- Email: [email protected]
Mechanisms of Complement-Mediated Damage in Hematological Disorders
Author’s Accepted Manuscript
image
Mechanisms of complement-mediated damage in hematological disorders Ronald P. Taylor, Margaret A. Lindorfer
www.elsevier.com/locate/bios
PII: DOI: Reference:
S0037-1963(18)30005-2 https://doi.org/10.1053/j.seminhematol.2018.02.003 YSHEM50938
To appear in: Seminars in Hematology Cite this article as: Ronald P. Taylor and Margaret A. Lindorfer, Mechanisms of complement-mediated damage in hematological disorders, Seminars in Hematology,doi:10.1053/j.seminhematol.2018.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Mechanisms of complement-mediated damage in hematological disorders Ronald P. Taylora and Margaret A. Lindorfera aDepartment
of Biochemistry and Molecular Genetics, University of Virginia School
of Medicine, Charlottesville, VA 22908 USA Address all correspondence to Dr. Ronald P. Taylor, Department of Biochemistry and Molecular Genetics, University of Virginia School of Medicine, 1340 Jefferson Park Avenue, Charlottesville, VA 22908 USA 1-434-924-2664 (voice) 1-434-9245069 (fax) email: [email protected] Declarations of interest: none
Abstract The complement cascade is an ancient defense system that destroys and eliminates threats to normal homeostasis in the bloodstream and tissues. Although multiple controls keep complement in check to minimize innocent bystander injury to normal cells and tissues, defects in complement regulation due to mutations in, or autoantibodies to, complement control proteins underlie the pathogenesis of several hemolytic diseases including paroxysmal nocturnal hemoglobinuria, and atypical hemolytic uremic syndrome. In autoimmune hemolytic anemias complement plays an important role in erythrocyte destruction mediated by anti-erythrocyte antibodies. The pathogenic mechanisms of these hemolytic diseases are discussed, with an emphasis on pivotal steps in complement activation. Abbreviations: aHUS, atypical hemolytic uremic syndrome; AIHA, autoimmune hemolytic anemia; AP, alternative pathway of complement; CAD, cold agglutinin disease; C1s*, C1r*, activated forms of C1s and C1r respectively; CFHR proteins, complement factor H related proteins; CLL, chronic lymphocytic leukemia; CP, classical pathway of complement; DAF, decay accelerating factor, CD55; IC, immune complex; MAC, membrane attack complex; MBL, mannan binding lectin pathway of complement; MCP, membrane co-factor protein, CD46; MIRL, membrane inhibitor of reactive lysis, CD59; PNH, paroxysmal nocturnal hemoglobinuria; THBD, thrombomodulin.
1
1. Introduction The complement cascade is an ancient anti-microbial defense system that has evolved to synergize with several facets of the vertebrate immune system [1-7]. Complement provides an early warning system of “first responders” that react, in a rapid and exponential manner, to destroy and eliminate threats to normal homeostasis in the bloodstream and in tissues. Key elements in the initiation of complement activation include plasma proteins which specifically recognize dangerous microorganisms (exogenous threats) as well as potentially noxious apoptotic cells and cellular debris (endogenous threats). Complement also provides defense in association with adaptive immunity by rapidly clearing antigen-antibody immune complexes (IC). Regardless of the initiating event, the downstream consequences of complement activation are characterized by numerous cytotoxic reactions and production of several potent mediators of inflammation. As a defense system complement works quite effectively in eliminating pathogens, but this incendiary system also has real potential for causing injury to host cells and tissue. In fact, low level, non-specific activation of complement in normal cells and tissue can occur in the absence of frank external stimuli. There is also the possibility of the generation of cytotoxic factors against normal tissue when nearby pathogens are targeted by complement. Microorganisms that are recognized as bonafide threats by complement are subject to substantial levels of complement activation. For these agents several pivotal steps in the complement cascades easily reach and surpass activation thresholds. However, intrinsic to the complement system is a series of controls that provide defense of normal cells and tissue against these reactions [13]. Indeed, complement is most efficiently controlled based on thresholds for several different reaction steps in the pathways. Under normal homeostasis these complement controls maintain the level of complement activation on normal cells and tissue considerably below threshold values. However, in most complementmediated hemolytic diseases, complement is not properly regulated [8]. In this chapter we will focus on key defects in complement control mechanisms in hemolytic diseases, in which complement is inappropriately activated and pivotal thresholds are exceeded, thus leading to substantial damage to normal cells and tissues. 1.1 The Classical Pathway (CP) There are three pathways in which molecular challenges/stimuli promote complement activation, and all three are characterized by sequential proteolysis steps. Figure 1 is an abbreviated representation of the complement cascade that highlights key control points which are most important in hemolytic diseases. In the CP, C1q, a component of the C1 complex binds to IC containing IgG1, IgG3 or IgM [1,2,9]. C1q (as part of the C1 complex) can also bind directly to certain substrates and activate the CP on targeted cells in the absence of bound immunoglobulins; these substrates include bacterial lipopolysaccharides, C-reactive protein, pentraxins, and degradation products of cells, e.g. DNA. As a consequence of C1q 2
binding, C1r and C1s are consecutively activated in proteolytic steps, and activated C1s (now a protease) cleaves C4 to C4a and C4b. In nascent activated C4b, an internal thioester bond is exposed to solvent and therefore C4b can react with water or form a covalent amide or ester bond with acceptor sites on the IC, thus additionally tagging it as “foreign”. The C4bC1s complex then cleaves C2, and the newly-generated C4bC2a complex, the CP C3 convertase (C2a is a protease), cleaves C3 to produce C3a and C3b. In activated C3b an internal thioester bond is also exposed to solvent which can either react with water or form a covalent bond with acceptor sites on the IC. In solution phase the classical pathway is subject to downregulation due to the action of factor H (see Section 1.4) as well as due to the dissociation and dilution of labile newly-formed complexes. This phenomenon is illustrated in the enhanced ability of the CD20-specific mAb ofatumumab to activate complement on binding to the surface of B cells compared to the action of the CD20specific mAb rituximab [10,11]. Ofatumumab binds to a target site on CD20 much closer to the cell membrane than the site targeted by rituximab. When cells are opsonized with ofatumumab there is more effective concentration of activated complement components on the cell membrane, demonstrated by more successful covalent binding of nascent C4b and C3b to the targeted cell, and decreased nonproductive reaction of these components with water. These findings have now been extended and generalized in model systems which revealed that mAbs specific for CD20 and CD52 were most effective in promoting CDC of cells in which the targets were expressed closest to the cell membrane [12]. If there is sufficient (threshold) C3 activation, and enough C3b is able to covalently opsonize the substrate IC, the C5 convertase (C4bC2aC3b) will assemble (C2a continues to provide protease activity) and cleave C5 to C5b and C5a. The newlyformed C5b, released at the target surface, can then consecutively attract molecules of C6-C9 to produce the membrane attack complex (MAC). If a sufficient number of MAC are generated and penetrate the cell membrane, the cell will be killed due to rapid influx of Ca+2 [13]. Inherent in this pathway is a series of amplification steps, due to the multiple enzymes that are produced from their zymogen precursors [9]. The C3b activation step is particularly important with respect to destruction of cells because if threshold amounts of C3b are bound to the substrate cell, the cell will become susceptible to phagocytosis by effector cells that express receptors for C3b or for its downstream fragment iC3b. Thus, even if the cell under attack is not killed by the MAC, C3b opsonization can still lead to its elimination. This is one of the main mechanisms by which bacteria are successfully targeted by complement; after opsonization with C3b and iC3b, neutrophils and macrophages eliminate them by phagocytosis [1,4,5]. Other important inflammatory agents produced at the C3 and C5 activation steps are C3a and C5a, respectively, anaphylatoxins that recruit and activate effector cells at sites of complement activation. Chelation by C5a of the C5a receptor on effector cells (e.g. mast cells, monocytes, macrophages, neutrophils) enhances the “inflammatory poise” of the cells, which is manifested by increases in expression of activating Fcγ receptors and complement receptor 3 and down-modulation of the 3
inhibitory receptor FcγRIIb [3,14]. The subsequent cytotoxic inflammatory responses include phagocytosis and release of cytotoxic agents such as superoxide, hydrolytic enzymes, cytokines, and chemokines [2]. C5a can also induce endothelial cell activation by promoting release of histamine from basophils and mast cells, thereby increasing vascular permeability and contributing to vascular injury. 1.2 The Alternative Pathway (AP) The AP is initiated by spontaneous reaction of small amounts of C3 with water (~0.3% per hour) [15,16]. This fluid phase endogenous tickover reaction involves a conformational change in C3 comparable to that which occurs when C3 is converted to C3b, thus exposing the thioester bond to water and generating C3(H20) (Figure 1). Factor B then binds to C3(H20) and is cleaved and activated by Factor D, generating C3(H20)Bb. Bb within this complex has proteolytic activity, and the complex can bind and cleave C3, giving rise to nascent C3b. Then, C3b can either react with water, or form a covalent bond with surfaces of bacteria and yeast which can active the AP, or with PNH E that cannot down-modulate the AP. As noted earlier, under normal conditions the fluid phase activation of complement is also more easily down-regulated due to dilution and dissociation of reacting factors, but the AP will be considerably more active when a “safe harbor” on a substrate surface forms the nexus for continued complement activation. Indeed, under these conditions substrate-bound C3b can bind additional Factor B, and in the presence of Factor D produce more Bb, thereby giving rise to C3bBb (the AP C3 convertase) in an autocatalytic pathway capable of rapid production of substrate-bound C3b molecules (Figure 1, amplification step). This convertase is metastable, but can be stabilized by Properdin. C3b produced during CP activation can also feed into the AP, contributing even more “fuel” to the AP C3 convertase. In analogy to the requirements in the CP, if enough C3b is produced and captured by the AP C3 convertase, the C5 convertase of the AP is formed (C3bBbC3b) which cleaves C5 to produce C5a and C5b, thus allowing for production of the MAC. All complement convertases are metastable and decompose rapidly, and this insures some natural control of complement activation and presumably limits possible innocent bystander damage to nearby host tissue. Any agents or gain of function mutations (e.g. in C3 or Factor B) that tend to stabilize the convertases could have the effect of promoting complement-mediated damage to normal cells and tissue, and several of these mutations play primary roles in induction of certain hemolytic anemias (see Section 3).
4
1.3 The Mannan Binding Lectin Pathway (MBL) This pathway is a “hard-wired” innate pathway specific for carbohydrates expressed on microorganisms including yeast and bacteria [1,7]. In analogy to the CP, after the recognition molecules (MBL, ficolins, collectins) bind to the substrate surface, mannan-binding lectin associated proteases are activated, and their actions then feed into the classical pathway at the C4 activation step. There is little evidence that this pathway plays an important role in hemolytic diseases. 1.4 Control of Complement Almost all steps in complement activation are subject to control by either soluble factors or by membrane-associated proteins on potentially vulnerable cells (Figure 1). Loss-of-function mutations in these proteins or inhibition of their activities due to autoantibodies can lead to hemolytic diseases. As noted, both the CP and MBL pathway feed into the AP, and the AP promotes exponential and rapid amplification of complement activation starting at the C3 stage. Therefore, it is not surprising that several severe diseases associated with defective control of complement can be attributed to lesions that induce loss of control of the AP [3]. Factor H, a soluble protein, is the most important regulator of the AP [1,2,7,17]. By binding to C3b (and competing with Factor B) it accelerates decay of the AP C3 convertase. In addition Factor H serves as a co-factor for Factor I, a protease that directly degrades C3b, converting it to iC3b, its inactive form [1,16]. Factor H also down-regulates complement on cell surfaces; it binds to cell membranes containing negatively charged molecules including sialic acid and glycosoaminoglycans such as heparin sulfate, thus enabling it to recognize and bind to normal “self” cells that express negatively charged molecules [2,3]. Therefore, cell-associated Factor H serves a sentinel function to down-regulate incipient complement activation after C3b deposition on self-tissue. On the other hand, when covalent deposition of C3b is initiated on susceptible pathogens lacking negative cell-surface molecules, Factor H does not bind to such non-self-tissue or block C3b deposition, and therefore the pathogen becomes rapidly coated with C3b, exceeding the thresholds needed to initiate activation of C5 or to promote phagocytosis mediated by complement receptors on effector cells. That is, Factor H “ignores” these microorganisms and does not interfere with C3b deposition, therefore the AP is free to deposit C3b at a rapid clip, and to continue downstream to production of the MAC. Five similar proteins that lack the first four N-terminal complement regulatory domains of Factor H (CFHR1 to CFHR5), as well as a spliced variant of factor H [FHL-1] have been described and the potential role of these proteins in either complement control or activation in several diseases including aHUS (see section 3) is under active investigation [18]. Two important plasma membrane-associated complement control proteins that are expressed on blood cells are CD55 (DAF, decay accelerating Factor) and CD59 (MIRL, membrane inhibitor of reactive lysis) [15,16,19]. CD55 binds to C4b and 5
interferes with formation of the C3 CP convertase, and also accelerates decay of the C3 and C5 convertases of the AP and CP. CD59 blocks formation of the MAC by binding to C5b-8 and preventing incorporation of C9. When these control proteins are not present or are expressed at lower levels on blood cells (see section 2), then the cells are subject to attack by the AP, and at best soluble Factor H can slow down the attack, but not block it [20]. CD46, (MCP, membrane co-factor protein) is expressed on most cells other than E, and serves as a co-factor for Factor I in the conversion of active C3b to inactivated iC3b, thus down-regulating complement activation on cells and surfaces [21]. Finally, thrombomodulin, an endothelial cellassociated protective protein, regulates complement activation in the vasculature [22,23]. First, it enhances the activity of Factor I and its associated co-factors in degrading C3b to iC3b. In addition it down-modulates production of C5a by inhibiting the action of thrombin (which can directly activate C5 to C5a and C5b). It also stimulates synthesis of carboxypeptidase B, which degrades C3a and C5a, thus inhibiting their activities. 2. Paroxysmal Nocturnal Hemoglobinuria (PNH) The pathogenesis of PNH has been the subject of intense investigations which have aided in the elucidation of several molecular details of the complement cascades [24-27]. PNH is most readily characterized by severe anemia (Coombs negative), due to complement-mediated intravascular hemolysis of E. The plasma membranes of E of patients with PNH have substantially reduced levels of complement control proteins CD55 and CD59 which renders the cells quite sensitive to AP-mediated lysis, especially at lower pH. Indeed, the tickover reaction of the AP, including its potential for exponential activation, was elegantly demonstrated by MuellerEberhard’s group based on examination of the hemolysis of PNH E in serum at low pH under conditions in which only the AP could be activated [15,25]. Studies led by T. Kinoshita established that the origin of sensitivity of PNH E to complement could be traced to acquired (but rare) somatic mutations in one or more multipotent hematopoietic stem cells that gave rise to E as well as other blood cells [28]. The mutation is in the X-chromosome phosphatidylinositol glycan class A (PIG-A) gene which codes for a key step in synthesis of glycosyl phosphatidylinositol (GPI) groups that serve to anchor certain proteins to cell membranes. Hematopoietic stem cells with these mutations appear to have a paradoxical survival advantage due to immune escape from T cell-mediated attack [29-31]. Progeny derived from these mutant stem cells, which include E, platelets, monocytes and granulocytes, have considerably reduced levels of CD55 and CD59 which normally are attached to cells with the GPI link. Three populations of E, at varying levels, are demonstrable in PNH: PNH I E, normal expression of CD55 and CD59; PNH II E, with considerably reduced levels of CD55 and CD59; PNH III E, no expression of either CD55 or CD59. PNH is quite often associated with aplastic anemia, which may be related to the original lesion in the bone marrow that leads to generation and survival of the mutant cells [26]. In view of the marked action of Factor H in down-regulating complement activation on cell surfaces, one might 6
expect that E lacking CD55 and CD59 would still be adequately protected from complement by Factor H. In fact, Ferreira and Pangburn reported in vitro experiments demonstrating that Factor H provides some protection of PNH E from complement-mediated lysis, but in vivo the protection is not adequate [20]. 2.1 Downstream consequence of hemolysis in PNH The major cause of mortality in PNH is due to thromboses, which are most likely produced as a consequence of a “perfect storm” of inflammation associated with both complement activation and E lysis followed by the release into the circulation of procoagulant E microparticles, free hemoglobin and its breakdown product heme [2,24,32,33]. The link between hemolysis and thromboses is not limited to PNH, and appears to be complex and not fully understood. However, as noted by Cappellini, “hemolysis per se, whatever the cause, seems to be a pro-coagulant condition” [34]. Although the body has mechanisms for clearing hemoglobin and heme from the bloodstream, these processes can be saturated and overwhelmed by high levels of hemolysis [32,33]. Free hemoglobin in the circulation can promote several inflammatory reactions leading to vascular injury, many of which are induced due to consumption of nitric oxide by hemoglobin. Moreover, free heme can also activate the AP, thus producing C3a and C5a, which activate and aggregate platelets, especially because platelets lack, or poorly express, CD55 and CD59 in PNH. This high level of complement activation and inflammation, including production of the MAC, are also likely to damage endothelial tissue [35], thus additionally contributing to thrombosis. There are additional links between the complement system and the clotting system [7]. Ritis et al have reported in a model system that complement activation and the production of C5a causes neutrophils to express Tissue Factor, which initiates the blood clotting cascade [36]. The investigators suggested that with regard to normal defense, e.g., response to a bacterial infection, formation of a blood clot would serve to contain and prevent spread of the infection. Recent studies by Gremmel et al have demonstrated that ADP released from lysed E can also directly activate platelets and promote their aggregation [37]. 2.2 PNH and Eculizumab The neutralizing anti-C5 mAb eculizumab has proven to be a potent agent for treatment of PNH and aHUS by blocking activation of C5 to C5a and C5b [27]. Indeed, treatment of PNH patients with eculizumab reduces intravascular hemolysis and the risk of thromboembolisms, thus providing a link between E lysis and downstream coagulation [35]. Moreover, observations on patients treated with eculizumab speak to the kinetics of complement activation and down-regulation: In vitro reaction of PNH E with acidified serum in the presence of eculizumab leads to deposition of C3b/iC3b fragments on the cells. After a 24 hour reaction in vitro these fragments decay to C3d, but the “transition” from C3b/iC3b to C3d in vivo is much faster. When PNH patients receive eculizumab therapy only C3d is detected on E [8,38,39]. We suggest that if blood samples are taken immediately after
7
complement activation that is triggered subsequent to eculizumab blockade, then it should be possible to detect C3b/iC3b-opsonized PNH E. The classic studies of Atkinson and Frank on the fate of IgM-opsonized, C3b/iC3bcoated E demonstrated that these E are cleared from the bloodstream due to phagocytosis mediated by macrophages with receptors for C3b/iC3b [40]. It is likely that during eculizumab therapy, the PNH E that are heavily coated with C3b/iC3b are cleared similarly. Indeed, Risitano’s group has provided clear evidence that after eculizumab therapy, extravascular clearance of PNH E is mediated by the spleen [38]. However, for PNH E that are not cleared (perhaps due to lower levels of opsonization) the C3b/iC3b likely rapidly decays to C3d and these E continue to circulate. C3dg-coated E are also subject to phagocytosis mediated by complement receptors on macrophages [41], but this may be a less efficient process because C3dtagged E are demonstrable for some time in the circulation of eculizumab-treated PNH patients [8,39]. These phenomena are quite similar to the reactions that occur when the anti-CD20 mAbs rituximab or ofatumumab are infused into the circulation of patients with chronic lymphocytic leukemia (CLL). Although very early on the CLL B cells contain C3b/iC3b fragments, the transition to C3d is quite rapid [10]. 2.3 Role of CD59 Conventional wisdom argues that the fundamental defect in protective mechanisms in PNH E has been elucidated: decreased levels of both CD55 and CD59 presage complement-mediated hemolysis [8,26,27]. Lack of CD55 allows build-up of threshold amounts of C3b activation fragments, and then E, heavily opsonized with C3b, are excellent substrates for the AP C5 convertase, allowing deposition of lethal amounts of the MAC. However, the situation may be more complicated. Yamashina et al reported on a patient with a PNH phenotype who had an inherited complete deficiency of CD59 but normal levels of CD55 [42], and the patient’s E were lysed in acidified serum. The question is: why did CD55 not “protect” the patient’s E by blocking C3b deposition, thus preventing downstream generation of the MAC? Parker’s group reported that E from patients with the rare Inab phenotype (lack of CD55) could be lysed in acidified serum when the action of CD59 was also blocked [19]. They found that C3b deposition on the E was increased considerably, suggesting that CD59 may “modulate” the C3 convertase of the AP. Recently Risitano and Notaro found that when the activity of CD59 on normal E is blocked with a neutralizing mAb, and the E are reacted in acidified serum with Mg-EGTA containing eculizumab, then C3b fragments are deposited on the E via the AP [43]. This result suggests that active CD59 can inhibit C3b deposition on E, a provocative finding at odds with a substantial literature. When CD59 is blocked, sublytic attack by the MAC on the E membrane may expose phosphatidyl serine on the cell’s outer membrane, which could lead to activation of the AP and C3b deposition [44,45]. Resolution of this question may prove to be most interesting and may lead to new paradigms for controlling complement activation in certain diseases.
8
3. Atypical Hemolytic Uremic Syndrome (aHUS) The thrombotic microangiopathies include thrombotic thrombocytopenia purpura, (TTP), hemolytic uremia syndrome (HUS) and its atypical form (aHUS). Due to space limitations we will only discuss aHUS. In this disease there is substantial pathology associated with capillary injury as a consequence of platelet aggregation and inflammation concentrated on endothelial cells [46-48]. Moreover, E are damaged during transit within the compromised and narrowed microvasculature, giving rise to non-immunological, Coombs-negative “mechanical” hemolytic anemia in which schistocytes (damaged E) are demonstrable in peripheral blood smears [32,46,48]. As in PNH, hemolysis of circulating E leads to release of large amounts of hemoglobin followed by downstream elaboration of heme, both of which contribute to inflammation due to nitric oxide consumption, complement activation and increased vascular injury [32,33]. In aHUS, defective AP regulation on surfaces can occur as a consequence of several independent factors. The most common lesion is a mutation in Factor H which inhibits its ability to bind to surfaces [49,50]. As a consequence, tickover-mediated deposition of C3b on susceptible endothelial cells progresses to threshold levels of C3b deposition, allowing for generation of C5b, and subsequent damage by the MAC. In aHUS this pathologic mechanism is most prominently associated with kidney injury; Pickering et al were able to replicate this pathology in an aHUS mouse model by engineering Factor H lacking surface recognition domains [51]. Certain segments of the kidney express low levels of membrane-bound complement control proteins [52], and therefore proper functioning of Factor H may be necessary to protect the kidney from inappropriate complement activation. Variations or outright deletions of certain of the CFHR proteins can also affect the activity of Factor H and lead to aHUS. For example, loss of expression of CHFR1 and CFHR3 appears to induce generation of autoantibodies to Factor H, and subsequent development of aHUS [17]. Recently Csincsi et al reported that CFHR5 has several other activities, in addition to competing with Factor H for C3b binding, all of which should increase complement activation in the kidney [53]. These activities include blocking binding of Factor H to pentraxin, C-reactive protein, and the extracellular matrix ,and promoting formation of the AP C3b convertase. These activities, of either wild-type or mutated forms of CFHR5, may play a role in development of aHUS. The phenotype and tissue damage observed in patients with aHUS can be induced by mutations affecting the function of other complement control proteins that regulate the AP including Factor I, MCP (CD46), and thrombomodulin [17,22]. In addition, gain of function mutations in Factor B or C3, which increase the activity of the AP on surfaces can also give rise to aHUS [54,55]. There are potentially redundant independent factors that regulate the AP, and it is not surprising that with respect to aHUS, the disease penetrance of these mutations averages about 50% [2,4,17]. Therefore, if only one complement control factor is defective, other complement control factors may compensate, and disease may not be expressed. Alternatively, environmental triggers may be required to convert an aHUS 9
predisposing factor (e.g., Factor H mutation or autoantibodies to Factor H) into fullblown disease [17]. 3.1 aHUS and Targets The anti-C5 mAb eculizumab was approved for aHUS treatment [56], thereby validating the hypothesis that as in PNH, defective AP regulation and MAC production underlie aHUS pathogenesis. However, although both diseases are characterized by complement activation and E lysis, the targets of complement are different: E with severely reduced expression of CD55 and CD59 are targets in PNH, while endothelial cells are targets and sites of inflammation in aHUS [17,24,26,46,48]. Moreover, there is a profound difference in activity of sera from patients with these respective diseases. Sera from aHUS patients promotes C3b deposition and MAC attack on cultured vascular endothelial cells, which is consistent with its presumed mode of action in vivo [46]. However, aHUS sera do not lyse normal human E, which apparently have sufficient levels of CD55 and CD59 to overcome the lack of AP regulation found in these sera. As noted, E destruction in aHUS is a consequence of “mechanical “ lysis, because E are damaged during transit through partially occluded capillaries [17,47,48]. On the other hand, aHUS sera in which Factor H is defective can lyse non-opsonized sheep E. In normal human serum the sialic acid on sheep E serves as a binding site for Factor H which protects the sheep E from lysis. However, if the aHUS serum has a defect in Factor H activity (mutation or autoantibodies) then the sheep E will be lysed because of a lack of Factor H protective activity at the cell surface [57]. The unique nature of aHUS sera has been employed in a modified Ham test which differentiates between sera from patients with aHUS from sera obtained from patients with TTP [58]. Only sera from aHUS patients could lyse cultured cells in which complement control proteins had been removed by treatment with phospholipase C. 4. Autoimmune Hemolytic Anemias (AIHA) Complement plays a key pathologic role in autoimmune hemolytic anemias, but its role is not due to dysregulation [34,59,60]. Rather, complement (CP) is “normally” activated on the surface of E opsonized with either IgM which binds to E at low temperatures in acral areas of the circulation (CAD, cold agglutinin disease) or E can be opsonized with IgG (warm AIHA). These E can be cleared from the circulation by macrophages in liver and spleen that recognize E with bound IgG and/or tagged with large amounts of C3b. In a minority of cases there can be intravascular hemolysis of E, and the phenotype of the disease may then more closely resemble PNH because intravascular lysis can promote thrombosis as well. In CAD C3d-tagged E are demonstrable in the bloodstream, in marked similarity to the observation of C3d-tagged E in PNH patients treated with eculizumab. It is reasonable to infer that, for some E from patients with CAD, there will be insufficient levels of deposited C3b/iC3b, or the C3b/iC3b will be rapidly degraded to C3d, and then the thresholds for clearance and lysis will not be reached [40].
10
Figure 1 Legend The classical and alternative complement activation pathways contribute to hemolytic disease pathologies. The key control steps are marked in gray; key amplification steps are marked in blue; and threshold steps are marked in green. These designations are based, in part, on [9,16]. See text for details. The most important controls are based on blocking the action of C3b by decay acceleration of the C3 and C5 convertases (FH, DAF), or by Factor I mediated inactivation of C3b by converting it to iC3b. FH, MCP, and THBD provide co-factor activity for FI. Plasmamembrane associated MIRL plays a key role in blocking formation of the MAC on cells.
References
[1] Wagner E, Frank MM. Therapeutic potential of complement modulation. Nat Rev Drug Disc 2010;9:43-56. [2] Merle NS, Church SE, Fremeaux-Bacchi V, Roumenina LT. Complement system part I: molecular mechanisms of activation and regulation. Front Immunol 2015;6:1-30. [3] Mathern DR, Heeger PS. Molecules great and small: The complement system. Clin J Am Soc Nephrol 2015;10:1636-50. [4] Merle NC, Noe R, Halbwachs-Mecarelli L, Fremeaux-Bacchi V, Roumenina LT. Complement system part II: role in immunity. Front Immunol 2015;6:1-26. [5] Ricklin D, Reis ES, Mastellos DC, Gros P, Lambris JD. Complement component C3 - the "Swiss army knife" of innate immunity and host defense. Immunol Rev 2016;274:33-58. [6] Ricklin D, Mastellos DC, Reis ES, Lambris JD. The renaissance of complement therapeutics. Nat Rev Nephrol 2018;14:26-47. [7] Baines AC, Broksky RA. Complementopathies. Blood Rev 2017;31:213-23. [8] Risitano AM, Marotta S. Therapeutic complement inhibition in complementmediated hemolytic anemias: Past, present and future. Sem Immunol 2016;28:223-40. [9] Hughes-Jones NC. The classical pathway. In: Ross GD, editors. Immunobiology of the complement system. Orlando: Academic Press; 1986, pp. 21-44.
11
[10] Beurskens FJ, Lindorfer MA, Farooqui M et al. Exhaustion of cytotoxic effector systems may limit monoclonal antibody-based immunotherapy in cancer patients. J Immunol 2012;188:3532-41. [11] Melis JPM, Strumane K, Ruuls SR et al. Complement in therapy and disease: regulating the complement system with antibody-based therapeutics. Mol Immunol 2015;67:117-30. [12] Cleary KLS, Chan HTC, James S, Glennis MJ, Cragg MS. Antibody distance from the cell membrane regulates antibody effector mechanisms. J Immunol 2017;198:3999-4011. [13] Lindorfer MA, Cook EM, Tupitza JC et al. Real-time analysis of the detailed sequence of cellular events in mAb-mediated complement-dependent cytotoxicity of B-cell lines and of CLL B-cells. Mol Immunol 2016;70:13-23. [14] Lindorfer MA, Kohl J, Taylor RP. Interactions between the complement system and Fcg receptors. In: Ackerman ME, Nimmerjahn F, editors. Antibody Fc: Linking adaptive and innate immunity. Philadelphia, PA: Elsevier; 2014, pp. 49-74. [15] Pangburn M, Schreiber R, Müller-Eberhard H. Formation of the initial C3 convertase of the alternative complement pathway. Acquisition of C3b-like activities by spontaneous hydrolysis of the putative thioester in native C3. J Exp Med 1981;154:856-67. [16] Pangburn MK. The alternative pathway. In: Ross GD, editors. Immunobiology of the complement system. Orlando: Academic Press; 1986, pp. 21-44. [17] Nester CM, Barbour T, de Cordoba SR et al. Atypical aHUS: state of the art. Mol Immunol 2015;67:31-42. [18] Skerka C, Chen Q, Fremeaux-Bacchi V, Roumenina L.T. Complement factor H related proteins (CFHRs). Mol Immunol 2013;56:170-80. [19] Holguin MH, Martin CB, Bernshaw NJ, Parker CJ. Analysis of the effects of activation of the alternative pathway of complement on erthrocytes with an isolated deficiency of DAF. J Immunol 1992;148:498-502. [20] Ferreira VP, Pangburn MK. Factor H-mediated cell surface protection from complement is critical for the survival of PNH erythrocytes. Blood 2007;110:2190-2. [21] Seya T, Turner JR, Atkinson JP. Purification and characterization of a membrane protein (gp45-70) that is a cofactor for cleavage of C3b and C4b. J Exp Med 1986;163:837-55.
12
[22] Delvaeye M, Noris M, De Vriese A et al. Thrombomodulin mutations in atypical hemolytic-uremic syndrome. NEJM 2009;361:345-57. [23] Noris M, Mescia F, Remuzzi G. STEC-HUS, atypical HUS and TTP are all diseases of complement activation. Nat Rev Nephrol 2012;8:622-33. [24] Hillmen P, Lewis SM, Bessler M, Luzzatto DJV. Natural history of PNH. NEJM 1995;333:1253-8. [25] Fishelson Z, Horstmann RD, Muller-Eberhard HJ. Regulation of the alternative pathway of complement by pH. J Immunol 1987;138:3392-5. [26] Parker C, Omine M, Richards S et al. Diagnosis and management of paroxysmal nocturnal hemoglobinuria. Blood 2005;106:3699-709. [27] Rother RP, Rollins SA, Mojcik CF, Brodsky RA, Bell L. Discovery and development of the complement inhibitor eculizumab for the treatment of PNH. Nat Biotech 2007;25:1256-64. [28] Takeda J, Miyata T, Kawagoe K et al. Deficiency of the GPI anchor caused by a somatic mutation of the PIG-A gene in PNH. Cell 1993;73:703-11. [29] Gargiulo L, Papaioannou M, Sica M et al. Glycosylphosphatidylinositolspecific, CD1d-restricted T cells in paroxysmal nocturnal hemoglobinuria. Blood 2013;121:2753-61. [30] Rotoli B, Luzzatto L. Paroxysmal nocturnal hemoglobinuria. Semin Hematol 1989;26:201-7. [31] Young NS, Maciejewski JP. Genetic and environmental effects in paroxysmal nocturnal hemoglobinuria: this little PIG-A goes "Why? Why? Why?". J Clin Invest 2000;106:637-41. [32] Frimat M, Tabarin F, Dimitrov JD et al. Complement activation by heme as a secondary hit for atypical hemolytic uremic syndrome. Blood 2013;122:28292. [33] Rother RP, Bell L, Hillmen P, Gladwin MT. The clinical sequelae of intravascular hemolysis and extracellular plasma hemoglobin. JAMA 2005;293:1653-62. [34] Cappellini MD. Coagulation in the pathophysiology of hemolytic anemias.Am. Soc. Hematol. Educ. Program: Am. Soc. Hematol.; 2007, pp. 74-8. [35] Hillmen P, Muus P, Duhrsen U et al. Effect of the complement inhibitor eculizumab on thromboembolism in patients with PNH. Blood 2007;110:4123-8.
13
[36] Ritis K, Doumas M, Mastellos D et al. A novel C5a receptor-tissue factor crosstalk in neutrophils links innate immunity to coagulation pathways. J Immunol 2006;177:4794-802. [37] Gremmel T, Fedrizzi S, Weigel G, Eichelberer B, Panzer S. Underlying mechanism and specific prevention of hemolysis-induced platelet activation. Platelets 2017;28:555-9. [38] Risitano AM, Marando L, Seneca E, Rotoli B. Hemoglobin normalization after splenectomy in a PNH patient treated by eculizumab. Blood 2008;112:44950. [39] Risitano AM, Notaro R, Pascariello C et al. The complement receptor 2/factor H fusion protein TT30 protects PNH erythrocytes from complementmediated hemolysis and C3 fragment opsonization. Blood 2012;119:630716. [40] Atkinson JP, Frank MM. Studies on the in vivo effects of antibody. Interaction of IgM antibody and complement in the immune clearance and destruction of erythrocytes in man. J Clin Invest 1974;54:339-48. [41] Lin Z, Schmidt CQ, Koutsogiannaki S et al. Complement C3dg-mediated erythrophagocytosis: implications for paroxysmal nocturnal hemoglobinuria. Blood 2015;126:891-4. [42] Yamashina M, Ueda E, Kinoshita T et al. Inherited complete deficiency of 20kilodalton homologous restriction factor (CD59) as a cause of PNH. NEJM 1990;1184-9. [43] Sica M, Rondelli T, Ricci P et al. CD59 deficiency is critical for C3 binding on red blood cells of patients with paroxysmal nocturnal hemoglobinuria (PNH) during anti-C5 treatment (eculizumab). Blood 2016;128:401. [44] Beum PV, Lindorfer MA, Peek EM et al. Penetration of antibody-opsonized cells by the membrane attack complex of complement promotes Ca2+ influx and induces streamers. Eur J Immunol 2011;41:2436-46. [45] Wang RH, Phillips Jr. G, Medof ME, Mold C. Activation of the alternative complement pathway by exposure of phosphatidylethanolamine and phosphatidylserine on erythrocytes from sickle cell disease patients. J Clin Invest 1993;92:1326-35. [46] Noris M, Galbusera M, Gastoldi S et al. Dynamics of complement activation in aHUS and how to monitor eculizumab therapy. Blood 2014;124:1715-26. [47] Noris M, Remuzzi G. Atypical hemolytic-uremic syndrome. NEJM 2009;361:1676-87.
14
[48] Atkinson JP, Goodship THJ. Complement factor H and the hemolytic uremic syndrome. J Exptl Med 2007;204:1245-8. [49] Manuelian T, Hellwage J, Meri S et al. Mutations in factor H reduce binding affinity to C3b and heparin and surface attachment to endothelial cells in hemolytic uremic syndrome. J Clin Invest 2003;111:1181-90. [50] Jokiranta TS, Jaakola V-P, Lehtinen MJ et al. Structure of complement factor H carboxyl-terminus reveals molecular basis of atypical haemolytic uremic syndrome. EMBO J 2006;25:1784-94. [51] Pickering MC, de Jorge EG, Martinez-Barricarte R et al. Spontaneous hemolytic uremic syndrome triggered by complement factor H lacking surface recognition domains. J Exptl Med 2007;204:1249-56. [52] Nangaku M. Complement regulatory proteins in glomerular diseases. Kid Intl 1998;54:1419-28. [53] Csincsi AI, Kopp A, Zoldi M et al. Factor H-related protein 5 interqacts with pentraxin 3 and the extracellular matric and modulates complement activiation. J Immunol 2015;194:4963-73. [54] de Jorge EG, Harris CL, Esparza-Gordillo J et al. Gain-of-function mutations in complement factor B are associated with atypical hemolytic uremic syndrome. PNAS 2007;104:240-5. [55] Martinez-Barricarte R, Heurich M, Lopez-Perrote A et al. The molecular and structural bases for the association of complement C3 mutations with atypical hemolytic uremic syndrome. Mol Immunol 2015;66:263-73. [56] Legendre CM, Licht C, Muus P, Greenbaum LA, et al. Terminal complement inhibitor eculizumab in atypical hemolytic-uremic syndrome. NEJM 2013;368:2169-81. [57] Sanchez-Corral P, Gonzalez-Rubio C, Rodriguez de Cordoba S, Lopez-Trascasa M. Functional analysis in serum from atypical hemolytic uremic syndrome patients reveal impaired protection of host cells associated with mutations in Factor H. Mol Immunol 2004;41:81-4. [58] Gavriilaki E, Yuan X, Ye Z et al. Modified Ham test for atypical hemolytic uremic syndrome. Blood 2015;125:3637-46. [59] Berentsen S, Randen U, Tjonnfjord GE. Cold agglutinin-mediated autoimmune hemolytic anemia. Hematol Oncol Clin N Am 2015;29:455-71. [60] Liebman HA, Weitz IC. Autoimmune hemolytic anemia. Med Clin N Am 2017;101:351-9.
15
Classical pathway
Alternative pathway
Immune Complexes
Tickover
C1qC1r2C1s2
C3
C1qC1r2*C1s2*
C4 C4a C4b
C3(H2O)
Amplification step
FB FD
C4bC1s*
Ba
C2
C3(H2O)Bb (Fluid-phase C3 convertase)
C2b
C4bC2a (CP C3 convertase)
FB FD
C3a
C3b
C3 FH,DAF
C3bBb (AP C3 convertase)
C3a
C3
C4bC2aC3b (C5 convertase)
C3b
FI,FH,MCP, DAF,THBD
C3bBbC3b (C5 convertase)
C5
C5a
C5b C6 C7 C8 C9
C3b
FI,FH,MCP, DAF,THBD
Inhibitor MIRL
Amplification Threshold
MAC
Figure 1 Taylor and Lindorfer
image
Mechanisms of complement-mediated damage in hematological disorders Ronald P. Taylor, Margaret A. Lindorfer
www.elsevier.com/locate/bios
PII: DOI: Reference:
S0037-1963(18)30005-2 https://doi.org/10.1053/j.seminhematol.2018.02.003 YSHEM50938
To appear in: Seminars in Hematology Cite this article as: Ronald P. Taylor and Margaret A. Lindorfer, Mechanisms of complement-mediated damage in hematological disorders, Seminars in Hematology,doi:10.1053/j.seminhematol.2018.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Mechanisms of complement-mediated damage in hematological disorders Ronald P. Taylora and Margaret A. Lindorfera aDepartment
of Biochemistry and Molecular Genetics, University of Virginia School
of Medicine, Charlottesville, VA 22908 USA Address all correspondence to Dr. Ronald P. Taylor, Department of Biochemistry and Molecular Genetics, University of Virginia School of Medicine, 1340 Jefferson Park Avenue, Charlottesville, VA 22908 USA 1-434-924-2664 (voice) 1-434-9245069 (fax) email: [email protected] Declarations of interest: none
Abstract The complement cascade is an ancient defense system that destroys and eliminates threats to normal homeostasis in the bloodstream and tissues. Although multiple controls keep complement in check to minimize innocent bystander injury to normal cells and tissues, defects in complement regulation due to mutations in, or autoantibodies to, complement control proteins underlie the pathogenesis of several hemolytic diseases including paroxysmal nocturnal hemoglobinuria, and atypical hemolytic uremic syndrome. In autoimmune hemolytic anemias complement plays an important role in erythrocyte destruction mediated by anti-erythrocyte antibodies. The pathogenic mechanisms of these hemolytic diseases are discussed, with an emphasis on pivotal steps in complement activation. Abbreviations: aHUS, atypical hemolytic uremic syndrome; AIHA, autoimmune hemolytic anemia; AP, alternative pathway of complement; CAD, cold agglutinin disease; C1s*, C1r*, activated forms of C1s and C1r respectively; CFHR proteins, complement factor H related proteins; CLL, chronic lymphocytic leukemia; CP, classical pathway of complement; DAF, decay accelerating factor, CD55; IC, immune complex; MAC, membrane attack complex; MBL, mannan binding lectin pathway of complement; MCP, membrane co-factor protein, CD46; MIRL, membrane inhibitor of reactive lysis, CD59; PNH, paroxysmal nocturnal hemoglobinuria; THBD, thrombomodulin.
1
1. Introduction The complement cascade is an ancient anti-microbial defense system that has evolved to synergize with several facets of the vertebrate immune system [1-7]. Complement provides an early warning system of “first responders” that react, in a rapid and exponential manner, to destroy and eliminate threats to normal homeostasis in the bloodstream and in tissues. Key elements in the initiation of complement activation include plasma proteins which specifically recognize dangerous microorganisms (exogenous threats) as well as potentially noxious apoptotic cells and cellular debris (endogenous threats). Complement also provides defense in association with adaptive immunity by rapidly clearing antigen-antibody immune complexes (IC). Regardless of the initiating event, the downstream consequences of complement activation are characterized by numerous cytotoxic reactions and production of several potent mediators of inflammation. As a defense system complement works quite effectively in eliminating pathogens, but this incendiary system also has real potential for causing injury to host cells and tissue. In fact, low level, non-specific activation of complement in normal cells and tissue can occur in the absence of frank external stimuli. There is also the possibility of the generation of cytotoxic factors against normal tissue when nearby pathogens are targeted by complement. Microorganisms that are recognized as bonafide threats by complement are subject to substantial levels of complement activation. For these agents several pivotal steps in the complement cascades easily reach and surpass activation thresholds. However, intrinsic to the complement system is a series of controls that provide defense of normal cells and tissue against these reactions [13]. Indeed, complement is most efficiently controlled based on thresholds for several different reaction steps in the pathways. Under normal homeostasis these complement controls maintain the level of complement activation on normal cells and tissue considerably below threshold values. However, in most complementmediated hemolytic diseases, complement is not properly regulated [8]. In this chapter we will focus on key defects in complement control mechanisms in hemolytic diseases, in which complement is inappropriately activated and pivotal thresholds are exceeded, thus leading to substantial damage to normal cells and tissues. 1.1 The Classical Pathway (CP) There are three pathways in which molecular challenges/stimuli promote complement activation, and all three are characterized by sequential proteolysis steps. Figure 1 is an abbreviated representation of the complement cascade that highlights key control points which are most important in hemolytic diseases. In the CP, C1q, a component of the C1 complex binds to IC containing IgG1, IgG3 or IgM [1,2,9]. C1q (as part of the C1 complex) can also bind directly to certain substrates and activate the CP on targeted cells in the absence of bound immunoglobulins; these substrates include bacterial lipopolysaccharides, C-reactive protein, pentraxins, and degradation products of cells, e.g. DNA. As a consequence of C1q 2
binding, C1r and C1s are consecutively activated in proteolytic steps, and activated C1s (now a protease) cleaves C4 to C4a and C4b. In nascent activated C4b, an internal thioester bond is exposed to solvent and therefore C4b can react with water or form a covalent amide or ester bond with acceptor sites on the IC, thus additionally tagging it as “foreign”. The C4bC1s complex then cleaves C2, and the newly-generated C4bC2a complex, the CP C3 convertase (C2a is a protease), cleaves C3 to produce C3a and C3b. In activated C3b an internal thioester bond is also exposed to solvent which can either react with water or form a covalent bond with acceptor sites on the IC. In solution phase the classical pathway is subject to downregulation due to the action of factor H (see Section 1.4) as well as due to the dissociation and dilution of labile newly-formed complexes. This phenomenon is illustrated in the enhanced ability of the CD20-specific mAb ofatumumab to activate complement on binding to the surface of B cells compared to the action of the CD20specific mAb rituximab [10,11]. Ofatumumab binds to a target site on CD20 much closer to the cell membrane than the site targeted by rituximab. When cells are opsonized with ofatumumab there is more effective concentration of activated complement components on the cell membrane, demonstrated by more successful covalent binding of nascent C4b and C3b to the targeted cell, and decreased nonproductive reaction of these components with water. These findings have now been extended and generalized in model systems which revealed that mAbs specific for CD20 and CD52 were most effective in promoting CDC of cells in which the targets were expressed closest to the cell membrane [12]. If there is sufficient (threshold) C3 activation, and enough C3b is able to covalently opsonize the substrate IC, the C5 convertase (C4bC2aC3b) will assemble (C2a continues to provide protease activity) and cleave C5 to C5b and C5a. The newlyformed C5b, released at the target surface, can then consecutively attract molecules of C6-C9 to produce the membrane attack complex (MAC). If a sufficient number of MAC are generated and penetrate the cell membrane, the cell will be killed due to rapid influx of Ca+2 [13]. Inherent in this pathway is a series of amplification steps, due to the multiple enzymes that are produced from their zymogen precursors [9]. The C3b activation step is particularly important with respect to destruction of cells because if threshold amounts of C3b are bound to the substrate cell, the cell will become susceptible to phagocytosis by effector cells that express receptors for C3b or for its downstream fragment iC3b. Thus, even if the cell under attack is not killed by the MAC, C3b opsonization can still lead to its elimination. This is one of the main mechanisms by which bacteria are successfully targeted by complement; after opsonization with C3b and iC3b, neutrophils and macrophages eliminate them by phagocytosis [1,4,5]. Other important inflammatory agents produced at the C3 and C5 activation steps are C3a and C5a, respectively, anaphylatoxins that recruit and activate effector cells at sites of complement activation. Chelation by C5a of the C5a receptor on effector cells (e.g. mast cells, monocytes, macrophages, neutrophils) enhances the “inflammatory poise” of the cells, which is manifested by increases in expression of activating Fcγ receptors and complement receptor 3 and down-modulation of the 3
inhibitory receptor FcγRIIb [3,14]. The subsequent cytotoxic inflammatory responses include phagocytosis and release of cytotoxic agents such as superoxide, hydrolytic enzymes, cytokines, and chemokines [2]. C5a can also induce endothelial cell activation by promoting release of histamine from basophils and mast cells, thereby increasing vascular permeability and contributing to vascular injury. 1.2 The Alternative Pathway (AP) The AP is initiated by spontaneous reaction of small amounts of C3 with water (~0.3% per hour) [15,16]. This fluid phase endogenous tickover reaction involves a conformational change in C3 comparable to that which occurs when C3 is converted to C3b, thus exposing the thioester bond to water and generating C3(H20) (Figure 1). Factor B then binds to C3(H20) and is cleaved and activated by Factor D, generating C3(H20)Bb. Bb within this complex has proteolytic activity, and the complex can bind and cleave C3, giving rise to nascent C3b. Then, C3b can either react with water, or form a covalent bond with surfaces of bacteria and yeast which can active the AP, or with PNH E that cannot down-modulate the AP. As noted earlier, under normal conditions the fluid phase activation of complement is also more easily down-regulated due to dilution and dissociation of reacting factors, but the AP will be considerably more active when a “safe harbor” on a substrate surface forms the nexus for continued complement activation. Indeed, under these conditions substrate-bound C3b can bind additional Factor B, and in the presence of Factor D produce more Bb, thereby giving rise to C3bBb (the AP C3 convertase) in an autocatalytic pathway capable of rapid production of substrate-bound C3b molecules (Figure 1, amplification step). This convertase is metastable, but can be stabilized by Properdin. C3b produced during CP activation can also feed into the AP, contributing even more “fuel” to the AP C3 convertase. In analogy to the requirements in the CP, if enough C3b is produced and captured by the AP C3 convertase, the C5 convertase of the AP is formed (C3bBbC3b) which cleaves C5 to produce C5a and C5b, thus allowing for production of the MAC. All complement convertases are metastable and decompose rapidly, and this insures some natural control of complement activation and presumably limits possible innocent bystander damage to nearby host tissue. Any agents or gain of function mutations (e.g. in C3 or Factor B) that tend to stabilize the convertases could have the effect of promoting complement-mediated damage to normal cells and tissue, and several of these mutations play primary roles in induction of certain hemolytic anemias (see Section 3).
4
1.3 The Mannan Binding Lectin Pathway (MBL) This pathway is a “hard-wired” innate pathway specific for carbohydrates expressed on microorganisms including yeast and bacteria [1,7]. In analogy to the CP, after the recognition molecules (MBL, ficolins, collectins) bind to the substrate surface, mannan-binding lectin associated proteases are activated, and their actions then feed into the classical pathway at the C4 activation step. There is little evidence that this pathway plays an important role in hemolytic diseases. 1.4 Control of Complement Almost all steps in complement activation are subject to control by either soluble factors or by membrane-associated proteins on potentially vulnerable cells (Figure 1). Loss-of-function mutations in these proteins or inhibition of their activities due to autoantibodies can lead to hemolytic diseases. As noted, both the CP and MBL pathway feed into the AP, and the AP promotes exponential and rapid amplification of complement activation starting at the C3 stage. Therefore, it is not surprising that several severe diseases associated with defective control of complement can be attributed to lesions that induce loss of control of the AP [3]. Factor H, a soluble protein, is the most important regulator of the AP [1,2,7,17]. By binding to C3b (and competing with Factor B) it accelerates decay of the AP C3 convertase. In addition Factor H serves as a co-factor for Factor I, a protease that directly degrades C3b, converting it to iC3b, its inactive form [1,16]. Factor H also down-regulates complement on cell surfaces; it binds to cell membranes containing negatively charged molecules including sialic acid and glycosoaminoglycans such as heparin sulfate, thus enabling it to recognize and bind to normal “self” cells that express negatively charged molecules [2,3]. Therefore, cell-associated Factor H serves a sentinel function to down-regulate incipient complement activation after C3b deposition on self-tissue. On the other hand, when covalent deposition of C3b is initiated on susceptible pathogens lacking negative cell-surface molecules, Factor H does not bind to such non-self-tissue or block C3b deposition, and therefore the pathogen becomes rapidly coated with C3b, exceeding the thresholds needed to initiate activation of C5 or to promote phagocytosis mediated by complement receptors on effector cells. That is, Factor H “ignores” these microorganisms and does not interfere with C3b deposition, therefore the AP is free to deposit C3b at a rapid clip, and to continue downstream to production of the MAC. Five similar proteins that lack the first four N-terminal complement regulatory domains of Factor H (CFHR1 to CFHR5), as well as a spliced variant of factor H [FHL-1] have been described and the potential role of these proteins in either complement control or activation in several diseases including aHUS (see section 3) is under active investigation [18]. Two important plasma membrane-associated complement control proteins that are expressed on blood cells are CD55 (DAF, decay accelerating Factor) and CD59 (MIRL, membrane inhibitor of reactive lysis) [15,16,19]. CD55 binds to C4b and 5
interferes with formation of the C3 CP convertase, and also accelerates decay of the C3 and C5 convertases of the AP and CP. CD59 blocks formation of the MAC by binding to C5b-8 and preventing incorporation of C9. When these control proteins are not present or are expressed at lower levels on blood cells (see section 2), then the cells are subject to attack by the AP, and at best soluble Factor H can slow down the attack, but not block it [20]. CD46, (MCP, membrane co-factor protein) is expressed on most cells other than E, and serves as a co-factor for Factor I in the conversion of active C3b to inactivated iC3b, thus down-regulating complement activation on cells and surfaces [21]. Finally, thrombomodulin, an endothelial cellassociated protective protein, regulates complement activation in the vasculature [22,23]. First, it enhances the activity of Factor I and its associated co-factors in degrading C3b to iC3b. In addition it down-modulates production of C5a by inhibiting the action of thrombin (which can directly activate C5 to C5a and C5b). It also stimulates synthesis of carboxypeptidase B, which degrades C3a and C5a, thus inhibiting their activities. 2. Paroxysmal Nocturnal Hemoglobinuria (PNH) The pathogenesis of PNH has been the subject of intense investigations which have aided in the elucidation of several molecular details of the complement cascades [24-27]. PNH is most readily characterized by severe anemia (Coombs negative), due to complement-mediated intravascular hemolysis of E. The plasma membranes of E of patients with PNH have substantially reduced levels of complement control proteins CD55 and CD59 which renders the cells quite sensitive to AP-mediated lysis, especially at lower pH. Indeed, the tickover reaction of the AP, including its potential for exponential activation, was elegantly demonstrated by MuellerEberhard’s group based on examination of the hemolysis of PNH E in serum at low pH under conditions in which only the AP could be activated [15,25]. Studies led by T. Kinoshita established that the origin of sensitivity of PNH E to complement could be traced to acquired (but rare) somatic mutations in one or more multipotent hematopoietic stem cells that gave rise to E as well as other blood cells [28]. The mutation is in the X-chromosome phosphatidylinositol glycan class A (PIG-A) gene which codes for a key step in synthesis of glycosyl phosphatidylinositol (GPI) groups that serve to anchor certain proteins to cell membranes. Hematopoietic stem cells with these mutations appear to have a paradoxical survival advantage due to immune escape from T cell-mediated attack [29-31]. Progeny derived from these mutant stem cells, which include E, platelets, monocytes and granulocytes, have considerably reduced levels of CD55 and CD59 which normally are attached to cells with the GPI link. Three populations of E, at varying levels, are demonstrable in PNH: PNH I E, normal expression of CD55 and CD59; PNH II E, with considerably reduced levels of CD55 and CD59; PNH III E, no expression of either CD55 or CD59. PNH is quite often associated with aplastic anemia, which may be related to the original lesion in the bone marrow that leads to generation and survival of the mutant cells [26]. In view of the marked action of Factor H in down-regulating complement activation on cell surfaces, one might 6
expect that E lacking CD55 and CD59 would still be adequately protected from complement by Factor H. In fact, Ferreira and Pangburn reported in vitro experiments demonstrating that Factor H provides some protection of PNH E from complement-mediated lysis, but in vivo the protection is not adequate [20]. 2.1 Downstream consequence of hemolysis in PNH The major cause of mortality in PNH is due to thromboses, which are most likely produced as a consequence of a “perfect storm” of inflammation associated with both complement activation and E lysis followed by the release into the circulation of procoagulant E microparticles, free hemoglobin and its breakdown product heme [2,24,32,33]. The link between hemolysis and thromboses is not limited to PNH, and appears to be complex and not fully understood. However, as noted by Cappellini, “hemolysis per se, whatever the cause, seems to be a pro-coagulant condition” [34]. Although the body has mechanisms for clearing hemoglobin and heme from the bloodstream, these processes can be saturated and overwhelmed by high levels of hemolysis [32,33]. Free hemoglobin in the circulation can promote several inflammatory reactions leading to vascular injury, many of which are induced due to consumption of nitric oxide by hemoglobin. Moreover, free heme can also activate the AP, thus producing C3a and C5a, which activate and aggregate platelets, especially because platelets lack, or poorly express, CD55 and CD59 in PNH. This high level of complement activation and inflammation, including production of the MAC, are also likely to damage endothelial tissue [35], thus additionally contributing to thrombosis. There are additional links between the complement system and the clotting system [7]. Ritis et al have reported in a model system that complement activation and the production of C5a causes neutrophils to express Tissue Factor, which initiates the blood clotting cascade [36]. The investigators suggested that with regard to normal defense, e.g., response to a bacterial infection, formation of a blood clot would serve to contain and prevent spread of the infection. Recent studies by Gremmel et al have demonstrated that ADP released from lysed E can also directly activate platelets and promote their aggregation [37]. 2.2 PNH and Eculizumab The neutralizing anti-C5 mAb eculizumab has proven to be a potent agent for treatment of PNH and aHUS by blocking activation of C5 to C5a and C5b [27]. Indeed, treatment of PNH patients with eculizumab reduces intravascular hemolysis and the risk of thromboembolisms, thus providing a link between E lysis and downstream coagulation [35]. Moreover, observations on patients treated with eculizumab speak to the kinetics of complement activation and down-regulation: In vitro reaction of PNH E with acidified serum in the presence of eculizumab leads to deposition of C3b/iC3b fragments on the cells. After a 24 hour reaction in vitro these fragments decay to C3d, but the “transition” from C3b/iC3b to C3d in vivo is much faster. When PNH patients receive eculizumab therapy only C3d is detected on E [8,38,39]. We suggest that if blood samples are taken immediately after
7
complement activation that is triggered subsequent to eculizumab blockade, then it should be possible to detect C3b/iC3b-opsonized PNH E. The classic studies of Atkinson and Frank on the fate of IgM-opsonized, C3b/iC3bcoated E demonstrated that these E are cleared from the bloodstream due to phagocytosis mediated by macrophages with receptors for C3b/iC3b [40]. It is likely that during eculizumab therapy, the PNH E that are heavily coated with C3b/iC3b are cleared similarly. Indeed, Risitano’s group has provided clear evidence that after eculizumab therapy, extravascular clearance of PNH E is mediated by the spleen [38]. However, for PNH E that are not cleared (perhaps due to lower levels of opsonization) the C3b/iC3b likely rapidly decays to C3d and these E continue to circulate. C3dg-coated E are also subject to phagocytosis mediated by complement receptors on macrophages [41], but this may be a less efficient process because C3dtagged E are demonstrable for some time in the circulation of eculizumab-treated PNH patients [8,39]. These phenomena are quite similar to the reactions that occur when the anti-CD20 mAbs rituximab or ofatumumab are infused into the circulation of patients with chronic lymphocytic leukemia (CLL). Although very early on the CLL B cells contain C3b/iC3b fragments, the transition to C3d is quite rapid [10]. 2.3 Role of CD59 Conventional wisdom argues that the fundamental defect in protective mechanisms in PNH E has been elucidated: decreased levels of both CD55 and CD59 presage complement-mediated hemolysis [8,26,27]. Lack of CD55 allows build-up of threshold amounts of C3b activation fragments, and then E, heavily opsonized with C3b, are excellent substrates for the AP C5 convertase, allowing deposition of lethal amounts of the MAC. However, the situation may be more complicated. Yamashina et al reported on a patient with a PNH phenotype who had an inherited complete deficiency of CD59 but normal levels of CD55 [42], and the patient’s E were lysed in acidified serum. The question is: why did CD55 not “protect” the patient’s E by blocking C3b deposition, thus preventing downstream generation of the MAC? Parker’s group reported that E from patients with the rare Inab phenotype (lack of CD55) could be lysed in acidified serum when the action of CD59 was also blocked [19]. They found that C3b deposition on the E was increased considerably, suggesting that CD59 may “modulate” the C3 convertase of the AP. Recently Risitano and Notaro found that when the activity of CD59 on normal E is blocked with a neutralizing mAb, and the E are reacted in acidified serum with Mg-EGTA containing eculizumab, then C3b fragments are deposited on the E via the AP [43]. This result suggests that active CD59 can inhibit C3b deposition on E, a provocative finding at odds with a substantial literature. When CD59 is blocked, sublytic attack by the MAC on the E membrane may expose phosphatidyl serine on the cell’s outer membrane, which could lead to activation of the AP and C3b deposition [44,45]. Resolution of this question may prove to be most interesting and may lead to new paradigms for controlling complement activation in certain diseases.
8
3. Atypical Hemolytic Uremic Syndrome (aHUS) The thrombotic microangiopathies include thrombotic thrombocytopenia purpura, (TTP), hemolytic uremia syndrome (HUS) and its atypical form (aHUS). Due to space limitations we will only discuss aHUS. In this disease there is substantial pathology associated with capillary injury as a consequence of platelet aggregation and inflammation concentrated on endothelial cells [46-48]. Moreover, E are damaged during transit within the compromised and narrowed microvasculature, giving rise to non-immunological, Coombs-negative “mechanical” hemolytic anemia in which schistocytes (damaged E) are demonstrable in peripheral blood smears [32,46,48]. As in PNH, hemolysis of circulating E leads to release of large amounts of hemoglobin followed by downstream elaboration of heme, both of which contribute to inflammation due to nitric oxide consumption, complement activation and increased vascular injury [32,33]. In aHUS, defective AP regulation on surfaces can occur as a consequence of several independent factors. The most common lesion is a mutation in Factor H which inhibits its ability to bind to surfaces [49,50]. As a consequence, tickover-mediated deposition of C3b on susceptible endothelial cells progresses to threshold levels of C3b deposition, allowing for generation of C5b, and subsequent damage by the MAC. In aHUS this pathologic mechanism is most prominently associated with kidney injury; Pickering et al were able to replicate this pathology in an aHUS mouse model by engineering Factor H lacking surface recognition domains [51]. Certain segments of the kidney express low levels of membrane-bound complement control proteins [52], and therefore proper functioning of Factor H may be necessary to protect the kidney from inappropriate complement activation. Variations or outright deletions of certain of the CFHR proteins can also affect the activity of Factor H and lead to aHUS. For example, loss of expression of CHFR1 and CFHR3 appears to induce generation of autoantibodies to Factor H, and subsequent development of aHUS [17]. Recently Csincsi et al reported that CFHR5 has several other activities, in addition to competing with Factor H for C3b binding, all of which should increase complement activation in the kidney [53]. These activities include blocking binding of Factor H to pentraxin, C-reactive protein, and the extracellular matrix ,and promoting formation of the AP C3b convertase. These activities, of either wild-type or mutated forms of CFHR5, may play a role in development of aHUS. The phenotype and tissue damage observed in patients with aHUS can be induced by mutations affecting the function of other complement control proteins that regulate the AP including Factor I, MCP (CD46), and thrombomodulin [17,22]. In addition, gain of function mutations in Factor B or C3, which increase the activity of the AP on surfaces can also give rise to aHUS [54,55]. There are potentially redundant independent factors that regulate the AP, and it is not surprising that with respect to aHUS, the disease penetrance of these mutations averages about 50% [2,4,17]. Therefore, if only one complement control factor is defective, other complement control factors may compensate, and disease may not be expressed. Alternatively, environmental triggers may be required to convert an aHUS 9
predisposing factor (e.g., Factor H mutation or autoantibodies to Factor H) into fullblown disease [17]. 3.1 aHUS and Targets The anti-C5 mAb eculizumab was approved for aHUS treatment [56], thereby validating the hypothesis that as in PNH, defective AP regulation and MAC production underlie aHUS pathogenesis. However, although both diseases are characterized by complement activation and E lysis, the targets of complement are different: E with severely reduced expression of CD55 and CD59 are targets in PNH, while endothelial cells are targets and sites of inflammation in aHUS [17,24,26,46,48]. Moreover, there is a profound difference in activity of sera from patients with these respective diseases. Sera from aHUS patients promotes C3b deposition and MAC attack on cultured vascular endothelial cells, which is consistent with its presumed mode of action in vivo [46]. However, aHUS sera do not lyse normal human E, which apparently have sufficient levels of CD55 and CD59 to overcome the lack of AP regulation found in these sera. As noted, E destruction in aHUS is a consequence of “mechanical “ lysis, because E are damaged during transit through partially occluded capillaries [17,47,48]. On the other hand, aHUS sera in which Factor H is defective can lyse non-opsonized sheep E. In normal human serum the sialic acid on sheep E serves as a binding site for Factor H which protects the sheep E from lysis. However, if the aHUS serum has a defect in Factor H activity (mutation or autoantibodies) then the sheep E will be lysed because of a lack of Factor H protective activity at the cell surface [57]. The unique nature of aHUS sera has been employed in a modified Ham test which differentiates between sera from patients with aHUS from sera obtained from patients with TTP [58]. Only sera from aHUS patients could lyse cultured cells in which complement control proteins had been removed by treatment with phospholipase C. 4. Autoimmune Hemolytic Anemias (AIHA) Complement plays a key pathologic role in autoimmune hemolytic anemias, but its role is not due to dysregulation [34,59,60]. Rather, complement (CP) is “normally” activated on the surface of E opsonized with either IgM which binds to E at low temperatures in acral areas of the circulation (CAD, cold agglutinin disease) or E can be opsonized with IgG (warm AIHA). These E can be cleared from the circulation by macrophages in liver and spleen that recognize E with bound IgG and/or tagged with large amounts of C3b. In a minority of cases there can be intravascular hemolysis of E, and the phenotype of the disease may then more closely resemble PNH because intravascular lysis can promote thrombosis as well. In CAD C3d-tagged E are demonstrable in the bloodstream, in marked similarity to the observation of C3d-tagged E in PNH patients treated with eculizumab. It is reasonable to infer that, for some E from patients with CAD, there will be insufficient levels of deposited C3b/iC3b, or the C3b/iC3b will be rapidly degraded to C3d, and then the thresholds for clearance and lysis will not be reached [40].
10
Figure 1 Legend The classical and alternative complement activation pathways contribute to hemolytic disease pathologies. The key control steps are marked in gray; key amplification steps are marked in blue; and threshold steps are marked in green. These designations are based, in part, on [9,16]. See text for details. The most important controls are based on blocking the action of C3b by decay acceleration of the C3 and C5 convertases (FH, DAF), or by Factor I mediated inactivation of C3b by converting it to iC3b. FH, MCP, and THBD provide co-factor activity for FI. Plasmamembrane associated MIRL plays a key role in blocking formation of the MAC on cells.
References
[1] Wagner E, Frank MM. Therapeutic potential of complement modulation. Nat Rev Drug Disc 2010;9:43-56. [2] Merle NS, Church SE, Fremeaux-Bacchi V, Roumenina LT. Complement system part I: molecular mechanisms of activation and regulation. Front Immunol 2015;6:1-30. [3] Mathern DR, Heeger PS. Molecules great and small: The complement system. Clin J Am Soc Nephrol 2015;10:1636-50. [4] Merle NC, Noe R, Halbwachs-Mecarelli L, Fremeaux-Bacchi V, Roumenina LT. Complement system part II: role in immunity. Front Immunol 2015;6:1-26. [5] Ricklin D, Reis ES, Mastellos DC, Gros P, Lambris JD. Complement component C3 - the "Swiss army knife" of innate immunity and host defense. Immunol Rev 2016;274:33-58. [6] Ricklin D, Mastellos DC, Reis ES, Lambris JD. The renaissance of complement therapeutics. Nat Rev Nephrol 2018;14:26-47. [7] Baines AC, Broksky RA. Complementopathies. Blood Rev 2017;31:213-23. [8] Risitano AM, Marotta S. Therapeutic complement inhibition in complementmediated hemolytic anemias: Past, present and future. Sem Immunol 2016;28:223-40. [9] Hughes-Jones NC. The classical pathway. In: Ross GD, editors. Immunobiology of the complement system. Orlando: Academic Press; 1986, pp. 21-44.
11
[10] Beurskens FJ, Lindorfer MA, Farooqui M et al. Exhaustion of cytotoxic effector systems may limit monoclonal antibody-based immunotherapy in cancer patients. J Immunol 2012;188:3532-41. [11] Melis JPM, Strumane K, Ruuls SR et al. Complement in therapy and disease: regulating the complement system with antibody-based therapeutics. Mol Immunol 2015;67:117-30. [12] Cleary KLS, Chan HTC, James S, Glennis MJ, Cragg MS. Antibody distance from the cell membrane regulates antibody effector mechanisms. J Immunol 2017;198:3999-4011. [13] Lindorfer MA, Cook EM, Tupitza JC et al. Real-time analysis of the detailed sequence of cellular events in mAb-mediated complement-dependent cytotoxicity of B-cell lines and of CLL B-cells. Mol Immunol 2016;70:13-23. [14] Lindorfer MA, Kohl J, Taylor RP. Interactions between the complement system and Fcg receptors. In: Ackerman ME, Nimmerjahn F, editors. Antibody Fc: Linking adaptive and innate immunity. Philadelphia, PA: Elsevier; 2014, pp. 49-74. [15] Pangburn M, Schreiber R, Müller-Eberhard H. Formation of the initial C3 convertase of the alternative complement pathway. Acquisition of C3b-like activities by spontaneous hydrolysis of the putative thioester in native C3. J Exp Med 1981;154:856-67. [16] Pangburn MK. The alternative pathway. In: Ross GD, editors. Immunobiology of the complement system. Orlando: Academic Press; 1986, pp. 21-44. [17] Nester CM, Barbour T, de Cordoba SR et al. Atypical aHUS: state of the art. Mol Immunol 2015;67:31-42. [18] Skerka C, Chen Q, Fremeaux-Bacchi V, Roumenina L.T. Complement factor H related proteins (CFHRs). Mol Immunol 2013;56:170-80. [19] Holguin MH, Martin CB, Bernshaw NJ, Parker CJ. Analysis of the effects of activation of the alternative pathway of complement on erthrocytes with an isolated deficiency of DAF. J Immunol 1992;148:498-502. [20] Ferreira VP, Pangburn MK. Factor H-mediated cell surface protection from complement is critical for the survival of PNH erythrocytes. Blood 2007;110:2190-2. [21] Seya T, Turner JR, Atkinson JP. Purification and characterization of a membrane protein (gp45-70) that is a cofactor for cleavage of C3b and C4b. J Exp Med 1986;163:837-55.
12
[22] Delvaeye M, Noris M, De Vriese A et al. Thrombomodulin mutations in atypical hemolytic-uremic syndrome. NEJM 2009;361:345-57. [23] Noris M, Mescia F, Remuzzi G. STEC-HUS, atypical HUS and TTP are all diseases of complement activation. Nat Rev Nephrol 2012;8:622-33. [24] Hillmen P, Lewis SM, Bessler M, Luzzatto DJV. Natural history of PNH. NEJM 1995;333:1253-8. [25] Fishelson Z, Horstmann RD, Muller-Eberhard HJ. Regulation of the alternative pathway of complement by pH. J Immunol 1987;138:3392-5. [26] Parker C, Omine M, Richards S et al. Diagnosis and management of paroxysmal nocturnal hemoglobinuria. Blood 2005;106:3699-709. [27] Rother RP, Rollins SA, Mojcik CF, Brodsky RA, Bell L. Discovery and development of the complement inhibitor eculizumab for the treatment of PNH. Nat Biotech 2007;25:1256-64. [28] Takeda J, Miyata T, Kawagoe K et al. Deficiency of the GPI anchor caused by a somatic mutation of the PIG-A gene in PNH. Cell 1993;73:703-11. [29] Gargiulo L, Papaioannou M, Sica M et al. Glycosylphosphatidylinositolspecific, CD1d-restricted T cells in paroxysmal nocturnal hemoglobinuria. Blood 2013;121:2753-61. [30] Rotoli B, Luzzatto L. Paroxysmal nocturnal hemoglobinuria. Semin Hematol 1989;26:201-7. [31] Young NS, Maciejewski JP. Genetic and environmental effects in paroxysmal nocturnal hemoglobinuria: this little PIG-A goes "Why? Why? Why?". J Clin Invest 2000;106:637-41. [32] Frimat M, Tabarin F, Dimitrov JD et al. Complement activation by heme as a secondary hit for atypical hemolytic uremic syndrome. Blood 2013;122:28292. [33] Rother RP, Bell L, Hillmen P, Gladwin MT. The clinical sequelae of intravascular hemolysis and extracellular plasma hemoglobin. JAMA 2005;293:1653-62. [34] Cappellini MD. Coagulation in the pathophysiology of hemolytic anemias.Am. Soc. Hematol. Educ. Program: Am. Soc. Hematol.; 2007, pp. 74-8. [35] Hillmen P, Muus P, Duhrsen U et al. Effect of the complement inhibitor eculizumab on thromboembolism in patients with PNH. Blood 2007;110:4123-8.
13
[36] Ritis K, Doumas M, Mastellos D et al. A novel C5a receptor-tissue factor crosstalk in neutrophils links innate immunity to coagulation pathways. J Immunol 2006;177:4794-802. [37] Gremmel T, Fedrizzi S, Weigel G, Eichelberer B, Panzer S. Underlying mechanism and specific prevention of hemolysis-induced platelet activation. Platelets 2017;28:555-9. [38] Risitano AM, Marando L, Seneca E, Rotoli B. Hemoglobin normalization after splenectomy in a PNH patient treated by eculizumab. Blood 2008;112:44950. [39] Risitano AM, Notaro R, Pascariello C et al. The complement receptor 2/factor H fusion protein TT30 protects PNH erythrocytes from complementmediated hemolysis and C3 fragment opsonization. Blood 2012;119:630716. [40] Atkinson JP, Frank MM. Studies on the in vivo effects of antibody. Interaction of IgM antibody and complement in the immune clearance and destruction of erythrocytes in man. J Clin Invest 1974;54:339-48. [41] Lin Z, Schmidt CQ, Koutsogiannaki S et al. Complement C3dg-mediated erythrophagocytosis: implications for paroxysmal nocturnal hemoglobinuria. Blood 2015;126:891-4. [42] Yamashina M, Ueda E, Kinoshita T et al. Inherited complete deficiency of 20kilodalton homologous restriction factor (CD59) as a cause of PNH. NEJM 1990;1184-9. [43] Sica M, Rondelli T, Ricci P et al. CD59 deficiency is critical for C3 binding on red blood cells of patients with paroxysmal nocturnal hemoglobinuria (PNH) during anti-C5 treatment (eculizumab). Blood 2016;128:401. [44] Beum PV, Lindorfer MA, Peek EM et al. Penetration of antibody-opsonized cells by the membrane attack complex of complement promotes Ca2+ influx and induces streamers. Eur J Immunol 2011;41:2436-46. [45] Wang RH, Phillips Jr. G, Medof ME, Mold C. Activation of the alternative complement pathway by exposure of phosphatidylethanolamine and phosphatidylserine on erythrocytes from sickle cell disease patients. J Clin Invest 1993;92:1326-35. [46] Noris M, Galbusera M, Gastoldi S et al. Dynamics of complement activation in aHUS and how to monitor eculizumab therapy. Blood 2014;124:1715-26. [47] Noris M, Remuzzi G. Atypical hemolytic-uremic syndrome. NEJM 2009;361:1676-87.
14
[48] Atkinson JP, Goodship THJ. Complement factor H and the hemolytic uremic syndrome. J Exptl Med 2007;204:1245-8. [49] Manuelian T, Hellwage J, Meri S et al. Mutations in factor H reduce binding affinity to C3b and heparin and surface attachment to endothelial cells in hemolytic uremic syndrome. J Clin Invest 2003;111:1181-90. [50] Jokiranta TS, Jaakola V-P, Lehtinen MJ et al. Structure of complement factor H carboxyl-terminus reveals molecular basis of atypical haemolytic uremic syndrome. EMBO J 2006;25:1784-94. [51] Pickering MC, de Jorge EG, Martinez-Barricarte R et al. Spontaneous hemolytic uremic syndrome triggered by complement factor H lacking surface recognition domains. J Exptl Med 2007;204:1249-56. [52] Nangaku M. Complement regulatory proteins in glomerular diseases. Kid Intl 1998;54:1419-28. [53] Csincsi AI, Kopp A, Zoldi M et al. Factor H-related protein 5 interqacts with pentraxin 3 and the extracellular matric and modulates complement activiation. J Immunol 2015;194:4963-73. [54] de Jorge EG, Harris CL, Esparza-Gordillo J et al. Gain-of-function mutations in complement factor B are associated with atypical hemolytic uremic syndrome. PNAS 2007;104:240-5. [55] Martinez-Barricarte R, Heurich M, Lopez-Perrote A et al. The molecular and structural bases for the association of complement C3 mutations with atypical hemolytic uremic syndrome. Mol Immunol 2015;66:263-73. [56] Legendre CM, Licht C, Muus P, Greenbaum LA, et al. Terminal complement inhibitor eculizumab in atypical hemolytic-uremic syndrome. NEJM 2013;368:2169-81. [57] Sanchez-Corral P, Gonzalez-Rubio C, Rodriguez de Cordoba S, Lopez-Trascasa M. Functional analysis in serum from atypical hemolytic uremic syndrome patients reveal impaired protection of host cells associated with mutations in Factor H. Mol Immunol 2004;41:81-4. [58] Gavriilaki E, Yuan X, Ye Z et al. Modified Ham test for atypical hemolytic uremic syndrome. Blood 2015;125:3637-46. [59] Berentsen S, Randen U, Tjonnfjord GE. Cold agglutinin-mediated autoimmune hemolytic anemia. Hematol Oncol Clin N Am 2015;29:455-71. [60] Liebman HA, Weitz IC. Autoimmune hemolytic anemia. Med Clin N Am 2017;101:351-9.
15
Classical pathway
Alternative pathway
Immune Complexes
Tickover
C1qC1r2C1s2
C3
C1qC1r2*C1s2*
C4 C4a C4b
C3(H2O)
Amplification step
FB FD
C4bC1s*
Ba
C2
C3(H2O)Bb (Fluid-phase C3 convertase)
C2b
C4bC2a (CP C3 convertase)
FB FD
C3a
C3b
C3 FH,DAF
C3bBb (AP C3 convertase)
C3a
C3
C4bC2aC3b (C5 convertase)
C3b
FI,FH,MCP, DAF,THBD
C3bBbC3b (C5 convertase)
C5
C5a
C5b C6 C7 C8 C9
C3b
FI,FH,MCP, DAF,THBD
Inhibitor MIRL
Amplification Threshold
MAC
Figure 1 Taylor and Lindorfer