ACP medicine, 3rd Edition


Innate Immunity

  1. Kathryn Liszewski B.A.1

Wayne M. Yokoyama M.D.2

John P. Atkinson M.D., F.A.C.P.3

1Washington University School of Medicine

2Washington University School of Medicine

3Samuel B. Grant Professor, Professor of Medicine and Molecular Microbiology, Washington University School of Medicine St. Louis, Missouri

The authors have no commercial relationships with manufacturers of products or providers of services discussed in this chapter.

June 2004

The body uses two forms of immunity to recognize and respond to infectious microorganisms: innate immunity, which is nonspecific, and adaptive immunity, which is highly specific. The innate immune system is ancient, being present in all multicellular organisms. Adaptive immunity developed from the innate system, beginning about 400 million years ago in primitive fishes such as the lamprey. In vertebrates, the two systems cooperate to produce an immune response. Genetically, the innate system is hardwired and inflexible. In contrast, the adaptive system splices DNA to make new genes that can code for an almost infinite array of immunoglobulin and T cell receptors.

The innate immune system is particularly active at the interface between the environment and those surfaces of the body that are lined with epithelial cells—namely, the skin and the gastrointestinal, genitourinary, and sinopulmonary tracts. Intact physical barriers are, of course, critically important for preventing infections.

The epithelial cells' innate immune response to disturbance by a pathogen has only recently begun to be deciphered. For example, Toll-like receptors—the mammalian counterpart of Toll receptors in fruit flies—were not discovered until 1997. Engagement of these epithelial cell receptors results in the activation of genes that encode inflammatory cytokines, such as tumor necrosis factor (TNF), interferons, and interleukin-1 (IL-1) and IL-12, as well as genes that encode antimicrobial peptides (“Nature's antibiotics”).

In addition to the epithelial barrier itself, the fluids in these tracts contain mucus, natural antibodies (IgG and IgA), a complement system, and lectins. Mucus itself is a protective film that traps organisms and debris and also contains antibacterial substances. The complement system in secretions is present at about 10% to 20% of the concentration found in plasma. The lectins in these secretions bind sugars on pathogens and thereby activate the lectin pathway of complement activation. It is probably safe to say that for every major human pathogen, there is a lectin that recognizes a sugar structure on the pathogen's surface. Granulocytes undergo margination in small blood vessels throughout much of these barrier tissues and are available for rapid recruitment to a site of possible infection. Monocytes/macrophages are also present in secretions and in most tissues, where they phagocytose unwanted microbes.

From an evolutionary standpoint, the goal of the immune system is for the individual to survive to reproductive age. This concept suggests that a so-called hyperreactive immune system in early life may provide a survival advantage, even if it subsequently predisposes to allergies or autoimmunity. A second concept is that innate immunity provides guidance to the adaptive immune system regarding which infectious disease antigens it should respond to and what strategy to employ for the most efficient destruction of pathogens. In other words, it tells the adaptive immune system when and how to respond to a pathogen. A third point is that, in contrast to normally sterile areas of the body, such as the bloodstream or spinal fluid, epithelial surfaces are inhabited by large numbers of commensal organisms, against which an immune response would be unnecessary and even potentially harmful. Consequently, at these barriers, the immune system must not only discriminate between self and nonself but, faced with a sea of nonself microorganisms, must single out the dangerous ones. Failure to meet this substantial challenge is not uncommon and results in allergies (reaction to a foreign but benign substance) and autoimmunity (reaction to self).

The innate immune system also accounts for many aspects of the sepsis syndrome. Most house officers vividly remember their first patient with bacteremia or septicemia.1 Neither textbooks nor attending physicians could provide sound explanations for the signs and symptoms in these patients. Instead, the explanation offered would usually involve “too many bugs in the blood.” Of course, microbes in the bloodstream are the primary cause, but the innate immune system's reaction to those organisms is what produces the clinical syndrome. The simplified paradigm is that the bacteria and their products become bound—either directly or through soluble proteins (i.e., lectins, natural antibodies, complement proteins)—to receptors on endothelial and epithelial cells, and this binding signals cytokine and chemokine release (creating a so-called cytokine storm). Most of the clinical signs and symptoms of sepsis result from the effects of excessive quantities of these mediators on vascular endothelium and epithelial cells.

Epithelial Cell Receptors and Their Signaling Pathways


The breakthrough in the field of epithelial cell receptors took place when Janeway and colleagues recognized that the Toll (German slang for “way out” or “crazy”) receptor of the fruit fly had intracytoplasmic signaling motifs with homology to those found in mammalian IL-1 and TNF receptors.2,3,4,5,6 Sequencing of the fruit fly genome led to the discovery of 10 to 20 of these receptors, as well as related soluble binding proteins. These proteins directly recognize key structural components on microbial pathogens—for example, lipopolysaccharides in the outer membrane of gram-negative organisms or peptidoglycans in the cell wall of gram-positive organisms [see Table 1]. These structural components are not easily modified in a substantial fashion by the organism. In many cases, soluble binding proteins make first contact with the organism, which then facilitates engagement of the receptor. The subsequent cellular signaling involves nuclear factor-κB (NF-κB) and its intracellular signaling cohorts; these cohorts are analogous to those that mammalian cells use to set off an inflammatory response. Once activated, NF-κB travels to the nucleus, where it triggers the transcription of genes that prepare the host cell to do battle with the infecting organism. There is specificity to the response in the sense that the cytokines and the defensins elaborated are tailored to the class of microbe. For example, the cell elaborates different antibacterial defensins for a gram-negative organism than for a gram-positive organism, and it elaborates antifungal peptides distinct from those liberated in an antibacterial response.

Table 1 Toll-like Receptors and Their Ligands4




Lipoproteins, lipoarabinomannan, LPS from Leptospira and Porphyromonas gingivalis, peptidoglycan from gram-positive bacteria, zymosan from yeast, GPI anchor fromTreponema cruzi


LPS from gram-negative bacteria (except Leptospira and P. gingivalis), plant products (Taxol), viral products (RSV), host-derived products (HSP60, fibronectin fragments)





GPI—glycosylphosphatidylinositol HSP60—heat shock protein 60 LPS—lipopolysaccharide RSV—respiratory syncytial virus TLR—Toll-like receptor

Toll-like receptors and pattern recognition receptors (PRRs) [see Table 2] were eventually found on human epithelial cells and monocytes/macrophages. Amazingly, this same general scheme (i.e., the Toll-like receptor and its signaling pathway) is used by tomato plants to ward off the mosaic virus.7

Table 2 Pattern-Recognition Receptors36

Pattern-Recognition Receptor (PRR) Type


Protein/Domain Family



Secreted PRRs


C-type lectin

Terminal mannose residents

Activation of the lectin pathway of complement



Phosphorylcholine on microbial membranes

Opsonization; activation of classical complement pathway


Lipid-transfer protein family


LPS recognition

Cell-surface PRRs


Leucine-rich repeats, C-type lectin

LPS, peptidoglycan

Coreceptor for TLRs

Macrophage mannose receptor

C-type lectin

Terminal mannose residues


Macrophage scavenger receptor

Scavenger receptor, cysteine-rich domain

LDL, anionic polymers

Phagocytosis, LPS clearance, and lipid homeostasis


Scavenger receptor, cysteine-rich domain

Bacterial cell walls


Intracellular PRRs


dsRNA-binding domain, protein kinase domain


Activation of NF-κB and MAP kinases; inhibition of translation and induction of apoptosis in virally infected and stressed cells


Leucine-rich repeats, nucleotide-binding domain, CARD domain

Ligands for most Nod proteins are unknown; Nodl and Nod2 recognize LPS

Activation of NF-κB and MAP kinases; some family members may be involved in the induction of apoptosis; the exact function is unknown

CARD—caspase-recruitment domain CRP—C-reactive protein dsRNA—double-stranded RNA LBP—lipopolysaccharide-binding protein LDL—low-density lipoprotein LPS—lipopolysaccharide MAP—mitogen-activated protein MARCO—macrophage receptor with collagenous structure MBL—mannan-binding lectin NF-kB—nuclear factor-kB Nod—nucleotide-binding oligomerization domain PKR—dsRNA-activated protein kinase SAP—serum amyloid protein TLRs—Toll-like receptors

Innate Immune Cells


Natural killer (NK) cells constitute the third major population of lymphocytes, after T cells and B cells.8 They were initially identified because they spontaneously (i.e., naturally) kill certain tumor cells, a process that does not require prior exposure to the target. Like T cells and B cells, NK cells are involved in host immune defense. They are more closely related to T cells than to B cells in that they share effector functions, including the same killing mechanism and the capacity to produce cytokines. Unlike other lymphocytes, NK cells are components of innate immunity—they respond early against infections (and possibly tumors) and do not require gene rearrangement for maturation and function. Much remains to be clarified regarding the nature of NK cell receptors and their ligands. NK cells are negative for the T cell receptor (CD3) and B cell receptor (membrane immunoglobulin). Most NK cells in human peripheral blood are CD56+; this feature can be used for their identification, because expression of this adhesion-promoting molecule is restricted to NK cells and a small population of T cells. NK cells also express the transmembrane form of the low-affinity receptor for IgG (CD16 or FcγRIIIA) that is absent on mature T cells. When CD16 on the NK cell binds the Fc portion of IgG that is coating a target, this receptor activates release of cytoplasmic granules containing molecules that form pores in the target cell membrane and others that mediate apoptosis, resulting in antibody-dependent cellular cytotoxicity. Natural killing is mediated by the same mechanism, although CD16 is not required. Identification of the receptors that initiate natural killing is a topic of active research.

Of note, NK cells are better able to kill cells that lack major histocompatibility complex (MHC) or human leukocyte antigen (HLA) class I molecules, such as may result from tumorigenesis or viral infection. Although decreased expression of MHC class I molecules may allow targets to evade destruction by MHC class I-restricted cytotoxic T cells, it makes the targets more susceptible to killing by NK cells. This finding led to the so-called missing-self hypothesis, which holds that NK cells survey tissues for MHC class I molecules, which are normally expressed on most nucleated cells in the body.9 If MHC class I expression is decreased or absent, the NK cells are released from the negative influence of MHC class I and kill the target. This process may provide a fail-safe mechanism to protect the body from disease processes that evade acquired, specific T cell immunity.

Ongoing studies indicate that NK cells express a multitude of inhibitory receptors that guide their capacity to kill tumor and virus-infected cells.10 These receptors, termed killer immuno globulin-like receptors and CD94/NKG2 heterodimers, bind to HLA molecules on their targets. Subsequently, specific tyrosine residues are phosphorylated within immunoreceptor tyrosine-based inhibitory motifs (ITIMs) in the cytoplasmic domains. This results in the recruitment and activation of cytoplasmic phosphatases that dephosphorylate molecules in the activation cascade, hence inhibiting NK cell stimulation. NK cells also express related molecules lacking ITIMs. These molecules are prime candidates for activation receptors that bind as yet uncharacterized molecules on the surface of the target cell.

Early studies of NK cells focused on their role in tumor surveillance (i.e., the eradication of cancers before they become clinically apparent). NK cells can be expanded in tissue culture by exposure to high concentrations of the lymphokine IL-2, which results in production of lymphokine-activated killer (LAK) cells. Adoptive transfer of LAK cells into patients with radiation- and chemotherapy-resistant tumors can lead to remissions.11 However, high concentrations of IL-2 have to be simultaneously administered, increasing the risk of potentially serious complications such as pulmonary edema and capillary leak syndromes.

Subsequent studies have shown that NK cells also play a critical role in early innate immune responses to viral infections. Persons who lack NK cells suffer recurrent, severe systemic viral infections, particularly from herpesviruses.12 Depletion of NK cells has also been described in patients with advanced HIV infection and AIDS. This depletion, which apparently results from infection of the NK cell itself by herpesvirus 6 and HIV,13 may partially account for these patients' susceptibility to opportunistic infections such as those from herpesvirus and cytomegalovirus.

NK cell lymphomas and leukemias are rarely found in Western populations14 but are more common in Asian populations. NK cell lymphomas often present as nasal tumors, are associated with Epstein-Barr virus infections, and may occur in conjunction with autoimmune syndromes.

Natural killing by peripheral blood NK cells is altered in a variety of conditions. However, the significance of such findings is unclear. On the other hand, as more reagents become available to definitively detect NK cells in clinical specimens, the apparent role of NK cells in disease pathogenesis will become clearer, as suggested by studies indicating marked expansion of NK cells in synovial fluid of patients with early rheumatoid arthritis.14


T cells were traditionally thought to respond only to peptide antigens presented in the context of self-MHC. However, CD1-restricted T cells respond to a variety of nonpeptide antigens, especially glycolipids. These cells appear to be more closely related to NK cells than to traditional T cells, because they have a very restricted T cell receptor repertoire and often express molecules usually found on NK cells; hence their designation as NKT cells. NKT cells may have arisen to combat pathogens that feature lipid antigens, such as mycobacteria. Increasing evidence points to populations of T cells as being specifically designed to present complex lipids and related antigens for an immune response.

γδ T cells are found in large numbers in the mucous membranes of many tissues, including the skin, small intestine, female reproductive tract, and lung. These locations suggest that they are a first line of defense against invading microorganisms. These cells appear to mediate cellular immune functions without requiring antigen processing. Thus, these cells are probably important in triggering and controlling the local immune response to pathogens such as Mycobacterium tuberculosis and Listeria monocytogenes. In so doing, they shape the local inflammatory reaction to the invading pathogen, which includes playing a role in maintaining host tissue integrity. There is still much to learn about how these NKT cell and γδ T cell populations function in the immune response. One hypothesis is that they are the T cell equivalents of natural antibodies and lectins, insofar as they are programmed to provide an innate immune response to certain organisms.


Macrophages are present in body secretions and in tissue sites bordering epithelial tissues. In these locations, they respond to pathogens with phagocytosis and the release of cytokines. They accomplish this by expressing many types of receptors, including PRR, Toll, lectin, scavenger, Fcγ, and complement.15 In general, these receptors can be divided into two groups. Opsonic receptors recognize microbes coated with immunoglobulins and complement. In contrast, the nonopsonic receptors directly recognize a structure on the surface of the infecting organism. Most of the PRR, Toll, and lectin receptors are nonopsonic. The principles that govern the function of the Toll receptors and PRR (see above) also apply to macrophages, including the signaling pathways and cellular response of cytokine production. Of course, a major difference is that macrophages can phagocytose and kill an organism through oxidative and other microbicidal means not available to epithelial cells. The secreted cytokines, like those from epithelial cells, are designed to activate nearby cells for an inflammatory/immune response. The macrophage lectin receptors recognize sugar moieties (e.g., mannose, galactose, β-glucan, galactose, N-acetylglucosamine, and other oligosaccharides). The fine structure of these moieties differs sufficiently from that of host sugars that receptors recognize them but not host sugars. One complement receptor, CR3, like several of the lectin receptors, has multiple specificities: it recognizes C3bi (an opsonic receptor), as well as β-D-glucose, some mannoses, and N-acetylglucosamine-containing sugars. In most cases, the recognition of microbes by macrophages leads to phagocytosis and an immune response. However, some organisms use these receptors to gain entrance to intracellular compartments, which can be favorable locations.


The immune system has developed a system for capturing, processing, and then displaying antigens to lymphocytes. Macrophages can phagocytose and present antigens, but they do so much less efficiently than the so-called professional antigen-presenting cells (APCs, or dendritic cells). Most antigens enter via epithelial barriers, where they encounter a cell type bearing long dendritic processes. In the skin, these dendritic cells are called Langerhans cells. The capture of antigens by receptors on APCs (most such receptors are similar to those on macrophages, as described above) occurs by phagocytosis and pinocytosis. The APCs are relatively inefficient (i.e., immature) in their baseline state, but in an environment rich in inflammatory cytokines from epithelial cells or macrophages (as part of the innate immune response), these immature dendritic cells gradually turn into exceptionally efficient APCs. After antigen capture, the immature or maturing APCs migrate to regional lymph nodes, where they encounter T cells; this encounter initiates an adaptive immune response.

Cytokines of the Innate Immune System

In response to invading microbes, epithelial cells and macrophages secrete cytokines that mediate immune and inflammatory reactions. Cytokines communicate to the cells that produce these signals (autocrine action) and to nearby cells (paracrine action) through cytokine receptors. Through this process, cytokines prepare the local environment to engage microbes and to prepare immunocompetent cells for an immune response. Cytokines commonly released by macrophages and epithelial cells include TNF, IL-1, IL-6, IL-12, IL-15, interferon alfa, and interferon beta. Some of these cytokines of innate immunity also provide second signals for activation of B and T cells [see 6:III Adaptive Immunity: Antigens, Antibodies, and T Cell and B Cell Receptors].

The Complement System

The complement system lies at the interface between innate and acquired immunity. As a key component of innate immunity, it promotes the inflammatory response and attacks and destroys foreign substances.16,17,18,19 In this process, it is a facilitator and instructor of the adaptive immune response against the foreign antigen. Further, it provides one of the main effector mechanisms for antibody-mediated host defense [see Table 3].

Table 3 Roles of the Complement System in Immunity

First line of defense (innate immunity)
   Mediates an inflammatory response
   Modifies membranes of invading microbes
Instructor of adaptive immunity
   Antigen identification, processing, transportation, and retention
   Cellular activation and induction of costimulatory molecules, cytokines, and other mediators of immune response (lowers threshold for B cell signaling)
Effector arm of humoral immunity
   Promotes microbial killing via opsonization and lysis
   Removes antigens by immune complex formation

From its initial description more than 100 years ago as a heat-labile lytic substance that “complemented” antibodies in destroying bacteria, we now know that the complement system consists of three initiating arms, as well as many regulators and receptors [see Figure 1]. The early-reaction sequences behave as biologic cascades in which, by limited proteolysis, one component activates the next. This produces a rapid and robust amplification of the activation process. The primary goal is to attach clusters of C3b on a target (opsonization) so that phagocytes bearing C3 receptors can adhere to and then ingest the material. To counteract the potential for self-injury, almost half of the complement proteins serve as regulators or inhibitors. Despite this tight control, complement is an important contributor to tissue injury in ischemia-reperfusion injury and disease states, especially those featuring autoantibodies and immune complexes. Inherited deficiencies of components in the activating cascade predispose to infectious diseases, primarily of a pyogenic type; surprisingly, such deficiencies also predispose to autoimmunity, especially systemic lupus erythematosus (SLE). Deficiencies of regulators lead to excessive activation and tissue injury. Complement determinations facilitate the diagnosis and management of a number of illnesses. Although no agents for inhibiting complement activation are as yet commercially available, two promising candidates are in clinical trials, and more are in the pipeline.


Figure 1. Function of the Complement System

Function of the complement system. The early part of the complement system comprises three branches: the classical pathway (a), the lectin pathway (b), and the alternative pathway (c). These pathways converge to produce membrane pertubation (d) on bacterial surfaces. The most important function of complement is opsonization—the coating of a pathogen with clusters of complement activation fragments. These fragments, in turn, facilitate interactions with complement receptors and in some cases (e.g., certain gram-negative bacteria and viruses) lead to lysis. The second critical function of complement is to activate cells so as to promote inflammatory and immune responses. Through opsonization and cell activation, complement serves as nature's adjuvant to prepare, facilitate, and instruct the host for an adaptive immune response.34 (MASPs—mannan-binding lectin-associated serine proteases)


The early part of the complement system is divided into three branches: the antibody-initiated classical pathway (CP), the antibody-independent (i.e., innate) alternative pathway (AP), and the more recently described lectin pathway (LP). Although each branch is triggered differently, all share the common goal of depositing clusters of C3b on a target. This deposition results in the assembly of a common lytic mechanism, called the membrane attack complex (MAC) or C5b-9.

The CP proteins are identified by numbers (C1, C4, C2, and C3), and the AP proteins are designated by capital letters (factors B, D, and P) [see Table 4]. Each MAC consists of one each of C5b, C6, C7, and C8, and many C9s. In the early steps of the cascades, activated proteases cleave the next component in line to liberate a small fragment (these fragments are designated by a lower case “a” [e.g., C3a]), while the large fragment attaches to the target, after which it is designated “b” (e.g., C3b). Limited proteolytic cleavage produces further degradative fragments that are designated by lowercase letters; for example, C3b is cleaved to C3bi, and C3bi is in turn cleaved to C3c and C3dg [seeFigure 2]. Complement proteins expressed on cell membranes are regulators or receptors for activated components. They may be identified according to so-called cluster of differentiation (CD) designations or according to function. For example, complement receptor 1 (CR1) is also called CD35 and the C3b/C4b or immune adherence receptor.

Table 4 Plasma Components of the Complement Cascades



Serum Concentration (µg/ml)


Classical pathway (CP)


Part of C1 complex
Part of C1 complex
Part of C1 complex

Triggers CP
Binds Fc portion of IgG/IgM
Protease; cleaves C1s
Protease; cleaves C4 and C2
Opsonin,* C4a release
Protease; cleaves C3/C5

Alternative pathway (AP)

Factor B
Factor D

150 (wide range)

Binds sugars to trigger LP
Protease; cleaves C4 and C2
Protease; cleaves C3/C5
Protease; cleaves factor B
Stabilizes AP convertases

Central protein



Opsonin,* C3a release

Terminal pathway



MAC component (C5b), C5a release
MAC component
MAC component
MAC component for pore formation
MAC component for pore formation

C4b and C3b; also forms part of the C3 and C5 convertases.
LP—lectin pathway  MAC—membrane attack complex  MASP—MBL-associated serine protease  MBL—mannan-binding lectin


Figure 2. C3 Activation and Degradation

C3 activation and degradation. The boxed activation fragments are those that are covalently attached to the target. The other fragments are released into the surrounding milieu. C3a is a potent anaphylatoxin. C3f may be a marrow-mobilizing factor for neutrophils. C3c and C3g have no known function.


The AP is an ancient pathway of innate immunity. Unlike the CP, the AP does not require antibody for initiation. Rather, the natural breakdown (i.e., low-grade turnover) of plasma C3 via spontaneous cleavage of a highly reactive thioester bond allows C3 to attach to any nearby host or foreign surface. Regulatory proteins on host cells protect cells by inactivating such fragments. However, foreign membranes usually do not possess such inhibitors, so amplification (the feedback loop of AP) becomes engaged. Target-bound C3b binds to plasma component factor B. The latter undergoes proteolytic cleavage, mediated by plasma protein serine protease factor D, to produce Bb + Ba. The AP C3 convertase thus formed, C3bBb, is stabilized by properdin, which increases the half-life of the enzyme complex. As the stabilized convertase cleaves more C3 to C3b, a feedback loop becomes engaged for autoamplification. Through this mechanism, the AP can deposit several million C3bs on a bacterial surface in a few minutes.20

Membrane Attack Complex

As a further assault against a pathogen, the AP assembles the MAC. In this case, the C5 convertase (C3bBbC3bP) cleaves C5 to C5b. This promotes assembly of C6 + C7 + C8 and multiple C9s to allow perforation (i.e., channel or pore formation) of the foreign membrane. MAC assembly occurs through protein-protein interactions (i.e., no proteolysis is involved after C5 cleavage) to form the lytic complex. Cleavage of C5 also releases the anaphylatoxin C5a, which has potent inflammatory and chemotactic properties.


Four concepts underlie the function of the CP: initiation, enzyme activation, amplification, and attack.


The CP is triggered by an interaction between the C1q subcomponent [see Figure 3] of the C1 complex after it attaches to the Fc portion of antibody (an immune complex). IgM and IgG subclasses 1, 2, and 3 activate the CP, whereas IgA, IgD, IgE, and IgG4 do not. Additionally, C-reactive protein is an acute-phase reactant that activates the CP.21


Figure 3. Structure of C1q and MBL

Similarity of structure between C1q and mannan-binding lectin (MBL, also called mannose-binding protein [MBP]).19 Ficins, surfactins, and other members of the collectin family of proteins share this general structure. Many activate the lectin pathway upon interaction with their ligands.

Enzyme Activation

After C1 binds to antibody, the C1r subcomponent undergoes an autoactivation process, activating C1s by cleavage. C4 and C2 are then cleaved by C1s. Large proteolytic fragments derived from C4 (C4b) and C2 (C2a) (an exception to the nomenclature whereby smaller fragments are designated with an “a”) attach to the target surface, as well as to antibodies, to assemble the C3 convertase, C4b2a. This enzyme complex rapidly converts C3 to C3b. C4, a structural cousin of C3, also possesses the remarkable post-translational thioester modification. Activated C4b (produced by cleavage of the parent protein) has the transient ability to “glue” covalently onto a nearby surface via a hydroxyl group (i.e., to form an ester linkage).


Each activated C1 produces large numbers of C4b and C2a. Most of the C4b serve as opsonins to coat nearby surfaces, whereas others form convertases that rapidly and robustly amplify the system via C3b. Clusters of C3b deposited on a surface serve as ligands for complement receptors.


Whereas some C3bs serve as ligands for complement receptors, others form C5 convertases (i.e., C4b2a3b). These convertases cleave C5 to initiate the MAC.


Lectins are carbohydrate-binding proteins that were initially identified by their capability of agglutinating red blood cells. They are now recognized as important players in innate immunity. The activation scheme of the lectin pathway is similar to that of the CP, except that lectins substitute for antibodies, and associated proteases replace the C1r/C1s subcomponent of C1. In particular, mannan-binding lectin (MBL, also called mannose-binding protein [MBP]) is a plasma protein that preferentially binds to repeating mannoses and other sugars on pathogens [see Figure 3]. The resulting MBL-associated serine proteases (MASPs) cleave C4 and C2. MBL levels are elevated as part of the acute-phase response.22,23


Many of the effects of the complement system (e.g., immune adherence, phagocytosis, cell signaling) result from the interaction of receptors and activation fragments [see Tables 5 and 6].24 For example, complement receptors mediate clearance of immune complexes bearing complement proteins.25 Complement receptor 1 (CR1; CD35) on erythrocytes binds C3b/C4b-coated immune complexes for processing and transport to the liver and spleen. In these organs, immune complexes are transferred from erythrocytes to tissue macrophages, and the erythrocytes then return to the circulation. CR1 on granulocytes and monocytes promotes immune complex adherence and phagocytosis, whereas CR1 on B cells, tissue macrophages, and follicular-dendritic cells (FDCs) facilitates trapping and processing of immune complexes in lymphoid organs. Other receptors for C3b fragments help to localize immune complexes to FDC-rich and B cell-rich areas of the spleen and lymph nodes, where they guide the adaptive immune response. Receptors for anaphylatoxins exert vasomodulatory and chemotactic effects after the binding of C3a and C5a. Finally, receptors in the lectin group enhance phagocytosis.

Table 5 Complement Receptors for C3 and C4

Name (CD Number)

Primary Ligand



CR1 (CD35)


Peripheral blood cells (except for platelets), FDC, B cells, podocytes

Immune adherence, phagocytosis, antigen localization



B cells, FDC

Coreceptor for B cell signaling, antigen localization

CR3 (CD11b/CD18), CR4 (CD11c/CD18)


Myeloid lineage

Phagocytosis, adherence

FDC—follicular dendritic cells

Table 6 Complement Receptors for Anaphylatoxins







Myeloid lineage, including mast cells, smooth muscle, epithelial/endothelial cells

Cell activation including granule exocytosis, upregulation of adhesins, chemotaxis, cytoskeletal effects



Similar to C3aR

Similar to C3a but with more chemotactic effects

Complement regulatory proteins [see Table 7] provide the checks and balances to the system.24 Unregulated, the system would fire to exhaustion, a point well illustrated by inborn errors of several regulatory proteins. The system is designed so that the initial activation is unencumbered on foreign surfaces, yet rapidly inhibited on self-tissue.

Table 7 Complement Regulatory Proteins

Regulation Site


Tissue Distribution


Disease Associated with Deficiency

Initiation of complement cascade

C1 inhibitor


Inactivates C1r, C1s, and MASPs; a SERPIN

Hereditary angioedema


Factor I


Cleaves C3b and C4b; requires a cofactor protein

Infection (secondary to low C3 levels)

Membrane cofactor protein

Most cells

Cofactor for cleavage of C4b and C3b

Hemolytic-uremic syndrome

Decay-accelerating factor

Most cells

Destabilizes C3 and C5 convertases


C4b-binding protein


Cofactor for cleavage of C4b; decays CP C3 and C5 convertases


Factor H


Cofactor for cleavage of C3b; destabilizes AP C3 and C5 convertases

Hemolytic-uremic syndrome; glomerulonephritis

Complement receptor type 1 (CR1)

Blood cells

Receptor for C3b and C4b; cofactor activity for C3b and C4b and decays C3 and C5 convertases



S protein


Blocks fluid-phase MAC



Most cells

Blocks MAC on host cells

Paroxysmal nocturnal hemoglobinuria


Anaphylatoxin inactivator


Inactivates C3a, C4a, and C5a


*Too few cases of complete deficiency to establish an association.
A complete deficiency has not been reported.
AP—alternative pathway CP—classical pathway MAC—membrane attack complex MASPs—mannan-binding lectin-associated serine proteases SERPIN—serine protease inhibitor

Regulation of the early phase of the CP is provided by C1 inhibitor (C1-Inh), which prevents excessive plasma C1 activation. The C3/C5 convertases are regulated by a family of proteins that include two membrane proteins, decay accelerating factor (DAF; CD55) and membrane cofactor protein (MCP; CD46), as well as the serum inhibitors C4b-binding protein and factor H. These proteins function by disassembling the convertases (decay-accelerating activity), by facilitating proteolytic inactivation of C4b or C3b, or by both processes [see Figure 4]. Inactivation occurs in collaboration with the plasma serine protease factor I and is called cofactor activity. The MAC also has regulators in plasma and on cells. Vitronectin (S-protein) binds and inactivates MAC liberated into the fluid phase. On host cells, the glycolipid-anchored CD59 binds C8 and C9 to prevent proper MAC insertion.


Figure 4. Regulation of C3 and C5 Convertases

The membrane proteins decay-accelerating factor (DAF) and membrane cofactor protein (MCP) regulate the C3 and C5 convertases.35These proteins function by disassembling the convertases (decay-accelerating activity), by facilitating proteolytic inactivation, or by both processes. In the classical pathway, the components of C3 convertase are the proteases C4b and C2a. Decay-accelerating activity (a) occurs when DAF binds C4b, displacing the C2a. Proteolytic inactivation (b) occurs when MCP, in concert with the serine protease factor I, cleaves C4b; this prevents C4b from interacting with newly formed C2a. The residual bound C4d has no known biologic activity. In the classical pathway, C5 convertase (which consists of a C4bC2a with an attached C3b) is similarly inactivated by decay-accelerating activity. Although C3 and C5 convertases in the alternative pathway have a different structure, they are disassembled in an identical fashion by DAF and MCP.


In addition to complement's long-appreciated role as a mediator of antibody-directed events, complement activation is instrumental for generating a humoral immune response.26 More than 25 years ago, it was found that mice depleted of C3 by cobra venom factor had markedly impaired primary antibody responses. Similar results were associated with deficiencies or defects in C4, C2, C3, CR1, or CR2. The major findings of these studies were variable but often included a low primary IgM response; a failure to class switch; and, on a second antigenic challenge, a lack of recall (memory). The defective response could be overcome with larger doses of antigen.

These results have given rise to two hypotheses, which are not mutually exclusive.26 One hypothesis suggests that the binding of C3b to an antigen enhances the immune response by promoting transmittal of the antigen to a local lymph node or spleen for uptake, processing, and retention by APCs (especially FDCs). The second hypothesis suggests that the binding of CR2 with C3d complexed on antigen enhances signaling through the B cell antigen receptor. Reconstitution studies are helping to define mechanisms and determine the contributions of CR2 of B cells, relative to the contributions of FDCs. In such studies, B cell deficiency of CD21 (CR2) most closely resembled the knockout, although a contribution of FDCs to long-term memory was apparent.26


The complement system is a double-edged sword. Once unsheathed, it can attack in a robotic fashion. The pathophysiology of many inflammatory diseases involves the synthesis of autoantibodies and the presence of excessive quantities of immune complexes. If the host produces an antibody that reacts to a self-antigen (e.g., on an erythrocyte), the complement cascade becomes activated. Thus, just as complement can destroy a microbe, it may lyse an erythrocyte, opsonize a platelet, or disrupt a basement membrane. If immune complexes lodge in blood vessel walls in a particular tissue, they may activate complement to produce synovitis, vasculitis, dermatitis, or glomerulonephritis. Similarly, a powerful complement barrage may result from ischemia-reperfusion injury as the alternative pathway elicits C3b deposition on the damaged tissue, which is regarded as foreign.

Complement component deficiencies, although rare, predispose to autoimmune diseases (e.g., SLE) and bacterial infections [see Table 8].17,18,27,28 Deficiencies of complement regulatory proteins allow excessive activation of complement cascades [see Table 7]. These conditions are usually inherited as autosomal recessive traits, with the exception of deficiencies of C1-Inh (which is an autosomal dominant trait) and properdin (which is X-linked). The effects of these conditions are predictable, because the affected person experiences a loss of function of the deficient protein and of all the proteins that would ordinarily follow in the cascade. Deficiencies of early components (e.g., C1q, C1r/C1s, C4, and C2) predispose to SLE, whereas deficiency of C3, MBL, or MAC components leads to recurrent bacterial infections.

Table 8 Clinical Manifestations of Complement Deficiency in the Activation Pathways

Pathway Involved

Deficient Component

Clinical Syndrome

Classical pathway


SLE, infections*
SLE, infections*
SLE, infections*
SLE, infections*

Lectin pathway



Central component


Severe infections,* glomerulonephritis, SLE

Membrane perturbation

C5, C6, C7, C8, or C9

Neisseria infections

Alternative pathway

Properdin, factor D

Neisseria infections

*Typically, with commonly encountered pyogenic organisms.
MBL—mannan-binding lectin SLE—systemic lupus erythematosus

C1-Inh deficiency causes hereditary angioedema, whose symptoms range in severity from a minor inconvenience to life-threatening laryngeal edema [see 6:XIII Urticaria, Angioedema, and Anaphylaxis].29 Deficiency of factor H may lead to uncontrolled AP activation on, and damage to, endothelial cells, resulting in hemolytic uremic syndrome or glomerulonephritis.30

Acquired deficiencies of complement also predispose to illness. C1-Inh deficiency may occur as a result of excessive utilization of C1-Inh (usually because of a malignancy) or inactivation of C1-Inh by an autoantibody. An acquired deficiency of C3 may occur as a result of production of an autoantibody that binds and stabilizes the C3 convertase. In the AP, this antibody is called the C3 nephritic factor, whereas in the CP it is termed the C4 nephritic factor. Most patients with C3 nephritic factors are children; they may present with a combination of glomerulonephritis, partial lipodystrophy, and frequent infections with encapsulated bacteria.

An acquired hematopoietic stem cell disorder produces paroxysmal nocturnal hemoglobinuria [see 5:IV Hemoglobinopathies and Hemolytic Anemias]. A mutation in a stem cell prevents expression of an enzyme needed to produce so-called greasy foot (i.e., glycolipid) proteins. As a result, blood cells are deficient in proteins that have this cellular anchor. In particular, deficiencies of CD59 and DAF predispose erythrocytes to complement-mediated hemolytic anemia.


Complement levels can be assessed using either antigenic or functional assays [see Table 9]. The former are easier to perform and are most commonly employed for measuring C3 and C4 levels. The total hemolytic complement (THC or CH50) assay measures activation of the entire CP by assessing the ability of the test serum to lyse sheep erythrocytes optimally sensitized with rabbit antibody. Interpretation of the results is rather straightforward [see Table 10]. Decreased C4 and C3 levels almost always indicate CP activation, whereas AP activation is indicated by normal levels of C4 but low levels of C3 (and factor B, if measured). All nine components of the CP (C1 through C9) are needed to obtain a normal result on the CH50 assay. A CH50 of 200 means the tested serum lysed 50% of the antibody-coated sheep erythrocytes when assayed at a dilution of 1:200. A similar assay, AH50, measures the total alternative pathway, with the target for lysis being unsensitized rabbit red blood cells. Less widely used tests include measurement of the anaphylatoxins C5a and C3a or activation fragments, such as C3d and Bb. Tests showing increased levels of these substances have the advantage of reflecting ongoing activation and are more sensitive. In addition, specialized laboratories can determine the functional and antigenic levels of each of the complement components and regulators.

Table 9 Assays for Complement Activation in Human Disease




CH50 or THC

Screen for a component deficiency or activation of classical pathway

Functional assay; requires appropriate sample handling


Screen for component deficiency or activation of alternative pathway

Functional assay; requires appropriate sample handling

Antigenic (ELISA, immunodiffusion, nephelometry)

Standard method for C3, C4, factor B, C1-inhibitor, and MBL determinations

Widely available, easy to perform, reliable, inexpensive

Antigenic or hemolytic assay of individual components

To further define a suspected deficiency

Samples usually sent to laboratories specializing in complement assays

Activation fragments C3a, C5a, Bb C1-INH; C1r/C1s C5b-9 (neoantigen)

Additional tests to detect complement turnover

More sensitive than static levels; sample collection technique important; often available through commercial laboratories; expensive

C1-inhibitor function

When clinical picture is consistent with HAE but C1-inhibitor levels by antigenic assay are normal or elevated

15% of HAE patients have normal or elevated levels of a nonfunctional protein


Demonstration of complement activation fragments in tissue

C1q, C4, and C3 are most commonly studied in kidney and skin biopsy specimens

Antiglobulin testing (nongamma Coombs)

Demonstration of C3 fragments on erythrocytes

Usual fragment detected is C3d

AP50—alternative pathway equivalent of THC or CH50 ELISA—enzyme-linked immunosorbent assay HAE—hereditary angioedema MBL—mannan-binding lectin
THC—total hemolytic assay for classical pathway (also called CH50)

Table 10 Interpretation of Complement Assays

Test result


THC (units/ml)

C4 (mg/dl)

C3 (mg/dl)

< 10 or 0

< 8


Normal range
Acute-phase response
CP activation
AP activation
Inherited deficiency or in vitro activation*
Partial C4 deficiency or fluid-phase activation

*In vitro activation is more common than an inherited deficiency state. The lack of activity (< 10 THC) in the setting of normal C4 and C3 antigenic levels suggests (1) an improperly handled sample, (2) cold activation (such as by cryoglobulins) after collection of the sample, or (3) homozygous component deficiency (most commonly C2 with a lupus presentation or, if a Neisseria infection is present, of an AP or membrane-attack-complex component).
Detectable THC excludes a complete deficiency of C4. A partial C4 deficiency, such as of C4A, could give this result. Some types of immune complexes, especially cryoglobulins, and a deficiency of the C1-inhibitor (hereditary angioedema) also give this pattern. In these cases, measurement of C2 is often helpful: a low value suggests activation, whereas a normal value suggests an inherited, partial C4 deficiency. Also, C4A and C4B alleles can be assessed by commercial laboratories.
AP—alternative pathway CP—classical pathway THC—total hemolytic complement (also called CH50)


Currently, no therapeutic agent is commercially available to block the deleterious effects of pathologic complement activation.31,32,33 Two plasma-based therapies are undergoing clinical trials. A humanized monoclonal antibody to C5 adopts the straightforward strategy of blocking the function of a single component; specifically, it blocks cleavage of C5 by C5 convertases. A significant advantage of this approach is that recombinantly produced humanized monoclonal antibodies are already an established therapy in clinical medicine (e.g., anti-TNF). Recombinant monoclonal antibodies have a relatively long half-life and, if humanized, are usually nonimmunogenic. However, it is unclear how much tissue damage in clinical syndromes results from the activation of C5 as compared with that caused by C3. An advantage of this therapeutic approach is that activation up to and including C3 remains intact.

A second compound in clinical trials is a solubilized version of CR1 (sCR1), which degrades C3b/C4b and decays C3/C5 convertases. Proof of principle has been established in animal models, as it has been for the monoclonal antibody to C5. Potential barriers to the use of sCR1 in clinical practice include cost, lack of selectivity, the need for intravenous administration, and the potential for infectious complications.

The search for a complement therapeutic has led to two unanticipated observations. First, complement activation contributes to tissue damage in ischemia-reperfusion injury syndromes such as myocardial infarctions and stroke; second, C5b-C9 and C5a bear responsibility for more of the tissue damage during complement activation than predicted.


Figures 1 through 4 Seward Hung.


  1. Cross AS, Opal SM: A new paradigm for the treatment of sepsis: is it time to consider combination therapy? Ann Intern Med 138:502, 2003
  2. Janeway CAJ: Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb Symp Quant Biol 54:1, 1989
  3. Medzhitov R, Preston-Hurlburt P, Janeway CAJ: A human homologue of the DrosophilaToll protein signals activation of adaptive immunity. Nature 388:394, 1997
  4. Akira S, Takeda K, Kaisho T: Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2:675, 2001
  5. Revillard J-P: Innate immunity. Eur J Dermatol 12:224, 2002
  6. Bendelac A, Fearon DT: Innate immunity: receptors and effectors of innate immunity. Curr Opin Immunol 12:11, 2000
  7. Ellis J, Dodds P, Pryor T: Structure, function and evolution of plant disease resistance genes. Curr Opin Plant Biol 3:278, 2000
  8. Yokoyama W: Natural killer cells. Fundamental Immunology Paul WE, Ed. Lippincott-Raven, New York, 1999, p 575
  9. Ljunggren HG, Karre K: In search of the ‘missing self’: MHC molecules and NK cell recognition. Immunol Today 11:237, 1990
  10. Rosenberg SA, Lotze MT, Muul LM, et al: A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high-dose interleukin-2 alone. N Engl J Med 316:889, 1987
  11. Biron CA, Byron KS, Sullivan JL: Severe herpesvirus infections in an adolescent without natural killer cells (comment). N Engl J Med 320:1731, 1989
  12. Lusso P, Malnati MS, Garzino-Demo A, et al: Infection of natural killer cells by human herpesvirus 6. Nature 362:458, 1993
  13. Jaffe ES, Lindskog S, Ahren B: Report of the workshop on nasal and related extra-nodal angiocentric T/natural killer cell lymphomas: definitions, differential diagnosis, and epidemiology. Am J Surg Pathol 20:103, 1996
  14. Tak PP, Kummer JA, Hack CE, et al: Granzyme-positive cytotoxic cells are specifically increased in early rheumatoid synovial tissue. Arthritis Rheum 37:1735, 1994
  15. Linehan SA, Martinez-Pomares L, Gordon S: Macrophage lectins in host defence. Microbes Infect 2:279, 2000
  16. Volanakis JE: Overview of the complement system. The Human Complement System in Health and Disease, 10th ed. Volanakis JE, Frank MM, Eds. Marcel Dekker, New York, 1998, p 97
  17. Walport MJ: Complement. First of two parts. N Engl J Med 344:1058, 2001
  18. Walport MJ: Complement. Second of two parts. N Engl J Med 344:1140, 2001
  19. Frank MM, Atkinson JP: Complement system. Samter's Immunologic Diseases, 6th ed. Austen KF, Frank MM, Atkinson JP, et al, Eds. Lippincott Williams & Wilkins, Baltimore, 2001, p 281
  20. Ollert MW, Kadlec JV, David K, et al: Antibody-mediated complement activation on nucleated cells: a quantitative analysis of the individual reaction steps. J Immunol 153:2213, 1994
  21. DuClos TW: Function of C-reactive protein. Ann Med 32:274, 2000
  22. Turner MW: Mannose-binding lectin: the pluripotent molecule of the innate immune system. Immunol Today 17:532, 1996
  23. Fujita T: Evolution of the lectin-complement pathway and its role in innate immunity. Nat Rev Immunol 2:346, 2002
  24. Liszewski MK, Farries TC, Lublin DM, et al: Control of the complement system. Adv Immunol 61:201, 1996
  25. Hebert LA: The clearance of immune complexes from the circulation of man and other primates. Am J Kidney Dis 17:353, 1991
  26. Nielsen CH, Leslie RG: Complement's participation in acquired immunity. J Leukoc Biol 72:249, 2002
  27. Barilla-LaBarca ML, Atkinson JP: Rheumatic syndromes associated with complement deficiency. Curr Opin Rheumatol 15:55, 2003
  28. Pickering MC, Botto M, Taylor PR, et al: Systemic lupus erythematosus, complement deficiency, and apoptosis. Adv Immunol 76:227, 2000
  29. Davis AE III: C1 inhibitor gene and hereditary angioedema. The Human Complement System in Health and Disease. Volanakis JE, Frank MM, Eds. Marcel Dekker, New York, 1998, p 455
  30. Zipfel PF: Hemolytic uremic syndrome: how do factor H mutants mediate endothelial damage? Trends Immunol 22:345, 2001
  31. Liszewski MK, Atkinson JP: Complement inhibition. Biological Therapy in Rheumatology. Smolen J, Lipsky P, Eds. Martin Dunitz, London, 2002, p 453
  32. Kirshfink M: Targeting complement in therapy. Immunol Rev 180:177, 2001
  33. Klickstein LB, Moore FD Jr, Atkinson JP: Therapeutic inhibition of complement activation with emphasis on drugs in clinical trials. Therapeutic Immunology Austen KF, Burakoff SJ, Strom TB, et al, Eds. Blackwell Science, Malden, MA, 2000, p 287
  34. Liszewski MK, Atkinson JP: The complement system. Fundamental Immunology 3rd ed. Paul WE, Ed. Raven Press, New York, 1993, p 917
  35. Liszewski MK, Farries TC, Lublin DM, et al: Control of the complement system. Adv Immunol 61:201, 1996
  36. Medzhitov R: Toll-like receptors and innate immunity. Nat Rev Immunol 1:135, 2001

Editors: Dale, David C.; Federman, Daniel D.