 in native Tris–Glycine Sample Buffer (Bio-rad Laboratories, Carlsbad, CA, USA). The gels were run for 30 min at

65 V, and then for 4 h at 100 V on ice until the dye reached the end of the gel. The gels were then dyed with Coomassie Brilliant Blue. HDL diameter was estimated by optical densitometry analysis using Multi Gauge (Fuji, Japan) analysis software. As reference proteins, high molecular weight proteins (thyroglobulin 17 nm, ferritin 12.2 nm, lactate 10.4 nm, dehydrogenase 8.2 nm and albumin 7.1 nm; GE healthcare, Amersham Pharmacia Biotech, Buckinghamshire, UK) were used.

Western blot ApoA1 protein expressions were studied with a standard western blot technique using anti-ApoA1 polyclonal antibody (1:500, Millipore, Temecula, CA, USA), as previously described [27]. Equal amounts of HDL (20 μg) from PE and NP were separated on native 15% PAGE. In this step, HDL stacks at the top of the gel because of its large diameter, and non-bound proteins separate by charge and size throughout the gel. The gels were run at 160 V and 24 mA for 1 h. The proteins were then transferred to a nitrocellulose membrane. The bound antibody was visualized with the Enhanced Chemiluminescent Reporter System (ECL, Santa Cruz, CA, USA). LAS-3500 (Fuji, Japan) was used to quantify protein expressions. Optical densities were normalized to general protein stains (Ponceau).

RNA extraction and reverse transcription (RT) PCR Total RNA was extracted from cells, as previously described [28, 29] using MasterPure RNA purification Kit (EPICENTRE, Madison, WI, USA). RNA integrity was assessed using NanoDrop (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). RNA (1 µg) was reverse transcribed into single-strand DNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems Inc., Foster City, CA, USA), according to the manufacturer’s instructions.

Real‑time polymerase chain reaction (PCR) Real-time PCR was performed using ABI real-time PCR (7500 Fast Real Time PCR System, Applied Biosystems, Inc., Foster City, CA, USA) using Syber Green I reaction mix (Applied Biosystems Inc., Foster City, CA, USA). Realtime PCR was performed to validate the expression pattern of selected genes: VCAM-1 and eNOS with glucuronidase beta (GUSB) as reference gene. Data were analyzed as accepted the ­2−ΔΔCt method [28, 29]. VCAM-1 primers: forward primer: 5′-CAA​GTC​TAC​ATA​TCA​CCC​AAG​ AAT​ACA​GTT​-3′ reverse primer: 5′-GGT​AGA​CCC​TCG​CTG​GAA​CA-3′

13



Archives of Gynecology and Obstetrics

eNOS primers forward primer: 5′-TGC​AGT​TGC​TGC​CAG​GTC​reverse primer: 5′-GGA​CTT​GCT​GCT​TTG​CAG​GT-3′ GUSB primers: forward primer: 5′-CAA​TAC​CTG​ACT​GAC​ACC​TCC​ AGT​A-3′ reverse primer: 5′-TGG​TGG​GTG​TCG​TGT​ACA​GAAGT3′

Data analysis The sample size was calculated based on the assumption of a twofold change in HDL composition, α = 5% and power = 80%. All data are expressed as mean ± SD or median (range). Variables were tested for normality (Shapiro–Wilk test). For the variables which were not normally distributed, median and range were presented. Paired t test or Wilcoxon signed-rank non-parametric test were used to evaluate the changes in HDL diameter, ApoA1 protein expression and eNOS and VCAM mRNA expression (each when appropriate). P values less than 0.05 were considered significant.

Results Characteristics of the study population The study included 19 women with PE and 6 with NP. Baseline clinical characteristics are shown in Table 1. None of the participants had a history of hypertension, diabetes or dyslipidemia. Gestation was shorter in women with PE (35.2 ± 3.1 weeks vs. 38.4 ± 1.2; P = 0.0035 ). Mean Table 1  Demographic and clinical characteristics of pregnant women with preeclampsia and healthy women during an uncomplicated pregnancy

systolic and diastolic blood pressures taken at the time of admission were found to be significantly higher in women with PE (144.9 ± 10.4  mmHg vs. 114.7 ± 4.9; P = 0.005 and 94.2 ± 11.6 mmHg vs. 74.3 ± 10.4, P = 0.005, respectively). Serum albumin levels were significantly lower in women with PE. The range of proteinuria among women with PE was 250–4800 mg/24 h, as compared with proteinuria within the normal range (< 200 mg/dl) in women with NP. Three women in the preeclamptic group had features of severe PE (systolic blood pressure > 160 mmHg, diastolic > 100  mmHg, thrombocytopenia < 100,000/µl, and elevated liver function tests).

HDL composition, ApoA1 expression and PON1 activity HDL from 19 women with PE and 6 with NP were separated in 4–12% gradient PAGE, as described in the Methods section. HDL diameter was analyzed according to the study of Cohen et al. [27] using Multi Gauge (Fuji, Japan) analysis software. In brief, a line was drawn to divide the HDL band into two sections (a and b in Fig. 1A) exactly in the middle of the control HDL. This makes the density ratio between them, about 1. The ratio is expected to be greater than 1 when the HDL diameter is increased. This occurs when more particles appear above the line drawn in the graph. HDL were significantly denser in women with PE as compared to NP (2.07 ± 1 vs. 1.19 ± 0.2, P = 0.013). The data are presented as a box plot in Fig. 1B. Released ApoA1 density was significantly higher in women with PE as compared to NP (3.53 ± 0.85 vs. 0.89 ± 0.35, P = 0.001) (Fig. 1C, D). The altered HDL properties of increased HDL diameter and ApoA1 release might affect HDL anti-atherogenic activity. PON1 activity measured using lactonase activity assay was significantly down-regulated in serum from women with

Characteristic

Preeclampsia (19)

Normal pregnancy (6)

P value

Age, years Gestational age, weeks Body mass index, kg/m2 Systolic blood pressure, mmHg Diastolic blood pressure, mmHg Total serum cholesterol, mg/dl Serum triglycerides, mg/dl Serum HDL cholesterol, mg/dl Serum LDL cholesterol, mg/dl Urine, mg protein/24 h Serum creatinine, mg/dl Serum albumin, g/dl

29 ± 7 35.2 ± 3.1 28.3 ± 4.8 144.9 ± 10.4 94.2 ± 11.6 268.5 (158–296) 246 (98–864) 68.3 ± 16.3 125.4 ± 31.1 521.5 (250–4800) 0.64 ± 0.16 3.0 ± 0.4

32 ± 3.8 38.4 ± 1.2 23.6 ± 4 114.7 ± 4.9 74.3 ± 10.4 265 (216–274) 219 (142–296) 82 ± 19.8 125.5 ± 27.6 < 200 0.66 ± 0.09 3.2 ± 0.2

NS 0.035 NS 0.001 0.005 NS NS NS NS

Data are expressed as mean ± SD or median (minimum–maximum) NS not significant (P > 0.05)

13

NS 0.036

Archives of Gynecology and Obstetrics

Fig. 1  HDL composition and APOA1 protein expression. A HDL from NP and PE were separated from plasma and run on native 4–12% PAGE. Band intensities above (a) and below (b) the lines drawn in the middle of each control HDL band were analyzed. B The ratio a/b was calculated for the PE HDL as compared with the NP-

HDL (box plot of data). C HDL from NP and PE run on 15% PAGE followed by western blot analysis using anti-Apo A1 antibody. D Apo A1 protein expression (densitometric analysis—box plot of data). Optical densities were normalized to general protein stain (Ponceau). * P < 0.05 compared to NP

eNOS and VCAM‑1 mRNA expression The effect of HDL on the transcriptional response of HUVEC stimulated with TNF-α, was evaluated based on the changes observed at the level of eNOS and VCAM-1 mRNA expression. HUVEC exposed to HDL from women with PE in vitro demonstrated decreased eNOS mRNA expression and increased VCAM-1 mRNA expression (Fig. 3).

Discussion Fig. 2  Lactonase PON1 activity. Serum PON1 was measured in women with PE and NP using lactonase activity assay. *P < 0.05 compared to NP

PE compared to those with NP (6.51 ± 0.73 vs. 9.98 ± 0.54; P = 0.015; Fig. 2).

This study provides evidence that HDL isolated from women with PE differ in composition and in function and demonstrate a loss of vasoprotective effects, as compared to HDL from women with normal pregnancies. HDL from women with PE was larger in diameter, released more APOA1 and demonstrated reduced PON1 activity. Moreover, the effects of PE-HDL and NP-HDL on the transcriptional response of HUVEC exposed to TNF-α at the level of eNOS and VCAM-1 mRNA expression, suggested alterations in other anti-atherogenic effects of HDL in the presence of PE.

13



Fig. 3  eNOS and VCAM-1 mRNA expression in HUVEC stimulated with HDL (PE and NP). HUVEC were pre-incubated with HDL (50 µg/ml, 60 m) and stimulated with TNF-α (0.1 ng/ml, 4 h). a eNOS mRNA expression; b VCAM mRNA expression. *P < 0.05 compared to control (no stimulation), **P < 0.05 compared to TNF-α stimulation, #P < 0.05 compared to TNF-α stimulation with PE-HDL

Several anti-atherogenic effects of HDL have been demonstrated, including cholesterol efflux from cells, mainly macrophages, prevention of LDL-oxidation, augmentation of NO synthesis, inhibition of adhesion molecule expression and of endothelial apoptosis and thrombotic activation [19]. The pathogenesis of preeclampsia remains ambiguous, but irregularities in placental perfusion resulting in increased oxidative stress have been shown to have a fundamental role in producing the endothelial dysfunction observed in PE. It has been speculated that changes in HDL composition and function could contribute to this endothelial dysfunction. HDL function is directly related to its proteome and lipid content: both dictate HDL diameter and function [30, 31]. ApoA1 is the main protein in HDL particles, comprising 70% of its proteome. Decreased APOA1 caused by its release from HDL particles has been described previously in patients with acute coronary syndrome, and was associated with increased cardiovascular risk [19]. Release of ApoA1 was also reported in HDL isolated from homogenate atherosclerotic plaque. Hedrick et al. [32] demonstrated that as the ratio of ApoA2 to ApoA1 increases, the HDL become larger

13

Archives of Gynecology and Obstetrics

due to inhibition of hepatic lipase (HL) and they lose their anti-atherogenic properties. Van der Steeg analyzed data from the large IDEAL and EPIC-Norfolk clinical trials and found an inverse correlation between ApoA1 and coronary artery disease (CAD), whereas high-plasma HDL-C and very large HDL particles were associated with increased CAD risk [33]. The larger diameter of HDL among women with PE, as well as the release of ApoA1 from HDL, indicates loss of the vasoprotective properties of HDL. The causes and consequences of these phenomena are yet to be defined [19]. Our results demonstrated decreased PON1 activity in women with PE as compared to those with normal pregnancies. The anti-oxidant activity of HDL is primarily a result of PON1, which is in the serum and binds to HDL-C in a calcium-dependent manner [7]. PON1 retards or reverses atherosclerosis by preventing low-density lipoprotein cholesterol (LDL-C) oxidation [8–10]. PON1 activity is inversely correlated with the risk of cardiovascular disease and is decreased in several clinical conditions associated with atherosclerosis and oxidative stress [34, 35]. Studies in knockout mice have demonstrated the protective effects of PON1. Shih et al. [36] found that HDL isolated from PON1-deficient mice was unable to prevent LDL-oxidation in a co-cultured cell model, and both HDL and LDL isolated from these mice were more susceptible to oxidation. Several studies reported lower PON1 activity in women with PE as compared to normal pregnancies [37–40]. One explanation for this decrease in PON1 activity might be related to liver damage that occurs with PE [39, 40]. The results of our study were compatible with those of reports suggesting decreased serum PON1 activity in preeclampsia. Inactivation of PON1 in HDL from healthy subjects also affects NO production by endothelial cells. Furthermore, HDL isolated from PON1-deficient mice failed to stimulate endothelial cells to produce NO. Ramet et al. [41] reported a three-fold increase in eNOS protein expression in cultured endothelial cells after incubation with HDL for 24 h. However, this increase was not associated with steady-state increments in mRNA levels. TNF-α reduces endothelium-dependent vasorelaxation in vivo and ex vivo. In the setting of atherosclerosis, in which steady-state eNOS mRNA is down-regulated, the contribution of TNF-α is considered crucial to the pathogenesis of the disease [42]. The current study evaluated the effect of HDL on the eNOS transcription response in HUVEC exposed to TNF-α. Endothelial cells stimulated with TNF-α showed down-regulation of eNOS mRNA expression. The increase in eNOS mRNA expression in endothelial cells seen with the addition of HDL from normal pregnancies was blunted in women with PE. These results indicate that, at the level of mRNA expression, improved endothelial function obtained from HDL from

Archives of Gynecology and Obstetrics

women with normal pregnancies was absent in HDL from women with PE, and therefore, might contribute to the endothelial dysfunction observed in PE. Atherosclerosis is a chronic inflammatory disorder characterized by accumulation of macrophages and T lymphocytes in the arterial intima, accompanied by oxidation. An early step in this inflammatory process is the expression of adhesion molecules on endothelial cells, which allows inflammatory cells to migrate into the intima. HDL inhibits adhesion molecule expression on endothelial cells [43]. VCAM-1 is an important adhesion molecule that is upregulated during endothelial activation. Several studies have demonstrated that HDL from subjects with normal kidney function reduced VCAM-1 gene expression in endothelial cells in response to TNF-α stimulation, while HDL from patients with CKD increased VCAM-1 expression [20, 44, 45]. Ortiz-Munoz et al. [45] showed that HDL isolated from patients with acute cerebral infarction were less effective in inhibiting TNF-α induced VCAM-1 expression by blood brain barrier endothelial cells as compared to HDL from healthy subjects. This suggests that dysfunctional HDL could be involved in the inappropriate protection of endothelial cells in acute stroke. Szarka et al. [46] found elevated levels of pro-inflammatory cytokines IL-6 and TNF-α; chemokines IL-8, IP-10 and MCP-1; and adhesion molecules ICAM-1 and VCAM-1 in PE as compared to normal pregnancies. They suggested that these findings might contribute to the systemic pro-inflammatory state observed in PE. The current study evaluated the functionality of HDL at the level of RNA transcription. We found that HDL from women with PE was less effective in reducing VCAM-1 mRNA expression in endothelial cells when compared to HDL from normal pregnancies. This information could contribute to the understanding of the pro-inflammatory state seen in PE at the level of transcription response. In the current study, serum levels of total cholesterol, triglycerides, LDL-C and HDL-C were similar in women with PE or NP. There is no consensus regarding the reasons serum levels of cholesterol and triglycerides vary in PE compared to NP. Some studies have reported higher levels of cholesterol and triglycerides in PE [47, 48], but as in our study, other reports did not find differences in these levels [18, 39, 49]. This study was limited by the relatively few participants, which prevented subgroup analysis between the severe, early-onset and moderate, late-onset forms of preeclampsia (diagnosis before or after 34 weeks’ gestation). Women with PE were not followed after delivery to observe whether HDL composition returned to normal values. In addition, we recruited women with PE only in the third trimester, when the signs of PE were more severe. At that stage, changes in HDL particles might be more prominent as compared to earlier stages of gestation, which were not assessed.

Conclusion HDL from women with PE demonstrates decreased PON1 activity and increased APOA1 release. These processes are associated with larger HDL particle diameters, which suggest impaired HDL anti-oxidant activity. Whether these changes contribute to the increased cardiovascular morbidity seen among these women later in life, remains to be determined. Acknowledgements  This study is part of the basic science requirements of M. Agassi-Zaitler, Department of Obstetrics and Gynecology, Meir Medical Center, Kfar Saba, Israel. We thank the delivery room staff of Meir Medical Center for helping recruit women to the study. We thank Prof. M. Aviram and N. Volkova from the Lipid Research Laboratory Technion, Rappaport Faculty of Medicine, Rambam Medical Center, Haifa, Israel for teaching us the HDL isolation method. We thank D. Atrahimovich from the Laboratory of Oxidative Stress and Human Diseases in the Migdal-Galilee Technology Center, Israel for helping with the PON1 activity assay. We thank F. Schreiber, MSc for editing the manuscript and N. Jelin, MA for assistance with the statistical analysis. They are both employees of Meir Medical Center. Author contributions  YE, TBS, JV, SB and TZG conception and design of research; YE, KTG, MAZ and TZG data and blood collection; MO, KTG, MAZ, SK and TZG performed experiments; TZG, MO and SK analyzed data, interpreted results and prepared figures; YE, TBS, JV, SB and TZG drafted manuscript; YE, SB and TZG edited and revised manuscript; all authors approved the final version of manuscript. Funding  This work was supported by the Mintz-Law Foundation (Y. Einbinder) from the Sackler Faculty of Medicine, Tel Aviv University, Israel.

Compliance with ethical standards  Conflict of interest  All authors have no conflicts of interest.

References 1. Noris M, Perico N, Remuzzi G (2005) Mechanisms of disease: pre-eclampsia. Nat Clin Pract Nephrol 1:98–114 2. Sibai B, Dekker G, Kupferminc M (2005) Pre-eclampsia. Lancet 365:785–799 3. Myatt L, Webster RP (2009) Vascular biology of preeclampsia. J Thromb Haemost 7:375–384 4. Scantlebury DC, Hayes SN (2014) How does preeclampsia predispose to future cardiovascular disease? Curr Hypertens Rep 16:472 5. Kopecky C, Genser B, Drechsler C et al (2015) Quantification of HDL proteins, cardiac events, and mortality in patients with type 2 diabetes on hemodialysis. Clin J Am Soc Nephrol 10:224–231 6. Roberts JM (1998) Endothelial dysfunction in preeclampsia. Semin Reprod Endocrinol 16:5–15 7. Flavahan NA (1992) Atherosclerosis or lipoprotein-induced endothelial dysfunction. Potential mechanisms underlying reduction in EDRF/nitric oxide activity. Circulation 85:1927–1938

13

8. Ray K, Wainwright NW, Visser L et al (2012) Changes in HDL cholesterol and cardiovascular outcomes after lipid modification therapy. Heart 98:780–785 9. Toth PP, Barter PJ, Rosenson RS et al (2013) High-density lipoproteins: a consensus statement from the National Lipid Association. J Clin Lipidol 7:484–525 10. Gansevoort RT, Correa-Rotter R, Hemmelgarn BR et al (2013) Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet 382:339–352 11. Maynard SE, Min JY, Merchan J et al (2003) Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 111:649–658 12. Barker DJ (2004) The developmental origins of well-being. Philos Trans R Soc Lond B Biol Sci 359:1359–1366 13. Rich-Edwards JW, Fraser A, Lawlor DA, Catov JM (2014) Pregnancy characteristics and women’s future cardiovascular health: an underused opportunity to improve women’s health? Epidemiol Rev 36:57–70 14. Belo L, Caslake M, Gaffney D et al (2002) Changes in LDL size and HDL concentration in normal and preeclamptic pregnancies. Atherosclerosis 162:425–432 15. Gbandjaba NY, Ghalim N, Hassar M et al (2012) Paraoxonase activity in healthy, diabetic, and hemodialysis patients. Clin Biochem 45:470–474 16. Barter PJ, Nicholls S, Rye KA, Anantharamaiah GM, Navab M, Fogelman AM (2004) Antiinflammatory properties of HDL. Circ Res 95:764–772 17. Phillips JC, Wriggers W, Li Z, Jonas A, Schulten K (1997) Predicting the structure of apolipoprotein A-I in reconstituted highdensity lipoprotein disks. Biophys J 73:2337–2346 18. Demir B, Demir S, Atamer Y et al (2011) Serum levels of lipids, lipoproteins and paraoxonase activity in pre-eclampsia. J Int Med Res 39:1427–1431 19. Riwanto M, Landmesser U (2013) High density lipoproteins and endothelial functions: mechanistic insights and alterations in cardiovascular disease. J Lipid Res 54:3227–3243 20. Shroff R, Speer T, Colin S et al (2014) HDL in children with CKD promotes endothelial dysfunction and an abnormal vascular phenotype. J Am Soc Nephrol 25:2658–2668 21. Mackness MI, Durrington PN (1995) HDL, its enzymes and its potential to influence lipid peroxidation. Atherosclerosis 115:243–253 22. Demirci O, Tugrul AS, Dolgun N, Sozen H, Eren S (2011) Serum lipids level assessed in early pregnancy and risk of pre-eclampsia. J Obstet Gynaecol Res 37:1427–1432 23. Aviram M (1983) Plasma lipoprotein separation by discontinuous density gradient ultracentrifugation in hyperlipoproteinemic patients. Biochem Med 30:111–118 24. Khersonsky O, Tawfik DS (2006) Chromogenic and fluorogenic assays for the lactonase activity of serum paraoxonases. Chem BioChem 7:49–53 25. Rashid G, Benchetrit S, Fishman D, Bernheim J (2004) Effect of advanced glycation end-products on gene expression and synthesis of TNF-alpha and endothelial nitric oxide synthase by endothelial cells. Kidney Int 66:1099–1106 26. Gugliucci A, Caccavello R, Kotani K, Sakane N, Kimura S (2013) Enzymatic assessment of paraoxonase 1 activity on HDL subclasses: a practical zymogram method to assess HDL function. Clin Chim Acta 415:162–168 27. Cohen E, Aviram M, Khatib S, Volkova N, Vaya J (2016) Human carotid atherosclerotic plaque protein(s) change HDL protein(s) composition and impair HDL anti-oxidant activity. BioFactors 42:115–128 28. Zitman-Gal T, Golan E, Green J, Bernheim J, Benchetrit S (2012) Vitamin D receptor activation in a diabetic-like environment:

13

Archives of Gynecology and Obstetrics

29.

30.

31. 32.

33.

34.

35.

36. 37. 38.

39.

40.

41.

42. 43. 44. 45. 46.

potential role in the activity of the endothelial pro-inflammatory and thioredoxin pathways. J Steroid Biochem Mol Biol 132:1–7 Einbinder Y, Ohana M, Benchetrit S et al (2016) Glucagon-like peptide-1 and vitamin D: anti-inflammatory response in diabetic kidney disease in db/db mice and in cultured endothelial cells. Diabetes Metab Res Rev 32:805–815 Luscher TF, Landmesser U, von Eckardstein A, Fogelman AM (2014) High-density lipoprotein: vascular protective effects, dysfunction, and potential as therapeutic target. Circ Res 114:171–182 Feig JE, Hewing B, Smith JD, Hazen SL, Fisher EA (2014) Highdensity lipoprotein and atherosclerosis regression: evidence from preclinical and clinical studies. Circ Res 114:205–213 Hedrick CC, Castellani LW, Wong H, Lusis AJ (2001) In vivo interactions of apoA-II, apoA-I, and hepatic lipase contributing to HDL structure and antiatherogenic functions. J Lipid Res 42:563–570 van der Steeg WA, Holme I, Boekholdt SM et al (2008) High-density lipoprotein cholesterol, high-density lipoprotein particle size, and apolipoprotein A-I: significance for cardiovascular risk: the IDEAL and EPIC-Norfolk studies. J Am Coll Cardiol 51:634–642 Regieli JJ, Jukema JW, Doevendans PA et al (2009) Paraoxonase variants relate to 10-year risk in coronary artery disease: impact of a high-density lipoprotein-bound antioxidant in secondary prevention. J Am Coll Cardiol 54:1238–1245 Bhattacharyya T, Nicholls SJ, Topol EJ et al (2008) Relationship of paraoxonase 1 (PON1) gene polymorphisms and functional activity with systemic oxidative stress and cardiovascular risk. JAMA 299:1265–1276 Shih DM, Gu L, Xia YR et al (1998) Mice lacking serum paraoxonase are susceptible to organophosphate toxicity and atherosclerosis. Nature 394:284–287 Genc H, Uzun H, Benian A et al (2011) Evaluation of oxidative stress markers in first trimester for assessment of preeclampsia risk. Arch Gynecol Obstet 284:1367–1373 Uzun H, Benian A, Madazli R, Topcuoglu MA, Aydin S, Albayrak M (2005) Circulating oxidized low-density lipoprotein and paraoxonase activity in preeclampsia. Gynecol Obstet Invest 60:195–200 Hubel CA, McLaughlin MK, Evans RW, Hauth BA, Sims CJ, Roberts JM (1996) Fasting serum triglycerides, free fatty acids, and malondialdehyde are increased in preeclampsia, are positively correlated, and decrease within 48 hours post partum. Am J Obstet Gynecol 174:975–982 Rosing U, Samsioe G, Olund A, Johansson B, Kallner A (1989) Serum levels of apolipoprotein A-I, A-II and HDL-cholesterol in second half of normal pregnancy and in pregnancy complicated by pre-eclampsia. Horm Metab Res 21:376–382 Ramet ME, Ramet M, Lu Q et al (2003) High-density lipoprotein increases the abundance of eNOS protein in human vascular endothelial cells by increasing its half-life. J Am Coll Cardiol 41:2288–2297 Tai SC, Robb GB, Marsden PA (2004) Endothelial nitric oxide synthase: a new paradigm for gene regulation in the injured blood vessel. Arterioscler Thromb Vasc Biol 24:405–412 Mineo C, Shaul PW (2012) Novel biological functions of highdensity lipoprotein cholesterol. Circ Res 111:1079–1090 Kaseda R, Jabs K, Hunley TE et al (2015) Dysfunctional highdensity lipoproteins in children with chronic kidney disease. Metabolism 64:263–273 Ortiz-Munoz G, Couret D, Lapergue B et al (2016) Dysfunctional HDL in acute stroke. Atherosclerosis 253:75–80 Szarka A, Rigo J Jr, Lazar L, Beko G, Molvarec A (2010) Circulating cytokines, chemokines and adhesion molecules in normal pregnancy and preeclampsia determined by multiplex suspension array. BMC Immunol 11:59

Archives of Gynecology and Obstetrics 47. Fainaru O, Lichtenberg D, Pinchuk I, Almog B, Gamzu R, Kupferminc M (2003) Preeclampsia is associated with increased susceptibility of serum lipids to copper-induced peroxidation in vitro. Acta Obstet Gynecol Scand 82:711–715 48. Wakatsuki A, Ikenoue N, Okatani Y, Shinohara K, Fukaya T (2000) Lipoprotein particles in preeclampsia: susceptibility to oxidative modification. Obstet Gynecol 96:55–59

49. Sarandol E, Safak O, Dirican M, Uncu G (2004) Oxidizability of apolipoprotein B-containing lipoproteins and serum paraoxonase/ arylesterase activities in preeclampsia. Clin Biochem 37:990–996

13