Activation of the complement cascade and the stability of active complement proteins are tightly regulated to prevent complement activation on normal host cells and to limit the duration of complement activation even on microbial cells and antigen antibody complexes. Regulation of complement is mediated by several circulating and cell membrane proteins (Table 1). Many of these proteins belong to a family called regulators of complement activity (RCA) and are encoded by homologous genes that are tightly clustered on chromosome 1 at q3.2. RCA proteins include the cell membrane proteins, decay accelerating factor (DAF/CD55), membrane cofactor protein (MCP/CD46), complement receptor 1 (CR1/CD35), and complement receptor 2 (CR2/CD21). The circulating plasma RCA proteins include Factor H and C4-binding protein (C4BP).
Table1. Regulators of Complement Activation
Complement activation needs to be regulated for two reasons. First, low-level complement activation goes on spontaneously, and if such activation is allowed to proceed, the result can be damage to normal cells and tissues. Second, even when complement is activated where needed, such as on microbial cells or antigen-antibody complexes, it needs to be controlled because degradation products of complement proteins can diffuse to adjacent cells and injure them.
Different regulatory mechanisms inhibit the formation of C3 convertases in the early steps of complement activation, break down and inactivate C3 and C5 convertases, and inhibit formation of the MAC in the late steps of the complement pathway.
• The proteolytic activity of C1r, C1s, and MASP2 is inhibited by a plasma protein called C1 inhibitor (C1-INH). C1 INH is a serine protease inhibitor (serpin) that mimics the nor mal substrates of C1r and C1s. If C1q binds to an antibody and begins the process of complement activation, C1-INH becomes a target of the enzymatic activity of the bound C1r2 C1s2. C1-INH is cleaved by and becomes covalently attached to these complement proteins, and, as a result, the C1r2-C1s2 tetramer dissociates from C1q, thus stopping activation by the classical pathway (Fig. 1). In this way, C1-INH pre vents the accumulation of enzymatically active C1r2-C1s2 in the plasma and limits the time for which active C1r2-C1s2 is available to activate subsequent steps in the complement cascade. Similarly, by inactivating MASP2, C1-INH also dampens the lectin pathway. An autosomal dominant inherited disease called hereditary angioedema is due to a deficiency of C1-INH. Clinical manifestations of the disease include intermittent acute accumulation of edema fluid in the skin and mucosa, which causes abdominal pain, vomiting, diarrhea, and potentially life-threatening airway obstruction. In some of these patients, the plasma levels of C1-INH protein are sufficiently reduced (<20% –30% of normal) that activation of C1 by immune complexes is not properly controlled and increased breakdown of C4 and C2 occurs. The media tors of edema formation in patients with hereditary angioedema include a proteolytic fragment of C2, called C2 kinin, and bradykinin. C1-INH is an inhibitor of other plasma serine proteases besides C1, including kallikrein and coagulation factor XII, both of which can promote increased formation of bradykinin. Recombinant C1-INH is now one of the therapies used to treat patients with this deficiency.
• Assembly of the components of C3 and C5 convertases is inhibited by the binding of regulatory proteins of the RCA family to C3b and C4b deposited on cell surfaces (Fig. 2). If C3b is deposited on the surfaces of normal mammalian cells, it may be bound by several membrane proteins, including MCP (CD46), CR1, and DAF, and the plasma protein Factor H. C4b deposited on cell surfaces is similarly bound by DAF, CR1, MCP, and another plasma protein, C4BP. By binding to C3b or C4b, these proteins competitively inhibit the binding of other components of the C3 convertase, such as Bb of the alternative pathway and C2a of the classical pathway, thus blocking further progression of the complement cascade. (Factor H inhibits binding of only Bb to C3b and is thus a regulator of the alternative but not the classical pathway.) MCP, CR1, and DAF are produced by mammalian cells but not by microbes. Therefore, these regulators of complement selectively inhibit complement activation on host cells but complement activation proceeds on microbes. In addition, cell surfaces rich in sialic acid favor binding of the regulatory protein Factor H over the alternative pathway protein Factor B. Mammalian cells express higher levels of sialic acid than most microbes do, which is another reason that complement activation is prevented on normal host cells and permitted on microbes.
Fig1. Regulation of C1 activity by C1 inhibitor (INH). C1 INH displaces C1r2s2 from C1q and terminates classical pathway activation.
Fig2. Inhibition of the formation of C3 convertases. The classical pathway C3 convertase, C4b2a, or the alternative pathway C3 convertase, C3bBb, can be dissociated by the replacement of one com ponent with decay accelerating factor (DAF). Other regulatory proteins, such as membrane cofactor protein and complement receptor 1, function similarly to DAF.
DAF is a glycophosphatidylinositol (GPI)-linked mem brane protein expressed on endothelial cells and erythrocytes. A deficiency in hematopoietic stem cells of the enzyme required to form such protein-lipid linkages results in the failure to express many GPI-linked membrane proteins, including DAF and CD59 (see following), and causes a disease called paroxysmal nocturnal hemoglobinuria. This disease is characterized by recurrent bouts of intravascular hemolysis, at least partly attributable to unregulated complement activation on the surface of erythrocytes. Recurrent intravascular hemolysis in turn leads to chronic hemolytic anemia and venous thrombosis. An unusual feature of this disease is that the causative mutation in the gene that encodes an enzyme responsible for the generation of the GPI anchor is not inherited but is an acquired somatic mutation in hematopoietic stem cells.
• Cell-associated C3b is proteolytically degraded by a plasma serine protease called Factor I, which is active only in the presence of regulatory proteins (Fig. 3). MCP, Factor H, C4BP, and CR1 all serve as cofactors for Factor I–mediated cleavage of C3b (and C4b). Thus, these regulatory host cell proteins promote proteolytic degradation of complement proteins; as discussed earlier, the same regulatory proteins cause dissociation of C3b (and C4b)-containing complexes. Factor I–mediated cleavage of C3b generates the fragments called iC3b, C3d, and C3dg, which do not participate in complement activation but are recognized by receptors on phagocytes and B lymphocytes.
• Inflammation induced by C3a and C5a is regulated by the rapid cleavage of their C-terminal arginine residues by plasma carboxypeptidases. This results in the generation of C3a des-Arg and C5a des-Arg, which each have only approximately 10% of the activity of the native forms of these proteins.
• Formation of the MAC is inhibited by a membrane protein called CD59. CD59 is a GPI-linked protein expressed on many cell types. It works by incorporating itself into assembling MACs after the membrane insertion of C5b-8, thereby inhibiting the subsequent addition of C9 molecules (Fig. 4). CD59 is present on normal host cells, where it limits MAC formation, but it is not present on microbes. Formation of the MAC is also inhibited by plasma proteins such as S protein, which functions by binding to soluble C5b,6,7 complexes and thereby preventing their insertion into cell membranes near the site where the complement cascade was initiated. Growing MACs can insert into any neigh boring cell membrane besides the membrane on which they were generated. Inhibitors of the MAC in the plasma and in host cell membranes ensure that lysis of innocent bystander cells does not occur near the site of complement activation.
Fig3. Factor I–mediated cleavage of C3b. In the presence of cell membrane–bound cofactors (mem brane cofactor protein [MCP] or complement receptor 1 [CR1]), plasma factor I proteolytically cleaves C3b attached to cell surfaces, leaving an inactive form of C3b (iC3b). Factor H and C4-binding protein can also serve as cofactors for factor I–mediated cleavage of C3b. The same process is involved in the proteolysis of C4.
Fig4. Regulation of formation of the membrane attack complex (MAC). The MAC is formed on cell surfaces as an end result of complement activation. The membrane protein CD59 and S protein in the plasma inhibit formation of the MAC.
Much of the analysis of the function of complement regulatory proteins has relied on in vitro experiments, and most of these experiments have focused on assays that measure MAC mediated lysis of erythrocytes as an endpoint. On the basis of these studies, a hierarchy of importance for inhibiting complement activation is believed to be CD59 > DAF > MCP; this hierarchy may reflect the relative abundance of these proteins on cell surfaces.
The function of regulatory proteins may be overwhelmed by excessive activation of complement pathways. We have emphasized the importance of these regulatory proteins in preventing complement activation on normal cells. However, complement mediated phagocytosis and damage to normal cells are important pathogenic mechanisms in many immunologic diseases. In these diseases, large amounts of antibodies may be deposited on host cells, generating enough active complement proteins that the regulatory molecules are unable to control complement activation.
