REVIEW ARTICLE
Complement in Pregnancy: A Delicate Balance Kerina J. Denny1,2, Trent M. Woodruff1, Stephen M. Taylor1,2, Leonie K. Callaway2,3 1
School of Biomedical Sciences, The University of Queensland, Brisbane, QLD, Australia; School of Medicine, The University of Queensland, Brisbane, QLD, Australia; 3 Royal Brisbane and Women’s Hospital, Brisbane, QLD, Australia 2
Keywords C5a, preeclampsia, preterm birth, recurrent miscarriage Correspondence Kerina Denny, School of Biomedical Science, University of Queensland, Brisbane, QLD, Australia. E-mail:
[email protected] Submission May 15, 2012; accepted July 9, 2012. Citation Denny KJ, Woodruff TM, Taylor SM, Callaway LK. Complement in pregnancy: a delicate balance, Am J Reprod Immunol 2013; 69: 3–11
The complement system is a key component of innate host defence that, under normal conditions, is responsible for the opsonization and destruction of potential pathogens. However, inappropriate or excessive activation of complement can have a detrimental effect on the host and has been implicated in the pathophysiology of numerous disease states. Recently, there has been increasing evidence for a role of the complement system and, in particular, the potent pro-inflammatory anaphylatoxin complement component 5a (C5a) in both normal and complicated pregnancy. The following review describes the results of in vitro, animal, and human clinical studies investigating the role of the complement system in healthy pregnancy, recurrent miscarriage, preterm birth, and preeclampsia.
doi:10.1111/aji.12000
The complement system The complement system is an evolutionarily ancient component of innate immunity which, when activated, results in the recruitment and activation of immune cells leading to the rapid opsonization and destruction of pathogens.1–4 Complement activation occurs via three pathways: the classical pathway, triggered by antigen–antibody complexes; the alternative pathway which is spontaneously and continuously activated; and the lectin pathway, triggered by the binding of mannan-binding lectin to mannose residues on the surface of microorganisms. Independent of the mechanism of activation, all pathways converge to generate C3 convertases that cleave C3 into its active components, C3a and C3b. C3b is a principal effector of complement, responsible for tagging nonself cells for destruction by phagocytes. C3b additionally binds to C3 convertases to form a C5 convertase that subsequently cleaves C5 into C5b and the potent pro-inflammatory
mediator, C5a. C5b then associates with C6, C7, C8 and several units of C9 to form a lytic pore that inserts into cell membranes, the membrane attack complex (MAC; C5b-9; also known as the terminal complement complex). C3a and C5a, collectively known as the anaphylatoxins because of their originally described role in anaphylactic shock, facilitate pathogen clearance by increasing vascular permeability, inducing inflammatory cell chemotaxis, and releasing cytokines.5 C3a and C5a exert these pro-inflammatory effects by binding to their respective receptors, namely the C3a receptor (C3aR) and in the case of C5a to two receptors, C5a receptor (C5aR; CD88) and C5a receptorlike 2 receptor (C5L2). C5L2 shares homology with C5aR, although its exact biologic function remains unclear.6 Finally, various serine proteases belonging to the coagulation system are also able to activate the complement cascade, independently of the aforementioned pathways.7 Specifically, proteases such as
American Journal of Reproductive Immunology 69 (2013) 3–11 ª 2012 John Wiley & Sons A/S
3
DENNY ET AL.
thrombin can directly cleave both C3 and C5 leading to the generation of functional C3a and C5a.7 Given the multiple pathways of activation and the catalytic nature of the many steps, regulation of complement is necessarily complex. To protect against an autologous attack on self, the system makes use of both serum and cell surface regulatory proteins.8 Figure 1 summarizes both the activation and key regulatory proteins of the complement system. Complement in healthy pregnancy Healthy pregnancy is associated with a number of immunological changes compared with the nonpregnant state. One of the most thoroughly investigated immunological phenomena of pregnancy is the modulation of T-helper (Th) cell responses. Th cells can be classified into Th1 cells, which are involved in cellular immunity, and Th2 cells, which are
Fig. 1 Overview of the main effectors and regulators of the complement system. Complement is activated by the classical pathway, the lectin pathway, and the alternative pathway. The production of C3 convertase is the point at which the pathways converge and the main effector functions of complement are generated. Extrinsic proteases such as thrombin can also independently activate complement components. The complement cascade terminates with the formation of the membrane attack complex (MAC) leading to target cell lysis. Decay-accelerating factor (decay-accelerating factor), membrane cofactor protein (membrane cofactor protein), and CD59 are key complement regulators found on human placental tissue responsible for preventing inappropriate activation of complement.
involved in humoral immunity. In pregnancy, there is a ‘switch’ to predominately Th2-type immunity with relative suppression of Th1-type immune responses, an occurrence that presumably allows for tolerance of paternal alloantigens from the fetus.9 However, recent reports have argued that this Th1/ Th2 theory of immune tolerance in pregnancy is an oversimplification.10 Pregnancy is also associated with increased activation of the complement system.11–13 In both physiological and pathological pregnancy, complement products are deposited on placental tissue.14–16 In pregnancy, it is important to protect the mother and fetus against potential infection from pathogens invading the uteroplacental unit, and complement is presumed to have a key role in this at the feto-maternal interface.17–20 However, excessive complement activation in response to infection may contribute disease pathogenesis. For example, in placental malaria, complement activation in response to the malaria parasite has been hypothesized to impair angiogenesis, leading to placental insufficiency and subsequent intrauterine growth restriction.21,22 Experiments in mice underscore the importance of complement inhibition for successful pregnancy. In a seminal study by Xu et al.,23 deficiency of the murine complement receptor-1–related gene/protein Y (Crry) – a membrane-bound intrinsic complement inhibitory protein – was found to lead to universal neonatal mortality in the offspring. Mice lacking both Crry and C3 genes, however, were completely rescued, demonstrating that Crry / embryos die in utero because of an inability to suppress complement activation and consequent complement-mediated damage.23 In humans, excessive complement activation is prevented in successful pregnancy by three regulatory proteins present on trophoblast membranes – namely decay-accelerating factor (DAF), membrane cofactor protein (MCP), and CD59.24 DAF both inhibits C3convertase formation and accelerates the decay of preformed C3 convertases, whereas MCP facilitates the cleavage of C3b and C4b to their inactive forms. CD59 acts further downstream on the complement cascade, preventing the assembly of the MAC. Although some studies have demonstrated an association between levels of these regulatory proteins and pregnancy failure, the data remain scarce and inconsistent.25–27 Nevertheless, it is proposed that in the face of excessive complement activation, the capacity American Journal of Reproductive Immunology 69 (2013) 3–11
4
ª 2012 John Wiley & Sons A/S
THE ROLE OF COMPLEMENT IN PREGNANCY
of complement regulators may be overwhelmed, contributing to the development of pathological pregnancy. Complement in recurrent miscarriage Miscarriage, defined as the spontaneous loss of pregnancy before the fetus has reached viability,28 is a common complication of pregnancy, affecting approximately 15% of clinically recognized pregnancies.29 Approximately 1% of women will further suffer from recurrent miscarriage, defined as three or more consecutive pregnancy losses.28 There is increasing experimental and clinical evidence to suggest a pathogenic role for the complement system in both antibody-dependent and antibody-independent recurrent pregnancy loss. Antibody-dependent recurrent miscarriage has been extensively described in women with antiphospholipid syndrome (APS), a condition characterized by vascular thrombosis, thrombocytopenia, and recurrent fetal loss occurring in the presence of antiphospholipid (aPL) antibodies.30 APS-associated pregnancy loss accounts for 25–30% of immunologically mediated miscarriages,17 and current anticoagulant treatments are often ineffective and fraught with potential complications.31 Patients with APS have significantly elevated plasma levels of complement activation products.32 In a study by Girardi et al.,31 administration of aPL antibodies to pregnant mice resulted in complement activation, generating split products that mediated placental injury and subsequent fetal loss. C5-deficient mice (C5 / ) and wild-type mice treated with a monoclonal anti-C5 antibody or a specific C5aR peptide antagonist were protected from aPL antibody-induced miscarriage, indicating that C5aR activation is required for APS-associated pregnancy loss. In contrast, C6-deficient mice that cannot assemble the MAC are not protected from fetal death in the aPL antibody-mouse model,33 supporting the notion that the C5a-C5aR interaction is the chief effector of downstream pathogenic events. Additionally, heparin, the standard anticoagulant treatment employed in patients with APS,34 can prevent complement activation both in vitro and in vivo and protected mice from aPL antibodyinduced pregnancy complications.35 This was in contrast to fondaparinux or hirudin, anticoagulants that do not inhibit complement and were found not to prevent pregnancy loss in this model,17,35 highlight-
ing that anticoagulation alone is insufficient therapy for aPL antibody-induced miscarriage. There is also clinical evidence supporting a role for complement in the pathogenesis of antibody-independent pregnancy loss.36 Elevated plasma levels of C3 and C4 have been found to predict subsequent miscarriage in patients suffering from unexplained recurrent miscarriages.37 In contrast, a retrospective case–control study by Micheloud and others observed a relative hypocomplementemia in a group of women with recurrent miscarriage.38 Experiments in animal models support the hypothesis that complement is a causal factor in otherwise unexplained pregnancy failure. For example, in a well-established antibody-independent murine model of pregnancy loss, the DBA/2-mated CBA/J model, inhibition of complement activation results in fetal rescue.39 Notably, bacterial lipopolysaccharides (LPS) are an essential cofactor in spontaneous abortion in the DBA/2-mated CBA/J model40–43 and, in concordance with previous studies, inhibition of C5a generation also prevents LPS-driven fetal loss.44 One mechanism by which C5a may exert its harmful effects is by inducing the release of the potent anti-angiogenic factor, soluble vascular endothelial growth factor receptor-1 (sVEGFR-1, also known as sFlt-1).17 sVEGFR-1 sequesters both vascular endothelial growth factor (VEGF) and placental growth factor (PlGF), growth factors crucial for normal placental development and successful pregnancy. Circulating levels of sVEGFR-1 are elevated both in women with recurrent miscarriage and in murine models of pregnancy loss.39 In the DBA/ 2-mated CBA/J mouse, complement antagonists prevent monocyte recruitment, the rise in plasma sVEGFR-1 levels and the concordant fall in circulating VEGF and PlGF, and abnormal placentation, providing strong circumstantial evidence that sVEGFR-1 is a critical downstream effector of fetal injury in recurrent pregnancy loss. Intriguingly, administration of the lipoprotein lowering drug pravastatin has been found to significantly reduce the frequency of recurrent miscarriage in both antibody-dependent and antibody-independent mouse models45,46 – a phenomenon hypothesized to result from the anti-inflammatory effects of this drug. In support of this premise, it was found that monocytes pre-incubated with pravastatin released significantly less sVEGFR-1 in response to C5a than monocytes treated with C5a alone.46
American Journal of Reproductive Immunology 69 (2013) 3–11 ª 2012 John Wiley & Sons A/S
5
DENNY ET AL.
Similarly, pregnant DBA/2-mated CBA/J mice treated with pravastatin have reduced plasma VEGFR-1 levels.46 In summary, in both women with unexplained recurrent miscarriage and in women with recurrent pregnancy loss secondary to APS, uninhibited activation of complement, particularly C5a, is likely to contribute toward fetal rejection. Furthermore, it is proposed that C5a may exert its deleterious effects by means of impairing placental angiogenesis. Whether C5a inhibition alone is sufficient to prevent recurrent pregnancy loss in the clinical setting however remains to be realized. Complement in preterm birth Preterm deliveries – defined as those that occur at
