There are three major pathways of complement activation: the classical pathway, which is activated by certain classes of antibodies bound to antigens; the alternative pathway, which is activated on microbial cell surfaces in the absence of antibody; and the lectin pathway, which is activated by lectins that bind to surface carbohydrates on microbes (Fig. 1). The names classical and alternative arose because the classical pathway was discovered and characterized first, but the alternative pathway is phylogenetically older. Although the pathways of complement activation differ in how they are initiated, all of them result in cleavage of the most abundant complement protein, C3. The alternative and lectin pathways are effector mechanisms of innate immunity, whereas the classical pathway is a major mechanism of adaptive humoral immunity.
Fig1. The early steps of complement activation by the alternative, classical, and lectin pathways. The alternative pathway is activated by C3b binding to various activating surfaces, such as microbial cell walls; the classical pathway is initiated by C1 binding to antigen-antibody complexes; and the lectin pathway is activated by binding of a plasma lectin to carbohydrates on microbes. The C3b that is generated by the action of the C3 convertase binds to the microbial cell surface or the antibody and becomes a component of the enzyme that cleaves C5 (C5 convertase) and initiates the late steps of complement activation. The late steps of all three pathways are the same (not shown), and complement activated by all three pathways serves the same functions. Ig, Immunoglobulin.
The central event in complement activation is proteolysis of the complement protein C3 to generate biologically active products and the subsequent covalent attachment of a fragment of C3, called C3b, to microbial cell surfaces or to antibodies bound to antigens (see Fig. 1). Complement activation involves the generation of a proteolytic complex, the C3 convertase, which cleaves C3 into two fragments, called C3a and C3b. (By convention, the proteolytic products of each complement protein are identified by lowercase letter suffixes, “a” refers to the smaller product and “b” to the larger one; C2 is an exception, for historical reasons.) C3b becomes covalently attached to the microbial cell surface or to antibody molecules bound to antigens. All of the biologic functions of complement are dependent on the proteolytic cleavage of C3. For example, complement activation promotes phagocytosis because C3b becomes covalently linked to microbes, and phagocytes (neutrophils and macrophages) express receptors for C3b. When C3 is broken down, C3a is released. This soluble fragment stimulates inflammation.
C3 proteolysis also initiates the subsequent steps of the complement cascade. In all three pathways of complement activation, after the generation of C3b by the C3 convertase, a second enzyme complex called the C5 convertase is assembled, which cleaves C5 into C5a and C5b. The C5a fragment induces inflammation and C5b initiates the late steps of the complement cascade that culminate in the formation of pores in the membranes of microbial targets. The pathways of complement activation differ in how C3b is produced but follow a common sequence of reactions after the cleavage of C5.
With this background, we proceed to more detailed descriptions of the alternative, classical, and lectin pathways.
The Alternative Pathway
The alternative pathway of complement activation results in the proteolysis of C3 and the stable attachment of its break down product C3b to microbial surfaces, without a role for antibodies (Fig. 2 and Table 1). The C3 protein contains a reactive thioester bond that is buried in a region of the protein known as the thioester domain. Normally, C3 in plasma is continuously cleaved at a low rate (1%–2% of the total plasma C3 per hour) to generate C3b in a process that is called C3 tick over. Tickover involves spontaneous hydrolysis of the reactive thioester bond in C3 that facilitates cleavage of C3 by Factors B and D, as described below. When C3 is cleaved into C3b and C3a, the C3b molecule undergoes a dramatic conformational change (a large shift of approximately 85 Å) and the thioester domain flips out, exposing the previously hidden reactive thioester bond. A small amount of the C3b may become covalently attached to the surfaces of cells, including microbes, through the thioester domain, which reacts with the amino or hydroxyl groups of cell surface proteins or polysaccharides to form amide or ester bonds (Fig. 3). If these bonds are not formed, the C3b remains in the fluid phase, and the exposed reactive thioester bond is quickly hydrolyzed, rendering the protein inactive. As a result, further complement activation cannot proceed.
Fig2. The alternative pathway of complement activation. Spontaneous hydrolysis of plasma C3 leads to the formation of a fluid-phase C3 convertase and the generation of C3b, which is deposited on microbial surfaces, where it binds Factor B and forms the alternative pathway C3 convertase. This convertase (C3bBb) cleaves C3 to produce more C3b, which binds to the microbial surface for opsonization of the microbe, and also covalently attaches to the C3 convertase and thus participates in the formation of a C5 convertase. The C5 convertase, C3bBbC3b, cleaves C5 to generate C5b, the initiating molecule for the late steps of complement activation.
Table1. Proteins of the Alternative Pathway of Complement
Fig3. Internal thioester bonds of C3 molecules. Proteolytic cleavage of the α chain of C3 converts it into a metastable form in which the internal thioester bonds are exposed and susceptible to nucleophilic attack by oxygen atoms (as shown) or nitrogen atoms. The result is the formation of covalent bonds with proteins or carbohydrates on the cell surfaces. C4 is structurally homologous to C3 and has an identical thioester group.
When C3b undergoes its postcleavage conformational change, a binding site for a plasma protein called Factor B is exposed. Factor B then binds to the C3b protein that is now covalently tethered to the surface of the cell. Bound Factor B is in turn cleaved by a plasma serine protease called Factor D, releasing a small fragment called Ba and generating a larger fragment called Bb that remains attached to C3b. The C3bBb complex is the alternative pathway C3 convertase, and it functions to cleave more C3 molecules, thus setting up an amplification sequence. Even when C3b is generated by the classical or lectin pathway, it can form a complex with Bb, and this complex is able to cleave more C3. Thus, the alternative path way C3 convertase functions to amplify complement activation when it is initiated by the alternative, classical, or lectin pathway.
Alternative pathway activation readily occurs on microbial cell surfaces but not on mammalian cells. If the C3bBb complex is formed on mammalian cells, it is rapidly degraded, and the reaction is terminated by the action of several regulatory proteins present on these cells (discussed later). Lack of the regulatory proteins on microbial cells allows binding and activation of the alternative pathway C3 convertase. In addition, another protein of the alternative pathway, called properdin, can bind to and stabilize the C3bBb complex, and the attachment of properdin is favored on microbial as opposed to normal host cells. Properdin is released by activated neutrophils and can also be made by macrophages and some T cells.
Some of the C3b molecules generated by the alternative path way C3 convertase bind covalently to the convertase itself. This results in the formation of a complex containing one Bb moiety and two molecules of C3b, which functions as the alternative pathway C5 convertase, which cleaves C5 and initiates the late steps of complement activation.
The Classical Pathway
The classical pathway is initiated by binding of the complement protein C1 to the CH 2 domains of IgG or the CH 3 domains of IgM molecules that have bound antigen (Fig. 4 and Table 2). Among IgG antibodies, IgG1 and IgG3 (in humans) are more efficient activators of complement than are other subclasses. IgG2 has some ability to activate complement, but IgG4 does not. C1 is a large, multimeric protein complex composed of C1q, C1r, and C1s subunits; C1q binds to the antibody, and C1r and C1s are proteases. The C1q subunit is made up of an umbrella like radial array of six chains, each of which has a globular head connected by a collagen-like arm to a central stalk (Fig. 5). This hexamer performs the recognition function of the molecule and binds specifically to the Fc regions of µ and some γ heavy chains. As mentioned in Chapter 4, C1q also binds to pentraxins like C-reactive protein and serum amyloid protein, and can also bind to apoptotic bodies.
Fig4. The classical pathway of complement activation. Antigen antibody complexes that activate the classical pathway may be soluble, fixed on the surface of cells (as shown), or deposited on extracellular matrices. The classical pathway is initiated by the binding of C1 to anti gen-complexed antibody molecules, which leads to the production of C3 and C5 convertases attached to the surfaces where the antibody was deposited. The C5 convertase cleaves C5 to begin the late steps of complement activation.
Table2. Proteins of the Classical Pathway of Complement
Fig5. Structure of C1. C1q consists of six identical subunits arranged to form a central core and symmetrically projecting radial arms. The globular heads at the end of each arm, designated H, are the contact regions for immunoglobulin. C1r and C1s form a tetramer composed of two C1r and two C1s molecules. The ends of C1r and C1s contain the catalytic domains of these proteins. One C1r2s2 tetramer wraps around the radial arms of the C1q complex in a manner that juxta poses the catalytic domains of C1r and C1s. Ig, Immunoglobulin.
Only antibodies bound to antigens, and not free circulating antibodies, can initiate classical pathway activation (Fig. 6). The reason for this is that each C1q molecule must bind to at least two Ig heavy chains to be activated and each Ig Fc region has only a single C1q-binding site. Therefore, two or more Fc regions have to be accessible to C1 to initiate classical pathway activation. Because each IgG molecule has only one Fc region, multiple IgG molecules must be brought close together before C1q can bind, and multiple IgG antibodies are brought together only when they simultaneously attach to identical epitopes of a multivalent antigen or to several antigen molecules on a microbe, cell, or tissue surface. Even though free (circulating) IgM is pentameric, it does not bind C1q because the Fc regions of free IgM are in a conformation that is inaccessible to C1q. Binding of the IgM to an antigen induces a change that exposes the C1q binding sites in the Fc regions and allows C1q to bind. Because of its pentameric structure, a single molecule of IgM can bind two or more C1q molecules, and this is one reason that IgM is a more efficient complement-binding (also called complement-fixing) anti body than is IgG.
Fig6. C1 binding to the Fc portions of immunoglobulin M (IgM) and IgG. C1 must bind to two or more Fc portions to initiate the complement cascade. Soluble IgG molecules will not activate C1 because each IgG has only one Fc region (A), but after binding to cell surface antigens, adjacent IgG Fc portions can bind and activate C1 (B). The Fc portions of soluble pentameric IgM are not accessible to C1 (C). After IgM binds to surface-bound antigens, it undergoes a shape change that permits C1 binding and activation (D).
C1r and C1s are serine proteases that form a tetramer containing two molecules of each protein. The binding of two or more of the globular heads of C1q to the Fc regions of IgG or IgM leads to enzymatic activation of the associated C1r, which cleaves and activates C1s (see Fig. 4). Activated C1s cleave the next protein in the cascade, C4, to generate C4b. (The smaller C4a fragment is released and has biologic activities that are described later.) C4 is homologous to C3, and C4b contains an internal thioester bond, similar to that in C3b, that forms covalent amide or ester linkages with the antigen-antibody complex or with the adjacent surface of a cell to which the antibody is bound. This attachment of C4b ensures that classical pathway activation proceeds on a cell surface or immune complex. The next complement protein, C2, then forms a complex with the cell surface–bound C4b and is cleaved by a nearby C1s molecule to generate a soluble C2b fragment of unknown importance and a larger C2a fragment that remains physically associated with C4b on the cell surface. (Note that the nomenclature of C2 fragments is different from that of the other complement proteins because the attached, larger fragment is called the a piece and the released part is the b fragment.) The resulting C4b2a complex is the classical pathway C3 convertase; it has the ability to bind to and proteolytically cleave C3. The binding of C3 to the C3 convertase is mediated by the C4b component, and proteolysis of C3 is catalyzed by the C2a component. The cleavage of C3 results in removal of the small C3a fragment, and the larger C3b fragment can form covalent bonds with cell surfaces or with the antibody where complement activation was initiated. After C3b is deposited, it can bind Factor B and generate more C3 convertase, essentially using the alternative pathway discussed earlier as an amplification loop. The net effect of the multiple enzymatic steps and amplification is that millions of molecules of C3b can be deposited within minutes on the cell surface where complement is activated. The key early steps of the alternative and classical pathways are analogous: C3 in the alternative pathway is homologous to C4 in the classical path way, and Factor B is homologous to C2.
Some of the C3b molecules generated by the classical path way C3 convertase bind covalently to the convertase (as in the alternative pathway) and form a C4b2a3b complex. This com plex functions as the classical pathway C5 convertase; it cleaves C5 and initiates the late steps of complement activation.
The Lectin Pathway
The lectin pathway of complement activation is triggered by the binding of microbial polysaccharides to circulating lectins, such as plasma mannose-binding lectin (MBL), or to ficolins (Table 3). The soluble lectins are collagen-like proteins that structurally resemble C1q. MBL has an N-terminal collagen-like domain and a C-terminal carbohydrate recognition (lectin) domain and is thus a member of the collectin family of serum agglutinins. The ficolins, including L-, H-, and M-ficolin, have a similar structure, with an N-terminal collagen-like domain and a C-terminal fibrinogen-like domain. The collagen like domains help to assemble basic triple-helical structures that can form higher-order oligomers. MBL, L-ficolin, and H-ficolin are plasma proteins; M-ficolin is mainly secreted by activated macrophages in tissues. MBL binds to mannose residues on polysaccharides, and the fibrinogen-like domain of ficolin binds N-acetylglucosamine–containing glycans. These polysaccharides and glycans are abundant in bacteria and fungi. Ficolins bind to N-acetylated sugar residues of carbohydrates on bacterial and fungal cell walls. Both MBL and ficolins associate with MBL associated serine proteases (MASPs) including MASP1, MASP2, and MASP3 (see Table 3). The MASPs are structurally homologous to the C1r and C1s proteases and serve a similar function, namely, the cleavage of C4 and C2 to activate the complement pathway. Multimers of MBL associate with a protease complex made up of MASP1 associated with MASP2; MASP1 may activate MASP2 and MASP2 is the protease that cleaves C4 and C2. A related protease, MASP3, may be more relevant for the alternative pathway. Subsequent events in this pathway are identical to those that occur in the classical pathway.
Table3. Proteins of the Lectin Pathway of Complement
A fourth pathway of complement has been described that is linked to adaptive Immunity and that utilizes the same C3 and C5 convertases used by the classical and lectin pathways. Subsets of CD8+ and Th1-like CD4+ T cells secrete Granzyme K constitutively. Granzyme K can directly cleave C4 and C2 at tissue sites, activating complement.
Late Steps of Complement Activation
C5 convertases generated by the alternative, classical, or lectin pathway initiate activation of the late components of the complement system, which culminates in formation of the cytocidal membrane attack complex (MAC) (Table 4 and Fig.7). C5 convertases cleave C5 into a small C5a fragment that is released and a two-chain C5b fragment (containing an α and a β chain) that is also released and binds to plasma C6. C6 undergoes a conformational change, and the C5b-C6 complex binds to the cell membrane through both ionic and hydrophobic interactions. C5a has potent biologic effects on several cells that are discussed later. C7 from the plasma then binds to the α chain of C5b and forms the C5b-C6-C7 (C5b-7) complex. The bound C7 undergoes an amphiphilic transition and penetrates the plasma membrane and can contribute to the release of some phospholipid micelles from the membrane but does not form complete pores. Once the C5b-7 complex is in the lipid bilayer of the membrane, it recruits C8, a protein trimer composed of three distinct chains. This stably inserted C5b,6,7,8 complex (C5b-8) forms unstable pores that range from 0.4 to 3 nm in diameter, and while very large numbers of these C5b-8 complexes can lyse cells, this is an inefficient process. The formation of a fully active MAC is accomplished by the binding of C9, the final component of the complement cascades, to the C5b-8 complex. The rate-limiting step in MAC formation is the recruitment of the first C9 molecule, which rapidly recruits other C9 molecules from the plasma. C9 polymerizes at the site of the bound C5b-8 to form pores in plasma membranes that are made up of C5b-9 complexes containing C5b, C6, C7, C8, and up to 18 molecules of C9. These pores are approximately 20 nm in external diameter and 1 to 11 nm in internal diameter, with a height of approximately 15 nm, and they form channels that allow free movement of water and ions. The channel size varies based on the number of C9 molecules in the C5b-C9 complex. Tubular complexes of C9 alone may also form. The entry of water results in osmotic swelling and rupture of the cells on whose surface the MAC is deposited. The pores formed by polymerized C9 are similar to the membrane pores formed by perforin, the granule protein found in cytotoxic T lymphocytes and NK cells that promotes the cytolytic actions of these cells, and C9 is structurally homologous to perforin.
Table4. Proteins of the Late Steps of Complement Activation
Fig7. Late steps of complement activation and formation of the membrane attack complex (MAC). The cell-associated C5 convertase cleaves C5 and generates C5b, becomes bound to the convertase. C5b binds C6 and C7 sequentially, and the C5b-7 complex inserts into the plasma membrane, followed by the formation of the C5b-8 com plex, which forms unstable pores. The C5b-8 complex can form a pore with C9, and C9 can also be induced to homo-oligomerize by the C5b-8 complex. As many as 15 C9 molecules may polymerize to form the MAC, which creates pores in the membrane and induces cell lysis. C5a released on proteolysis of C5 stimulates inflammation.
