INCREASED MYELOPEROXIDASE IN THE PLACENTA AND CIRCULATION OF WOMEN WITH PREECLAMPSIA

Abstract

Myeloperoxidase (MPO) is a hemoprotein normally released from activated monocytes and neutrophils. Traditionally viewed as a microbicidal enzyme, MPO also induces LDL oxidation, activates metalloproteinases and oxidatively consumes endothelium-derived nitric oxide. Elevated plasma MPO level is a risk factor for myocardial events in patients with coronary artery disease. Patients with preeclampsia display evidence of the inflammation and endothelial dysfunction associated with oxidative stress in the circulation, vasculature and placenta. We hypothesized that MPO levels in the circulation and placental extracts from women with preeclampsia would be greater than levels in women with normal pregnancies. Placental extracts were prepared from placental villous biopsies from preeclamptic (n=27) and control (n=43) placentas. EDTA- plasma samples were obtained from gestationally age matched preeclamptic and control normal pregnancies. MPO concentrations were measured by ELISA. Immunohistochemistry was used to determine MPO localization in placenta. MPO levels in placental extracts from women with preeclampsia were significantly higher than the levels in normal controls (546±62 ng/mL vs 347 ±32 p=0.025). MPO was found in the floating villi and basal plate of placentas with a greater staining in the basal plates from preeclampsia placentas compared to normal pregnancies. Plasma MPO levels were 3-fold higher in patients with preeclampsia compared to normal controls (36.6± 7.6 ng/mL vs 11.0± 3.1 ng/mL p=0.003). In conclusion, MPO levels are significantly increased in the circulation and placenta of women with preeclampsia. We speculate that MPO may contribute to the oxidative damage reported in the endothelium and placenta of women with preeclampsia.

Introduction

Normal human pregnancy is associated with maternal systemic inflammation that is exaggerated in preeclampsia, the pregnancy disorder characterized by hypertension developing after the 20^th week of gestation and proteinuria. Preeclampsia is associated with increased neutrophil counts ¹. Peripheral blood leukocytes are activated in women with normal third trimester pregnancies compared to non-pregnant patients. Leukocytes in normal pregnancy have increased markers of activation (CD11b, CD14, and CD64, intracellular reactive oxygen species) which are further increases in preeclampsia.^{2, 3} Monocytes are also activated in pregnancy as evidenced by elevated surface expression of CD11b, CD14 and CD64 by the third trimester of pregnancy. ² Monocyte activation progressively increase and peak at 29–36 weeks of gestation.⁴ There are several lines of evidence of neutrophil activation during preeclampsia including: 1) greater basal³ and fMLP-induced superoxide production ⁵ 2) increased CD11b expression basally in neutrophils from women with preeclampsia compared to normal pregnancies ^{3, 5} 3) increased neutrophil elastase.⁶

Myeloperoxidase (MPO) is a hemoprotein normally produced and released by activated monocytes and neutrophils. It is traditionally recognized as a microbicidal enzyme primarily through the action of hypochlorus acid. MPO is associated with vascular dysfunction and is mechanistically linked to the pathophysiology of numerous vascular inflammatory diseases including arteriosclerosis and coronary artery disease^{7, 8}. Elevated circulating MPO levels are a risk factor of myocardial events in patients with coronary artery disease.^{9, 10} There is a strong relationship between serum MPO levels and endothelial dysfunction in humans.¹¹ Patients with the highest quartile of serum MPO were more than six times as likely to have endothelial dysfunction assessed by brachial artery flow-mediated dilation than patients in the lowest quartile.¹¹

MPO has been shown to mediate oxidation of lipoproteins, catalyze nitration of tyrosine residues and has been implicated in depletion of endothelial derive nitric oxide ^12–16. Once in the circulation, MPO can become sequestered in the subendothelial space through a process of heparin glycosaminoglycan dependent binding and endothelial transcytosis, resulting in its accumulation in the endothelial cell matrix ¹².There it can be a potent source of reactive oxygen and nitrogen species and consume antioxidants ^{12, 17, 18}. Activated MPO also catalyzes oxidation of nitrite and nitric oxide ultimately causing nitration of protein tyrosines ^12–15. Likewise, increased nitrotyrosine in blood vessels of the placenta and peripheral circulation of the mother ^{19, 20} is suggestive of MPO over-activity in the vasculature during PE. There are reports of increased or unchanged myeloperoxidase in the circulation of women with PE ²¹. Women with a history of preeclampsia are at increased risk for poor cardiovascular health later in life²². These observations prompted us to examine MPO levels in the placenta and circulation of women with preeclampsia.

Methods

Preeclampsia was defined using the criteria of gestational hypertension, proteinuria and hyperuricemia, and reversal of hypertension and proteinuria after delivery. Gestational hypertension was defined as systolic blood pressure > 140 mmHg systolic or diastolic blood pressure > 90 mmHg arising after 20 weeks' gestation in a previously normotensive woman. Proteinuria was defined as > 300 mg of protein in a 24-hour urine collection or > 2 + on a voided or > 1 + on a catheterized random urine sample, or a random urine protein/creatinine ratio of > 0.3. Hyperuricemia was defined as > 1 standard deviation above normal for the given gestational age [at term > 5.5 mg/dL (3.3 mM)]. Control groups were composed of women with uncomplicated, normotensive pregnancies and who delivered healthy, non-SGA babies. Patients with multiple gestations, chronic hypertension, diabetes, renal disease or other significant metabolic disorder, or a history of illicit drug use were excluded. The institutional review board approved the study and written informed consent was obtained. Procedures followed were in accordance with institutional guidelines.

Placental biopsies

were obtained immediately after Cesarean delivery from a site between the placental rim and cord insertion. Clinical data for these patients are provided in Table 1. The tissue was quickly washed three times in physiological saline, frozen in liquid nitrogen and stored at −70°C until use. Placental extracts were prepared from ~100 mg wet weight placental biopsy samples. Total proteins from placental tissues were extracted by homogenizing by sonication (Ultrasonic Processor, Tekmar, Cincinnati, OH, microprobe setting 70 for 30 seconds) in 4 volumes of 1X Laemmli buffer (50 mM Tris HCl, pH 6.8, 2% SDS, 10% Glycerol) containing 5 mM DTT, 0.5 mM PMSF, 1 mM sodium vanadate and 1.0 µl/ml of protease inhibitors cocktail.²³ Protein was determined by the Biorad protein assay. Samples were diluted into sample diluting buffer (20 mM phosphate buffer, pH 7.4 containing 2 mg/mL bovine serum albumin, 0.1% tween 20 and 0.2% sodium azide) for the MPO ELISA and MPO levels were as described below.

Table 1.

Patient data for placental samples

Characteristic (Placental Extracts)	Preeclampsia (n= 27)	Normal Pregnant (n= 43)
Age (years)	28.4±1.8*	23.1±1.1
Prepregnancy BMI (kg/m2)	25.5±1.3*	23.1±1.1
Weeks gestation at delivery	36.42±0.64	38.8±0.3
Blood pressure (mmHg) <20 weeks gestation
Systolic	118 ± 2	112 ± 2
Diastolic	71 ± 2	68 ± 1
Pre-delivery blood pressure (mmHg)
Systolic	159 ±7*	118 ± 3
Diastolic	96 ± 3*	72 ±1

^*p< 0.05 Preeclampsia vs. Normal Pregnancy

EDTA- plasma samples

were obtained from normal weight preeclamptic (n=11, 35.5 weeks gestation) and gestationally-age matched control “normal” pregnancies (n=13, 36 weeks gestation) with no sign of infection but delivering early, and a second set of controls (n=17) with two blood samples one during gestation (32.1 weeks) and a second near delivery (38 weeks) were used for comparison of MPO levels in the circulation during pregnancy and preeclampsia. Clinical data for these patients are provided in Table 2. MPO concentrations were measured by ELISA. The accuracy of measuring MPO in previously frozen EDTA plasma was confirmed using MPO ELISA from 2 companies (Calbiochem and Assay Designs). Our preliminary experiments indicated that EDTA plasma yielded comparable results to heparinized plasma or spiked plasma using fresh or 1 freeze-thaw plasma. The Calbiochem ELISA standard curve was linear from 1 to 25 ng of MPO per milliliter. The sensitivity of the assay was 1.5 ng/mL with interassay and intraassay variation of 8%. The Assay Designs (ELISA) standard curve was linear from 0.39 to 12 ng of MPO per milliliter with a sensitivity of 0.13 ng/mL with interassay variation of 7% and intra-assay variation of 5%.

Table 2.

Patient data for plasma samples

Characteristic (Plasma)	Preeclampsia (n=11)	Normal Pregnant (n=30)
Age (years)	22±2	27±8
Prepregnancy BMI (kg/m2)	25±1	23±8
Weeks gestation at blood collection	35±1	36±2
Weeks gestation at delivery	36±1*	37±1
Blood pressure (mmHg) <20 weeks gestation
Systolic	118±3	108±4
Diastolic	72±3	77±4
Pre-delivery blood pressure (mmHg) gestation
Systolic	165±4*	122±6
Diastolic	95±3*	84±5

^*p< 0.05 Preeclampsia vs. Normal Pregnancy

Immunohistochemistry

was used to determine MPO localization in placenta from control normal pregnancies (n=7) and preeclamptic pregnancies (n=5) using previously published protocols. Patient data for these samples is provided in Table 3 with the exception of early pregnancy samples from pregnancy terminations 7.5 and 24 weeks (n=2 at each time point). These patients had no complications, including normal blood pressures. Briefly, placental samples were fixed 3% paraformaldehyde for 30 minutes followed by 5, 10, and 15% sucrose PBS then embedded in OCT. Samples were stored at −80°C until sectioned. Anti-myeloperoxidase antibody (Rabbit polyclonal, Calbiochem, La Jolla, CA) at a 1:200 dilution was used to localize MPO ^{24, 25,26}. The primary antibody was omitted as a negative control. Gestational increases in MPO localization in the floating villi from normal pregnancy placenta was used as a positive control ²⁷. Cytokeratin was stained using at 1:200 antibody dilution (rat monoclonal IgG) ²⁶.

Table 3.

Patient data for placental immunohistochemistry

Characteristic (Immunostaining)	Preeclampsia (n=5)	Normal Pregnant (n=3)
Age (years)	28.5±5.5	28.2±6.8
Prepregnancy BMI (kg/m2)	25.8±1.4	23.4±1.2
Weeks gestation at delivery	29.8±3.1*	37.3±0.5
Blood pressure (mmHg) <20 weeks gestation
Systolic	117.7±2.2	111.8±2.1
Diastolic	70.5±2.3	69.1±1.7
Pre-delivery blood pressure (mmHg)
Systolic	160±8.1*	118±4
Diastolic	97.2±4.2*	72±2

^*p< 0.05 Preeclampsia vs. Normal Pregnancy

Heparin treatment of patients

16 women (age 25± 1.4 years, 365± 38 postpartum) and previous normal pregnancies at a mean of 1 year ± 38 days and 15 women with prior preeclamptic pregnancies (age 29±4 years, 388± 30 days postpartum) with normal blood pressures and BMI were recruited for treatment with heparin. All women fasted overnight and initial blood specimens were collected via 18 ga. intravenous line; this is the “pre-heparin” sample. If hematocrit and platelets were normal (hct>35%; platelets > 100,000), women received injection of 75 IU heparin/kg body weight intravenously. After 15 minutes, 20 cc of venous blood is collected as the “post-heparin” sample. Samples are stored at −70°C until use.

Statistical comparisons were by Students t-test, One or two-way ANOVA or Dunn’s test as required and mean± standard errors are reported.

Results

MPO in Placental Extracts

Placental extracts were prepared from preeclamptic (n=27, 36.4 weeks gestation) and control (n=43, 39.1 weeks gestation) placentas. Clinical data for these patients is provided in Table 1. MPO concentrations in placental extracts from women with preeclampsia were significantly higher than in normal controls 546±62 vs 347±32 ng/mL (p=0.025). Based upon previous reports that MPO levels in the placenta increase slowly with gestation, we limited our normal pregnancy group to match for gestational age of the sample (n=16, gestational age 37.5± 0.3), and again found MPO concentrations in normal pregnancy to be much lower (286.5±66 ng/mL, p=0.009).

Placental Immunolocalization of MPO

The localization of MPO within the placentas from normal and preeclamptic pregnancies was determined by immunohistochemical analysis. High levels of MPO were found in basal plates from 4 of 5 preeclampsia placentas compared to normal pregnancy placentas where high levels were found 1 in 7 of placenta, representative slides are shown in Figure 1B (Panel D vs Panel B). In the basal plate of women, degenerated cytotrophoblasts were MPO positive; however in PE basal plate there was also strong staining of cytotrophoblasts and some as yet unidentified maternal cells. In the floating villi, MPO levels increased through gestation in normal pregnancies and this increase was then compared to MPO in placentas from patients with preeclampsia. Two 3^rd trimester preeclampsia placentas had MPO staining greater than normal 3^rd trimester placenta and the remaining 3 preeclampsia placentas had MPO staining greater than normal term (Figure 1C, panels D vs B). In the floating villi, MPO staining increased through gestation in placenta from normal pregnancies consistent with previous work ²⁷. This is shown in a representative placental sample at 7.5 weeks gestation (Figure 1C panel B ) in comparison to 37 weeks (Figure 1C panel D). Figure 1C (panel F) demonstrates the increased MPO staining intensity in floating villi from a preeclampsia placenta delivering at 26 weeks gestation. Cell types in the villi staining positive for MPO were infiltrating white blood cells (cytokeratin -negative, CD45 positive, MPO positive cells, shown in the inset of Figure 1C panel D) and syncytiotrophoblast (cytokeratin and MPO positive cells). The positive staining of syncytiotrophoblasts was much greater in villi from preeclampsia placenta.

Figure 1. 1A. Myeloperoxidase Levels in Placental Extracts.

Myeloperoxidase (MPO) levels were determined in placental extracts from patients with preeclampsia (n=27), and normal pregnancies (n=43). MPO levels were significantly greater in placental extracts from preeclampsia (546±62 ng/mL) vs. normal controls with (287±66 ng/mL) or without (347±32 ng/mL) controlling for gestational age. Expressing MPO levels as ng/mL extract or ng/mg protein did not change these results. Myeloperoxidase was examined immunohistochemically in basal plate (Figure 1B) and floating villi (Figure 1C) from placental samples. In Figure 1B, basal plate from normal pregnancy placenta at 37 weeks and from preeclampsia placenta (26 weeks) is shown with cytokeratin staining in panels A and C. Myeloperoxidase immunolocalization in normal pregnancy is shown in panel B, and in preeclampsia (Panel D). There was much greater myeloperoxidase staining was in the basal plate of preeclampsia compared to normal pregnancy placenta. In the floating villi (Figure 1C), panels A,C,E are cytokeratin staining of samples from 7.5 week normal pregnancy (panel A), 37 week normal pregnancy (panel C) and preeclampsia placentas(panel E). In floating villi from normal pregnancies, myeloperoxidase staining increased with gestation from 7.5 weeks (panel B) to 37 weeks (panel D), while the preeclampsia placental sample had the highest levels of staining (panel F). MPO positive cells were identified as primarily infiltrating white blood cells (cytokeratin negative cells, shown in the inset of panel D) in normal pregnancy and white blood cells and syncytiotrophoblast (cytokeratin and MPO positive cells) in preeclampsia. Magnification is 100X with a scale of size as indicated on the figures.

MPO in Pregnancy Plasma

EDTA- plasma samples from gestational age matched preeclamptic (n=11, 35.5 weeks gestation) and control normal pregnancies (n=13, 36.1 weeks gestation) were obtained for determination of plasma MPO. Clinical data is provided in Table 2. Plasma MPO levels were 3-fold higher in patients with preeclampsia compared to normal controls (36.6± 7.6 ng/mL vs 11.0± 3.1 ng/mL p=0.003) (Figure 2A) by ELISA (Calbiochem). Controlling for gestational age or protein levels did not influence these results. A second set of control patients with two samples collected during gestation (20–36 weeks) and at term were identified and used to determine if MPO levels in plasma changed with normal gestation. These samples were again compared to samples from patients with preeclampsia using a second ELISA assay system (Assay Designs). Plasma MPO levels were not significantly different earlier in gestation compared to term in normal pregnancies (2.0 ± 0.8 vs 5.0±1.4 ng/mL), while MPO in preeclampsia plasma was significantly elevated (32.5±11 ng/mL). We did not find any trend of increasing plasma MPO levels during gestation to term in the normal pregnancies. In normal pregnancies, 36% of patients had no change in MPO with resampling, 27% of patients had increased MPO levels and 37% had decreased MPO levels with the second measurement. MPO concentrations measured by ELISA for preeclampsia patients were not significantly different between assays.

Figure 2. Myeloperoxidase Levels in Plasma.

A.MPO levels were significantly elevated in plasma of patients with preeclampsia (36.4 ± 8 ng/mL, n=11) compared to gestationally age matched normal pregnant controls (11±3 ng/mL, n=13 , p=0.003) using an ELISA. B. A second ELISA (Assay Designs) was used to confirm these results in the preeclampsia samples and a second group of normal pregnant controls late in gestation (n=17), and a set of samples gestationally age matched controls not near delivery (n=17). Again preeclampsia plasma samples had significantly more MPO than controls when matched for gestational age 32.5±11 vs. 2.0± 0.8 ng/mL, late in gestation samples were also lower than preeclampsia (5.0 ±1.4 ng/mL). Statistical analysis by Dunn’s method.

MPO in Postpartum Plasma

To determine the potential of MPO to bind to the vasculature of healthy reproductive age women, we examined MPO levels in plasma from women post-pregnancy before and after a heparin infusion (clinical data provided in Table 4), which would release heparin sulfate bound MPO into the circulation. Heparin infusion resulted in a greater than 5-fold increase in plasma MPO levels in the circulation 18.5± 4 to 103.7± 8 (n=16, p<0.001, Figure 3A) in women with normal pregnancies and 20.1± 5 to 114±10.3 in women with prior preeclampsia (n=15, p<0.001, Figure 3B). There was no significant difference in plasma MPO levels or the heparin releasable levels in women with prior preeclamptic pregnancies compared to women with prior normal pregnancies at approximately one year postpartum.

Table 4.

Heparin Treated Post-Partum Patients

	Previous Preeclampsia (n=15)	Previous Normal Pregnancy (n=16)
Age (years)	29.4±1	25.2±1.4
BMI	29.9±1.7	27.9±1.6
Blood Pressure (mmHg)
Systolic	119±2.6*	109±2.0
Diastolic	81±2.5*	73±2.2
Time Since Delivery	388.6±29.5	365.3±37.7
(days)

^*p< 0.05 Preeclampsia vs. Normal Pregnancy

Figure 3. Releasable Myeloperoxidase in Healthy Women with Previous Normal Pregnancies or Preeclampsia.

A.Heparin infusion of post-partum women with prior normal pregnancies demonstrated a significant increase in plasma MPO levels after heparin treatment. B. MPO release also resulted in a significant increase in MPO in the circulation of post-partum women with prior preeclampsia. There was no significant difference in plasma MPO levels initially, or after heparin treatment between women with prior normal pregnancies or preeclampsia at greater than one-year post-partum.

Discussion

MPO levels were significantly increased in placental extracts from women with preeclampsia. Placental MPO levels have been shown to increase with gestational age in placenta for normal pregnancies. ²⁷ MPO levels were significantly elevated when compared to gestationally-age matched placental samples without preeclampsia, or normal term placental samples. The normal pregnancy data was limited to a sample set with matching gestational age to the preeclampsia group, resulting in a group of patients with early deliveries but without any evidence of pregnancies complications related to infection. This is consistent with the low concentrations of MPO measured in these patients. The high levels of MPO in the placental extracts were confirmed immunohistochemically, however the samples used for the extract preparation did not include the basal plate, which was shown to have the most dramatic difference in MPO localization in the placental sections analyzed.

The previous data on circulating MPO levels in patients with preeclampsia was conflicting²⁸, our data agrees with studies reporting increased MPO in the circulation of women with preeclampsia.^{1, 29} Unlike the placenta, circulating MPO levels did not significantly increase with gestation. The most plausible source of MPO in the circulation is activated inflammatory cells either in the circulation or tissue. We speculate that MPO may contribute to the oxidative damage reported in the endothelium of women with preeclampsia.

Once in the circulation, MPO can become sequestered in the subendothelial space through a process of heparin glycosaminoglycan dependent binding and endothelial transcytosis, resulting in the accumulation of MPO in the endothelial cell matrix ¹². There it can be a potent source of reactive oxygen and nitrogen species. Functional significance of the vascular associated MPO has been shown in patients undergoing coronary angiography for treatment of coronary artery disease. Baldus et al found no significant difference in baseline plasma MPO levels, but upon heparin infusion, circulating MPO levels were significantly higher in patients with CAD. ³⁰ Heparin infusion was also associated with improved nitric oxide mediated endothelial function using flow-mediated dilation and responsiveness to acetylcholine. The heparin infusion was associated with a decrease in MPO burden in the vasculature of these patients ³⁰. We sought to determine the levels of heparin released in young women to assess the potential relevance of this mechanism in healthy patients, and to also validate the determination of MPO in EDTA-plasma samples from healthy patients. Our data also showed that by one-year post-partum, there was no significant difference in heparin releasable MPO in previously preeclamptic women compared to women who had a normal pregnancy.

Prospective

Elevations in MPO have not only been associated with cardiovascular events, but have also been shown to have predictive power in high-risk patients.⁹ Women with preeclampsia have numerous symptoms (hypertension, inflammation, and endothelial dysfunction) during pregnancy that can be associated with cardiovascular risk factors, elevations MPO in these patients may not only index the cardiovascular involvement of this disorder but may also play a causative role. Further evaluation of the time course for elevations in MPO relative to hypertension in these patients should provide insight into the predictive power of MPO in pregnant women, as well as if MPO may contribute mechanistically to the vascular dysfunction and hypertension in preeclampsia. These post-partum data indicate that there is no long-term mechanism of inflammation present in patients with preeclampsia that causes higher MPO levels, compared to women with previous normal pregnancies, which remains in the post-partum period.