Relationship between SARS-CoV-2 and ACE2

Angiotensin-converting enzyme (ACE) 2, a homologue of ACE, is a zinc metalloprotease with hydrolase activity (Ref. Reference Donoghue9) that is able to hydrolyse angiotensin (Ang) I and Ang II to generate Ang-(1–9) and Ang-(1–7), respectively (Ref. Reference Vickers10). Ang II and Ang-(1–7) hormones are the most important hormones produced in the renin-angiotensin system (RAS) and have opposite effects (Ref. Reference Simões, Silva and Teixeira11). Ang II induces vasoconstriction and inflammatory reactions (Ref. Reference Campbell-Boswell and Robertson12), promotes proliferation (Refs Reference Ray13, Reference Hiruma14) and facilitates fibrosis and tissue remodelling (Ref. Reference Bataller15), whereas Ang-(1–7) has anti-inflammatory properties (Ref. Reference Oliveira16), mediates vasodilation (Ref. Reference Brosnihan17) and alleviates cardiac and metabolic dysfunction (Refs Reference Iwata18–Reference Giani21). Thus, ACE2, a key component of the RAS, is essential to balance Ang II and Ang-(1–7) levels (Fig. 1).

Fig. 1. Components of RAS and its role in female ovarian function. Ang I, angiotensin I; Ang II, angiotensin II; ACE, angiotensin-converting enzyme; Ang-(1-9), angiotensin-(1–9); Ang-(1-7), angiotensin-(1–7); AT1, angiotensin II type 1; AT2, angiotensin II type 2; RAS, renin-angiotensin system.

The SARS-CoV-2 virus gains access to host cells via attachment to the ACE2 receptor (Ref. Reference Ashour22). Coronaviruses are spherical-enveloped viruses capsuled with positive single-stranded RNA. The structural proteins of SARS-CoV-2 are composed of spike (S), membrane (M), envelope (E) and nucleocapsid (N) proteins. The first three proteins are embedded in the viral envelope, whereas the N protein, a core component of the nucleocapsid, interacts with the viral RNA (Ref. Reference Wang23). Similar to SARS-CoV-1, the viral S protein of SARS-CoV-2 has a strong affinity for ACE2 (Ref. Reference Sriram and Insel24). Viral S proteins have two subunits, the S1 and S2 domains. The S1 domain directly binds to receptors of host cells, whereas the S2 domain mediates viral and host cell membrane fusion (Refs Reference Simmons25, Reference He26). This process is also facilitated by the proteolytic cleavage and activation of viral S proteins induced by the transmembrane protease serine 2 (TMPRSS2) in the cytoplasm (Ref. Reference Hoffmann27). Then, viral genomic RNA is released into the target cell cytoplasm and replicates using the host cell organelles, resulting in new virion release (Refs Reference Boopathi28, Reference Jiang29). SARS-CoV-2 infection has been proven to decrease the activity and downregulate the expression of ACE2, resulting in an increase of Ang II recruitment and a decrease in Ang-(1–7) production in circulation, which explains the inflammatory reactions investigated in COVID-19 patients (Refs Reference Verdecchia30, Reference Kuba31).

ACE2 and ovarian function

It has been reported that ACE2 exists in a variety of mammalian ovaries, including rats (Ref. Reference Pereira32) and cattle (Ref. Reference Tonellotto dos Santos33). Additionally, ACE2 be detected in ovaries of women of reproductive age (Ref. Reference Reis34). ACE2 is highly expressed in stromal cells, theca cells and granulosa cells, as well as oocytes (Refs Reference Pereira32, Reference Barreta35). In the female reproductive system, ACE2 is predominantly enriched in the ovary (Refs Reference Lee36, Reference Jing37), making it a potential target organ for SARS-CoV-2 infection. Moreover, previous studies have demonstrated that ACE2 has been detected in embryos before the 8-cell stage and in trophectoderm cells of late blastocysts, and TMPRSS2 exists in embryos in the late blastocyst stage (Ref. Reference Cheng38), revealing a high SARS-CoV-2 infection susceptibility in peri-implantation embryos (Ref. Reference Chen39).

Ang II, most abundantly expressed in granulosa cells (Ref. Reference Miyabayashi40), regulates steroid secretion (Ref. Reference Shuttleworth41), promotes follicle growth (Ref. Reference Ferreira42), facilitates oocyte maturation (Ref. Reference Yoshimura43), contributes to follicular atresia (Ref. Reference Obermüller44), affects the ovulation process (Ref. Reference Xu and Stouffer45) and induces corpus luteum angiogenesis (Ref. Reference Sugino46). Although Ang-(1–7), found predominantly in theca-interstitial cells, induces steroidogenesis, especially oestradiol and progesterone production (Ref. Reference Costa47), enhances ovulation (Ref. Reference Viana48), resumes oocyte meiosis (Ref. Reference Honorato-Sampaio49) and regulates oocyte maturation (Ref. Reference Cavallo50). In addition, ACE2 can be found in follicles in various developmental stages, and the expression levels are regulated by the secretion of gonadotrophin, revealing the possible relationship between ACE2 expression and female fertility. Moreover, the level of Ang-(1–7) in human follicular fluid has been proven to be positively related to the oocyte maturation rate. This evidence supports the significance of Ang-(1–7) levels in the oocyte maturation process (Ref. Reference Cavallo50). Furthermore, the decrease in ACE2 activity induced by SARS-CoV-2 infection can increase circulating Ang II, which might alter ovarian function, influence the biological process of oocyte development and maturation, impact oocyte quality and ultimately impair fertility function (Ref. Reference Lee36). In addition, Ang II recruitment also increases oxidative stress (Ref. Reference Pan51), which may lead to inflammatory reactions and affect ovarian physiology and reproductive ability (Fig. 1).

ACE2 and endometrial activity

The uterus – in particular, the endometrium – is essential for female fertility, and the components of the RAS can be found in the uterus, especially in endometrial epithelial and stromal cells (Refs Reference Vaz-Silva52, Reference Brosnihan53). This makes the endometrium more susceptible to viral damage (Ref. Reference Abhari and Kawwass54), which might induce embryo implantation impairment. Some studies have suggested that RAS component expression varies with the menstrual cycles (Refs Reference Jing37, Reference Vaz-Silva52). ACE2 expression has been proven to be more abundant in the secretory phase than in the proliferation phase, and lower in stromal cells than in epithelial cells (Refs Reference Herr55, Reference Vinson56). Moreover, the expression of ACE2 is reported to increase with female age (Refs Reference Abhari and Kawwass54, Reference Henarejos-Castillo57). This evidence indicates that older females in the secretory phase are likely to be more susceptible to endometrial infection compared with younger women in the proliferation phase.

Evidence has demonstrated that the maintenance of Ang II and Ang-(1–7) balance is critical for regulating menstrual cycles because of the significant role of RAS in angiogenesis and tissue remodelling. Ang II, with tissue remodelling properties, induces spiral artery vasoconstriction, facilitates endometrial regeneration, enhances stromal proliferation and initiates menstruation (Refs Reference Ahmed58–Reference Li and Ahmed60). SARS-CoV-2 infection in the uterus might disturb the Ang II/Ang-(1–7) balance, decrease Ang II expression levels and alter Ang II distributions in the uterus, which may cause severe endometrial and myometrial disorders (Refs Reference Vaz-Silva52, Reference Deliu61), such as dysfunctional uterine bleeding (Ref. Reference Li and Ahmed62). Moreover, several studies have reported an association between ACE2 expression and the prognosis of endometrial cancer (Refs Reference Watanabe63, Reference Delforce64), revealing the significant role of ACE2 and RAS in uterine function.

ACE2 and pregnancy

The placenta provides nutrient and oxygen exchange between the mother and foetus. Although limited studies have investigated and analysed RAS function in the placenta, all RAS components are expressed in the placenta (Ref. Reference Ito65), even in human placental cell lines (Ref. Reference Pan66). The RAS has been assumed to regulate placental function by several studies (Ref. Reference Anton and Brosnihan67). Additionally, ACE2 is ubiquitous in the human placenta (Ref. Reference Valdés68), the expression of which is even higher than that in the lung, indicating that the placenta might be a potential target for the viral infection. Interestingly, ACE2 levels differ in various areas of the placenta (Ref. Reference Valdés68). In placental villi, ACE2 expression levels are most abundantly detected in the syncytiotrophoblast, cytotrophoblast and vascular smooth muscle of primary and secondary villi (Ref. Reference Neves69), whereas in the maternal stroma, ACE2 is found predominantly in invading trophoblasts, intravascular trophoblasts and decidual cells (Ref. Reference Valdés68). ACE2 can be detected from 6 weeks of gestation until birth, but it is also expressed differently throughout foetal development (Ref. Reference Vaswani70). It has been proven that ACE2 levels increase in early gestation but decrease dramatically in late gestation (Refs Reference Pringle71, Reference Petit72). Furthermore, the most highly expressed areas transfer from the decidual zone, luminal epithelium and glandular epithelium to the labyrinth placenta, amniotic epithelium and yolk sac epithelium during gestation (Refs Reference Neves69, Reference Ghadhanfar73).

The RAS is mainly involved in balancing vasoconstriction and vasodilation and regulating foetal development during pregnancy, and RAS components are also reported to influence several other biological processes. Ang II facilitates trophoblast invasion and angiogenesis (Ref. Reference Hering74), and the overexpression of Ang II may result in gestational hypertension, preeclampsia and eclampsia (Ref. Reference Jing37). Excessive vasoconstriction in preeclamptic women induced by high Ang II levels is attributed to the reduction of blood and nutrition supply in foetuses (Refs Reference Anton75, Reference Shibata76). Similarly, decreased serum Ang-(1–7) and increased plasma Ang II levels can be observed in women diagnosed with preeclampsia (Ref. Reference Brosnihan77). Moreover, decreased ACE2 and Ang-(1–7) levels in the placenta may induce intrauterine growth restriction (Ref. Reference Ghadhanfar73). Additionally, ACE2 knockout in mice during pregnancy can result in placental function disorders, such as placental hypoxia, and finally foetal growth retardation (Refs Reference Bharadwaj78, Reference Yamaleyeva79). Furthermore, an aberrant Ang II/Ang-(1–7) ratio is associated with premature birth (Ref. Reference Chen80) and cardiovascular disorders in adult offspring (Ref. Reference Irving81), which could be attenuated by upregulating ACE2 in rats (Ref. Reference Bessa82).

COVID-19 and female fertility

Ovaries may be a potential target for SARS-CoV-2 infection (Ref. Reference Lee36), although until now, the impact of viral infections on female fertility has been debated. Ovarian reserve is used to evaluate female fertility, and basal antral follicle count, anti-Müllerian hormone (AMH) and sex hormones, such as follicle-stimulating hormone, luteinising hormone, oestradiol, progesterone and testosterone, are the most frequently utilised indicators of ovarian reserve (Ref. Reference Moolhuijsen and Visser83). In addition, a regular menstrual cycle can also reflect ovarian reserve in women of reproductive age (Ref. Reference Younis84). Li et al. analysed the clinical data from 237 females with a history of SARS-CoV-2 infection, and they found that nearly a quarter of the participants had menstrual cycle changes, including volume and duration changes, despite similar serum AMH and sex hormone concentrations in the compared cohorts (Ref. Reference Li85). Another study reported a negative association between serum levels of both AMH and oestradiol and the severity of viral infection (Ref. Reference Ding86). However, no significant differences have been observed in women with non-severe and severe COVID-19 in terms of status, volumes or phases of menstrual cycles. Of note, COVID-19 may act as a potential risk factor for ovarian function and cause ovarian injury, including decreased ovarian reserve and hormone disturbance, in infected women (Ref. Reference Ding87).

According to previous human oocyte transcriptome and proteome databases, ACE2 and TMPRSS2, the essential molecules for SARS-CoV-2 entry into host cells, are expressed in human oocytes from the in vitro fertilisation process (Ref. Reference Virant-Klun and Strle88). Immunohistochemistry analyses in human oocytes, as well as pre- and peri-implantation embryos, have also reinforced the strong expression of ACE2 in human oocytes and blastocysts (Ref. Reference Essahib89). Nevertheless, no studies have systematically evaluated and reviewed the impacts of SARS-CoV-2 infection on human oocyte development potential to date. However, in light of the susceptibility of SARS-CoV-2 infection to early embryonic development, great attention should be paid to embryonic development potential in infected women. Whether COVID-19 might cause oocyte and embryo impairments remains elusive and needs further evaluation.

COVID-19 and pregnancy

SARS-CoV-2 infection, which constitutes a threat to both the mother and foetus, may be associated with various pregnancy and neonatal complications (Refs Reference Chen90, Reference Chen91). Reduced ACE2 levels in gravidas after infection induce a rise in placental Ang II levels, which promotes vasoconstriction in the placenta, accompanied by an increasing risk of gestational hypertension, and ultimately preterm birth and intrauterine growth restriction (Ref. Reference Chen80). A recent systematic review also concluded that gravidas with COVID-19 have a higher risk of maternal death and preterm birth, and their babies are more likely to be hospitalised in the neonatal department (Ref. Reference Allotey92). Currently, no evidence has clearly proven that COVID-19 causes placental dysfunction, whereas to avoid possible obstetric risks, more obstetricians and gravidas reportedly prefer caesarean section (Refs Reference Chen91, Reference Engels Calvo93). Additionally, because of the high expression of ACE2 in the kidney, COVID-19-associated acute kidney injury is quite frequent (Ref. Reference Ronco94), and renal failure subsequently serves as a risk factor for death in hospitalised patents, particularly critically ill patients (Refs Reference Nadim95, Reference Gabarre96). A previous study has reported viral infection in renal tubular cells (Ref. Reference Pacciarini97) and increased ACE2 levels in the kidneys of pregnant mice (Ref. Reference Brosnihan98). Thus, maternal kidney function during pregnancy in infected women is worthy of our attention.

It has been reported that foetuses born to mothers diagnosed with COVID-19 tested positive for nucleic acid identification through nasopharyngeal or anal swabs a few days after birth (Ref. Reference Zeng99). Moreover, newborns of infected women exhibited elevated serum SARS-CoV-2 immunoglobulin (Ig) M levels 2 h after birth, indicating the probable occurrence of intrauterine infection (Refs Reference Egloff100, Reference Dong101). These cases suggest that infants may be infected during gestation. Nevertheless, a systematic review of 936 neonates with maternal infection has found that only 27 of them (2.9%) had a positive viral RNA test, revealing that vertical transmission of SARS-CoV-2 has a low incidence (Ref. Reference Kotlyar102).

According to GeneCards, ACE2 exists in the female breast, providing an entry site for SARS-CoV-2 infection (Ref. Reference Jing37). A study performed SARS-CoV-2 nuclei acid identification tests in breast milk specimens from three infected females, and one of them tested positive, revealing the possibility of maternal–infant transmission by breastfeeding (Refs Reference Jing37, Reference Mahyuddin103). Moreover, the immune system of neonates has not been fully established (Ref. Reference Basha104), and close contact during breastfeeding may lead to a higher risk of potential viral infection. Two cases of neonatal infection caused by unprotected breastfeeding by new mothers diagnosed with COVID-19 have also been reported (Ref. Reference Jing37). Thus, although breastfeeding can effectively reduce the risks of neonatal infections in the respiratory and gastrointestinal systems and metabolic disorders (Ref. Reference Salone105), we still strongly recommend artificial feeding to infected mothers. If the mothers insist on breastfeeding, adequate disinfection of hands and mask wearing should also be encouraged before and during breastfeeding to minimise the chance of viral transmission through close contact (Ref. Reference Baud106). In addition, a disinfected breast pump is also recommended.

COVID-19 and human gametes and embryos

Limited studies have provided direct evidence of the impact of SARS-CoV-2 infection on human gametes and embryos until now. Wang et al. found that SARS-CoV-2 infection in females might not negatively affect female fertility and embryo development by analysing assisted reproductive technology (ART) data (Ref. Reference Wang107). The study compared the embryo outcomes of females with and without a history of SARS-CoV-2 infection via propensity score matching and found that the ovarian reserves and ovarian responses between groups were similar, as were the proportions of mature oocytes, fertilised oocytes, high-quality embryos and available blastocysts. No significant differences were found in terms of clinical pregnancy rate or miscarriage rate. Although theoretically, human oocytes and embryos are at a high risk of viral damage, much about the crisis, including the impact on fertility, remains unidentified, and evidence of the direct impact of SARS-CoV-2 infection on gametes and embryos is lacking.