The Functional Role of the Renin-Angiotensin System in Pregnancy and Preeclampsia

Abstract

During normal pregnancy, the renin-angiotensin system (RAS) plays a vitally important role in salt balance and subsequent well-being of mother and fetus. In this balance, one must consider not only the classical renal RAS but also that of the uteroplacental unit, where both maternal and fetal tissues contribute to the signaling cascade. Many studies have shown that in normal pregnancy there is an increase in almost all of the components of the RAS. In derangements of pregnancy this delicate equilibrium can become unbalanced. Preeclampsia is one such case. It is a disorder of pregnancy characterized by hypertension, proteinuria and placental abnormalities associated with shallow trophoblast invasion and impaired spiral artery remodeling. Despite being a leading cause of maternal death and a major contributor to maternal and perinatal morbidity, the mechanisms responsible for the pathogenesis of preeclampsia are poorly understood. Immunological mechanisms and the RAS have been long considered to be involved in the development of preeclampsia. Numerous recent studies demonstrate the presence of the angiotensin II type I receptor agonistic autoantibody (AT₁-AA). This autoantibody can induce many key features of the disorder and upregulate molecules involved in the pathogenesis of preeclampsia. Here we review the functional role of the RAS during pregnancy and the impact of AT₁-AA on preeclampsia.

Introduction of the classical RAS pathway

The circulating renin-angiotensin system, herein RAS, is a signaling cascade that plays a key role in regulating blood pressure and electrolyte balance. It is classically described in the kidney. The enzyme renin is synthesized and released by juxtaglomerular cells of the afferent renal arterioles in response to low blood pressure and low circulating sodium chloride. Renin release is mediated in part by prostaglandins produced by cells of the kidney’s macula densa [1]. Renin enzymatically cleaves angiotensinogen, which is made in the liver, to angiotensin-1 (ANG I), a ten amino acid peptide. This is the rate-limiting step of the RAS cascade (Figure 1). ANG I is not biologically functional and is cleaved by angiotensin-converting enzyme (ACE), made primarily in lung endothelium, to the biologically active, eight amino acid effector molecule, angiotensin-II (ANG II).

Figure 1. Classic Renin-Angiotensin System Cascade.

ANG II, the key effector molecule of the RAS and potent vasoconstrictor, acts through AT₁ receptors to increase blood pressure. AT₁ receptors are found on many cell types. Abbreviations - ADH: Antidiuretic hormone, ANG II: Angiotensin II, AT₁ receptor: angiotensin-II type I receptor.

There are two major types of angiotensin receptors: AT₁ and AT₂. They belong to the seven transmembrane G-protein-coupled receptor family. They have thirty-four percent sequence dentity and have similar affinities for ANG II [2]. Most of the effects of ANG II are mediated through activation of AT₁ receptors which are expressed on the surface of vascular smooth muscle cells and adrenal glands, among others. The AT₁ receptor is coupled to the G_q protein, that functions in a signaling pathway to increase intracellular calcium. Its activation promotes vasoconstriction, sympathetic activity and aldosterone release. The AT₂ receptor is highly expressed in the fetal kidney and its expression decreases during the neonatal period [3]. In the adult kidney AT₂ is much less abundant than AT₁ [2] AT₂ stimulation can inhibit cell growth, increase apoptosis, cause vasodilation and is involved in fetal tissue development [4].

It should be noted that ACE, made by endothelial cells and others such as smooth muscle cells, is not the only enzyme that can generate ANG II from ANG I. Chymase, a chymotrypsin-like serine protease, is a non-ACE angiotensin generating enzyme that is produced by villous syncytiotrophoblasts [5]. Chymase is also found in great quantities in mast cells, as well as in the skin, heart and arteries and is a major contributor to the pool of ANG II found in these tissues [6,7].

A local RAS is present in the placenta

In addition to the classical view of the RAS there is accumulating evidence indicating components of the renin-angiotensin system are synthesized in many tissues, such as the brain, heart, ovary, and placenta [8,9].

One of the major extra-renal RAS during pregnancy is in the placenta. As early as 1967, Hodari et al. described a placental RAS and identified a renin-like substance in human placental tissue [10]. Renin expression in cultured chorionic cells was first reported by Symonds in 1968 [11]. Since then, pro-renin, angiotensinogen, ACE, ANG I and ANG II have all been identified in fetal placental tissues. AT₁ receptor expression in fetal placental vasculature has also been shown [12]. Many other experiments using first-trimester human decidua show expression of renin, angiotensinogen, ACE and AT₁ receptors [13]. More recent studies using human third trimester decidual cells also indicate the presence of angiotensinogen and renin [14]. Localization studies around the decidual spiral arteries show expression of angiotensinogen, renin, ACE and AT₁ receptors [15]. Thus, in the gravid woman, the maternal decidua and the fetal placental tissues each contain all the necessary components for a functional RAS.

Regulation of the RAS during pregnancy

In humans, the RAS undergoes major changes in response to pregnancy. There is an early increase in renin due to extra-renal local release by the ovaries and maternal decidua [16]. Angiotensinogen synthesis by the liver is increased by circulating estrogen produced by the growing placenta. This leads to increased serum ANG II and aldosterone levels [17]. ACE is the only RAS component that decreases during pregnancy [18]. Table 1 compares the levels of serum RAS components between non-pregnant women and normotensive pregnant women.

Table 1. Comparison of serum RAS components in normotensive and preeclamptic pregnancies versus non-pregnant women.

Overall, circulating levels of the RAS are increased in normal pregnancies versus the non-pregnant state. Though there are slight increases in the RAS serum components in preeclamptic women, they are decreased as compared to normotensive pregnant women.

Serum RAS Component	Normotensive Pregnancy	Preeclamptic Pregnancy	References
Renin	++	+	Hsueh [16], Langer [101]
Angiotensin I	++	+	Merrill [18], Langer [101]
ACE	−	+	Merrill [18], Langer [101]
Aldosterone	++	+	Brown [102], Langer [101]
ANG-(1–7)	++	−	Merrill [18]
Angiotensin II	++	+	Langer [101]
Angiotensin II sensitivity	−−(system refractory)	++(system highly sensitive)	Gant [20]
AT1-AA	Absent	Present	Wallukat [44]
AT1 receptor	Baseline production in maternal decidua	Upregulation in maternal decidua	Herse [35]

Legend: ++ Greatly increased over non-pregnant

+ Slightly increased over non-pregnant

− Decreased compared to non-pregnant

Although ANG II levels increase during pregnancy, normotensive pregnant women are actually refractory to its vasopressor effects. A historic study performed by Assali et al. showed that the pregnant woman requires twice as much ANG II intravenous infusion over a non-pregnant woman to achieve the same vasomotor response [19]. This is thought to be due to the presence of increased progesterone and prostacyclins which can decrease ANG II sensitivity [20]. Additionally in normal pregnancy, AT₁ receptors are monomeric and are inactivated by reactive oxygen species (ROS). This is in comparison to the heterodimeric state seen in ANG II sensitive conditions [21]. These facts help explain why a greater ANG II stimulus is required in order to achieve the appropriate vasomotor response in normotensive pregnancies.

Trophoblasts are rich in AT₁ receptors and thus are responsive to the changes in ANG II concentrations that occur during pregnancy. Recent studies demonstrate that multiple genes are regulated by AT₁ receptor signaling and include those encoding secreted proteins associated with trophoblast invasion (e.g., plasminogen activator inhibitor-1, PAI-I) and angiogenesis (soluble fms-like tyrosine receptor-1, sFlt-1). ANG II signaling also activates NF-kappa B and stimulates NADPH-oxidase synthesis by trophoblasts [22]. ANG II decreases system A amino acid transporter activity in human placental villous fragments through AT₁ receptor activation, a feature believed to contribute to IUGR in some cases [23]. In addition to its regulation of specific gene expression, the RAS is believed to play a critical regulatory role in feto-placental circulation, facilitating adequate placental blood flow for fetal oxygenation and maturation. Recent studies suggest that decidual tissue serves as a source of ANG II production and trophoblasts serve as paracrine targets of ANG II signaling through AT₁ receptor activation.

The results obtained with human tissue have been corroborated and extended with studies in mice. Takimoto et al. mated transgenic male mice carrying the human renin gene with female transgenic carrying the human angiotensinogen gene [24]. They showed the activation of renin gene expression in trophoblast cells late in pregnancy and that human renin is released by the placenta into the maternal circulation [24]. Xia et al. used two murine models to study renin gene expression during pregnancy. In ICR mice, high levels of renin gene expression occur at the maternal-fetus interface, first in the maternal decidua and then in the fetal placenta [25]. While ICR mice have two related renin genes, Ren1 and Ren2, C57Bl/6 mice have only one renin gene, Ren1. In these pregnant mice, minimal renin gene expression was observed in placentas but instead was upregulated in kidneys. Though in both ICR and C57Bl/6 mice there is an increase in renin in maternal circulation during pregnancy, they differ with regard to gestation-induced sites of increased renin gene expression.

Taken together, both human and animal studies indicate that the RAS undergoes specific and necessary changes during normal pregnancy.

Preeclampsia is characterized by significant alterations in the RAS

Preeclampsia is a pregnancy-specific syndrome of hypertension and proteinuria resulting in substantial maternal and neonatal morbidity and mortality. The condition is also characterized by placental abnormalities, such as decreased invasion by extravillous trophoblasts into maternal spiral artery endothelium. In advanced stages the clinical symptoms may include cerebral edema, renal failure and the HELLP (Hemolysis, Elevated Liver enzymes and Low Platelets) syndrome. The clinical management of preeclampsia is hampered by the lack of reliable diagnostic tests and effective therapy. Although the underlying pathogenic mechanisms of the disorder are not well understood, preeclampsia is largely believed to be associated with uteroplacental ischemia and the subsequent release of toxic factors from the placenta into the maternal circulation. Roberts and colleagues were among the first to propose that alterations in endothelial cell function by activating agents produced by the placenta initiate the clinical syndrome of preeclampsia [26]. This serious condition affects approximately 7% of pregnancies and is thus a major health concern [27].

Changes in the circulating RAS in preeclampsia

Several features of the RAS in preeclampsia differ from the normal pregnant state. Except for ACE, RAS components in the circulation increase in uncomplicated pregnancy. A study by Merrill et al. demonstrated that this is not the case in preeclamptic women. In preeclamptic women the circulating levels of renin, ANG-1 and aldosterone are lower than their normotensive counterparts (Table 1). There are two exceptions to the decreases observed. The ACE level is approximately the same and ANG-(1–7), a vasodilatory member of the RAS, is significantly reduced in preeclampsia [18]. The exact role ANG-(1–7) plays in uteroplacental blood flow is still uncertain. It is produced throughout the body by many tissues such as kidney, heart, hypothalamus and ovary. This peptide can be derived from ANG I independent of ACE or synthesized from ANG II by removal of the C-terminal phenylalanine by several enzymes, such as ACE-2, prolylendopeptidase and prolylcarboxypeptidase [28]. ANG-(1–7) not only interacts with AT₁ and AT₂ receptors, but there is growing evidence that it can act through its own specific receptor [29,30].

While normotensive pregnant women demonstrate decreased vascular sensitivity to ANG II, preeclamptic women exhibit increased sensitivity of the adrenal cortex and vascular system to ANG II [20,31]. This phenomenon could be due to heterodimerization of the AT₁ receptor. During normal pregnancy monomeric AT₁ receptors are inactivated by ROS leading to lower ANG II sensitivity [21]. In preeclampsia, however, the AT₁ receptor is found in the heterodimeric form with the bradykinin receptor (B2) [21,32]. These AT₁/B2 heterodimers show resistance to ROS-inactivation and remain active and hyper-responsive to ANG II [21,33,34]. Monitoring the presence and activity of these heterodimeric receptors as preeclamptic symptoms subside postpartum will be of future interest.

Changes in the local uteroplacental RAS in preeclampsia

The changes observed in the uteroplacental RAS are different from those in the circulation in preeclampsia. Recent studies by Herse et al. demonstrate that the only change in placental or decidual RAS component expression in preeclampsia is the upregulation of the AT₁ receptor in the maternal decidua [35]. They did not observe an increase in renin production in the decidua of preeclamptic over normotensive placentas. This is in contrast to the earlier findings of Shah et al., who demonstrate an increase in renin expression in the decidua vera of preeclamptic women over those of normotensive pregnant women [36]. They believe that the maternal decidua acts as an additional site of RAS activation and that the small amount of ANG II produced locally finds its way into maternal circulation and is sufficient to down-regulate ANG II production in the kidney as seen in preeclampsia. Another recent study by Anton et al. shows an increase in ANG II, but no increase in ANG-(1–7) in the chorionic villi of preeclamptic placentas as compared to placentas of normotensive pregnancies [37]. The varied results of placental RAS studies indicate that further investigation of the RAS during preeclampsia and in placental tissues is necessary.

In addition, chymase, a non-ACE ANG II producing enzyme, is upregulated in trophoblast cells in the placentas of preeclamptic women as compared to those of normotensive women [5]. This serine protease, released by mast cells and smooth muscle cells, has known roles in hypertensive and inflammatory diseases [38,39]. Overexpression of human vascular chymase in transgenic mice leads to increased blood pressure, vasoconstriction and hypertensive arteriopathy, all features of preeclampsia [40]. Chymase also cleaves big endothelin-1 (ET-1) to the 31-amino-acid length endothelin ET-1(1–31) [41], a vasoconstrictor that is increased in the myometrium of preeclamptic women [42]. ET-1-(1–31) is an especially potent vasoconstrictor in the umbilical artery and could play an important role in fetal circulation and the observed intrauterine growth restriction and hypertension observed in preeclampsia [43]. Investigation into the exact role and regulation of chymase in this disease will be of great interest in the near future.

The Angiotensin II type I Agonistic Autoantibody (AT₁-AA)

It is a puzzling feature that circulating ANG II levels are decreased in preeclamptic women as compared to normotensive pregnant women [20,31]. Despite this fact, women suffering from preeclampsia exhibit symptoms, such as hypertension and renal damage, which could be attributed to an excess of ANG II or AT₁ receptor activation.

A major advance in our understanding of preeclampsia was made in 1999 by Wallukat et al. who reported that women with preeclampsia harbor an autoantibody that stimulates the AT₁ receptor [44] This AT₁ receptor agonistic autoantibody (AT₁-AA) represents a major intrusion into the normal functioning RAS. Using affinity purification and peptide competition experiments they showed that the autoantibody binds to a seven amino acid sequence present on the second extracellular loop of the AT₁ receptor.

Since its discovery, there has been extensive investigation into the contribution of AT₁-AA to the pathogenesis of preeclampsia. Studies by the group above, ours and others show that AT₁-AA bind to AT₁ receptors on a variety of cells, including trophoblasts, and increase factors attributed to the pathogenesis of preeclampsia. Examples of the regulatory role of AT₁-AA in preeclampsia are reviewed below and summarized in Figure 2.

Figure 2. AT₁-AA interact with AT₁ receptors on trophoblasts which synergistically act with ANG II to impair placentation.

Normal RAS function and AT₁ receptor activation is imperative for normal placental development. AT₁-AA, found in the serum of preeclamptic women, function as ANG II by activating AT₁ receptors to increase production of sFlt-1, PAI-1 and NADPH oxidase in trophoblast cells. The 7-AA peptide corresponds to a sequence on the second extracellular loop of the AT₁ receptor. AT₁-AA-mediated effects can be neutralized by 7-AA. AT₁-AA: Angiotensin II type I receptor agonistic autoantibody.

Pathologic role of AT₁-AA in abnormal placental development and the maternal syndrome of preeclampsia

Role of AT₁-AA in placental abnormalities

Placentas of preeclamptic women are often small, exhibit shallow trophoblast invasion, aberrant spiral artery remodeling and reduced uteroplacental blood flow. The triggering factor and mechanism of these changes have not been determined. There is a growing body of evidence indicating that AT₁-AA and its derangement of the RAS could influence these transformations and contribute to the pathogenesis of preeclampsia.

a) AT₁-AA induces excess sFlt-1 secretion and impaired angiogenesis

Recently soluble fms-like tyrosine kinase-1 (sFlt-1) has been brought to the forefront of factors playing a role in placental development and preeclampsia [45,46]. It is a splice variant of VEGFR-1: a secreted soluble form of the VEGFR-1 receptor lacking the transmembrane and cytoplasmic domains. sFlt-1 acts as an antagonist of angiogenesis by binding to free vascular endothelial growth factor (VEGF) and placental growth factor (PlGF) and inhibiting their angiogenic actions [47–49]. In preeclamptic placentas, which are reportedly hypoxic (Roberts 1989, 1992), there is an increase of sFlt-1 secretion two- to five-fold greater than in the placentas of normotensive pregnancies [50–52]. Though both sFlt-1 and VEGF are hypoxia-inducible genes, Nagamatsu et al. have shown that in the hypoxic induction of sFlt-1 in the placenta is cell-specific and that hypoxia/ischemia of cytotrophoblast cells leads to excess sFlt-1 production alone [53]. Khaliq et al. have shown that hypoxia decreases PlGF expression in BeWo choriocarcinoma cells [54]. Ahmed et al. have also shown that there is a diametric expression of VEGF and PlGF during pregnancy [55]. In preeclampsia, Lash et al. report a cytotrophoblastic increase in VEGF and decrease in PlGF [56]. While total VEGF expression may increase during pregnancy, when high sFlt-1 levels are also present, free VEGF decreases, resulting in less angiogenesis [57]. Therefore, the overabundance of sFlt-1 production by the cytotrophoblast leads to a decrease of free VEGF on a background of low PlGF, resulting in an overall anti-angiogenic state in the placenta. This angiogenic imbalance could result in the small, hypoxic placentas described in preeclampsia.

Caniggia and others hypothesize that early decidual hypoxia globally alters placental gene expression and retards trophoblast invasion necessary for a healthy placenta [58]. They suggest that hypoxia-inducible factor-1α (HIF-1α) is upregulated in preeclampsia and this leads to increases in transforming growth factor-β3 (TGF-β3) which limits trophoblast invasion into maternal spiral arteries and decidua thereby generating further hypoxic conditions [58–60]. This mechanism of early placental hypoxia may be one explanation of excessive sFlt-1 production and the alterations in angiogenesis seen in preeclampsia.

Hypoxia, however, is only one plausible hypothesis for the overproduction of sFlt-1 in preeclampsia. The autoantibody, AT₁-AA, may also play an important role. During normal pregnancy the placenta produces sFlt-1 through ANG II stimulation of trophoblast cells via the calcineurin-NFAT pathway [61]. Therefore overstimulation of the AT₁ receptor by the autoantibody could lead to excessive sFlt-1 production. In this regard, Zhou et al.. have shown that AT₁-AA purified from preeclamptic patient serum can not only induce sFlt-1 secretion in both human placental villous explants and human trophoblast cells [61,62], but also in a pregnant mouse model [61]. This leads to the possibility that alongside ANG II and local placental hypoxia, AT₁-AA can additively contribute to the excess sFlt-1 secretion reported in preeclamptic patients.

Oversecretion of sFlt-1 secondary to both AT₁-AA and placental hypoxia could potentiate a positive feed-forward cycle: increased sFlt-1 could lead to the overly inhibited angiogenesis and further placental hypoxia observed in preeclampsia, with subsequently more placental sFlt-1 production. It will be important in future studies to further clarify the local AT₁ receptor-mediated mechanisms of sFlt-1 oversecretion in the placenta.

b) AT₁-AA stimulates excess PAI-1 secretion and shallow trophoblast invasion

PAI-1 is another important factor to consider in the pathogenesis of preeclampsia for two reasons. First, in the placenta it plays a role in impaired trophoblast invasion. It does this by inhibiting urokinase-like plasminogen activator (uPA) resulting in decreased conversion of plasminogen to plasmin. This leads to decreased fibrinolysis, less extracellular matrix digestion and shallow trophoblast invasion, a hallmark of the placenta in preeclampsia. AT₁-AA activates AT₁ receptors on trophoblast cells resulting in elevated PAI-1 levels [63,64]. It has also been shown to decrease trophoblast invasion in vitro using a matrigel assay [63,65]. Thus, activation of AT₁ receptors by AT₁-AA on human trophoblasts may contribute to increased PAI-1 production and shallow trophoblast invasion.

c) AT₁-AA increases ROS production

Free radical species, or reactive oxygen species (ROS) are a normal by-product of aerobic respiration and regulate cellular functions through redox reactions [66]. However, when excess ROS are present, a cell’s natural anti-oxidant defenses are unable to overcome the overload of non-specific damage of cellular DNA, proteins and lipids. In this regard, during pregnancy, oxidative stress could lead directly to placental tissue damage. The teratogenic effects of ROS could be detrimental to the developing fetus, especially during the critical period of organogenesis [67,68]. The generation of reactive oxygen species (ROS) is increased in preeclampsia and could play a role in aberrant placentation [69]. Dechend et al. found that AT₁-AA increases intracellular ROS via NADPH oxidase in placental trophoblast cells as well as vascular smooth muscle cells [22]. They also showed that AT₁-AA markedly upregulates NFκB as a downstream target and confirmed that there is increased ROS production in preeclamptic placentas in and around the blood vessels. Finally, they demonstrated that NADPH oxidases are elevated in the placentas of preeclamptic women. The authors suggest that AT₁-AA could contribute to ROS production in the placenta through activation of NADPH oxidase and in the inflammatory responses associated with preeclampsia [22].

Collectively, AT₁-AA-mediated sFlt-1, PAI-1 and NADPH oxidase induction may contribute to the pathological changes observed in the placentas of preeclamptic patients.

The Maternal Syndrome of Preeclampsia

In addition to its potential contribution to placental abnormalities, AT₁-AA also plays an important role in the maternal features of preeclampsia and contributes to the endothelial cell dysfunction and vascular damage which also characterize the disease [26,70]. Intuitively, AT₁ receptor stimulation would lead to vasoconstriction and subsequent hypertension. Furthermore, AT₁-AA has been linked to renal abnormalities and increased hypercoagulation, two other clinical features associated with the disorder.

a) AT₁-AA increases PAI-1 production in the kidneys

Plasminogen activator inhibitor-1 (PAI-1) is a serine protease inhibitor which not only decreases the conversion of plasminogen to plasmin leading to decreased fibrinolysis and increased fibrosis, but also indirectly inhibits extracellular matrix breakdown via matrix metalloproteinases [71]. Mesangial cell PAI-1 production is partially controlled by the action of ANG II on AT₁ receptors [72,73]. Bobst et al. have shown that AT₁ receptors on cultured human mesangial cells can be activated by AT₁-AA and increase PAI-1 secretion [63]. This accumulation of PAI-1 and decrease of available plasmin could potentially contribute to the kidney damage via decreased extracellular matrix degradation and sub-endothelial and sub-epithelial fibrin deposition observed in preeclampsia [74,75]. Excess fibrin deposition in the glomeruli decreases the kidney’s filtration capability and could contribute to proteinuria [76,77]. By acting through the RAS, AT₁-AA induces increases of PAI-1 in both the placenta and kidney, leading to decreased fibrinolysis and extracellular matrix breakdown that could manifest in the symptoms seen in preeclampsia.

b) AT₁-AA and increases calcium mobilization

Preeclampsia is associated with abnormalities in calcium metabolism and increased intracellular calcium levels in erythrocytes, lymphocytes and platelets [78–80]. Haller et al. showed an elevation in basal intracellular free Ca²⁺ in the platelets of preeclamptic patients in comparison to those of normotensive pregnant women. Similarly, intracellular free Ca²⁺ concentration is increased in the lymphocytes and erythrocytes of preeclamptic patients [79,81]. Thway et al. tested the possibility that AT₁-AA could increase intracellular concentration of free Ca²⁺ and the downstream activation of Ca²⁺ signaling pathways via AT₁ receptor stimulation [82]. They found that IgG from preeclamptic patients activated AT₁ receptors and increased intracellular free Ca²⁺ whereas IgG from normotensive women was incapable of doing so. They also showed that this increase in intracellular Ca²⁺ lead to the activation of the NFAT transcription factor [82]. These results suggest that AT₁-AA may contribute to the increased intracellular free Ca²⁺ concentration and downstream signaling associated with the systemic disorder of preeclampsia.

c) AT₁-AA induces Tissue Factor production

Tissue factor (TF) is a transmembrane protein that initiates the extrinsic pathway of coagulation and is found in high levels in the placentas of preeclamptic women. TF overexpression could cause additional vascular damage in the placenta and could contribute to the hypercoaguability experienced by some women with severe preeclampsia. Dechend et al. observed that AT₁-AA induces TF expression in vascular smooth muscle cells [83] and Dorfell et al. showed similar findings in monocytes [84]. These studies indicate that AT₁-AA, by activating AT₁ receptor signaling, initiates increased TF expression. Therefore, AT₁-AA may contribute to the hypercoaguability associated with preeclampsia by stimulating TF expression in human vascular smooth muscle cells and monocytes.

d) AT₁-AA-mediated release of sFlt-1 and renal impairment

A prevailing view is that toxic factors secreted by the placenta contribute to the maternal syndrome of preeclampsia. One such factor is sFlt-1. AT₁-AA induces sFlt-1 levels in human placental villous explants, human trophoblast cells [61,62] and in pregnant mice [61]. Maynard et al. have shown that treatment of pregnant rats with sFlt-1 induces a preeclamptic like state: increased blood pressure, proteinuria and histopathologic changes in the renal glomeruli parallel to those lesions observed humans, such as glomerular endotheliosis, a hallmark of endothelial cell injury [52]. Thus, AT₁-AA induction of sFlt-1 could lead to the kidney damage observed in preeclampsia.

Overall, multiple studies show that many features of preeclampsia can be explained by the ability of AT₁–AA to activate AT₁ receptors on a variety of cells and provoke biological responses relevant to the pathophysiology of preeclampsia.

Long Term Consequences of AT₁-AA

Though the symptoms of preeclampsia usually abate within 48 hours after the delivery of the baby and placenta, it will be important to evaluate the persistence of AT₁-AA in these women. The recent meta-analysis by Bellamy et al. highlights the long-term cardiovascular risks for preeclamptic patients [85]. They report that women who have had preeclampsia are at higher risk for persistent hypertension and future cardiovascular injury. Additionally, it is known that the same AT₁-AA found in serum of women with preeclampsia associates with patients who suffer from malignant hypertension [86]. It is therefore reasonable to suggest an association between the persistence of the autoantibody and the long-term derangement of the RAS in women with preeclampsia.

Key animal models used to study the RAS and hypertensive disorders of pregnancy

It has become clear that preeclampsia is a vascular disease involving the interaction of multiple cell types, including trophoblasts, endothelial cells, vascular smooth muscle cells and others. To fully understand the interplay among these cells in response to alterations in the RAS it will be necessary to decipher the intercellular signaling pathways involved. This is a major reason why animal models, in which complex cellular interactions can be studied, will play an especially important role in examining this disorder. Table 2 briefly summarizes the recent animal models concerning the RAS, pregnancy and preeclampsia mentioned in this review.

Table 2. Recent animal models used to investigate the RAS, pregnancy and preeclampsia.

Brief description of the key recent animal models reviewed in this manuscript and their references.

Description of model	Animal	Findings	Reference
Mated male transgenic mice (overexpressing human renin) with female transgenic mice (overexpressing human angiotensinogen)	Mice	Pregnant female mice demonstrated: increased human rennin production by trophoblasts and renin release in circulation, transient hypertension that resolved upon delivery, proteinuria and placental abnormalities.	Takimoto E, 1996 [24].
Mated male transgenic mice (overexpressing human renin) with female transgenic mice (overexpressing human angiotensinogen) and lacking the AT1a receptor	Mice	Pregnant female mice exhibited no cardiac, renal or placental defects, indicating the important role of the AT1a receptor in a preeclamptic-like state in mice.	Saito T, 2004 [88].
Mated male transgenic rats (overexpressing human renin) with female transgenic rats (overexpressing human angiotensinogen)	Rats	Pregnant female rats demonstrated: transient hypertension that resolved upon delivery, proteinuria, glomerular fibrin deposition and placental vascular defects.	Dechend R, 2005 [89].
Evaluated renin expression in ICR and C57Bl/6 pregnant mice	Mice	Two murine models differ in their gestation-induced sites of renin expression. ICR mice show increased renin the placenta. C57Bl/6 mice show increased renin expression in the maternal kidneys.	Xia Y, 2002 [25].
Overexpression of human vascular chymase in transgenic mice	Mice	Chymase overexpression in mice led to hypertension, vasoconstriction and hypertensive arteriopathy.	Ju H. 2001 [40].
AT1–AA was injected into pregnant mice	Mice	Circulating maternal sFlt-1 increased in AT1-AA injected mice.	Zhou CC, 2008 [61].
sFlt-1 injected directly into pregnant rats	Rats	sFlt-1 injected pregnant rats developed increased blood pressure, proteinuria and glomerular endotheliosis.	Maynard SE, 2003 [52].
Pregnant rats (1) surgically treated to reduce uterine blood flow via RUPP (Reduced Uterine Perfusion Pressure) method or (2) infused with low-dose TNF-alpha	Rats	RUPP- and TNF-alpha treated rats demonstrated similar findings: increased blood pressure, proteinuria, circulating sFlt-1, as well as AT1–AA production.	Granger JP, 2006 [90]. Dechend R, 2006. [91]

Without genetic manipulation, rodents do not naturally and spontaneously develop preeclampsia. Despite this, many animal models have been used to elucidate the role of RAS in this disorder. Rodent and human placentas share similar forms and vascular structure: discoid and chorioallantoic, respectively [87]. They differ in that the mouse has a labyrinth-type hemotrichorial interdigitation whereas humans have a monochorial villous maternal-fetal interface. The RAS of the rodent and human are remarkably similar. The mouse has two isotypes of the AT₁ receptor, AT_1a and AT_1b, whereas humans only have one form of AT₁. In general, there is an upregulation of the components of RAS in normal pregnancy in both rodents and humans [25].

Despite differences between rodent and human placentas, extensive knowledge about the RAS and pregnancy has been gained using both mouse and rat models. Takimoto et al. showed that transient hypertension was induced in pregnant transgenic mice expressing human angiotensinogen that were mated with male transgenic mice expressing the human renin gene [24]. The blood pressure in these females increased late in pregnancy and resolved to normal levels after delivery. They also demonstrated glomerular enlargement coupled with increased proteinuria, myocardial hypertrophy as well as necrotic and edematous changes in their placentas. This group showed an increase in human renin mRNA in chorionic trophoblasts and an increase in placental-derived human renin in the maternal circulation of the pregnant transgenic mice. This implies that secreted placental factors could play a role in the pathogenesis of hypertensive disorders in pregnancy.

The same group went on to investigate the role of the angiotensin receptors in the transgenic mice mentioned above. Female mice expressing the human angiotensinogen gene, but lacking the AT_1a receptor, were mated to male mice expressing the human renin gene. During pregnancy, they surprisingly did not have an increase in blood pressure despite having intact AT_1b receptors [88]. The other features induced in transgenic mice with human angiotensinogen expression alone were not observed in mice lacking the AT_1a receptor. In fact, the AT_1a receptor deficient mice demonstrated no renal, cardiac or placental abnormalities. These findings illustrate the importance of AT_1a receptors in the development of hypertension and other histopathologic changes in pregnancy in the setting of a dysregulated RAS.

Key animal models involving the AT₁-AA and preeclampsia

The above experiments did not highlight the role of AT₁-AA in the development of hypertension in pregnancy. To address this issue, Dechend et al. mated female transgenic rats expressing the human angiotensinogen gene with male rats expressing the human renin gene [89]. Much like the mice of Takimoto et al., these rat dams demonstrated hypertension and proteinuria late in pregnancy that resolved upon delivery. They also developed fibrin deposition in their glomeruli and their placentas demonstrated vascular defects, such as atherosis-like lesions in the spiral arteries of the placental beds. AT₁-AA, the same autoantibody produced by women with preeclampsia, was found in their serum [89]. The production of AT₁-AA and a dysregulated RAS implies a close relationship of these factors to preeclampsia.

Granger et al. used reduction in uterine perfusion pressure (RUPP) in rats to investigate the role of the ischemic placenta during pregnancy. Pregnant rats underwent a surgical procedure wherein small silver clips were placed around the aorta at the iliac bifurcation and on both ovarian arteries which reduced blood supply the uterus [90]. RUPP rats demonstrated a “preeclamptic-like state” with increases in blood pressure, proteinuria, sFlt-1, TNF-α, endothelin production and endothelial dysfunction. Notably, RUPP rats develop AT₁-AA, whereas unmanipulated pregnant rats do not [91]. This group also investigated the effect of TNF-α alone. Interestingly, when low-dose TNF-α was infused into pregnant rats, increased blood pressure and the production of AT₁-AA followed [91]. These effects were not observed in non-pregnant animals, implying that placental ischemia can lead to an inflammatory response that triggers the production of AT₁-AA. The appearance of the AT₁-AA in RUPP rats indicates a relationship between reduced placental perfusion and the derangement of RAS.

Recently Zhou et al. have shown that injection of AT₁-AA obtained from preeclamptic women into pregnant mice can induce hypertension, proteinuria and increased circulating sFlt-1 [61]. The placentas of these mice are small. The effects of injected AT₁-AA are diminished with co-injection of losartan, an AT₁-receptor blocker, or a short antibody-neutralizing epitope peptide [61]. These findings indicate that AT₁-AA contribute to the pathophysiology of preeclampsia via AT₁-receptor activation.

Placental hypoxia and ischemia: a possible explanation for autoantibody production in preeclampsia

Despite the growing body of work described above linking AT₁-AA to preeclampsia the cause for this autoantibody generation is unknown. In general, the etiology of autoimmune disease remains unidentified, however many factors have been proposed, including a genetic predisposition, a maladaptive immune response and environmental triggers [92–94]. In the case of preeclampsia, autoantibody generation could be secondary to reduced placental perfusion, leading to vascular injury that exposes the offending antigen, coupled with the increased inflammatory response associated with the disease [95-98]. RUPP-induced hypertension was markedly attenuated by antagonism of the AT₁ receptor [99,100], supporting the current finding that it relies on autoantibody stimulation. In addition, this group has shown that low-dose TNF-alpha infusion into pregnant rats induced autoantibody production [91]. In both cases of AT₁-AA production, the hallmark features of preeclampsia were also evident: hypertension and proteinuria. Therefore, these experimental models of preeclampsia may provide a valuable system to determine the immunological origin of AT₁-AA in the disorder.

Collectively, these studies indicate that placental ischemia, the associated vascular damage and inflammatory response may serve as important stimuli for AT₁-AA production during pregnancy. This also implies that during preeclamptic pregnancies, AT₁ receptor activation plays an important role in the hypertension produced subsequent to placental ischemia.

It will be important to assess all experimental animal models of preeclampsia for the presence of AT₁-AA. In order to do this, a reliable, high throughput assay must be developed to determine both the presence and relative activity of the autoantibody. By determining the exact triggers and environment which give rise to AT₁-AA in experimental animal models, insight will be gained on the possible mechanisms of autoantibody generation in the human gravid state.

Conclusions and significance

The renin-angiotensin system has long been thought to play an important role in placentation, normal pregnancy and the pathophysiology of preeclampsia. How all the components of the system interact during normal and abnormal pregnancy has yet to be entirely understood. While there is a general upregulation of the RAS in normal pregnancy, this delicate balance is lost preeclampsia.

The central role of the placenta in preeclampsia is undisputed. Specific factors, such as sFlt-1, liberated by the placenta are now at the forefront of efforts to decipher the pathogenesis of this disease. Therefore, the emerging role of AT₁-AA and its induction of these placental factors will be integral to our understanding of preeclampsia. As described, AT₁-AA can enhance the cascade of RAS by acting synergistically with ANG II and its biological effects can be blocked by a 7-AA peptide corresponding to a specific epitope on the second extracellular loop of the AT₁ receptor. This epitope consistency, i.e. the fact that AT₁-AA observed in women with preeclampsia all recognize the same epitope peptide sequence, suggests a common immunological origin of these autoantibodies. This has profound therapeutic implications which could be specifically targeted against AT₁-AA. Treatment for preeclampsia is currently limited, and severe cases often require premature delivery of the infant. If maternal circulating AT₁-AA play a vital role in preeclampsia, their timely removal from or inhibition in preeclamptic women may provide significant therapeutic benefit. Eventually, it may be possible to block autoantibody-mediated AT₁ receptor activation, thereby forestalling or preventing the onset of the symptoms of preeclampsia.

Abbreviations

ANG II

angiotensin II

AT₁-AA

angiotensin II type I receptor agonistic autoantibody

NT

normotensive

PE

preeclampsia

sFlt-1

soluble fms-like tyrosine kinase-1

RAS

Renin-Angiotensin System