Placenta
Volume 31, Issue 5 , Pages 392-400, May 2010

Effects of Labor on Placental Expression of Superoxide Dismutases in Preeclampsia

  • L. Roland

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • ,
  • D. Beauchemin

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • ,
  • G. Acteau

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • ,
  • C. Fradette

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • ,
  • I. St-Pierre

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • ,
  • J.-F. Bilodeau

      Affiliations

    • Axe reproduction, santé périnatale et santé de l'enfant, Centre de Recherche en Biologie de la Reproduction (CRBR), Centre de Recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL), Québec, Canada
    • Département d'Obstétrique et Gynécologie, Faculté de médecine, Université Laval, Québec, Canada
    • Corresponding Author InformationCorresponding author. Reproduction, perinatal health, and child health, CHUQ Research Center, Room T-1-49, 2705 Boulevard, Laurier, Québec, PQ, G1V 4G2, Canada. Tél.: +1 418 525 4444x46153; fax: +1 418 654 2765.

Accepted 9 February 2010. published online 15 March 2010.

Article Outline

Abstract 

A decreased antioxidant activity for superoxide dismutases (SODs) in the placenta was reported in preeclampsia (PE). However, it is unclear if this reduced enzymatic activity can be attributed to a specific SOD isoform. Moreover, the specific spatial SOD expression in the placenta and the impact of the mode of delivery on the latter are still lacking. There are three known SOD isoforms: SOD1 (cytosolic), SOD2 (mitochondrial) and SOD3 (extracellular). Our main objective was to characterize by RT-PCR, western blot and immunolocalization, the expression of SOD1, SOD2, and SOD3 in placentas of normotensive (n = 23) and PE pregnancies (n = 25) according to the presence or absence of labor, the sampling site (peri-insertion, mid-disc and periphery) and the placental layer: amnion-chorion, villi, and maternal side layer (MS). In absence of labor (cesarean), SOD1 expression in the placental villi and MS was lower in PE than in controls (p < 0.049). In presence of labor (vaginal deliveries), SOD1 expression in the amnion-chorion only was higher in PE than controls (p = 0.014). Additionally, SOD2 and SOD3 expression in presence of labor were higher in all three layers in PE than controls, with a strong positive correlation between these two SODs (mRNA; r > 0.65, p < 0.008). The sampling site and gestational age had no effect on SOD expression within the placenta. In this study, we showed that the reported decrease for SOD activity in PE may be attributed to SOD1 in absence of labor. Also, this is the first study characterizing specific SOD isoforms according to the mode of delivery. We demonstrated in PE that labor upregulates SOD1 in fetal membranes as well as SOD2 and SOD3 in the whole placenta.

Keywords: Oxidative stress, Amnion, Superoxide dismutase, Preeclampsia, Effect of labor, Human, Chorion, Villi

 

Back to Article Outline

1. Introduction 

Placental oxidative stress has been clearly associated with the pathogenesis of preeclampsia (PE) [1]. An oxidative stress is defined as an imbalance between the generation of reactive oxygen species (ROS) and the antioxidants that prevent oxidative damages [1]. Among ROS, the superoxide anion (O2−) can disturb vascular integrity and endothelial cell functions [2].

Superoxide dismutases (SODs) are metalloenzymes responsible for dismuting two molecules of O2− to produce hydrogen peroxide (H2O2) and molecular oxygen. Three isoforms of SOD can be found in mammalian tissues. All SODs contain redox active metal ions in their catalytic site, but they differ considerably in their primary structure [3] and amino acid sequence [4]. The SOD1 also called Cu,Zn-SOD, is the major cytosolic form. The SOD2 or Mn-SOD is specifically expressed in the mitochondria. The SOD3 or extracellular SOD (EC-SOD) contains Cu and Zn atoms like SOD1 [5], [6]. Unlike SOD1 and SOD2, the expression of SOD3 mRNA and protein is tissue and cell specific. Indeed, SOD3 is not only found in the placenta but also in the heart, lungs (lung macrophages and alveolar type II cells), blood vessels (vascular smooth muscle cells and endothelial cells), and kidneys (proximal to renal tubular cells) [6].

These past decades, placental activities of various endogenous antioxidants including SOD have extensively been studied. Previous studies have demonstrated that SOD placental activity and protein expression increased with gestation from 8 to 20 weeks [7]. Furthermore, most of the reports linking the level of SOD to preeclampsia, revealed a decreased enzymatic activity in preeclamptic placentas [8], [9], [10], [11] and circulation [12], [13], [14]. However, very few studies have examined these antioxidants on the gene expression level [8], [10]. Wang et al. (1996) were the first to demonstrate that SOD1 relative mRNA expression was significantly lower in preeclamptic placentas compared to the controls [8]. Nevertheless, the specific protein level of SOD1 and the expression of the other two SOD isoforms were not investigated in the latter report. Boggess et al. (1998) examined the placental distribution and activity of SOD3 and concluded that there was no difference between preeclamptic and normal placentas [15]. To our knowledge, SOD2 and SOD3 mRNA expression and protein levels in the preeclamptic placenta have never been studied. Though, it is still unclear if the reduced total activity previously reported by other studies is related to a specific decrease in mRNA expression in one of the SODs.

In our study, we hypothesized that the placenta would respond differentially to the oxidative stress in PE according to the labor and the spatial location of the biopsy. Specifically, we studied the mRNA expression, the protein level and the immunolocalization of antioxidants SOD1, SOD2 and SOD3 in normotensive and preeclamptic placentas obtained after vaginal deliveries (with labor) and cesarean (no labor). We controlled for labor, because it was shown that uterine contractions during deliveries are associated with a discontinuous utero-placental perfusion leading to an oxidative stress [16]. Finally, we also controlled for the sampling site, because it has been demonstrated that the activity of certain antioxidant enzymes like catalase varied within the placenta [17].

Back to Article Outline

2. Methods 

2.1. Patients recruitment 

The following study was approved by the local ethic committee and a written informed consent was obtained from all subjects. Pregnant women were recruited at the Centre Mère-Enfant du Centre Hospitalier Universitaire de Québec in accordance with institutional guidelines. Patients with normal blood pressure (≤140/90 mm Hg), absence of proteinuria, and no medical complication were considered normotensive pregnancies. Characteristics for preeclampsia, also called gestational hypertension with proteinuria, included a diastolic blood pressure of 90 mm Hg or more on two separate readings at least 4 h apart, with proteinuria of ≥0.3 g/d (+1 or greater on dipsticks) in accordance with the Canadian Hypertension Society Consensus [18]. Exclusion factors were extreme age (<18 years old or >40 years old), the intake of anticoagulant drugs or drugs affecting lipid metabolism, obesity (BMI>30 prior to pregnancy), patients with any pre-existing medical conditions (chronic hypertension, diabetes mellitus, kidney diseases, inflammatory intestinal diseases and blood clotting disorders) or any obstetrical complications, such as gestational diabetes. No eclampsia was observed in our patients. Table 1 sums up the clinical features of the two study groups. Small for gestational age (SGA) is defined by a birth weight below the 10th percentile for that gestational age and sex [19].

Table 1. Clinical features of recruited patients.
Control groupPreeclamptic group
No labor (cesarean)Labor (vaginal deliveries)No labor (cesarean)Labor (vaginal deliveries)
n =9141411
Maternal age (years)32.4 ± 1.529.9 ± 0.727.2 ± 1.4a31.7 ± 2.0
Gestational age at sampling (weeks)38.41 ± 0.2240.53 ± 0.1434.51 ± 0.94a37.75 ± 0.42a
Systolic blood pressure (mmHg)122 ± 2120 ± 2164 ± 5a161 ± 5a
Diastolic blood pressure (mmHg)73 ± 372 ± 2102 ± 3a99 ± 3a
Proteinuria (g/24 h)< 0.3< 0.33.93 ± 1.26a0.51 ± 0.04a
Placental weight (g)471 ± 25499 ± 34287 ± 27a451 ± 32
Birth weight (g)3306 ± 1983552 ± 1251993 ± 218a2902 ± 168a
Small for gestational age (SGA)0064

Values are means ± SEM. Statistical analysis was performed using the Student's t-test. A p value < 0.05 was considered significant.

aStatistically different from the corresponding control group (cesarean or vaginal deliveries).

2.2. Collection of placental samples 

Placental samples were collected within 20 min following delivery. The placenta is a heterogeneous organ composed by fetal membranes, villi and decidua. Bacon et al. (1986) demonstrated that sampling from a particular region will represent the whole organ only in the central region of the parabasal plate in normal term placentas [20]. So it is unreliable to assume that only one region would be representative of the pathological placenta. Consequently, to reduce sampling bias [21], [22], we constructed a circular template with a square grid of 16 holes in a 4 × 4 design to superimpose the placenta. We prepared several templates of various sizes to best fit placentas of different diameters. Indeed, the average transverse diameter expected for the control group was 18.2 ± 1.3 cm and for the PE group it was 14.0 ± 2.6 cm, (mean ± SD) [23]. Nine samples of tissues were systematically taken at three sampling sites on the horizontal plane of each placenta: peri-insertion (within 2 cm from the insertion of the umbilical cord), mid-disc (at midway between the umbilical cord insertion and the periphery on the longest diameter of the placenta) and periphery (at 2 cm of the placental margin). Three full thickness biopsies (1 cm3) were obtained by manual dissection and the samples were subsequently separated according to the placental layer: amnion-chorion (AC) membranes, villi (V) and on the maternal side (MS). The MS samples contained stem villi and a portion of decidua. After sampling, the biopsies were immediately washed in a solution of PBS 1× and snap frozen at −80 °C.

2.3. Quantitative RT-PCR 

Total RNA was extracted from the samples using TRIzol®, following the manufacturer's instructions (Invitrogen, Burlington, ON, Canada). Four micrograms of RNA were reverse transcribed using the Superscript II reverse transcriptase and random hexamer primers (Invitrogen) following the manufacturer's instructions. The cDNA at a concentration of 20 ng/μL was then used as a template in the quantitative real-time PCR reaction mixture.

Specific primers for human SOD1 (forward: 5′-tca-gga-gac-cat-tgc-atc-att-3′; reverse: 5′-cgc-ttt-cct-gtc-ttt-gta-ctt-tct-tc-3′) [24], SOD2 (forward: 5′-acc-tca-gcc-cta-acg-gtg-gt-3′; reverse: 5′-agc-cgt-cag-ctt-ctc-ctt-aaa-ct-3′) [24], and SOD3 (forward: 5′-tgc-tcc-aac-aga-cac-ctt-cca-c-3′; reverse: 5′-ggt-aca-aat-gga-ggc-ctt-cag-ac-3′) [25] were designed based on known sequences as described previously [26]. Based on the findings by Meller and al. [27], the three following housekeeping genes were used as internal controls in all PCR experiments: YWHAZ (tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide) (forward: 5′-ccg-cca-gga-caa-acc-agt-at-3′; reverse: 5′-act-ttt-ggt-aca-ttg-tgg-ctt-caa-3′), SDHA (Homo sapiens succinate dehydrogenase complex, subunit A) (forward: 5′-cgg-tcc-atg-act-ctg-gag-at-3′; reverse: 5′-agg-acc-tgc-ccc-ttg-tag-tt-3′), and hTBP (Homo sapiens TATA box binding protein) (forward: 5′-gaa-cca-cgg-cac-tga-ttt-tc-3′; reverse: 5′-ccc-cac-cat-gtt-ctg-aat-ct-3′). The PCR products were separated on a 1.2% agarose gel, extracted from the gel using the QIAquick Gel Extraction Kit from Qiagen (Mississauga, ON), and then sequenced. These specific PCR products were serially diluted from 0.5 ng/μL to 5.0 × 10−10 ng/μL to establish the standard curves. The quantitative RT-PCR reactions were carried out in a LightCycler (Roche Diagnostics, Laval, Québec). The 20 μL reaction mixture was composed of 5 μL cDNA or PCR product, 0.58 μM of each primer, 3 mM of MgCl2, 2 μL of FastStart Master SYBRGreen I mix (Roche Diagnostics) and PCR-grade water up to the final volume. The RT-PCR reactions consisted in 3 steps. The denaturation at 95 °C for 10 min was immediately followed by a 45 cycles amplification (95 °C for 0 s, annealing temperature for 5 s and 72 °C for 20 s) with a single acquisition of fluorescence at the end of the extension step. The annealing temperature was set up for each gene based on the primers (SOD1: 62 °C, SOD2: 64 °C, SOD3: 64 °C, YWHAZ: 63 °C, SDHA: 64 °C and hTBP: 62 °C). At the end of 40 amplification cycles, a melting curve was built. For that purpose, the samples were gradually heated (0.1 °C/sec) from 60 °C to 95 °C with stepped acquisition of fluorescence. Only one peak was observed in the melting curve. The specificity of each amplicon was further confirmed by agarose gel electrophoresis and sequencing (Genotyping and Sequencing platform, CHUL Research Centre, Québec, PQ, Canada). The quantification analysis of the data was performed using the LightCycler analysis software as previously described [26]. Each cDNA was run in duplicates. The GeNorm software version 3.5 was used for the calculation of the housekeeping genes stability measure, M as previsously described by Meller et al. [27]. Briefly, the principle underlying the GeNorm algorithm is that the ratio between constantly expressed and non-normalized housekeeping genes should remain stable. The “M” is the average pair-wise variation of housekeeping genes and it is calculated from the raw expression data. The most stable gene has the lowest M value and consequently, the gene with the highest M value (less stable) is removed at each step until the remaining genes could not be further compared. The GeNorm program can be found on the following website: http://medgen.ugent.be/∼jvdesomp/genorm/. The relative mRNA expression was expressed as the ratio of target genes to M values.

2.4. Western blot analysis 

We analyzed randomly sixteen samples from the two groups (control and PE). In each group, we selected eight samples for vaginal deliveries and cesarean. A hundred milligrams of each placental sample were mixed in ice-cold PBS buffer containing a protease inhibitor cocktail (1 mM EDTA, 120 μg/mL Leupeptin, 85 μg/mL PMSF) and homogenized with an Ultra Turrax T25 (IKA Works, Inc., Wilmington, NC, USA). Laemmli sample buffer (5% (v:v) 2-mercaptoethanol) was added to the homogenates and they were boiled for 10 min [28]. The quantity of proteins was determined using a BCA (bicinchoninic acid) protein assay, according to the manufacturer's instructions (Pierce, Rockford, IL, USA).

An aliquot (50 μg) of proteins from each placental section was loaded on a 12% polyacrylamide-SDS gel [28]. Proteins from the three different placental layers were analyzed on distinct gels. After electrophoresis, the proteins were transferred to a nitrocellulose membrane (0.2 μm; Bio-Rad Laboratories, Montréal, QC, Canada). The membranes were incubated for 1 h at room temperature with 2% ECL advance blocking agent (GE Healthcare Bio-Sciences, Baie d'Urfé, QC, Canada) in TBS-Tween 0.1%, then hybridized for 1 h at room temperature with primary rabbit polyclonal antibodies against SOD1 (1:10 000, Novus Biologicals, Littleton, CO, USA), SOD2 (1:5000; GeneTex, San Antonio, TX, USA), SOD3 (1:2500; Stressgen, Victoria, BC, Canada) or a mouse anti-β-actin antibody (1:10 000; Sigma–Aldrich, Oakville, ON, Canada). The signal was revealed on film using ECL advance (GE Healthcare Bio-Sciences, Baie d'Urfé QC, Canada). Western blotting with specific antibodies recognized a 23-kDa SOD1, a 25-kDa SOD2 and a 35-kDa SOD3 in placental homogenates. A densitometry analysis was performed on the signal obtained on film using an Alpha Imager 2000 (Alpha Innotech, San Leandro, CA, USA) and the IOD (Integrated Optical Density) was normalized using β-actin as previously described [28].

2.5. Immunohistochemistry 

Right after sampling, each portion of the placenta was washed in PBS and fixed in 4% paraformaldehyde for 36–48 h. Then, they were sent to the pathology laboratory of the Centre de Recherche du CHUL for paraffin embedding and mounting on microscope slides (4 μm). Immunohistochemistry analysis was performed using the Vectastain Elite ABC kit (Vector Laboratories Inc., Burlingame, CA, USA) in accordance with the manufacturer's instructions. All incubations were done at room temperature unless otherwise stated. A 30 min treatment with 0.3% H2O2 in methanol was used to block the endogenous peroxidase activity, and later, microwave oven heating for 15 min at 100 °C allowed a better retrieval of epitopes.

The slides were incubated for 1 h with 10% horse serum (Sigma–Aldrich, Oakville, ON, Canada) in PBS to block non-specific binding, and then hybridized overnight at 4 °C with one of the following rabbit polyclonal antibodies: anti-SOD1 (1:100; Novus Biologicals, Littleton, CO, USA), anti-SOD2 (1:500; GeneTex, San Antonio, TX, USA), anti-SOD3 (1:1000; Abcam, Cambridge, MA, USA). Corresponding non-specific IgG were used as a negative control and processed in parallel. After washing in PBS, the slides were incubated using a biotinylated goat anti-rabbit IgG (1:1500; Vector Laboratories Inc., Burlingame, CA, USA) for 1 h. The slides were incubated 30 min with the ABC elite reagent and after washing, the color reaction was developed with 3-amino-9-ethylcarbazole (AEC). The sections were then counterstained with Gill hematoxylin (Ricca Chemical Company, Arlington, TX, USA) and mounted with MOWIOL medium. An Axioskop 2 Plus microscope (Zeiss, Toronto, ON, Canada) coupled to a digital camera was used to capture color images.

2.6. Statistical analysis 

A three way analysis of variance with repeated measures was used to study the effects of the pathology (control, PE), the placental layer (AC, V, MS) and the distance from the umbilical cord (peri-insertion, mid-disc and periphery) on mRNA levels. The MIXED procedure of the SAS 9.2 software (SAS Institute Inc., 2008; Cary, NC, USA) was used with a repeated statement and a covariance structure that minimize the Akaike criterion. A logarithmic transformation was used on mRNA values in order to reach normality and the homogeneity of the variance. Pair-wise comparisons were then made using the protected Fisher LSD (least significant difference). A student's t-test was used in Table 1, as well as in the statistical analysis of the protein level (IODn). Correlation coefficients were determined by the Spearman test. For all analyses, a p value below 0.05 was considered significant. All data represented are means ± SEM.

Back to Article Outline

3. Results 

3.1. Characteristics of the study population 

Clinical features of the recruited mothers are shown in Table 1. We recruited 25 PE and 23 normotensive pregnant women (NP) as controls. In the control group, there were 14 vaginal deliveries and 9 elective cesarean respectively. In the PE group, there were 11 vaginal deliveries and 14 elective cesareans. In the vaginal deliveries, all patients underwent labor, whereas in cesareans there was no labor registered. The maternal age between controls and PE was similar for vaginal deliveries but differs of about 3 years for cesareans (p = 0.03). The gestational age at delivery for the PE group was 4 weeks lower for cesareans and three weeks lower for vaginal deliveries than the respective control groups (p < 0.001). Furthermore, the mean diastolic and systolic blood pressures were significantly higher in all PE (p < 0.001) compared to the controls. As expected, proteinuria was also elevated in all PE. The placental weight was only lower in PE cesareans when compared to corresponding controls. SGA was observed in 10 cases in the PE group (6 cesarean and 4 vaginal deliveries) respectively. No sign of chorioamnionitis or any other placental infections were observed after pathological examination.

3.2. SOD mRNA and protein expression in the placenta 

The sampling site, the gestational age and the occurrence of SGA had no impact on mRNA and protein expression after statistical analysis (data not shown). However, the mRNA and protein levels for SOD1, SOD2 and SOD3 were impacted by the placental layers (AC, V and MS), as well as by the pathological state (NP or PE).

Fig. 1 shows the mRNA and protein levels for SOD1 in the placenta. In absence of labor (cesarean), the expression of SOD1 mRNA in the AC membranes was very similar between the PE and the control group (Fig. 1A). However in the villi, there was a lower level of mRNA expression in PE than in normotensive pregnancies (V;−27%, p = 0.048). In addition, there was a trend toward a decrease in the maternal side layer (MS; −23%, p = 0.058) for SOD1 expression in PE compared to controls.

  • View full-size image.
  • Fig. 1 

    Placental SOD1 expression from normotensive (NP; □) and preeclamptic (PE; ■) pregnancies in absence (A, C) or presence of labor (B, D). The mRNA level was measured by RT-PCR (A, B). Three housekeeping genes (YWHAZ, SDHA, hTBP) were analyzed by the GeNorm algorithm. Relative gene expression values are expressed as a ratio of SOD1 mRNA/M value (cesareans: n = 9 for NP and n = 14 for PE; vaginal deliveries: n = 14 for NP and n = 11 for PE). SOD1 protein level was obtained by western blot and quantified by densitometry (IODn) (C, D). The loading control was β-actin (n = 8 for NP and PE). Data are means ± SEM. *: p < 0.05 and §: p < 0.1 for the PE group compared to the NP control group.

Patients were randomly picked for protein analysis (n = 8 for each mode of delivery in NP and PE groups). In absence of labor, no difference was noted in the fetal membranes for SOD1 protein level (Fig. 1C) as observed for the mRNA analysis (Fig. 1A). However, the SOD1 protein level was about 2-fold (p = 0.031) and 1.6-fold (p = 0.038) lower in PE than controls, in the villi and MS samples respectively (Fig. 1C). The latter confirms the trend observed with mRNA expression in the villi (Fig. 1A).

In presence of labor (vaginal deliveries, Fig. 1B), SOD1 mRNA expression in the preeclamptic AC membranes was almost 2-fold higher than in controls (p = 0.014). In the placental villi, we noted a trend toward a lower SOD1 mRNA level in PE (−29%, p = 0.063). The SOD1 protein expression was in accordance with the mRNA expression in the AC membranes. Indeed, SOD1 protein level was 2.3-fold higher in PE than in controls (p = 0.026). However, there was no difference between the two groups in the protein level of SOD1 in either the villi or the MS samples (Fig. 1D).

In absence of labor (Fig. 2A and C), there was no significant difference between controls and PE in all placental layers for SOD2 mRNA and protein levels. In contrast, both, SOD2 mRNA and protein expressions were higher in all PE placental layers than controls in presence of labor (Fig. 2B and D). Indeed, we measured higher SOD2 mRNA level in AC membranes, villi and MS samples by 1.9-fold (p = 0.020), 2.1-fold (p = 0.051) and 1.9-fold (p = 0.032) respectively (Fig. 2B). Accordingly, a higher SOD2 protein level was observed in PE by 1.8-fold in AC (p = 0.027), 1.6-fold in villi (p = 0.011) and MS (p = 0.047) than controls (Fig. 2D).

  • View full-size image.
  • Fig. 2 

    Placental SOD2 expression from normotensive (NP; □) and preeclamptic (PE; ■) pregnancies in absence (A, C) or presence of labor (B, D). The mRNA level was measured by RT-PCR (A, B). Three housekeeping genes (YWHAZ, SDHA, hTBP) were analyzed by the GeNorm algorithm. Relative gene expression values are expressed as a ratio of SOD2 mRNA/M value (cesareans: n = 9 for NP and n = 14 for PE; vaginal deliveries: n = 14 for NP and n = 11 for PE). SOD1 protein level was obtained by western blot and quantified by densitometry (IODn) (C, D). The loading control was β-actin (n = 8 for NP and PE). Data are means ± SEM. *: p < 0.05 and §: p < 0.1 for the PE group compared to the NP control group.

Similar to SOD2, no significant difference in SOD3 expression was found between control and PE placentas in absence of labor (Fig. 3A and C). Also, SOD3 mRNA and protein levels were in accordance with SOD2 expression in presence of labor (Fig. 2, Fig. 3). Indeed, in PE there were a 1.6-fold higher SOD3 mRNA expression (p = 0.014; Fig. 3B) and 2-fold higher SOD3 protein level (p = 0.015; Fig. 3D) in the AC than controls. Furthermore, we also noted higher SOD3 mRNA expression (2-fold, p = 0.003; Fig. 3B) and protein level (2.8-fold, p = 0.043; Fig. 3D) in the villi from PE placentas than controls. In the MS samples, the mRNA (2.3-fold, p = 0.004; Fig. 3B) and the protein level (2.2-fold, p = 0.013; Fig. 3D) were also more elevated in PE placentas when compared to controls.

  • View full-size image.
  • Fig. 3 

    Placental SOD3 expression from normotensive (NP; □) and preeclamptic (PE; ■) pregnancies in absence (A, C) or presence of labor (B, D). The mRNA level was measured by RT-PCR (A, B). Three housekeeping genes (YWHAZ, SDHA, hTBP) were analyzed by the GeNorm algorithm. Relative gene expression values are expressed as a ratio of SOD3 mRNA/M value (cesareans: n = 9 for NP and n = 14 for PE; vaginal deliveries: n = 14 for NP and n = 11 for PE). SOD3 protein levels was obtained by western blot and quantified by densitometry (IODn) (C, D). The loading control was β-actin (n = 8 for NP and PE). Data are means ± SEM. *: p < 0.05 and §: p < 0.1 for the PE group compared to the NP control group.

3.3. Strong relationship between SOD isoforms in the placenta 

Table 2 summarizes the correlations observed between the three SOD isoforms in control pregnancies at the mRNA level. In absence of labor, only one significant correlation was observed. This correlation was observed exclusively between SOD1 and SOD3 in villi (r = 0.88). In presence of labor however, we found a greater number of significant positive correlations between SODs. Indeed, SOD1 and SOD3 are correlated in the MS (r = 0.58), while SOD2 and SOD3 are correlated in all tissues (r = 0.82), and specifically in the AC (r = 0.83) and MS (r = 0.65).

Table 2. Correlations between SOD1, SOD2 and SOD3 mRNA expression in normotensive placentas and the impact of the mode of delivery.
CorrelationsNo labor (cesarean)Labor (vaginal deliveries)
SOD1 and SOD2
All tissuesnsns
Amnion-chorionnsns
Villinsr = 0.59 (p = 0.020)
Maternal side layernsns

SOD1 and SOD3
All tissuesnsns
Amnion-chorionnsns
Villir = 0.88 (p = 0.039)ns
Maternal side layernsr = 0.58 (p = 0.024)

SOD2 and SOD3
All tissuesnsr = 0.82 (p < 0.001)
Amnion-chorionnsr = 0.83 (p < 0.001)
Villinsns
Maternal side layernsr = 0.65 (p < 0.001)

Values (r) are Spearman correlation coefficients. ns = non-significant (p < 0.05).

In Table 3, the correlations between the three SODs were investigated in preeclampsia. In absence of labor, positive correlations were only observed between SOD2 and SOD3 in the whole placenta (r = 0.61), the villi (r = 0.56) and the MS (r = 0.54). In presence of labor this correlation was kept only in the whole placenta (r = 0.74). On the other hand, a new correlation appears between SOD1 and SOD2 in MS during labor (r = 0.79).

Table 3. Correlations between SOD1, SOD2 and SOD3 mRNA expression in preeclamptic placentas and the impact of the mode of delivery.
CorrelationsNo labor (cesarean)Labor (vaginal deliveries)
SOD1 and SOD2
All tissuesnsns
Amnion-chorionnsns
Villinsns
Maternal side layernsr = 0.79 (p = 0.004)

SOD1 and SOD3
All tissuesnsns
Amnion-chorionnsns
Villinsns
Maternal side layernsns

SOD2 and SOD3
All tissuesr = 0.61 (p = 0.020)r = 0.74 (p = 0.008)
Amnion-chorionnsns
Villir = 0.56 (p = 0.039)ns
Maternal side layerr = 0.54 (p = 0.045)ns

Values (r) are Spearman correlation coefficients. ns = non-significant (p < 0.05).

3.4. Placental immunolocalization of SOD isoforms 

Examination of placentas from both PE and control groups showed that SOD isoforms were differentially expressed in different cell types (Fig. 4). However, there was no subjective noticeable change in the localization of the SODs in PE placentas compared to the controls or according to the mode of delivery (3 placentas investigated per conditions).

  • View full-size image.
  • Fig. 4 

    Immunolocalization of SOD1, SOD2 and SOD3 in the placenta. Paraffin embedded sections of amnion-chorion, villi, and maternal side layer were stained using polyclonal antibodies specific for SOD1, SOD2 and SOD3. Since no difference in the localization was observed between NP and PE placentas, only the NP placentas are shown. The red color represents the positive result and the counterstain is blue. Corresponding non-specific IgGs were used as negative controls. Panel magnifications are as follow: A and D (5×); G, J, K and L (10×); B, C, E, F, H and I (20×). Amnion (A), chorion (C), blood vessels (BV), syncytiotrophoblast (ST), stromal cells (SC) and decidual cells (DC).

In the fetal membranes, SOD1 was predominantly found in the chorion (Fig. 4A) whereas SOD2 was mainly expressed in the amnion (Fig. 4D). SOD3 showed a faint staining in both amnion and chorion (Fig. 4G).

In the placental villi, SOD1 protein was strongly expressed in the syncytiotrophoblast and the vascular endothelium, but only a faint staining was observed in the stromal cells of the villous core (Fig. 4B). Inversely, SOD2 was mainly present in the stromal cells and was absent from the syncytiotrophoblast and the vascular endothelium (Fig. 4E). SOD3 has a similar localization as that of SOD1 and was mostly expressed in the syncytiotrophoblast and the vascular endothelium (Fig. 4H). In the MS samples of the placenta, the decidual cells were positively stained for all three SOD isoforms (Fig. 4C, F and I).

Back to Article Outline

4. Discussion 

In this study, we did not observe any significant difference in SODs mRNA and protein expression according to the sampling site. Supporting our results, Jendryczko et al. (1991) and Boggess et al. (1998) showed that both total SOD and SOD3 activities were not influenced by the distance from the umbilical cord in placental homogenates [15], [17]. It is also possible that intrauterine growth restriction may affect SOD expression in placentas. In our study, we did not observe any effect of SGA on SOD expression. The number of cases of SGA was probably too low to adequately address this question, and a Doppler analysis would be required for a direct association with intrauterine growth restriction [29]. Indeed, 10 PE cases were SGA below the 10th percentile but within these cases, only 2 were below the 3rd percentile (results not shown).

Immunolocalization experiments revealed that there was no subjective difference in the cellular relative expression of SODs between normotensive and PE placentas which is in accordance with previous publications [15], [30], [31]. However, no in-depth quantitative analysis was carried out in this study. In agreement with what we are reporting, Telfer et al. (1997) previously showed that there was a moderate staining in the chorion for SOD1, whereas SOD2 was found predominantly in the amnion [31]. They also reported an intense SOD1 immunoreactivity in the syncytiotrophoblast and a moderate staining in endothelial cells of blood vessels and villous stromal cells. Also, they detected SOD2 in the syncytiotrophoblast, the endothelium and the stromal cells of placental villi [31]. In contrast, we did not find any significant expression of SOD2 in the syncytiotrophoblast. Supporting our findings, Myatt et al. (1997) observed only a faint staining for SOD2 in the syncytiotrophoblast [30]. In accordance with our results, Telfer et al. (1997) also observed an intense staining for SOD1 and SOD2 in the decidual cells [31]. We are the first to report a faint staining for SOD3 in the amnio-chorion. In the villi, we report a similar staining for SOD3 and SOD1. Indeed, Boggess et al. (1998) also studied the distribution of SOD3 in the placental villi [15]. They concluded that SOD3 was localized within the villous extracellular matrix around arterioles.

In this study, we observed lower mRNA and protein levels for SOD1 in the villi and maternal side layer in PE compared to the controls only in absence of labor. This observation is concordant with the reports from Wang et al. (1996, 2001) who described a significantly decreased mRNA expression for SOD1 in trophoblast cells isolated from preeclamptic placentas [8], [10]. It has been mentioned that SOD1 represents 90% of the total SOD activity [32], [33], and therefore, we believe that its decreased expression is the cause of the reduced total activity observed by previous studies. Moreover, we are the first to report a significant increase of SOD1 mRNA and protein level in the fetal membranes of preeclamptic patients who went through labor. Of note, it would have been of interest to investigate the impact of higher placental SOD1 level in patients with Down syndrome (SOD1 is located on the chromosome 21) in relation to preeclampsia. Indeed, there is about a 1.5 higher level of SOD1 mRNA, protein and activity in isolated villous cytotrophoblasts from Down syndrome than control pregnancies [34].

We showed that the combined effects of labor and PE cause an increase in SOD2 mRNA and protein expression. This upregulation may be explained by the level of pro-inflammatory cytokines. Indeed, labor is often described as an inflammatory process [35]. Sugino et al. (1998) reported that SOD2 mRNA was highly induced by inflammatory cytokines such as TNF-α, IL-6 and IL-1β in a luteal cell line [36]. PE is considered as a pro-inflammatory state as well [22]. Our results suggest that the combined effects of labor and PE are required to induce SOD2 in the whole placenta.

It is also known that PE placentas produce more progesterone than NP placentas [37], and that SOD2 is upregulated by progesterone in human endometrial cells (ESC) [38]. Furthermore, Löfgren et al. (1997) showed that the progesterone concentration in maternal and umbilical sera were significantly higher following normal labor than after an elective cesarean [39]. We believe that the increase in SOD2 mRNA and protein expression in all placental layers from our vaginal delivery group could partly be attributed to progesterone and to pro-inflammatory cytokines as discussed above.

We observed in our study that SOD3 mRNA and protein expression in presence of labor was higher in PE placentas than in controls. We know that both labor and PE cause oxidative stress [1], [16]. It was shown that placental nitric oxide synthase (NOS) activity and nitric oxide (NO) production were significantly increased in PE, and were directly related to the severity of the pathology [40]. It was also reported that an increased NOS expression in mouse vessel endothelial cells upregulates the expression of SOD3 in adjacent smooth muscle via cGMP/PKG and p38MAP kinase–dependent pathways [41]. Indeed, O2− can have an indirect influence on the vascular tone by inactivating NO [2]. The O2− can react with NO to form peroxynitrite (ONOO−), a powerful oxidant that ultimately reduces the bioavailability of NO [28], [42]. Therefore, SODs may compete with NO for O2− and, as a result, lengthen the action of NO and promote its vasodilatory effects [15], [30]. Consequently, a possible mechanism that can explain the upregulation of SOD3 in PE in presence of labor is the NO level. However, there is no evidence suggesting that labor increases NO output, since endothelial and inducible NOS mRNA and protein did not differ in presence or absence of labor [43]. Again, only a combination of labor and PE is required to induce SOD3.

A strong positive correlation between SOD2 and SOD3 mRNA expression has been found in presence of labor in normotensive pregnancies. The latter correlation also exists in preeclampsia in presence or not of labor. These SODs are expressed in different layers and cells of the placenta. Indeed, SOD2 was found mainly in the amnion, the stromal cells and the decidua. This correlation suggests that mitochondrial SOD2 and extracellular SOD3 work together to protect the placenta against the oxidative damages caused by labor and/or preeclampsia.

In summary, the differential regulation of SODs in preeclampsia confirms their implication in the defense against oxidative stress in this syndrome. In light of the presented results, we can state that the mode of delivery is a major contributor to the regulation of SODs expression. It is worth mentioning that most of the previous studies did not take into account the effects of the mode of delivery as a potential bias in their research, and therefore it is of the utmost importance that this factor be considered in the upcoming studies.

Back to Article Outline

Acknowledgements 

This work was supported by a grant from the Canadian Institutes of Health Research (CIHR, grant|MOP-84219). J.-F.B is a recipient of a Canadian Institutes of Health Research (CIHR), Institute of Aging, New investigators award. Isabelle St-Pierre is a recipient of a scholarship from the Fondation Jeanne et Jean-Louis Lévesque. The authors would like to thank Ms Olga Gordynska and Hélène Crépeau (Laval University, Statistical service) for the statistical analysis.

Back to Article Outline

References 

  1. Myatt L, Cui X. Oxidative stress in the placenta. Histochem Cell Biol. 2004;122(4):369–382
  2. Gryglewski RJ, Palmer RM, Moncada S. Superoxide anion is involved in the breakdown of endothelium-derived vascular relaxing factor. Nature. 1986;320(6061):454–456
  3. Bannister JV, Bannister WH, Rotilio G. Aspects of the structure, function, and applications of superoxide dismutase. CRC Crit Rev Biochem. 1987;22(2):111–180
  4. Zelko IN, Mariani TJ, Folz RJ. Superoxide dismutase multigene family: a comparison of the CuZn-SOD (SOD1), Mn-SOD (SOD2), and EC-SOD (SOD3) gene structures, evolution, and expression. Free Radic Biol Med. 2002;33(3):337–349
  5. Perkins AV. Endogenous anti-oxidants in pregnancy and preeclampsia. Aust N Z J Obstet Gynaecol. 2006;46(2):77–83
  6. Nozik-Grayck E, Suliman HB, Piantadosi CA. Extracellular superoxide dismutase. Int J Biochem Cell Biol. 2005;37(12):2466–2471
  7. Ali Akbar S, Nicolaides KH, Brown PR. Measurement of Cu/Zn SOD in placenta, cultured cells, various fetal tissues, decidua and semen by ELISA. J Obstet Gynaecol. 1998;18(4):331–335
  8. Wang Y, Walsh SW. Antioxidant activities and mRNA expression of superoxide dismutase, catalase, and glutathione peroxidase in normal and preeclamptic placentas. J Soc Gynecol Investig. 1996;3(4):179–184
  9. Vanderlelie J, Venardos K, Clifton VL, Gude NM, Clarke FM, Perkins AV. Increased biological oxidation and reduced anti-oxidant enzyme activity in pre-eclamptic placentae. Placenta. 2005;26(1):53–58
  10. Wang Y, Walsh SW. Increased superoxide generation is associated with decreased superoxide dismutase activity and mRNA expression in placental trophoblast cells in pre-eclampsia. Placenta. 2001;22(2–3):206–212
  11. Wiktor H, Kankofer M. Superoxide dismutase activity in normal and preeclamptic placentas. Ginekol Pol. 1998;69(12):915–918
  12. Ademuyiwa O, Odusoga OL, Adebawo OO, Ugbaja RN. Endogenous antioxidant defences in plasma and erythrocytes of pregnant women during different trimesters of pregnancy. Acta Obstet Gynecol Scand. 2007;1–6
  13. Bulgan Kilicdag E, Ay G, Celik A, Ustundag B, Ozercan I, Simsek M. Oxidant-antioxidant system changes relative to placental-umbilical pathology in patients with preeclampsia. Hypertens Pregnancy. 2005;24(2):147–157
  14. Dordevic NZ, Babic GM, Markovic SD, Ognjanovic BI, Stajn AS, Zikic RV, et al. Oxidative stress and changes in antioxidative defense system in erythrocytes of preeclampsia in women. Reprod Toxicol. 2008;25(2):213–218
  15. Boggess KA, Oury TD, Kay HH, Crapo JD. Extracellular superoxide dismutase localization and activity within the human placenta. Placenta. 1998;19(5–6):417–422
  16. Cindrova-Davies T, Yung HW, Johns J, Spasic-Boskovic O, Korolchuk S, Jauniaux E, et al. Oxidative stress, gene expression, and protein changes induced in the human placenta during labor. Am J Pathol. 2007;171(4):1168–1179
  17. Hempstock J, Bao YP, Bar-Issac M, Segaren N, Watson AL, Charnock-Jones DS, et al. Intralobular differences in antioxidant enzyme expression and activity reflect the pattern of maternal arterial bloodflow within the human placenta. Placenta. 2003;24(5):517–523
  18. Helewa ME, Burrows RF, Smith J, Williams K, Brain P, Rabkin SW. Report of the Canadian hypertension Society Consensus Conference: 1. Definitions, evaluation and classification of hypertensive disorders in pregnancy. Cmaj. 1997;157(6):715–725
  19. Kramer MS, Platt RW, Wen SW, Joseph KS, Allen A, Abrahamowicz M, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108(2):E35
  20. Bacon BJ, Gilbert RD, Longo LD. Regional anatomy of the term human placenta. Placenta. 1986;7(3):233–241
  21. Mayhew TM, Burton GJ. Methodological problems in placental morphometry: apologia for the use of stereology based on sound sampling practice. Placenta. 1988;9(6):565–581
  22. Benyo DF, Smarason A, Redman CW, Sims C, Conrad KP. Expression of inflammatory cytokines in placentas from women with preeclampsia. J Clin Endocrinol Metab. 2001;86(6):2505–2512
  23. Teasdale F, Jean-Jacques G. Intrauterine growth retardation: morphometry of the microvillous membrane of the human placenta. Placenta. 1988;9(1):47–55
  24. Vanderlelie J, Gude N, Perkins AV. Antioxidant gene expression in preeclamptic placentae: a preliminary investigation. Placenta. 2008;29(6):519–522
  25. Chen X, Niroomand F, Liu Z, Zankl A, Katus HA, Jahn L, et al. Expression of nitric oxide related enzymes in coronary heart disease. Basic Res Cardiol. 2006;101(4):346–353
  26. Lapointe J, Kimmins S, Maclaren LA, Bilodeau JF. Estrogen selectively up-regulates the phospholipid hydroperoxide glutathione peroxidase in the oviducts. Endocrinology. 2005;146(6):2583–2592
  27. Meller M, Vadachkoria S, Luthy DA, Williams MA. Evaluation of housekeeping genes in placental comparative expression studies. Placenta. 2005;26(8–9):601–607
  28. Roy M, Gauvreau D, Bilodeau JF. Expression of superoxide dismutases in the bovine oviduct during the estrous cycle. Theriogenology. 2008;70(5):836–842
  29. Ertan AK, Schanz S, Tanriverdi HA, Meyberg R, Schmidt W. Doppler examinations of fetal and uteroplacental blood flow in AGA and IUGR fetuses before and after maternal physical exercise with the bicycle ergometer. J Perinat Med. 2004;32(3):260–265
  30. Myatt L, Eis AL, Brockman DE, Kossenjans W, Greer IA, Lyall F. Differential localization of superoxide dismutase isoforms in placental villous tissue of normotensive, pre-eclamptic, and intrauterine growth-restricted pregnancies. J Histochem Cytochem. 1997;45(10):1433–1438
  31. Telfer JF, Thomson AJ, Cameron IT, Greer IA, Norman JE. Expression of superoxide dismutase and xanthine oxidase in myometrium, fetal membranes and placenta during normal human pregnancy and parturition. Hum Reprod. 1997;12(10):2306–2312
  32. Crapo JD, Oury T, Rabouille C, Slot JW, Chang LY. Copper, zinc superoxide dismutase is primarily a cytosolic protein in human cells. Proc Natl Acad Sci U S A. 1992;89(21):10405–10409
  33. Sturtz LA, Diekert K, Jensen LT, Lill R, Culotta VC. A fraction of yeast Cu, Zn-superoxide dismutase and its metallochaperone, CCS, localize to the intermembrane space of mitochondria. A physiological role for SOD1 in guarding against mitochondrial oxidative damage. J Biol Chem. 2001;276(41):38084–38089
  34. Frendo JL, Therond P, Guibourdenche J, Vidaud M, Evain-Briona D. Implication of copper zinc superoxide dismutase (SOD-1) in human placenta development. Ann N Y Acad Sci. 2002;973:297–301
  35. Osman I, Young A, Ledingham MA, Thomson AJ, Jordan F, Greer IA, et al. Leukocyte density and pro-inflammatory cytokine expression in human fetal membranes, decidua, cervix and myometrium before and during labour at term. Mol Hum Reprod. 2003;9(1):41–45
  36. Sugino N, Telleria CM, Gibori G. Differential regulation of copper-zinc superoxide dismutase and manganese superoxide dismutase in the rat corpus luteum: induction of manganese superoxide dismutase messenger ribonucleic acid by inflammatory cytokines. Biol Reprod. 1998;59(1):208–215
  37. Walsh SW. Progesterone and estradiol production by normal and preeclamptic placentas. Obstet Gynecol. 1988;71(2):222–226
  38. Sugino N, Karube-Harada A, Kashida S, Takiguchi S, Kato H. Differential regulation of copper-zinc superoxide dismutase and manganese superoxide dismutase by progesterone withdrawal in human endometrial stromal cells. Mol Hum Reprod. 2002;8(1):68–74
  39. Lofgren M, Backstrom T. High progesterone is related to effective human labor. Study of serum progesterone and 5alpha-pregnane-3,20-dione in normal and abnormal deliveries. Acta Obstet Gynecol Scand. 1997;76(5):423–430
  40. Shaamash AH, Elsonosy ED, Zakhari MM, Radwan SH, El-Dien HM. Placental nitric oxide synthase (NOS) activity and nitric oxide (NO) production in normal pregnancy, pre-eclampsia and eclampsia. Int J Gynaecol Obstet. 2001;72(2):127–133
  41. Fukai T, Siegfried MR, Ushio-Fukai M, Cheng Y, Kojda G, Harrison DG. Regulation of the vascular extracellular superoxide dismutase by nitric oxide and exercise training. J Clin Invest. 2000;105(11):1631–1639
  42. Chaudiere J, Ferrari-Iliou R. Intracellular antioxidants: from chemical to biochemical mechanisms. Food Chem Toxicol. 1999;37(9–10):949–962
  43. Kakui K, Sagawa N, Itoh H, Yura S, Korita D, Takemura M, et al. Expression of nitric oxide synthase isoforms in the human placenta is not altered by labor. Endocr J. 2003;50(5):535–544

PII: S0143-4004(10)00086-X

doi:10.1016/j.placenta.2010.02.007

Placenta
Volume 31, Issue 5 , Pages 392-400, May 2010

Article Tools

* Image Usage & Resolution