Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2009 Jun 11;150(9):4316–4325. doi: 10.1210/en.2009-0076

The Uterine Placental Bed Renin-Angiotensin System in Normal and Preeclamptic Pregnancy

Lauren Anton 1, David C Merrill 1, Liomar A A Neves 1, Debra I Diz 1, Jenny Corthorn 1, Gloria Valdes 1, Kathryn Stovall 1, Patricia E Gallagher 1, Cheryl Moorefield 1, Courtney Gruver 1, K Bridget Brosnihan 1
PMCID: PMC2736074  PMID: 19520788

Abstract

Previously, we demonstrated activation of the renin-angiotensin system in the fetal placental chorionic villi, but it is unknown whether the immediately adjacent area of the maternal uterine placental bed is regulated similarly. This study measured angiotensin peptides, renin-angiotensin system component mRNAs, and receptor binding in the fundus from nonpregnant subjects (n = 19) and in the uterine placental bed from normal (n = 20) and preeclamptic (n = 14) subjects. In the uterine placental bed from normal pregnant women, angiotensin II peptide levels and angiotensinogen, angiotensin-converting enzyme, angiotensin receptor type 1 (AT1), AT2, and Mas mRNA expression were lower as compared with the nonpregnant subjects. In preeclamptic uterine placental bed, angiotensin II peptide levels and renin and angiotensin-converting enzyme mRNA expression were significantly higher than normal pregnant subjects. The AT2 receptor was the predominant receptor subtype in the nonpregnant fundus, whereas all angiotensin receptor binding was undetectable in normal and preeclamptic pregnant uterine placental bed compared with nonpregnant fundus. These findings suggest that the maternal uterine placental bed may play an endocrine role by producing angiotensin II, which acts in the adjacent placenta to vasoconstrict fetal chorionic villi vessels where we have shown previously that AT1 receptors predominate. This would lead to decreased maternal-fetal oxygen exchange and fetal nutrition, a known characteristic of preeclampsia.

In the preeclamptic uterine placental bed, Ang II, a component of the renin angiotensin system, is increased; however, the finding of decreased uterine angiotensin receptors suggests that Ang II may be acting in an endocrine manner on the adjacent fetal placenta, causing vasoconstriction and consequently defects in maternal-fetal oxygen and nutrient exchange.

Preeclampsia is the second leading cause of maternal mortality in the United States, affecting 7–10% of all pregnancies, and contributes significantly to neonatal mortality and morbidity (1); however, despite the large amount of research focused on this disease, the pathophysiological causes of preeclampsia remain uncertain. Recent studies suggest that in preeclamptic women, abnormal placentation leads to shallow trophoblast invasion of the maternal spiral arteries in the endometrium of the uterus, resulting in improper remodeling of the uterine spiral arteries (2,3), and consequently, placental hypoxia activates and releases several placental factors into the maternal circulation leading to hypertension, endothelial dysfunction, and proteinuria (3,4,5).

Several recent studies have suggested that the renin angiotensin system (RAS) may be playing a role in the development of preeclampsia. During normal pregnancy, the RAS is activated due to increased estrogen levels, which consequently cause levels of angiotensinogen and renin (6,7) to rise. The activation of the RAS ultimately leads to increases in angiotensin (Ang) II levels (8,9); however, previous studies illustrated that pregnant women are resistant to the pressor effects of Ang II resulting in normal or decreased blood pressure levels (10,11,12). The role of the RAS in normal pregnancy is incompletely understood; however, dysregulation of the RAS has been hypothesized to play a role in the pathophysiology of preeclampsia. A previously published study in our laboratory found that many of the components of the circulating RAS in women with preeclampsia are down-regulated, including plasma Ang I, Ang II, Ang-(1-7), and plasma renin activity (13). In contrast to the down-regulation of the circulating RAS in preeclampsia, a second study done in our laboratory showed that Ang II levels and angiotensinogen and AT1 receptor mRNA levels were increased in the chorionic villi of women with preeclampsia (14). Because the chorionic villi are an essential component of the placenta that is responsible for maternal-fetal oxygen and nutrient transport, our findings are consistent with increased concentration of Ang II causing increased vasoconstriction within the chorionic villi, which may be contributing to the pathophysiology of preeclampsia by decreasing maternal-fetal exchange of oxygen and nutrients (14). After completing the investigation of RAS components in the chorionic villi or fetal side of the placenta, we were interested in investigating the expression of RAS components in the maternal side or fundus (placental bed site) of the uterus in normal and preeclamptic women. We measured angiotensin peptide levels, RAS component mRNAs, angiotensin receptor binding, and the distribution and localization of Ang-(1-7) and angiotensin-converting enzyme 2 (ACE2) in the uterine fundus from nonpregnant, normal pregnant, and preeclamptic subjects.

Materials and Methods

Human subjects

These experiments were conducted using human fundus or placental bed uterine tissue collected from three separate groups of subjects. Group 1 (n = 19) consisted of nonpregnant women undergoing hysterectomy for benign fibroids or endometriosis. Normal uterine tissue adjacent to the fibroids or endometriosis was removed. This group was used as a nonpregnant control with recognition of its limitations. Group 2 (n = 20) consisted of normal pregnant subjects who remained normotensive throughout pregnancy (blood pressure <140/90 mm Hg), have no history of chronic blood pressure elevation, and have an absence of proteinuria. Group 3 (n = 14) consisted of preeclamptic subjects who developed new-onset hypertension (blood pressure >140/90 mm Hg) and proteinuria measured as greater than or equal to 1+ on a urine dipstick (>30 mg/dl) after the 20th week of gestation. Subjects in the pregnant groups were all nulliparous and underwent cesarean section. Patients with evidence of chorioamnionitis were excluded. Women in all three groups were over age 18 yr and less than age 50 yr and were free of other known cardiovascular, renal, or connective tissue diseases, diabetes, cancer, or hyperplasia.

The study was approved by the institutional review boards at both Wake Forest University School of Medicine and Forsyth Medical Center.

Experimental procedures

For both normal pregnant and preeclamptic subjects, immediately after delivery of the baby and placenta, three tissue sections containing both endometrium and myometrium were taken from the uterine placental bed (placenta attachment site). The total amount of time from the removal of the uterine samples until the tissues were processed did not exceed 15 min. For women undergoing hysterectomy, the uterus was removed and three tissue sections containing both endometrium and myometrium were taken from the uterine fundus in the operating room pathology lab. The first set of tissue sections were immediately snap frozen in liquid nitrogen and stored at −80 C for analysis of angiotensin peptides by RIAs or for quantification of angiotensinogen, renin, ACE, ACE2, neprilysin (NEP), and AT1, AT2, and Mas receptor mRNA by real-time PCR. The second set of tissue sections were frozen in O.C.T. (optimum cutting temperature) formulation (Tissue-Tek, Northbrook, IL) and stored at −80 C for AT1, AT2, and AT1-7 receptor density measurements by autoradiography. The third set of tissue sections were placed in 4% formalin for immunocytochemical analysis of Ang-(1-7) and ACE2.

Tissue concentration of angiotensin peptides

Frozen tissues were rapidly weighed and homogenized as previously described (15). Tissue homogenates were extracted using Sep-Pak columns as previously described (15). The eluate was divided for three RIAs [Ang I, Ang II, and Ang-(1-7)], and the solvent was evaporated. Ang I and Ang II were measured using a Peninsula RIA kit (San Carlos, CA) and an Alpco Diagnostic kit (Windham, NH), respectively. Ang-(1-7) was measured using an antibody produced in our laboratory (15). The minimum detectable levels of the assays were 1.0 fmol/ml for Ang I, 0.8 fmol/ml for Ang II, and 2.8 fmol/ml for Ang-(1-7). The intra- and interassay coefficients of variation for Ang I RIA are 18 and 22%, for Ang II are 12 and 22%, and for Ang-(1-7) are 8 and 20%, respectively.

RNA isolation and real-time RT-PCR

RNA was isolated from the uterine fundus and the placental bed, as described previously (14), using TRIZOL reagent (GIBCO Invitrogen, Carlsbad, CA), as directed by the manufacturer. The RNA concentration and integrity were assessed using an Agilent 2100 Bioanalyzer with an RNA 6000 Nano LabChip (Agilent Technologies, Palo Alto, CA). Approximately 1 μg total RNA was reverse transcribed using avian myeloblastosis virus reverse transcriptase as previously described (14). Human primer/probe sets for angiotensinogen, renin, ACE, NEP, AT1, AT2, and Mas receptors were purchased from Applied Biosystems (Foster City, CA) except for ACE2, which was our design (forward primer 5′-CCCAGAGAACAGTGGACCAAAA-3′, reverse primer 5′-GCTCCACCACACCAACGAT-3′, and probe 5′-FAM-CTCCCGCTTCATCTCC-3′). All reactions were performed in triplicate, and 18S rRNA, amplified using the TaqMan rRNA control kit (Applied Biosystems), served as an internal control. RAS component expression comparisons were made as previously published (14); the normal pregnant group was set as the basis of comparison.

In vitro receptor autoradiography

Uterine fundus or placental bed tissues frozen in O.C.T. freezing formulation (Tissue-Tek) were sectioned at 14 μm, and receptor autoradiography was performed using [125I](sarcosine1, threonine8) Ang II ([125I]SarThran) at 0.6 nm to determine the apparent maximal density of receptors (15,16). A lower concentration of [125I]SarThran (0.2 nm) was used in the presence or absence of 3 μm losartan, PD123,319, or d-Ala7-Ang-(1-7) (A779 or d-ala) to determine percentage of each receptor subtype present. Sections were exposed to film, and films were analyzed using a computerized densitometry system (Micro Computer Imaging Device, Ontario, Canada) (17). Data for binding density are expressed as the amount of total binding attributed to each receptor subtype as determined by the competition study.

Immunocytochemistry

Human uterine fundus tissues (n = 6 per group) were paraffin embedded, and Ang-(1-7) and ACE2 immunostaining was performed using the avidin-biotin method (18). The primary antibodies were an affinity-purified rabbit polyclonal antibody to Ang-(1-7) generated at Wake Forest University School of Medicine (1:25) and a mouse monoclonal antibody to ACE2 obtained as a gift from Dr. Josef Penninger (1:300). The purification and characterization of the Ang-(1-7) antibody has been described (19). The secondary antibody was biotinylated antirabbit or antimouse diluted 1:400 in 1% BSA. To assess the specificity of the staining, tissue sections were incubated without the primary antibody or by preabsorption of the antibody with 10 μm Ang-(1-7) or 10 μm of an ACE2 peptide (20,21,22). Antibodies to cytokeratin (mouse monoclonal anti-cytokeratin, 1:100) (Biomedia, Foster City, CA) and vimentin (goat antihuman-vimentin, 1:2000) (Sigma Chemical Co., St. Louis, MO) were used to characterize cells as cytotrophoblasts or decidual cells, respectively.

Statistics and data analysis

Based on our previous study in the chorionic villi (14), we made predetermined comparisons of normal pregnant and preeclamptic groups using the Student’s t test (GraphPad Software, San Diego, CA). Additional comparisons between nonpregnant and normal pregnant were made using the Student’s t test. A P value <0.05 was considered statistically different. All arithmetic means are presented ± sem.

Results

Clinical profile of the patient population

Table 1 shows the clinical profile of the three groups; the preeclamptic subjects had significant hypertension as shown by higher systolic, diastolic, and mean blood pressure (P < 0.0001) vs. normal pregnant women. No differences in blood pressure were seen between normal pregnant and nonpregnant women. The preeclamptic subjects had proteinuria measured as greater than or equal to 1+ on a urine dipstick (>30 mg/dl). The birth weight of the babies born to preeclamptic subjects was significantly lower than those of the normal pregnant subjects (P < 0.01). In addition, the preeclamptic subjects had a significantly shorter gestational term when compared with normal pregnant women (P < 0.01). There was no difference in patient age between normal pregnant and preeclamptic subjects, but patient age was significantly higher in the nonpregnant women (P < 0.001) vs. the pregnant subjects.

Table 1.

Clinical profile of the study population

Patient clinical characteristics Nonpregnant Normal pregnancy Preeclamptic pregnancy
n 19 20 14
Age (yr) 40.8 ± 1.4 27.4 ± 1.3 25.3 ± 1.4
Body mass index (kg/m2) 30.8 ± 2.3 28.5 ± 1.5 27.6 ± 1.9
Birth weight (g) 3017 ± 175 1901 ± 318a
Gestational age (wk) 37.5 ± 0.7 32.9 ± 1.3a
Systolic blood pressure (mm Hg) 125 ± 4 123 ± 2 165 ± 4b
Diastolic blood pressure (mm Hg) 77 ± 3 74 ± 2 97 ± 3b
Mean blood pressure (mm Hg) 93 ± 3 90 ± 2 120 ± 3b
Proteinuria (0–4+) None None >1+
Number of patients receiving anti-hypertensive medications None None 3c
Open in a new tab

Values are expressed as mean ± sem. PCO 

a

P < 0.01; 

b

P < 0.0001, normal pregnant vs. preeclamptic. 

c

Three preeclamptic patients were treated with hydralazine as antihypertensive therapy. 

Immunohistochemical localization of Ang-(1-7) and ACE2 in the uterus

Ang-(1-7) and ACE2 staining was found in the epithelial cells of the endometrial glands of the nonpregnant uterine fundus (Fig. 1, A and B). In the uterine placental bed from normal pregnant and preeclamptic women, we found Ang-(1-7) and ACE2 staining in the invading trophoblasts (Fig. 1, E and F) and in the trophoblast cells lining the uterine spiral arteries (Fig. 1, H and I). Trophoblast cells were positively identified by the cell-specific stain cytokeratin (Fig. 1, G and J). No difference in trophoblast staining was observed between normal and preeclamptic uterine tissues. Ang-(1-7) and ACE2 staining was seen in the normal and preeclamptic uterine decidual cells (Fig. 1, K and L), which are identified by the cell-specific stain vimentin (Fig. 1M). No Ang-(1-7) or ACE2 staining was seen in the myometrium (not shown) in the nonpregnant uterine fundus or in the uterine placental bed.

Figure 1.

Figure 1

Open in a new tab

Immunohistochemical expression of Ang-(1-7) and ACE2 in the uterus from nonpregnant, normal pregnant, and preeclamptic subjects. Immunohistochemical expression of Ang-(1-7) (A, E, H, and K) and ACE2 (B, F, I, and L) was measured in the glandular epithelium (A and B) in the uterine fundus from nonpregnant women and in the invading trophoblasts (E and F), intraarterial trophoblasts (H and I), and decidual cells (K and L) in the uterine placental bed from normal pregnant and preeclamptic women. Tissue sections stained with the cell-specific marker cytokeratin positively identified invading (G) and intraarterial (J) trophoblasts, whereas tissues stained with vimentin positively identified decidual cells (M). Tissue sections incubated in the absence of the primary antibody showed no Ang-(1-7) (C) or ACE2 (D) staining. Positive staining is brown, and nuclear counterstaining is blue. Scale bar, 100 μm.

Angiotensin peptide levels in the uterus

Ang II peptide levels were significantly lower in normal pregnant women (P < 0.0001) vs. nonpregnant women (Fig. 2B). In the preeclamptic uterine placental bed, Ang II was significantly higher (P < 0.05) vs. normal pregnant uterus. These results also show that Ang II is the predominant peptide in the fundus of nonpregnant women. However, Ang-(1-7) is the predominant peptide in the uterine placental bed from both normal pregnant and preeclamptic subjects. There were no significant differences in either Ang I or Ang-(1-7) peptide levels in the fundus or placental bed from the three groups of subjects (Fig. 2, A and C).

Figure 2.

Figure 2

Open in a new tab

Angiotensin peptide levels in the uterus from nonpregnant, normal pregnant, and preeclamptic women. Angiotensin peptide levels were measured by RIA of Ang I (A), Ang II (B), and Ang-(1-7) (C) in the uterine fundus of nonpregnant women and the uterine placental bed of normal pregnant and preeclamptic women. Data are expressed as the mean ± sem. *, P < 0.01 nonpregnant uterine fundus vs. normal pregnant uterine placental bed; #, P < 0.05 normal vs. preeclamptic uterus.

RAS gene expression in the uterus

Angiotensinogen mRNA was found to be lower in the placental bed in normal pregnant women vs. nonpregnant uterine fundus (P < 0.01) (Fig. 3A). There was no difference in angiotensinogen mRNA between normal pregnant and preeclamptic uterine bed. There was no difference in renin mRNA between nonpregnant fundus and normal pregnant placental bed. However, renin expression was significantly higher in the placental bed from preeclamptic women (P < 0.001) vs. normal pregnant subjects (Fig. 3A). There was significantly lower ACE mRNA in normal pregnant women vs. nonpregnant (P < 0.0001). ACE mRNA was higher in the preeclamptic (P < 0.001) uterus vs. normal pregnant women. NEP mRNA was significantly higher in the uterine placental bed from normal pregnant women vs. tissue from nonpregnant (P < 0.05) women. No difference in NEP expression was seen in the uterine placental bed from preeclamptic vs. normal pregnant women. No differences in ACE2 mRNA were seen among groups (Fig. 3B). AT1 (P < 0.05), AT2 (P < 0.0001), and Mas (P < 0.0001) receptor expression were all lower in the uterine placenta bed from normal pregnant women vs. nonpregnant women. No significant differences were seen in AT1, AT2, and Mas receptor mRNA between normal pregnant and preeclamptic women (Fig. 4, A–C).

Figure 3.

Figure 3

Open in a new tab

Gene expression of angiotensinogen, renin, ACE, ACE2, and NEP in the uterus from nonpregnant, normal pregnant, and preeclamptic women. Relative gene expression as determined by real-time RT-PCR of angiotensinogen and renin (A) and ACE, ACE2, and NEP (B) in the uterine fundus of nonpregnant women and the uterine placental bed of normal pregnant and preeclamptic women. Values for mRNA were normalized to the normal pregnant uterus as was done in our previous publication where RAS component mRNA expression was compared between normal and preeclamptic chorionic villi (14). Data are expressed as the mean ± sem. *, P < 0.05 nonpregnant uterine fundus vs. normal pregnant uterine placental bed; #, P < 0.01 normal pregnant vs. preeclamptic uterine placental bed.

Figure 4.

Figure 4

Open in a new tab

Gene expression of angiotensin receptors in the uterus from nonpregnant, normal pregnant, and preeclamptic women. Relative gene expression as determined by real-time RT-PCR of AT1 receptor (A), AT2 receptor (B), and Mas receptor (C) in the uterine fundus of nonpregnant women and the uterine placental bed of normal pregnant and preeclamptic women. *, P < 0.05 nonpregnant uterine fundus vs. normal pregnant uterine placental bed.

Angiotensin receptor binding in the uterus

The myometrium of the nonpregnant uterus had significantly higher total angiotensin receptor binding vs. the endometrium (P < 0.001) (Fig. 5, A, B, and E). Competition by the AT2 receptor antagonist PD123,319 was significantly higher than competition with either the AT1 receptor antagonist losartan (P < 0.05) or the AT1-7 receptor antagonist d-ala (P < 0.05) in the myometrium of nonpregnant women (Fig. 5, A and F). PD123,319 competed for 100% of the angiotensin receptors. But unexpectedly, losartan, the AT1 receptor antagonist, and d-ala, the AT1-7 antagonist, also competed for 60 and 40%, respectively. In the endometrium of the fundus from nonpregnant women (Fig. 5, A, B, and E), angiotensin receptor binding was detectable, but the values were so low that differential binding between the three receptors could not be determined. Receptor binding in both the decidua and myometrium of the normal pregnant and preeclamptic placental bed was undetectable (Fig. 5, C and D).

Figure 5.

Figure 5

Open in a new tab

Angiotensin receptor binding in the uterus from nonpregnant, normal pregnant, and preeclamptic women. AT1, AT2, and AT1-7 receptor density was measured by receptor autoradiography in the uterine fundus from nonpregnant women and in the uterine placental bed from normal pregnant and preeclamptic women as assessed by the displacement of [125I]SarThran binding by the AT1 receptor antagonist losartan (3 μm), the AT2 receptor antagonist PD123,319 (PD; 3 μm), and the AT1-7 receptor antagonist d-ala (3 μm) (A). Quantification of receptors in the endometrium and myometrium of the nonpregnant uterine fundus showed higher total angiotensin receptor binding in the myometrium vs. the endometrium (E). PD123,319 competed for the highest amount of angiotensin receptor binding in the myometrium of nonpregnant women (P < 0.05) when compared with competition with losartan and d-ala (F). Angiotensin receptors in the endometrium of the nonpregnant uterus, as identified by hematoxylin and eosin staining (B), are present, but the values were too low to quantify individual angiotensin receptor subtypes. Specific angiotensin receptor binding in the uterine placental bed of normal pregnant and preeclamptic women was undetectable as shown in the photomicrographs where total binding was similar to the nonspecific staining seen with SarThran competition (C and D). Data are expressed as the mean ± sem. *, P < 0.05, PD123,319 vs. losartan receptor density in the myometrium of nonpregnant uterine fundus; #, P < 0.05, PD123,319 vs. d-ala receptor density in the myometrium of nonpregnant uterine fundus.

Discussion

This study shows a lower expression of RAS components and subsequent Ang II down-regulation during normal pregnancy, which could contribute to a spiral artery vasodilatory state thus allowing increased blood flow into the uterus. In contrast, during preeclampsia, the uterine placental bed RAS components are significantly higher than normal pregnancy, with increases in renin and ACE mRNA, which are likely responsible for the observed increase in Ang II levels. This would provide support for failure to suppress the RAS in preeclamptic patients. Increased Ang II vasoconstriction within the uterine spiral arteries in the placental bed of the preeclamptic uterus would likely contribute to a reduction in placental blood flow seen in women with preeclampsia. However, interestingly, all three angiotensin receptor subtypes, AT1, AT2, and Mas mRNA, were found to be substantially lower in both normal pregnancy and preeclampsia. In agreement with these results, in vitro receptor autoradiography showed that angiotensin receptor binding was so low that it was undetectable in the placental bed of normal pregnant and preeclamptic women when compared with nonpregnant women. Although there is an increase in Ang II production in the uterine placental bed from preeclamptic women, the decrease in all three angiotensin receptors would provide evidence for diminished functionality of the RAS as a paracrine hormone in the placental bed of the uterus. The elevated Ang II in the face of undetectable angiotensin receptors in the preeclamptic uterus is consistent with Ang II playing an endocrine role whereby it would influence the adjacent placental chorionic villi. In fact, we previously found elevated AT1 receptors in chorionic villi of preeclamptic subjects that could contribute to the vasoconstriction of the fetal vessels and lead to growth restriction.

Angiotensinogen mRNA was present in the nonpregnant fundus as well as normal pregnant and preeclamptic placental bed tissues, in agreement with previous studies investigating angiotensinogen expression in first- and second-trimester and term decidual tissue (23). The lower angiotensinogen may contribute to lower production of Ang II during normal pregnancy. Our finding of no difference in angiotensinogen expression between normal and preeclamptic tissue agrees with studies by Herse et al. (24) in vacuum-suctioned decidual tissues.

Renin expression was reported in the decidua from both early pregnancy and term tissues (23,25). Although the tissue collection methods were different among studies, there is agreement that decidual tissue provides a greater source of renin expression than the fetal placenta (25). We also found renin expression in uterine tissue, consistent with total and active renin activity in endometrial nonpregnant uterine samples (25). Our samples from pregnant women contain both decidua and myometrium, but based on previous studies that demonstrated no renin mRNA or protein in myometrium (25), we predict that the renin present is primarily from decidua. Total and active renin were higher in decidual tissues than nonpregnant endometrium, whereas we found no differences in renin mRNA between nonpregnant and normal pregnant tissue. Although we are unable to resolve the difference between these studies without renin protein determinations, one consideration is that the normal pregnant tissues also contained myometrium which may be diluting renin mRNA from the decidua. However, a significantly higher renin mRNA was found in preeclamptic uterine placental bed tissues vs. normal pregnant tissues. The elevated renin in the preeclamptic uterus is consistent with the higher Ang II in the uterine placental bed.

ACE mRNA was lower in normal pregnant uterine placental bed tissue vs. nonpregnant uterine fundus. A previous study using immunohistochemical and Western blots for protein expression in the nonpregnant endometrium showed cyclic changes in ACE based on the menstrual cycle (26). ACE immunohistochemistry showed expression in the glandular epithelium throughout the menstrual cycle. ACE expression has been localized to the spiral artery endothelial cells and perivascular decidual stromal cells during the first trimester of pregnancy (27). Herse et al. (24) showed no differences in decidual or placental ACE mRNA between normal pregnant and preeclamptic subjects, which agrees with our finding in the placenta (14). However, ACE expression was elevated in preeclamptic uterine placental bed when compared with normal pregnant uterine placental bed, potentially contributing to the increased Ang II in the uterus of these subjects.

The gene expression of ACE2, an enzyme involved in Ang-(1-7) formation, was similar in the uterine bed from all three groups. This is the first demonstration of ACE2 by immunohistochemistry and real-time RT-PCR in nonpregnant, normal pregnant, and preeclamptic uterine samples. ACE2 was colocalized with Ang-(1-7) in the glandular epithelial cells of the endometrium in the nonpregnant uterus and in the invading trophoblasts as well as the trophoblast cells surrounding the uterine spiral arteries during pregnancy. The presence of ACE2 in these cells indicates that Ang-(1-7) may be playing a role in trophoblast invasion and uterine spiral artery remodeling during pregnancy. Ang-(1-7) and ACE2 staining was detected in the decidual cells of the uterine placental bed in both normal and preeclamptic pregnancies. The fact that no differences in ACE2 mRNA occurred between normal pregnant and preeclamptic uterine bed tissues may account for similar Ang-(1-7) levels.

Uterine NEP mRNA was significantly higher during normal pregnancy when compared with the nonpregnant uterus. This up-regulation of NEP in the uterus during pregnancy is consistent with studies showing that estrogen up-regulates uterine NEP mRNA and activity in normotensive and hypertensive rats (28). NEP is an enzyme with the potential to convert Ang I or Ang-(1–9) to Ang-(1-7). Because Ang-(1-7) and Ang I peptide levels were unchanged in normal and preeclamptic pregnancy, the contribution of NEP to Ang-(1-7) formation in the uterus during pregnancy awaits metabolism studies using specific NEP inhibitors, together with activity and protein measurements.

All three major peptides of the RAS are present in the uterine fundus from nonpregnant women and in the placental bed from normal pregnant and preeclamptic women. A previous study described Ang II in the endometrium of nonpregnant women throughout the menstrual cycle. Specifically, Ang II immunoreactivity was found in the glandular epithelium and the stroma during the proliferative phase and around the spiral arterioles in the secretory phase (29). Naruse et al. (30) found high levels of Ang II measured by RIA in the nonpregnant uterus from women undergoing hysterectomy, conditions similar to the nonpregnant subjects in our study. They found that the uterus had much higher levels than other tissues, including the adrenal. Although in both studies care was taken in obtaining normal tissue, neither study could eliminate the possibility of an influence on Ang II resulting from the adjacent abnormal tissue. Based on comparison with nonpregnant tissue, Ang II levels in the normal pregnant uterus were significantly down-regulated. However, with preeclampsia, Ang II levels were higher in the uterus. In addition, we found no significant effect of maternal age on Ang II expression in the uterus of nonpregnant vs. pregnant women.

In addition to Ang II, Ang-(1-7) was also present at detectable levels by both RIA and immunohistochemistry. Ang-(1-7) was the predominant peptide in the uterine placental bed from normal and preeclamptic pregnancies. The presence of both Ang-(1-7) in the invading and intraarterial trophoblast cells in the uterine placental bed suggests that it may be playing a role in the regulation of trophoblast invasion of the maternal uterine spiral arteries. Because Ang II has also been found surrounding the spiral arteries, Ang-(1-7) may act to counterbalance the vasoconstrictor properties of Ang II during normal pregnancy. Previous studies have measured AT1 and AT2 receptors in the uterus of nonpregnant and pregnant women (29); however, not many have focused specifically on uterine placental bed tissues, nor have they investigated differences between normal pregnant and preeclamptic women. This is the first study to report the presence of the Mas/AT1-7 receptor in the human uterus.

The mRNA for all three angiotensin receptors was present in nonpregnant uterine tissues that contained both endometrium and myometrium. We observed significant [125I]SarThran binding in the myometrium and low-density binding in the endometrium of the nonpregnant uterus. In agreement with the presence of mRNA for all three receptors, all three selective antagonists competed for binding using in vitro receptor autoradiography of the fundus from nonpregnant women. In the myometrium, the AT2 antagonist competed for 100% of the specific binding. Previous studies by others have shown that AT2 receptors are significantly higher than AT1 receptors in the nonpregnant uterine myometrium (31,32,33). Cox et al. (31) reported that the AT2 receptor antagonist competed for 90% of the total angiotensin receptor binding in the myometrium of the nonpregnant uterus. The remaining sites in their studies were competed for by losartan. In contrast, the presence of only AT2 receptors and not AT1 receptors in the myometrium of the nonpregnant uterus has also been reported (34). Although each of the above studies is consistent with AT2 receptors as the predominant subtype in the myometrium, there appears to be controversy over the presence of AT1 receptors in this tissue. Unexpectedly, in our study, there was significant (40–60%) competition with losartan and d-ala in the myometrium, leading to greater than 100% of specific binding, and combinations of the antagonists did not show additive effects. Because specific binding is quite high in this tissue (>85%), competition for nonspecific binding by one of the antagonists cannot account for the results. Thus, the overlap in recognition of these antagonists within the [125I]SarThran binding could indicate a loss of selectivity of the antagonists for their respective angiotensin receptors. In our previous study using these same techniques, the chorionic villi of the placenta expressed predominately AT1 receptors with losartan competing for the majority of [125I]SarThran binding and the expected additive total of 100%. Therefore, it appears unlikely that the antagonists, specifically PD123,319 or d-ala, are inherently not selective under the conditions used. Instead, the data argue for differences in the pharmacology of the receptors in the myometrium. In fact, the presence of higher levels of mRNA for both the AT1-7 receptor and the AT2 receptor compared with the chorionic villi of the placenta, as is seen in the myometrium of the uterus, suggests that the presence of these receptor subtypes rather than the AT1 receptor may account for the relative nonselectivity of the competitors. The discrepancies could result from methodological differences. We performed in vitro receptor autoradiography that measures total angiotensin receptor binding in the tissue including the nuclear membrane, cytoplasm, and cell membrane vs. the ligand binding assay done by Saridogan et al. (34), which measures only angiotensin receptors in the plasma membranes. However, in our study, there was also significant competition by d-ala. No previous studies using this antagonist have been done to investigate the presence of AT1-7 receptors in the nonpregnant uterus. However, studies by Ahmed et al. (29) identified a novel non-AT1/non-AT2 receptor that was insensitive to both losartan and PD123,177 in the endometrium of the nonpregnant uterus that accounted for 16% of angiotensin receptor-specific total binding. Therefore, additional studies are warranted using full competition curves to fully understand the pharmacology of the angiotensin receptors in this tissue.

Several studies have identified functions of the AT2 receptor including inhibition of cell growth and DNA synthesis (35), neuronal cell differentiation and nerve regeneration (36), and cGMP production (37). In the uterus, the functional actions of the AT2 receptor are largely unknown. Although Ang II functions have been hypothesized to be attributed to uterine contraction (34), decidualization (38), and the regeneration of the endothelium after menstruation (29), these actions are all thought to be mediated by the AT1 receptor and not the AT2 receptor. Therefore, additional studies are warranted to understand the function of the AT2 receptor within the nonpregnant uterus.

The lower AT1, AT2, and Mas/AT1-7 receptor mRNA and angiotensin receptor density observed during pregnancy agrees with previous studies that have shown down-regulation of Ang receptors in the myometrium of the pregnant uterus. Total angiotensin receptor binding density was 92% lower in pregnancy including 97% lower AT2 receptors (31). In addition, a 50-fold lower Ang II receptor density is reported in the myometrium during pregnancy that was attributed entirely to the fall in AT2 receptors (32). Some have accredited the lower AT2 receptors to a negative feedback loop based on the increase in circulating Ang II levels seen during pregnancy. Another possible reason for a down-regulation of the AT2 receptor is hormonal regulation during pregnancy by the sex steroid hormones including estrogen and progesterone (33). No differences in angiotensin receptor binding density or mRNA were seen between uterine placental bed tissues in normal pregnancy and preeclampsia. Despite the higher Ang II peptide levels in the uterine placental bed from preeclamptic women, the decrease in all three angiotensin receptors in tissues from normal and preeclamptic subjects would argue for a loss of function of angiotensin peptides in the uterine placental bed in pregnancy. Previously, we showed the presence of a local tissue-specific RAS in the placental chorionic villi where Ang II peptide levels and angiotensinogen and AT1 receptor mRNAs were increased in preeclamptic tissue (14). The findings from these two studies would suggest that during normal and preeclamptic pregnancy, uterine Ang II would continue to act in an endocrine manner at receptors in the adjacent placenta where maternal-fetal nutrient and oxygen exchange is regulated.

In addition, recent preeclampsia research has focused on the presence and function of AT1 receptor autoantibodies that have been found in the serum of preeclamptic women (39). AT1 receptor autoantibodies act as agonists of the AT1 receptor, therefore mimicking the actions of Ang II. Although these autoantibodies may play a role in the pathophysiology of preeclampsia, it would seem unlikely that they would be having any effect in the preeclamptic uterus because in this study, we found that the AT1 receptor is significantly down-regulated in the preeclamptic uterine placental bed. Therefore, the contribution of AT1 autoantibodies to the development of preeclampsia may be limited to the fetal placenta.

In summary, we provide evidence for the presence of a local tissue-specific RAS in the uterine fundus from nonpregnant women and the uterine placental bed from normal pregnant and preeclamptic women. A local paracrine RAS is present in the nonpregnant uterus that is characterized by markedly higher levels of Ang II and predominately AT2 receptors. During pregnancy, Ang II peptide levels and angiotensinogen, ACE, AT1, AT2, and Mas receptor mRNA expression along with total angiotensin receptor binding are all significantly lower, indicating that the uterine placental bed RAS is down-regulated during normal pregnancy. In women with preeclampsia, Ang II peptide levels and renin and ACE mRNA are significantly higher, providing potential for an activated uterine placental bed RAS. However, the substantially lower levels of all three angiotensin receptor mRNAs along with lower angiotensin receptor binding during both normal and preeclamptic pregnancy would indicate that although Ang II is increased, the receptors needed to mediate its actions are not detectable in this local environment. Therefore, we propose that angiotensin peptides from the uterine placental bed are acting in an endocrine manner in the fetal placental chorionic villi, an essential part of the placenta that is responsible for maternal-fetal nutrient and oxygen exchange, where we have previously shown a significant upregulation in AT1 receptors (14).

Acknowledgments

We gratefully acknowledge the technical support of the Hypertension and Vascular Research Center Biochemistry and Molecular Biology Core Laboratories.

Footnotes

This work was supported by grants from the National Institutes of Health, NHLBI-P01 HL51952, and HL070130. L.A was supported in part by a predoctoral grant awarded by the Mid-Atlantic American Heart Association (AHA0515221U). We gratefully acknowledge grant support in part provided by Unifi, Inc., Greensboro, NC, Farley-Hudson Foundation, Jacksonville, NC, and Fondo Nacional de Desarollo Científico y Tecnológico.

Disclosure Summary: The authors have nothing to disclose.

First Published Online June 11, 2009

Abbreviations: ACE2, Angiotensin-converting enzyme 2; Ang, angiotensin; NEP, neprilysin; RAS, renin angiotensin system; [125I]SarThran, [125I](sarcosine1, threonine8 Ang II.

References

  1. The Working Group on High Blood Pressure in Pregnancy 1991 National High Blood Pressure Education Program (NHPEP). Working group report on high blood pressure in pregnancy. Pub. No. 91–3029. Washington, DC: U.S. Department of Health and Human Services; 1–46 [Google Scholar]
  2. Brosens I, Robertson WB, Dixon HG 1967 The physiological response of the vessels of the placental bed to normal pregnancy. J Pathol Bacteriol 93:569–579 [DOI] [PubMed] [Google Scholar]
  3. Fisher SJ, Roberts JM 1999 Defects in placentation and placental perfusion. In: Linheimer M, Roberts JM, Cunningham FG, eds. Chesley’s hypertensive disorders in pregnancy. 2nd ed. Stanford, CA: Appleton, Lange; 377–394 [Google Scholar]
  4. Granger JP, Alexander BT, Llinas MT, Bennett WA, Khalil RA 2002 Pathophysiology of preeclampsia: linking placental ischemia/hypoxia with microvascular dysfunction. Microcirculation 9:147–160 [DOI] [PubMed] [Google Scholar]
  5. Taylor RN, Roberts JM 1999 Endothelial cell dysfunction. In: Linheimer M, Roberts JM, Cunningham FG, eds. Chesley’s hypertensive disorders in pregnancy. 2nd ed. Stanford, CA: Appleton, Lange; 395–429 [Google Scholar]
  6. Hsueh WA, Luetscher JA, Carlson EJ, Grislis G, Fraze E, McHargue A 1982 Changes in active and inactive renin throughout pregnancy. J Clin Endocrinol Metab 54:1010–1016 [DOI] [PubMed] [Google Scholar]
  7. August P, Lenz T, Ales KL, Druzin ML, Edersheim TG, Hutson JM, Müller FB, Laragh JH, Sealey JE 1990 Longitudinal study of the renin-angiotensin-aldosterone system in hypertensive pregnant women: deviations related to the development of superimposed preeclampsia. Am J Obstet Gynecol 163:1612–1621 [DOI] [PubMed] [Google Scholar]
  8. Brown MA, Wang J, Whitworth JA 1997 The renin-angiotensin-aldosterone system in pre-eclampsia. Clin Exp Hypertens 19:713–726 [DOI] [PubMed] [Google Scholar]
  9. Baker PN, Broughton Pipkin F, Symonds EM 1990 Platelet angiotensin II binding and plasma renin concentration, plasma renin substrate and plasma angiotensin II in human pregnancy. Clin Sci 79:403–408 [DOI] [PubMed] [Google Scholar]
  10. Baker PN, Broughton Pipkin F, Symonds EM 1992 Comparative study of platelet angiotensin II binding and the angiotensin II sensitivity test as predictors of pregnancy-induced hypertension. Clin Sci 83:89–95 [DOI] [PubMed] [Google Scholar]
  11. Gant NF, Daley GL, Chand S, Whalley PJ, MacDonald PC 1973 A study of angiotensin II pressor response throughout primigravid pregnancy. J Clin Invest 52:2682–2689 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chesley LC, Talledo E, Bohler CS, Zuspan FP 1965 Vascular reactivity to angiotensin II and norepinephrine in pregnant and nonpregnant women. Am J Obstet Gynecol 91:837–842 [DOI] [PubMed] [Google Scholar]
  13. Merrill DC, Karoly M, Chen K, Ferrario CM, Brosnihan KB 2002 Angiotensin-(1-7) in normal and preeclamptic pregnancy. Endocrine 18:239–245 [DOI] [PubMed] [Google Scholar]
  14. Anton L, Merrill DC, Neves LA, Stovall K, Gallagher PE, Diz DI, Moorefield C, Gruver C, Ferrario CM, Brosnihan KB 2008 Activation of local chorionic villi angiotensin II levels but not angiotensin (1-7) in preeclampsia. Hypertension 51:1066–1072 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Allred AJ, Chappell MC, Ferrario CM, Diz DI 2000 Differential actions of renal ischemic injury on the intrarenal angiotensin system. Am J Physiol Renal Physiol 279:F636–F645 [DOI] [PubMed] [Google Scholar]
  16. Chappell MC, Diz DI, Jacobsen DW 1992 Pharmacological characterization of angiotensin II binding sites in the canine pancreas. Peptides 13:313–318 [DOI] [PubMed] [Google Scholar]
  17. Diz DI, Barnes KL, Ferrario CM 1986 Contribution of the vagus nerve to angiotensin II binding sites in the canine medulla. Brain Res Bull 17:497–505 [DOI] [PubMed] [Google Scholar]
  18. Ferrario CM, Averill DB, Brosnihan KB, Chappell MC, Iskandar SS, Dean RH, Diz DI 2002 Vasopeptidase inhibition and Ang-(1-7) in the spontaneously hypertensive rat. Kidney Int 62:1349–1357 [DOI] [PubMed] [Google Scholar]
  19. Moriguchi A, Tallant EA, Matsumura K, Reilly TM, Walton H, Ganten D, Ferrario CM 1995 Opposing actions of angiotensin-(1-7) and angiotensin II in the brain of transgenic hypertensive rats. Hypertension 25:1260–1265 [DOI] [PubMed] [Google Scholar]
  20. Krob HA, Vinsant SL, Ferrario CM, Friedman DP 1998 Angiotensin-(1-7) immunoreactivity in the hypothalamus of the (mRen-2d)27 transgenic rat. Brain Res 798:36–45 [DOI] [PubMed] [Google Scholar]
  21. Averill DB, Ishiyama Y, Chappell MC, Ferrario CM 2003 Cardiac angiotensin-(1-7) in ischemic cardiomyopathy. Circulation 108:2141–2146 [DOI] [PubMed] [Google Scholar]
  22. Crackower MA, Sarao R, Oudit GY, Yagil C, Kozieradzki I, Scanga SE, Oliveira-dos-Santos AJ, da Costa J, Zhang L, Pei Y, Scholey J, Ferrario CM, Manoukian AS, Chappell MC, Backx PH, Yagil Y, Penninger JM 2002 Angiotensin-converting enzyme 2 is an essential regulator of heart function. Nature 417:822–828 [DOI] [PubMed] [Google Scholar]
  23. Li C, Ansari R, Yu Z, Shah D 2000 Definitive molecular evidence of renin-angiotensin system in human uterine decidual cells. Hypertension 36:159–164 [DOI] [PubMed] [Google Scholar]
  24. Herse F, Dechend R, Harsem NK, Wallukat G, Janke J, Qadri F, Hering L, Muller DN, Luft FC, Staff AC 2007 Dysregulation of the circulating and tissue-based renin-angiotensin system in preeclampsia. Hypertension 49:604–611 [DOI] [PubMed] [Google Scholar]
  25. Shaw KJ, Do YS, Kjos S, Anderson PW, Shinagawa T, Dubeau L, Hsueh WA 1989 Human decidua is a major source of renin. J Clin Invest 83:2085–2092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Li XF, Ahmed A 1997 Compartmentalization and cyclic variation of immunoreactivity of renin and angiotensin converting enzyme in human endometrium throughout the menstrual cycle. Hum Reprod 12:2804–2809 [DOI] [PubMed] [Google Scholar]
  27. Morgan T, Craven C, Ward K 1996 Human spiral artery renin-angiotensin system. Br J Obstet Gynecol 103:879–886 [DOI] [PubMed] [Google Scholar]
  28. Neves LAA, Chappell MC, Ferrario CM, Gallagher PE, Ganten D, Brosnihan KB 2006 Effect of estrogen on neprilysin expression in uterus and kidney of hypertensive and normotensive rats. Peptides 27:2912–2918 [DOI] [PubMed] [Google Scholar]
  29. Ahmed A, Li XF, Shams M, Gregory J, Rollason T, Barnes NM, Newton JR 1995 Localization of the angiotensin II and its receptor subtype expression in human endometrium and identification of a novel high-affinity angiotensin II binding site. J Clin Invest 96:848–857 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Naruse M, Naruse K, Kurimoto F, Sakurai H, Yoshida S, Toma H, Ishii T, Obana K, Demura H, Inagami T, Shizume K 1985 Evidence for the existence of des-Asp1-angiotensin II in human uterine and adrenal tissues. J Clin Endocrinol Metab 61:480–483 [DOI] [PubMed] [Google Scholar]
  31. Cox BE, Word RA, Rosenfeld CR 1996 Angiotensin II receptor characteristics and subtype expression in uterine arteries and myometrium during pregnancy. J Clin Endocrinol Metab 81:49–58 [DOI] [PubMed] [Google Scholar]
  32. de Gasparo M, Whitebread S, Kalenga MK, de Hertogh R, Crevoisier P, Thomas K 1994 Down regulation of the angiotensin II receptor subtype AT2 in human myometrium during pregnancy. Regul Pept 53:39–45 [DOI] [PubMed] [Google Scholar]
  33. Matsumoto T, Sagawa N, Mukoyama M, Tanaka I, Itoh H, Goto M, Itoh H, Horiuchi M, Dzau VJ, Mori T, Nakao K 1996 Type 2 angiotensin II receptor is expressed in human myometrium and uterine leiomyoma and is down-regulated during pregnancy. J Clin Endocrinol Metab 81:4366–4372 [DOI] [PubMed] [Google Scholar]
  34. Saridogan E, Djahanbakhch O, Puddefoot JR, Demetroulis C, Dawda R, Hall AJ, Vinson GP 1996 Type 1 angiotensin II receptors in human endometrium. Mol Hum Reprod 2:659–664 [DOI] [PubMed] [Google Scholar]
  35. Stoll M, Steckelings UM, Paul M, Bottari SP, Metzger R, Unger T 1995 The angiotensin AT2-receptor mediates inhibition of cell proliferation in coronary endothelial cells. J Clin Invest 95:651–657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Meffert S, Stoll M, Steckelings UM, Bottari SP, Unger T 1996 The angiotensin II AT2 receptor inhibits proliferation and promotes differentiation in PC12W cells. Mol Cell Endocrinol 122:59–67 [DOI] [PubMed] [Google Scholar]
  37. Siragy HM, Carey RM 1996 The subtype-2 (AT2) angiotensin receptor regulates renal cyclic guanosine 3′, 5′-monophosphate and AT1 receptor-mediated prostaglandin E2 production in conscious rats. J Clin Invest 97:1978–1982 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Squires PM, Kennedy TG 1992 Evidence for a role for a uterine renin-angiotensin system in decidualization in rats. J Reprod Fertil 95:791–802 [DOI] [PubMed] [Google Scholar]
  39. Dechend R, Viedt C, Müller DN, Ugele B, Brandes RP, Wallukat G, Park JK, Janke J, Barta P, Theuer J, Fiebeler A, Homuth V, Dietz R, Haller H, Kreuzer J, Luft FC 2003 AT1 receptor agonistic antibodies from preeclamptic patients stimulate NADPH oxidase. Circulation 107:1632–1639 [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES