The complement system

The complement system is a main branch of the humoral immune response that plays an important role in the elimination of pathogens and altered self-structures. It may be activated through three distinct pathways: the classical, lectin and the alternative pathways. The former two are triggered by pattern recognition receptors bound to antibodies or surface-carbohydrate structures, while the alternative pathway (AP) is continuously activated by a low-rate self-activation of the C3 molecule. The three activation cascades converge on the level of C3 which is cleaved into the C3a and C3b fragments by the C3-convertase complexes. The C3b fragment attaches to foreign surfaces, initiates further C3 activation and subsequent terminal pathway activation, which causes inflammation and tissue damage (Fig. 1). Since the AP is continuously activated, its inefficient regulation may lead to overactivation of the complement system with substantial endothelial injury, such as seen in thrombotic microangiopathies (TMAs).

Fig. 1
figure 1

Basic mechanisms of alternative pathway activation and complement regulation. a Complement fragment C3b binds to surfaces and factor B, which is then cleaved by factor D. The newly formed C3bBb complex (alternative pathway C3-convertase) is able to cleave multiple further C3 molecules. Thus, the alternative pathway is not only capable of spontaneous activation, but can enhance complement activation triggered through any pathway, functioning as a positive feedback loop. b Human surfaces are protected from the detrimental effects of complement activation by numerous surface-bound (MCP, DAF, thrombomodulin) and soluble (factor H and I) regulators. Factor I cleaves and inactivates C3b in the presence of its cofactors, factor H and MCP; thrombomodulin potentiates C3b cleavage. Factor H binds to sialic acid present on human cell membranes, but not on most pathogens, and serves as a cofactor for factor I. In addition, factor H and DAF facilitate the decay of the C3-convertases. c If the above regulation is defective, excess C3b binds to C3-convertases, turning them into C5-convertases, which cleave C5 to C5a and C5b, initiating the common terminal pathway. C5a is a potent anaphylatoxin (similar to C3a), while surface-bound C5b assembles the C6-9 molecules, forming a pore in the membrane, called the membrane attack complex. MCP membrane cofactor protein, DAF decay accelerating factor, THBD thrombomodulin. Complement factors are marked by letters. Complement components C6–C9 are marked by numbers only

Full size image

Forms of TMA

TMAs are rare but life-threatening disorders characterized by microvascular hemolytic anemia and acute thrombocytopenia with or without organ damage (for example signs of neurological or kidney injury). While the bed-side, immediate diagnosis of TMA is based on the clinical picture and routine laboratory results, it may cover various subgroups of diseases, the pathogenesis of which is shortly summarized in Table 1. Despite similarities in the initial presentation, TMA differential diagnostics has an impact on both long-term patient follow-up, and the determination of the optimal therapeutic choice from early on. Hemolytic uremic syndrome (HUS) usually presents with the triad of hemolytic anemia, thrombocytopenia and acute renal failure (a minority of patients may suffer from neurologic symptoms as well). Typical HUS (90–95% of the cases) is preceded by a gastrointestinal infection with Shiga-toxin-producing bacteria, and it represents the most common cause of acute kidney failure in children. Atypical HUS (aHUS) may also present with gastrointestinal symptoms that could be misleading at the initial phase; however, endothelial injury in aHUS mainly results from the dysregulation of the alternative complement pathway.

Table 1 Key pathogenetic factors and the potential role of complement in various TMA forms
Full size table

Patients with thrombotic thrombocytopenic purpura (TTP) usually show characteristic clinical symptoms such as critical thrombocytopenia and microangiopathic hemolytic anemia along with neurological symptoms, mental status changes, or sometimes signs of kidney injury. All are severe conditions that require immediate attention and diagnostic workup for complement abnormalities. To underline the characteristics of these subgroups, their distinct pathogenic features are detailed in this review, with emphasis on the role of complement (Table 1) and on the potential novel biomarkers (Table 2) of these diseases.

Table 2 Characteristic laboratory findings and potential biomarkers of various TMA forms
Full size table

Complement mediated atypical hemolytic uremic syndrome (aHUS)

A number of different alterations in the complement genes encoding activators and regulators of the AP are known to be directly associated with relapsing (atypical) HUS. Since the first description of a causative factor H mutation [1], numerous further disease-causing mutations have been described in the complement genes, such as CFH, CFI, CD46, C3, CFB, THBD and CFHR5 [2,3,4]. All these pathogenic variants account for around 50–60% of aHUS cases [4] but in the remainder of patients no disease-causing variations can be identified. This observation is consistent with the fact that the combined presence of a genetic predisposition and environmental trigger factors is needed to provoke an aHUS episode [3]. Pathogenic aHUS-associated mutations in the complement regulator genes lead to overactivation of the AP. CFH mutations mostly affect the C‑terminus of the protein, which is responsible for binding to surface-associated C3b and self-specific glycosaminoglycans. Hence, the altered regulator is unable to recognize its targets and to act as a cofactor of factor I mediated C3b degradation (Fig. 1). This results in impaired complement regulation on host cell surfaces and, upon a triggering event, may lead to extensive C3b deposition on the vascular endothelium, thus, inducing cellular injury and microvascular thrombosis. Disease-causing mutations in CD46 mainly account for reduced expression or decreased binding to C3b, whereas mutations in CFI lead to impaired C3b inactivation. On the other hand, gain-of-function mutations in C3 or CFB result in a hyperactive, regulation-resistant C3-convertase.

It is noteworthy that the mutation penetrance in pedigrees is incomplete, which supports the hypothesis that the additive effect of multiple mutations and risk variations is often necessary to cause a clinically manifest disease. This together with the high number of patients with unidentified disease-associated genetic alterations highlights the fact that additional, yet unexplored genetic abnormalities may also contribute to the development of aHUS.

Genetic predisposing factors are also present in the autoimmune form of aHUS, where autoantibodies directed against factor H are the key pathogenic factors of disease manifestation. This TMA subgroup is strongly associated with the polymorphic homozygous deletion of the complement factor H-related gene CFHR1 [5].

STEC-HUS

Diarrhea-associated or typical HUS is the most common form of HUS. The condition is usually evoked by a Shiga toxin (Stx)-producing Escherichia coli (STEC) infection. Its initial symptoms are related to the bacterial colonization of the gastrointestinal tract causing intestinal inflammation and—often bloody—diarrhea. Stx1 and Stx2 released by the adhered bacteria are the primary cause of microangiopathy through their globotriaosylceramide (Gb3) receptor mediated internalization and blockade of protein synthesis within the endothelial cells [6].

Accordingly, STEC-HUS is primarily not a complement-mediated disorder; however, increased levels of the complement-degradation products C3a(desArg), C3d, Bb, C3bBbP and sC5b-9, detected in the circulation during the acute phase of the disease provide clear evidence for an increased complement activity in this form of HUS [7,8,9]. This can most probably be attributed to the direct or indirect effects of Stx and bacterial lipopolysaccharide (LPS) on complement activation and coagulation. In vitro studies demonstrated that Stx can activate the alternative pathway in the fluid phase, while upon binding to the surface recognition sites of factor H, Stx may delay its inhibitory effect and promote complement activation on the cellular surface [10]. Besides, Stx—particularly in the presence of LPS—was shown to induce the formation of platelet–leukocyte aggregates and the release of blood cell-derived microparticles coated with C3 and C9 [8, 11]. Furthermore, Stx was shown to promote the upregulation of the membrane adhesion molecule P‑selectin on microvascular endothelium, which—by acting as a C3b-binding protein—increases C3 deposits and favors platelet thrombus formation, thus increasing the circulatory C3a level [12]. Involvement of the alternative pathway in the microvascular processes was also supported by in vivo experiments: Thrombotic effects of Stx/LPS treatment could be diminished in factor B‑deficient mice or could be inhibited by the admission of a C3a receptor antagonist [12] in animal models of STEC-HUS.

Thrombotic thrombocytopenic purpura (TTP)

TTP is caused by the deficiency of the ADAMTS13. The role of ADAMTS13 is to cleave the ultra-large form of von Willebrand factor (ULVWF), which is secreted by activated endothelial cells, and can spontaneously bind and activate platelets.

ADAMTS13 deficiency is necessary, but not enough, to provoke TTP, since ADAMTS13 deficiency may also be present in its convalescence. Similarly to aHUS, the onset of TTP is often associated with a triggering event. Endothelial activating conditions like pregnancy or infections may cause expression of ULVWF from endothelial cells, which, in combination with ADAMTS13 deficiency, may result in the increased presence of ULVWF molecules and consequent initiation of platelet thrombus formation. Activated platelets and endothelial cells express P‑selectin, which is able to bind C3b and activate the complement system [13]. ULVWF is also able to directly bind complement factors and trigger complement alternative and terminal pathway activation [14], thus, leading to increased levels of complement activation products C3a, C5a, and sC5b9 in acute phase TTP patients [15, 16]. These complement activation products further activate endothelial cells, initiating the vicious circle of increased ULVWF and P‑selectin expression, and decreased thrombomodulin expression [14]. Furthermore, complement activation leads to granulocyte activation [17] and subsequent cellular adherence to the endothelium facilitated by the increased P‑selectin expression. Production of reactive oxygen species and proteases by the attached granulocytes further enhances endothelial dysfunction [18]. In summary, complement activation is part of an amplification loop: it augments the prothrombotic changes in the microvasculature that leads to a full-blown thrombotic microangiopathy (Fig. 2).

Fig. 2
figure 2

Summary of thrombotic microangiopathy (TMA) pathogenesis. a In case of ADAMTS13 deficiency, uncleaved ultra-large form of von Willebrand factor (ULVWF) provides a surface for platelet aggregation and thrombus formation. b Antibody–antigen complexes, ULVWF, and other molecules on activated endothelial cells and platelets trigger complement activation. c, d The activated terminal pathway can in turn activate neutrophil granulocytes and facilitate their binding to endothelial cells, leading to endothelial injury and prothrombotic changes in the endothelium. The activated neutrophil cells can release neutrophil extracellular traps (NET), which provides a surface for thrombus formation and complement activation. The above events are present to a different extent in distinct forms of TMA, with the complement system connecting them to form a vicious circle. Numbers in circles indicate the order of events in each section. C5b-9 C5b-9 complex, or membrane attack complex, ET-1 endothelin‑1, WPB Weibel-Palade body

Full size image

Secondary TMA

Secondary forms of TMA represent a heterogeneous group of disorders that all emerge on the basis of a preexisting condition. Secondary TMA may be associated with infections and septic conditions, allogenic hematopoietic stem cell (HSC) or solid organ transplantation, systemic autoimmune diseases, pregnancy, tumor progression or malignant hypertension. Even though their etiology may vary, it is common in secondary TMA that overactivation and subsequent consumption of both classical and alternative pathway complement components, and decreased ADAMTS13 activity are present in these conditions.

The involvement of the complement alternative pathway dysregulation has been recently suggested in the pathogenesis of post-HSCT-TMA. In this condition, an elevated systemic sC5b-9 level was associated with worse long-term outcome [19]. Gloude et al. also suggested that chemotherapy, radiation and infections leading to endothelial injury during HSCT provoke complement activation through neutrophil activation and neutrophil extracellular trap (NET) release [20, 21]. In line with these findings, increased activation of all three complement pathways was observed in our series of secondary TMA patients with various etiologies and elevated sC5b-9 and C3a concentrations were associated with a poor patient outcome [22]. Although the pathophysiology of these TMA forms has not entirely been explored, the clear involvement of complement may support future plans to study complement inhibitors such as eculizumab in these conditions.

Discussion

Our understanding of the pathophysiology and characteristic course of various TMA forms has improved in recent years with novel genes, pathways and mechanisms described as a result of intensive research of this field. As highlighted in this review, complement activation may represent an important amalgamating process in all of these conditions, since it is able to link activation and damage of multiple involved cell types, such as endothelial cells, platelets, and neutrophils (Fig. 2). The recent knowledge on the pathophysiology of TMAs was translated into clinical use and has reached the clinical care of patients, too, since more and more laboratories provide appropriate tests for clinical diagnostics (Table 1). In addition, future research will help to clarify if biomarkers of the above described pathways (Table 2) are appropriate tools for prediction of disease exacerbation or severity.

The current management of various TMAs largely relies on supportive care, infection control, immunosuppression, cytostatics and therapeutic plasma exchange. There are only a few targeted therapies available for TMA patients that include B‑cell depletion by the anti-CD20 monoclonal antibody rituximab, inhibition of platelet adhesion by caplacizumab, a nanobody targeting von Willebrand factor administered in TTP or complement inhibition with anti-C5 monoclonal antibody eculizumab for patients with aHUS. The accumulated knowledge on the role of neutrophils, endothelial cells, platelets, and the complement system in the pathophysiology of TMAs may open new avenues for research on additional targeted therapies, including the blockage of neutrophil activation and degranulation with colchicine, inhibition of complement activation (for example with drugs limiting C3 activation and alternative pathway amplification) or its action (such as C5a receptor blockade), or preparations that restore endothelial function.