Pregnancy‐associated venous insufficiency course with placental and systemic oxidative stress

Abstract The development of lower extremity venous insufficiency (VI) during pregnancy has been associated with placental damage. VI is associated with increased oxidative stress in venous wall. We have investigated potential disturbance/dysregulation of the production of reactive oxygen species (ROS) in placenta and its eventual systemic effects through the measurement of malondialdehyde (MDA) plasma levels in women with VI. A total of 62 women with VI and 52 healthy controls (HCs) were studied. Levels of nicotinamide adenine dinucleotide phosphate‐oxidase 1 (NOX1), 2 (NOX2), inducible nitric oxide synthase (iNOS), endothelial (eNOS), poly(ADP‐ribose) polymerase PARP (PARP) and ERK were measured in placental tissue with immunohistochemistry and RT‐qPCR. Plasma and placental levels of MDA were determined by colorimetry at the two study times of 32 weeks of gestation and post‐partum. Protein and gene expression levels of NOX1, NOX2, iNOS, PARP and ERK were significantly increased in placentas of VI. eNOS activity was low in both study groups, and there were no significant differences in gene or protein expression levels. Women with VI showed a significant elevation of plasma MDA levels at 32 weeks of gestation, and these levels remained elevated at 32 weeks post‐partum. The MDA levels were significantly higher in placentas of women with VI. Placental damage that was found in the women with VI was characterized by overexpression of oxidative stress markers NOX1, NOX2, and iNOS, as well as PARP and ERK. Pregnant women with VI showed systemic increases in oxidative stress markers such as plasma MDA levels. The foetuses of women with VI had a significant decrease in their venous pH as compared to those from HC women. The situation of oxidative stress and cellular damage created in the placenta is in coexpression with the production of a pH acidification.


| INTRODUC TI ON
Lower extremity venous insufficiency (VI) is a vascular disorder that is characterized by alteration of the peripheral venous system that reduces or impedes venous return. 1,2 Various epidemiological studies worldwide provide evidence that VI is one of the chronic diseases that shows the greatest variability in its incidence and prevalence. 3 The yearly incidence of varicose veins in women is 2.6%, according to the Framingham study. 4 Pregnancy is one of the factors that increases the risk of VI development. 5 Consequently, VI is a common complication of pregnancy and affects a third of women during their first pregnancy. 6,7 Several pathogenic mechanisms appear to be involved in this association during pregnancy, including an increased abdominal pressure and blood volume and an enhanced cardiac output with a secondary augmentation of the venous pressure in the lower extremities. 8,9 Furthermore, the hormonal changes experienced by pregnant women may also contribute to venous damage. 10 Age, obesity, sex hormones and family history of VI increase the risk of varicose veins during pregnancy. 7,[10][11][12] Interestingly, pregnancy-associated VI appears to involve more venous territories than those that are clinically evident in the lower extremities. An unexpected association of VI with the involvement and damage of the placenta has been recently shown. 9 The placental tissues of women with pregnancy-associated VI show structural remodelling and hypoxic cellular stress with a higher number of villi and syncytial knots and enhanced apoptotic cellular death. 9,13 The formation of the placental lesions is driven by the compromised arterial blood flow, but its pathogenesis is complex, and several molecular pathways appear to be involved. One of these pathways is the overexpression of different enzymes that produce the components of reactive oxygen species (ROS) in the villi of the placenta, which can be damaged by arterial diseases such as pre-eclampsia and arterial hypertension. [14][15][16] Furthermore, ROS overexpression is found in the venous wall of VI patients. 17 An imbalance between ROS production and antioxidant defence mechanisms results in an oxidative stress that can produce different molecular alteration phenomena such as the oxidation of lipids, proteins and nucleic acids, enzyme inactivation and tissue inflammation. [18][19][20] Furthermore, the interaction of ROS with nitric oxide (NO) promotes its impact on the cellular ageing process. 21,22 NO is considered to be a critical mediator of the cellular damage that is induced by oxidative stress. 22,23 The induced isoform (iNOS) and endothelial nitric oxide synthase (eNOS) are the main sources of cellular NO. 19,24 Disturbances in iNOS and eNOS have been described in the placentas of patients with pre-eclampsia. 25,26 Cellular oxidative stress is also linked to the increased generation of ROS through the activation of the transmembrane nicotinamide adenine dinucleotide phosphate-oxidase (NOX) enzyme family. 27 The isoforms NOX1 and NOX2 are mayor players in the production of superoxide anion. 19 It has been shown that both isoforms have increased expression in the placentas of women with vascular hypertension such as pre-eclampsia, showing an alteration of its expression and activity in these vascular pathologies. 28,29 Authors have observed how oxidative stress and metabolic changes can create DNA damage that is associated with an increase in poly(ADP-ribose) polymerase PARP (PARP) repair activity. [30][31][32] In this sense, the mitogen-activated protein kinase/ extracellular signal-regulated (MAPK/ERK) pathway has been mentioned as a point of great importance in cellular damage processes. Signalling pathways (MAPK/ERK) play fundamental roles in a wide range of cellular processes and are often deregulated in disease states. An important mode of action for these pathways is to control gene expression, in particular by regulating transcription with great tissue impact. 33,34 One of the points that is pointed out in an important way to know the damage that oxidative stress causes in cells is the production of malondialdehyde (MDA) as a result of lipid peroxidation.
The particular reaction of ROS with lipids is generally known as 'lipid peroxidation'. Isoprostane MDA is a widely accepted biomarker of oxidative stress with biological consequences. 35 It is possible that the placental hypoxaemic damage that is found in women with pregnancy-associated VI might be associated with an abnormal ROS regulation in the placental tissue. This potential oxidative stress in the placenta might also have systemic effects such as increased levels of oxidative metabolites. In this work, we investigated the expression levels of the iNOS, eNOS, NOX1, NOX2 and PARP enzymes in the different cellular components of the placenta from women with pregnancy-associated VI. We studied the placentas of women with pregnancy-associated VI in parallel with the healthy control (HC) group. In addition, we analysed the placental tissue and peripheral blood MDA levels as biomarkers of the oxidative stress and physiological stress that occurs in women with systemic increases in oxidative stress markers such as plasma MDA levels. The foetuses of women with VI had a significant decrease in their venous pH as compared to those from HC women. The situation of oxidative stress and cellular damage created in the placenta is in coexpression with the production of a pH acidification.

K E Y W O R D S
ERK, inducible nitric oxide synthase, Lower extremity venous insufficiency, malondialdehyde, NOX1, NOX2, PARP, Pregnancy pregnancy-associated VI and the HCs, both during pregnancy and after delivery.

| Study design
In this prospective work, we studied 62 women who were diagnosed as having VI without a concomitant disease during 32 weeks of their pregnancy (women with pregnancy-associated VI). The study point at 32 weeks of gestational age is an international consensus standard in the study of placental insufficiency. The 32-week study is due to foetal maturity and other maternal-foetal structures. 36,37 Women in the VI were newly diagnosed without and previous clinical evidence of the disease. In parallel, we studied 52 women who were healthy for the same 32 weeks of pregnancy, and we also studied them throughout their remaining pregnancy (HC).
This study was conducted in women that visited the obstetric consultations in the 32 weeks of gestation. Once they signed the informed consent form, a clinical history, general physical examination and laboratory measurements were obtained, and an exploration of their lower limbs was performed with an Eco-Doppler (portable M-Turbo Eco-Doppler; SonoSite, Inc.) transducer at 7.5 MHz. The examination was performed, while the women were in a standing position and the leg was explored in external rotation and with support in the contralateral leg. The study included the major saphenous axis from the inguinal region to the ankle and femoral veins.
The classification of VI in the pregnant women under study was performed according to the Classification System for Chronic Venous Disorders (CEAP). 38 The pregnancy was routinely monitored at the centre, and placental samples were obtained from the women at their time of delivery.

| Samples
The placental tissue biopsies were taken once the placenta was expelled and used to immunohistochemical, genetic and molecular studies. In all cases, five fragments of the placenta were taken with a scalpel to ensure that the samples included several cotyledons Blood samples were obtained from the study population by puncturing the superficial vein of the elbow fold after placing a compressor on the arm (the extraction was done in the morning and fasting for 10 hours). Plasma was obtained from said blood samples for MDA study. The entire volume of the blood sample was transferred from the heparin tubes to the sterile centrifuge tubes, where they were centrifuged at 1500 g for 15 minutes. Next, the plasma was collected and transferred to 1.5-mL Eppendorf tubes, where they were kept frozen at −80°C until they were used for the study.
The haematological and biochemical parameters were analysed by the Clinical Analysis Service of the University Hospital according to standardized protocols.

| Genetic or molecular studies
The amounts of cDNA in each sample were quantified with real-time polymerase chain reaction (qPCR) of the following genes of interest: NOX1, NOX2, iNOS, eNOS and PARP. The results were normalized using the constitutively expressed gene TBP (TATA box-binding protein). De novo primers or primers that were specific for all genes of interest were designed (Table S1) using the online applications Primer-BLAST 39 and AutoDimer. 40 RNA was extracted using the guanidine-phenol-chloroform isothiocyanate method described by Ortega et al. 41 The qPCR was performed in a StepOnePlus™ System (Applied Biosystems-Life Technologies) using the relative standard curve method. The reaction was performed as follows: 5 µl of each sample containing cDNA diluted 1/20 in nuclease-free water was mixed with 10 µL of iQ™ SYBR ® Green Supermix (Bio-Rad Laboratories), 1 µL of forward primer (6 μmol/L), 1 µL of reverse primer (6 μmol/L) and 3 µL of DNase-and RNase-free water in a MicroAmp ® 96-well plate (Applied Biosystems-Life Technologies) for a total reaction volume of 20 µL.

| Immunohistochemical studies
The samples that were preserved in MEM were washed/hydrated several times with medium without antibiotic to remove blood cells and were cut into fragments that were kept in different fixatives, including F13 (60% ethanol, 20% methanol, 7% polyethylene glycol and 13% distilled H 2 O). After the samples were fixed for the necessary amount of time in each fixing solution, they were dehydrated following standardized protocols. 42 The antigen-antibody reaction was detected with the ABC (avidin-biotin complex) method with peroxidase or alkaline phosphatase as a chromogen, according to the following protocol: 1. wash the samples three times with 1× PBS 1× for 5 minutes each; 2. block the non-specific binding sites with 3% bovine serum albumin (BSA) in PBS for 30 minutes at room temperature; 3. incubate with the primary antibody (Table S2)  For each of the patients in the established groups, 5 sections and 10 fields per section were examined by random selection in placental villi and decidual cells. The patients were described as positive when the average marked area in the analysed sample was greater than or equal to 5% of the total, following the anatomopathological protocol of Cristobal et al. 42 Positive cells for immunohistochemical studies were counted under a microscope (1000×) in 10 random areas (0.5 mm 2 per patient). The preparations were examined under a Zeiss Axiophot optical microscope (Carl Zeiss) that was equipped with an AxioCam HRc digital camera (Carl Zeiss).

| Tissue samples
A total of 20 mg of placental tissue was used for the analysis. The protocol that was used was as follows: 1. rinse the tissue in cold PBS; 2. prepare the lysis solution: 300 mL of lysis buffer with MDA and 3 µL of BHT (100X); 4. homogenize the tissue in 303 µL of lysis solution (buffer + BHT) with 10-15 passes with a homogenizer on ice; and 5. centrifuge at 13 000 g for 10 minutes to remove the insoluble material and collect the supernatant.

| Statistical analysis
For the statistical analysis, the GraphPad Prism ® 6.0 program was used, and the Mann-Whitney U test was applied. The data are expressed as the median with interquartile range (IQR). The significance was set at P < .05 (*). If the study variables were not quantitative, we used Pearson's chi-squared test or Fisher's exact test, when applicable.

| Placentas from women with pregnancyassociated VI show increased iNOS but normal eNOS expression
We

| Placentas from women with pregnancyassociated VI show increased PARP gene and protein expression
We analysed the gene expression of PARP by RT-qPCR in the placentas of women with pregnancy-associated VI and in the HCs. Our results showed a significant increase in the level of PARP expression in the placentas from the women with pregnancy-associated VI

| Placentas from women with pregnancyassociated VI show increased MAPK/ERK gene and protein expression
We analysed the gene and protein expression of MAPK, specific  Figure 5, Panels B-C).  Figure 6A).

| Increased MDA levels in the plasma and placental tissue of women with pregnancy-associated VI
Next, we studied systemic lipid peroxidation, as measured by plasma MDA levels, in both groups of women at the time of inclusion in the study at 32 weeks of pregnancy and at 32 weeks after delivery. We found that the plasma MDA levels in the women with pregnancy-associated VI were significantly higher than those found in the HC women at both times of the study (VI = 14.55 [6.08- levels at pregnancy were significantly higher than those after delivery in both group of women (P = .0402 compared VI group, P = .0350 compared HC group; Figure 6B).

| Foetuses of mothers with pregnancy-associated VI show a decrease in the umbilical cord venous pH
We investigated the umbilical cord venous pH of foetuses from women with pregnancy-associated VI and from HC women ( Figure 6C). We observed that the foetuses of women with  Pregnancy is a neuroendocrine and mechanical challenge to women that can induce or favour the expression of different disorders, including cardiovascular diseases such as VI. 43,44 The development of pregnancy-associated VI has been associated with the development of placental lesions. 9,13 In this paper, we investigated the pathogenic mechanisms that were involved in the induction of the cellular lesions that were found in the placentas from women with pregnancy-associated VI. Oxidative stress is a critical mechanism for cellular damage. 19 We found a specific pattern of overexpression of NOX1 and NOX2 enzymes in the placentas from women with pregnancy-associated VI. A widespread increased expression of NOX1 in the four different placenta cellular types was observed, as was an enhanced NOX2 expression that was specific to the syncytiotrophoblast and decidual cells in these damaged placentas. Numerous authors have shown how placental oxidative stress has very important role in pregnancy complications. [45][46][47][48][49] Placentas from women with arterial pregnancy-associated diseases such as pre-eclampsia show increased expression of NOX1 in the syncytiotrophoblast and decidual cells. 27 Furthermore, the increased expression of NOX1 and NOX2 has been described in the syncytiotrophoblast and homeostasis. 25 We also investigated the severity of the oxidative stress that is present in the placentas of women with pregnancy-associated VI by indirectly analysing the damage to the cellular DNA. This molecular event triggers the response of different protective mechanisms, including the activation of the PARP enzyme. 55,60 Thus, PARP expression is a biomarker of DNA damage. 32 Indeed, increased PARP expression has been observed in cytotrophoblast cells suffering from oxidative stress. 49 Our findings demonstrate an enhanced PARP expression in the syncytiotrophoblast and decidual cells in the placentas of women with pregnancy-associated VI. However, we cannot ascribe the induction of this DNA repair mechanism solely to oxidative stress, since other mechanisms can also be involved in DNA damage. Recently, it has been shown that a culture medium with a low pH can induce PARP expression in decidual cells. 61 We found a low pH in the umbilical cord venous blood of women with is a significant increase in plasma levels of MDA. 72 The persistence of the increased MDA levels in the women with pregnancy-associated VI favours a mechanism of oxidative stress that is not solely dependent on the placenta. It is plausible to suggest that women with pregnancy-associated VI suffer from a more generalized venous disease that remains after their pregnancy.

| D ISCUSS I ON
F I G U R E 6 A, Malondialdehyde (MDA) levels in pmol/mg in the placental tissues of pregnant women with lower extremity venous insufficiency (VI) and healthy control (HC) pregnant women. B, Plasma levels of malondialdehyde (MDA) in μmol/L of pregnant women with VI and HCs at 32 wk of gestation and at 32 wk post-partum. C, Significant decrease of foetal venous pH in mothers with pregnancy-associated VI. Data shown median and IQR. *P-value < .05 (Mann-Whitney U test) Future studies with a much larger sample size are necessary to be able to take into account determining factors in foetal well-being, this point is limiting. Future works should establish the potential clinical relevance of pregnancy-associated VI and the value of preventive strategies, including the role of angiogenic and proangiogenic factors at a systematic level. Our data support the hypothesis that these women show severe oxidative stress in their placentas as well as a systemic and long period of increased generation of oxidative markers.

CO N FLI C T S O F I NTE R E S T
None declared.