ACE2: a key modulator of the renin-angiotensin system and pregnancy.

Angiotensin-converting enzyme 2 (ACE2) is a membrane-bound protein containing 805 amino acids. ACE2 shows approximately 42% sequence similarity to somatic ACE but has different biochemical activities. The key role of ACE2 is to catalyze the vasoconstrictor peptide angiotensin (ANG) II to Ang-(1-7), thus regulating the two major counterbalancing pathways of the renin-angiotensin system (RAS). In this way, ACE2 plays a protective role in end-organ damage by protecting tissues from the proinflammatory actions of ANG II. The circulating RAS is activated in normal pregnancy and is essential for maintaining fluid and electrolyte homeostasis and blood pressure. Renin-angiotensin systems are also found in the conceptus. In this review, we summarize the current knowledge on the regulation and function of circulating and uteroplacental ACE2 in uncomplicated and complicated pregnancies, including those affected by preeclampsia and fetal growth restriction. Since ACE2 is the receptor for SARS-CoV-2, and COVID-19 in pregnancy is associated with more severe disease and increased risk of abnormal pregnancy outcomes, we also discuss the role of ACE2 in mediating some of these adverse consequences. We propose that dysregulation of ACE2 plays a critical role in the development of preeclampsia, fetal growth restriction, and COVID-19-associated pregnancy pathologies and suggest that human recombinant soluble ACE2 could be a novel therapeutic to treat and/or prevent these pregnancy complications.

INTRODUCTION

During pregnancy, the maternal and fetal circulating and tissue renin-angiotensin systems (RASs) act in concert to ensure a successful pregnancy outcome. The circulating RAS regulates maternal blood pressure and fluid volume, whereas the intrauterine tissue RAS is thought to promote placentation and regulate uteroplacental blood flow (1–3). Angiotensin-converting enzyme 2 (ACE2), a key component of the RAS, transforms angiotensin (ANG) II to Ang-(1–7) and protects against ANG II-induced oxidative stress and inflammation (4–7). The expression of ACE2 thereby influences the relative amounts of ANG II and Ang-(1–7). It is perhaps not surprising then that a potential imbalance in the ANG II and Ang-(1–7) pathways that favor the ANG II pathway are reported in women with preeclampsia (PE), and other common pregnancy complications such as fetal growth restriction (FGR) (8–12). The protective role of ACE2 in counterbalancing the ANG II-mediated vasoconstrictor, proinflammatory, and oxidative stress-inducing arm of the RAS has been extensively investigated in the heart, lung, and kidney (13–16); however, its actions in pregnancy have not yet been fully investigated.

ACE2 also acts as a receptor for the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is responsible for COVID-19 (17–19). By binding and endocytosing ACE2 (20) or increasing its shedding from the cell surface (21), SARS-CoV-2 decreases the expression of ACE2 on the cell membrane, likely decreasing the local metabolism of ANG II and production of Ang-(1–7). This would lead to an increase in the ANG II-to-Ang-(1–7) ratio, which causes inflammation, hypertension, and coagulopathies (22) that are also seen in PE (23). Indeed, women with COVID-19 in pregnancy are more likely to develop pregnancy complications such as PE and FGR (24–30). This review summarizes the current evidence on the role of ACE2 both in healthy and pathological pregnancies, including those affected by COVID-19, and the potential for ACE2 to be targeted therapeutically in pregnancy.

THE RENIN-ANGIOTENSIN SYSTEM

The RAS is a multifactorial signaling cascade that regulates blood pressure, salt and water homeostasis, and tissue growth. (31). Tissue RASs can act either in an autonomous manner or congruous with the circulating RAS, to regulate a variety of physiological and pathological functions, including cell proliferation, invasion, and differentiation, as well as inflammation and oxidative stress (32–34). These functions of the RAS are controlled by balancing the key pathways that produce the two major peptides ANG II and Ang-(1–7) (Fig. 1). The rate-limiting enzyme, renin (REN; which is produced by the kidney), converts angiotensinogen (AGT; which is produced in the liver) to ANG I. ACE then cleaves ANG I to form the potent vasoconstrictor peptide, ANG II. ANG II has well-known physiological roles in stimulating aldosterone release, increasing sodium reabsorption and maintaining blood pressure, and stimulating cell proliferation and angiogenesis (1, 8, 35, 36); functions that can become pathological if ANG II activity via its type I receptor (AT1R) is uncontrolled. Importantly, ANG II can also bind to its type 2 receptor, AT2R, which has actions that oppose those effects mediated by ANG II binding to the AT1R (37). In addition, the other dominant RAS peptide, Ang-(1–7), which is produced by ACE2 cleavage of ANG II, also counter-regulates the functions of ANG II/AT1R signaling (38). Ang-(1–7) can also be produced by other pathways including via the cleavage of ANG I by neprilysin (NEP) (39), or ACE2. Specifically, cleavage of ANG I by ACE2 forms Ang-(1–9), which is subsequently converted into Ang-(1–7) by ACE and/or NEP (Fig. 1) (40, 41). Ang-(1–7) counter-regulates the actions of the ANG II/AT1R pathway by binding to the MAS receptor (42).

Figure 1.

Schematic representation of the renin-angiotensin system (RAS). ACE, angiotensin-converting enzyme; ACE2, angiotensin-converting enzyme 2; AGT, angiotensinogen; ANG, angiotensin; AT1R, angiotensin II type 1 receptor, AT2R; angiotensin II type 2 receptor; BP, blood pressure; MasR, Mas receptor; NEP, neprilysin.

DISCOVERY, STRUCTURE, TISSUE DISTRIBUTION, AND REGULATION OF ACE2

ACE2 is an 805 amino acid containing transmembrane protein that shows ∼42% sequence similarity to somatic ACE (43). ACE2 belongs to the type I integral membrane protein family. It consists of a large extracellular domain that acts as a catalyst in which the zinc-binding motif, HEXXH, is located, a membrane-spanning domain securing ACE2 to the plasma membrane, and a short cytoplasmic domain of 22 amino acids. ACE2 acts entirely as a monocarboxypeptidase and therefore has different biochemical activities from ACE (38). Although ANG II is the major substrate for ACE2, apelin-13, dynorphin A, and des-Arg bradykinin are also potent substrates for ACE2 (2, 38). The predominant role of ACE2 is to catalyze the vasoconstrictor peptide ANG II to Ang-(1–7), implicating ACE2 as a regulator of the RAS (2, 38, 44, 45). ACE2 has an almost 500-fold greater catalytic efficiency in producing Ang-(1–7) from ANG II than from ANG I [which must undergo cleavage by multiple enzymes to produce Ang-(1–7); see Fig. 1] (38). ACE2 is not blocked by ACE inhibitors such as captopril, lisinopril, and enalapril (40). Furthermore, research suggests that ACE2 acts as a protective component of the RAS to oppose the overactivity of ANG II, which is responsible for vasoconstriction (46), reactive oxygen species (ROS) production (47), and inflammation, and thus exerts a protective role in end-organ damage (48, 49).

ACE2 is most abundantly expressed in the kidney, lung, testis, heart, gastrointestinal tract, liver, placenta, and uterus (43, 50, 51). Notably, ACE2 and Ang-(1–7) are abundantly expressed in placental syncytiotrophoblast, cytotrophoblast, and villous stroma (1, 52).

The abundance/activity of ACE2 could be regulated in part by the levels of ANG II (53–55). However, the involvement of ANG II in regulating ACE2 is not fully understood, as studies have found that ACE2 expression is reduced in response to ANG II in human kidney tubular cells (53, 55), whereas ANG II increases levels of ACE2 in hepatic cells (56). AT1R signaling is thought to be involved in ANG II-mediated downregulation of ACE2. Koka et al. (53) have shown that blockade of ERK1/2 or p38 MAP Kinase abolished the ANG II-induced decrease in ACE2 in human kidney tubular cells. Similar results were found in human umbilical artery smooth muscle cells, where blockade of ERK1/2 and c-Jun NH2-terminal kinase (JNK) phosphorylation prevented ANG II-induced ACE2 downregulation (57). Furthermore, it has been reported that AT1R inhibition in rat aortic vascular smooth muscle cells (VSMCs) and renal cortical cells increase levels of ACE2 (55, 58). To the best of our knowledge, these regulatory mechanisms have not yet been investigated in pregnancy.

Insight into the transcriptional regulation of ACE2 is still in its infancy. Although ANG peptides might regulate the expression of ACE2, some studies have shown that oxygen levels (59) as well as hormones such as aldosterone or endothelin-1 (60) might also be involved in regulating its expression. For example, while hypoxia diminishes the transcription of ACE2 in pulmonary artery smooth muscle cells (59), further investigation has shown that hypoxia-induced HIF-1α augments ACE expression causing elevated levels of ANG II, which then facilitates a reduction in ACE2 (59).

In the placenta, oxygen has also been shown to regulate ACE2 expression. Rats exposed to antenatal maternal hypoxia have decreased placental ACE2 mRNA but increased ACE2 protein levels (61). Similarly, treatment of BeWo trophoblast cell lines with cobalt chloride to mimic hypoxia causes increased ACE2 mRNA and protein levels (62). In humans, placental hypoxia, as assessed by examination of monochorionic twin anemia-polycythemia (TAPS) placentae, a model of discordant placental oxygenation, is also associated with increased ACE2 protein levels (63).

ANTIOXIDANT PROPERTIES OF ACE2

ACE2 reduces ANG II-induced oxidative stress by producing Ang-(1–7). ANG II increases the expression and activation of NAD(P)H oxidases, a major source of superoxide anion (Fig. 2) (64). Superoxide reacts with the vasodilator nitric oxide (NO), resulting in decreased NO bioavailability and endothelial dysfunction (64). Ang-(1–7), which is produced by ACE2, when bound to the Mas receptor (4), decreases the production of ROS, increases endothelial NO synthesis (64), and attenuates the activity of NAD(P)H oxidase (Fig. 2) (65), improving the antioxidant capacity of tissues (64) and preventing endothelial dysfunction (64). In addition, studies have shown that genetic deletion of ACE2 causes a dramatic decrease in the expression of endothelial NO synthase (eNOS) and a decrease in NO concentration (66). ACE2 knockout mice have decreased superoxide dismutase (SOD) activity and increased lipid peroxidation, indicating impaired antioxidant capacity (66). MAS-deficient mice also display a reduction in SOD and catalase (CAT) activity and an increase in lipid peroxidation, isoprostanes, and ROS levels accompanied by impaired endothelial function and increased blood pressure (67).

Figure 2.

Angiotensin-driven oxidative stress and inflammation pathways. ANG II/AT1R signaling enhances NF-κB-mediated translation of proinflammatory factors and, in turn, increases inflammatory cell recruitment. ANG II/AT1R signaling enhances NAD(P)H oxidase activation to increase reactive oxygen species (ROS) and enhance markers of oxidative stress. Increased ROS levels further enhance NF-κB-mediated translation. The actions of ACE2 in degrading ANG II and producing Ang-(1–7) will result in activation of the Ang-(1–7)/MasR pathway. Ang-(1–7)/MasR signaling reduces inflammation by inhibiting NAD(P)H oxidase and/or NF-κB-mediated oxidative stress and inflammation. Furthermore, Ang-(1–7)/MasR signaling enhances Nrf-2-mediated antioxidant and cytoprotective gene translation. ACE, angiotensin-converting enzyme; ACE2, angiotensin-converting enzyme 2; ANG, angiotensin; AT1R, angiotensin II type 1 receptor; eNOS, endothelial nitric oxide synthase; HO-1, hemoxygenase 1; IL, interleukin; MasR, Mas receptor; MCP-1, monocyte chemoattractant protein-1; NF-κB, nuclear factor-κB; Nrf2, nuclear factor-erythroid-2-related factor; NQO1, NAD(P)H quinone dehydrogenase 1; ROS, reactive oxygen species; SOD, superoxide dismutase; TNF-α, tumor necrosis factor-α; TGF-β, transforming growth factor-β.

In contrast, spontaneously hypertensive rats overexpressing ACE2 in vascular smooth muscle cells show improved endothelial function and reduced vasoconstriction in response to ANG II (68). More recently, ACE2 has been shown to activate the antioxidant nuclear factor-erythroid-2-related factor (Nrf2)/heme oxygenase-1 (HO-1)/NAD(P)H quinine oxidoreductase 1 (NQO1) pathway to reduce hyperoxic lung injury (Fig. 2) (69). Diabetic spontaneously hypertensive rats treated with Ang-(1–7) also have improved endothelial dysfunction, mediated primarily by a decrease in NAD(P)H-mediated oxidative stress (70). We suggest that placental ACE2 and, hence the placental ACE2/Ang-(1–7)/Mas axis, has similar antioxidant actions and thus protects pregnant women from placental dysfunction that can lead to severe pregnancy pathologies including PE and FGR. To date, this has not been investigated.

ANTI-INFLAMMATORY PROPERTIES OF ACE2

The RAS has both proinflammatory and anti-inflammatory actions. Most proinflammatory actions of the RAS appear to be due to the effects of ANG II acting on the AT1R (Fig. 2). Indeed, ANG II activates several pathways related to tissue injury, inflammation and fibrosis, activation of protein kinases and nuclear transcription factors, recruitment of inflammatory cells, adhesion of monocytes and neutrophils, and stimulation of expression, synthesis and release of cytokines, and chemokines (Fig. 1) (7). The ACE2/Ang-(1–7)/MAS axis suppresses inflammation by inhibiting leukocyte migration, cytokine expression and release, and profibrotic pathways (7) (Fig. 2).

Ang-(1–7), the product of ACE2 (a key enzyme of RAS), has been shown to play a role in inhibiting the activity of inflammatory signaling molecules such as MAPK, protein kinase C (PKC), and c-SRC kinase (5, 71–73). A previous study has also shown that Ang-(1–7) has significant anti-inflammatory actions in the pathogenesis of inflammatory bowel disease (IBD) (6). Ang-(1–7) administration or ACE2 overexpression both inhibit neutrophil accumulation and leukocyte adhesion, inhibit the proinflammatory cytokines interleukin (IL)-6, IL-1β, and tumor necrosis factor (TNF)-α and enhance the anti-inflammatory cytokine, IL-10 (Fig. 2) (7). ACE2-deficient animals express higher levels of IL-6, IL-1β, and TNF-α and neutrophil accumulation in a variety of models (summarized in Ref. 7). Although the regulation of inflammatory processes in pregnancy has been extensively investigated, the role of the RAS and, in particular, the protective role of the ACE2/Ang-(1–7)/Mas receptor axis in maintaining the balance between the proinflammatory and anti-inflammatory RAS pathways in pregnancy has yet to be explored.

SHEDDING OF ACE2

The ectodomain of ACE2 can be cleaved by the metalloprotease ADAM 17 to form soluble ACE2 (sACE2; see Fig. 3) (77). The site for ADAM 17 cleavage is debated with its location thought to be between amino acids 716 and 741 (21, 74, 77) or 708 and 709 (75). In addition, recent studies suggest that the arginine/lysine residues located at amino acids 652–659 function as a signal for ADAM 17 to recognize the downstream cleavage site and are essential for ACE2 shedding by ADAM 17 (76). Jia et al. also showed that ADAM 10 (74) could cause shedding of ACE2; however, this is disputed (78). It is important to note that while studies have shown that TMPRSS2 (transmembrane protease serine S1 member 2) and HAT (human airway trypsin-like protease) also cleave ACE2 (at amino acids 697–716), whether this cleavage results in shedding of a biologically stable form of the sACE2 is unclear (76). sACE2 retains catalytic activity and can still regulate the conversion of ANG II to Ang-(1–7), (74). Soluble ACE2 levels and activity can be detected in both plasma and urine (79).

Figure 3.

Angiotensin-converting enzyme 2 (ACE2) shedding. A: full-length ACE2 protein contains 805 amino acids and consists of a signal peptide (SP) at the NH2 terminus, a large carboxypeptidase (catalytic) domain containing a zinc-binding motif “HEXXH,” and a collectrin domain containing a protease cleavage (PC) site, transmembrane (TM) domain, and a short 22 amino -acid cytoplasmic tail at the COOH terminus. The extracellular component of ACE2 can be cleaved and released into the extracellular space as soluble ACE2 (sACE2). A and B: The extracellular compartment of ACE2 can be cleaved by α-disintegrin and metalloprotease (ADAM) 17 between amino acids 716–741 (74) and/or 708–709 (75). Recent studies by Heurich et.al. (76) suggest that the arginine/lysine (Arg/Lys) residues located at amino acids 652–659 are essential for ACE2 shedding by ADAM 17 because they act as a signal to initiate ADAM 17 recognition of the downstream cleavage site. In addition, while disputed in the literature, Jia et.al.(74) have shown that ADAM 10 may also be involved in ACE2 ectodomain shedding. Type II transmembrane serine proteases TMPRSS2 and HAT cleave full-length ACE2 at amino acids 697–716; however, whether this cleavage results in a biologically stable form of sACE2 has not been confirmed (76).

CIRCULATING AND TISSUE ACE2 LEVELS IN UNCOMPLICATED PREGNANCIES

We have shown that sACE2 levels and activity are greatly increased in the maternal circulation during pregnancy compared with healthy nonpregnant women and they remain high throughout gestation (11). In agreement with a previous study (80), we have also found that Ang-(1–7) levels are increased in pregnant compared with nonpregnant women (11). Furthermore, we have shown that both sACE2 levels and activity are positively correlated with each other and that Ang-(1–7) levels are positively correlated with sACE2 levels and activity (11). This implies that sACE2 is active and the amount and activity of sACE2 in the circulation is rate-limiting in terms of the production of Ang-(1–7). A summary of the reported changes in the key circulating RAS peptides and enzymes during pregnancy is reported in Table 1. The elevated levels and activity of sACE2 in pregnancy also suggests that it could play a critical role in the maintenance of maternal blood pressure by producing Ang-(1–7). A previous study has shown a significant negative correlation between maternal blood pressure and Ang-(1–7) levels (80). Furthermore, pregnant ACE2 knockout mice have increased blood pressure (82). In the ACE2 gene specifically, the major allele of rs2074192 and the minor allele of rs233575 have been associated with an increased risk of hypertension among males of European ancestry. Furthermore, the ACE2 rs2106809 polymorphism (males: T vs. C, females: TT vs. CC/CT) is associated with hypertension in both males and females of Indian ancestry (83, 84). To the best of our knowledge; however, there are no data demonstrating any correlations between ACE2 gene polymorphisms and the activity of ACE2. Therefore, further research is required to fully understand the role of ACE2 gene polymorphisms in hypertension.

Table 1.

Maternal circulating RAS components in healthy, preeclamptic, and SGA pregnancies

Serum RAS Component	Healthy Pregnancy	Preeclamptic Pregnancy	SGA Pregnancy	References
ACE	↑/↓	↑/↓	↑	Merrill et al. (80); Langer et al. (81); Zhou et al. (9); Tamanna et al. (11)
ANG II	++	↓	?	Merrill et al. (80)
ACE2	++	↓	↑	Tamanna et al. (11)
Ang-(1–7)	++	↓	=	Merrill et al. (80); Tamanna et al. (11)
NEP	=	↓	=	Tamanna et al. (11)

ACE, angiotensin-converting enzyme; ANG, angiotensin; NEP, neprilysin; RAS, renin-angiotensin system; SGA, small for gestational age; ↑/↓, increased or decreased compared with nonpregnant or healthy pregnant; ++ greatly increased compared with nonpregnant women; =, not significantly different to nonpregnant; ?, no data available. Note: preeclamptic and SGA pregnancies were compared with healthy pregnant only.

Very limited studies have been carried out to examine the expression of ACE2 and its product, Ang-(1–7), in the uteroplacental unit in humans. We have shown that ACE2 is localized to the syncytiotrophoblast (STB) of the placental villi (1). As the STB is the major secretory cell of the placenta, this suggests that in the STB, ACE2 could act to release Ang-(1–7) into the maternal circulation and thereby contribute to the vasodilation of the maternal vasculature. Previous studies carried out in rat and human placenta demonstrated that Ang-(1–7) is colocalized with ACE2 in the placenta (52, 85). Furthermore, immunohistochemical analysis by Valdes et al. (85) has shown that ACE2 and Ang-(1–7) are expressed in the cytotrophoblast, STB, and invading and intravascular trophoblasts of the human placenta. The expression of Ang-(1–7) and ACE2 in the luminal and glandular epithelium of the uterus (52) also suggests a role for the RAS in the process of implantation and decidualization (86).

EVIDENCE FOR THE ROLE OF ACE2 IN PREGNANCY FROM ANIMAL MODELS

Previous studies have shown that there is significant upregulation of ACE and ACE2 expression in placentas collected from Brown Norway (BN) rats, which present with natural placental insufficiency, a characteristic feature of preeclampsia, compared with Sprague–Dawley controls (87). On the contrary, in the uterus and placentas obtained from a rat model of pregnancy-induced hypertension, the reduced uterine perfusion pressure (RUPP) model, Ang-(1–7), the product of ACE2, was significantly decreased (52). Reduced levels of ACE2 mRNA were also found in the uterus of RUPP rats (52). However, placental ACE2 mRNA was unchanged in RUPP animals compared with controls (52).

Pregnant ACE2 KO mice are reported to exhibit increased systolic blood pressure, placental hypoxia, reduced umbilical blood flow velocity, decreased pup weights, and increased pup resorptions (82, 88). In addition, they have lower circulating maternal Ang-(1–7) and higher ANG II levels than their pregnant wild-type counterparts. It is unclear, however, whether these changes are due to placental or systemic ACE2 deficiency as no placental-specific ACE2 KO models currently exist.

Decreases in placental ACE2 expression can also be found in various animal models of FGR. For example, FGR caused by maternal protein restriction in rats is associated with reduced placental ACE2 expression (89). Dexamethasone administration causes a reduction in fetal weight and labyrinthine zone weight accompanied by a decrease in ACE2 mRNA and protein levels and a decrease in Ang-(1–7) levels in the placenta (90). Furthermore, ACE2 KO mice have reduced fetal weights and a reduction in the fetal-to-placental weight ratio, an indirect measure of placental insufficiency (82).

CHANGES IN THE CIRCULATING AND UTEROPLACENTAL ACE2 IN PATHOLOGICAL PREGNANCIES

We have shown that maternal circulating ACE2 levels and activity, as well as Ang-(1–7) levels are decreased in the third trimester of women with PE (11). Merrill et al. (80) also showed that Ang-(1–7) levels were reduced in women with PE, as were ANG I and ANG II levels. However, Valdes et al. (85) and Gao et al. (91) showed that placental levels of Ang-(1–7) and ACE2 were not significantly different between normal term or preeclamptic pregnancies, except for enhanced expression of ACE2 in umbilical arterial endothelium in women with PE. These suggest that control of placental ACE2 expression and, therefore, the activity of ACE2 may differ from the regulation of the circulating RAS.

We have shown that human FGR placentae, similar to PE placenta, have reduced ACE2 expression (10). We have also shown that mothers giving birth to small for gestational age (SGA) babies have significantly higher levels of sACE2 in the circulation at 13, 18, and 30 wk of pregnancy but not in late pregnancy (36 wk) (11). This could act to counteract the actions of the increased sACE in early gestation in women who gave birth to SGA babies (11) and regulate the levels of ANG II. A recent study has demonstrated that if the fetus, but not their mothers, had the ACE2 rs2074192 T allele polymorphism they were more likely to be born SGA, which could contribute to their susceptibility to metabolic syndrome and cardiovascular disease in later life (92).

Together, these studies suggest that reduced ACE2 and Ang-(1–7) in the placenta may both contribute to PE and FGR. A summary of the reported changes in the key RAS peptides and enzymes, including ANG II, ACE, and NEP, in preeclamptic, SGA, and FGR pregnancies are reported in Tables 1 and 2.

Table 2.

Alterations in the expression of uteroplacental RAS components in preeclamptic and FGR pregnancies compared with healthy pregnancies

Placental RAS Component	Model	Preeclampsia	FGR	References
ACE	Human	=/↑	=	Kalenga et al. (93)Ito et al. (2)Delforce et al. (10)
ANG II	Human	↑	?	Gao et al. (91)
ACE2	Human and rat	↑/=	↓	Valdes et al. (85)Delforce et al. (10)
Ang-(1–7)	RUPP rat and human	↓/=	?	Neves et al. (52)Gao et al. (91)
NEP	Human	?	↓	Delforce et al. (10)

ACE, angiotensin-converting enzyme; ANG, angiotensin; FGR, fetal growth restriction; NEP, neprilysin; RAS, renin-angiotensin system; RUPP, reduced uterine perfusion pressure; =, not significantly different to healthy pregnant; ↑/↓, increased or decreased compared with healthy pregnant; ?, no data available.

LINK BETWEEN ACE2 AND SARS-CoV-2 AND ITS POTENTIAL IMPACT IN PREGNANCY

ACE2 is the major receptor for SARS-CoV-2, which is responsible for COVID-19 (17–19). The SARS-CoV-2 spike protein binds with human ACE2 with substantially higher affinity than does the spike protein of SARS-CoV-2 (18). Binding of the SARS-CoV-2 spike protein to membrane-bound ACE2 causes endocytosis of ACE2 alongside viral particles into endosomes and reduces surface ACE2 expression (94). The spike protein of SARS-CoV also induces ADAM 17-dependent shedding of the ACE2 ectodomain (21), resulting in loss of the protective actions of the ACE2/Ang-(1–7)/Mas receptor arm of the RAS, accumulation of ANG II, and unopposed ANG II-induced inflammation and oxidative stress. Recently, Verma et al. (20) demonstrated that colonization of SARS-CoV-2 was found to be the greatest in maternal decidua, fetal trophoblasts, Hofbauer cells, and in placentas delivered prematurely. SARS-CoV-2 was localized to cells expressing ACE2 and infected placentas had significantly lower levels of ACE2 (20). This is, however, under contention, with new reports suggesting that placental ACE2 mRNA expression is increased in severe cases of COVID-19 (95). Regardless, it is likely that SARS-CoV-2 infection in pregnant women has adverse effects on the balance between the ACE/ANG II/AT1R and ACE2/Ang-(1–7)/MasR axes in the placenta resulting in poor pregnancy outcomes for both the mother and fetus. Whether or not the virus alters maternal systemic RAS pathways have not yet been investigated and in nonpregnant individuals reports on the effect of the virus on circulating ACE, ACE2, and ANG peptide levels are conflicting (96–102). Furthermore, since SARS-CoV-2 binding to ACE2 induces ADAM 17-dependent shedding of the ACE2 ectodomain (21), circulating levels are not likely to reflect local levels in tissues.

Emerging evidence suggests that COVID-19 infection in pregnancy is associated with adverse neonatal outcomes, including preterm birth and FGR with several reports of neonatal or intrauterine death (103). Importantly, hypertensive disorders of pregnancy, including PE, have also been noted in women with COVID-19 (20, 103–105).

Several studies showed that placental infection with SARS-CoV-2 can occur (106–110). Hosier et al. (106) by analyzing a case study, reported a case where a pregnant woman contracted COVID-19 in the second trimester of pregnancy and presented with severe hypertension, coagulopathy, and PE. There was clear evidence of viral infection of the placenta, which was positive for SARS-CoV-2 RNA [in situ hybridization (ISH)] and spike protein detected by immunohistochemistry (IHC). Viral particles identified by electron microscopy were also predominantly located in the cytosol of the syncytiotrophoblast. The placenta was infiltrated by T lymphocytes and macrophages, which was considered a response to the inflammatory cascade caused by the viral infection. Menter et al. (107) also showed that the term placenta from women with SARS-CoV-2 had notable lymphohistiocytic inflammatory infiltrate causing chronic lymphohistiocytic villitis and intervillositis accompanied by vasculitis of fetal vessels and focal thrombosis. Shane et al. (108) showed that third-trimester placentas from pregnant women with severe SARS-CoV-2 had greater rates of decidual arteriopathy and other maternal vascular malperfusion features, more specifically, abnormal or injured maternal vessels, and intervillous thrombi compared with controls. However, they did not find evidence of higher acute and/or chronic inflammation compared with controls. Similarly, Smithgall et al. (109) showed that there was maternal/fetal vascular malperfusion in third trimester and term placentas, respectively, from women positive for SARS-CoV-2. However, ISH and IHC reports showed no evidence of direct viral involvement. Vázquez et al. (110) showed fetal and placental infection with SARS-CoV-2 can also occur during early pregnancy.

Recent findings also suggest that in JEG-3 placental trophoblast cells, treatment with recombinant SARS-CoV-2 spike protein or a live modified virus with a vesicular stomatitis viral backbone expressing spike protein (VSV-S), significantly diminished the expression of ACE2 and enhanced the expression of AT1R compared with untreated control cells (20). This was concomitant with an increase in the antiangiogenic soluble fms-like tyrosine kinase-1(sFlt1), a notable marker of PE. The SARS-CoV-2 viral infection was associated with decreased proangiogenic factors including AT2R and placental growth factor (PlGF) (20), the latter of which competitively binds to sFlt1, quenching its actions. Infected pregnant women had increased levels of serum sFlt1 and AT1R autoantibodies before delivery, these being critical markers of PE (20). In cases of infection with SARS-CoV-2, decreased placental ACE2 activity, leading to increased levels of ANG II could unmask PE and/or impair fetal growth, especially in pregnancies predisposed to these complications.

Importantly, recombinant human soluble ACE2 [(rh)sACE2)] has been shown to block binding of the SARS-CoV-2 spike protein to membrane-bound ACE2, thereby preventing cellular infection (111). Likewise, Monteil et al. (112) demonstrated that (rh)sACE2 might lessen SARS-CoV-2 load by a factor of 1,000–5,000 in in vitro cell culture experiments and engineered organoids, suggesting that (rh)sACE2 can effectively neutralize SARS-CoV-2.

Furthermore, treatment of a SARS-CoV-2-infected patient with (rh)sACE2 significantly decreased ANG II, increased the ACE2 products, Ang-(1–7) and Ang-(1–9), and reduced inflammatory markers such as IL-6 and chemokine IL-8, as well as ferritin (113). In addition, clinical trials have shown that (rh)sACE2 is safe in healthy adults and trials are underway to test its use in treating acute respiratory distress syndrome. Therefore, (rh)sACE2 may be a novel treatment option for women with SARS-CoV-2 infection in pregnancy to prevent adverse pregnancy outcomes such as FGR and PE.

THE ACE2/Ang-(1–7)/MAS AXIS AS A POTENTIAL THERAPEUTIC TARGET IN PREGNANCY

It is clear that there is an imbalance in the two opposing RAS pathways in pregnancy complications like preeclampsia that favors the ANG II/AT1R pathway (8–12). Since drugs that inhibit this pathway (i.e., ACE inhibitors and angiotensin receptor blockers) are contraindicated in pregnancy, we believe that attention now needs to focus on therapeutic activation of the protective ACE2/Ang-(1–7)/MasR arm (Fig. 4). There are several ways in which this pathway could be targeted.

Figure 4.

Overarching hypothesis. Angiotensin converting enzyme 2 (ACE2) protects against placental inflammation and oxidative stress. Downregulation of placental ACE2 (e.g., because of inhibition or SARS-CoV-2 spike protein binding) results in placental inflammation and oxidative stress, which limits fetal growth and causes the classical symptoms of preeclampsia (hypertension and proteinuria). Treatment with recombinant human (rh)sACE2 and/or activation of ACE2 may prevent SARS-CoV-2 binding to the placenta (in the case of (rh)sACE2) and protect against growth restriction and preeclampsia.

First, (rh)sACE2 could be administered to high-risk women to replace the diminished sACE2 levels seen in women with preeclampsia and SGA pregnancies. This may also be beneficial for women with COVID-19 in pregnancy to effectively “mop-up” circulating viral particles. Alternatively, the activity of ACE2 could be stimulated with an ACE2 agonist such as diminazene aceturate (DIZE). DIZE has been shown to exert beneficial effects in experimental models of hypertension, myocardial infarction, type 1 diabetes, and atherosclerosis (114). Furthermore, DIZE prevents inflammation, ROS generation, NFκ-B activation, and Nrf2 inhibition in animal models of acute lung injury (69).

An alternative approach would be to treat women with Ang-(1–7), or an analog of Ang-(1–7) to improve its stability. In a rat model of preeclampsia (RUPP), Ang-(1–7) administration significantly attenuated the rise in systolic blood pressure and urinary protein, inhibited the increase in sFlt-1, IL-6, and decrease in VEGF, NO, eNOS, IL-10, and total antioxidant capacity in preeclamptic animals (115). Furthermore, in in vitro studies, Ang-(1–7) has been shown to attenuate podocyte injury in podocytes treated with preeclamptic serum (116). Spontaneously hypertensive rats (SHR) have been used to show that stimulation of the ACE2/Ang-(1–7)/Mas axis with Ang-(1–7) or DIZE prevents hypertension, cardiomyocyte hypertrophy, and left ventricular and renal fibrosis in adult male offspring (117). To date, no one has examined the beneficial effects of these therapies in pregnancy, however, there are several clinical trials currently underway to study the efficacy of intravenous Ang-(1–7) for the treatment of COVID-19, hypertension, and metabolic syndrome in nonpregnant individuals.

FUTURE PERSPECTIVES

Based on the existing evidence, we propose that ACE2 is important for a successful pregnancy outcome and that decreased expression and/or activity of ACE2 in pregnancy could lead to pregnancy pathologies such as PE and/or FGR (Fig. 4). Elucidation of the regulatory mechanisms for ACE2, and the functional aspects of the enzyme in distinct tissue compartments such as the placenta and uterus remain two key areas for continuing study. Though the levels of ACE2 have been measured in healthy and pathological pregnancies, further studies are essential to elucidate the protective role of ACE2 in counteracting the actions of ANG II mediated by the AT1R in pregnancy. As well, further research is also required to demonstrate that ACE2 and, hence the placental ACE2/Ang-(1–7)/MAS axis, protects against inflammation and oxidative stress in the placenta. FGR and PE might result from reduced expression/activity of ACE2 and a disorder of the placental ACE2/Ang-(1–7)/MAS axis. Future studies are also required to better understand the consequences of SARS-CoV-2 infection in pregnancy and the role that ACE2 plays in this. Furthermore, studies are needed to explore whether treatment with (rh)sACE2 or drugs that activate the ACE2/Ang-(1–7)/MAS pathway could improve pregnancy outcomes in at-risk women, including those with COVID-19 (Fig. 4).

SUMMARY

In conclusion, we have reviewed recent advances in investigations of ACE2 with respect to its critical role in understanding pregnancy pathologies and developing new approaches for pregnancy complications. So far, our knowledge of the role of ACE2 during pregnancy is inadequate and a better understanding of the role of circulating and local tissue ACE2 throughout gestation could be beneficial to target the RAS in preventing FGR, PE accompanied by placental insufficiency, oxidative stress, and inflammation. Further studies are needed to pave the way for future preclinical and clinical studies to examine the benefits of using ACE2 and other ACE2 modulating compounds to treat pregnancy complications.

GRANTS

S.T. was supported by an Australian Commonwealth-funded Research Training Program stipend. S.K.M. was supported by an Australian Government Research Training Program Scholarship and is now supported by Hunter Medical Research Institute researcher bridging funds. K.G.P. was supported by an Australian Research Council Future Fellowship (FT150100179).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.T., E.R.L., and K.G.P. conceived and designed research; S.K.M. and S.J.D. prepared figures; S.T., E.R.L., and K.G.P. drafted manuscript; S.T., E.R.L., S.K.M., S.J.D., and K.G.P. edited and revised manuscript; S.T., E.R.L., S.K.M., S.J.D., and K.G.P. approved final version of manuscript.