Preeclampsia Is Associated with Lower Production of Vascular Endothelial Growth Factor by Peripheral Blood Mononuclear Cells

Archives of Medical Research 45 (2014) 561e569

ORIGINAL ARTICLE

Preeclampsia Is Associated with Lower Production of Vascular Endothelial Growth Factor by Peripheral Blood Mononuclear Cells Marıa G. Cardenas-Mondragon,a,b Gabriela Vallejo-Flores,b,c Jose Delgado-Dominguez,d Juan F. Romero-Arauz,e Alejandro Gomez-Delgado,b Guadalupe Aguilar-Madrid,f Juan J. Sanchez-Barriga,g and Janeth Marquez-Acostaa,e a

Unidad de Investigacion Medica en Medicina Reproductiva, UMAE Hospital de Ginecologıa y Obstetricia No. 4 ‘‘Luis Castelazo Ayala’’, Instituto Mexicano del Seguro Social (IMSS), Mexico, D.F., Mexico b Unidad de Investigacion Medica en Enfermedades Infecciosas y Parasitarias, UMAE Hospital de Pediatrıa, ‘‘Dr. Silvestre Frenk Freund’’, CMN Siglo XXI, IMSS, Mexico, D.F., Mexico c Departamento de Inmunologıa, Escuela Nacional de Ciencias Biologicas, Instituto Politecnico Nacional, Mexico, D.F., Mexico d Departamento de Medicina Experimental, Facultad de Medicina, Universidad Nacional Autonoma de Mexico, Mexico, D.F., Mexico e Servicio de Complicaciones Hipertensivas del Embarazo, UMAE Hospital de Ginecologıa y Obstetricia No. 4 ‘‘Luis Castelazo Ayala’’, IMSS, Mexico, D.F., Mexico f Unidad de Investigacion en Salud en el Trabajo, CMN Siglo XXI, IMSS, Mexico, D.F., Mexico g Direcci on de Investigacion Operativa en Epidemiologıa, Direccion General de Epidemiologıa, Secretaria de Salud, Mexico, D.F., Mexico Received for publication March 19, 2014; accepted October 17, 2014 (ARCMED-D-14-00163).

Background. Recent studies show that vascular endothelial growth factor (VEGF) downregulation is implicated in preeclampsia (PE) pathophysiology. This study assessed the relationship between PE and VEGF levels produced by peripheral blood mononuclear cells (PBMCs) and their serum levels. Methods. A cross-sectional design was performed in 36 patients who had hypertensive disorders during pregnancy. We also used a longitudinal design with 12 pregnant women with risk factors for PE development and/or abnormal uterine arteries by Doppler study. VEGF and soluble fms-like tyrosine kinase-1 (sFlt-1) levels were measured for all patients in both designs. Results. sFlt-1 serum was higher in preeclamptic patients (n 5 26), whereas VEGF produced by stimulated PBMCs was lower than in healthy pregnant women and VEGF levels produced by stimulated PBMCs were even lower ( p !0.003) in severe PE (n 5 16). The receiver-operating characteristic curve analysis allowed establishing a cut-off value to identify patients with PE. VEGF production by PBMCs was 339.87 pg/mL. In addition, a robust linear regression model was performed to adjust the variance in VEGF levels. The patients’ age decreased VEGF levels and was adjusted by weeks of gestation (WG) in our model. In the longitudinal study, 7/12 patients developed PE. VEGF produced by PBMCs cells was significantly lower in PE at 24e26 WG. Conclusions. VEGF production by PBMCs is inhibited during PE, creating a downregulation of the microenvironment; this deficiency may contribute to the pathogenesis of disease. Ó 2014 IMSS. Published by Elsevier Inc. Key Words: VEGF, sFlt-1, Preeclampsia, Hypertensive disorders of pregnancy.

Introduction Address reprint requests to: Marıa G. Cardenas-Mondragon, PhD, Unidad de Investigaci on en Medicina Reproductiva, Av. Rio Magdalena 289,   Col Tizapan San Angel, Del. Alvaro Obregon, 1090 Mexico, D.F., Mexico; Phone: (þ52) (55) 56276900 ext: 22407; FAX: (þ52) (55) 46276900 ext: 22407; E-mail: [email protected]

Preeclampsia (PE) is a clinical syndrome present in 5 to 8% of all pregnant women and remains as a major cause of maternal and neonatal mortality and morbidity worldwide (1e3). It is characterized by endothelial dysfunction,

0188-4409/$ - see front matter. Copyright Ó 2014 IMSS. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arcmed.2014.10.004

562

Cardenas-Mondragon et al./ Archives of Medical Research 45 (2014) 561e569

hypertension and proteinuria after 20 weeks of gestation (WG) and/or a fetal syndrome (1,2,4e6). PE also remains a leading cause of prematurity as placenta delivery is currently the only way to successfully treat the disorder. The etiology and pathogenesis of this pregnancy-specific syndrome are unclear. Accumulated evidence suggests that PE may be the result of an imbalance in angiogenic factors (7e14). Recent studies have hypothesized that vascular endothelial growth factor (VEGF) and its receptors are implicated in the pathophysiology of PE (10,12,14e18). VEGF is a homodimeric 34e42 kDa glycoprotein with potent angiogenic activity (19). During pregnancy, VEGF is essential for normal placenta development, for trophoblast proliferation and for the development of embryonic vasculature. Three VEGF receptors have been described: VEGFR-1, also known as fms-like tyrosine kinase (Flt-1) (20,21); VEGFR-2, also known as kinase-insert domaincontaining receptor (KDR and Flk-1) (21,22) and VEGFR-3, also known as Flt-4 (23). A soluble truncated form of the Flt-1 (soluble fms-like tyrosine kinase-1, sFlt1) receptor produces alternative splicing and contains an extracellular domain that binds to the ligand but lacks the transmembrane and cytoplasmic domains (24), acting as a natural potent VEGF and placental growth factor (PIGF) antagonist (7) by binding them and blocking the interaction with their receptors Flt-1 and Flk-1 on the cellular membrane. Low serum levels of free VEGF (25,26) and elevated sFlt-1 serum levels have been reported in PE (10,12,13,16,17,27). sFlt-1 elevation correlates with severity of the disease. An in vivo study has demonstrated that overexpression of sFlt-1 in pregnant rats produced a PE-like syndrome with hypertension, proteinuria and glomerular endotheliosis. This anti-angiogenic state can be rescued by administration of VEGF and PIGF (12). A 50% reduction of VEGF in transgenic mice resulted in a clinical entity similar to human PE, characterized by proteinuria and endotheliosis (28). Patients receiving antiVEGF antibodies as a treatment for metastatic renal cancer developed hypertension and proteinuria (29,30). Hypertension and proteinuria decreased after therapy cessation (29). VEGF and sFlt-1 are produced fundamentally by the placenta but these molecules and their receptors are also synthesized and secreted by endothelial cells and peripheral blood mononuclear cells (PBMCs) (24,31e33). PBMCs obtained from preeclamptic women produced significantly higher amounts of sFlt-1 under normal tissue culture conditions compared with PBMCs from normal pregnant women (33). Pregnant non-human primates undergoing a reduction in placental blood flow by induction of uteroplacental ischemia (UPI) resulted in blood pressure elevation, development of proteinuria, and histological changes, identical to human PE. UPI also resulted in an increase in circulating sFlt-1 and expressed significantly higher sFlt-1 mRNA levels in PBMCs compared with the sham group (34). On

the other hand, macrophages are key regulators of the angiogenic switch in tumors; they may promote angiogenesis by producing angiogenic growth factors including VEGF (35). Macrophages produced by VEGF may recruit and interact with other cells in the tumor microenvironment (e.g., neutrophils) (35,36). A previous study reported that the proportion of T and NK cells that shows intracellular VEGF expression in the peripheral blood was markedly decreased in patients with PE when compared to healthy pregnant women (36). The cause and clinical significance of these findings remain to be determined. Previous studies have evaluated the use of Doppler as a test to identify patients at risk of developing PE, reporting that sensitivity improves when the test is performed after 24e26 WG and when a diastolic constant notch is one of the criteria (37e41). When combining uterine artery Doppler test with maternal history and mean blood pressure, a detection rate of approximately 90 for 10% of false positives cases is achieved (42). In this study we assessed sFlt-1 and VEGF levels in serum samples and in PBMC culture supernatant from PE patients and from patients with risk factors for PE development. We found that although VEGF levels were significantly reduced in stimulated PBMCs in PE, serum Flt-1 was significantly elevated compared with their levels in normal pregnancy. These deficient VEGF levels may contribute to the pathogenesis of PE.

Materials and Methods Patient Details The IMSS Scientific and Ethics Committee approved this study. Participants were informed regarding the nature of the study and written informed consent was obtained from all participants. These women were patients admitted to the Hypertensive Diseases of Pregnancy Clinic of the Gynecology Hospital ‘‘Luis Castelazo Ayala’’ of the IMSS. Study participants were at gestational age $20 weeks and had new-onset hypertension. Hypertension was defined as systolic blood pressure $140 mm Hg and/or diastolic blood pressure $90 mm Hg, measured twice at least 6 h apart and that returned to normal values within 3 months after delivery. Hypertensive disorder of pregnancy was defined according to the criteria of the American College of Obstetrics and Gynecology (6). Gestational hypertension (GH) was defined as isolated hypertension without significant proteinuria and mild preeclampsia (mPE) as hypertension and significant proteinuria ($300 mg of protein in a 24 h urine specimen or a protein:creatinine ratio $0.30 in a random urine sample) (43,44). Severe PE (sPE) was considered when either HELLP syndrome (hemolysis, elevated liver enzymes, low platelet count), eclampsia, or PE with severe hypertension (systolic blood pressure $160 mmHg and/or diastolic blood pressure $110 mmHg on at least

VEGF Produced by PBMCs Is Lower in Preeclampsia

563

two occasions 6 h apart) or massive proteinuria ($5 g/day or a protein:creatinine ratio $3.87) (43) were present. Other parameters included were persistent headache or visual disturbances, presence of hemolysis or abnormal liver enzyme levels with right upper quadrant or epigastric pain, thrombocytopenia (!100,000/mL in at least two samples), serum creatinine O1.0 mg/dL, pulmonary edema, oliguria (!500 mL per day), or oligohydramnios. Women with any form of diabetes mellitus or other concurrent medical complications before or developed during pregnancy were not considered eligible for the study.

centrifuged at 2000 rpm for 20 min with the brake off. Cultured PBMCs from the interface were transferred into a fresh tube and the cell suspension was centrifuged at 1200 rpm for 15 min at room temperature, the supernatant was discarded and the cell pellet was washed with PBS and centrifuged for 10 min at 1200 rpm. The supernatants were discarded and the cells were resuspended in RPMI medium containing 10% fetal bovine serum, 50 mM b-mercaptoethanol, 2 mM L-glutamine, 100 U/mL penicillin and 100 mg/mL streptomycin.

Study Design

Cell Culture

In a cross-sectional design, VEGF and sFlt-1 levels produced by PBMCs and its serum levels were determined in pregnant women: patients with GH, patients with PE (with mPE and with sPE), and healthy pregnant (HP) women, as well as a group of healthy and non-pregnant women (HnoP). A longitudinal design was used for the prediction of PE. Protein levels were determined at 24e26, 28, and 32 WG in patients with risk factors for PE and/or abnormal uterine arteries by Doppler study between 24 and 26 WG [to evaluate a modified uterine artery score based on the pulsatility index (PI $1.4) or notching presence in the Doppler velocity waveform]. The uterine artery score was constructed assigning one point to each abnormal parameter—high pulsatility index and presence of notch—thus ranging from 0 (normal findings in both uterine arteries) to 4 (notch and high pulsatility index in both uterine arteries)].

To determine human VEGF levels produced by PBMCs, these human cells were cultured immediately after isolation by Lymphoprep: 1  106 cells/mL were plated in 24-well culture plates and cultured in RPMI supplemented with 10% fetal calf serum, 50 mM b-mercaptoethanol, 2 mM L-glutamine, 100 U/mL penicillin and 100 mg/mL streptomycin sulfate. The cultured cells were or were not stimulated with 10 mg/mL phytohemagglutinin (PHA) for 1 and 5 days under normoxic conditions. For sFlt-1 levels, cells were cultured in RPMI supplemented with IL-2 (10 ng/mL), stimulated with 10 mg/mL PHA or unstimulated and cultured for 1 and 6 days under normoxic conditions. The cultured medium was centrifuged at 1500 rpm for 10 min and the supernatant was quickly frozen.

Study Limitations

Immunoreactive VEGF and sFLT-1 in serum and cell culture supernatants were determined by corresponding sandwich enzyme immunoassay technique kits (clone 26503, Quantikine Kit, R&D Systems, Minneapolis, MN) following manufacturer’s instructions. All samples were measured in duplicate. A subgroup of samples was done in quadruplicate using different plates for reproducibility evaluation.

In both cross-sectional and longitudinal designs, the sample size was small, but for cross-sectional design the results for sFlt-1 serum levels and VEGF serum levels are in line with previous studies with a larger sample size. Samples A sample of venous blood was drawn under basal conditions and without hormonal or drug stimulus. Peripheral blood samples were collected without anticoagulants for obtaining sera or in heparin tubes for obtaining PBMCs. Sera was separated by centrifugation at 3000 rpm for 15 min at 4 C, and the serum obtained was immediately stored in aliquots at 80 C until assayed. All samples were collected before either induction of labor or cesarean delivery. PBMC Isolation PBMCs were obtained using Lymphoprep following the manufacturer’s instructions. (Axis-Shield PoC AS, Oslo, Norway). Briefly, 8 mL of the blood sample was diluted 1:3 with PBS and overlaid on 4 mL Lymphoprep and

VEGF and sFLT-1 Immunoassay

Statistical Analysis Differences among $3 groups were compared by one-way analysis of variance (ANOVA) followed by post-hoc procedures (Scheffe’s test) or by the Kruskal-Wallis test followed by Mann-Whitney U test for non-normally distributed variables. For VEGF and sFlt-1 levels we used the last test due to non-normal distribution. A two-tailed p !0.05 was considered statistically significant. The performance of screening cases with sPE was evaluated by receiveroperating characteristic (ROC) curve analysis. The ROC curve was used to establish a cut-off value for VEGF levels produced by PBMCs and finally a robust linear regression model was run to adjust the variance of the levels of VEGF.

564

Cardenas-Mondragon et al./ Archives of Medical Research 45 (2014) 561e569

Table 1. Clinical and demographic characteristics of pregnant women with hypertensive disorders of pregnancy and healthy pregnant women

Variable Age, years, mean  SD Gravid, median (range) Maternal weight, kg, mean  SD Gestational age at enrollment, wk, mean  SD Gestational age at delivery, wk, mean  SD Infant’s birth weight, g, mean  SD Frequency of IUGR (%) Systolic pressure Diastolic pressure

Healthy pregnant women (n [ 10)

Gestational hypertension (n [ 10)

30.2  6.44 1 (1e5) 77.48  18.20 27.36  4.21

32.4  5.83 2 (1e5) 70.97  22.26 32.72  2.63

38.42  1.05 3005.88  581.623 10 110.8  9.1 64.8  23.6

Mild PE (n [ 10)

pa

0.910 0.007a

26.5  5.15 2 (1e3) 83.85  13.39 32.72  3.28

35.71  2.32

0.276

34.55  3.7

2325.00  496.956 10 130.0  13.3 88.0  10.3

0.163 1.0 0.041a 0.104

0.928

Severe PE (n [ 16)

pa

2266.50  831.080 10 141.0  15.9 86.8  30.0

pa

0.915 0.007a

30.33  5.52 2 (1e3) 80.33  16.6 29.31  3.12

0.989 0.636

0.048a

30.35  3.49

0.0001a

0.099 1.0 0.0001a 0.133

1257.94  652.348 31.3 164.3  13.6 101.8  12.23

0.0001a 0.352 0.0001a 0.0001a

0.655

1.0

IUGR, intrauterine growth restriction (45); PE, preeclampsia. a Indicates a significant difference between healthy pregnant and the group marked. Displayed significance set at p !0.05.

Results General Description of the Study Population in the CrossSectional Design Forty six pregnant women at $20 WG age were included in the cross-sectional study as well as ten HnoP women. Ten pregnant women did not meet the criteria for hypertensive disorders of pregnancy in further evaluations and were considered as HP; the remaining women (n 5 36) were diagnosed as having hypertensive disorders of pregnancy [10 with GH, 26 with PE (10 with mPE and 16 with sPE)]. No difference was observed in age between HP and patients with hypertensive disease of pregnancy. Patients with mPE were younger than in HP; however, the difference was not significant ( p 5 0.655) (Table 1). Probably no statistically significant differences could be explained due to the small sample size. Maternal weight was also analyzed and a similar result was observed in all groups. The sPE group had higher blood pressure compared with HP control group [both systolic and diastolic ( p !0.0001)], lower gestational age at delivery ( p !0.0001), delivered infants with lower birth weights

and had a greater proportion of preterm deliveries. The frequency of intrauterine growth restriction (IUGR) (45) was higher in sPE than in HP; however, the difference was not significant (Table 1). VEGF Levels Produced by PBMCs and Serum Levels The production of VEGF by PBMCs was significantly different between groups analyzed ( p !0.0096 by Kruskal-Wallis when comparing all groups). When cells are stimulated with PHA for 5 days under normal culture conditions (without stimuli there is no production of VEGF at 24 h or at 5 days), PBMCs from women with PE produced significantly lower VEGF levels ( p !0.006, Mann-Whitney U test) (median 18 pg/mL [range 0e2316]) (Table 2 and Figure 1 right) compared with levels from HP control group (median, 689.3 pg/mL [range 0e3059]). The diminution of VEGF levels produced by PBMCs was more pronounced ( p !0.003) in sPE (median, 9.5 pg/mL [range 0e526]) (Table 2 and Figure 1 left). VEGF levels produced by PBMCs from HP control group were significantly ( p !0.05, Mann-Whitney U test) higher than levels

Table 2. Supernatant levels of VEGF and sFLT-1 as well as serum levels from pregnant women with hypertensive disorders of pregnancy and control group, median (range)

Levels (pg/mL) Serum VEGF Supernatant VEGF 24 h Supernatant VEGF 5 days Serum sFlt-1 Supernatant sFlt-1 24 h Supernatant sFlt-1 6 days

Non-pregnant women (n [ 10) 58.38 0 2.3 0 2829 8866

(26.21e589.5) (0e28.9) (0e30)a (0e406)a (2098e5328) (6600e25744)

Healthy pregnant women (n [ 10) 22.5 0.47 689.3 3533 5574 15173

(0e144.4) (0e22.7) (0e3059) (1145e7648) (1334e6218) (7938e22136)

Gestational hypertension (n [ 10) 16.7 14 20 6618 2279 15038

(0e464.7) (0e73.2) (0e3573)a (579e25164) (0e11123) (1813e29110)

Mild PE (n [ 10) 14.3 7.5 57 7328 6218 17884

(0e84.9) (0e24.5) (0e2316) (0e46499) (1717e13540) (5741e21524)

PE, preeclampsia. Indicates a significant difference between healthy pregnant and the group marked. Displayed significance set at p !0.05. b p !0.001. a

Severe PE (n [ 16) 29.7 2.84 9.5 20414 2279 11278

(0e211.63) (0e19.87) (0e526)b (16e59 562)a (45e6218) (4834e21524)

VEGF Produced by PBMCs Is Lower in Preeclampsia

565

Figure 1. VEGF levels produced by stimulated cells from patients with hypertensive disease (right) (PE; n 5 26) and according to the different grades of preeclampsia (left); mild preeclampsia (mPE; n 5 10) and severe preeclampsia (sPE; n 5 16).

produced by PBMCs from patients with GH (median 19.6 pg/mL [range 0e3573]) and from ( p !0.004, MannWhitney U test) HnoP control group (median 2.3 pg/mL [range 0e30]). VEGF levels produced by PBMCs from patients with mPE were lower but not significantly (median 57 pg/mL [range 0e2316]). Probably there are no statistically significant differences that could be explained because the sample size is small. Construction of the ROC curve allowed establishing the cut-off value for identification of patients with PE. VEGF production by PBMCs was calculated at 339.87 pg/mL (area under the curve, 0.76; p !0.01), and for sPE (area under the curve, 0.8; p !0.01) with a sensitivity of 93.7% and a specificity of 60% (data not shown), which means that women with sPE have lower VEGF levels produced by PBMCs than the cut-off point. In addition, a robust linear regression model was run to adjust the variance of VEGF levels, finding that variables that predict VEGF levels in supernatants are useful for the diagnosis of hypertensive diseases of pregnancy. VEGF levels decrease 621.6 pg/mL in GH, 948.3 pg/mL in mPE and 1102.4 pg/mL in sPE, and only the last one was significant. Age also decreases these levels and finally adjusted by the WG (Table 3) (R2 was 26.6%). In contrast, no significant differences were detected in serum VEGF levels in women with HP (median 22.5 pg/ mL) compared with a wide spectrum of hypertensive disorder of pregnancy (GH median 16.7 pg/mL, PE median 23.8 pg/mL, mPE median 14.3 pg/mL, sPE median 29.7 pg/mL) (Table 2). These data are in line with previous studies. Serum levels do not show a relation between VEGF levels produced by PBMCs under normal culture conditions or with PHA stimulation for 24 h and for 5 days.

pg/mL [range 7938e22136]) and preeclamptic patients (median 14285 pg/mL [range 4834e21524]) or women with GH (median 15038 pg/mL [range 1813e29110]) as well as 24 h cultured PHA stimulated cells (Table 2). In contrast, in maternal serum sFlt-1 there was significant ( p !0.0001) difference when we compared all groups (by Kruskal-Wallis). Serum levels of sFlt-1 were significantly ( p !0.026 Mann-Whitney U test) higher in patients with PE (median 17320 pg/mL [range 0e59562]) than in HP (median 3533 pg/mL [range 1145e7648]) and were higher ( p !0.005 Mann-Whitney U test) in sPE (median 20414 pg/mL [range 16e59562]). On the other hand, there was no significant difference when we compared serum levels of sFlt-1 in sPE with mPE (median 7328 pg/mL [range 0e46499]) and GH (median 6618 pg/mL [579e25164]) but they were significantly ( p !0.001) higher than HnoP women (median 0 pg/mL [range 0e406]) (Table 2).

sFlt-1 Levels Produced by PBMNCs and Serum Levels

PE, preeclampsia. Note: VEGF levels decrease in HG, in mPE and in sPE, only in this last was significant. Age also decreases these levels and finally was adjusted by the weeks of gestation (WG). Displayed significance set at p !0.05.

There was no significant difference in sFlt-1 levels produced by PBMCs in cultured PHA stimulated or not (for 6 days) cells between control groups, HP (median 15173

Longitudinal Study In the longitudinal study seven of 12 patients with risk factors for PE and/or an abnormal artery Doppler study developed PE and five pregnant women did not develop PE. Patients Table 3. Robust model for adjustment of the variance of supernatant levels of VEGF produced by stimulated PBMCs for 5 days from pregnant hypertensive women Adjusted VEGF Standard in supernatant Coefficient error GH mPE sPE Age WG

621.6 948.3 1102.4 47.5 7.8

607.8 553.4 447.9 32.1 57.5

1.02 1.71 2.46 1.48 0.14

p

95% CI

0.313 1849.9 606.8 0.094 2066.8 170.3 0.018 2007.5 197.2 0.147 112.4 17.5 0.892 108.3 123.9

566

Cardenas-Mondragon et al./ Archives of Medical Research 45 (2014) 561e569

developed PE at 31 WG (median, range 27e38). All patients who developed PE had an abnormal uterine artery Doppler study at 24e26 WG with a score $3, whereas 6/7 patients who developed PE were sPE and only one was mPE. A significant elevation ( p !0.047 Mann-Whitney U test) in VEGF levels produced by PBMCs was found in the HP control group (median 555.6 pg/mL [range 0e1433]) at the first sampling at 24e26 WG, whereas in patients who developed PE no increase in VEGF levels produced by PBMCs (median 6.3 pg/mL [range 0e32]) was seen. In contrast, no significant differences were detected in free VEGF serum levels in women with HP and in patients (Table 4). In the longitudinal study no significant difference was found in increase in sFlt-1 levels in serum as observed in the cross-sectional study. At the first sample at 24e26 weeks, serum sFlt-1 levels for HP control group were 4047.47 pg/mL (median, [range 1553.47e7647.64 pg/ mL]) and for the group who developed PE the levels were 9706.18 pg/mL (median, [range 7e23939 pg/mL]) (Table 4). However, when we analyze according to weeks before the appearance of clinical diagnosis of PE, a significant increase ( p !0.01) was found at 1e3 WG before the onset of PE (18464 pg/mg), serum sFlt-1 levels before 4 weeks of PE and at 1 week before clinical diagnosis of PE were 3918.65 and 3643 pg/mg, respectively, and similar to HP control group. On the other hand, there was no significant difference of sFlt-1 levels produced by PBMCs in culture with or without PHA stimulated for 24 h or 6 days between control groups and women who developed PE. Discussion Accumulated evidence suggests that PE may be the result of an imbalance in angiogenic factors (12,14) causing maternal vascularity damage and endothelial dysfunction. Previous studies in patients with PE reported elevated serum levels of sFlt-1 (10,12,14,16,46e48), Our results

are in line with these studies. Other studies have reported elevated production of sFlt-1 under normal tissue culture conditions by PBMCs (33) or low serum VEGF levels (12,25). In this study we determined that there were significantly lower VEGF levels produced by stimulated PBMCs in cell culture and a significant increase of serum sFlt-1 levels during PE. This decrease in VEGF decrease levels produced by PBMCs and serum elevation of sFlt-1 correlate with disease severity. PE requires knowledge of its etiology and pathogenesis as well as the ability to predict or to identify at an early stage those women who are high-risk. In contrast with Rajakumar (33), we determined that sFlt-1 levels produced by PBMCs do not differ between HP control group and PE patients or women with a wide spectrum of hypertensive disorder of pregnancy, suggesting that circulating sFlt-1 excess does not come from cells. Moreover, sFlt-1 produced by PBMCS does not correlate with their serum levels. Serum sFlt-1 level results are consistent with previous reports (10,12); however, in other studies it is reported that increase in serum sFlt-1 levels is detectable 5e8 weeks before the onset of clinically evident PE (14,46). Our results showed that sFlt-1 serum levels began to increase at 4 weeks before the onset of clinically apparent PE and this increase is greater 1 week before the clinical disease. This result could be important for early prediction of the development of PE. Previous studies reported that serum levels of VEGF in preeclamptic women are low or even undetectable as measured by immunoassay. We found that stimulation with PHA for 5 days of mononuclear cells allows us to observe a difference in VEGF production in groups of women studied and allows observing cells of women with healthy pregnancy if they have the production capacity, whereas PBMCs from PE patients do not have this capacity. VEGF levels produced by the PBMCs from pregnant women in culture open the possibility to use them as biomarkers in order to identify those patients who may be at risk for adverse maternal conditions.

Table 4. Supernatant levels of VEGF and sFLT-1 as well as serum levels from pregnant women who developed PE and control group (median, range)

Levels (pg/mL) WG Serum VEGF Supernatant VEGF 24 h Supernatant VEGF5 days Serum sFlt-1 Supernatant sFlt-1 24 h Supernatant sFlt-1 6 days

Healthy pregnant womena (n [ 5)

Development of PE (n [ 7)

24 (24e26) 22.3 (0e24) 1 (0e22)

26 (22e26) 20.8 (0e211.63) 0 (0e20)

Healthy pregnant women (n [ 5) 28 (28e28) 22.3 (0e43) 0 (0e28.5)

555.6 (0e1433)

6.32 (0e32)b

141.5 (0e822)

4047 (1553e7647) 3009 (1043e6218)

9706 (57e23939) 3543 (856e6218)

4822 (2051e1112) 3277 (45e5574)

11702 (11478e19722) 17350 (1)

15252 (0)

Development of PE (n [ 3)

Healthy pregnant women (n [ 5)

28 (27e29) 19.87 (17e60.96) 18 (0e24)

32 (32e36) 34 (0e69) 0 (0e32)

17 (40e167)

48 (0e2914)

6882 (16e28932) 14706 (5118e25378) 3009 (1622e4998) 2919 (1526e6218) 11414 (1)

16736 (4)

Development of PE (n [ 3) 32 (32e36) 49 (2e137) 17 (0e24) 147.36 (36e2315) 18007 (15280e19870) 2279 (1999e13540) 20138

PE, preeclampsia. Control group. b Indicates a significant difference between healthy pregnant women and the group indicated at similar WG. Displayed significance set at p !0.047. a

VEGF Produced by PBMCs Is Lower in Preeclampsia

Decrease in VEGF levels produced by PBMCS in PE cannot be observed in the VEGF serum levels, probably because a certain number of days of incubation with PHA are needed to observe this phenomenon. The robust model explains a percentage of the variability of VEGF in the supernatant, noting a decreased gradient from GH to sPE. Age also decreases these levels and is finally adjusted by WG, but due to the sample size cannot be adjusted for other variables such as maternal weight, diet or smoking. Our finding has not been reported by any other authors. Abundant evidence suggests that immunological factors are crucial for the development of PE (2,49). Some studies demonstrated that the immunoregulatory system is downregulated in preeclampsia (50,51). Lower VEGF production by PBMCs from PE patients stimulated with PHA could be because these cells (T cells and monocytes) are activated but probably produced by inflammatory molecules instead of VEGF. This could also indicate that cells that produce VEGF in normal pregnancy have immunological activity in culture and probably are T or NK cells (36) or monocytes. Furthermore, to date, the molecular mechanisms of VEGF production are well characterized. Hypoxiainducible transcription factor 1 (HIF-1) is a central mediator of the homeostatic response that enables cells to survive and differentiate under normoxic and low-oxygen conditions. Genetically altered mice have been used to identify important roles for HIF-1 as well as VEGF. HIF1 induces the transcription of O100 enzymes and proteins including VEGF. Recently it was reported that HIF-1 regulates the balance between regulatory T cell (Treg) and Th17 differentiation. HIF-1 induces Th17 differentiation and attenuates Treg development by binding Foxp3 and targeting it for ubiquitination and proteasome degradation (52). Some authors reported a significant diminution in number of Treg cell in women with PE compared to HP women (50,53,54) and an upregulation of the Th17 immune response in PE (51,55). Moreover, the cell expression of free HIF-1a in maternal peripheral blood has been proposed as a marker for PE (56). The results found in this study indicate an altered cell production of VEGF in PE. We need to explore in the future the role of a HIF-1 and mononuclear VEGF cell production (macrophages and T cells, especially pathogenic subtype Th17) in the regulation of the production of VEGF in normal pregnancy and PE (35,50,51,57). VEGF is essential for normal placenta development; VEGF produced by stimulated PBMCs in normal pregnancy may be another extra-placental source of VEGF protein necessary for adequate angiogenesis in pregnancy. Moreover, this decrease in the production of VEGF by stimulated PBMCs in women with PE can also be happening in other nonplacental cells that are producing VEGF. Probably in PE, VEGF-producing cells are inhibited. In contrast to what happens in the tumor microenvironment where the cell-produced VEGF stimulates tumor

567

vasculature development and recruits and interacts with other cells presumably in PE, the microenvironment is inhibitory or downregulated. Reduction of VEGF produced by PBMCs, coupled with the reduction in the production from other sources (placenta or others) of this angiogenic factor and the higher sFlt-1 serum levels, caused an antiangiogenic state. Possibly this is the cause of the extent and intensity of damage to the maternal endothelium in PE. In summary, we present evidence of PBMCs from women with severe PE that are not capable of producing VEGF levels similar to normal pregnancy and higher sFlt-1 serum levels than those present in PE. This study demonstrates that VEGF is involved in the pathophysiology of PE. Molecular mechanisms for the production of VEGF in normal pregnancy and in PE need to be studied. These results will have implications for the correct diagnosis, development, treatment and mechanisms that originate this disease.

Acknowledgments This study was supported by grant FIS/IMSS/PROT/C2007/007 (to M.G.C-M) from the Fondo de Investigaci on en Salud (FIS)IMSS, Mexico. Disclosure statement: The authors have nothing to disclose.

References 1. Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 2000;183:S1eS22. 2. Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet 2005;365: 785e799. 3. Steegers EA, von Dadelszen P, Duvekot JJ, et al. Pre-eclampsia. Lancet 2010;376:631e644. 4. Roberts JM, Cooper DW. Pathogenesis and genetics of preeclampsia. Lancet 2001;357:53e56. 5. Sibai BM. Diagnosis, prevention, and management of eclampsia. Obstet Gynecol 2005;105:402e410. 6. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Number 33, January 2002. Obstet Gynecol 2002;99: 159e167. 7. Kendall RL, Thomas KA. Inhibition of vascular endothelial cell growth factor activity by an endogenously encoded soluble receptor. Proc Natl Acad Sci USA 1993;90:10705e10709. 8. Ferrara N. Role of vascular endothelial growth factor in regulation of physiological angiogenesis. Am J Physiol Cell Physiol 2001;280: C1358eC1366. 9. Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006; 355:992e1005. 10. Koga K, Osuga Y, Yoshino O, et al. Elevated serum soluble vascular endothelial growth factor receptor 1 (sVEGFR-1) levels in women with preeclampsia. J Clin Endocrinol Metab 2003;88:2348e2351. 11. Levine RJ, Karumanchi SA. Circulating angiogenic factors in preeclampsia. Clin Obstet Gynecol 2005;48:372e386. 12. Maynard SE, Min JY, Merchan J, et al. Excess placental soluble fmslike tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649e658.

568

Cardenas-Mondragon et al./ Archives of Medical Research 45 (2014) 561e569

13. Molvarec A, Szarka A, Walentin S, et al. Circulating angiogenic factors determined by electrochemiluminescence immunoassay in relation to the clinical features and laboratory parameters in women with pre-eclampsia. Hypertens Res 2010;33:892e898. 14. Levine RJ, Maynard SE, Qian C, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672e683. 15. Sugimoto H, Hamano Y, Charytan D, et al. Neutralization of circulating vascular endothelial growth factor (VEGF) by anti-VEGF antibodies and soluble VEGF receptor 1 (sFlt-1) induces proteinuria. J Biol Chem 2003;278:12605e12608. 16. Tsatsaris V, Goffin F, Munaut C, et al. Overexpression of the soluble vascular endothelial growth factor receptor in preeclamptic patients: pathophysiological consequences. J Clin Endocrinol Metab 2003;88: 5555e5563. 17. Chaiworapongsa T, Romero R, Espinoza J, et al. Evidence supporting a role for blockade of the vascular endothelial growth factor system in the pathophysiology of preeclampsia. Young Investigator Award. Am J Obstet Gynecol 2004;190:1541e1547. discussion 7e50. 18. Shibata E, Rajakumar A, Powers RW, et al. Soluble fms-like tyrosine kinase 1 is increased in preeclampsia but not in normotensive pregnancies with small-for-gestational-age neonates: relationship to circulating placental growth factor. J Clin Endocrinol Metab 2005; 90:4895e4903. 19. Yancopoulos GD, Davis S, Gale NW, et al. Vascular-specific growth factors and blood vessel formation. Nature 2000;407:242e248. 20. de Vries C, Escobedo JA, Ueno H, et al. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 1992;255: 989e991. 21. Robinson CJ, Stringer SE. The splice variants of vascular endothelial growth factor (VEGF) and their receptors. J Cell Sci 2001;114: 853e865. 22. Terman BI, Dougher-Vermazen M, Carrion ME, et al. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem Biophys Res Commun 1992;187: 1579e1586. 23. Galland F, Karamysheva A, Pebusque MJ, et al. The FLT4 gene encodes a transmembrane tyrosine kinase related to the vascular endothelial growth factor receptor. Oncogene 1993;8:1233e1240. 24. Hornig C, Barleon B, Ahmad S, et al. Release and complex formation of soluble VEGFR-1 from endothelial cells and biological fluids. Lab Invest 2000;80:443e454. 25. Polliotti BM, Fry AG, Saller DN, et al. Second-trimester maternal serum placental growth factor and vascular endothelial growth factor for predicting severe, early-onset preeclampsia. Obstet Gynecol 2003;101:1266e1274. 26. Taylor RN, Grimwood J, Taylor RS, et al. Longitudinal serum concentrations of placental growth factor: evidence for abnormal placental angiogenesis in pathologic pregnancies. Am J Obstet Gynecol 2003;188:177e182. 27. Zhou Y, McMaster M, Woo K, et al. Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol 2002;160:1405e1423. 28. Eremina V, Sood M, Haigh J, et al. Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases. J Clin Invest 2003;111:707e716. 29. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349:427e434. 30. Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol 2003;21:60e65.

31. Kendall RL, Wang G, Thomas KA. Identification of a natural soluble form of the vascular endothelial growth factor receptor, FLT-1, and its heterodimerization with KDR. Biochem Biophys Res Commun 1996;226:324e328. 32. Vuorela P, Lintula S, Stenman UH, et al. Expression of vascular endothelial growth factor in peripheral blood cells of preeclamptic women. Hypertens Pregnancy 2003;22:193e201. 33. Rajakumar A, Michael HM, Rajakumar PA, et al. Extra-placental expression of vascular endothelial growth factor receptor-1, (Flt-1) and soluble Flt-1 (sFlt-1), by peripheral blood mononuclear cells (PBMCs) in normotensive and preeclamptic pregnant women. Placenta 2005;26:563e573. 34. Makris A, Thornton C, Thompson J, et al. Uteroplacental ischemia results in proteinuric hypertension and elevated sFLT-1. Kidney Int 2007;71:977e984. 35. Lin EY, Li JF, Gnatovskiy L, et al. Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 2006;66: 11238e11246. 36. Molvarec A, Ito M, Shima T, et al. Decreased proportion of peripheral blood vascular endothelial growth factor-expressing T and natural killer cells in preeclampsia. Am J Obstet Gynecol 2010;203: 567.e1e567.e8. 37. Hernandez-Andrade E, Brodszki J, Lingman G, et al. Uterine artery score and perinatal outcome. Ultrasound Obstet Gynecol 2002;19: 438e442. 38. Espinoza J, Romero R, Nien JK, et al. Identification of patients at risk for early onset and/or severe preeclampsia with the use of uterine artery Doppler velocimetry and placental growth factor. Am J Obstet Gynecol 2007;196:326.e1e326.e13. 39. Papageorghiou AT, Yu CK, Nicolaides KH. The role of uterine artery Doppler in predicting adverse pregnancy outcome. Best Pract Res Clin Obstet Gynaecol 2004;18:383e396. 40. Plasencia W, Maiz N, Poon L, et al. Uterine artery Doppler at 11 þ 0 to 13 þ 6 weeks and 21 þ 0 to 24 þ 6 weeks in the prediction of pre-eclampsia. Ultrasound Obstet Gynecol 2008;32:138e146. 41. Yu CK, Khouri O, Onwudiwe N, et al. Prediction of pre-eclampsia by uterine artery Doppler imaging: relationship to gestational age at delivery and small-for-gestational age. Ultrasound Obstet Gynecol 2008;31:310e313. 42. Scazzocchio E, Figueras F. Contemporary prediction of preeclampsia. Curr Opin Obstet Gynecol 2011;23:65e71. 43. Leanos-Miranda A, Marquez-Acosta J, Romero-Arauz F, et al. Protein:creatinine ratio in random urine samples is a reliable marker of increased 24-hour protein excretion in hospitalized women with hypertensive disorders of pregnancy. Clin Chem 2007;53: 1623e1628. 44. Leanos-Miranda A, Marquez-Acosta J, Cardenas-Mondragon GM, et al. Urinary prolactin as a reliable marker for preeclampsia, its severity, and the occurrence of adverse pregnancy outcomes. J Clin Endocrinol Metab 2008;93:2492e2499. 45. Avery GB, MacDonald GM, Mullett MD, et al. Growth parameters. Appendix D. Avery’s Neonatology: Pathophysiology & Management of the Newborn. 6th ed Philadelphia: Lippincott Williams & Wilkins; 2005. p. 1659. 46. Chaiworapongsa T, Romero R, Kim YM, et al. Plasma soluble vascular endothelial growth factor receptor-1 concentration is elevated prior to the clinical diagnosis of pre-eclampsia. J Matern Fetal Neonatal Med 2005;17:3e18. 47. Park CW, Park JS, Shim SS, et al. An elevated maternal plasma, but not amniotic fluid, soluble fms-like tyrosine kinase-1 (sFlt-1) at the time of mid-trimester genetic amniocentesis is a risk factor for preeclampsia. Am J Obstet Gynecol 2005;193:984e989. 48. Lim JH, Kim SY, Park SY, et al. Effective prediction of preeclampsia by a combined ratio of angiogenesis-related factors. Obstet Gynecol 2008;111:1403e1409.

VEGF Produced by PBMCs Is Lower in Preeclampsia 49. Gleicher N. Why much of the pathophysiology of preeclampsiaeclampsia must be of an autoimmune nature. Am J Obstet Gynecol 2007;196:5.e1e5.e7. 50. Sasaki Y, Darmochwal-Kolarz D, Suzuki D, et al. Proportion of peripheral blood and decidual CD4(þ) CD25(bright) regulatory T cells in pre-eclampsia. Clin Exp Immunol 2007;149:139e145. 51. Darmochwal-Kolarz D, Kludka-Sternik M, Tabarkiewicz J, et al. The predominance of Th17 lymphocytes and decreased number and function of Treg cells in preeclampsia. J Reprod Immunol 2012;93: 75e81. 52. Dang EV, Barbi J, Yang HY, et al. Control of T(H)17/T(reg) balance by hypoxia-inducible factor 1. Cell 2011;146:772e784. 53. Darmochwal-Kolarz D, Saito S, Rolinski J, et al. Activated T lymphocytes in pre-eclampsia. Am J Reprod Immunol 2007;58:39e45.

569

54. Toldi G, Saito S, Shima T, et al. The frequency of peripheral blood CD4þ CD25high FoxP3þ and CD4þ CD25- FoxP3þ regulatory T cells in normal pregnancy and pre-eclampsia. Am J Reprod Immunol 2012;68:175e180. 55. Toldi G, Rigo J Jr, Stenczer B, et al. Increased prevalence of IL-17producing peripheral blood lymphocytes in pre-eclampsia. Am J Reprod Immunol 2011;66:223e229. 56. Ashur-Fabian O, Yerushalmi GM, Mazaki-Tovi S, et al. Cell free expression of hif1alpha and p21 in maternal peripheral blood as a marker for preeclampsia and fetal growth restriction. PLoS One 2012;7:e37273. 57. Lin EY, Li JF, Bricard G, et al. Vascular endothelial growth factor restores delayed tumor progression in tumors depleted of macrophages. Mol Oncol 2007;1:288e302.